From umari at MIT.EDU  Tue Aug  1 10:25:17 2006
From: umari at MIT.EDU (Paolo Umari)
Date: Tue, 01 Aug 2006 04:25:17 -0400
Subject: [Pw_forum] Re: AlAs in field
Message-ID: <20060801042517.46hwfp4veb1wc84k@webmail.mit.edu>

Dear Vivek

Unfortunately I couldn't get your files from the
attached file. Could you attach them  in a text form,
so I can try to see where is the problem?
Does the code crash also without electric field?

Paolo



From hqzhou at nju.edu.cn  Wed Aug  2 08:04:09 2006
From: hqzhou at nju.edu.cn (Huiqun Zhou)
Date: Wed, 2 Aug 2006 14:04:09 +0800
Subject: [Pw_forum] Crash when running pw.x for relaxing a structure
References: <000f01c6aa53$f8073ff0$1d00a8c0@solarflare> <200607241825.12293.giannozz@nest.sns.it>
Message-ID: <000b01c6b5f9$78579e10$1d00a8c0@solarflare>

Paolo,

Thanks for your response. I have made more calculations these days. Below 
are
some facts:

(1) The system I'm investigating is with orthohombic unit cell. The volume 
is fixed
  at 1570.0 bohr^3, and the calculations were carried on by changing
  b/a = 0.275 - 0.400 : 0.025
  c/a = 0.975 - 1.200 : 0.025
(2) Not all calculations will fail at charge density extrapolation. For 
example, when
  b/a = 0.375, the calculation will fail only when c/a = 1.200. This error 
is highly
  reproducible.
(3) The error occurrs only in the calculation on 4 CPU cores, and it's no 
problem
  when running on one or two CPU cores. This is true for all failed cases.
(4) This problem can be duplicated on compute nodes with different dual core
  processors, such as Intel dempsey (Xeon 5060), woodcrest (Xeon 5140), and
 AMD opteron 280.

The OS of my cluster is RHEL4 U3 (kernel 2.6.9-34). I'm using Intel FORTRAN
9.0, MKL 8.0 and FFTW 2.1.5. I have no access to other commercial compilers,
and failed to compile QE with g95.

The attached zip file includes input for b/a=0.375, c/a=1.200, and output 
files of the
results on 1 core (successful) and 4 cores (failure).

Thanks again for your help.

Huiqun Zhou


----- Original Message ----- 
From: "Paolo Giannozzi" <giannozz at nest.sns.it>
To: <pw_forum at pwscf.org>
Sent: Tuesday, July 25, 2006 12:25 AM
Subject: Re: [Pw_forum] Crash when running pw.x for relaxing a structure


> On Tuesday 18 July 2006 12:21, Huiqun Zhou wrote:
>
>> I'm doing structural optimization for chromite with calcium ferrite
>> structure while changing b/a and c/a at fixed volume. But for every
>> run with different pair of b/a and c/a, I alway got following error
>> after 3-5 rounds of SCF calculations:
>
>>      Writing output data file fecr2o4-cf-relax.save
>>
>>      second order charge density extrapolation
>> rank 1 in job 170  woodcrest_32906   caused collective abort of all ranks
>>   exit status of rank 1: return code 220
>
>> The job was running parallely on one compute node with 4 CPU cores
>> (Intel woodcrest).
>>
>> Did I do anything wrong?
>
> difficult to say. Is it reproducible? does it happen on other machines
> or with other compilers or in serial execution ? If it is not reproducible
> it may not be related to the code itself
>
> P.
> -- 
> Paolo Giannozzi             Phone:   +39/050-509876
> DEMOCRITOS and SNS          Fax:     +39/050-563513
> Piazza dei Cavalieri 7      I-56126 Pisa, Italy
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
> 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: fecr2o4-cf.zip
Type: application/octet-stream
Size: 25406 bytes
Desc: not available
Url : /pipermail/attachments/20060802/54688685/attachment.obj 

From hqzhou at nju.edu.cn  Wed Aug  2 12:09:33 2006
From: hqzhou at nju.edu.cn (Huiqun Zhou)
Date: Wed, 2 Aug 2006 18:09:33 +0800
Subject: [Pw_forum] Re: Woodcrest vs Opteron performance in pwscf calc.
References: <1154501849.2719.7.camel@m01>
Message-ID: <000c01c6b61b$c10d3bc0$1d00a8c0@solarflare>

Hi, Alexander,

A short answer is: Woodcrest (2.66 GHz) > Dempsey (3.2 GHz) > Opteron 280 
(2.6 GHz).

The compute nodes with woodcrest and dempsey in my cluster have 4GB memory, 
one SATA
disk. The nodes with opteron 280 are SUN Fire x4100, each has 8GB memory and 
two SAS
disks.

The test case is total energy calculation of MgAl2O4 (calcium ferrite 
structure). The system has
28 atoms (orthohombic system, Z=4). It took the same number of iterations to 
reach SCF
convergency on the 3 different machines and got physically same results. The 
number in
parantheses is the elapsed time given by PWscf.

Here are numbers
(1) woodcrest (2.66 GHz):
1 core : 3m57s (3m55.86s)
2 cores: 2m11s (2m10.44s)
4 cores: 1m23s (1m17.73s)

(2) dempsey (3.2 GHz)
1 core : (6m26.90s)
2 cores: (3m16.47s)
4 cores: (1m39.74s)

(3) opteron 280 (2.6 GHz)
1 core : 7m13s (7m09.71s)
2 cores: 3m56s (3m52.70s)
4 cores: 2m26s (2m16.72s)

It seems that woodcrest and dempsey are much faster than opteron. The 
scalability of
dempsey is the best, woodcrest is the worst. Despite of the amazing 
performance per
core of woodcrest, it drops to the same level of its predecessor, dempsey, 
when taking
the machine as a unit to evaluate its performance.

But remember one thing: the number for opteron may not be fair. I compiled 
the program
using Intel fortran, Intel MPI 2.0. However, I ever used both Intel and 
PathScale to
compile FFTW and its test cases on opteron machine, I didn't find any 
impressive
differences.

Hope it help you.


Huiqun Zhou


----- Original Message ----- 
From: "Alexander Shaposhnikov" <shaposh at isp.nsc.ru>
To: "Huiqun Zhou" <hqzhou at nju.edu.cn>
Sent: Wednesday, August 02, 2006 2:57 PM
Subject: Woodcrest vs Opteron performance in pwscf calc.


> Hello dear sir,
>
> i have noticed from your posts to PW_forum that you
> have access to both newest Intel Woodcrest platform and
> AMD Opteron. Could you compare the performance of these two
> platforms, for single-threaded, and parallel 4-core pwscf calculations?
>
> Thank you in advance,
> Best Regards,
> A. Shaposhnikov
>
>
>
> 



From shaposh at isp.nsc.ru  Wed Aug  2 12:46:28 2006
From: shaposh at isp.nsc.ru (Alexander Shaposhnikov)
Date: Wed, 02 Aug 2006 17:46:28 +0700
Subject: [Pw_forum] Re: Woodcrest vs Opteron performance in pwscf calc.
In-Reply-To: <000c01c6b61b$c10d3bc0$1d00a8c0@solarflare>
References: <1154501849.2719.7.camel@m01>
	 <000c01c6b61b$c10d3bc0$1d00a8c0@solarflare>
Message-ID: <1154515588.2719.19.camel@m01>

Thanks for the answer. 
I don't think this topic is relevant to the PW_forum goals. 

Could you send me the job's input file ? I'd like to test this and
compare with my own Opteron system. We have several Opteron machines,
and if the performance difference is really that big, probably its the
time to move on to the new platform. But my experience with the prev.
generation of Intel processors showed, that Opteron is faster is most
cases, especially then it comes to multi-threaded calculations.

Best Regards,
A. Shaposhnikov


On Wed, 2006-08-02 at 18:09 +0800, Huiqun Zhou wrote:
> Hi, Alexander,
> 
> A short answer is: Woodcrest (2.66 GHz) > Dempsey (3.2 GHz) > Opteron 280 
> (2.6 GHz).
> 
> The compute nodes with woodcrest and dempsey in my cluster have 4GB memory, 
> one SATA
> disk. The nodes with opteron 280 are SUN Fire x4100, each has 8GB memory and 
> two SAS
> disks.
> 
> The test case is total energy calculation of MgAl2O4 (calcium ferrite 
> structure). The system has
> 28 atoms (orthohombic system, Z=4). It took the same number of iterations to 
> reach SCF
> convergency on the 3 different machines and got physically same results. The 
> number in
> parantheses is the elapsed time given by PWscf.
> 
> Here are numbers
> (1) woodcrest (2.66 GHz):
> 1 core : 3m57s (3m55.86s)
> 2 cores: 2m11s (2m10.44s)
> 4 cores: 1m23s (1m17.73s)
> 
> (2) dempsey (3.2 GHz)
> 1 core : (6m26.90s)
> 2 cores: (3m16.47s)
> 4 cores: (1m39.74s)
> 
> (3) opteron 280 (2.6 GHz)
> 1 core : 7m13s (7m09.71s)
> 2 cores: 3m56s (3m52.70s)
> 4 cores: 2m26s (2m16.72s)
> 
> It seems that woodcrest and dempsey are much faster than opteron. The 
> scalability of
> dempsey is the best, woodcrest is the worst. Despite of the amazing 
> performance per
> core of woodcrest, it drops to the same level of its predecessor, dempsey, 
> when taking
> the machine as a unit to evaluate its performance.
> 
> But remember one thing: the number for opteron may not be fair. I compiled 
> the program
> using Intel fortran, Intel MPI 2.0. However, I ever used both Intel and 
> PathScale to
> compile FFTW and its test cases on opteron machine, I didn't find any 
> impressive
> differences.
> 
> Hope it help you.
> 
> 
> Huiqun Zhou
> 
> 
> ----- Original Message ----- 
> From: "Alexander Shaposhnikov" <shaposh at isp.nsc.ru>
> To: "Huiqun Zhou" <hqzhou at nju.edu.cn>
> Sent: Wednesday, August 02, 2006 2:57 PM
> Subject: Woodcrest vs Opteron performance in pwscf calc.
> 
> 
> > Hello dear sir,
> >
> > i have noticed from your posts to PW_forum that you
> > have access to both newest Intel Woodcrest platform and
> > AMD Opteron. Could you compare the performance of these two
> > platforms, for single-threaded, and parallel 4-core pwscf calculations?
> >
> > Thank you in advance,
> > Best Regards,
> > A. Shaposhnikov
> >
> >
> >
> > 
> 
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum



From akohlmey at vitae.cmm.upenn.edu  Wed Aug  2 14:07:34 2006
From: akohlmey at vitae.cmm.upenn.edu (Axel Kohlmeyer)
Date: Wed, 2 Aug 2006 08:07:34 -0400 (EDT)
Subject: [Pw_forum] Re: Woodcrest vs Opteron performance in pwscf calc.
In-Reply-To: <1154515588.2719.19.camel@m01>
Message-ID: <Pine.LNX.4.44.0608020749470.8867-100000@localhost.localdomain>

On Wed, 2 Aug 2006, Alexander Shaposhnikov wrote:

hi,

AS> Thanks for the answer. 
AS> I don't think this topic is relevant to the PW_forum goals. 

actually, it is. some people here spend a lot of money 
on new machines and figuring out what is the best deal
for a specific application needs a lot of testing. so
every contribution is important.

[...]
AS> time to move on to the new platform. But my experience with the prev.
AS> generation of Intel processors showed, that Opteron is faster is most
AS> cases, especially then it comes to multi-threaded calculations.

really?? are you talking about OpenMP multi-threaded or explicit
multi-threaded? in my experience so far, OpenMP on an opteron system
was a serious letdown, and it was usually much better to use 
MPI parallelism, even within the nodes.

[...]

AS> > It seems that woodcrest and dempsey are much faster than opteron. The 
AS> > scalability of
AS> > dempsey is the best, woodcrest is the worst. Despite of the amazing 
AS> > performance per
AS> > core of woodcrest, it drops to the same level of its predecessor, dempsey, 
AS> > when taking
AS> > the machine as a unit to evaluate its performance.

one thing to check when using intel MKL is, whether it is running in
multi-threaded mode and thus getting better results on a 'half-loaded'
machines. for that, you may want to re-run the jobs with the 
environment variable OMP_NUM_THREADS set to 1. secondly, memory 
contention is a problem, so it would be interesting to see the 
performance, if you run 4 serial jobs at the same time.

AS> > But remember one thing: the number for opteron may not be fair. I compiled 
AS> > the program
AS> > using Intel fortran, Intel MPI 2.0. However, I ever used both Intel and 
AS> > PathScale to
AS> > compile FFTW and its test cases on opteron machine, I didn't find any 
AS> > impressive
AS> > differences.

the intel compiler ususally does a good job on opteron. especially,
since for floating point intensive jobs don't benefit a lot if at
all from using the atomated vectorization with SSE. usually get
the best performance on opteron and P4 with '-O2 -tpp6 -unroll' 
not using any vectorization. that however is a different story
when it comes to BLAS/LAPACK: using ACML > 2.7 is essential to
get good performance on dual-core opteron machines.

there is a way to make (the gcc) ACML compatible with the intel
compiler (at least for packages that use only double precision
functions), see:
https://www.liniac.upenn.edu/wiki/tiki-index.php?page=acml+for+CMM

it would be nice to see, how using ACML would affect 
the performance in this case.


best regards,
    axel.

-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.



From claudio.attaccalite at gmail.com  Wed Aug  2 15:58:08 2006
From: claudio.attaccalite at gmail.com (Claudio Attaccalite)
Date: Wed, 2 Aug 2006 15:58:08 +0200
Subject: [Pw_forum] Band Strcture calculation with PW
Message-ID: <6b6a635f0608020658y79f8932ek75be90f556e3c06c@mail.gmail.com>

Hi I just
stared to use PWSCF and I have a problem to calculate the band structure of
an hexagonal BN sheet
It seems that the conduction states do not converge, any hints?
This are my input files:


**************** BN SCF ****************
 &control
    calculation = 'scf'
    restart_mode='from_scratch',
    prefix='hbn',
    pseudo_dir = '/home/attacc/espresso-3.1.1/pseudo/',
    outdir='/home/attacc/tmp/'
 /
 &system
    ibrav=  4
    celldm(1) = 4.7177372151
    celldm(3) = 10
    nat=  2
    ntyp= 2
    ecutwfc =25.0,
 /
 &electrons
    diagonalization='cg'
    mixing_mode = 'plain'
    mixing_beta = 0.7
    conv_thr =  1.0d-8
 /
ATOMIC_SPECIES
 B   10.811     B.pz-vbc.UPF
 N   14.00674   N.pz-vbc.UPF
ATOMIC_POSITIONS crystal
 B 0 0 0
 N 0.333333333333333 0.66666666666666666 0
K_POINTS automatic
  6 6 1 0 0 0
*****************************************************

********************* BN bands ******************
 &control
    calculation='nscf'
    pseudo_dir = '/home/attacc/espresso-3.1.1/pseudo/',
    outdir='/home/attacc/tmp/',
    prefix='hbn'
    verbosity = 'high'
 /
 &system
    ibrav=  4
    celldm(1) = 4.7177372151
    celldm(3) = 10
    nat=  2
    ntyp= 2
    ecutwfc =25.0,
    nbnd = 8
 /
 &electrons
  diago_full_acc = .TRUE.
 /
ATOMIC_SPECIES
 B   10.811     B.pz-vbc.UPF
 N   14.00674   N.pz-vbc.UPF
ATOMIC_POSITIONS crystal
 B 0 0 0
 N 0.333333333333333 0.66666666666666666 0
K_POINTS
25
0 0 0 1
.0333333333     .0333333333 0 1
.0666666666     .0666666666 0 1
.0999999999     .0999999999 0 1
.1333333332     .1333333332 0 1
.1666666665     .1666666665 0 1
.1999999998     .1999999998 0 1
.2333333331     .2333333331 0 1
.2666666664     .2666666664 0 1
.2999999997     .2999999997 0 1
.3333333330     .3333333330 0 1
.277777777      .361111111 0 1
.222222222      .388888888 0 1
.166666666      .416666666 0 1
.111111111      .444444444 0 1
.055555555      .472222222 0 1
0               0.5        0 1
0       .4375000000 0 1
0       .3750000000 0 1
0       .3125000000 0 1
0       .2500000000 0 1
0       .1875000000 0 1
0       .1250000000 0 1
0       .0625000000 0 1
0 0 0  1

Thanks in advance
Claudio


==========================================================
Claudio Attaccalite

Institute for Electronics, Microelectronics, and Nanotechnology
Dept. ISEN
B.P. 60069
59652 Villeneuve d'Ascq Cedex France
Skype: claudioattaccalite
web site: www.attaccalite.altervista.org
===========================================================

Freely download scientific books from: www.freescience.info
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060802/cf21a3ae/attachment.htm 

From swblelia at ehu.es  Thu Aug  3 10:05:02 2006
From: swblelia at ehu.es (Aritz Leonardo Liceranzu)
Date: Thu, 3 Aug 2006 10:05:02 +0200
Subject: [Pw_forum] plotting psi2
Message-ID: <200608031005.02611.swblelia@ehu.es>

Hello All:

I pretend to plot the charge-density of the surface state at M point for a 
slab of Mg.

First question, is it possible?

From the input of PP I guess I should pick 
plot_num=7

But, how do I specify the K point and the energy of the state I would like to 
draw?

Thanks a lot
Aritz


From hqzhou at nju.edu.cn  Thu Aug  3 10:41:25 2006
From: hqzhou at nju.edu.cn (Huiqun Zhou)
Date: Thu, 3 Aug 2006 16:41:25 +0800
Subject: [Pw_forum] Re: Woodcrest vs Opteron performance in pwscf calc.
References: <1154501849.2719.7.camel@m01> <000c01c6b61b$c10d3bc0$1d00a8c0@solarflare> <1154515588.2719.19.camel@m01>
Message-ID: <006e01c6b6d8$9b733e10$1d00a8c0@solarflare>

Here they are. Looking forward to hearing your story.

Huiqun Zhou

----- Original Message ----- 
From: "Alexander Shaposhnikov" <shaposh at isp.nsc.ru>
To: <pw_forum at pwscf.org>
Sent: Wednesday, August 02, 2006 6:46 PM
Subject: Re: [Pw_forum] Re: Woodcrest vs Opteron performance in pwscf calc.


> Thanks for the answer.
> I don't think this topic is relevant to the PW_forum goals.
>
> Could you send me the job's input file ? I'd like to test this and
> compare with my own Opteron system. We have several Opteron machines,
> and if the performance difference is really that big, probably its the
> time to move on to the new platform. But my experience with the prev.
> generation of Intel processors showed, that Opteron is faster is most
> cases, especially then it comes to multi-threaded calculations.
>
> Best Regards,
> A. Shaposhnikov
>
>
> On Wed, 2006-08-02 at 18:09 +0800, Huiqun Zhou wrote:
>> Hi, Alexander,
>>
>> A short answer is: Woodcrest (2.66 GHz) > Dempsey (3.2 GHz) > Opteron 280
>> (2.6 GHz).
>>
>> The compute nodes with woodcrest and dempsey in my cluster have 4GB 
>> memory,
>> one SATA
>> disk. The nodes with opteron 280 are SUN Fire x4100, each has 8GB memory 
>> and
>> two SAS
>> disks.
>>
>> The test case is total energy calculation of MgAl2O4 (calcium ferrite
>> structure). The system has
>> 28 atoms (orthohombic system, Z=4). It took the same number of iterations 
>> to
>> reach SCF
>> convergency on the 3 different machines and got physically same results. 
>> The
>> number in
>> parantheses is the elapsed time given by PWscf.
>>
>> Here are numbers
>> (1) woodcrest (2.66 GHz):
>> 1 core : 3m57s (3m55.86s)
>> 2 cores: 2m11s (2m10.44s)
>> 4 cores: 1m23s (1m17.73s)
>>
>> (2) dempsey (3.2 GHz)
>> 1 core : (6m26.90s)
>> 2 cores: (3m16.47s)
>> 4 cores: (1m39.74s)
>>
>> (3) opteron 280 (2.6 GHz)
>> 1 core : 7m13s (7m09.71s)
>> 2 cores: 3m56s (3m52.70s)
>> 4 cores: 2m26s (2m16.72s)
>>
>> It seems that woodcrest and dempsey are much faster than opteron. The
>> scalability of
>> dempsey is the best, woodcrest is the worst. Despite of the amazing
>> performance per
>> core of woodcrest, it drops to the same level of its predecessor, 
>> dempsey,
>> when taking
>> the machine as a unit to evaluate its performance.
>>
>> But remember one thing: the number for opteron may not be fair. I 
>> compiled
>> the program
>> using Intel fortran, Intel MPI 2.0. However, I ever used both Intel and
>> PathScale to
>> compile FFTW and its test cases on opteron machine, I didn't find any
>> impressive
>> differences.
>>
>> Hope it help you.
>>
>>
>> Huiqun Zhou
>>
>>
>> ----- Original Message ----- 
>> From: "Alexander Shaposhnikov" <shaposh at isp.nsc.ru>
>> To: "Huiqun Zhou" <hqzhou at nju.edu.cn>
>> Sent: Wednesday, August 02, 2006 2:57 PM
>> Subject: Woodcrest vs Opteron performance in pwscf calc.
>>
>>
>> > Hello dear sir,
>> >
>> > i have noticed from your posts to PW_forum that you
>> > have access to both newest Intel Woodcrest platform and
>> > AMD Opteron. Could you compare the performance of these two
>> > platforms, for single-threaded, and parallel 4-core pwscf calculations?
>> >
>> > Thank you in advance,
>> > Best Regards,
>> > A. Shaposhnikov
>> >
>> >
>> >
>> >
>>
>> _______________________________________________
>> Pw_forum mailing list
>> Pw_forum at pwscf.org
>> http://www.democritos.it/mailman/listinfo/pw_forum
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
> 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mgal2o4-cf.scf.in
Type: application/octet-stream
Size: 1571 bytes
Desc: not available
Url : /pipermail/attachments/20060803/36b160cd/attachment.obj 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: test.sh
Type: application/octet-stream
Size: 183 bytes
Desc: not available
Url : /pipermail/attachments/20060803/36b160cd/attachment-0001.obj 

From hqzhou at nju.edu.cn  Thu Aug  3 11:10:54 2006
From: hqzhou at nju.edu.cn (Huiqun Zhou)
Date: Thu, 3 Aug 2006 17:10:54 +0800
Subject: [Pw_forum] Re: Woodcrest vs Opteron performance in pwscf calc.
References: <Pine.LNX.4.44.0608020749470.8867-100000@localhost.localdomain>
Message-ID: <007301c6b6dc$b9e26480$1d00a8c0@solarflare>

Hi, Axel and list users,

>
> actually, it is. some people here spend a lot of money
> on new machines and figuring out what is the best deal
> for a specific application needs a lot of testing. so
> every contribution is important.
>

So, I continue to post here ;-)

>
> one thing to check when using intel MKL is, whether it is running in
> multi-threaded mode and thus getting better results on a 'half-loaded'
> machines. for that, you may want to re-run the jobs with the
> environment variable OMP_NUM_THREADS set to 1. secondly, memory
> contention is a problem, so it would be interesting to see the
> performance, if you run 4 serial jobs at the same time.
>

OMP_NUM_THREADS had been set to 1 in all my tests.

>
> there is a way to make (the gcc) ACML compatible with the intel
> compiler (at least for packages that use only double precision
> functions), see:
> https://www.liniac.upenn.edu/wiki/tiki-index.php?page=acml+for+CMM
>
> it would be nice to see, how using ACML would affect
> the performance in this case.
>

If anyone can contribute a makefile for using PathScale compiler, the test 
will
be handy. I have pathscale version of ACML 2.8. Unfortunately, I haven't
succeded in compiling QE with this compiler yet.

Best Regards,


Huiqun Zhou



From shaposh at isp.nsc.ru  Thu Aug  3 12:55:41 2006
From: shaposh at isp.nsc.ru (Alexander Shaposhnikov)
Date: Thu, 03 Aug 2006 17:55:41 +0700
Subject: [Pw_forum] Re: Woodcrest vs Opteron performance in pwscf calc.
In-Reply-To: <007301c6b6dc$b9e26480$1d00a8c0@solarflare>
References: <Pine.LNX.4.44.0608020749470.8867-100000@localhost.localdomain>
	 <007301c6b6dc$b9e26480$1d00a8c0@solarflare>
Message-ID: <1154602541.6549.11.camel@m01>

My experience with Pathscale compiler shows that it  does not give any
substantial improvement over Intel compiler for AMD, in most cases.
Intel compiler is also more compatible with different FORTRAN codes,
and, last but not least, it is free for non-commercial use.
By the way, gcc-compiled ACML is  compatible with Intel compiler.
I use it. Intel compiler can even be used to compile and run code on AMD
with -xP optimization, i.e. with SSE3 instructions. But performance
improvement (if  any) is  negligible.

I will run tests and post results as soon as i get access to the
machines.

Best Regards,
A. Shaposhnikov

On Thu, 2006-08-03 at 17:10 +0800, Huiqun Zhou wrote:
> Hi, Axel and list users,
> 
> >
> > actually, it is. some people here spend a lot of money
> > on new machines and figuring out what is the best deal
> > for a specific application needs a lot of testing. so
> > every contribution is important.
> >
> 
> So, I continue to post here ;-)
> 
> >
> > one thing to check when using intel MKL is, whether it is running in
> > multi-threaded mode and thus getting better results on a 'half-loaded'
> > machines. for that, you may want to re-run the jobs with the
> > environment variable OMP_NUM_THREADS set to 1. secondly, memory
> > contention is a problem, so it would be interesting to see the
> > performance, if you run 4 serial jobs at the same time.
> >
> 
> OMP_NUM_THREADS had been set to 1 in all my tests.
> 
> >
> > there is a way to make (the gcc) ACML compatible with the intel
> > compiler (at least for packages that use only double precision
> > functions), see:
> > https://www.liniac.upenn.edu/wiki/tiki-index.php?page=acml+for+CMM
> >
> > it would be nice to see, how using ACML would affect
> > the performance in this case.
> >
> 
> If anyone can contribute a makefile for using PathScale compiler, the test 
> will
> be handy. I have pathscale version of ACML 2.8. Unfortunately, I haven't
> succeded in compiling QE with this compiler yet.
> 
> Best Regards,
> 
> 
> Huiqun Zhou
> 
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum



From cazzato at democritos.it  Thu Aug  3 14:53:18 2006
From: cazzato at democritos.it (cazzato at democritos.it)
Date: Thu, 03 Aug 2006 14:53:18 +0200
Subject: [Pw_forum] Re: Woodcrest vs Opteron performance in pwscf calc.
In-Reply-To: <007301c6b6dc$b9e26480$1d00a8c0@solarflare>
References: <Pine.LNX.4.44.0608020749470.8867-100000@localhost.localdomain>
	<007301c6b6dc$b9e26480$1d00a8c0@solarflare>
Message-ID: <20060803145318.bcsdk1nddssg0488@mail.democritos.it>

Quoting Huiqun Zhou <hqzhou at nju.edu.cn>:

...

> If anyone can contribute a makefile for using PathScale compiler, the 
> test will
> be handy. I have pathscale version of ACML 2.8. Unfortunately, I haven't
> succeded in compiling QE with this compiler yet.
>

Have you tried using configure ? the last version of QE (3.1.1) (should) have
added support for Pathscale compiler

paolo cazzato



----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.



From peterwiney at gmail.com  Thu Aug  3 17:41:44 2006
From: peterwiney at gmail.com (Peter Winey)
Date: Thu, 3 Aug 2006 11:41:44 -0400
Subject: [Pw_forum] a question about phonon dos
Message-ID: <dac27ff50608030841j54bf509cpf18ef047d58f1673@mail.gmail.com>

Dear all,

For those of you who knows program "matdyn.x" well, I have a question.

In the description part (head) of the program, I noticed the following:
"dos if .true. calculate phonon dos using tetrahedra and a uniform q-point
grid. NB: may not work properly in noncubic materials."

Could someone kindly explain to me what "using tetrahedra" means?
if the program does not work properly for noncubic materials, how to get the
proper phonon dos?

Thanks in advance!

-Peter
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060803/96e529a4/attachment.htm 

From limkr at umich.edu  Thu Aug  3 23:15:06 2006
From: limkr at umich.edu (Donghee Lim)
Date: Thu, 03 Aug 2006 17:15:06 -0400
Subject: [Pw_forum] Warning messages
Message-ID: <20060803171506.m1qkxgz3ko0gwcsw@web.mail.umich.edu>

Dear All,

I was wondering what the following "warning" messages are for.
The calculation was normally completed. Do I need to do anything for 
those warning messages?

Thank you,
Donghee Lim

************************************************************************
Warning:   FFTs have been heavily restructured, those using sunperf are 
untested
Warning:   Please locate and remove the following line in 
Modules/fft_scalar.f90
:
            call errore('cft_1z','sunperf untested, please test',1)
Warning:   Please recompile, test, report if it works

     Initial potential from superposition of free atoms

     starting charge   15.49992, renormalised to   16.00000
Warning:   FFTs have been heavily restructured, those using sunperf are 
untested
Warning:   Please locate and remove the following line in 
Modules/fft_scalar.f90
:
            call errore('cft_1z','sunperf untested, please test',1)
Warning:   Please recompile, test, report if it works
     Starting wfc are atomic

     total cpu time spent up to now is     47.51 secs
************************************************************************


From eyvaz_isaev at yahoo.com  Fri Aug  4 00:25:01 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Thu, 3 Aug 2006 15:25:01 -0700 (PDT)
Subject: [Pw_forum] Warning messages
In-Reply-To: <20060803171506.m1qkxgz3ko0gwcsw@web.mail.umich.edu>
Message-ID: <20060803222501.51442.qmail@web60322.mail.yahoo.com>

Hi, 

It seems you are using the CVS version, because the
same message has been reported quite recently, but
with  stop of the program. Following the recipe
prescribed the compilation was successful. 
Here is a link for this message:

http://www.democritos.it/pipermail/pw_forum/2006-July/004558.html

So, I assume you should not be worried about these
messages if you did some tests from /examples and got
the same (may be with small deviations due to
different  libraries, CPUs, OS, etc. used).

Good luck,
Eyvaz.

--- Donghee Lim <limkr at umich.edu> wrote:

> Dear All,
> 
> I was wondering what the following "warning"
> messages are for.
> The calculation was normally completed. Do I need to
> do anything for 
> those warning messages?
> 
> Thank you,
> Donghee Lim
> 
>
************************************************************************
> Warning:   FFTs have been heavily restructured,
> those using sunperf are 
> untested
> Warning:   Please locate and remove the following
> line in 
> Modules/fft_scalar.f90
> :
>             call errore('cft_1z','sunperf untested,
> please test',1)
> Warning:   Please recompile, test, report if it
> works
> 
>      Initial potential from superposition of free
> atoms
> 
>      starting charge   15.49992, renormalised to  
> 16.00000
> Warning:   FFTs have been heavily restructured,
> those using sunperf are 
> untested
> Warning:   Please locate and remove the following
> line in 
> Modules/fft_scalar.f90
> :
>             call errore('cft_1z','sunperf untested,
> please test',1)
> Warning:   Please recompile, test, report if it
> works
>      Starting wfc are atomic
> 
>      total cpu time spent up to now is     47.51
> secs
>
************************************************************************
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From eyvaz_isaev at yahoo.com  Fri Aug  4 00:36:13 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Thu, 3 Aug 2006 15:36:13 -0700 (PDT)
Subject: [Pw_forum] a question about phonon dos
In-Reply-To: <dac27ff50608030841j54bf509cpf18ef047d58f1673@mail.gmail.com>
Message-ID: <20060803223613.30353.qmail@web60311.mail.yahoo.com>

Hi, 

For the tetrahedra method please find a paper of
Blochl, Jepsen, and Andersen in Phys.Rev. B49,16223
(1994), and then an original paper of Lehman and Taut
(see references in the abovementioned ref.).

If you have a noncubic system you can use special
k-points technique with a broadening scheme you like. 
See PRB13, 5188. For an additional information please
have a look at /Doc/INPUT_PW file (section K_POINTS). 

Bests, 
Eyvaz.

--- Peter Winey <peterwiney at gmail.com> wrote:

> Dear all,
> 
> For those of you who knows program "matdyn.x" well,
> I have a question.
> 
> In the description part (head) of the program, I
> noticed the following:
> "dos if .true. calculate phonon dos using tetrahedra
> and a uniform q-point
> grid. NB: may not work properly in noncubic
> materials."
> 
> Could someone kindly explain to me what "using
> tetrahedra" means?
> if the program does not work properly for noncubic
> materials, how to get the
> proper phonon dos?
> 
> Thanks in advance!
> 
> -Peter
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From eyvaz_isaev at yahoo.com  Fri Aug  4 02:11:18 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Thu, 3 Aug 2006 17:11:18 -0700 (PDT)
Subject: [Pw_forum] ������� �� PWscf
In-Reply-To: <1983393525.20060730135107@fromru.com>
Message-ID: <20060804001118.83835.qmail@web60323.mail.yahoo.com>

Privet, Vitaly!

Ia rekomenduiu vam citat' obzor Allan, Teter , Arias,
Johannopoulos iz Rev.Mod.Phys. i obzor Pickett iz
(kajetsa?) Comp.Phys.Comm. (gde-to nacalo 90-x, no ne
Comp.mater.Science.)

Esli u vas net dostupa k etim journalam, mogu
pereslat' ih po e-mail.

Eyvaz.


--- Vitaliy <twin at fromru.com> wrote:

> ????????????????????????????????????,
???????????????!
> 
>   ???????????? ????????? ???
????????????????????????: ??????????????? ??????
?????? ????????????????????? ?????????, ????????????
> ??? ????????? ????????????,
???????????????-??????????????????
????????????????????????????????????
> ??????????????????????????? ??????
?????????????????? PWscf ?????????
?????????????????????, ????????? ??????
???????????????
> ???????????????. ??????????????????????????? ??????
?????????????????????????????????
> ???????????????????????????????????? ???
readme-?????????????????? ??? ?????? ????????????
???????????????????????? ?????????
> ?????????????????????????????????
??????????????????. ??????????????? ?????????
???????????????????????????
> ??????????????????????????? ?????????????????????
???????????????????????????, ????????????????????????,
???????????????????????????
> ????????????????????????????????????
????????????????????? ?????????
???????????????????????? ?????????????????????
> ecutwfc) ??? ??????????????????. ???????????????
???????????? ????????????????????? ??? ??????
????????????????????????
> ???????????????????????????. ??? ???????????????,
??????????????? ????????????????????????
> ?????? ?????????????????????
???????????????????????? ?????? ???????????????
??????????????????.
> 
> ?????? ???????????????
??????????????????????????????!!!
> -- 
> ??? ???????????????????????????,
>  ?????????????????????                         
mailto:
> twin at fromru.com
> 
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From hqzhou at nju.edu.cn  Fri Aug  4 07:30:30 2006
From: hqzhou at nju.edu.cn (Huiqun Zhou)
Date: Fri, 4 Aug 2006 13:30:30 +0800
Subject: [Pw_forum] Re: Woodcrest vs Opteron performance in pwscf calc.
References: <Pine.LNX.4.44.0608020749470.8867-100000@localhost.localdomain><007301c6b6dc$b9e26480$1d00a8c0@solarflare> <20060803145318.bcsdk1nddssg0488@mail.democritos.it>
Message-ID: <00fa01c6b787$19d0f370$1d00a8c0@solarflare>

I tried , but failed.

Huiqun Zhou

----- Original Message ----- 
From: <cazzato at democritos.it>
To: <pw_forum at pwscf.org>
Sent: Thursday, August 03, 2006 8:53 PM
Subject: Re: [Pw_forum] Re: Woodcrest vs Opteron performance in pwscf calc.


> Quoting Huiqun Zhou <hqzhou at nju.edu.cn>:
>
> ...
>
>> If anyone can contribute a makefile for using PathScale compiler, the 
>> test will
>> be handy. I have pathscale version of ACML 2.8. Unfortunately, I haven't
>> succeded in compiling QE with this compiler yet.
>>
>
> Have you tried using configure ? the last version of QE (3.1.1) (should) 
> have
> added support for Pathscale compiler
>
> paolo cazzato
>
>
>
> ----------------------------------------------------------------
> This message was sent using IMP, the Internet Messaging Program.
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
> 



From lanhaiping at gmail.com  Fri Aug  4 08:32:43 2006
From: lanhaiping at gmail.com (lan haiping)
Date: Fri, 4 Aug 2006 14:32:43 +0800
Subject: [Pw_forum] higher cutoff, lower symmetry?
Message-ID: <c92002fa0608032332k7b3efbd9v2ffbf16bf29acd6b@mail.gmail.com>

Hi, All,

  I did 2 'relax' calculations for the same crystal structure with 2 cutoff
wfc : 20.0 Ry and 30.0 Ry.
The output of 30.0 Ry calculation gave such information :

    *  warning: symmetry operation #  2 not allowed.   fractional
translation:
       0.0000000  0.0000000 -0.5000000  in crystal coordinates
     warning: symmetry operation #  4 not allowed.   fractional translation:
       0.0000000  0.0000000  0.5000000  in crystal coordinates
*
while  no such warnings for 20.0 Ry cutoff calculation.

 Is it expected in PWSCF calculations ?

Regards,

Hai-Ping
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060804/bec0611d/attachment.htm 

From eyvaz_isaev at yahoo.com  Fri Aug  4 09:39:10 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Fri, 4 Aug 2006 00:39:10 -0700 (PDT)
Subject: [Pw_forum] higher cutoff, lower symmetry?
In-Reply-To: <c92002fa0608032332k7b3efbd9v2ffbf16bf29acd6b@mail.gmail.com>
Message-ID: <20060804073910.58599.qmail@web60317.mail.yahoo.com>

Hi,

Yes, sometime it happens and please have a look at the
mail archive to know why it happens and how to fix it.


Bests,

Eyvaz.

--- lan haiping <lanhaiping at gmail.com> wrote:

> Hi, All,
> 
>   I did 2 'relax' calculations for the same crystal
> structure with 2 cutoff
> wfc : 20.0 Ry and 30.0 Ry.
> The output of 30.0 Ry calculation gave such
> information :
> 
>     *  warning: symmetry operation #  2 not allowed.
>   fractional
> translation:
>        0.0000000  0.0000000 -0.5000000  in crystal
> coordinates
>      warning: symmetry operation #  4 not allowed.  
> fractional translation:
>        0.0000000  0.0000000  0.5000000  in crystal
> coordinates
> *
> while  no such warnings for 20.0 Ry cutoff
> calculation.
> 
>  Is it expected in PWSCF calculations ?
> 
> Regards,
> 
> Hai-Ping
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From eyvaz_isaev at yahoo.com  Fri Aug  4 09:49:56 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Fri, 4 Aug 2006 00:49:56 -0700 (PDT)
Subject: [Pw_forum] higher cutoff, lower symmetry?
In-Reply-To: <c92002fa0608032332k7b3efbd9v2ffbf16bf29acd6b@mail.gmail.com>
Message-ID: <20060804074956.40879.qmail@web60321.mail.yahoo.com>

Please also have a look at a pdf-version of the Users
Guide (page 56 for the v.3.1)

Bests,
Eyvaz.

--- lan haiping <lanhaiping at gmail.com> wrote:

> Hi, All,
> 
>   I did 2 'relax' calculations for the same crystal
> structure with 2 cutoff
> wfc : 20.0 Ry and 30.0 Ry.
> The output of 30.0 Ry calculation gave such
> information :
> 
>     *  warning: symmetry operation #  2 not allowed.
>   fractional
> translation:
>        0.0000000  0.0000000 -0.5000000  in crystal
> coordinates
>      warning: symmetry operation #  4 not allowed.  
> fractional translation:
>        0.0000000  0.0000000  0.5000000  in crystal
> coordinates
> *
> while  no such warnings for 20.0 Ry cutoff
> calculation.
> 
>  Is it expected in PWSCF calculations ?
> 
> Regards,
> 
> Hai-Ping
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From cazzato at democritos.it  Fri Aug  4 11:42:49 2006
From: cazzato at democritos.it (cazzato at democritos.it)
Date: Fri, 04 Aug 2006 11:42:49 +0200
Subject: [Pw_forum] Re: Woodcrest vs Opteron performance in pwscf calc.
In-Reply-To: <00fa01c6b787$19d0f370$1d00a8c0@solarflare>
References: <Pine.LNX.4.44.0608020749470.8867-100000@localhost.localdomain><007301c6b6dc$b9e26480$1d00a8c0@solarflare>
	<20060803145318.bcsdk1nddssg0488@mail.democritos.it>
	<00fa01c6b787$19d0f370$1d00a8c0@solarflare>
Message-ID: <20060804114249.l72evjorzeqswg8s@mail.democritos.it>

Where did you fail? Did configure stop with error ? Or was make.sys wrong and
compilation failed? Could you provide the error message?

paolo cazzato

Quoting Huiqun Zhou <hqzhou at nju.edu.cn>:

> I tried , but failed.
>
...
>>
>> Have you tried using configure ? the last version of QE (3.1.1) 
>> (should) have
>> added support for Pathscale compiler
>>


----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.



From peterwiney at gmail.com  Fri Aug  4 22:42:51 2006
From: peterwiney at gmail.com (Peter Winey)
Date: Fri, 4 Aug 2006 16:42:51 -0400
Subject: [Pw_forum] a question about phonon dos
In-Reply-To: <20060803223613.30353.qmail@web60311.mail.yahoo.com>
References: <dac27ff50608030841j54bf509cpf18ef047d58f1673@mail.gmail.com>
	 <20060803223613.30353.qmail@web60311.mail.yahoo.com>
Message-ID: <dac27ff50608041342u60457417o2e7a32bf0f31ada5@mail.gmail.com>

Gotcha. Thanks a lot for the references!

-Peter
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060804/1fa81e27/attachment.htm 

From cesards at msi.umn.edu  Sat Aug  5 00:04:35 2006
From: cesards at msi.umn.edu (cesards at msi.umn.edu)
Date: Fri, 4 Aug 2006 17:04:35 -0500 (CDT)
Subject: [Pw_forum] Re: Re: Re: Woodcrest vs Opteron performance in pwscf calc.
 (Huiqun Zhou)
Message-ID: <57258.128.101.191.226.1154729075.squirrel@www.msi.umn.edu>

Dear Huiqun Zhou,

In your job there is a line

NUM_OMP_THREADS=1

Should be

OMP_NUM_THREADS=1

I don't know how many threads your system defaults to, but if it is more
than one, your tests are flowed.

Yours,

Cesar R.S. da Silva



From tijuthomas at jncasr.ac.in  Sat Aug  5 05:50:31 2006
From: tijuthomas at jncasr.ac.in (Tiju Thomas)
Date: Sat, 5 Aug 2006 09:20:31 +0530 (IST)
Subject: [Pw_forum] response
Message-ID: <47983.202.41.111.151.1154749831.squirrel@202.41.111.151>

Dear Aritz
It is possible to take psi k ^2 using PWSCF

What you need to do is  1)  run the scf run
                        2)  set up an input file for the executable pp.x
   An example is given below:
    &inputpp
prefix = 'WC',
outdir = '/home/tijuthomas/PWSCF_2.0.1/6HEnergySurfaceSF1/tmpSF1minbands//',
filplot = 'SF1AvgKchdensin',
plot_num = 7 ,
kpoint = 11,
kband = 49 ,

/

You should know the k point index at which pisk^2 needs to be plotted and
the band index in which you are interested, just in case your scf run was
done using a monkhorst pack mesh which was automatically generated, hence
the k point that interests you did not feature in the k point mesh
generated, please rerun the file using a specification of k points that
suits you- like the one done in a band structure calculation, since you
are mentioning the kpoints in this case, you are sure to get the k point
of interest in the list!! also note the label of the k point- it is that
which has to be supplied for kpoint.


Please feel free to write to me if you still have trouble,

Regards,
Tiju Thomas
MS (Engg)
Theoretical Sciences Unit
JNCASR
Bangalore
The kband can be easily decided I suppose,  the plot num option should be
set correctly, please note this apart you can also plot the local ionic
potential etc in the 3-D form whch is compatible with the xcrysden
software. The prefix is the prefix that you give for your pw input file.






From hqzhou at nju.edu.cn  Sat Aug  5 09:55:21 2006
From: hqzhou at nju.edu.cn (Huiqun Zhou)
Date: Sat, 5 Aug 2006 15:55:21 +0800
Subject: [Pw_forum] Re: Re: Re: Woodcrest vs Opteron performance in pwscf calc. (Huiqun Zhou)
References: <57258.128.101.191.226.1154729075.squirrel@www.msi.umn.edu>
Message-ID: <002001c6b864$80576790$1d00a8c0@solarflare>

Cesar and list-users,

I'm sorry for my fault. You're right. As there is only 2 cores on one of
my woodcrest nodes are spare right now, I re-run the tests on them and
the opteron node.

Here are the results:
(1) Opteron 285 (2.6 GHz)
  Nearly same as my previous reported numbers.

(2) Woodcrest (2.66 GHz)
  1 core : 4m44s (4m42.74s)
  2 cores: 2m52s (2m49.49s)
  The results were got while another pwscf job was running on other
  two cores.

I'll run the same test on the node next week (maybe) when it becomes
totally clean and spare. So, we can get an impression of how memory
contention would influence the speed as Axel prompted.

Have a nice weekend!

Huiqun Zhou

----- Original Message ----- 
From: <cesards at msi.umn.edu>
To: <pw_forum at pwscf.org>
Sent: Saturday, August 05, 2006 6:04 AM
Subject: [Pw_forum] Re: Re: Re: Woodcrest vs Opteron performance in pwscf 
calc. (Huiqun Zhou)


> Dear Huiqun Zhou,
>
> In your job there is a line
>
> NUM_OMP_THREADS=1
>
> Should be
>
> OMP_NUM_THREADS=1
>
> I don't know how many threads your system defaults to, but if it is more
> than one, your tests are flowed.
>
> Yours,
>
> Cesar R.S. da Silva
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
> 



From swblelia at ehu.es  Mon Aug  7 09:23:28 2006
From: swblelia at ehu.es (Aritz Leonardo Liceranzu)
Date: Mon, 7 Aug 2006 09:23:28 +0200
Subject: [Pw_forum] plotting psi2
Message-ID: <200608070923.28192.swblelia@ehu.es>

Hello All:

I pretend to plot the charge-density of the surface state at M point for a
slab of Mg.

First question, is it possible?

From the input of PP I guess I should pick
plot_num=7

But, how do I specify the K point and the energy of the state I would like to
draw?

Thanks a lot
Aritz



From ongphuongvu at yahoo.com  Mon Aug  7 09:31:05 2006
From: ongphuongvu at yahoo.com (vu ongphuong)
Date: Mon, 7 Aug 2006 00:31:05 -0700 (PDT)
Subject: [Pw_forum] the unit of Dielectric Tensor in .dynG  output file
Message-ID: <20060807073105.87381.qmail@web50604.mail.yahoo.com>

Dear all users,
   
  I have 2 questions:
  1. what is the unit of (electronic) dielectric tensor in .dynG output file; for example:
   
  "Dielectric Tensor:
            6.197084791261          0.000000000000          0.000000000000
          0.000000000000          6.197084791261          0.000000000000
          0.000000000000          0.000000000000          5.774979436500"
   
  2. As I know, in the .dynG file masses are in atomic rydberg units (mass of one electron = 0.5); eigendisplacements are in units  of masses^(-1/2). If so, I want to calculate the phonon contribution dielectric constant; which units of phonon frequencies (omega) I should chose?
   
  Thank you very much.
  Ong Phuong Vu 

 		
---------------------------------
Yahoo! Music Unlimited - Access over 1 million songs.Try it free. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060807/6487cb2f/attachment.htm 

From hqzhou at nju.edu.cn  Mon Aug  7 11:09:59 2006
From: hqzhou at nju.edu.cn (Huiqun Zhou)
Date: Mon, 7 Aug 2006 17:09:59 +0800
Subject: [Pw_forum] Re: Re: Re: Woodcrest vs Opteron performance in pwscf calc. (Huiqun Zhou)
References: <57258.128.101.191.226.1154729075.squirrel@www.msi.umn.edu> <002001c6b864$80576790$1d00a8c0@solarflare>
Message-ID: <005c01c6ba01$428628e0$1d00a8c0@solarflare>

Hi, list-users,

Here I report my test result on a completely clean woodcrest node:
1 cores : 3m54s (3m53.51s)
2 cores: 2m12s (2m10.61s)
4 cores: 1m22s (1m17.38s)

Comparing the results of this test and that in my previous post, we can 
conclude
(1) Setting OMP_NUM_THREADS to 1 or not has no influence on the speed.
  This may be due to the fact that Intel BLAS and LAPACK libraries assume
  that it's equal to 1 while the Scalapack and sparse solver in MKL library 
set
  OMP_NUM_THREADS=2.
(2) It seems that the memory contention is significant:
  1 core : 3m54s (clean) vs. 4m44s (loaded on other 2 cores)
  2 cores: 2m12s (clean vs. 2m52s (loaed on other 2 cores)

Huiqun Zhou

----- Original Message ----- 
From: "Huiqun Zhou" <hqzhou at nju.edu.cn>
To: <pw_forum at pwscf.org>
Sent: Saturday, August 05, 2006 3:55 PM
Subject: Re: [Pw_forum] Re: Re: Re: Woodcrest vs Opteron performance in 
pwscf calc. (Huiqun Zhou)


> Cesar and list-users,
>
> I'm sorry for my fault. You're right. As there is only 2 cores on one of
> my woodcrest nodes are spare right now, I re-run the tests on them and
> the opteron node.
>
> Here are the results:
> (1) Opteron 285 (2.6 GHz)
>  Nearly same as my previous reported numbers.
>
> (2) Woodcrest (2.66 GHz)
>  1 core : 4m44s (4m42.74s)
>  2 cores: 2m52s (2m49.49s)
>  The results were got while another pwscf job was running on other
>  two cores.
>
> I'll run the same test on the node next week (maybe) when it becomes
> totally clean and spare. So, we can get an impression of how memory
> contention would influence the speed as Axel prompted.
>
> Have a nice weekend!
>
> Huiqun Zhou
>
> ----- Original Message ----- 
> From: <cesards at msi.umn.edu>
> To: <pw_forum at pwscf.org>
> Sent: Saturday, August 05, 2006 6:04 AM
> Subject: [Pw_forum] Re: Re: Re: Woodcrest vs Opteron performance in pwscf 
> calc. (Huiqun Zhou)
>
>
>> Dear Huiqun Zhou,
>>
>> In your job there is a line
>>
>> NUM_OMP_THREADS=1
>>
>> Should be
>>
>> OMP_NUM_THREADS=1
>>
>> I don't know how many threads your system defaults to, but if it is more
>> than one, your tests are flowed.
>>
>> Yours,
>>
>> Cesar R.S. da Silva
>>
>> _______________________________________________
>> Pw_forum mailing list
>> Pw_forum at pwscf.org
>> http://www.democritos.it/mailman/listinfo/pw_forum
>>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
> 



From hqzhou at nju.edu.cn  Mon Aug  7 12:55:44 2006
From: hqzhou at nju.edu.cn (Huiqun Zhou)
Date: Mon, 7 Aug 2006 18:55:44 +0800
Subject: [Pw_forum] Re: Woodcrest vs Opteron performance in pwscf calc.
References: <Pine.LNX.4.44.0608020749470.8867-100000@localhost.localdomain><007301c6b6dc$b9e26480$1d00a8c0@solarflare><20060803145318.bcsdk1nddssg0488@mail.democritos.it><00fa01c6b787$19d0f370$1d00a8c0@solarflare> <20060804114249.l72evjorzeqswg8s@mail.democritos.it>
Message-ID: <00cd01c6ba10$0828ea20$1d00a8c0@solarflare>

Hi, Paolo,

Sorry I haven't noticed your response. I just simply got the folowing error
when I tried to configure QE:
  ......
  checking for Fortran 77 compiler default output file name... configure 
error:
  Fortrn 77 compiler cannot create excutables
  ......

I'm using the command line:

./configure CC=pathcc F77=pathf90 F90=pathf90 MPIF77=mpif77 MPIF90=mpif90 \
                 LIBDIRS=/opt/acml3.1.0/pathscale64/lib

BTW, the PSC compiler is an evaluation copy. If the command line I typed is 
not
wrong, it may means that the compiler has expired.

Huiqun Zhou


----- Original Message ----- 
From: <cazzato at democritos.it>
To: <pw_forum at pwscf.org>
Sent: Friday, August 04, 2006 5:42 PM
Subject: Re: [Pw_forum] Re: Woodcrest vs Opteron performance in pwscf calc.


>
> Where did you fail? Did configure stop with error ? Or was make.sys wrong 
> and
> compilation failed? Could you provide the error message?
>
> paolo cazzato
>
> Quoting Huiqun Zhou <hqzhou at nju.edu.cn>:
>
>> I tried , but failed.
>>
> ...
>>>
>>> Have you tried using configure ? the last version of QE (3.1.1) (should) 
>>> have
>>> added support for Pathscale compiler
>>>
>
>
> ----------------------------------------------------------------
> This message was sent using IMP, the Internet Messaging Program.
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
> 



From khkreddy at ipc.iisc.ernet.in  Mon Aug  7 15:38:13 2006
From: khkreddy at ipc.iisc.ernet.in (K.Hari Krishna Reddy)
Date: Mon, 7 Aug 2006 19:08:13 +0530 (IST)
Subject: [Pw_forum] Error while running metadynamics.
Message-ID: <2843.10.50.18.26.1154957893.squirrel@ipc.iisc.ernet.in>

Hi
I'm doing metadynamics on dimethyl-ether(as a test case) using the pwscf
code(-3.1.1 version). The following are some details from the input file

COLLECTIVE_VARS
6
"distance"  1   2
"distance"  2   3
"distance"  3   1
"distance"  1   5
"distance"  1   4
"distance"  1   6
.....................

ATOMIC_POSITIONS { angstrom }
C       -1.171165000     -0.196607000      0.000000000
O        0.000000000      0.587847000      0.000000000
C        1.174595000     -0.195947000      0.000000000
H       -1.242668000     -0.832333000      0.897697000
H       -1.242668000     -0.832333000     -0.897697000
H       -2.020165000      0.493833000      0.000000000
H        1.231895000     -0.836253000     -0.896096000
H        1.231895000     -0.836253000      0.896096000
H        2.021127000      0.495894000      0.000000000

While running the calculation is stopped showing the following error message

" from check_constraint : error #         1
     on some constraint g = 0 is not satisfied"

am a beginner in running dynamics..could anybody help me regarding this
case. And I also want some explanation regarding the keyword "delta_t" in
the name list &IONS.
Thank You

-- 
K.Hari Krishna Reddy (Int.Ph.D)
Prof. E.D.Jemmis' group (080-22933348)
Inorganic and Physical Chemistry
Indian Institute of Science


From konstantin_kudin at yahoo.com  Mon Aug  7 18:00:14 2006
From: konstantin_kudin at yahoo.com (Konstantin Kudin)
Date: Mon, 7 Aug 2006 09:00:14 -0700 (PDT)
Subject: [Pw_forum] Re: Woodcrest vs Opteron performance in pwscf calc.
In-Reply-To: <000c01c6b61b$c10d3bc0$1d00a8c0@solarflare>
Message-ID: <20060807160014.35192.qmail@web52004.mail.yahoo.com>

 Hi Huiqun,

 Very interesting results on the performance! And, at this point one
can offer a simplistic explanation for the relative speeds. See my
comments below.

> Here are numbers
> (1) woodcrest (2.66 GHz):
> 1 core : 3m57s (3m55.86s)
> 2 cores: 2m11s (2m10.44s)
> 4 cores: 1m23s (1m17.73s)
> 
> (2) dempsey (3.2 GHz)
> 1 core : (6m26.90s)
> 2 cores: (3m16.47s)
> 4 cores: (1m39.74s)
> 
> (3) opteron 280 (2.6 GHz)
> 1 core : 7m13s (7m09.71s)
> 2 cores: 3m56s (3m52.70s)
> 4 cores: 2m26s (2m16.72s)
> 
> It seems that woodcrest and dempsey are much faster than opteron. The
> 
> scalability of
> dempsey is the best, woodcrest is the worst. Despite of the amazing 
> performance per
> core of woodcrest, it drops to the same level of its predecessor,
> dempsey, 
> when taking
> the machine as a unit to evaluate its performance.

 A while ago Nicola Marzari was doing extensive benchmarking, and I
joined the effort at some point as well. Basically, at the time CP
timings were judged to be BLAS determined. It appears now that the same
thing is applicable to PWSCF. The fastest BLAS library at this moment
is probably the GOTO BLAS.

 Dempsey and Opterons do 2 BLAS operations per cycle, while Woodcrest
does 4. So effectively you get these frequencies for BLAS (per core):
 Woodcrest (4x2.66=10.6), Dempsey (3.2x2=6.4), Opteron ( 2.6x2=5.2).
That is exactly the order you get in terms of performance. Your Opteron
scaling is not too good, which either suggests that there is not enough
memory bandwidth, or you do not have NUMA turned on.

 Now, the theoretical performance would translate into the real world
if the memory is fast enough. I think both Dempsey and Woodcrest use
the same chipset with 2 buses, so earlier memory contention issues with
multiple Intel chips are mostly gone for now. Still, you see that with
4 Woodcrest cores the speedups are worse then for Dempsey, which
suggests that perhaps the optimal purchase for QE would be lower
frequency chips, such as 2.0 or 2.33 Ghz since 4 2.66 Ghz cores are too
fast for the memory.

 Kostya


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From shaposh at isp.nsc.ru  Tue Aug  8 06:14:30 2006
From: shaposh at isp.nsc.ru (Alexander Shaposhnikov)
Date: Tue, 08 Aug 2006 11:14:30 +0700
Subject: [Pw_forum] Re: Woodcrest vs Opteron performance in pwscf calc.
In-Reply-To: <006e01c6b6d8$9b733e10$1d00a8c0@solarflare>
References: <1154501849.2719.7.camel@m01>
	 <000c01c6b61b$c10d3bc0$1d00a8c0@solarflare> <1154515588.2719.19.camel@m01>
	 <006e01c6b6d8$9b733e10$1d00a8c0@solarflare>
Message-ID: <1155010470.2722.2.camel@m01>

Hi,
my numbers for Opteron with intel ifort 9.0 and ACML
are close to those you have obtained. 

Best Regards,
A. Shaposhnikov




On Thu, 2006-08-03 at 16:41 +0800, Huiqun Zhou wrote:
> Here they are. Looking forward to hearing your story.
> 
> Huiqun Zhou
> 
> ----- Original Message ----- 
> From: "Alexander Shaposhnikov" <shaposh at isp.nsc.ru>
> To: <pw_forum at pwscf.org>
> Sent: Wednesday, August 02, 2006 6:46 PM
> Subject: Re: [Pw_forum] Re: Woodcrest vs Opteron performance in pwscf calc.
> 
> 
> > Thanks for the answer.
> > I don't think this topic is relevant to the PW_forum goals.
> >
> > Could you send me the job's input file ? I'd like to test this and
> > compare with my own Opteron system. We have several Opteron machines,
> > and if the performance difference is really that big, probably its the
> > time to move on to the new platform. But my experience with the prev.
> > generation of Intel processors showed, that Opteron is faster is most
> > cases, especially then it comes to multi-threaded calculations.
> >
> > Best Regards,
> > A. Shaposhnikov
> >
> >
> > On Wed, 2006-08-02 at 18:09 +0800, Huiqun Zhou wrote:
> >> Hi, Alexander,
> >>
> >> A short answer is: Woodcrest (2.66 GHz) > Dempsey (3.2 GHz) > Opteron 280
> >> (2.6 GHz).
> >>
> >> The compute nodes with woodcrest and dempsey in my cluster have 4GB 
> >> memory,
> >> one SATA
> >> disk. The nodes with opteron 280 are SUN Fire x4100, each has 8GB memory 
> >> and
> >> two SAS
> >> disks.
> >>
> >> The test case is total energy calculation of MgAl2O4 (calcium ferrite
> >> structure). The system has
> >> 28 atoms (orthohombic system, Z=4). It took the same number of iterations 
> >> to
> >> reach SCF
> >> convergency on the 3 different machines and got physically same results. 
> >> The
> >> number in
> >> parantheses is the elapsed time given by PWscf.
> >>
> >> Here are numbers
> >> (1) woodcrest (2.66 GHz):
> >> 1 core : 3m57s (3m55.86s)
> >> 2 cores: 2m11s (2m10.44s)
> >> 4 cores: 1m23s (1m17.73s)
> >>
> >> (2) dempsey (3.2 GHz)
> >> 1 core : (6m26.90s)
> >> 2 cores: (3m16.47s)
> >> 4 cores: (1m39.74s)
> >>
> >> (3) opteron 280 (2.6 GHz)
> >> 1 core : 7m13s (7m09.71s)
> >> 2 cores: 3m56s (3m52.70s)
> >> 4 cores: 2m26s (2m16.72s)
> >>
> >> It seems that woodcrest and dempsey are much faster than opteron. The
> >> scalability of
> >> dempsey is the best, woodcrest is the worst. Despite of the amazing
> >> performance per
> >> core of woodcrest, it drops to the same level of its predecessor, 
> >> dempsey,
> >> when taking
> >> the machine as a unit to evaluate its performance.
> >>
> >> But remember one thing: the number for opteron may not be fair. I 
> >> compiled
> >> the program
> >> using Intel fortran, Intel MPI 2.0. However, I ever used both Intel and
> >> PathScale to
> >> compile FFTW and its test cases on opteron machine, I didn't find any
> >> impressive
> >> differences.
> >>
> >> Hope it help you.
> >>
> >>
> >> Huiqun Zhou
> >>
> >>
> >> ----- Original Message ----- 
> >> From: "Alexander Shaposhnikov" <shaposh at isp.nsc.ru>
> >> To: "Huiqun Zhou" <hqzhou at nju.edu.cn>
> >> Sent: Wednesday, August 02, 2006 2:57 PM
> >> Subject: Woodcrest vs Opteron performance in pwscf calc.
> >>
> >>
> >> > Hello dear sir,
> >> >
> >> > i have noticed from your posts to PW_forum that you
> >> > have access to both newest Intel Woodcrest platform and
> >> > AMD Opteron. Could you compare the performance of these two
> >> > platforms, for single-threaded, and parallel 4-core pwscf calculations?
> >> >
> >> > Thank you in advance,
> >> > Best Regards,
> >> > A. Shaposhnikov
> >> >
> >> >
> >> >
> >> >
> >>
> >> _______________________________________________
> >> Pw_forum mailing list
> >> Pw_forum at pwscf.org
> >> http://www.democritos.it/mailman/listinfo/pw_forum
> >
> > _______________________________________________
> > Pw_forum mailing list
> > Pw_forum at pwscf.org
> > http://www.democritos.it/mailman/listinfo/pw_forum
> > 



From tijuthomas at jncasr.ac.in  Tue Aug  8 09:16:31 2006
From: tijuthomas at jncasr.ac.in (Tiju Thomas)
Date: Tue, 8 Aug 2006 12:46:31 +0530 (IST)
Subject: [Pw_forum] psik^2 is possible Aritz
Message-ID: <60282.202.41.111.151.1155021391.squirrel@202.41.111.151>

Dear Aritz
It is possible to take psi k ^2 using PWSCF

What you need to do is  1)  run the scf run
                        2)  set up an input file for the executable pp.x
   An example is given below:
    &inputpp
prefix = 'WC',
outdir = '/home/tijuthomas/PWSCF_2.0.1/6HEnergySurfaceSF1/tmpSF1minbands//',
filplot = 'SF1AvgKchdensin',
plot_num = 7 ,
kpoint = 11,
kband = 49 ,

/

You should know the k point index at which pisk^2 needs to be plotted and
the band index in which you are interested, just in case your scf run was
done using a monkhorst pack mesh which was automatically generated, hence
the k point that interests you did not feature in the k point mesh
generated, please rerun the file using a specification of k points that
suits you- like the one done in a band structure calculation, since you
are mentioning the kpoints in this case, you are sure to get the k point
of interest in the list!! also note the label of the k point- it is that
which has to be supplied for kpoint.


Please feel free to write to me if you still have trouble,

Regards,
Tiju Thomas
MS (Engg)
Theoretical Sciences Unit
JNCASR
Bangalore
The kband can be easily decided I suppose,  the plot num option should be
set correctly, please note this apart you can also plot the local ionic
potential etc in the 3-D form whch is compatible with the xcrysden
software. The prefix is the prefix that you give for your pw input file.









From tr.kelkar at ncl.res.in  Tue Aug  8 13:37:05 2006
From: tr.kelkar at ncl.res.in (Tuhina Kelkar)
Date: Tue, 8 Aug 2006 17:07:05 +0530 (IST)
Subject: [Pw_forum] phonon dispersion calculations
In-Reply-To: <31710563.1155033620029.JavaMail.root@webmail.ncl.res.in>
Message-ID: <24364520.1155037025264.JavaMail.root@webmail.ncl.res.in>

Hi! I am a new user of QEspresso. I want to use it to calculate the phonon dispersion for MgH2 along the following high symmetry points: Gamma to X to M to Gamma to Z to R to A to Z.
But I am not being able to construct the matdyn.in file. How does one decide on the number of q-points? And how to go about creating this list of q-points to be used in matdyn.in?
I have already obtained the IFC's and the following is the resulting q2r.out file:
  reading grid info from file dyn0
  reading force constants from file dyn1
  Dielectric Tensor not found
  nqs=            1
 q=   0.00000000  0.00000000  0.00000000
  reading force constants from file dyn2
  nqs=            1
 q=   0.00000000  0.00000000 -0.74850299
  reading force constants from file dyn3
  nqs=            2
 q=   0.00000000 -0.50000000  0.00000000
 q=   0.50000000  0.00000000  0.00000000
  reading force constants from file dyn4
  nqs=            2
 q=   0.00000000 -0.50000000 -0.74850299
 q=   0.50000000  0.00000000 -0.74850299
  reading force constants from file dyn5
  nqs=            1
 q=  -0.50000000 -0.50000000  0.00000000
  reading force constants from file dyn6
  nqs=            1
 q=  -0.50000000 -0.50000000 -0.74850299
      q-space grid ok, #points =    8
      fft-check success (sum of imaginary terms < 10^-12)
Thanks in advance.
Tuhina
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060808/a86b2335/attachment.htm 

From swblelia at ehu.es  Tue Aug  8 14:34:51 2006
From: swblelia at ehu.es (Aritz Leonardo Liceranzu)
Date: Tue, 8 Aug 2006 14:34:51 +0200
Subject: [Pw_forum] psik^2 is possible Aritz
In-Reply-To: <60282.202.41.111.151.1155021391.squirrel@202.41.111.151>
References: <60282.202.41.111.151.1155021391.squirrel@202.41.111.151>
Message-ID: <200608081434.51137.swblelia@ehu.es>

Thank you very much Tiju!

Now I know how to draw psi^2

Regards
Aritz

On Tuesday 08 August 2006 09:16, Tiju Thomas wrote:
> Dear Aritz
> It is possible to take psi k ^2 using PWSCF
>
> What you need to do is  1)  run the scf run
>                         2)  set up an input file for the executable pp.x
>    An example is given below:
>     &inputpp
> prefix = 'WC',
> outdir =
> '/home/tijuthomas/PWSCF_2.0.1/6HEnergySurfaceSF1/tmpSF1minbands//', filplot
> = 'SF1AvgKchdensin',
> plot_num = 7 ,
> kpoint = 11,
> kband = 49 ,
>
> /
>
> You should know the k point index at which pisk^2 needs to be plotted and
> the band index in which you are interested, just in case your scf run was
> done using a monkhorst pack mesh which was automatically generated, hence
> the k point that interests you did not feature in the k point mesh
> generated, please rerun the file using a specification of k points that
> suits you- like the one done in a band structure calculation, since you
> are mentioning the kpoints in this case, you are sure to get the k point
> of interest in the list!! also note the label of the k point- it is that
> which has to be supplied for kpoint.
>
>
> Please feel free to write to me if you still have trouble,
>
> Regards,
> Tiju Thomas
> MS (Engg)
> Theoretical Sciences Unit
> JNCASR
> Bangalore
> The kband can be easily decided I suppose,  the plot num option should be
> set correctly, please note this apart you can also plot the local ionic
> potential etc in the 3-D form whch is compatible with the xcrysden
> software. The prefix is the prefix that you give for your pw input file.
>
>
>
>
>
>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum


From akohlmey at cmm.upenn.edu  Tue Aug  8 14:36:40 2006
From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer)
Date: Tue, 8 Aug 2006 08:36:40 -0400
Subject: [Pw_forum] phonon dispersion calculations
In-Reply-To: <24364520.1155037025264.JavaMail.root@webmail.ncl.res.in>
References: <31710563.1155033620029.JavaMail.root@webmail.ncl.res.in>
	 <24364520.1155037025264.JavaMail.root@webmail.ncl.res.in>
Message-ID: <7b6913e90608080536i2a625e9dqa60ae2198687d5e6@mail.gmail.com>

On 8/8/06, Tuhina Kelkar <tr.kelkar at ncl.res.in> wrote:
>
>
> Hi! I am a new user of QEspresso. I want to use it to calculate the phonon

hi,

you may want to take a look at the tutorial material that was put
up for the recent QE tutorials in minneapolis and urbana.

http://www.vlab.msi.umn.edu/events/lecture.shtml
http://mccweb1.mcc.uiuc.edu/software/display/ss2006/Quantum+Espresso+Lab

cheers,
   axel.


> dispersion for MgH2 along the following high symmetry points: Gamma to X to
> M to Gamma to Z to R to A to Z.
>
> But I am not being able to construct the matdyn.in file. How does one decide
> on the number of q-points? And how to go about creating this list of
> q-points to be used in matdyn.in?
>
> I have already obtained the IFC's and the following is the resulting q2r.out
> file:
>
>   reading grid info from file dyn0
>   reading force constants from file dyn1
>   Dielectric Tensor not found
>   nqs=            1
>  q=   0.00000000  0.00000000  0.00000000
>   reading force constants from file dyn2
>   nqs=            1
>  q=   0.00000000  0.00000000 -0.74850299
>   reading force constants from file dyn3
>   nqs=            2
>  q=   0.00000000 -0.50000000  0.00000000
>  q=   0.50000000  0.00000000  0.00000000
>   reading force constants from file dyn4
>   nqs=            2
>  q=   0.00000000 -0.50000000 -0.74850299
>  q=   0.50000000  0.00000000 -0.74850299
>   reading force constants from file dyn5
>   nqs=            1
>  q=  -0.50000000 -0.50000000  0.00000000
>   reading force constants from file dyn6
>   nqs=            1
>  q=  -0.50000000 -0.50000000 -0.74850299
>       q-space grid ok, #points =    8
>
>       fft-check success (sum of imaginary terms < 10^-12)
>
> Thanks in advance.
> Tuhina
>


-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
  Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.


From akohlmey at cmm.upenn.edu  Tue Aug  8 14:41:17 2006
From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer)
Date: Tue, 8 Aug 2006 08:41:17 -0400
Subject: [Pw_forum] psik^2 is possible Aritz
In-Reply-To: <200608081434.51137.swblelia@ehu.es>
References: <60282.202.41.111.151.1155021391.squirrel@202.41.111.151>
	 <200608081434.51137.swblelia@ehu.es>
Message-ID: <7b6913e90608080541l2c801c76sc1ae26ab189a4f7@mail.gmail.com>

On 8/8/06, Aritz Leonardo Liceranzu <swblelia at ehu.es> wrote:
>
> Thank you very much Tiju!
>
> Now I know how to draw psi^2

depending on the output options in the &plot namelist
you can create an .xsf file which can be read into either
xcrysden or VMD, or a .plt file for gOpenMol or a cube
file for many other visualization programs.

cheers,
   axel.

>
> Regards
> Aritz
>
> On Tuesday 08 August 2006 09:16, Tiju Thomas wrote:
> > Dear Aritz
> > It is possible to take psi k ^2 using PWSCF
> >
> > What you need to do is  1)  run the scf run
> >                         2)  set up an input file for the executable pp.x
> >    An example is given below:
> >     &inputpp
> > prefix = 'WC',
> > outdir =
> > '/home/tijuthomas/PWSCF_2.0.1/6HEnergySurfaceSF1/tmpSF1minbands//', filplot
> > = 'SF1AvgKchdensin',
> > plot_num = 7 ,
> > kpoint = 11,
> > kband = 49 ,
> >
> > /
> >
> > You should know the k point index at which pisk^2 needs to be plotted and
> > the band index in which you are interested, just in case your scf run was
> > done using a monkhorst pack mesh which was automatically generated, hence
> > the k point that interests you did not feature in the k point mesh
> > generated, please rerun the file using a specification of k points that
> > suits you- like the one done in a band structure calculation, since you
> > are mentioning the kpoints in this case, you are sure to get the k point
> > of interest in the list!! also note the label of the k point- it is that
> > which has to be supplied for kpoint.
> >
> >
> > Please feel free to write to me if you still have trouble,
> >
> > Regards,
> > Tiju Thomas
> > MS (Engg)
> > Theoretical Sciences Unit
> > JNCASR
> > Bangalore
> > The kband can be easily decided I suppose,  the plot num option should be
> > set correctly, please note this apart you can also plot the local ionic
> > potential etc in the 3-D form whch is compatible with the xcrysden
> > software. The prefix is the prefix that you give for your pw input file.
> >
> >
> >
> >
> >
> >
> >
> > _______________________________________________
> > Pw_forum mailing list
> > Pw_forum at pwscf.org
> > http://www.democritos.it/mailman/listinfo/pw_forum
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
>


-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
  Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.


From eyvaz_isaev at yahoo.com  Tue Aug  8 16:02:49 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Tue, 8 Aug 2006 07:02:49 -0700 (PDT)
Subject: [Pw_forum] phonon dispersion calculations
In-Reply-To: <24364520.1155037025264.JavaMail.root@webmail.ncl.res.in>
Message-ID: <20060808140250.79198.qmail@web60315.mail.yahoo.com>

Hi,

You can generate these q-points for yourself. 
The number of q-points depends on how the phonon
spectrum is complex (how many intersections, softening
of phonon branches). 20 for each high symmetry
direction looks reasonable.

Just have a look Bradley-Cracknell textbook about BZ
vertices and then divide each direction as much as you
like (see above). Do not forget that you have to
transform these q-points to cartesian. 

Bests,
Eyvaz.


--- Tuhina Kelkar <tr.kelkar at ncl.res.in> wrote:

> Hi! I am a new user of QEspresso. I want to use it
> to calculate the phonon dispersion for MgH2 along
> the following high symmetry points: Gamma to X to M
> to Gamma to Z to R to A to Z.
> But I am not being able to construct the matdyn.in
> file. How does one decide on the number of q-points?
> And how to go about creating this list of q-points
> to be used in matdyn.in?
> I have already obtained the IFC's and the following
> is the resulting q2r.out file:
>   reading grid info from file dyn0
>   reading force constants from file dyn1
>   Dielectric Tensor not found
>   nqs=            1
>  q=   0.00000000  0.00000000  0.00000000
>   reading force constants from file dyn2
>   nqs=            1
>  q=   0.00000000  0.00000000 -0.74850299
>   reading force constants from file dyn3
>   nqs=            2
>  q=   0.00000000 -0.50000000  0.00000000
>  q=   0.50000000  0.00000000  0.00000000
>   reading force constants from file dyn4
>   nqs=            2
>  q=   0.00000000 -0.50000000 -0.74850299
>  q=   0.50000000  0.00000000 -0.74850299
>   reading force constants from file dyn5
>   nqs=            1
>  q=  -0.50000000 -0.50000000  0.00000000
>   reading force constants from file dyn6
>   nqs=            1
>  q=  -0.50000000 -0.50000000 -0.74850299
>       q-space grid ok, #points =    8
>       fft-check success (sum of imaginary terms <
> 10^-12)
> Thanks in advance.
> Tuhina
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From sbraccia at sissa.it  Tue Aug  8 22:38:38 2006
From: sbraccia at sissa.it (carlo sbraccia)
Date: Tue, 08 Aug 2006 16:38:38 -0400
Subject: [Pw_forum] Error while running metadynamics.
In-Reply-To: <2843.10.50.18.26.1154957893.squirrel@ipc.iisc.ernet.in>
References: <2843.10.50.18.26.1154957893.squirrel@ipc.iisc.ernet.in>
Message-ID: <44D8F64E.7040406@sissa.it>

What you posted is not enough to understand what is wrong in your 
calculation: you should send (to the forum or to me directly) the entire 
input&output files.

 From the Doc/INPUT_PW file:

delta_t        REAL ( default = 1.D0 )
               delta_t = 1                 : every 'nraise' step the actual
                                             temperature is rescaled to 
tempw
               delta_t /= 1 && delta_T > 0 : at each step the temperature is
                                             multiplied by delta_t; this is
                                             done rescaling all the 
velocities.
               delta_t < 0                 : every 'nraise' steps 
temperature
                                             is reduced by -delta_T

Unless you want to change dynamically the temperature (not to be done in 
a meta-dynamics run) you shouldn't modify the default value of this 
keyword. Also, to run a finite temperature meta-dynamics simulation I 
suggest you to set nraise=1.

I add few more comments about meta-dynamics in Quantum-ESPRESSO:

1) The study of a 6-dimensional free-energy surface (FES) is an 
EXTREMELY expensive task (in general the computational cost of exploring 
an N-dim landscape grows exponentially with N). Maybe that, by using 
good collective variables (such as coordination numbers) instead of 
simple distances, you can reduce the dimensionality of the FES. In any 
case you should consider reducing the number of collective variables.

2) So far meta-dynamics has been used at T=0 only. Finite temperature 
simulations are possible, but not yet tested. Moreover, neither the 
mean-force corrections nor the dynamical corrections that arise at 
finite temperature are implemented (Ciccotti et al.; ChemPhysChem, 6, 
1809, 2005). This makes the finite-temperature meta-dynamics which is 
implemented in Quantum-ESPRESSO a method only suited for qualitative 
studies of the FES.

3) PWscf still lacks a  good thermostat like a Nose-Hoover chain. A good 
alternative can be the over-damped Langevin dynamics (see keyword 
ion_temperature in Doc/INPUT_PW) which, however, still requires some 
tuning. The Car-Parrinello code (cp.x) represents an excellent 
alternative for finite-temperature simulations.

Regards,

carlo

K.Hari Krishna Reddy wrote:
> Hi
> I'm doing metadynamics on dimethyl-ether(as a test case) using the pwscf
> code(-3.1.1 version). The following are some details from the input file
>
> COLLECTIVE_VARS
> 6
> "distance"  1   2
> "distance"  2   3
> "distance"  3   1
> "distance"  1   5
> "distance"  1   4
> "distance"  1   6
> .....................
>
> ATOMIC_POSITIONS { angstrom }
> C       -1.171165000     -0.196607000      0.000000000
> O        0.000000000      0.587847000      0.000000000
> C        1.174595000     -0.195947000      0.000000000
> H       -1.242668000     -0.832333000      0.897697000
> H       -1.242668000     -0.832333000     -0.897697000
> H       -2.020165000      0.493833000      0.000000000
> H        1.231895000     -0.836253000     -0.896096000
> H        1.231895000     -0.836253000      0.896096000
> H        2.021127000      0.495894000      0.000000000
>
> While running the calculation is stopped showing the following error message
>
> " from check_constraint : error #         1
>      on some constraint g = 0 is not satisfied"
>
> am a beginner in running dynamics..could anybody help me regarding this
> case. And I also want some explanation regarding the keyword "delta_t" in
> the name list &IONS.
> Thank You
>
>   



From giannozz at nest.sns.it  Tue Aug  8 22:55:54 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Tue, 8 Aug 2006 22:55:54 +0200
Subject: [Pw_forum] phonon dispersion calculations
In-Reply-To: <24364520.1155037025264.JavaMail.root@webmail.ncl.res.in>
References: <24364520.1155037025264.JavaMail.root@webmail.ncl.res.in>
Message-ID: <200608082255.54183.giannozz@nest.sns.it>

On Tuesday 08 August 2006 13:37, Tuhina Kelkar wrote:

> I want [...] to calculate the phonon dispersion for MgH2 along the 
> following high symmetry points: Gamma to X to M to Gamma to Z
> to R to A to Z. But I am not being able to construct the matdyn.in file.
> How does one decide on the number of q-points?  

it depends on what you want to do with them. If you just want nice
dispersions, you need as many q-points as needed for a nice plot

> And how to go about creating this list of q-points to be used in matdyn.in?

one after the other, in units of 2pi/a

> I have already obtained the IFC's and the following is the resulting q2r.out
> file: 
> reading grid info from file dyn0
>   reading force constants from file dyn1
>   Dielectric Tensor not found

and thus no LO-TO splitting. You need epsil=.true. in the
input of the phonon calculation at q=0

Paolo
-- 
Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From giannozz at nest.sns.it  Tue Aug  8 23:28:08 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Tue, 8 Aug 2006 23:28:08 +0200
Subject: [Pw_forum] Electric field potential in pp.x version 3.1
In-Reply-To: <1153431568.8270.25.camel@almaz.mit.edu>
References: <1153431568.8270.25.camel@almaz.mit.edu>
Message-ID: <200608082328.08636.giannozz@nest.sns.it>

On Thursday 20 July 2006 23:39, Boris Kozinsky wrote:

> I tried plotting the electric potential using plot_num=11 by running a
> job first with and then without electric field and then manually
> subtracting the output files from pp.x. I expected to see the sawtooth
> potential in the vacuum region, but instead it has a smooth sinusoidal
> shape with vanishing amplitude.
>
> The same input works fine and reproduces a sawtooth potential in earlier
> versions 2.1.3 and 2.1.4. The induced dipoles etc come out the same, so
> the bug must be somewhere in the pp routine.
>
> Can someone help and investigate this?

sure: YOU can help and investigate this. It is not that difficult to
understand what is going on in pp.x

Paolo
-- 
Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From giannozz at nest.sns.it  Tue Aug  8 23:34:42 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Tue, 8 Aug 2006 23:34:42 +0200
Subject: [Pw_forum] how to get Interatomic force constants
In-Reply-To: <44C49DE1.0000BA.29829@bj163app94.163.com>
References: <44C49DE1.0000BA.29829@bj163app94.163.com>
Message-ID: <200608082334.42583.giannozz@nest.sns.it>

On Monday 24 July 2006 12:16, ??? wrote:

> I calculated Pb2MgTeO6, following example06. But I do not know how to get
> the IFC?s. As README in example06 tells, the IFC?s are contained in the
> Pb2MgTeO6.fc file, but in this file there are only ten matrix(see below),
> represents atom1-atom10.  How can I know the IFC between Pb-O, Te-O, 
> Mg-O, from these matrix?

you cannot: what you sent is not the file containing interatomic force
constants in real space, C(R). It looks like the file containing the dynamical
matrices, i.e. C(q) (apart from the mass factors), but it actually contains
only the dielectric tensor and the effective charges (so it should be q=0).
Are you really sure you correctly followed example 06?

Paolo
-- 
Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From giannozz at nest.sns.it  Tue Aug  8 23:43:13 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Tue, 8 Aug 2006 23:43:13 +0200
Subject: [Pw_forum] Warning messages
In-Reply-To: <20060803171506.m1qkxgz3ko0gwcsw@web.mail.umich.edu>
References: <20060803171506.m1qkxgz3ko0gwcsw@web.mail.umich.edu>
Message-ID: <200608082343.13547.giannozz@nest.sns.it>

On Thursday 03 August 2006 23:15, Donghee Lim wrote:
\
> I was wondering what the following "warning" messages are for.
> The calculation was normally completed. Do I need to do anything for
> those warning messages?

> Warning:   FFTs have been heavily restructured, those using sunperf are
> untested 

...etc etc. How did you manage to have those messages AND the code 
running? in the unmodified code there is a stop after the warnings, 
because these messages are there to prevent people from running 
something untested. 

Paolo
-- 
Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From xshen at ic.sunysb.edu  Wed Aug  9 07:06:15 2006
From: xshen at ic.sunysb.edu (xshen)
Date: Wed, 09 Aug 2006 05:06:15 GMT
Subject: [Pw_forum] Fermi Energy and DOS
Message-ID: <44d96d47.1bf.26ad.7113@ic.sunysb.edu>

Hi:

I am doing calculation of DOS of a supercell system. I did the scf
calculation first and then a nscf calculation for the DOS. The scf and
nscf calculations give me similar value of Fermi energy. however when I
checked  the total DOS state, I found that the  "Int dos" at the given
Fermi energy is not equal to the total number of electrons in my system.
 Should I use the Fermi energy the program gives me, or pick the energy
where the "Int dos" is equal to my # of electrons?

The data is here .
scf calculation gives : Fermi energy is    -0.3788 ev
nscf calculation gives : Fermi energy is    -0.3892 ev

My system has 234 electrons, however at E= -0.4, the INT dos is 225.2.
If I want INT dos =234, the energy is between 2.500 and 2.600 ev.

Part of the Dos file:

#  E (eV)   dos(E)     Int dos(E)
 -0.500  0.8034E-83  0.2252E+03
 -0.400  0.2168E-78  0.2252E+03
 -0.300  0.3285E-41  0.2252E+03
 -0.200  0.9263E-16  0.2252E+03
 -0.100  0.4941E-02  0.2252E+03
  0.000  0.2989E+01  0.2255E+03
  0.100  0.1309E+01  0.2256E+03
  0.200  0.5846E+00  0.2257E+03
  0.300  0.9425E+00  0.2258E+03
  0.400  0.9223E+00  0.2259E+03
  0.500  0.5458E+00  0.2259E+03
  0.600  0.1596E+00  0.2259E+03
  0.700  0.4433E-01  0.2259E+03
  0.800  0.2465E+00  0.2260E+03
  0.900  0.1201E+01  0.2261E+03
  1.000  0.2066E+01  0.2263E+03
  1.100  0.9347E+00  0.2264E+03
  1.200  0.1511E+00  0.2264E+03
  1.300  0.8143E+01  0.2272E+03
  1.400  0.4891E+01  0.2277E+03
  1.500  0.4694E+01  0.2282E+03
  1.600  0.4475E+01  0.2286E+03
  1.700  0.2586E+01  0.2289E+03
  1.800  0.1381E+01  0.2290E+03
  1.900  0.2355E+01  0.2293E+03
  2.000  0.3072E+01  0.2296E+03
  2.100  0.3584E+01  0.2299E+03
  2.200  0.4034E+01  0.2303E+03
  2.300  0.4071E+01  0.2307E+03
  2.400  0.1592E+02  0.2323E+03
  2.500  0.1043E+02  0.2334E+03
  2.600  0.9686E+01  0.2343E+03
  2.700  0.7500E+01  0.2351E+03
  2.800  0.4397E+01  0.2355E+03




From giannozz at nest.sns.it  Wed Aug  9 10:34:02 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Wed, 9 Aug 2006 10:34:02 +0200
Subject: [Pw_forum] the unit of Dielectric Tensor in .dynG  output file
In-Reply-To: <20060807073105.87381.qmail@web50604.mail.yahoo.com>
References: <20060807073105.87381.qmail@web50604.mail.yahoo.com>
Message-ID: <200608091034.02687.giannozz@nest.sns.it>

On Monday 07 August 2006 09:31, vu ongphuong wrote:

>   1. what is the unit of (electronic) dielectric tensor

"unit"? dielectric tensor is adimensional

>   2. As I know, in the .dynG file masses are in atomic rydberg units 
>   (mass of one electron = 0.5); eigendisplacements are in units  of
>   masses^(-1/2).  If so, I want to calculate the phonon contribution
>   dielectric constant; which units of phonon frequencies (omega) 
>   I should chose? 

with the Lyddane-Sachs-Teller formula? it doesn't matter,
the frequency enters as ratio of LO and TO frequencies

-- 
Paolo Giannozzi             Phone:   +39/050-509876
DEMOCRITOS and SNS          Fax:     +39/050-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From giannozz at nest.sns.it  Wed Aug  9 10:42:47 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Wed, 9 Aug 2006 10:42:47 +0200
Subject: [Pw_forum] a question about phonon dos
In-Reply-To: <dac27ff50608030841j54bf509cpf18ef047d58f1673@mail.gmail.com>
References: <dac27ff50608030841j54bf509cpf18ef047d58f1673@mail.gmail.com>
Message-ID: <200608091042.47397.giannozz@nest.sns.it>

On Thursday 03 August 2006 17:41, Peter Winey wrote:

> In the description part (head) of the program, I noticed the following:
> "dos if .true. calculate phonon dos using tetrahedra and a uniform 
> q-point  grid. NB: may not work properly in noncubic materials."
>
> Could someone kindly explain to me what "using tetrahedra" means?

it means "using the tetrahedron method to perform sums over
the Brillouin Zone". See for instance:
P.E. Bloechl et al, PRB49, 16223 (1994)

> if the program does not work properly for noncubic materials, 
> how to get the proper phonon dos?

it is not that it doesn't work at all, but there will be a small
error in the contribution of the region around q=0. It is not
clear to me how to take into account the dependence of LO
phonons at q=0 upon the direction of q=>0 

Paolo
-- 
Paolo Giannozzi             Phone:   +39/050-509876
DEMOCRITOS and SNS          Fax:     +39/050-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From tijuthomas at jncasr.ac.in  Wed Aug  9 12:16:22 2006
From: tijuthomas at jncasr.ac.in (Tiju Thomas)
Date: Wed, 9 Aug 2006 15:46:22 +0530 (IST)
Subject: [Pw_forum] dynmat.x error 1
Message-ID: <56144.202.41.111.151.1155118582.squirrel@202.41.111.151>

Dear pw users,
Greetings to all
The following is the input file I set up for dynmat.x

     &input
fildyn='SiCprim.dynG1',
amass(1)= 28.0855,
amass(2)=12.000 ,
asr= 'crystal',
filout='SiCprim.out',
filxsf= 'SiCprim.axsf',
/

and the dynamical matrix   SiCprim.dynG1  seems to be generated correctly,
but in the output file I get the following:-

 Reading Dynamical Matrix from file SiCprim.dynG1

 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    from latgen : error #         1
     wrong at for ibrav=0
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

     stopping ...



:-
Everything was running without any ibrav complaint so far, I do'nt know
why the code is complaing now, please help as soon as possible,


Thanking you in anticipation
Regards
Tiju Thomas
MS (Engg)
Theoretical Sciences Unit
JNCASR, Bangalore, India




From limkr at umich.edu  Wed Aug  9 13:53:46 2006
From: limkr at umich.edu (Donghee Lim)
Date: Wed, 09 Aug 2006 07:53:46 -0400
Subject: [Pw_forum] Re:Re: Waring message
Message-ID: <20060809075346.u1se0a9500804w8g@web.mail.umich.edu>

Dear Paolo
I didn't do anything to complete my calculation after the waring 
message appeared, but the calculation was completed. Am I supposed to 
do modify the QE code to complete the calculation after the waring 
message appeared? If so, how was my calculation completed without any 
modification?

> > >> Warning:   FFTs have been heavily restructured, those using sunperf are
> > >> untested


Thank you,
Donghee Lim

> > > --__--__--
> > >
> > > Message: 2
> > > From: Paolo Giannozzi <giannozz at nest.sns.it>
> > > Organization: Scuola Normale Superiore
> > > To: pw_forum at pwscf.org
> > > Subject: Re: [Pw_forum] Warning messages
> > > Date: Tue, 8 Aug 2006 23:43:13 +0200
> > > Reply-To: pw_forum at pwscf.org
> > >
> > > On Thursday 03 August 2006 23:15, Donghee Lim wrote:
> > > \
> > >> I was wondering what the following "warning" messages are for.
> > >> The calculation was normally completed. Do I need to do anything for
> > >> those warning messages?
> > >
> > >> Warning:   FFTs have been heavily restructured, those using sunperf are
> > >> untested
> > >
> > > ...etc etc. How did you manage to have those messages AND the code
> > > running? in the unmodified code there is a stop after the warnings,
> > > because these messages are there to prevent people from running
> > > something untested.
> > >
> > > Paolo
> > > --
> > > Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
> > > Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513
> > > Piazza dei Cavalieri 7      I-56126 Pisa, Italy
> >
>



From degironc at sissa.it  Wed Aug  9 16:59:43 2006
From: degironc at sissa.it (degironc)
Date: Wed, 09 Aug 2006 16:59:43 +0200
Subject: [Pw_forum] Fermi Energy and DOS
In-Reply-To: <44d96d47.1bf.26ad.7113@ic.sunysb.edu>
References: <44d96d47.1bf.26ad.7113@ic.sunysb.edu>
Message-ID: <44D9F85F.5080200@sissa.it>

are you using the dos.f90 program in PP ?
Integration is made with a very poor algorithm ... I would not trust it 
of  deltaE is large .
Try to reduce deltaE and see if results improves
stefano

xshen wrote:

>Hi:
>
>I am doing calculation of DOS of a supercell system. I did the scf
>calculation first and then a nscf calculation for the DOS. The scf and
>nscf calculations give me similar value of Fermi energy. however when I
>checked  the total DOS state, I found that the  "Int dos" at the given
>Fermi energy is not equal to the total number of electrons in my system.
> Should I use the Fermi energy the program gives me, or pick the energy
>where the "Int dos" is equal to my # of electrons?
>
>The data is here .
>scf calculation gives : Fermi energy is    -0.3788 ev
>nscf calculation gives : Fermi energy is    -0.3892 ev
>
>My system has 234 electrons, however at E= -0.4, the INT dos is 225.2.
>If I want INT dos =234, the energy is between 2.500 and 2.600 ev.
>
>Part of the Dos file:
>
>#  E (eV)   dos(E)     Int dos(E)
> -0.500  0.8034E-83  0.2252E+03
> -0.400  0.2168E-78  0.2252E+03
> -0.300  0.3285E-41  0.2252E+03
> -0.200  0.9263E-16  0.2252E+03
> -0.100  0.4941E-02  0.2252E+03
>  0.000  0.2989E+01  0.2255E+03
>  0.100  0.1309E+01  0.2256E+03
>  0.200  0.5846E+00  0.2257E+03
>  0.300  0.9425E+00  0.2258E+03
>  0.400  0.9223E+00  0.2259E+03
>  0.500  0.5458E+00  0.2259E+03
>  0.600  0.1596E+00  0.2259E+03
>  0.700  0.4433E-01  0.2259E+03
>  0.800  0.2465E+00  0.2260E+03
>  0.900  0.1201E+01  0.2261E+03
>  1.000  0.2066E+01  0.2263E+03
>  1.100  0.9347E+00  0.2264E+03
>  1.200  0.1511E+00  0.2264E+03
>  1.300  0.8143E+01  0.2272E+03
>  1.400  0.4891E+01  0.2277E+03
>  1.500  0.4694E+01  0.2282E+03
>  1.600  0.4475E+01  0.2286E+03
>  1.700  0.2586E+01  0.2289E+03
>  1.800  0.1381E+01  0.2290E+03
>  1.900  0.2355E+01  0.2293E+03
>  2.000  0.3072E+01  0.2296E+03
>  2.100  0.3584E+01  0.2299E+03
>  2.200  0.4034E+01  0.2303E+03
>  2.300  0.4071E+01  0.2307E+03
>  2.400  0.1592E+02  0.2323E+03
>  2.500  0.1043E+02  0.2334E+03
>  2.600  0.9686E+01  0.2343E+03
>  2.700  0.7500E+01  0.2351E+03
>  2.800  0.4397E+01  0.2355E+03
>
>
>_______________________________________________
>Pw_forum mailing list
>Pw_forum at pwscf.org
>http://www.democritos.it/mailman/listinfo/pw_forum
>  
>



From baroni at sissa.it  Wed Aug  9 18:02:08 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Wed, 9 Aug 2006 18:02:08 +0200
Subject: [Pw_forum] the unit of Dielectric Tensor in .dynG  output file
In-Reply-To: <20060807073105.87381.qmail@web50604.mail.yahoo.com>
References: <20060807073105.87381.qmail@web50604.mail.yahoo.com>
Message-ID: <8CBCE46B-D60A-4FBA-AA46-54735ECBE6F0@sissa.it>

Dear Ong Phuong Vu:

please take a glance at this freshly written wiki article that you  
may find interesting. It is not guaranteed to be typo-free, so please  
report any mis-typed formulas, or else any other errors that you or  
others may find (very soon the Quantum ESPRESSO wiki, which I would  
like to advertize, will be freely editable by everybody).

http://www.quantum-espresso.org/wiki/index.php/Dielectric_constant

Cheers - Stefano

On Aug 7, 2006, at 9:31 AM, vu ongphuong wrote:

> Dear all users,
>
> I have 2 questions:
> 1. what is the unit of (electronic) dielectric tensor in .dynG  
> output file; for example:
>
> "Dielectric Tensor:
>           6.197084791261          0.000000000000           
> 0.000000000000
>           0.000000000000          6.197084791261           
> 0.000000000000
>           0.000000000000          0.000000000000           
> 5.774979436500"
>
> 2. As I know, in the .dynG file masses are in atomic rydberg units  
> (mass of one electron = 0.5); eigendisplacements are in units  of  
> masses^(-1/2). If so, I want to calculate the phonon contribution  
> dielectric constant; which units of phonon frequencies (omega) I  
> should chose?
>
> Thank you very much.
> Ong Phuong Vu
>
> Yahoo! Music Unlimited - Access over 1 million songs. Try it free.

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060809/6a0ac1c2/attachment.htm 

From giannozz at nest.sns.it  Wed Aug  9 18:27:17 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Wed, 9 Aug 2006 18:27:17 +0200
Subject: [Pw_forum] Re:Re: Waring message
In-Reply-To: <20060809075346.u1se0a9500804w8g@web.mail.umich.edu>
References: <20060809075346.u1se0a9500804w8g@web.mail.umich.edu>
Message-ID: <200608091827.17709.giannozz@nest.sns.it>

On Wednesday 09 August 2006 13:53, Donghee Lim wrote:
> Dear Paolo
> I didn't do anything to complete my calculation after the waring
> message appeared, but the calculation was completed. Am I supposed to
> do modify the QE code to complete the calculation after the waring
> message appeared? If so, how was my calculation completed without any
> modification?

the quantum-espresso distribution has a lot of nice and exciting features,
but it doesn't (yet) modify and compile itself. Whoever compiled it has
removed two lines that were stopping the code after the warning message. 
So whoever compiled the code should also remove the lines issuing the 
warning messages (in Modules/fft_scalar.f90) and recompile (provided 
that the code yields the correct results, in addition to a zillion warning
messages). Of course you may choose to just ignore the warning messages, 
but the readability of your output files will suffer.

Paolo
-- 
Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From xshen at ic.sunysb.edu  Wed Aug  9 20:14:15 2006
From: xshen at ic.sunysb.edu (xshen)
Date: Wed, 09 Aug 2006 18:14:15 GMT
Subject: [Pw_forum] Fermi Energy and DOS (problem solved)
Message-ID: <44da25f7.363.3f1a.23677@ic.sunysb.edu>

Thanks. I change the deltaE from 0.1 eV to 0.001 eV and the problem is
solved.

xshen

>are you using the dos.f90 program in PP ?
>Integration is made with a very poor algorithm ... I would not trust it
>of  deltaE is large .
>Try to reduce deltaE and see if results improves
>stefano




From reboredofa at ornl.gov  Thu Aug 10 21:50:16 2006
From: reboredofa at ornl.gov (Fernando A Reboredo)
Date: Thu, 10 Aug 2006 15:50:16 -0400
Subject: [Pw_forum] postdoc in electronic structure
References: <44da25f7.363.3f1a.23677@ic.sunysb.edu>
Message-ID: <00a701c6bcb6$34006a30$b7305ba0@ornl.gov>


Postdoctoral Research Associate Position

Theoretical-Computational

Oak Ridge National Laboratory, Oak Ridge, Tennessee



The Materials Science and Technology Division (MST) of Oak Ridge National 
Laboratory announces an opening for a theoretical-computational postdoctoral 
appointment for the applications and development of advanced Quantum 
Many-Body techniques.  The postdoctoral research will include the 
calculation of optical properties of nanostructures with Quantum Monte Carlo 
(QMC) and Density Functional Theory.  Calculations will target materials 
properties that overlap strongly with experimental programs carried out in 
the MST division and in the Center for Nanophase Materials Sciences (CNMS). 
The applicant should have a Ph.D. in Physics, Materials Science, Chemistry, 
or related field.  A strong background in some ab-initio electronic 
structure method is highly preferred.  Depending on the candidate interest 
and skills, algorithmic and method development activities are possible.  The 
applicant is further expected to be capable of conducting world-class 
research with the exceptional supercomputers available in the ORNL National 
Center for Computational Sciences and must have a demonstrated capability to 
communicate results in English by publication in peer-reviewed journals and 
presentations at scientific meetings.



Qualified candidates are invited to send a complete CV and publication list, 
a statement of research interests, and the names of at least three 
references to Dr. Fernando Reboredo, Oak Ridge National Laboratory, P. O. 
Box 2008, MS-6032, Oak Ridge, TN 37831-6032, or by e-mail to 
reboredofa at ornl.gov .



ORNL is located in Eastern Tennessee on the Oak Ridge Reservation at 20 
miles from Knoxville and within driving distance to the Great Smoky 
Mountains National Park and numerous waterways. ORNL's goal is to develop 
technologies that will enrich and protect humankind.  ORNL, a multiprogram 
research facility managed by UT-Battelle, LLC, for the U.S. Department of 
Energy, is an equal opportunity employer committed to building and 
maintaining a diverse work force.  Additional information about ORNL is 
available at http://www.ornl.gov



This position will be offered through the Oak Ridge Institute for Science 
and Education (ORISE) Oak Ridge National Laboratory Postdoctoral Research 
Associates program

http://www.orau.gov/orise/edu/ornl/ornl-pd/ornlpdoc.htm .  This position is 
open to all qualified U.S. and non-U.S. citizens without regard to race, 
color, age, religion, sex, national origin, physical or mental disability, 
or status as a Vietnam-era veteran or disabled veteran.



From hqzhou at nju.edu.cn  Wed Aug  9 13:22:06 2006
From: hqzhou at nju.edu.cn (Huiqun Zhou)
Date: Wed, 9 Aug 2006 19:22:06 +0800
Subject: [Pw_forum] Re: Woodcrest vs Opteron performance in pwscf calc.
References: <20060807160014.35192.qmail@web52004.mail.yahoo.com>
Message-ID: <011401c6bcea$6f736340$1d00a8c0@solarflare>

Kostya and list-users,

Thanks for your comment and recommendation. Considering the 
price/performance
of machines with dempsey and woodcrest, dempsey may be a good choice too,
especially if you have no problem to pay the electricity bill ;-)

You mentioned NUMA enabling on opteron machines, I wonder if it's a default
function of kernel 2.6.9-xx. If it's not, I need to turn it on in 
re-configuration of
the kernel and recompile, right?

Thanks,

Huiqun

----- Original Message ----- 
From: "Konstantin Kudin" <konstantin_kudin at yahoo.com>
To: <pw_forum at pwscf.org>
Sent: Tuesday, August 08, 2006 12:00 AM
Subject: Re: [Pw_forum] Re: Woodcrest vs Opteron performance in pwscf calc.


>
> Dempsey and Opterons do 2 BLAS operations per cycle, while Woodcrest
> does 4. So effectively you get these frequencies for BLAS (per core):
> Woodcrest (4x2.66=10.6), Dempsey (3.2x2=6.4), Opteron ( 2.6x2=5.2).
> That is exactly the order you get in terms of performance. Your Opteron
> scaling is not too good, which either suggests that there is not enough
> memory bandwidth, or you do not have NUMA turned on.
>
> Now, the theoretical performance would translate into the real world
> if the memory is fast enough. I think both Dempsey and Woodcrest use
> the same chipset with 2 buses, so earlier memory contention issues with
> multiple Intel chips are mostly gone for now. Still, you see that with
> 4 Woodcrest cores the speedups are worse then for Dempsey, which
> suggests that perhaps the optimal purchase for QE would be lower
> frequency chips, such as 2.0 or 2.33 Ghz since 4 2.66 Ghz cores are too
> fast for the memory.
>
> Kostya
>



From hqzhou at nju.edu.cn  Fri Aug 11 04:18:37 2006
From: hqzhou at nju.edu.cn (Huiqun Zhou)
Date: Fri, 11 Aug 2006 10:18:37 +0800
Subject: Fw: [Pw_forum] Re: Woodcrest vs Opteron performance in pwscf calc.
Message-ID: <014801c6bcec$7450e480$1d00a8c0@solarflare>

It seems that maintanence work of our campus network messed up our mail 
server
these two days. My reply to the pw forum is in my "Sent" box, but I haven't 
seen
its arrival in the mailing list although three days have passed.

Here I forward my post. if you by any means had received the mail 2 days 
ago, I
apologize for the inconvenience to you.

Huiqun Zhou

----- Original Message ----- 
From: "Huiqun Zhou" <hqzhou at nju.edu.cn>
To: <pw_forum at pwscf.org>
Sent: Wednesday, August 09, 2006 7:22 PM
Subject: Re: [Pw_forum] Re: Woodcrest vs Opteron performance in pwscf calc.


> Kostya and list-users,
>
> Thanks for your comment and recommendation. Considering the 
> price/performance
> of machines with dempsey and woodcrest, dempsey may be a good choice too,
> especially if you have no problem to pay the electricity bill ;-)
>
> You mentioned NUMA enabling on opteron machines, I wonder if it's a 
> default
> function of kernel 2.6.9-xx. If it's not, I need to turn it on in 
> re-configuration of
> the kernel and recompile, right?
>
> Thanks,
>
> Huiqun
>
> ----- Original Message ----- 
> From: "Konstantin Kudin" <konstantin_kudin at yahoo.com>
> To: <pw_forum at pwscf.org>
> Sent: Tuesday, August 08, 2006 12:00 AM
> Subject: Re: [Pw_forum] Re: Woodcrest vs Opteron performance in pwscf 
> calc.
>
>
>>
>> Dempsey and Opterons do 2 BLAS operations per cycle, while Woodcrest
>> does 4. So effectively you get these frequencies for BLAS (per core):
>> Woodcrest (4x2.66=10.6), Dempsey (3.2x2=6.4), Opteron ( 2.6x2=5.2).
>> That is exactly the order you get in terms of performance. Your Opteron
>> scaling is not too good, which either suggests that there is not enough
>> memory bandwidth, or you do not have NUMA turned on.
>>
>> Now, the theoretical performance would translate into the real world
>> if the memory is fast enough. I think both Dempsey and Woodcrest use
>> the same chipset with 2 buses, so earlier memory contention issues with
>> multiple Intel chips are mostly gone for now. Still, you see that with
>> 4 Woodcrest cores the speedups are worse then for Dempsey, which
>> suggests that perhaps the optimal purchase for QE would be lower
>> frequency chips, such as 2.0 or 2.33 Ghz since 4 2.66 Ghz cores are too
>> fast for the memory.
>>
>> Kostya
>>
> 



From konstantin_kudin at yahoo.com  Fri Aug 11 22:51:46 2006
From: konstantin_kudin at yahoo.com (Konstantin Kudin)
Date: Fri, 11 Aug 2006 13:51:46 -0700 (PDT)
Subject: [Pw_forum] Re: Woodcrest vs Opteron performance in pwscf calc.
In-Reply-To: <011401c6bcea$6f736340$1d00a8c0@solarflare>
Message-ID: <20060811205146.9583.qmail@web52011.mail.yahoo.com>

> You mentioned NUMA enabling on opteron machines, I wonder if it's a
> default
> function of kernel 2.6.9-xx. If it's not, I need to turn it on in 
> re-configuration of
> the kernel and recompile, right?

 I think there could be BIOS settings that affect this (motherboard
specific), and also the kernel needs to be set properly too. Your
version looks fresh enough to have it. 

 We have here duals with NUMA on, and quads with NUMA off. So 2 duals
at 2.0Ghz over 1Gbit beat the quad at 2.2Ghz by like 20-30% (the quad
needs NUMA off due to problems with another package).

 Also, consider using the GOTO blas. When I checked, for CP it gave
almost miraculous speed-ups over stuff like ACML, which appears to be
nothing more than a bunch of standard fortran routines compiled with
some optimizations :-) Even Atlas BLAS might be more advanced than that
... Just link whatever BLAS you have first, and then something else to
have LAPACK. Something like this:

../flib/ptools.a ../flib/flib.a ../clib/clib.a ../iotk/src/libiotk.a 
/home/kostya/lib/libgoto64.so -lpthread
/home/kostya/lib/acml/gnu64/lib/libacml.so -lg2c
-L/home/kostya/lib/fftw-2.1.5v/fftw/.libs -lfftw

 Kostya
 

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From tijuthomas at jncasr.ac.in  Mon Aug 14 06:25:19 2006
From: tijuthomas at jncasr.ac.in (Tiju Thomas)
Date: Mon, 14 Aug 2006 09:55:19 +0530 (IST)
Subject: [Pw_forum] electron-phonon coupling at a certain k point
Message-ID: <59012.202.41.111.151.1155529519.squirrel@202.41.111.151>

Dear Profs  Giannozzi, Gironcoli,Corso, Baroni , and pw users

I need to calculate the electron-phonon coupling for phonons in my system
only along certain directions, that is for the phonons having definite q
values.

 Hence I do'nt need to calculate the electron-phonon coupling for all the
phonons. Is there a prescription that you can suggest to do this in the
fastest manner (without doing the dense q point calculation suggested in
example 7 for phonons (input to ph.x there uses a 4*4*4 q point mesh
which will be very expensive for me), since I will need electron-phonon
coupling only along some directions. Most computational effort will go a
waste since I will not use the data on electron-phonon coupling at most
of the points).

For getting the electron-phonon coupling, I realize that in the electronic
scf step- the k point mesh needed should be very dense in this case, but
the kpoint step involving ph.x is very time consuming, so may be I could
just mention the q points in particular and get the information on
electron-phonon coupling, but if I do that do I get accurate results?

Thanking you in anticipation,

Yours faithfully,
Tiju Thomas
MS
Bangalore


From sareh_zaree at yahoo.com  Mon Aug 14 09:35:22 2006
From: sareh_zaree at yahoo.com (sareh zarei)
Date: Mon, 14 Aug 2006 08:35:22 +0100 (BST)
Subject: [Pw_forum] superposition of atomic density
Message-ID: <20060814073522.84369.qmail@web31009.mail.mud.yahoo.com>

Dear all 
  I am trying to plot difference electron density (difference between electron density of crystal and superposition of atomic density) but I do not know how can I calculate superposition of atomic density .
  
  can you suggest me some way for doing it.
   regards
  sareh
  
  Iran,Isfahan
  Isfahan University of Technology
 Send instant messages to your online friends http://uk.messenger.yahoo.com 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060814/cc6f40a5/attachment.htm 

From sagars_daya at yahoo.co.in  Mon Aug 14 18:06:00 2006
From: sagars_daya at yahoo.co.in (daya sagar)
Date: Mon, 14 Aug 2006 17:06:00 +0100 (BST)
Subject: [Pw_forum] Doubt regarding the bandstructure calculations for nanowires
Message-ID: <20060814160600.6569.qmail@web8314.mail.in.yahoo.com>


  Hi

   I am using vasp for simulating nanowires. I have a problem with bandstructure 
calculation. I was not able to figure out solution for calculating the bandstructure
of  nanowires. Can anyone suggest me how to do the simulation using VASP. How should the input files INCAR, KPOINTS, POSCAR should be arranged.

Thank you.

Sincerely,
Dayasagar.

 				
---------------------------------
 Here's a new way to find what you're looking for - Yahoo! Answers 
Send FREE SMS to your friend's mobile from Yahoo! Messenger Version 8. Get it NOW
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060814/19699d2e/attachment.htm 

From stewart at cnf.cornell.edu  Mon Aug 14 18:07:33 2006
From: stewart at cnf.cornell.edu (stewart at cnf.cornell.edu)
Date: Mon, 14 Aug 2006 12:07:33 -0400
Subject: [Pw_forum] Using bands.x to extra band structure info
In-Reply-To: <00a701c6bcb6$34006a30$b7305ba0@ornl.gov>
References: <44da25f7.363.3f1a.23677@ic.sunysb.edu>
            <00a701c6bcb6$34006a30$b7305ba0@ornl.gov>
Message-ID: <20060814160733.15723.qmail@xuxa.iecc.com>

Hi all, 

   I would like to use bands.x to extract the band structure data from a 
nscf calculation.  However, when I try to run it, it states that it needs 
the potential file.  How do I insure this is generated in the nscf 
calculation.  Currently it just outputs the wavefunction and charge files.  
I am using version 3.0 for these calculations. 

Thanks in advanced, 

Derek 


################################
Derek Stewart, Ph. D.
Scientific Computation Associate
250 Duffield Hall
Cornell Nanoscale Facility (CNF)
Ithaca, NY 14853
stewart (at) cnf.cornell.edu
(607) 255-2856 



From akohlmey at cmm.upenn.edu  Mon Aug 14 18:16:41 2006
From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer)
Date: Mon, 14 Aug 2006 12:16:41 -0400
Subject: [Pw_forum] Doubt regarding the bandstructure calculations for nanowires
In-Reply-To: <20060814160600.6569.qmail@web8314.mail.in.yahoo.com>
References: <20060814160600.6569.qmail@web8314.mail.in.yahoo.com>
Message-ID: <7b6913e90608140916g32954a47l2fcdd3eebf18868a@mail.gmail.com>

On 8/14/06, daya sagar <sagars_daya at yahoo.co.in> wrote:
>
>   Hi

hi,

since this is the forum for pwscf and the quantum espresso
package, people here could probably give you some hints
on using pw.x to get the band structure for your nanowires. ;-)
but if you only want information about vasp, please contact
vasp user forums or the vasp authors.

best regards,
    axel.

>    I am using vasp for simulating nanowires. I have a problem with
> bandstructure
> calculation. I was not able to figure out solution for calculating the
> bandstructure
> of  nanowires. Can anyone suggest me how to do the simulation using VASP.
> How should the input files INCAR, KPOINTS, POSCAR should be arranged.
>
> Thank you.
>
> Sincerely,
> Dayasagar.
>
>
>  ________________________________
>  Here's a new way to find what you're looking for - Yahoo! Answers
>  Send FREE SMS to your friend's mobile from Yahoo! Messenger Version 8. Get
> it NOW
>
>


-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
  Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.


From sridevi at uprrp.edu  Tue Aug 15 00:35:47 2006
From: sridevi at uprrp.edu (Krishnan  Sridevi)
Date: Mon, 14 Aug 2006 18:35:47 -0400 (AST)
Subject: [Pw_forum] compling quantum espresso
Message-ID: <2517.136.145.86.102.1155594947.squirrel@wmail.uprrp.edu>

Hi,

I am trying to compile Quantum Espresso 3.1.1 in SGI Origin 300, IRIX 650 OS.
I ran ./configure and got a message "configure success".

while i run make all i get the following error message
"don't know how to make @phcom@ (bu42).
*** Error code 1 (bu21)"

I could not find any fix from the mailing list.

thanks in advance
Sridevi






From akohlmey at vitae.cmm.upenn.edu  Tue Aug 15 00:53:19 2006
From: akohlmey at vitae.cmm.upenn.edu (Axel Kohlmeyer)
Date: Mon, 14 Aug 2006 18:53:19 -0400 (EDT)
Subject: [Pw_forum] compling quantum espresso
In-Reply-To: <2517.136.145.86.102.1155594947.squirrel@wmail.uprrp.edu>
Message-ID: <Pine.LNX.4.44.0608141849250.26882-100000@localhost.localdomain>

On Mon, 14 Aug 2006, Krishnan  Sridevi wrote:

KS> Hi,
KS> 
KS> I am trying to compile Quantum Espresso 3.1.1 in SGI Origin 300, IRIX 650 OS.
KS> I ran ./configure and got a message "configure success".
KS> 
KS> while i run make all i get the following error message
KS> "don't know how to make @phcom@ (bu42).
KS> *** Error code 1 (bu21)"

KS> I could not find any fix from the mailing list.

have a look at the following thread:
http://www.democritos.it/pipermail/pw_forum/2006-June/004411.html

(first hit with searching google on 'know how to make @phcom@')

there are at least three suggestions...

cheers,
    axel.

KS> 
KS> 
KS> thanks in advance
KS> Sridevi
KS> 
KS> 
KS> 
KS> 
KS> _______________________________________________
KS> Pw_forum mailing list
KS> Pw_forum at pwscf.org
KS> http://www.democritos.it/mailman/listinfo/pw_forum
KS> 

-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.



From matteoc at MIT.EDU  Wed Aug 16 00:14:03 2006
From: matteoc at MIT.EDU (Matteo Cococcioni)
Date: Tue, 15 Aug 2006 18:14:03 -0400
Subject: [Pw_forum] error during compilatio
Message-ID: <20060815181403.58y93oppl9a9wgk0@webmail.mit.edu>


Dear all,

I am trying to compile the code on a MacBook laptop (running Mac OS X but with
an intel processor) and I'm experiencing some troubles. I'm using the intel
compiler for Mac OS X (temporary licence) and here is what I did (aftr
installing Xcode containing gcc and the command "make"):

1) since straight configure doesn't recognize the architecture I've used
./configure.old pc_ifc

2) I tried to adapt the obtained make.sys using as a model the make.sys that
works fine on a regular linux pc. The result is attached to this email.

I think I got pretty close to compile pwscf. But at the end it gives the
following error:

ld: Undefined symbols:
_errore_
_ylmr2_
_dylmr2_
_simpson_
_zgefa_
_zgedi_
_sph_bes_
_sph_dbes_
_erf_
_flush_unit_
_int_to_char_
_invmat_
_infomsg_
.
.
.
_iotk_write_dat_character1_0_
_iotk_write_attr_integer1_0_
_iotk_write_attr_integer1_1_
_iotk_write_dat_logical1_0_
_iotk_write_attr_integer1_2_
_iotk_write_attr_real2_0_
_iotk_write_dat_complex2_1_
_iotk_scan_dat_complex2_1_
make[1]: *** [pw.x] Error 1
make: *** [pw] Error 2

which I don't know how to solve. Does anybody have any idea/experience about it?
Of course it's not urgent problem to solve, but maybe this could help having the
make.sys for another kind of architecture.
By the way: the compiler seems to be affected by an internal error which one
finds trying to compile cp (make cp). But the issue with the compilation of pw
seems to be unrelated to that.

I would appreciate any help/suggestion/comment. Thank you.

Matteo
-------------- next part --------------
A non-text attachment was scrubbed...
Name: make.sys
Type: application/octet-stream
Size: 2879 bytes
Desc: not available
Url : /pipermail/attachments/20060815/35d39470/attachment.obj 

From akohlmey at vitae.cmm.upenn.edu  Wed Aug 16 00:35:51 2006
From: akohlmey at vitae.cmm.upenn.edu (Axel Kohlmeyer)
Date: Tue, 15 Aug 2006 18:35:51 -0400 (EDT)
Subject: [Pw_forum] error during compilatio
In-Reply-To: <20060815181403.58y93oppl9a9wgk0@webmail.mit.edu>
Message-ID: <Pine.LNX.4.44.0608151828540.12074-100000@localhost.localdomain>

On Tue, 15 Aug 2006, Matteo Cococcioni wrote:

MC> 
MC> Dear all,

ciao matteo,

MC> I think I got pretty close to compile pwscf. But at the end it gives the
MC> following error:
MC> 
MC> ld: Undefined symbols:
MC> _errore_
MC> _ylmr2_
MC> _dylmr2_
MC> _simpson_

MC> _iotk_scan_dat_complex2_1_
MC> make[1]: *** [pw.x] Error 1
MC> make: *** [pw] Error 2
MC> 
MC> which I don't know how to solve. Does anybody have any idea/experience about it?

no experience, but it looks, as if the linker is not able to resolve
the library files. on many machine, order matters for linking libraries 
and if you put them in LDFLAGS, they will be loaded
_first_  (which will have them ignored, since there is yet no
undefined symbol.). please compare to the regular make.sys file, where
those libraries are in LIBOBJS which will have them loaded _after_
all code.

MC> Of course it's not urgent problem to solve, but maybe this could help having the
MC> make.sys for another kind of architecture.

MC> By the way: the compiler seems to be affected by an internal error which one
MC> finds trying to compile cp (make cp). But the issue with the compilation of pw
MC> seems to be unrelated to that.

no surprise here. cp.x has a few files that strain the optimizer to the 
max. try recompiling the failing files with lower optimization.

hope that helps,
    axel.


MC> 
MC> I would appreciate any help/suggestion/comment. Thank you.
MC> 
MC> Matteo

-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.



From baroni at sissa.it  Wed Aug 16 09:29:47 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Wed, 16 Aug 2006 09:29:47 +0200
Subject: [Pw_forum] error during compilatio
In-Reply-To: <20060815181403.58y93oppl9a9wgk0@webmail.mit.edu>
References: <20060815181403.58y93oppl9a9wgk0@webmail.mit.edu>
Message-ID: <642DD185-7A06-44EB-A11B-E2809E0A33E7@sissa.it>

Matteo: you finally bought it! Is it a Mac Book, or a Mac Book Pro?
Not that I am any expert in this kind of issues, but it looks like  
the old damn story of the number of underscores appended by compilers  
to routine names. It usually only affects routines written in a  
different language than the main program. This does not seem to be  
the case here, but may be the loader is confused by the existence of  
routines compiled with gcc _and_ fortran .?... You may want to try a  
small program consisting of just one main program, one fortran  
routine and, possibly, one c routine?
Take care - Stefano

On Aug 16, 2006, at 12:14 AM, Matteo Cococcioni wrote:

>
> Dear all,
>
> I am trying to compile the code on a MacBook laptop (running Mac OS  
> X but with
> an intel processor) and I'm experiencing some troubles. I'm using  
> the intel
> compiler for Mac OS X (temporary licence) and here is what I did (aftr
> installing Xcode containing gcc and the command "make"):
>
> 1) since straight configure doesn't recognize the architecture I've  
> used
> ./configure.old pc_ifc
>
> 2) I tried to adapt the obtained make.sys using as a model the  
> make.sys that
> works fine on a regular linux pc. The result is attached to this  
> email.
>
> I think I got pretty close to compile pwscf. But at the end it  
> gives the
> following error:
>
> ld: Undefined symbols:
> _errore_
> _ylmr2_
> _dylmr2_
> _simpson_
> _zgefa_
> _zgedi_
> _sph_bes_
> _sph_dbes_
> _erf_
> _flush_unit_
> _int_to_char_
> _invmat_
> _infomsg_
> .
> .
> .
> _iotk_write_dat_character1_0_
> _iotk_write_attr_integer1_0_
> _iotk_write_attr_integer1_1_
> _iotk_write_dat_logical1_0_
> _iotk_write_attr_integer1_2_
> _iotk_write_attr_real2_0_
> _iotk_write_dat_complex2_1_
> _iotk_scan_dat_complex2_1_
> make[1]: *** [pw.x] Error 1
> make: *** [pw] Error 2
>
> which I don't know how to solve. Does anybody have any idea/ 
> experience about it?
> Of course it's not urgent problem to solve, but maybe this could  
> help having the
> make.sys for another kind of architecture.
> By the way: the compiler seems to be affected by an internal error  
> which one
> finds trying to compile cp (make cp). But the issue with the  
> compilation of pw
> seems to be unrelated to that.
>
> I would appreciate any help/suggestion/comment. Thank you.
>
> Matteo
> <make.sys>

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060816/f4709c7a/attachment.htm 

From limkr at umich.edu  Wed Aug 16 16:29:11 2006
From: limkr at umich.edu (Donghee Lim)
Date: Wed, 16 Aug 2006 10:29:11 -0400
Subject: [Pw_forum] Negative or zero norm error
Message-ID: <20060816102911.aaqbw8agzocs4ckg@web.mail.umich.edu>

Dear all

I wonder in which case the following error occurs. Could you please let 
me know how I can overcome that kind of error. My input option is as 
follows;

Thank you,
Donghee Lim

<Error message>
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     from cgramg1_k : error #         1
     negative or zero norm in S
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

<Input option>
&CONTROL
                 calculation = 'relax' ,
                     tprnfor = .true. ,
/
&SYSTEM
                       ibrav = 8,
                   celldm(1) = 10.79865D0,
                   celldm(2) = 1.41421D0,
                   celldm(3) = 3.15706D0,
                         nat = 36,
                        ntyp = 3,
                     ecutwfc = 35.D0 ,
                     ecutrho = 350.D0 ,
                 occupations = 'smearing' ,
                     degauss = 0.01D0 ,
                    smearing = 'methfessel-paxton' ,
                       nspin = 2 ,     starting_magnetization(1) = 0.5D0,
/
&ELECTRONS
                    conv_thr = 1.D-6 ,
                 mixing_mode = 'local-TF',
                 mixing_beta = 0.3D0 ,
             diagonalization = 'david_overlap',
            diago_david_ndim = 8,
/
&IONS
                   bfgs_ndim = 3 ,
           pot_extrapolation = 'second_order' ,
           wfc_extrapolation = 'second_order' ,
/
ATOMIC_SPECIES
   Fe   55.8457  Fe.pbe-sp-van_ak.UPF   C    12.0107  C.pbe-van_ak.UPF
   Cl   35.4530  Cl.pbe-n-van.UPF







From matteoc at MIT.EDU  Wed Aug 16 18:23:56 2006
From: matteoc at MIT.EDU (Matteo Cococcioni)
Date: Wed, 16 Aug 2006 12:23:56 -0400
Subject: [Pw_forum] error during compilatio
In-Reply-To: <642DD185-7A06-44EB-A11B-E2809E0A33E7@sissa.it>
References: <20060815181403.58y93oppl9a9wgk0@webmail.mit.edu>
	<642DD185-7A06-44EB-A11B-E2809E0A33E7@sissa.it>
Message-ID: <20060816122356.nnpr1xqcnen4kc4g@webmail.mit.edu>


Dear Axel and Stefano,

thank you for your comments and suggestions. It works! I just followed Axel's
instructions and was able to obtain a make.sys that compiles a working pw.x.
I didn't menage to compile cp.x though. I even tried to compile runcp.o with
lower optimization level but didn't work.
The make.sys that works is attached to this email. I think it can be used to
have this architecture recognized by the standard configure (instead
configure.old pc_ifc).
Stefano: yes it's a MacBook and I'm pretty happy so far. I even like 
the glossy
lcd screen. it seems pretty fast too (running the code) though I will do more
extensive benchmark.
Does anybody know if the intel compiler for Linux (freely available) 
would work
on this machine? I probably will try after the licence expires.

Thanks again!

Matteo


Quoting Stefano Baroni <baroni at sissa.it>:

> Matteo: you finally bought it! Is it a Mac Book, or a Mac Book Pro?
> Not that I am any expert in this kind of issues, but it looks like  
> the old damn story of the number of underscores appended by compilers 
>  to routine names. It usually only affects routines written in a  
> different language than the main program. This does not seem to be  
> the case here, but may be the loader is confused by the existence of  
> routines compiled with gcc _and_ fortran .?... You may want to try a  
> small program consisting of just one main program, one fortran  
> routine and, possibly, one c routine?
> Take care - Stefano
>
> On Aug 16, 2006, at 12:14 AM, Matteo Cococcioni wrote:
>
>>
>> Dear all,
>>
>> I am trying to compile the code on a MacBook laptop (running Mac OS  
>> X but with
>> an intel processor) and I'm experiencing some troubles. I'm using  the intel
>> compiler for Mac OS X (temporary licence) and here is what I did (aftr
>> installing Xcode containing gcc and the command "make"):
>>
>> 1) since straight configure doesn't recognize the architecture I've  used
>> ./configure.old pc_ifc
>>
>> 2) I tried to adapt the obtained make.sys using as a model the  
>> make.sys that
>> works fine on a regular linux pc. The result is attached to this  email.
>>
>> I think I got pretty close to compile pwscf. But at the end it  gives the
>> following error:
>>
>> ld: Undefined symbols:
>> _errore_
>> _ylmr2_
>> _dylmr2_
>> _simpson_
>> _zgefa_
>> _zgedi_
>> _sph_bes_
>> _sph_dbes_
>> _erf_
>> _flush_unit_
>> _int_to_char_
>> _invmat_
>> _infomsg_
>> .
>> .
>> .
>> _iotk_write_dat_character1_0_
>> _iotk_write_attr_integer1_0_
>> _iotk_write_attr_integer1_1_
>> _iotk_write_dat_logical1_0_
>> _iotk_write_attr_integer1_2_
>> _iotk_write_attr_real2_0_
>> _iotk_write_dat_complex2_1_
>> _iotk_scan_dat_complex2_1_
>> make[1]: *** [pw.x] Error 1
>> make: *** [pw] Error 2
>>
>> which I don't know how to solve. Does anybody have any idea/ 
>> experience about it?
>> Of course it's not urgent problem to solve, but maybe this could  
>> help having the
>> make.sys for another kind of architecture.
>> By the way: the compiler seems to be affected by an internal error  
>> which one
>> finds trying to compile cp (make cp). But the issue with the  
>> compilation of pw
>> seems to be unrelated to that.
>>
>> I would appreciate any help/suggestion/comment. Thank you.
>>
>> Matteo
>> <make.sys>
>
> ---
> Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  Trieste
> [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)
>
> Please, if possible, don't  send me MS Word or PowerPoint attachments
> Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html
>
>
>
>


-------------- next part --------------
A non-text attachment was scrubbed...
Name: make.sys_mac
Type: application/octet-stream
Size: 2360 bytes
Desc: not available
Url : /pipermail/attachments/20060816/3fc7e7b5/attachment.obj 

From vranjan at ncsu.edu  Thu Aug 17 00:34:03 2006
From: vranjan at ncsu.edu (Vivek Ranjan)
Date: Wed, 16 Aug 2006 18:34:03 -0400 (EDT)
Subject: [Pw_forum] segmentation fault
Message-ID: <52613.152.14.74.117.1155767643.squirrel@webmail.ncsu.edu>

I am running a job. The programme stops in the first scf iteration without
any error message and writes the following on the stdout :


Traceback for process 387481(msp mode, ssp 2) apid 387481.0 on node 0
     cegterg_para at cegterg_+0x1578 (0x1059118) at cegterg.f90
                  cegterg_+0x0488 (0x10657A8) at :127
        c_bands_k at c_bands_+0x235C (0x103F61C) at :707
                  c_bands_+0x04B4 (0x10413D4) at :105
                electrons_+0x1814 (0x112E0F4) at :269
                    pwscf_+0x074C (0x1003A2C) at pwscf.f90:105
Fault: Attempt to dereference null pointer: 0x0
Segmentation fault

Can someone please suggest what could be the problem ?

Thank you



From akohlmey at cmm.upenn.edu  Thu Aug 17 01:09:20 2006
From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer)
Date: Wed, 16 Aug 2006 19:09:20 -0400
Subject: [Pw_forum] segmentation fault
In-Reply-To: <52613.152.14.74.117.1155767643.squirrel@webmail.ncsu.edu>
References: <52613.152.14.74.117.1155767643.squirrel@webmail.ncsu.edu>
Message-ID: <7b6913e90608161609u7ffbc802p563fb8fa4a72e179@mail.gmail.com>

vivek,

it is basically impossible to give _any_ advice without knowing
what kind of machine this is on, whether this is parallel or serial,
whether there is other output (and what it is) and especially,
what the input is.

cheers,
   axel.

On 8/16/06, Vivek Ranjan <vranjan at ncsu.edu> wrote:
> I am running a job. The programme stops in the first scf iteration without
> any error message and writes the following on the stdout :
>
>
> Traceback for process 387481(msp mode, ssp 2) apid 387481.0 on node 0
>      cegterg_para at cegterg_+0x1578 (0x1059118) at cegterg.f90
>                   cegterg_+0x0488 (0x10657A8) at :127
>         c_bands_k at c_bands_+0x235C (0x103F61C) at :707
>                   c_bands_+0x04B4 (0x10413D4) at :105
>                 electrons_+0x1814 (0x112E0F4) at :269
>                     pwscf_+0x074C (0x1003A2C) at pwscf.f90:105
> Fault: Attempt to dereference null pointer: 0x0
> Segmentation fault
>
> Can someone please suggest what could be the problem ?
>
> Thank you
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
>


-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
  Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.


From mnunez at ncsu.edu  Thu Aug 17 10:12:43 2006
From: mnunez at ncsu.edu (mnunez at ncsu.edu)
Date: Thu, 17 Aug 2006 04:12:43 -0400 (EDT)
Subject: [Pw_forum] Re: Negative or zero norm error
In-Reply-To: <20060816231001.19184.30669.Mailman@democritos.sissa.it>
References: <20060816231001.19184.30669.Mailman@democritos.sissa.it>
Message-ID: <2833.65.87.176.133.1155802363.squirrel@webmail.ncsu.edu>

Donghee,

I got the same error for relaxations on other systems  using pwscf-3.1.1
and pwscf-3.1.
I posted it in the forum and Paolo answered  the following:

----------------------------------------------------------------------
On Tuesday 11 July 2006 13:24, mnunez at ncsu.edu wrote:

>      from cgramg1_k : error #         1
>      negative or zero norm in S

S is the overlap matrix in the Ultrasoft pseudopotential formalism.
It has to be positive definite. Likely one (or more than one) of the
pseudopotentials is bad

Paolo
-- 
Paolo Giannozzi             Phone:   +39/050-509876
DEMOCRITOS and SNS          Fax:     +39/050-563513
Piazza dei Cavalieri 7      I-56126 Pisa, Italy
-------------------------------------------------------------------

the interesting thing is that  if I use pwscf-3.0 with the same
pseudopotentials I do not have this error   and the relaxation seems to
converge.

Matias

 Message: 2
> Date: Wed, 16 Aug 2006 10:29:11 -0400
> From: Donghee Lim <limkr at umich.edu>
> To: pw_forum <pw_forum at pwscf.org>
> Subject: [Pw_forum] Negative or zero norm error
> Reply-To: pw_forum at pwscf.org
>
> Dear all
>
> I wonder in which case the following error occurs. Could you please let
> me know how I can overcome that kind of error. My input option is as
> follows;
>
> Thank you,
> Donghee Lim
>
> <Error message>
> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>      from cgramg1_k : error #         1
>      negative or zero norm in S
> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>
> <Input option>
> &CONTROL
>                  calculation = 'relax' ,
>                      tprnfor = .true. ,
> /
> &SYSTEM
>                        ibrav = 8,
>                    celldm(1) = 10.79865D0,
>                    celldm(2) = 1.41421D0,
>                    celldm(3) = 3.15706D0,
>                          nat = 36,
>                         ntyp = 3,
>                      ecutwfc = 35.D0 ,
>                      ecutrho = 350.D0 ,
>                  occupations = 'smearing' ,
>                      degauss = 0.01D0 ,
>                     smearing = 'methfessel-paxton' ,
>                        nspin = 2 ,     starting_magnetization(1) = 0.5D0,
> /
> &ELECTRONS
>                     conv_thr = 1.D-6 ,
>                  mixing_mode = 'local-TF',
>                  mixing_beta = 0.3D0 ,
>              diagonalization = 'david_overlap',
>             diago_david_ndim = 8,
> /
> &IONS
>                    bfgs_ndim = 3 ,
>            pot_extrapolation = 'second_order' ,
>            wfc_extrapolation = 'second_order' ,
> /
> ATOMIC_SPECIES
>    Fe   55.8457  Fe.pbe-sp-van_ak.UPF   C    12.0107  C.pbe-van_ak.UPF
>    Cl   35.4530  Cl.pbe-n-van.UPF
>
>
>




From gnuplot at 163.com  Thu Aug 17 10:44:28 2006
From: gnuplot at 163.com (gnuplot)
Date: Thu, 17 Aug 2006 16:44:28 +0800
Subject: [Pw_forum] How to compile the cpvib.x vdw.x?
Message-ID: <200608171644269379006@163.com>

Dear all,

I.m want to compile the codes of cpvib.x in the source of espresso-3.1.1/VIB, but the neede file 'makeile.cpvib' does not exist. How to do this task?

Thanks

J.L. Tang


gnuplot
2006-08-17





From eyvaz_isaev at yahoo.com  Thu Aug 17 11:54:20 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Thu, 17 Aug 2006 02:54:20 -0700 (PDT)
Subject: [Pw_forum] How to compile the cpvib.x vdw.x?
In-Reply-To: <200608171644269379006@163.com>
Message-ID: <20060817095420.89851.qmail@web60324.mail.yahoo.com>

Hi, 

You can find makefile.cpvib and makefile.pwlib files
in the CVS version. Please read  README.cvs how to get
the CVS version.

No extra makefiles are required for VdW, just type
make in /VdW.

Bests,
Eyvaz.

--- gnuplot <gnuplot at 163.com> wrote:

> 
> Dear all,
> 
> I.m want to compile the codes of cpvib.x in the
> source of espresso-3.1.1/VIB, but the neede file
> 'makeile.cpvib' does not exist. How to do this task?
> 
> Thanks
> 
> J.L. Tang
> 
> 
> gnuplot
> 2006-08-17
> 
> 
> 
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From minu1379 at yahoo.com  Thu Aug 17 16:11:23 2006
From: minu1379 at yahoo.com (Meena Talati)
Date: Thu, 17 Aug 2006 07:11:23 -0700 (PDT)
Subject: [Pw_forum] Q.Espresso-3.1.1: Error in compilation
Message-ID: <20060817141123.94241.qmail@web51115.mail.yahoo.com>

        Hi,
   
  I wish to run Q.Espresso-3.1.1 on IBM laptop. For the same, when I configured Q.Espresso in root directory , and ran 'make all' command after editing make.sys file and inserting the path for BLASS and LAPACK lib. and '-m32' in CFLAGS. I came accross some errors finally. 
  Here is the output where the errors ocurred:
   
  o/p:
   
  ifort error: fatal error in /opt/intel/fc/9.0/bin/fortcom,
  terminated by terminal signal
  compilation aborted for smcp.f90 (code1)
  make[1]: *** [smcp.o] Error1
  make[1]: Leaving directory 'root/workshop/espresso-3.1.1/CPV'
  make: *** [CP] Error2
   
   
  I seek guidance and some suggestions in this regard.
   
  Best wishes,
   
  Mina Talati
   
   






====================================================
Ms.Mina Talati
Senior Research Fellow (CSIR)
Computational Condensed Matter Physics Lab.(CCMP)
Department of Physics, Faculty of Science,
The M.S.University of Baroda, Vadodara - 390 002.
Ph.No: +91-265-279 5339 (O)  extn: 30-25 
====================================================
 		
---------------------------------
Yahoo! Messenger with Voice. Make PC-to-Phone Calls to the US (and 30+ countries) for 2?/min or less.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060817/314ade43/attachment.htm 

From tr.kelkar at ncl.res.in  Fri Aug 18 11:21:32 2006
From: tr.kelkar at ncl.res.in (Tuhina Kelkar)
Date: Fri, 18 Aug 2006 14:51:32 +0530 (IST)
Subject: [Pw_forum] doing QHA calculations
Message-ID: <28022696.1155892892132.JavaMail.root@webmail.ncl.res.in>

Hi everyone!
First of all thanks to all those who gave me their inputs on phonon dispersion calculations. I am having another problem now. 
After calculating the Phonon DOS for MgH2 at different alat values when I run fqha.x I get the following message everytime "wrong grid" and the program stops. What does it mean by wrong grid and how do I correct this error?
Please reply asap. Thanks in advance.
Tuhina
---------------------------------------------------------------------------------------------------------------
Tuhina Kelkar
Junior Research Fellow
Physical Chemistry Division,
National Chemical Laboratory,
Pune - 411 008
INDIA
Contact: 020-25902298
alternative email: tuhina at gmail.com
---------------------------------------------------------------------------------------------------------------
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060818/75971b6e/attachment.htm 

From shemep at rpi.edu  Fri Aug 18 17:08:42 2006
From: shemep at rpi.edu (Philip Shemella)
Date: Fri, 18 Aug 2006 11:08:42 -0400
Subject: [Pw_forum] Re:  dynmat.x error 1
Message-ID: <44E5D7FA.9050503@rpi.edu>

Hi All,
I was having a similar problem as Tiju with dynmat.x for molecular 
systems (i.e. a water molecule).

There seems to be a really simple solution:  In the input file for the 
SCF job it appears that some symmetry must be defined.  For the water 
molecule I now use simple cubic:

  ibrav=1,celldm(1)=20.0

rather than manually defining CELL_PARAMETERS and making it simple cubic 
by writing in the 3x3 matrix.

Hope it helps,  Phil Shemella


p.s.  Thanks to Prof's Baroni, Marzari, Fratesi, and Giannozzi for 
giving great seminars and labs at the Urbana summer school.




> Dear pw users,
> Greetings to all
> The following is the input file I set up for dynmat.x
> 
>      &input
> fildyn='SiCprim.dynG1',
> amass(1)= 28.0855,
> amass(2)=12.000 ,
> asr= 'crystal',
> filout='SiCprim.out',
> filxsf= 'SiCprim.axsf',
> /
> 
> and the dynamical matrix   SiCprim.dynG1  seems to be generated correctly,
> but in the output file I get the following:-
> 
>  Reading Dynamical Matrix from file SiCprim.dynG1
> 
>  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>     from latgen : error #         1
>      wrong at for ibrav=0
>  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
> 
>      stopping ...
> 
> 
> 
> :-
> Everything was running without any ibrav complaint so far, I do'nt know
> why the code is complaing now, please help as soon as possible,
> 
> 
> Thanking you in anticipation
> Regards
> Tiju Thomas
> MS (Engg)
> Theoretical Sciences Unit
> JNCASR, Bangalore, India




From xzheng2 at ncsu.edu  Fri Aug 18 18:11:50 2006
From: xzheng2 at ncsu.edu (xzheng2 at ncsu.edu)
Date: Fri, 18 Aug 2006 12:11:50 -0400 (EDT)
Subject: [Pw_forum] Complitation error
Message-ID: <35248.152.14.74.125.1155917510.squirrel@webmail.ncsu.edu>

Dear all,

When I compile the code espresso-3.1.1, the following error occured:

[zxh at DualXeon0 espresso-3.1.1]$ make pw
test -d bin || mkdir bin
if test -d iotk ; then \
( cd iotk ; if test "make" = "" ; then make  TLDEPS= libiotk.a ; \
else make  TLDEPS= libiotk.a ; fi ) ; fi
make[1]: Entering directory `/home/zxh/METHOD/espresso-3.1.1/iotk'
cd src ; make libiotk.a
make[2]: Entering directory `/home/zxh/METHOD/espresso-3.1.1/iotk/src'
make[2]: `libiotk.a' is up to date.
make[2]: Leaving directory `/home/zxh/METHOD/espresso-3.1.1/iotk/src'
make[1]: Leaving directory `/home/zxh/METHOD/espresso-3.1.1/iotk'
( cd Modules ; if test "make" = "" ; then make  TLDEPS= all ; \
else make  TLDEPS= all ; fi )
make[1]: Entering directory `/home/zxh/METHOD/espresso-3.1.1/Modules'
cpp -P -traditional -D__LINUX -D__PGI -D__FFTW -D__USE_INTERNAL_FFTW
-I../include  constraints_module.f90 -o constraints_module.F90
pgf90 -fast -r8 -D__LINUX -D__PGI -D__FFTW -D__USE_INTERNAL_FFTW
-I../include  -I. -I../Modules -I../PW -I../PH -I../iotk/src -I../CPV -c
constraints_module.F90 -o constraints_module.o
PGF90-F-0004-Corrupt or Old Module file ./kinds.mod
(constraints_module.F90: 110)
PGF90/any Linux/x86 5.0-1: compilation aborted
make[1]: *** [constraints_module.o] Error 2
make[1]: Leaving directory `/home/zxh/METHOD/espresso-3.1.1/Modules'
make: *** [mods] Error 2

Who can tell me how to deal with it?



From akohlmey at cmm.upenn.edu  Fri Aug 18 18:34:21 2006
From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer)
Date: Fri, 18 Aug 2006 12:34:21 -0400
Subject: [Pw_forum] Complitation error
In-Reply-To: <35248.152.14.74.125.1155917510.squirrel@webmail.ncsu.edu>
References: <35248.152.14.74.125.1155917510.squirrel@webmail.ncsu.edu>
Message-ID: <7b6913e90608180934h4e90a539qa281b5d68bcaedb8@mail.gmail.com>

On 8/18/06, xzheng2 at ncsu.edu <xzheng2 at ncsu.edu> wrote:
> Dear all,
>
> When I compile the code espresso-3.1.1, the following error occured:
>
> cpp -P -traditional -D__LINUX -D__PGI -D__FFTW -D__USE_INTERNAL_FFTW
> -I../include  constraints_module.f90 -o constraints_module.F90
> pgf90 -fast -r8 -D__LINUX -D__PGI -D__FFTW -D__USE_INTERNAL_FFTW
> -I../include  -I. -I../Modules -I../PW -I../PH -I../iotk/src -I../CPV -c
> constraints_module.F90 -o constraints_module.o
> PGF90-F-0004-Corrupt or Old Module file ./kinds.mod
> (constraints_module.F90: 110)
> PGF90/any Linux/x86 5.0-1: compilation aborted

hi, please try a 'make clean' and then another 'make all'
perhaps there are some modules left from previous compilation
attempts. please also note, that PGI v5.x compilers are known
to miscompile parts of quantum esperesso (besides being horribly
outdated). so please consider updgrading or changing to a
different compiler.

cheers,
   axel.

> make[1]: *** [constraints_module.o] Error 2
> make[1]: Leaving directory `/home/zxh/METHOD/espresso-3.1.1/Modules'
> make: *** [mods] Error 2
>
> Who can tell me how to deal with it?
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
>


-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
  Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.


From eyvaz_isaev at yahoo.com  Fri Aug 18 20:43:57 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Fri, 18 Aug 2006 11:43:57 -0700 (PDT)
Subject: [Pw_forum] doing QHA calculations
In-Reply-To: <28022696.1155892892132.JavaMail.root@webmail.ncl.res.in>
Message-ID: <20060818184357.96034.qmail@web60315.mail.yahoo.com>

Hi, 

> I get the  following message everytime "wrong grid" 
> and the program stops. What does it mean by wrong 
> grid and  how do I correct this error?

If you are completely sure what you are doing just
comment the line in fqha.f90 and recompile it.

> Please reply asap. 

The fastest way is to understand why it happens.
Otherwise please  be patient.

Bests,
Eyvaz.

Thanks in advance.
> Tuhina
>
---------------------------------------------------------------------------------------------------------------
> Tuhina Kelkar
> Junior Research Fellow
> Physical Chemistry Division,
> National Chemical Laboratory,
> Pune - 411 008
> INDIA
> Contact: 020-25902298
> alternative email: tuhina at gmail.com
>
---------------------------------------------------------------------------------------------------------------
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From eyvaz_isaev at yahoo.com  Fri Aug 18 20:53:37 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Fri, 18 Aug 2006 11:53:37 -0700 (PDT)
Subject: [Pw_forum] Q.Espresso-3.1.1: Error in compilation
In-Reply-To: <20060817141123.94241.qmail@web51115.mail.yahoo.com>
Message-ID: <20060818185337.84802.qmail@web60313.mail.yahoo.com>

Hi,

It might be for two reasons: 

1) buggy compiler. It is known that there is (at
least, there was) some problem with IFC 9.0. Try IFC
8.1.
On my Fujitsu-Siemens laptop there was no problem with
8.1 ans SuSe 10.1. 

2) sometime it depends on compiler flags (for
preposseing). 

3) The information you provided it is not enough to
suggest more.

Bests,
Eyvaz.

--- Meena Talati <minu1379 at yahoo.com> wrote:

>         Hi,
>    
>   I wish to run Q.Espresso-3.1.1 on IBM laptop. For
> the same, when I configured Q.Espresso in root
> directory , and ran 'make all' command after editing
> make.sys file and inserting the path for BLASS and
> LAPACK lib. and '-m32' in CFLAGS. I came accross
> some errors finally. 
>   Here is the output where the errors ocurred:
>    
>   o/p:
>    
>   ifort error: fatal error in
> /opt/intel/fc/9.0/bin/fortcom,
>   terminated by terminal signal
>   compilation aborted for smcp.f90 (code1)
>   make[1]: *** [smcp.o] Error1
>   make[1]: Leaving directory
> 'root/workshop/espresso-3.1.1/CPV'
>   make: *** [CP] Error2
>    
>    
>   I seek guidance and some suggestions in this
> regard.
>    
>   Best wishes,
>    
>   Mina Talati
>    
>    
> 
> 
> 
> 
> 
> 
> ====================================================
> Ms.Mina Talati
> Senior Research Fellow (CSIR)
> Computational Condensed Matter Physics Lab.(CCMP)
> Department of Physics, Faculty of Science,
> The M.S.University of Baroda, Vadodara - 390 002.
> Ph.No: +91-265-279 5339 (O)  extn: 30-25 
> ====================================================
>  		
> ---------------------------------
> Yahoo! Messenger with Voice. Make PC-to-Phone Calls
> to the US (and 30+ countries) for 2?/min or less.


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From xzheng2 at ncsu.edu  Fri Aug 18 23:39:27 2006
From: xzheng2 at ncsu.edu (xzheng2 at ncsu.edu)
Date: Fri, 18 Aug 2006 17:39:27 -0400 (EDT)
Subject: [Pw_forum] insufficient virtual memory! What to do?
In-Reply-To: <20060818185337.84802.qmail@web60313.mail.yahoo.com>
References: <20060818185337.84802.qmail@web60313.mail.yahoo.com>
Message-ID: <35886.152.14.74.125.1155937167.squirrel@webmail.ncsu.edu>

Dear all,

When I run pw.x, the following mistake happens. Could you tell me what to
do, please? Thank you very much!

Sincerely,
Xiaohong

[zxh at localhost pwscf]$ run.sh
  running the geometry relaxation for Au16Si...\c
forrtl: severe (41): insufficient virtual memory
Image              PC        Routine            Line        Source
pw.x               0835666C  Unknown               Unknown  Unknown
pw.x               08354B6D  Unknown               Unknown  Unknown
pw.x               08321F86  Unknown               Unknown  Unknown
pw.x               082E2464  Unknown               Unknown  Unknown
pw.x               08307D99  Unknown               Unknown  Unknown
pw.x               08307CD3  Unknown               Unknown  Unknown
pw.x               080D3705  Unknown               Unknown  Unknown
pw.x               0815C08D  Unknown               Unknown  Unknown
pw.x               0815BD81  Unknown               Unknown  Unknown
 done
  cleaning /data/work/pwscf/tmp...\c
 done




From eyvaz_isaev at yahoo.com  Fri Aug 18 23:47:03 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Fri, 18 Aug 2006 14:47:03 -0700 (PDT)
Subject: [Pw_forum] insufficient virtual memory! What to do?
In-Reply-To: <35886.152.14.74.125.1155937167.squirrel@webmail.ncsu.edu>
Message-ID: <20060818214704.48130.qmail@web60323.mail.yahoo.com>

Hi,

Try reduce number of k-points or ecutrho. If you use
ultrasoft pseudopotentials keep ecutwfc around 30Ry. 
Otherwise, use a parallel computer.

Bests,
Eyvaz.

--- xzheng2 at ncsu.edu wrote:

> Dear all,
> 
> When I run pw.x, the following mistake happens.
> Could you tell me what to
> do, please? Thank you very much!
> 
> Sincerely,
> Xiaohong
> 
> [zxh at localhost pwscf]$ run.sh
>   running the geometry relaxation for Au16Si...\c
> forrtl: severe (41): insufficient virtual memory
> Image              PC        Routine            Line
>        Source
> pw.x               0835666C  Unknown              
> Unknown  Unknown
> pw.x               08354B6D  Unknown              
> Unknown  Unknown
> pw.x               08321F86  Unknown              
> Unknown  Unknown
> pw.x               082E2464  Unknown              
> Unknown  Unknown
> pw.x               08307D99  Unknown              
> Unknown  Unknown
> pw.x               08307CD3  Unknown              
> Unknown  Unknown
> pw.x               080D3705  Unknown              
> Unknown  Unknown
> pw.x               0815C08D  Unknown              
> Unknown  Unknown
> pw.x               0815BD81  Unknown              
> Unknown  Unknown
>  done
>   cleaning /data/work/pwscf/tmp...\c
>  done
> 
> 
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From ezadshojaee at hotmail.com  Sat Aug 19 10:09:51 2006
From: ezadshojaee at hotmail.com (Ezad Shojaee)
Date: Sat, 19 Aug 2006 08:09:51 +0000
Subject: [Pw_forum] phonon at Gamma point
Message-ID: <BAY124-F26D6306CF7801FFB9DCFA9AA430@phx.gbl>

hi
i am trying to calculate phonon frequencies at gamma point of TiO2(anatase) 
and i have relaxed the structure enough (i think )but performing PH.X after 
scf, 2of 18 frequencies  are minus!
i want to know that what technical problem do i have? does anyone calculate 
these beore?
any suggestion will be appreciated

_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today it's FREE! 
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/



From lanhaiping at gmail.com  Sat Aug 19 14:38:06 2006
From: lanhaiping at gmail.com (lan haiping)
Date: Sat, 19 Aug 2006 20:38:06 +0800
Subject: [Pw_forum] phonon at Gamma point
In-Reply-To: <BAY124-F26D6306CF7801FFB9DCFA9AA430@phx.gbl>
References: <BAY124-F26D6306CF7801FFB9DCFA9AA430@phx.gbl>
Message-ID: <c92002fa0608190538y24a9f4devbaadcee0ecefd55e@mail.gmail.com>

Dear All,
I also have calculated gamma phonon frequencies of anatase TiO2 .
I run ph.x caculation after relaxing the structure, but  9 of 18 frequencies
are minus.

Would you please give me some hints and help?

input settings are below, thank you in advance !

*phonons of TiO2 at Gamma
 &inputph
  tr2_ph=1.0d-14,
  recover=.true.,
  epsil=.true.,
  prefix='tio2',
  fildyn='TiO2-gamma.dyn',
  outdir='/home/haiping/tmp/'
 /
0.0 0.0 0.0*
**
*Relaxing calculation input :*
*&CONTROL
             title = 'TiO2' ,
       calculation = 'relax',
         verbosity = 'default',
           tprnfor = .true.,
            outdir = '/home/haiping/tmp/',
            prefix = 'tio2',
           disk_io = 'default',
             nstep = 200,
        pseudo_dir = '/home/haiping/espresso/pseudo/',
/
&SYSTEM
                ibrav= 14,
               celldm(1)=10.2856,celldm(2)=1.0,celldm(3)=1.0,
               celldm(4)=-0.759357,celldm(5)=-0.759357,
               celldm(6)=0.518714
               nat = 6 ,
              ntyp = 2 ,
              nbnd = 30 ,
           ecutwfc = 30.0000000000,
           ecutrho = 160.000,
       occupations = 'fixed' ,
/


&ELECTRONS
  electron_maxstep = 100,
          conv_thr = 1.0e-07,
       mixing_mode = 'plain' ,
       startingpot = 'atomic' ,
       startingwfc = 'atomic' ,
       mixing_beta = 0.300000000,
       mixing_ndim = 8,
   diagonalization = 'david_overlap' ,
  diago_david_ndim = 8,
/

&IONS
             ion_dynamics = 'bfgs',
             pot_extrapolation = 'first_order',
             wfc_extrapolation = 'first_order',

/

ATOMIC_SPECIES
O  15.999 O.pw91-van_ak.UPF
Ti 47.867  Ti.pw91-nsp-van.UPF
ATOMIC_POSITIONS crystal
Ti    0.00000   0.00000   0.00000
Ti    0.75000   0.25000   0.50000
O     0.20800   0.20800  -0.00000
O     0.95800   0.45800   0.50000
O     0.54200   0.04200   0.50000
O     0.79200   0.79200  -0.00000
K_POINTS   automatic
6 6 7   0 0 0

*


On 8/19/06, Ezad Shojaee <ezadshojaee at hotmail.com> wrote:
>
> hi
> i am trying to calculate phonon frequencies at gamma point of
> TiO2(anatase)
> and i have relaxed the structure enough (i think )but performing PH.Xafter
> scf, 2of 18 frequencies  are minus!
> i want to know that what technical problem do i have? does anyone
> calculate
> these beore?
> any suggestion will be appreciated
>
> _________________________________________________________________
> Express yourself instantly with MSN Messenger! Download today it's FREE!
> http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060819/d99530e8/attachment.htm 

From eyvaz_isaev at yahoo.com  Sun Aug 20 00:30:10 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Sat, 19 Aug 2006 15:30:10 -0700 (PDT)
Subject: [Pw_forum] phonon at Gamma point
In-Reply-To: <c92002fa0608190538y24a9f4devbaadcee0ecefd55e@mail.gmail.com>
Message-ID: <20060819223010.77410.qmail@web60322.mail.yahoo.com>

Hi,

It seems your lattice choice is not correct. Please
have a look at 
http://cst-www.nrl.navy.mil/lattice/struk/c5.html

Bests,
Eyvaz.

--- lan haiping <lanhaiping at gmail.com> wrote:

> Dear All,
> I also have calculated gamma phonon frequencies of
> anatase TiO2 .
> I run ph.x caculation after relaxing the structure,
> but  9 of 18 frequencies
> are minus.
> 
> Would you please give me some hints and help?
> 
> input settings are below, thank you in advance !
> 
> *phonons of TiO2 at Gamma
>  &inputph
>   tr2_ph=1.0d-14,
>   recover=.true.,
>   epsil=.true.,
>   prefix='tio2',
>   fildyn='TiO2-gamma.dyn',
>   outdir='/home/haiping/tmp/'
>  /
> 0.0 0.0 0.0*
> **
> *Relaxing calculation input :*
> *&CONTROL
>              title = 'TiO2' ,
>        calculation = 'relax',
>          verbosity = 'default',
>            tprnfor = .true.,
>             outdir = '/home/haiping/tmp/',
>             prefix = 'tio2',
>            disk_io = 'default',
>              nstep = 200,
>         pseudo_dir =
> '/home/haiping/espresso/pseudo/',
> /
> &SYSTEM
>                 ibrav= 14,
>               
> celldm(1)=10.2856,celldm(2)=1.0,celldm(3)=1.0,
>               
> celldm(4)=-0.759357,celldm(5)=-0.759357,
>                celldm(6)=0.518714
>                nat = 6 ,
>               ntyp = 2 ,
>               nbnd = 30 ,
>            ecutwfc = 30.0000000000,
>            ecutrho = 160.000,
>        occupations = 'fixed' ,
> /
> 
> 
> &ELECTRONS
>   electron_maxstep = 100,
>           conv_thr = 1.0e-07,
>        mixing_mode = 'plain' ,
>        startingpot = 'atomic' ,
>        startingwfc = 'atomic' ,
>        mixing_beta = 0.300000000,
>        mixing_ndim = 8,
>    diagonalization = 'david_overlap' ,
>   diago_david_ndim = 8,
> /
> 
> &IONS
>              ion_dynamics = 'bfgs',
>              pot_extrapolation = 'first_order',
>              wfc_extrapolation = 'first_order',
> 
> /
> 
> ATOMIC_SPECIES
> O  15.999 O.pw91-van_ak.UPF
> Ti 47.867  Ti.pw91-nsp-van.UPF
> ATOMIC_POSITIONS crystal
> Ti    0.00000   0.00000   0.00000
> Ti    0.75000   0.25000   0.50000
> O     0.20800   0.20800  -0.00000
> O     0.95800   0.45800   0.50000
> O     0.54200   0.04200   0.50000
> O     0.79200   0.79200  -0.00000
> K_POINTS   automatic
> 6 6 7   0 0 0
> 
> *
> 
> 
> On 8/19/06, Ezad Shojaee <ezadshojaee at hotmail.com>
> wrote:
> >
> > hi
> > i am trying to calculate phonon frequencies at
> gamma point of
> > TiO2(anatase)
> > and i have relaxed the structure enough (i think
> )but performing PH.Xafter
> > scf, 2of 18 frequencies  are minus!
> > i want to know that what technical problem do i
> have? does anyone
> > calculate
> > these beore?
> > any suggestion will be appreciated
> >
> >
>
_________________________________________________________________
> > Express yourself instantly with MSN Messenger!
> Download today it's FREE!
> >
>
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
> >
> > _______________________________________________
> > Pw_forum mailing list
> > Pw_forum at pwscf.org
> > http://www.democritos.it/mailman/listinfo/pw_forum
> >
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From yay451 at mail.usask.ca  Sun Aug 20 00:36:25 2006
From: yay451 at mail.usask.ca (Yansun Yao)
Date: Sat, 19 Aug 2006 16:36:25 -0600
Subject: [Pw_forum] phonon at Gamma point
In-Reply-To: <c92002fa0608190538y24a9f4devbaadcee0ecefd55e@mail.gmail.com>
References: <BAY124-F26D6306CF7801FFB9DCFA9AA430@phx.gbl> <c92002fa0608190538y24a9f4devbaadcee0ecefd55e@mail.gmail.com>
Message-ID: <1156026985.44e7926954412@webmail.usask.ca>

Dear Haiping,

      Other than the internal coordinates, did you optimize the unit cell?

      Cheers
                                            Yansun


Quoting lan haiping <lanhaiping at gmail.com>:

> Dear All,
> I also have calculated gamma phonon frequencies of anatase TiO2 .
> I run ph.x caculation after relaxing the structure, but  9 of 18 frequencies
> are minus.
>
> Would you please give me some hints and help?
>
> input settings are below, thank you in advance !
>
> *phonons of TiO2 at Gamma
>  &inputph
>   tr2_ph=1.0d-14,
>   recover=.true.,
>   epsil=.true.,
>   prefix='tio2',
>   fildyn='TiO2-gamma.dyn',
>   outdir='/home/haiping/tmp/'
>  /
> 0.0 0.0 0.0*
> **
> *Relaxing calculation input :*
> *&CONTROL
>              title = 'TiO2' ,
>        calculation = 'relax',
>          verbosity = 'default',
>            tprnfor = .true.,
>             outdir = '/home/haiping/tmp/',
>             prefix = 'tio2',
>            disk_io = 'default',
>              nstep = 200,
>         pseudo_dir = '/home/haiping/espresso/pseudo/',
> /
> &SYSTEM
>                 ibrav= 14,
>                celldm(1)=10.2856,celldm(2)=1.0,celldm(3)=1.0,
>                celldm(4)=-0.759357,celldm(5)=-0.759357,
>                celldm(6)=0.518714
>                nat = 6 ,
>               ntyp = 2 ,
>               nbnd = 30 ,
>            ecutwfc = 30.0000000000,
>            ecutrho = 160.000,
>        occupations = 'fixed' ,
> /
>
>
> &ELECTRONS
>   electron_maxstep = 100,
>           conv_thr = 1.0e-07,
>        mixing_mode = 'plain' ,
>        startingpot = 'atomic' ,
>        startingwfc = 'atomic' ,
>        mixing_beta = 0.300000000,
>        mixing_ndim = 8,
>    diagonalization = 'david_overlap' ,
>   diago_david_ndim = 8,
> /
>
> &IONS
>              ion_dynamics = 'bfgs',
>              pot_extrapolation = 'first_order',
>              wfc_extrapolation = 'first_order',
>
> /
>
> ATOMIC_SPECIES
> O  15.999 O.pw91-van_ak.UPF
> Ti 47.867  Ti.pw91-nsp-van.UPF
> ATOMIC_POSITIONS crystal
> Ti    0.00000   0.00000   0.00000
> Ti    0.75000   0.25000   0.50000
> O     0.20800   0.20800  -0.00000
> O     0.95800   0.45800   0.50000
> O     0.54200   0.04200   0.50000
> O     0.79200   0.79200  -0.00000
> K_POINTS   automatic
> 6 6 7   0 0 0
>
> *
>
>
> On 8/19/06, Ezad Shojaee <ezadshojaee at hotmail.com> wrote:
> >
> > hi
> > i am trying to calculate phonon frequencies at gamma point of
> > TiO2(anatase)
> > and i have relaxed the structure enough (i think )but performing PH.Xafter
> > scf, 2of 18 frequencies  are minus!
> > i want to know that what technical problem do i have? does anyone
> > calculate
> > these beore?
> > any suggestion will be appreciated
> >
> > _________________________________________________________________
> > Express yourself instantly with MSN Messenger! Download today it's FREE!
> > http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
> >
> > _______________________________________________
> > Pw_forum mailing list
> > Pw_forum at pwscf.org
> > http://www.democritos.it/mailman/listinfo/pw_forum
> >
>





From marzari at MIT.EDU  Sun Aug 20 01:16:48 2006
From: marzari at MIT.EDU (Nicola Marzari)
Date: Sat, 19 Aug 2006 19:16:48 -0400
Subject: [Pw_forum] phonons
In-Reply-To: <1156026985.44e7926954412@webmail.usask.ca>
References: <BAY124-F26D6306CF7801FFB9DCFA9AA430@phx.gbl> <c92002fa0608190538y24a9f4devbaadcee0ecefd55e@mail.gmail.com> <1156026985.44e7926954412@webmail.usask.ca>
Message-ID: <44E79BE0.1080308@mit.edu>



Dear Haiping,


there are two possibilities that lead to negative phonon frequencies at
gamma.

The first one is less likely:

1) you have not carefully relaxed the *internal* degrees of freedom - 
i.e. the ionic coodinates. Then, the phonon code discovers that moving 
the atoms along a certain pattern (an "eigenvector" of the dynamical
matrix at gamma) the energy goes down. Perfect - just relax the ions
more. Relaxation of the unit cell has nothing to do with this

The second one is more likely:

2) the calculation is not converged with respect
to the plane wave cutoff needed, and the number of k-points needed,
for the small or zero frequencies at gamma. Obtaining an accurate,
faithful description of low energy modes requires cutoffs and especially
duals much higher than anything you are used to (to give you an example,
for an ultrasoft C for diamond, you need 100 Ry in the wavefunction and
2800 Ry in the charge density to get a convergence, at gamma,
of the zero modes down to the last cm-1). Does it mean you need
to use 100/2800 Ry for diamond ? No ! By enforcing the translation sum
rules that the interatomic force constants must satisfy, you can monitor
(after enforcing said rules) the convergence of the low frequency modes,
and discover you can use much more reasonable cutoffs.

Still, even more than a regular calculations, phonons require a
through control on the parameters that define your calculation.

Note that, as a warm-up, it is easier to converge LDA calculations
than GGA calculations (the exchange-correlation terms, calculated in
real space, introduce very high frequencies in reciprocal space).

			nicola


>> Dear All,
>> I also have calculated gamma phonon frequencies of anatase TiO2 .
>> I run ph.x caculation after relaxing the structure, but  9 of 18 frequencies
>> are minus.
>>
>> Would you please give me some hints and help?
>>
>> input settings are below, thank you in advance !
>>
>> *phonons of TiO2 at Gamma
>>  &inputph
>>   tr2_ph=1.0d-14,
>>   recover=.true.,
>>   epsil=.true.,
>>   prefix='tio2',
>>   fildyn='TiO2-gamma.dyn',
>>   outdir='/home/haiping/tmp/'
>>  /
>> 0.0 0.0 0.0*


---------------------------------------------------------------------
Prof Nicola Marzari   Department of Materials Science and Engineering
13-5066   MIT   77 Massachusetts Avenue   Cambridge MA 02139-4307 USA
tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu


From lanhaiping at gmail.com  Sun Aug 20 02:43:08 2006
From: lanhaiping at gmail.com (lan haiping)
Date: Sun, 20 Aug 2006 08:43:08 +0800
Subject: [Pw_forum] phonon at Gamma point
In-Reply-To: <20060819223010.77410.qmail@web60322.mail.yahoo.com>
References: <c92002fa0608190538y24a9f4devbaadcee0ecefd55e@mail.gmail.com>
	 <20060819223010.77410.qmail@web60322.mail.yahoo.com>
Message-ID: <c92002fa0608191743h458f45f5ubf539f93e99ed772@mail.gmail.com>

Dear Eyvaz,
Thank you.
I checked the strcuture again using XCRYSDEN,
and found no problems related to the  lattices.

Regards,

Hai-Ping



On 8/20/06, Eyvaz Isaev <eyvaz_isaev at yahoo.com> wrote:
>
> Hi,
>
> It seems your lattice choice is not correct. Please
> have a look at
> http://cst-www.nrl.navy.mil/lattice/struk/c5.html
>
> Bests,
> Eyvaz.
>
> --- lan haiping <lanhaiping at gmail.com> wrote:
>
> > Dear All,
> > I also have calculated gamma phonon frequencies of
> > anatase TiO2 .
> > I run ph.x caculation after relaxing the structure,
> > but  9 of 18 frequencies
> > are minus.
> >
> > Would you please give me some hints and help?
> >
> > input settings are below, thank you in advance !
> >
> > *phonons of TiO2 at Gamma
> >  &inputph
> >   tr2_ph=1.0d-14,
> >   recover=.true.,
> >   epsil=.true.,
> >   prefix='tio2',
> >   fildyn='TiO2-gamma.dyn',
> >   outdir='/home/haiping/tmp/'
> >  /
> > 0.0 0.0 0.0*
> > **
> > *Relaxing calculation input :*
> > *&CONTROL
> >              title = 'TiO2' ,
> >        calculation = 'relax',
> >          verbosity = 'default',
> >            tprnfor = .true.,
> >             outdir = '/home/haiping/tmp/',
> >             prefix = 'tio2',
> >            disk_io = 'default',
> >              nstep = 200,
> >         pseudo_dir =
> > '/home/haiping/espresso/pseudo/',
> > /
> > &SYSTEM
> >                 ibrav= 14,
> >
> > celldm(1)=10.2856,celldm(2)=1.0,celldm(3)=1.0,
> >
> > celldm(4)=-0.759357,celldm(5)=-0.759357,
> >                celldm(6)=0.518714
> >                nat = 6 ,
> >               ntyp = 2 ,
> >               nbnd = 30 ,
> >            ecutwfc = 30.0000000000,
> >            ecutrho = 160.000,
> >        occupations = 'fixed' ,
> > /
> >
> >
> > &ELECTRONS
> >   electron_maxstep = 100,
> >           conv_thr = 1.0e-07,
> >        mixing_mode = 'plain' ,
> >        startingpot = 'atomic' ,
> >        startingwfc = 'atomic' ,
> >        mixing_beta = 0.300000000,
> >        mixing_ndim = 8,
> >    diagonalization = 'david_overlap' ,
> >   diago_david_ndim = 8,
> > /
> >
> > &IONS
> >              ion_dynamics = 'bfgs',
> >              pot_extrapolation = 'first_order',
> >              wfc_extrapolation = 'first_order',
> >
> > /
> >
> > ATOMIC_SPECIES
> > O  15.999 O.pw91-van_ak.UPF
> > Ti 47.867  Ti.pw91-nsp-van.UPF
> > ATOMIC_POSITIONS crystal
> > Ti    0.00000   0.00000   0.00000
> > Ti    0.75000   0.25000   0.50000
> > O     0.20800   0.20800  -0.00000
> > O     0.95800   0.45800   0.50000
> > O     0.54200   0.04200   0.50000
> > O     0.79200   0.79200  -0.00000
> > K_POINTS   automatic
> > 6 6 7   0 0 0
> >
> > *
> >
> >
> > On 8/19/06, Ezad Shojaee <ezadshojaee at hotmail.com>
> > wrote:
> > >
> > > hi
> > > i am trying to calculate phonon frequencies at
> > gamma point of
> > > TiO2(anatase)
> > > and i have relaxed the structure enough (i think
> > )but performing PH.Xafter
> > > scf, 2of 18 frequencies  are minus!
> > > i want to know that what technical problem do i
> > have? does anyone
> > > calculate
> > > these beore?
> > > any suggestion will be appreciated
> > >
> > >
> >
> _________________________________________________________________
> > > Express yourself instantly with MSN Messenger!
> > Download today it's FREE!
> > >
> >
> http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
> > >
> > > _______________________________________________
> > > Pw_forum mailing list
> > > Pw_forum at pwscf.org
> > > http://www.democritos.it/mailman/listinfo/pw_forum
> > >
> >
>
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060820/cba810e3/attachment.htm 

From lanhaiping at gmail.com  Sun Aug 20 02:47:29 2006
From: lanhaiping at gmail.com (lan haiping)
Date: Sun, 20 Aug 2006 08:47:29 +0800
Subject: [Pw_forum] phonon at Gamma point
In-Reply-To: <1156026985.44e7926954412@webmail.usask.ca>
References: <BAY124-F26D6306CF7801FFB9DCFA9AA430@phx.gbl>
	 <c92002fa0608190538y24a9f4devbaadcee0ecefd55e@mail.gmail.com>
	 <1156026985.44e7926954412@webmail.usask.ca>
Message-ID: <c92002fa0608191747n4617f960k40b22acd127da89b@mail.gmail.com>

Dear Yansun

Thank you !
Since i just use  experimental  values of the lattices,
i didnot  optimize the lattices and just relaxed the internal coordinates.
 Should I optimize the lattices using "vc-relax" , and then run the
ph.xcalculation again ?

Regards,

Hai-Ping



On 8/20/06, Yansun Yao <yay451 at mail.usask.ca> wrote:
>
> Dear Haiping,
>
>      Other than the internal coordinates, did you optimize the unit cell?
>
>      Cheers
>                                            Yansun
>
>
> Quoting lan haiping <lanhaiping at gmail.com>:
>
> > Dear All,
> > I also have calculated gamma phonon frequencies of anatase TiO2 .
> > I run ph.x caculation after relaxing the structure, but  9 of 18
> frequencies
> > are minus.
> >
> > Would you please give me some hints and help?
> >
> > input settings are below, thank you in advance !
> >
> > *phonons of TiO2 at Gamma
> >  &inputph
> >   tr2_ph=1.0d-14,
> >   recover=.true.,
> >   epsil=.true.,
> >   prefix='tio2',
> >   fildyn='TiO2-gamma.dyn',
> >   outdir='/home/haiping/tmp/'
> >  /
> > 0.0 0.0 0.0*
> > **
> > *Relaxing calculation input :*
> > *&CONTROL
> >              title = 'TiO2' ,
> >        calculation = 'relax',
> >          verbosity = 'default',
> >            tprnfor = .true.,
> >             outdir = '/home/haiping/tmp/',
> >             prefix = 'tio2',
> >            disk_io = 'default',
> >              nstep = 200,
> >         pseudo_dir = '/home/haiping/espresso/pseudo/',
> > /
> > &SYSTEM
> >                 ibrav= 14,
> >                celldm(1)=10.2856,celldm(2)=1.0,celldm(3)=1.0,
> >                celldm(4)=-0.759357,celldm(5)=-0.759357,
> >                celldm(6)=0.518714
> >                nat = 6 ,
> >               ntyp = 2 ,
> >               nbnd = 30 ,
> >            ecutwfc = 30.0000000000,
> >            ecutrho = 160.000,
> >        occupations = 'fixed' ,
> > /
> >
> >
> > &ELECTRONS
> >   electron_maxstep = 100,
> >           conv_thr = 1.0e-07,
> >        mixing_mode = 'plain' ,
> >        startingpot = 'atomic' ,
> >        startingwfc = 'atomic' ,
> >        mixing_beta = 0.300000000,
> >        mixing_ndim = 8,
> >    diagonalization = 'david_overlap' ,
> >   diago_david_ndim = 8,
> > /
> >
> > &IONS
> >              ion_dynamics = 'bfgs',
> >              pot_extrapolation = 'first_order',
> >              wfc_extrapolation = 'first_order',
> >
> > /
> >
> > ATOMIC_SPECIES
> > O  15.999 O.pw91-van_ak.UPF
> > Ti 47.867  Ti.pw91-nsp-van.UPF
> > ATOMIC_POSITIONS crystal
> > Ti    0.00000   0.00000   0.00000
> > Ti    0.75000   0.25000   0.50000
> > O     0.20800   0.20800  -0.00000
> > O     0.95800   0.45800   0.50000
> > O     0.54200   0.04200   0.50000
> > O     0.79200   0.79200  -0.00000
> > K_POINTS   automatic
> > 6 6 7   0 0 0
> >
> > *
> >
> >
> > On 8/19/06, Ezad Shojaee <ezadshojaee at hotmail.com> wrote:
> > >
> > > hi
> > > i am trying to calculate phonon frequencies at gamma point of
> > > TiO2(anatase)
> > > and i have relaxed the structure enough (i think )but performing
> PH.Xafter
> > > scf, 2of 18 frequencies  are minus!
> > > i want to know that what technical problem do i have? does anyone
> > > calculate
> > > these beore?
> > > any suggestion will be appreciated
> > >
> > > _________________________________________________________________
> > > Express yourself instantly with MSN Messenger! Download today it's
> FREE!
> > > http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
> > >
> > > _______________________________________________
> > > Pw_forum mailing list
> > > Pw_forum at pwscf.org
> > > http://www.democritos.it/mailman/listinfo/pw_forum
> > >
> >
>
>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060820/f3c74e50/attachment.htm 

From lanhaiping at gmail.com  Sun Aug 20 03:15:39 2006
From: lanhaiping at gmail.com (lan haiping)
Date: Sun, 20 Aug 2006 09:15:39 +0800
Subject: [Pw_forum] phonons
In-Reply-To: <44E79BE0.1080308@mit.edu>
References: <BAY124-F26D6306CF7801FFB9DCFA9AA430@phx.gbl>
	 <c92002fa0608190538y24a9f4devbaadcee0ecefd55e@mail.gmail.com>
	 <1156026985.44e7926954412@webmail.usask.ca> <44E79BE0.1080308@mit.edu>
Message-ID: <c92002fa0608191815j45d52622xc6c80ff9a10e9068@mail.gmail.com>

Dear  Prof Nicola,

Thanks for your long suggestion.

I may not follow some of your words, would you please clarify  ?
 1,  How to relax the ions more ?
 2, I knew from most refs that ecutwfs  30 Ry for ultrasoft-PP is enough in
pw calculations , Is it right ? But i donot have much idear about ecutwfs in
ph calculations .  Would you please give me some refs of more detail
settings about PH calculations ?


I am trying to find more problems related to this wrong result following
your suggestion.
Thank you again, it really helps me much.

p.s.  It is nice for me listening  your lectures in Beijing , i really learn
much from your 3 lectures .


Regards,

Hai-Ping


On 8/20/06, Nicola Marzari <marzari at mit.edu> wrote:
>
>
>
>
> Dear Haiping,
>
>
> there are two possibilities that lead to negative phonon frequencies at
> gamma.
>
> The first one is less likely:
>
> 1) you have not carefully relaxed the *internal* degrees of freedom -
> i.e. the ionic coodinates. Then, the phonon code discovers that moving
> the atoms along a certain pattern (an "eigenvector" of the dynamical
> matrix at gamma) the energy goes down. Perfect - just relax the ions
> more. Relaxation of the unit cell has nothing to do with this
>
> The second one is more likely:
>
> 2) the calculation is not converged with respect
> to the plane wave cutoff needed, and the number of k-points needed,
> for the small or zero frequencies at gamma. Obtaining an accurate,
> faithful description of low energy modes requires cutoffs and especially
> duals much higher than anything you are used to (to give you an example,
> for an ultrasoft C for diamond, you need 100 Ry in the wavefunction and
> 2800 Ry in the charge density to get a convergence, at gamma,
> of the zero modes down to the last cm-1). Does it mean you need
> to use 100/2800 Ry for diamond ? No ! By enforcing the translation sum
> rules that the interatomic force constants must satisfy, you can monitor
> (after enforcing said rules) the convergence of the low frequency modes,
> and discover you can use much more reasonable cutoffs.
>
> Still, even more than a regular calculations, phonons require a
> through control on the parameters that define your calculation.
>
> Note that, as a warm-up, it is easier to converge LDA calculations
> than GGA calculations (the exchange-correlation terms, calculated in
> real space, introduce very high frequencies in reciprocal space).
>
>                        nicola
>
>
> >> Dear All,
> >> I also have calculated gamma phonon frequencies of anatase TiO2 .
> >> I run ph.x caculation after relaxing the structure, but  9 of 18
> frequencies
> >> are minus.
> >>
> >> Would you please give me some hints and help?
> >>
> >> input settings are below, thank you in advance !
> >>
> >> *phonons of TiO2 at Gamma
> >>  &inputph
> >>   tr2_ph=1.0d-14,
> >>   recover=.true.,
> >>   epsil=.true.,
> >>   prefix='tio2',
> >>   fildyn='TiO2-gamma.dyn',
> >>   outdir='/home/haiping/tmp/'
> >>  /
> >> 0.0 0.0 0.0*
>
>
> ---------------------------------------------------------------------
> Prof Nicola Marzari   Department of Materials Science and Engineering
> 13-5066   MIT   77 Massachusetts Avenue   Cambridge MA 02139-4307 USA
> tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060820/5d74ae87/attachment.htm 

From baroni at sissa.it  Sun Aug 20 10:26:59 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Sun, 20 Aug 2006 10:26:59 +0200
Subject: [Pw_forum] phonons
In-Reply-To: <c92002fa0608191815j45d52622xc6c80ff9a10e9068@mail.gmail.com>
References: <BAY124-F26D6306CF7801FFB9DCFA9AA430@phx.gbl> <c92002fa0608190538y24a9f4devbaadcee0ecefd55e@mail.gmail.com> <1156026985.44e7926954412@webmail.usask.ca> <44E79BE0.1080308@mit.edu> <c92002fa0608191815j45d52622xc6c80ff9a10e9068@mail.gmail.com>
Message-ID: <6599977F-D6E1-4C8B-8EB8-65B57F0810B2@sissa.it>


On Aug 20, 2006, at 3:15 AM, lan haiping wrote:

> Dear  Prof Nicola,

just a note about the "western" usage of names. (I would be glad to  
hear an explanation on how names are used in Chinese!). "Nicola" is  
the "given" name, i.e. it is the name that parents give to the  
newborn baby. It is also called "first" name because in written it is  
the first name that must be used. The other name "Marzari" is the  
family name which is transmitted from one generation to the next in a  
patrilinear way (a baby gets the fathers' family name by default). It  
is a form of confidentiality to address somebody by his/her first  
name. It is perfect here because: i) it is common usage on the web  
and; ii) it is common practice among scientists (particularly in the  
US). However, addressing somebody by his/her title is _very_ formal.  
Not very appropriate in this forum, but above all kind of funny if  
accompanied by the much more confidential usage of the first name.  
Resuming: "Dear Nicola" is confidential and "Dear Prof. Marzari" is  
formal - both ar OK, the first more appropriate to this forum. "Dear  
Prof Nicola" is just funny! ... I was not joking saying that I am  
curious about _your_ usage of "given", "family", or whatever names.  
If you have a  few spare minutes, just let me know ...

Science, now!

>
> Thanks for your long suggestion.
>
> I may not follow some of your words, would you please clarify  ?
>  1,  How to relax the ions more ?

The code must have some default threshold for the atomic forces (and,  
maybe, stress too if you also relax the unit cell) such that  when  
forces are smaller, it stops. The threshold is usually more than  
adequate for most practical purposes, _but to learn_ What I mean is  
that it is a good idea to practice and to see _with one's own eyes_  
how a given property (phonon frequencies, in the present case) depend  
on external parameter and/or convergence criteria. As it is possible  
to vary by hands the values of the convergence thresholds, it is a  
good idea to try a same calculation with different values of the  
relevant thresholds and to see how the frequencies depend on them.  I  
have a comment, though. As Nicola said in a previous post, it is  
unlikely (though not impossible) that phonons go soft because the  
structure is poorely optimized. It is worth stopping a while and  
understand carefully why it is both unlikely _and_ possible. It is  
possible because squared phonon frequencies are related to the  
Hessian (matrix of second derivatives) of the energy with respect to  
the nuclear coordinates. It is perfectly possible that, out of  
equilibrium, one or more of the eigenvalues of this matrix are  
negative. I even believe (but would like not to comment on this  
statement) that the ratio of the configuration space corresponding to  
all positive eigenvalues to the total configuration space decreases  
exponentially with the number of atoms. HOWEVER, phonon frequencies  
only depend on second derivatives of the energy, and not directly on  
first derivatives. So, if you are not too far from the minimum, not  
too good a convergence (i.e. the existence of nonvanishing forces)  
may certainly affect the accuracy of phonon calculations (it would  
not if the crystal were perfectly harmonic), but not the stability of  
the system (i.e. the sign of any of the squared frequencies).

>     2, I knew from most refs that ecutwfs  30 Ry for ultrasoft-PP  
> is enough in pw calculations , Is it right ?

This statement is meaningless. 30 Ry may be too large or too small a  
cutoff, depending on the PP that you are using. It is true that many  
US PP calculations are done with cutoffs in this range, but whether  
or not this value is appropriate in your case ONLY YOU CAN TELL. i.e.  
you have to perform careful convergence tests in order to be sure  
that your PW basis set is adequate to describe the properties that  
you want with the accuracy that you need (the same holds, of course,  
for many other devil's details of the calculation: k-points, gaussian  
smearing, etc.)

> But i donot have much idear about ecutwfs in ph calculations .

By and large, (squared) phonon frequencies are second derivatives of  
the crystal energy for a given kinetic-energy cutoff. If a cutoff is  
OK for geometry optimization, it  is also so for phonon calculation  
(see below)

> Would you please give me some refs of more detail settings about PH  
> calculations ?

S. Baroni, A. Dal Corso, S. de Gironcoli, and P. Giannozzi, Rev. Mod.  
Phys. 73, 515 (2001).

Enjoy!

Stefano B.

PS: for us, it is considered polite to sign with our first name, or  
with the first name followed by the family name. It would be unpolite  
(except for formal business correspondence) to prepend the title to  
the signature - "Stefano" or "Stefano B.": confidential, OK; "Stefano  
Baroni" a bit less confidential, also OK in most cases; "Prof.  
Stefano Baroni": very formal and kind of assuming and not-so-polite  
in many cases - OK in a few

>
>
> I am trying to find more problems related to this wrong result  
> following your suggestion.
> Thank you again, it really helps me much.
>
> p.s.  It is nice for me listening  your lectures in Beijing , i  
> really learn much from your 3 lectures .
>
>
> Regards,
>
> Hai-Ping
>
>
> On 8/20/06, Nicola Marzari <marzari at mit.edu> wrote:
>
>
>
> Dear Haiping,
>
>
> there are two possibilities that lead to negative phonon  
> frequencies at
> gamma.
>
> The first one is less likely:
>
> 1) you have not carefully relaxed the *internal* degrees of freedom -
> i.e. the ionic coodinates. Then, the phonon code discovers that moving
> the atoms along a certain pattern (an "eigenvector" of the dynamical
> matrix at gamma) the energy goes down. Perfect - just relax the ions
> more. Relaxation of the unit cell has nothing to do with this
>
> The second one is more likely:
>
> 2) the calculation is not converged with respect
> to the plane wave cutoff needed, and the number of k-points needed,
> for the small or zero frequencies at gamma. Obtaining an accurate,
> faithful description of low energy modes requires cutoffs and  
> especially
> duals much higher than anything you are used to (to give you an  
> example,
> for an ultrasoft C for diamond, you need 100 Ry in the wavefunction  
> and
> 2800 Ry in the charge density to get a convergence, at gamma,
> of the zero modes down to the last cm-1). Does it mean you need
> to use 100/2800 Ry for diamond ? No ! By enforcing the translation sum
> rules that the interatomic force constants must satisfy, you can  
> monitor
> (after enforcing said rules) the convergence of the low frequency  
> modes,
> and discover you can use much more reasonable cutoffs.
>
> Still, even more than a regular calculations, phonons require a
> through control on the parameters that define your calculation.
>
> Note that, as a warm-up, it is easier to converge LDA calculations
> than GGA calculations (the exchange-correlation terms, calculated in
> real space, introduce very high frequencies in reciprocal space).
>
>                        nicola
>
>
> >> Dear All,
> >> I also have calculated gamma phonon frequencies of anatase TiO2 .
> >> I run ph.x caculation after relaxing the structure, but  9 of 18  
> frequencies
> >> are minus.
> >>
> >> Would you please give me some hints and help?
> >>
> >> input settings are below, thank you in advance !
> >>
> >> *phonons of TiO2 at Gamma
> >>  &inputph
> >>   tr2_ph=1.0d-14,
> >>   recover=.true.,
> >>   epsil=.true.,
> >>   prefix='tio2',
> >>   fildyn='TiO2-gamma.dyn ',
> >>   outdir='/home/haiping/tmp/'
> >>  /
> >> 0.0 0.0 0.0*
>
>
> ---------------------------------------------------------------------
> Prof Nicola Marzari   Department of Materials Science and Engineering
> 13-5066   MIT   77 Massachusetts Avenue   Cambridge MA 02139-4307 USA
> tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060820/054cdf5f/attachment.htm 

From eyvaz_isaev at yahoo.com  Sun Aug 20 16:59:55 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Sun, 20 Aug 2006 07:59:55 -0700 (PDT)
Subject: [Pw_forum] phonon at Gamma point
In-Reply-To: <c92002fa0608191743h458f45f5ubf539f93e99ed772@mail.gmail.com>
Message-ID: <20060820145956.23780.qmail@web60312.mail.yahoo.com>

OK, let me hint more. 

Anatase has a Pearson symbol tI12 or space group
I4_1/amd. So, ibrav should be 7, but you have chosen a
monoclinic lattice, i.e. ibrav=14. The space group has
16 symmetry elements, but your choice after correcting
ecutrho gives only 4 symmetry elements . 

Imagine  FCC lattice and its conventional basis
vectors now. According to you now we have to put
ibrav=4 or 5. But ibrav=2 for FCC lattice. 

In addition, I also checked your input file via
XCrysDen before I replied previously. Yes, there is no
problem with it, but a figure does not look like 
anatase though  you used  nonzero ibrav. XcrysDen
usually replicates atomic positions  if  ibrav.neq.0. 
 For example, if you put ibrav=0  for  FCC cell and
specify conventional CELL_PARAMETERS you will find  a
rhombohedron, but using ibrav=2 and  default basis
vectors you find a cubic cell. 

Bests,
Eyvaz.


--- lan haiping <lanhaiping at gmail.com> wrote:

> Dear Eyvaz,
> Thank you.
> I checked the strcuture again using XCRYSDEN,
> and found no problems related to the  lattices.
> 
> Regards,
> 
> Hai-Ping
> 
> 
> 
> On 8/20/06, Eyvaz Isaev <eyvaz_isaev at yahoo.com>
> wrote:
> >
> > Hi,
> >
> > It seems your lattice choice is not correct.
> Please
> > have a look at
> > http://cst-www.nrl.navy.mil/lattice/struk/c5.html
> >
> > Bests,
> > Eyvaz.
> >
> > --- lan haiping <lanhaiping at gmail.com> wrote:
> >
> > > Dear All,
> > > I also have calculated gamma phonon frequencies
> of
> > > anatase TiO2 .
> > > I run ph.x caculation after relaxing the
> structure,
> > > but  9 of 18 frequencies
> > > are minus.
> > >
> > > Would you please give me some hints and help?
> > >
> > > input settings are below, thank you in advance !
> > >
> > > *phonons of TiO2 at Gamma
> > >  &inputph
> > >   tr2_ph=1.0d-14,
> > >   recover=.true.,
> > >   epsil=.true.,
> > >   prefix='tio2',
> > >   fildyn='TiO2-gamma.dyn',
> > >   outdir='/home/haiping/tmp/'
> > >  /
> > > 0.0 0.0 0.0*
> > > **
> > > *Relaxing calculation input :*
> > > *&CONTROL
> > >              title = 'TiO2' ,
> > >        calculation = 'relax',
> > >          verbosity = 'default',
> > >            tprnfor = .true.,
> > >             outdir = '/home/haiping/tmp/',
> > >             prefix = 'tio2',
> > >            disk_io = 'default',
> > >              nstep = 200,
> > >         pseudo_dir =
> > > '/home/haiping/espresso/pseudo/',
> > > /
> > > &SYSTEM
> > >                 ibrav= 14,
> > >
> > > celldm(1)=10.2856,celldm(2)=1.0,celldm(3)=1.0,
> > >
> > > celldm(4)=-0.759357,celldm(5)=-0.759357,
> > >                celldm(6)=0.518714
> > >                nat = 6 ,
> > >               ntyp = 2 ,
> > >               nbnd = 30 ,
> > >            ecutwfc = 30.0000000000,
> > >            ecutrho = 160.000,
> > >        occupations = 'fixed' ,
> > > /
> > >
> > >
> > > &ELECTRONS
> > >   electron_maxstep = 100,
> > >           conv_thr = 1.0e-07,
> > >        mixing_mode = 'plain' ,
> > >        startingpot = 'atomic' ,
> > >        startingwfc = 'atomic' ,
> > >        mixing_beta = 0.300000000,
> > >        mixing_ndim = 8,
> > >    diagonalization = 'david_overlap' ,
> > >   diago_david_ndim = 8,
> > > /
> > >
> > > &IONS
> > >              ion_dynamics = 'bfgs',
> > >              pot_extrapolation = 'first_order',
> > >              wfc_extrapolation = 'first_order',
> > >
> > > /
> > >
> > > ATOMIC_SPECIES
> > > O  15.999 O.pw91-van_ak.UPF
> > > Ti 47.867  Ti.pw91-nsp-van.UPF
> > > ATOMIC_POSITIONS crystal
> > > Ti    0.00000   0.00000   0.00000
> > > Ti    0.75000   0.25000   0.50000
> > > O     0.20800   0.20800  -0.00000
> > > O     0.95800   0.45800   0.50000
> > > O     0.54200   0.04200   0.50000
> > > O     0.79200   0.79200  -0.00000
> > > K_POINTS   automatic
> > > 6 6 7   0 0 0
> > >
> > > *
> > >
> > >
> > > On 8/19/06, Ezad Shojaee
> <ezadshojaee at hotmail.com>
> > > wrote:
> > > >
> > > > hi
> > > > i am trying to calculate phonon frequencies at
> > > gamma point of
> > > > TiO2(anatase)
> > > > and i have relaxed the structure enough (i
> think
> > > )but performing PH.Xafter
> > > > scf, 2of 18 frequencies  are minus!
> > > > i want to know that what technical problem do
> i
> > > have? does anyone
> > > > calculate
> > > > these beore?
> > > > any suggestion will be appreciated
> > > >
> > > >
> > >
> >
 
--- lan haiping <lanhaiping at gmail.com> wrote:

> Dear Eyvaz,
> Thank you.
> I checked the strcuture again using XCRYSDEN,
> and found no problems related to the  lattices.
> 
> Regards,
> 
> Hai-Ping
> 
> 
> 
> On 8/20/06, Eyvaz Isaev <eyvaz_isaev at yahoo.com>
> wrote:
> >
> > Hi,
> >
> > It seems your lattice choice is not correct.
> Please
> > have a look at
> > http://cst-www.nrl.navy.mil/lattice/struk/c5.html
> >
> > Bests,
> > Eyvaz.
> >
> > --- lan haiping <lanhaiping at gmail.com> wrote:
> >
> > > Dear All,
> > > I also have calculated gamma phonon frequencies
> of
> > > anatase TiO2 .
> > > I run ph.x caculation after relaxing the
> structure,
> > > but  9 of 18 frequencies
> > > are minus.
> > >
> > > Would you please give me some hints and help?
> > >
> > > input settings are below, thank you in advance !
> > >
> > > *phonons of TiO2 at Gamma
> > >  &inputph
> > >   tr2_ph=1.0d-14,
> > >   recover=.true.,
> > >   epsil=.true.,
> > >   prefix='tio2',
> > >   fildyn='TiO2-gamma.dyn',
> > >   outdir='/home/haiping/tmp/'
> > >  /
> > > 0.0 0.0 0.0*
> > > **
> > > *Relaxing calculation input :*
> > > *&CONTROL
> > >              title = 'TiO2' ,
> > >        calculation = 'relax',
> > >          verbosity = 'default',
> > >            tprnfor = .true.,
> > >             outdir = '/home/haiping/tmp/',
> > >             prefix = 'tio2',
> > >            disk_io = 'default',
> > >              nstep = 200,
> > >         pseudo_dir =
> > > '/home/haiping/espresso/pseudo/',
> > > /
> > > &SYSTEM
> > >                 ibrav= 14,
> > >
> > > celldm(1)=10.2856,celldm(2)=1.0,celldm(3)=1.0,
> > >
> > > celldm(4)=-0.759357,celldm(5)=-0.759357,
> > >                celldm(6)=0.518714
> > >                nat = 6 ,
> > >               ntyp = 2 ,
> > >               nbnd = 30 ,
> > >            ecutwfc = 30.0000000000,
> > >            ecutrho = 160.000,
> > >        occupations = 'fixed' ,
> > > /
> > >
> > >
> > > &ELECTRONS
> > >   electron_maxstep = 100,
> > >           conv_thr = 1.0e-07,
> > >        mixing_mode = 'plain' ,
> > >        startingpot = 'atomic' ,
> > >        startingwfc = 'atomic' ,
> > >        mixing_beta = 0.300000000,
> > >        mixing_ndim = 8,
> > >    diagonalization = 'david_overlap' ,
> > >   diago_david_ndim = 8,
> > > /
> > >
> > > &IONS
> > >              ion_dynamics = 'bfgs',
> > >              pot_extrapolation = 'first_order',
> > >              wfc_extrapolation = 'first_order',
> > >
> > > /
> > >
> > > ATOMIC_SPECIES
> > > O  15.999 O.pw91-van_ak.UPF
> > > Ti 47.867  Ti.pw91-nsp-van.UPF
> > > ATOMIC_POSITIONS crystal
> > > Ti    0.00000   0.00000   0.00000
> > > Ti    0.75000   0.25000   0.50000
> > > O     0.20800   0.20800  -0.00000
> > > O     0.95800   0.45800   0.50000
> > > O     0.54200   0.04200   0.50000
> > > O     0.79200   0.79200  -0.00000
> > > K_POINTS   automatic
> > > 6 6 7   0 0 0
> > >
> > > *
> > >
> > >
> > > On 8/19/06, Ezad Shojaee
> <ezadshojaee at hotmail.com>
> > > wrote:
> > > >
> > > > hi
> > > > i am trying to calculate phonon frequencies at
> > > gamma point of
> > > > TiO2(anatase)
> > > > and i have relaxed the structure enough (i
> think
> > > )but performing PH.Xafter
> > > > scf, 2of 18 frequencies  are minus!
> > > > i want to know that what technical problem do
> i
> > > have? does anyone
> > > > calculate
> > > > these beore?
> > > > any suggestion will be appreciated
> > > >
> > > >
> > >
> >
>
_________________________________________________________________
> > > > Express yourself instantly with MSN Messenger!
> > > Download today it's FREE!
> > > >
> > >
> >
>
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
> > > >
> > > >
> _______________________________________________
> > > > Pw_forum mailing list
> > > > Pw_forum at pwscf.org
> > > >
> http://www.democritos.it/mailman/listinfo/pw_forum
> > > >
> > >
> >
> >
> > __________________________________________________
> > Do You Yahoo!?
> > Tired of spam?  Yahoo! Mail has the best spam
> protection around
> > http://mail.yahoo.com
> > _______________________________________________
> > Pw_forum mailing list
> > Pw_forum at pwscf.org
> > http://www.democritos.it/mailman/listinfo/pw_forum
> >
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From xzheng2 at ncsu.edu  Sun Aug 20 19:36:40 2006
From: xzheng2 at ncsu.edu (xzheng2 at ncsu.edu)
Date: Sun, 20 Aug 2006 13:36:40 -0400 (EDT)
Subject: [Pw_forum] charge is wrong
In-Reply-To: <c92002fa0608191815j45d52622xc6c80ff9a10e9068@mail.gmail.com>
References: <BAY124-F26D6306CF7801FFB9DCFA9AA430@phx.gbl>
    <c92002fa0608190538y24a9f4devbaadcee0ecefd55e@mail.gmail.com>
    <1156026985.44e7926954412@webmail.usask.ca> <44E79BE0.1080308@mit.edu>
    <c92002fa0608191815j45d52622xc6c80ff9a10e9068@mail.gmail.com>
Message-ID: <57282.152.14.74.125.1156095400.squirrel@webmail.ncsu.edu>

Dear all,

In my calculation,the following error happened:

===========================================================
     WARNING: integrated charge=   186.30000000, expected=   187.00000000

 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     from electrons : error #         1
     charge is wrong
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

     stopping ...
===========================================================

It stopped here. I am a new user of PWscf. I do not know what to do now.
Which parameter should I modify so that the calculation will continue,
please? Thank you very much!



From akohlmey at cmm.upenn.edu  Sun Aug 20 20:08:10 2006
From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer)
Date: Sun, 20 Aug 2006 14:08:10 -0400
Subject: [Pw_forum] charge is wrong
In-Reply-To: <57282.152.14.74.125.1156095400.squirrel@webmail.ncsu.edu>
References: <BAY124-F26D6306CF7801FFB9DCFA9AA430@phx.gbl>
	 <c92002fa0608190538y24a9f4devbaadcee0ecefd55e@mail.gmail.com>
	 <1156026985.44e7926954412@webmail.usask.ca> <44E79BE0.1080308@mit.edu>
	 <c92002fa0608191815j45d52622xc6c80ff9a10e9068@mail.gmail.com>
	 <57282.152.14.74.125.1156095400.squirrel@webmail.ncsu.edu>
Message-ID: <7b6913e90608201108j75cac4fcr46c3f9b5d70eee3d@mail.gmail.com>

On 8/20/06, xzheng2 at ncsu.edu <xzheng2 at ncsu.edu> wrote:
> Dear all,
>
> In my calculation,the following error happened:

hi,

these kind of things usually happen, when you have an
error in your input (e.g. you didn't do a spin polarized calculation
with an odd number of electrons). other than that there is little
else to help you without seeing your input file.

> ===========================================================
>      WARNING: integrated charge=   186.30000000, expected=   187.00000000
>
>  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>      from electrons : error #         1
>      charge is wrong
>  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>
>      stopping ...
> ===========================================================
>
> It stopped here. I am a new user of PWscf. I do not know what to do now.
> Which parameter should I modify so that the calculation will continue,


you not only want to get your calculation to continue, you want the
calculation to produce a _correct result_. note that your calculation
can finish and still produce a totally bogus result. with DFT quite a
few things are mathematically correct, so can be computed, but still
have no physical meaning. so you always have to make sure you
know what you are doing and double check everything.

cheers,
    axel.

> please? Thank you very much!
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
>


-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
  Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.


From xzheng2 at ncsu.edu  Sun Aug 20 20:39:33 2006
From: xzheng2 at ncsu.edu (Zheng Xiaohong)
Date: Sun, 20 Aug 2006 14:39:33 -0400 (EDT)
Subject: [Pw_forum] charge is wrong
In-Reply-To: <7b6913e90608201108j75cac4fcr46c3f9b5d70eee3d@mail.gmail.com>
References: <BAY124-F26D6306CF7801FFB9DCFA9AA430@phx.gbl>
    <c92002fa0608190538y24a9f4devbaadcee0ecefd55e@mail.gmail.com>
    <1156026985.44e7926954412@webmail.usask.ca> <44E79BE0.1080308@mit.edu>
    <c92002fa0608191815j45d52622xc6c80ff9a10e9068@mail.gmail.com>
    <57282.152.14.74.125.1156095400.squirrel@webmail.ncsu.edu>
    <7b6913e90608201108j75cac4fcr46c3f9b5d70eee3d@mail.gmail.com>
Message-ID: <57363.152.14.74.125.1156099173.squirrel@webmail.ncsu.edu>

Thank you very much!! My input file is as follows:

===========================================================
&CONTROL
  calculation  = "relax",
  prefix       = "Au17",
  pseudo_dir   = "/spin/home/zheng/espresso-3.1.1/pseudo",
  outdir       = "/tmp/work/zheng/pwscf/tmp",
/
&SYSTEM
  ibrav     = 1,
  celldm(1) = 60.D0,
  nat       = 17,
  ntyp      = 1,
  ecutwfc   = 24.D0,
  ecutrho   = 144.D0,
/
&ELECTRONS
  conv_thr    = 1.D-5,
  mixing_beta = 0.7D0,
/
&IONS
  pot_extrapolation = "second_order",
  wfc_extrapolation = "second_order",
/
ATOMIC_SPECIES
Au 197.0  Au.pbe-nd-van.UPF
ATOMIC_POSITIONS {angstrom}
Au     4.23819    3.15126    3.98519
Au     2.79854    4.30837    6.75063
Au     1.54105    3.26386    4.27488
Au     4.19307    1.13230    6.09571
Au     0.48371    0.92178    4.84325
Au     2.91535    0.57625    3.82944
Au     2.77156    5.50267    4.15187
Au     5.54670    0.59657    3.84844
Au     6.66741    1.86328    7.07452
Au     4.20412    6.53544    6.69729
......

K_POINTS {Gamma}

======================================================

Do you think this is a bug of this program. Why will it stop for odd
number of electrons? This will not happen in some other software, e.g.
WIEN2K or VASP.

If I do spin-polarized calculation, will it go on?

Xiaohong


> On 8/20/06, xzheng2 at ncsu.edu <xzheng2 at ncsu.edu> wrote:
>> Dear all,
>>
>> In my calculation,the following error happened:
>
> hi,
>
> these kind of things usually happen, when you have an
> error in your input (e.g. you didn't do a spin polarized calculation
> with an odd number of electrons). other than that there is little
> else to help you without seeing your input file.
>
>> ===========================================================
>>      WARNING: integrated charge=   186.30000000, expected=
>> 187.00000000

>>
>>  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>>      from electrons : error #         1
>>      charge is wrong
>>  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>>
>>      stopping ...
>> ===========================================================
>>
>> It stopped here. I am a new user of PWscf. I do not know what to do now.
>> Which parameter should I modify so that the calculation will continue,
>
>
> you not only want to get your calculation to continue, you want the
> calculation to produce a _correct result_. note that your calculation
> can finish and still produce a totally bogus result. with DFT quite a
> few things are mathematically correct, so can be computed, but still
> have no physical meaning. so you always have to make sure you
> know what you are doing and double check everything.
>
> cheers,
>     axel.
>
>> please? Thank you very much!
>>
>> _______________________________________________
>> Pw_forum mailing list
>> Pw_forum at pwscf.org
>> http://www.democritos.it/mailman/listinfo/pw_forum
>>
>>
>
>
> --
> =======================================================================
> Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
>   Center for Molecular Modeling   --   University of Pennsylvania
> Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
> tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
> =======================================================================
> If you make something idiot-proof, the universe creates a better idiot.
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>




From eyvaz_isaev at yahoo.com  Sun Aug 20 21:20:52 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Sun, 20 Aug 2006 12:20:52 -0700 (PDT)
Subject: [Pw_forum] charge is wrong
In-Reply-To: <57363.152.14.74.125.1156099173.squirrel@webmail.ncsu.edu>
Message-ID: <20060820192052.74710.qmail@web60311.mail.yahoo.com>

Hi, 

Your system is considered as a semicoductor until you
did not specify occupations. For metallic systems like
Au bands at the  Fermi level are occupied  partially,
but for semiconductros these bands are filled
completely. 
Presumably, it would be nice if you read special
sections of Kittel or Ashkroft&Mermin's textbook about
 metals and semiconductors. 

Well, now you have to add 3 more keynotes in &system:

occupations="smearing"
smearing='m-v' (Nikola likes very much this. Think
about, why?)
degauss=0.025

Besides, increase ecutwfc up to 30, play with ecutrho
and decrease conv_thr down to -6 or better.

Please read /Doc/INPUT_PW what they are mean.

Bests,
Eyvaz.
 

--- Zheng Xiaohong <xzheng2 at ncsu.edu> wrote:

> Thank you very much!! My input file is as follows:
> 
>
===========================================================
> &CONTROL
>   calculation  = "relax",
>   prefix       = "Au17",
>   pseudo_dir   =
> "/spin/home/zheng/espresso-3.1.1/pseudo",
>   outdir       = "/tmp/work/zheng/pwscf/tmp,
> /
> &SYSTEM
>   ibrav     = 1,
>   celldm(1) = 60.D0,
>   nat       = 17,
>   ntyp      = 1,
>   ecutwfc   = 24.D0,
>   ecutrho   = 144.D0,
> /
> &ELECTRONS
>   conv_thr    = 1.D-5,
>   mixing_beta = 0.7D0,
> /
> &IONS
>   pot_extrapolation = "second_order",
>   wfc_extrapolation = "second_order",
> /
> ATOMIC_SPECIES
> Au 197.0  Au.pbe-nd-van.UPF
> ATOMIC_POSITIONS {angstrom}
> Au     4.23819    3.15126    3.98519
> Au     2.79854    4.30837    6.75063
> Au     1.54105    3.26386    4.27488
> Au     4.19307    1.13230    6.09571
> Au     0.48371    0.92178    4.84325
> Au     2.91535    0.57625    3.82944
> Au     2.77156    5.50267    4.15187
> Au     5.54670    0.59657    3.84844
> Au     6.66741    1.86328    7.07452
> Au     4.20412    6.53544    6.69729
> ......
> 
> K_POINTS {Gamma}
> 
>
======================================================
> 
> Do you think this is a bug of this program. Why will
> it stop for odd
> number of electrons? This will not happen in some
> other software, e.g.
> WIEN2K or VASP.
> 
> If I do spin-polarized calculation, will it go on?
> 
> Xiaohong
> 
> 
> > On 8/20/06, xzheng2 at ncsu.edu <xzheng2 at ncsu.edu>
> wrote:
> >> Dear all,
> >>
> >> In my calculation,the following error happened:
> >
> > hi,
> >
> > these kind of things usually happen, when you have
> an
> > error in your input (e.g. you didn't do a spin
> polarized calculation
> > with an odd number of electrons). other than that
> there is little
> > else to help you without seeing your input file.
> >
> >>
>
===========================================================
> >>      WARNING: integrated charge=   186.30000000,
> expected=
> >> 187.00000000
> 
> >>
> >> 
>
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
> >>      from electrons : error #         1
> >>      charge is wrong
> >> 
>
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
> >>
> >>      stopping ...
> >>
>
===========================================================
> >>
> >> It stopped here. I am a new user of PWscf. I do
> not know what to do now.
> >> Which parameter should I modify so that the
> calculation will continue,
> >
> >
> > you not only want to get your calculation to
> continue, you want the
> > calculation to produce a _correct result_. note
> that your calculation
> > can finish and still produce a totally bogus
> result. with DFT quite a
> > few things are mathematically correct, so can be
> computed, but still
> > have no physical meaning. so you always have to
> make sure you
> > know what you are doing and double check
> everything.
> >
> > cheers,
> >     axel.
> >
> >> please? Thank you very much!
> >>
> >> _______________________________________________
> >> Pw_forum mailing list
> >> Pw_forum at pwscf.org
> >>
> http://www.democritos.it/mailman/listinfo/pw_forum
> >>
> >>
> >
> >
> > --
> >
>
=======================================================================
> > Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu  
> http://www.cmm.upenn.edu
> >   Center for Molecular Modeling   --   University
> of Pennsylvania
> > Department of Chemistry, 231 S.34th Street,
> Philadelphia, PA 19104-6323
> > tel: 1-215-898-1582,  fax: 1-215-573-6233, 
> office-tel: 1-215-898-5425
> >
>
=======================================================================
> > If you make something idiot-proof, the universe
> creates a better idiot.
> > _______________________________________________
> > Pw_forum mailing list
> > Pw_forum at pwscf.org
> > http://www.democritos.it/mailman/listinfo/pw_forum
> >
> 
> 
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From akohlmey at cmm.upenn.edu  Sun Aug 20 21:38:09 2006
From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer)
Date: Sun, 20 Aug 2006 15:38:09 -0400
Subject: [Pw_forum] charge is wrong
In-Reply-To: <57363.152.14.74.125.1156099173.squirrel@webmail.ncsu.edu>
References: <BAY124-F26D6306CF7801FFB9DCFA9AA430@phx.gbl>
	 <c92002fa0608190538y24a9f4devbaadcee0ecefd55e@mail.gmail.com>
	 <1156026985.44e7926954412@webmail.usask.ca> <44E79BE0.1080308@mit.edu>
	 <c92002fa0608191815j45d52622xc6c80ff9a10e9068@mail.gmail.com>
	 <57282.152.14.74.125.1156095400.squirrel@webmail.ncsu.edu>
	 <7b6913e90608201108j75cac4fcr46c3f9b5d70eee3d@mail.gmail.com>
	 <57363.152.14.74.125.1156099173.squirrel@webmail.ncsu.edu>
Message-ID: <7b6913e90608201238j2cad548du5df4894a7da5f4dd@mail.gmail.com>

On 8/20/06, Zheng Xiaohong <xzheng2 at ncsu.edu> wrote:
>   nat       = 17,

[...]

> ATOMIC_SPECIES
> Au 197.0  Au.pbe-nd-van.UPF

> Do you think this is a bug of this program. Why will it stop for odd
> number of electrons? This will not happen in some other software, e.g.
> WIEN2K or VASP.

well, do you get a correct result in that case?
you should check the output whether the code automatically
switches to LSDA.

> If I do spin-polarized calculation, will it go on?

apart from the following the recommendations of eyvaz,
you should realize that you have an odd number of atoms
for an element with an odd number of electrons...

i guess you can take it from here. ;-)

axel.


>
> Xiaohong
>
>


-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
  Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.


From silviu at Princeton.EDU  Sun Aug 20 21:46:18 2006
From: silviu at Princeton.EDU (Silviu Zilberman)
Date: Sun, 20 Aug 2006 22:46:18 +0300
Subject: [Pw_forum] restart cp on Blue Gene
Message-ID: <44E8BC0A.5070806@Princeton.EDU>

Hi,

I am using the latest cvs on an IBM  Blue Gene machine. It seems that 
simple restart is not working, and I get the following error message:

   reading restart file: .//CO_inhib_CN_up_DTN_cubane_50.save
########################################################################################################################
# UNRECOVERABLE ERROR (ierr=1)
# ERROR IN: iotk_scan_dat (iotk_dat.spp:688)
# CVS Revision: 1.3
########################################################################################################################

I did not experience similar problems on our Linux machines and it used 
to work in the past.

Any ideas?

Below is my input file.

Thanks, Silviu.

-----------------------------------------------------------------------------------
INPUT FILE:

  &CONTROL
    calculation='scf',
    restart_mode = 'restart',
    iprint = 10000,
    isave  = 25000,
    dt     = 7.0,
    pseudo_dir = '/home0/silviu/Pot/',
    outdir = '/scr0/silviu/CO_inhib_CN_up_DTN_cubane',
    prefix = 'CO_inhib_CN_up_DTN_cubane',
    nstep  = 100000,
    etot_conv_thr = 1.d-5,
    ekin_conv_thr = 1.d-6,
 /
  &SYSTEM
    ibrav = 1,
    celldm(1) = 35.0d0,
    nat = 53,
    ntyp = 6,
    ecutwfc =30.0,
    ecutrho =180.0,
    nr1b=18, nr2b=18,  nr3b=18,
    nspin=2,
    tot_charge=-3.0,
    multiplicity=2,
  /
 &ELECTRONS
    emass = 700.d0,
    emass_cutoff = 0.8,
    orthogonalization = 'ortho',
    ortho_eps = 1.e-8,
    ortho_max = 100,
    electron_dynamics= 'damp',
    electron_damping =0.08,
    electron_velocities = 'zero',
 /
 &IONS
    ion_dynamics = 'none',
    ion_damping = 0.001
  /
AUTOPILOT
    on_step = 20 : dt = 10
    on_step = 30 : dt = 15
    on_step = 200 : dt = 7
    on_step = 200 : ion_dynamics = 'damp'
ENDRULES
ATOMIC_SPECIES
   Fe 55.845  fe_PBE_filippo.van
   S  31.9720 016-S-gpbe--bm.van
   O  15.9994 008-O-gpbe--bm.van
   N  14.0030 007-N-gpbe--bm.van
   C  12.0107 006-C-gpbe--bm.van
   H   1.0079 001-H-gpbe--bm.van
ATOMIC_POSITIONS bohr
....



From akohlmey at cmm.upenn.edu  Sun Aug 20 21:57:14 2006
From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer)
Date: Sun, 20 Aug 2006 15:57:14 -0400
Subject: [Pw_forum] restart cp on Blue Gene
In-Reply-To: <44E8BC0A.5070806@Princeton.EDU>
References: <44E8BC0A.5070806@Princeton.EDU>
Message-ID: <7b6913e90608201257p67239facp7d7d1a80992dee9b@mail.gmail.com>

On 8/20/06, Silviu Zilberman <silviu at princeton.edu> wrote:
> Hi,


hi.

> I am using the latest cvs on an IBM  Blue Gene machine. It seems that
> simple restart is not working, and I get the following error message:

can you read other restart files?
did you use the make.sys file provided in the distribution?

>    reading restart file: .//CO_inhib_CN_up_DTN_cubane_50.save
> ########################################################################################################################
> # UNRECOVERABLE ERROR (ierr=1)
> # ERROR IN: iotk_scan_dat (iotk_dat.spp:688)
> # CVS Revision: 1.3
> ########################################################################################################################
>
> I did not experience similar problems on our Linux machines and it used
> to work in the past.

this is most likely an issue with the ibm compilers. i've been through
that very recently on an IBM p690 AIX machine. you error looks very
similar, but is not quite the same. so it would be interesting to know,
how you compiled it.

ciao,
   axel.

>
> Any ideas?
>
> Below is my input file.
>
> Thanks, Silviu.
>

-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
  Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.


From eyvaz_isaev at yahoo.com  Sun Aug 20 23:01:18 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Sun, 20 Aug 2006 14:01:18 -0700 (PDT)
Subject: [Pw_forum] charge is wrong
In-Reply-To: <7b6913e90608201238j2cad548du5df4894a7da5f4dd@mail.gmail.com>
Message-ID: <20060820210118.68611.qmail@web60313.mail.yahoo.com>

Hi again,

> > Do you think this is a bug of this program. 
NO, it is your own "bug", hopefully.

> Why will it stop for odd  number of electrons? 
Because you have 11*17=187 electrons, but due to an
integration schem you provided the program can not
collect all these electrons. ODD NUMBER of ELECTRONS
DEFINITELY is for metallic systems. 

> This will not happen in some  other software, e.g. 
> WIEN2K or VASP.
At least in VASP by default ISMEAR=1, i.e.
Methfessel-Paxton broadening technique is used. So,
your system was considered as a metallic. More likely,
a similar schem was used for WIEN2K.

> If I do spin-polarized calculation, will it go on?
I do not think so if you treat  your system as a
semiconductor. In such a case you will find total
number of electrons with spin up to be equal  total
number of spin down electrons. I,e. total number of
electrons is even.   

Besides, spin-polarized calculation is required if you
have a magnetic atom in your system. Otherwise it is
useless.  

Bests,
Eyvaz.

> well, do you get a correct result in that case?
> you should check the output whether the code
> automatically switches to LSDA.
> 



> apart from the following the recommendations of
> eyvaz,
> you should realize that you have an odd number of
> atoms
> for an element with an odd number of electrons...
> 
> i guess you can take it from here. ;-)
> 
> axel.
> 
> 
> >
> > Xiaohong
> >
> >
> 
> 
> -- 
>
=======================================================================
> Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu  
> http://www.cmm.upenn.edu
>   Center for Molecular Modeling   --   University of
> Pennsylvania
> Department of Chemistry, 231 S.34th Street,
> Philadelphia, PA 19104-6323
> tel: 1-215-898-1582,  fax: 1-215-573-6233, 
> office-tel: 1-215-898-5425
>
=======================================================================
> If you make something idiot-proof, the universe
> creates a better idiot.
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From tr.kelkar at ncl.res.in  Mon Aug 21 07:54:13 2006
From: tr.kelkar at ncl.res.in (Tuhina Kelkar)
Date: Mon, 21 Aug 2006 11:24:13 +0530 (IST)
Subject: [Pw_forum] doing QHA calculations (contd)
In-Reply-To: <20060819053803.15930.30563.Mailman@democritos.sissa.it>
Message-ID: <1956221.1156139653516.JavaMail.root@webmail.ncl.res.in>

I checked my .phdos and found that the nu(i) values are very large. That is why even after commenting the "wrong grid" line in fqha.f90 and recompiling (as suggested by Eyvaz) I am getting Ftot as NaN in the output.
Where am I going wrong in the phonon DOS calcuations? What can I do to rectify this? Could you suggest some books and papers that I could refer to develop a better understanding of phonon DOS and quasi harmonic approximation?
Regards,
Tuhina


    
        
            
-- Original Message --
From: pw_forum-request at pwscf.org
To: pw_forum at pwscf.org
Date: Sat, 19 Aug 2006 07:38:03 +0200
Subject: Pw_forum digest, Vol 1 #1094 - 8 msgs
Message: 1
Date: Fri, 18 Aug 2006 14:51:32 +0530 (IST)
From: Tuhina Kelkar <tr.kelkar at ncl.res.in>
To: pw_forum at pwscf.org
Subject: [Pw_forum] doing QHA calculations
Reply-To: pw_forum at pwscf.org
Message: 5
Date: Fri, 18 Aug 2006 11:43:57 -0700 (PDT)
From: Eyvaz Isaev <eyvaz_isaev at yahoo.com>
Subject: Re: [Pw_forum] doing QHA calculations
To: pw_forum at pwscf.org
Reply-To: pw_forum at pwscf.org
Hi, 
> I get the  following message everytime "wrong grid" 
> and the program stops. What does it mean by wrong 
> grid and  how do I correct this error?
If you are completely sure what you are doing just
comment the line in fqha.f90 and recompile it.
> Please reply asap. 
The fastest way is to understand why it happens.
Otherwise please  be patient.
Bests,
Eyvaz.
Thanks in advance.
> Tuhina
>
---------------------------------------------------------------------------------------------------------------
> Tuhina Kelkar
> Junior Research Fellow
> Physical Chemistry Division,
> National Chemical Laboratory,
> Pune - 411 008
> INDIA
> Contact: 020-25902298
> alternative email: tuhina at gmail.com
>
---------------------------------------------------------------------------------------------------------------
> 
http://www.democritos.it/mailman/listinfo/pw_forum
End of Pw_forum Digest
*****************************************************************
This email is virus free by TrendMicro Inter Scan Security Suite.
*****************************************************************

        
    

-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060821/86401094/attachment.htm 

From eyvaz_isaev at yahoo.com  Mon Aug 21 09:49:33 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Mon, 21 Aug 2006 00:49:33 -0700 (PDT)
Subject: [Pw_forum] doing QHA calculations (contd)
In-Reply-To: <1956221.1156139653516.JavaMail.root@webmail.ncl.res.in>
Message-ID: <20060821074933.47976.qmail@web60316.mail.yahoo.com>

Hi, 

start  a loop (over nu) in fqha.f90 from 2 instead of
1. nu(1) is zero, so, ln(nu) is not defined.

Bests,
Eyvaz
--- Tuhina Kelkar <tr.kelkar at ncl.res.in> wrote:

> I checked my .phdos and found that the nu(i) values
> are very large. That is why even after commenting
> the "wrong grid" line in fqha.f90 and recompiling
> (as suggested by Eyvaz) I am getting Ftot as NaN in
> the output.
> Where am I going wrong in the phonon DOS
> calcuations? What can I do to rectify this? Could
> you suggest some books and papers that I could refer
> to develop a better understanding of phonon DOS and
> quasi harmonic approximation?
> Regards,
> Tuhina
> 
> 
>     
>         
>             
> -- Original Message --
> From: pw_forum-request at pwscf.org
> To: pw_forum at pwscf.org
> Date: Sat, 19 Aug 2006 07:38:03 +0200
> Subject: Pw_forum digest, Vol 1 #1094 - 8 msgs
> Message: 1
> Date: Fri, 18 Aug 2006 14:51:32 +0530 (IST)
> From: Tuhina Kelkar <tr.kelkar at ncl.res.in>
> To: pw_forum at pwscf.org
> Subject: [Pw_forum] doing QHA calculations
> Reply-To: pw_forum at pwscf.org
> Message: 5
> Date: Fri, 18 Aug 2006 11:43:57 -0700 (PDT)
> From: Eyvaz Isaev <eyvaz_isaev at yahoo.com>
> Subject: Re: [Pw_forum] doing QHA calculations
> To: pw_forum at pwscf.org
> Reply-To: pw_forum at pwscf.org
> Hi, 
> > I get the  following message everytime "wrong
> grid" 
> > and the program stops. What does it mean by wrong 
> > grid and  how do I correct this error?
> If you are completely sure what you are doing just
> comment the line in fqha.f90 and recompile it.
> > Please reply asap. 
> The fastest way is to understand why it happens.
> Otherwise please  be patient.
> Bests,
> Eyvaz.
> Thanks in advance.
> > Tuhina
> >
>
---------------------------------------------------------------------------------------------------------------
> > Tuhina Kelkar
> > Junior Research Fellow
> > Physical Chemistry Division,
> > National Chemical Laboratory,
> > Pune - 411 008
> > INDIA
> > Contact: 020-25902298
> > alternative email: tuhina at gmail.com
> >
>
---------------------------------------------------------------------------------------------------------------
> > 
> http://www.democritos.it/mailman/listinfo/pw_forum
> End of Pw_forum Digest
>
*****************************************************************
> This email is virus free by TrendMicro Inter Scan
> Security Suite.
>
*****************************************************************
> 
>         
>     
> 
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From roberta.cuozzo at st.com  Mon Aug 21 09:58:23 2006
From: roberta.cuozzo at st.com (Roberta CUOZZO)
Date: Mon, 21 Aug 2006 09:58:23 +0200
Subject: [Pw_forum] (no subject)
Message-ID: <004801c6c4f7$941db960$343082a4@ptc.st.com>

Please I wish to delete my subscription to your forum.
Thanks in advance
Roberta Cuozzo



From tr.kelkar at ncl.res.in  Mon Aug 21 11:10:16 2006
From: tr.kelkar at ncl.res.in (Tuhina Kelkar)
Date: Mon, 21 Aug 2006 14:40:16 +0530 (IST)
Subject: [Pw_forum] doing QHA calculations (contd)
Message-ID: <3266061.1156151416512.JavaMail.root@webmail.ncl.res.in>

It struck me with your mention of ln(nu)... a lot of my initial nu(i) values are negative.  Am I doing something wrong? Would it be ok to just discard these values? 
-Tuhina
>Hi,
>start  a loop (over nu) in fqha.f90 from 2 instead of
>1. nu(1) is zero, so, ln(nu) is not defined.
>Bests,
>Eyvaz
--- Tuhina Kelkar <<../../servlet/Compose?To=tr.kelkar at ncl.res.in>tr.kelkar at ncl.res.in> wrote:
> I checked my .phdos and found that the nu(i) values
> are very large. That is why even after commenting
> the "wrong grid" line in fqha.f90 and recompiling
> (as suggested by Eyvaz) I am getting Ftot as NaN in
> the output.
> Where am I going wrong in the phonon DOS
> calcuations? What can I do to rectify this? Could
> you suggest some books and papers that I could refer
> to develop a better understanding of phonon DOS and
> quasi harmonic approximation?
> Regards,
> Tuhina
---------------------------------------------------------------------------------------------------------------
Tuhina Kelkar
Junior Research Fellow
Physical Chemistry Division,
National Chemical Laboratory,
Pune - 411 008
INDIA
Contact: 020-25902298
alternative email: tuhina at gmail.com
---------------------------------------------------------------------------------------------------------------
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060821/7724c6a4/attachment.htm 

From eyvaz_isaev at yahoo.com  Mon Aug 21 11:58:55 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Mon, 21 Aug 2006 02:58:55 -0700 (PDT)
Subject: [Pw_forum] doing QHA calculations (contd)
In-Reply-To: <3266061.1156151416512.JavaMail.root@webmail.ncl.res.in>
Message-ID: <20060821095855.8764.qmail@web60320.mail.yahoo.com>

In this case I guess you can not calculate the free
energy. I am aware that somebody just neglected this
(imaginary) part when they got imaginary frequences,
but the correctness of this approximation is under
question, in my personal opinion. At least, you can
calculate zero point energy. 

Bests,
Eyvaz.  

--- Tuhina Kelkar <tr.kelkar at ncl.res.in> wrote:

> It struck me with your mention of ln(nu)... a lot of
> my initial nu(i) values are negative.  Am I doing
> something wrong? Would it be ok to just discard
> these values? 
> -Tuhina
> >Hi,
> >start  a loop (over nu) in fqha.f90 from 2 instead
> of
> >1. nu(1) is zero, so, ln(nu) is not defined.
> >Bests,
> >Eyvaz
> --- Tuhina Kelkar
>
<<../../servlet/Compose?To=tr.kelkar at ncl.res.in>tr.kelkar at ncl.res.in>
> wrote:
> > I checked my .phdos and found that the nu(i)
> values
> > are very large. That is why even after commenting
> > the "wrong grid" line in fqha.f90 and recompiling
> > (as suggested by Eyvaz) I am getting Ftot as NaN
> in
> > the output.
> > Where am I going wrong in the phonon DOS
> > calcuations? What can I do to rectify this? Could
> > you suggest some books and papers that I could
> refer
> > to develop a better understanding of phonon DOS
> and
> > quasi harmonic approximation?
> > Regards,
> > Tuhina
>
---------------------------------------------------------------------------------------------------------------
> Tuhina Kelkar
> Junior Research Fellow
> Physical Chemistry Division,
> National Chemical Laboratory,
> Pune - 411 008
> INDIA
> Contact: 020-25902298
> alternative email: tuhina at gmail.com
>
---------------------------------------------------------------------------------------------------------------
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From lukethulin at netscape.net  Mon Aug 21 14:13:35 2006
From: lukethulin at netscape.net (Luke Thulin)
Date: Mon, 21 Aug 2006 08:13:35 -0400
Subject: [Pw_forum] phonon at Gamma point
In-Reply-To: <20060820145956.23780.qmail@web60312.mail.yahoo.com>
References: <20060820145956.23780.qmail@web60312.mail.yahoo.com>
Message-ID: <44E9A36F.7000003@netscape.net>

I believe Eyvaz is correct, for anatase use:

ibrav = 7
A choice of cartesian coordinates for measured values:
Ti (0,0,0)
Ti (0,a/2,c/4)
O (0,0,-uc)
O (0,0,+uc)
O (0,a/2,c/4-uc)
O (0,a/2,c/4+uc)

where a = 3.785, c=9.514, u=.2081

Luke
eyvaz_isaev at yahoo.com wrote:

>OK, let me hint more. 
>
>Anatase has a Pearson symbol tI12 or space group
>I4_1/amd. So, ibrav should be 7, but you have chosen a
>monoclinic lattice, i.e. ibrav=14. The space group has
>16 symmetry elements, but your choice after correcting
>ecutrho gives only 4 symmetry elements . 
>
>Imagine  FCC lattice and its conventional basis
>vectors now. According to you now we have to put
>ibrav=4 or 5. But ibrav=2 for FCC lattice. 
>
>In addition, I also checked your input file via
>XCrysDen before I replied previously. Yes, there is no
>problem with it, but a figure does not look like 
>anatase though  you used  nonzero ibrav. XcrysDen
>usually replicates atomic positions  if  ibrav.neq.0. 
> For example, if you put ibrav=0  for  FCC cell and
>specify conventional CELL_PARAMETERS you will find  a
>rhombohedron, but using ibrav=2 and  default basis
>vectors you find a cubic cell. 
>
>Bests,
>Eyvaz.
>
>
>--- lan haiping <lanhaiping at gmail.com> wrote:
>
>  
>
>>Dear Eyvaz,
>>Thank you.
>>I checked the strcuture again using XCRYSDEN,
>>and found no problems related to the  lattices.
>>
>>Regards,
>>
>>Hai-Ping
>>
>>
>>
>>On 8/20/06, Eyvaz Isaev <eyvaz_isaev at yahoo.com>
>>wrote:
>>    
>>
>>>Hi,
>>>
>>>It seems your lattice choice is not correct.
>>>      
>>>
>>Please
>>    
>>
>>>have a look at
>>>http://cst-www.nrl.navy.mil/lattice/struk/c5.html
>>>
>>>Bests,
>>>Eyvaz.
>>>
>>>--- lan haiping <lanhaiping at gmail.com> wrote:
>>>
>>>      
>>>
>>>>Dear All,
>>>>I also have calculated gamma phonon frequencies
>>>>        
>>>>
>>of
>>    
>>
>>>>anatase TiO2 .
>>>>I run ph.x caculation after relaxing the
>>>>        
>>>>
>>structure,
>>    
>>
>>>>but  9 of 18 frequencies
>>>>are minus.
>>>>
>>>>Would you please give me some hints and help?
>>>>
>>>>input settings are below, thank you in advance !
>>>>
>>>>*phonons of TiO2 at Gamma
>>>> &inputph
>>>>  tr2_ph=1.0d-14,
>>>>  recover=.true.,
>>>>  epsil=.true.,
>>>>  prefix='tio2',
>>>>  fildyn='TiO2-gamma.dyn',
>>>>  outdir='/home/haiping/tmp/'
>>>> /
>>>>0.0 0.0 0.0*
>>>>**
>>>>*Relaxing calculation input :*
>>>>*&CONTROL
>>>>             title = 'TiO2' ,
>>>>       calculation = 'relax',
>>>>         verbosity = 'default',
>>>>           tprnfor = .true.,
>>>>            outdir = '/home/haiping/tmp/',
>>>>            prefix = 'tio2',
>>>>           disk_io = 'default',
>>>>             nstep = 200,
>>>>        pseudo_dir =
>>>>'/home/haiping/espresso/pseudo/',
>>>>/
>>>>&SYSTEM
>>>>                ibrav= 14,
>>>>
>>>>celldm(1)=10.2856,celldm(2)=1.0,celldm(3)=1.0,
>>>>
>>>>celldm(4)=-0.759357,celldm(5)=-0.759357,
>>>>               celldm(6)=0.518714
>>>>               nat = 6 ,
>>>>              ntyp = 2 ,
>>>>              nbnd = 30 ,
>>>>           ecutwfc = 30.0000000000,
>>>>           ecutrho = 160.000,
>>>>       occupations = 'fixed' ,
>>>>/
>>>>
>>>>
>>>>&ELECTRONS
>>>>  electron_maxstep = 100,
>>>>          conv_thr = 1.0e-07,
>>>>       mixing_mode = 'plain' ,
>>>>       startingpot = 'atomic' ,
>>>>       startingwfc = 'atomic' ,
>>>>       mixing_beta = 0.300000000,
>>>>       mixing_ndim = 8,
>>>>   diagonalization = 'david_overlap' ,
>>>>  diago_david_ndim = 8,
>>>>/
>>>>
>>>>&IONS
>>>>             ion_dynamics = 'bfgs',
>>>>             pot_extrapolation = 'first_order',
>>>>             wfc_extrapolation = 'first_order',
>>>>
>>>>/
>>>>
>>>>ATOMIC_SPECIES
>>>>O  15.999 O.pw91-van_ak.UPF
>>>>Ti 47.867  Ti.pw91-nsp-van.UPF
>>>>ATOMIC_POSITIONS crystal
>>>>Ti    0.00000   0.00000   0.00000
>>>>Ti    0.75000   0.25000   0.50000
>>>>O     0.20800   0.20800  -0.00000
>>>>O     0.95800   0.45800   0.50000
>>>>O     0.54200   0.04200   0.50000
>>>>O     0.79200   0.79200  -0.00000
>>>>K_POINTS   automatic
>>>>6 6 7   0 0 0
>>>>
>>>>*
>>>>
>>>>
>>>>On 8/19/06, Ezad Shojaee
>>>>        
>>>>
>><ezadshojaee at hotmail.com>
>>    
>>
>>>>wrote:
>>>>        
>>>>
>>>>>hi
>>>>>i am trying to calculate phonon frequencies at
>>>>>          
>>>>>
>>>>gamma point of
>>>>        
>>>>
>>>>>TiO2(anatase)
>>>>>and i have relaxed the structure enough (i
>>>>>          
>>>>>
>>think
>>    
>>
>>>>)but performing PH.Xafter
>>>>        
>>>>
>>>>>scf, 2of 18 frequencies  are minus!
>>>>>i want to know that what technical problem do
>>>>>          
>>>>>
>>i
>>    
>>
>>>>have? does anyone
>>>>        
>>>>
>>>>>calculate
>>>>>these beore?
>>>>>any suggestion will be appreciated
>>>>>
>>>>>
>>>>>          
>>>>>
> 
>--- lan haiping <lanhaiping at gmail.com> wrote:
>
>  
>
>>Dear Eyvaz,
>>Thank you.
>>I checked the strcuture again using XCRYSDEN,
>>and found no problems related to the  lattices.
>>
>>Regards,
>>
>>Hai-Ping
>>
>>
>>
>>On 8/20/06, Eyvaz Isaev <eyvaz_isaev at yahoo.com>
>>wrote:
>>    
>>
>>>Hi,
>>>
>>>It seems your lattice choice is not correct.
>>>      
>>>
>>Please
>>    
>>
>>>have a look at
>>>http://cst-www.nrl.navy.mil/lattice/struk/c5.html
>>>
>>>Bests,
>>>Eyvaz.
>>>
>>>--- lan haiping <lanhaiping at gmail.com> wrote:
>>>
>>>      
>>>
>>>>Dear All,
>>>>I also have calculated gamma phonon frequencies
>>>>        
>>>>
>>of
>>    
>>
>>>>anatase TiO2 .
>>>>I run ph.x caculation after relaxing the
>>>>        
>>>>
>>structure,
>>    
>>
>>>>but  9 of 18 frequencies
>>>>are minus.
>>>>
>>>>Would you please give me some hints and help?
>>>>
>>>>input settings are below, thank you in advance !
>>>>
>>>>*phonons of TiO2 at Gamma
>>>> &inputph
>>>>  tr2_ph=1.0d-14,
>>>>  recover=.true.,
>>>>  epsil=.true.,
>>>>  prefix='tio2',
>>>>  fildyn='TiO2-gamma.dyn',
>>>>  outdir='/home/haiping/tmp/'
>>>> /
>>>>0.0 0.0 0.0*
>>>>**
>>>>*Relaxing calculation input :*
>>>>*&CONTROL
>>>>             title = 'TiO2' ,
>>>>       calculation = 'relax',
>>>>         verbosity = 'default',
>>>>           tprnfor = .true.,
>>>>            outdir = '/home/haiping/tmp/',
>>>>            prefix = 'tio2',
>>>>           disk_io = 'default',
>>>>             nstep = 200,
>>>>        pseudo_dir =
>>>>'/home/haiping/espresso/pseudo/',
>>>>/
>>>>&SYSTEM
>>>>                ibrav= 14,
>>>>
>>>>celldm(1)=10.2856,celldm(2)=1.0,celldm(3)=1.0,
>>>>
>>>>celldm(4)=-0.759357,celldm(5)=-0.759357,
>>>>               celldm(6)=0.518714
>>>>               nat = 6 ,
>>>>              ntyp = 2 ,
>>>>              nbnd = 30 ,
>>>>           ecutwfc = 30.0000000000,
>>>>           ecutrho = 160.000,
>>>>       occupations = 'fixed' ,
>>>>/
>>>>
>>>>
>>>>&ELECTRONS
>>>>  electron_maxstep = 100,
>>>>          conv_thr = 1.0e-07,
>>>>       mixing_mode = 'plain' ,
>>>>       startingpot = 'atomic' ,
>>>>       startingwfc = 'atomic' ,
>>>>       mixing_beta = 0.300000000,
>>>>       mixing_ndim = 8,
>>>>   diagonalization = 'david_overlap' ,
>>>>  diago_david_ndim = 8,
>>>>/
>>>>
>>>>&IONS
>>>>             ion_dynamics = 'bfgs',
>>>>             pot_extrapolation = 'first_order',
>>>>             wfc_extrapolation = 'first_order',
>>>>
>>>>/
>>>>
>>>>ATOMIC_SPECIES
>>>>O  15.999 O.pw91-van_ak.UPF
>>>>Ti 47.867  Ti.pw91-nsp-van.UPF
>>>>ATOMIC_POSITIONS crystal
>>>>Ti    0.00000   0.00000   0.00000
>>>>Ti    0.75000   0.25000   0.50000
>>>>O     0.20800   0.20800  -0.00000
>>>>O     0.95800   0.45800   0.50000
>>>>O     0.54200   0.04200   0.50000
>>>>O     0.79200   0.79200  -0.00000
>>>>K_POINTS   automatic
>>>>6 6 7   0 0 0
>>>>
>>>>*
>>>>
>>>>
>>>>On 8/19/06, Ezad Shojaee
>>>>        
>>>>
>><ezadshojaee at hotmail.com>
>>    
>>
>>>>wrote:
>>>>        
>>>>
>>>>>hi
>>>>>i am trying to calculate phonon frequencies at
>>>>>          
>>>>>
>>>>gamma point of
>>>>        
>>>>
>>>>>TiO2(anatase)
>>>>>and i have relaxed the structure enough (i
>>>>>          
>>>>>
>>think
>>    
>>
>>>>)but performing PH.Xafter
>>>>        
>>>>
>>>>>scf, 2of 18 frequencies  are minus!
>>>>>i want to know that what technical problem do
>>>>>          
>>>>>
>>i
>>    
>>
>>>>have? does anyone
>>>>        
>>>>
>>>>>calculate
>>>>>these beore?
>>>>>any suggestion will be appreciated
>>>>>
>>>>>
>>>>>          
>>>>>
>_________________________________________________________________
>  
>
>>>>>Express yourself instantly with MSN Messenger!
>>>>>          
>>>>>
>>>>Download today it's FREE!
>>>>        
>>>>
>http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
>  
>
>>>>>          
>>>>>
>>_______________________________________________
>>    
>>
>>>>>Pw_forum mailing list
>>>>>Pw_forum at pwscf.org
>>>>>
>>>>>          
>>>>>
>>http://www.democritos.it/mailman/listinfo/pw_forum
>>    
>>
>>>__________________________________________________
>>>Do You Yahoo!?
>>>Tired of spam?  Yahoo! Mail has the best spam
>>>      
>>>
>>protection around
>>    
>>
>>>http://mail.yahoo.com
>>>_______________________________________________
>>>Pw_forum mailing list
>>>Pw_forum at pwscf.org
>>>http://www.democritos.it/mailman/listinfo/pw_forum
>>>
>>>      
>>>
>
>
>__________________________________________________
>Do You Yahoo!?
>Tired of spam?  Yahoo! Mail has the best spam protection around 
>http://mail.yahoo.com 
>_______________________________________________
>Pw_forum mailing list
>Pw_forum at pwscf.org
>http://www.democritos.it/mailman/listinfo/pw_forum
>  
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060821/fae74701/attachment.htm 

From lanhaiping at gmail.com  Mon Aug 21 14:43:40 2006
From: lanhaiping at gmail.com (lan haiping)
Date: Mon, 21 Aug 2006 20:43:40 +0800
Subject: [Pw_forum] phonons
In-Reply-To: <6599977F-D6E1-4C8B-8EB8-65B57F0810B2@sissa.it>
References: <BAY124-F26D6306CF7801FFB9DCFA9AA430@phx.gbl>
	 <c92002fa0608190538y24a9f4devbaadcee0ecefd55e@mail.gmail.com>
	 <1156026985.44e7926954412@webmail.usask.ca> <44E79BE0.1080308@mit.edu>
	 <c92002fa0608191815j45d52622xc6c80ff9a10e9068@mail.gmail.com>
	 <6599977F-D6E1-4C8B-8EB8-65B57F0810B2@sissa.it>
Message-ID: <c92002fa0608210543r1721a20ds4a8acc5b68abc1b@mail.gmail.com>

Dear Stefano   :

   Thanks for  your  instructions on  science , usage of  names  and skills
of web communications.
    Though i had learned some  western customs in my English class years
ago, I think i throwed off  most of  them back to teachers. I should
appoligize  for my naive mistake . En,  for Chinese , take my name for
example ,   in most situations,  others would call me ' lan hai-ping' ,
while 'lan' is my family name which means *blue,* and 'hai-ping' is  given
name . i think it is the most difference between  Chinese and  western names
,since our family names are the first names in written  and oral usages .



Regards,

Hai-Ping









On 8/20/06, Stefano Baroni <baroni at sissa.it> wrote:
>
>
>  On Aug 20, 2006, at 3:15 AM, lan haiping wrote:
>
>  Dear  Prof Nicola,
>
>
>
> just a note about the "western" usage of names. (I would be glad to hear
> an explanation on how names are used in Chinese!). "Nicola" is the "given"
> name, i.e. it is the name that parents give to the newborn baby. It is
> also called "first" name because in written it is the first name that must
> be used. The other name "Marzari" is the family name which is transmitted
> from one generation to the next in a patrilinear way (a baby gets the
> fathers' family name by default). It is a form of confidentiality to address
> somebody by his/her first name. It is perfect here because: i) it is common
> usage on the web and; ii) it is common practice among scientists
> (particularly in the US). However, addressing somebody by his/her title is
> _very_ formal. Not very appropriate in this forum, but above all kind of
> funny if accompanied by the much more confidential usage of the first name.
> Resuming: "Dear Nicola" is confidential and "Dear Prof. Marzari" is formal -
> both ar OK, the first more appropriate to this forum. "Dear Prof Nicola" is
> just funny! ... I was not joking saying that I am curious about _your_ usage
> of "given", "family", or whatever names. If you have a  few spare minutes,
> just let me know ...
>
>
> Science, now!
>
>
> Thanks for your long suggestion.
>
> I may not follow some of your words, would you please clarify  ?
>  1,  How to relax the ions more ?
>
>
>
>
> The code must have some default threshold for the atomic forces (and,
> maybe, stress too if you also relax the unit cell) such that  when forces
> are smaller, it stops. The threshold is usually more than adequate for most
> practical purposes, _but to learn_ What I mean is that it is a good idea to
> practice and to see _with one's own eyes_ how a given property (phonon
> frequencies, in the present case) depend on external parameter and/or
> convergence criteria. As it is possible to vary by hands the values of the
> convergence thresholds, it is a good idea to try a same calculation with
> different values of the relevant thresholds and to see how the frequencies
> depend on them.  I have a comment, though. As Nicola said in a previous
> post, it is unlikely (though not impossible) that phonons go soft because
> the structure is poorely optimized. It is worth stopping a while and
> understand carefully why it is both unlikely _and_ possible. It is possible
> because squared phonon frequencies are related to the Hessian (matrix of
> second derivatives) of the energy with respect to the nuclear coordinates.
> It is perfectly possible that, out of equilibrium, one or more of the
> eigenvalues of this matrix are negative. I even believe (but would like not
> to comment on this statement) that the ratio of the configuration space
> corresponding to all positive eigenvalues to the total configuration space
> decreases exponentially with the number of atoms. HOWEVER, phonon
> frequencies only depend on second derivatives of the energy, and not
> directly on first derivatives. So, if you are not too far from the minimum,
> not too good a convergence (i.e. the existence of nonvanishing forces) may
> certainly affect the accuracy of phonon calculations (it would not if the
> crystal were perfectly harmonic), but not the stability of the system (i.e.
> the sign of any of the squared frequencies).
>
>      2, I knew from most refs that ecutwfs  30 Ry for ultrasoft-PP is
> enough in pw calculations , Is it right ?
>
>
>
>
> This statement is meaningless. 30 Ry may be too large or too small a
> cutoff, depending on the PP that you are using. It is true that many US PP
> calculations are done with cutoffs in this range, but whether or not this
> value is appropriate in your case ONLY YOU CAN TELL. i.e. you have to
> perform careful convergence tests in order to be sure that your PW basis set
> is adequate to describe the properties that you want with the accuracy that
> you need (the same holds, of course, for many other devil's details of the
> calculation: k-points, gaussian smearing, etc.)
>
>  But i donot have much idear about ecutwfs in ph calculations .
>
>
>
>
> By and large, (squared) phonon frequencies are second derivatives of the
> crystal energy for a given kinetic-energy cutoff. If a cutoff is OK for
> geometry optimization, it  is also so for phonon calculation (see below)
>
>
> Would you please give me some refs of more detail settings about PH
> calculations ?
>
>
>
> S. Baroni, A. Dal Corso, S. de Gironcoli, and P. Giannozzi, Rev. Mod.
> Phys. 73, 515 (2001).
>
>
> Enjoy!
>
>
> Stefano B.
>
>
> PS: for us, it is considered polite to sign with our first name, or with
> the first name followed by the family name. It would be unpolite (except for
> formal business correspondence) to prepend the title to the signature -
> "Stefano" or "Stefano B.": confidential, OK; "Stefano Baroni" a bit less
> confidential, also OK in most cases; "Prof. Stefano Baroni": very formal and
> kind of assuming and not-so-polite in many cases - OK in a few
>
>
>
> I am trying to find more problems related to this wrong result following
> your suggestion.
> Thank you again, it really helps me much.
>
> p.s.  It is nice for me listening  your lectures in Beijing , i really
> learn much from your 3 lectures .
>
>
> Regards,
>
> Hai-Ping
>
>
> On 8/20/06, Nicola Marzari <marzari at mit.edu> wrote:
> >
> >
> >
> >
> > Dear Haiping,
> >
> >
> > there are two possibilities that lead to negative phonon frequencies at
> > gamma.
> >
> > The first one is less likely:
> >
> > 1) you have not carefully relaxed the *internal* degrees of freedom -
> > i.e. the ionic coodinates. Then, the phonon code discovers that moving
> > the atoms along a certain pattern (an "eigenvector" of the dynamical
> > matrix at gamma) the energy goes down. Perfect - just relax the ions
> > more. Relaxation of the unit cell has nothing to do with this
> >
> > The second one is more likely:
> >
> > 2) the calculation is not converged with respect
> > to the plane wave cutoff needed, and the number of k-points needed,
> > for the small or zero frequencies at gamma. Obtaining an accurate,
> > faithful description of low energy modes requires cutoffs and especially
> > duals much higher than anything you are used to (to give you an example,
> > for an ultrasoft C for diamond, you need 100 Ry in the wavefunction and
> > 2800 Ry in the charge density to get a convergence, at gamma,
> > of the zero modes down to the last cm-1). Does it mean you need
> > to use 100/2800 Ry for diamond ? No ! By enforcing the translation sum
> > rules that the interatomic force constants must satisfy, you can monitor
> > (after enforcing said rules) the convergence of the low frequency modes,
> >
> > and discover you can use much more reasonable cutoffs.
> >
> > Still, even more than a regular calculations, phonons require a
> > through control on the parameters that define your calculation.
> >
> > Note that, as a warm-up, it is easier to converge LDA calculations
> > than GGA calculations (the exchange-correlation terms, calculated in
> > real space, introduce very high frequencies in reciprocal space).
> >
> >                        nicola
> >
> >
> > >> Dear All,
> > >> I also have calculated gamma phonon frequencies of anatase TiO2 .
> > >> I run ph.x caculation after relaxing the structure, but  9 of 18
> > frequencies
> > >> are minus.
> > >>
> > >> Would you please give me some hints and help?
> > >>
> > >> input settings are below, thank you in advance !
> > >>
> > >> *phonons of TiO2 at Gamma
> > >>  &inputph
> > >>   tr2_ph=1.0d-14,
> > >>   recover=.true.,
> > >>   epsil=.true.,
> > >>   prefix='tio2',
> > >>   fildyn='TiO2-gamma.dyn ',
> > >>   outdir='/home/haiping/tmp/'
> > >>  /
> > >> 0.0 0.0 0.0*
> >
> >
> > ---------------------------------------------------------------------
> > Prof Nicola Marzari   Department of Materials Science and Engineering
> > 13-5066   MIT   77 Massachusetts Avenue   Cambridge MA 02139-4307 USA
> > tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu
> > _______________________________________________
> > Pw_forum mailing list
> > Pw_forum at pwscf.org
> > http://www.democritos.it/mailman/listinfo/pw_forum
> >
>
>
>  ---
> Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center - Trieste
> [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)
>
> Please, if possible, don't  send me MS Word or PowerPoint attachments
> Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html
>
>
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060821/8509ecda/attachment.htm 

From lanhaiping at gmail.com  Mon Aug 21 14:48:46 2006
From: lanhaiping at gmail.com (lan haiping)
Date: Mon, 21 Aug 2006 20:48:46 +0800
Subject: [Pw_forum] phonon at Gamma point
In-Reply-To: <20060820145956.23780.qmail@web60312.mail.yahoo.com>
References: <c92002fa0608191743h458f45f5ubf539f93e99ed772@mail.gmail.com>
	 <20060820145956.23780.qmail@web60312.mail.yahoo.com>
Message-ID: <c92002fa0608210548tce33296jbbca656e666a23e9@mail.gmail.com>

Thank you  , Eyvaz.
 I still have a confusion : why does the modified ecutrho  affect the
system's symmetry ?

Regards

Hai-Ping

On 8/20/06, Eyvaz Isaev <eyvaz_isaev at yahoo.com> wrote:
>
> OK, let me hint more.
>
> Anatase has a Pearson symbol tI12 or space group
> I4_1/amd. So, ibrav should be 7, but you have chosen a
> monoclinic lattice, i.e. ibrav=14. The space group has
> 16 symmetry elements, but your choice after correcting
> ecutrho gives only 4 symmetry elements .
>
> Imagine  FCC lattice and its conventional basis
> vectors now. According to you now we have to put
> ibrav=4 or 5. But ibrav=2 for FCC lattice.
>
> In addition, I also checked your input file via
> XCrysDen before I replied previously. Yes, there is no
> problem with it, but a figure does not look like
> anatase though  you used  nonzero ibrav. XcrysDen
> usually replicates atomic positions  if  ibrav.neq.0.
> For example, if you put ibrav=0  for  FCC cell and
> specify conventional CELL_PARAMETERS you will find  a
> rhombohedron, but using ibrav=2 and  default basis
> vectors you find a cubic cell.
>
> Bests,
> Eyvaz.
>
>
> --- lan haiping <lanhaiping at gmail.com> wrote:
>
> > Dear Eyvaz,
> > Thank you.
> > I checked the strcuture again using XCRYSDEN,
> > and found no problems related to the  lattices.
> >
> > Regards,
> >
> > Hai-Ping
> >
> >
> >
> > On 8/20/06, Eyvaz Isaev <eyvaz_isaev at yahoo.com>
> > wrote:
> > >
> > > Hi,
> > >
> > > It seems your lattice choice is not correct.
> > Please
> > > have a look at
> > > http://cst-www.nrl.navy.mil/lattice/struk/c5.html
> > >
> > > Bests,
> > > Eyvaz.
> > >
> > > --- lan haiping <lanhaiping at gmail.com> wrote:
> > >
> > > > Dear All,
> > > > I also have calculated gamma phonon frequencies
> > of
> > > > anatase TiO2 .
> > > > I run ph.x caculation after relaxing the
> > structure,
> > > > but  9 of 18 frequencies
> > > > are minus.
> > > >
> > > > Would you please give me some hints and help?
> > > >
> > > > input settings are below, thank you in advance !
> > > >
> > > > *phonons of TiO2 at Gamma
> > > >  &inputph
> > > >   tr2_ph=1.0d-14,
> > > >   recover=.true.,
> > > >   epsil=.true.,
> > > >   prefix='tio2',
> > > >   fildyn='TiO2-gamma.dyn',
> > > >   outdir='/home/haiping/tmp/'
> > > >  /
> > > > 0.0 0.0 0.0*
> > > > **
> > > > *Relaxing calculation input :*
> > > > *&CONTROL
> > > >              title = 'TiO2' ,
> > > >        calculation = 'relax',
> > > >          verbosity = 'default',
> > > >            tprnfor = .true.,
> > > >             outdir = '/home/haiping/tmp/',
> > > >             prefix = 'tio2',
> > > >            disk_io = 'default',
> > > >              nstep = 200,
> > > >         pseudo_dir =
> > > > '/home/haiping/espresso/pseudo/',
> > > > /
> > > > &SYSTEM
> > > >                 ibrav= 14,
> > > >
> > > > celldm(1)=10.2856,celldm(2)=1.0,celldm(3)=1.0,
> > > >
> > > > celldm(4)=-0.759357,celldm(5)=-0.759357,
> > > >                celldm(6)=0.518714
> > > >                nat = 6 ,
> > > >               ntyp = 2 ,
> > > >               nbnd = 30 ,
> > > >            ecutwfc = 30.0000000000,
> > > >            ecutrho = 160.000,
> > > >        occupations = 'fixed' ,
> > > > /
> > > >
> > > >
> > > > &ELECTRONS
> > > >   electron_maxstep = 100,
> > > >           conv_thr = 1.0e-07,
> > > >        mixing_mode = 'plain' ,
> > > >        startingpot = 'atomic' ,
> > > >        startingwfc = 'atomic' ,
> > > >        mixing_beta = 0.300000000,
> > > >        mixing_ndim = 8,
> > > >    diagonalization = 'david_overlap' ,
> > > >   diago_david_ndim = 8,
> > > > /
> > > >
> > > > &IONS
> > > >              ion_dynamics = 'bfgs',
> > > >              pot_extrapolation = 'first_order',
> > > >              wfc_extrapolation = 'first_order',
> > > >
> > > > /
> > > >
> > > > ATOMIC_SPECIES
> > > > O  15.999 O.pw91-van_ak.UPF
> > > > Ti 47.867  Ti.pw91-nsp-van.UPF
> > > > ATOMIC_POSITIONS crystal
> > > > Ti    0.00000   0.00000   0.00000
> > > > Ti    0.75000   0.25000   0.50000
> > > > O     0.20800   0.20800  -0.00000
> > > > O     0.95800   0.45800   0.50000
> > > > O     0.54200   0.04200   0.50000
> > > > O     0.79200   0.79200  -0.00000
> > > > K_POINTS   automatic
> > > > 6 6 7   0 0 0
> > > >
> > > > *
> > > >
> > > >
> > > > On 8/19/06, Ezad Shojaee
> > <ezadshojaee at hotmail.com>
> > > > wrote:
> > > > >
> > > > > hi
> > > > > i am trying to calculate phonon frequencies at
> > > > gamma point of
> > > > > TiO2(anatase)
> > > > > and i have relaxed the structure enough (i
> > think
> > > > )but performing PH.Xafter
> > > > > scf, 2of 18 frequencies  are minus!
> > > > > i want to know that what technical problem do
> > i
> > > > have? does anyone
> > > > > calculate
> > > > > these beore?
> > > > > any suggestion will be appreciated
> > > > >
> > > > >
> > > >
> > >
>
> --- lan haiping <lanhaiping at gmail.com> wrote:
>
> > Dear Eyvaz,
> > Thank you.
> > I checked the strcuture again using XCRYSDEN,
> > and found no problems related to the  lattices.
> >
> > Regards,
> >
> > Hai-Ping
> >
> >
> >
> > On 8/20/06, Eyvaz Isaev <eyvaz_isaev at yahoo.com>
> > wrote:
> > >
> > > Hi,
> > >
> > > It seems your lattice choice is not correct.
> > Please
> > > have a look at
> > > http://cst-www.nrl.navy.mil/lattice/struk/c5.html
> > >
> > > Bests,
> > > Eyvaz.
> > >
> > > --- lan haiping <lanhaiping at gmail.com> wrote:
> > >
> > > > Dear All,
> > > > I also have calculated gamma phonon frequencies
> > of
> > > > anatase TiO2 .
> > > > I run ph.x caculation after relaxing the
> > structure,
> > > > but  9 of 18 frequencies
> > > > are minus.
> > > >
> > > > Would you please give me some hints and help?
> > > >
> > > > input settings are below, thank you in advance !
> > > >
> > > > *phonons of TiO2 at Gamma
> > > >  &inputph
> > > >   tr2_ph=1.0d-14,
> > > >   recover=.true.,
> > > >   epsil=.true.,
> > > >   prefix='tio2',
> > > >   fildyn='TiO2-gamma.dyn',
> > > >   outdir='/home/haiping/tmp/'
> > > >  /
> > > > 0.0 0.0 0.0*
> > > > **
> > > > *Relaxing calculation input :*
> > > > *&CONTROL
> > > >              title = 'TiO2' ,
> > > >        calculation = 'relax',
> > > >          verbosity = 'default',
> > > >            tprnfor = .true.,
> > > >             outdir = '/home/haiping/tmp/',
> > > >             prefix = 'tio2',
> > > >            disk_io = 'default',
> > > >              nstep = 200,
> > > >         pseudo_dir =
> > > > '/home/haiping/espresso/pseudo/',
> > > > /
> > > > &SYSTEM
> > > >                 ibrav= 14,
> > > >
> > > > celldm(1)=10.2856,celldm(2)=1.0,celldm(3)=1.0,
> > > >
> > > > celldm(4)=-0.759357,celldm(5)=-0.759357,
> > > >                celldm(6)=0.518714
> > > >                nat = 6 ,
> > > >               ntyp = 2 ,
> > > >               nbnd = 30 ,
> > > >            ecutwfc = 30.0000000000,
> > > >            ecutrho = 160.000,
> > > >        occupations = 'fixed' ,
> > > > /
> > > >
> > > >
> > > > &ELECTRONS
> > > >   electron_maxstep = 100,
> > > >           conv_thr = 1.0e-07,
> > > >        mixing_mode = 'plain' ,
> > > >        startingpot = 'atomic' ,
> > > >        startingwfc = 'atomic' ,
> > > >        mixing_beta = 0.300000000,
> > > >        mixing_ndim = 8,
> > > >    diagonalization = 'david_overlap' ,
> > > >   diago_david_ndim = 8,
> > > > /
> > > >
> > > > &IONS
> > > >              ion_dynamics = 'bfgs',
> > > >              pot_extrapolation = 'first_order',
> > > >              wfc_extrapolation = 'first_order',
> > > >
> > > > /
> > > >
> > > > ATOMIC_SPECIES
> > > > O  15.999 O.pw91-van_ak.UPF
> > > > Ti 47.867  Ti.pw91-nsp-van.UPF
> > > > ATOMIC_POSITIONS crystal
> > > > Ti    0.00000   0.00000   0.00000
> > > > Ti    0.75000   0.25000   0.50000
> > > > O     0.20800   0.20800  -0.00000
> > > > O     0.95800   0.45800   0.50000
> > > > O     0.54200   0.04200   0.50000
> > > > O     0.79200   0.79200  -0.00000
> > > > K_POINTS   automatic
> > > > 6 6 7   0 0 0
> > > >
> > > > *
> > > >
> > > >
> > > > On 8/19/06, Ezad Shojaee
> > <ezadshojaee at hotmail.com>
> > > > wrote:
> > > > >
> > > > > hi
> > > > > i am trying to calculate phonon frequencies at
> > > > gamma point of
> > > > > TiO2(anatase)
> > > > > and i have relaxed the structure enough (i
> > think
> > > > )but performing PH.Xafter
> > > > > scf, 2of 18 frequencies  are minus!
> > > > > i want to know that what technical problem do
> > i
> > > > have? does anyone
> > > > > calculate
> > > > > these beore?
> > > > > any suggestion will be appreciated
> > > > >
> > > > >
> > > >
> > >
> >
> _________________________________________________________________
> > > > > Express yourself instantly with MSN Messenger!
> > > > Download today it's FREE!
> > > > >
> > > >
> > >
> >
> http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
> > > > >
> > > > >
> > _______________________________________________
> > > > > Pw_forum mailing list
> > > > > Pw_forum at pwscf.org
> > > > >
> > http://www.democritos.it/mailman/listinfo/pw_forum
> > > > >
> > > >
> > >
> > >
> > > __________________________________________________
> > > Do You Yahoo!?
> > > Tired of spam?  Yahoo! Mail has the best spam
> > protection around
> > > http://mail.yahoo.com
> > > _______________________________________________
> > > Pw_forum mailing list
> > > Pw_forum at pwscf.org
> > > http://www.democritos.it/mailman/listinfo/pw_forum
> > >
> >
>
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060821/eae516a7/attachment.htm 

From lanhaiping at gmail.com  Mon Aug 21 14:52:50 2006
From: lanhaiping at gmail.com (lan haiping)
Date: Mon, 21 Aug 2006 20:52:50 +0800
Subject: [Pw_forum] phonon at Gamma point
In-Reply-To: <44E9A36F.7000003@netscape.net>
References: <20060820145956.23780.qmail@web60312.mail.yahoo.com>
	 <44E9A36F.7000003@netscape.net>
Message-ID: <c92002fa0608210552g345e2b4fqa32a4b03da29c90@mail.gmail.com>

Thank you , Luke.
I am  calculating phonons with this settings.

Regards,
Hai-Ping


On 8/21/06, Luke Thulin <lukethulin at netscape.net> wrote:
>
>  I believe Eyvaz is correct, for anatase use:
>
> ibrav = 7
> A choice of cartesian coordinates for measured values:
> Ti (0,0,0)
> Ti (0,a/2,c/4)
> O (0,0,-uc)
> O (0,0,+uc)
> O (0,a/2,c/4-uc)
> O (0,a/2,c/4+uc)
>
> where a = 3.785, c=9.514, u=.2081
>
>
> Luke
>
> eyvaz_isaev at yahoo.com wrote:
>
> OK, let me hint more.
>
> Anatase has a Pearson symbol tI12 or space group
> I4_1/amd. So, ibrav should be 7, but you have chosen a
> monoclinic lattice, i.e. ibrav=14. The space group has
> 16 symmetry elements, but your choice after correcting
> ecutrho gives only 4 symmetry elements .
>
> Imagine  FCC lattice and its conventional basis
> vectors now. According to you now we have to put
> ibrav=4 or 5. But ibrav=2 for FCC lattice.
>
> In addition, I also checked your input file via
> XCrysDen before I replied previously. Yes, there is no
> problem with it, but a figure does not look like
> anatase though  you used  nonzero ibrav. XcrysDen
> usually replicates atomic positions  if  ibrav.neq.0.
>  For example, if you put ibrav=0  for  FCC cell and
> specify conventional CELL_PARAMETERS you will find  a
> rhombohedron, but using ibrav=2 and  default basis
> vectors you find a cubic cell.
>
> Bests,
> Eyvaz.
>
>
> --- lan haiping <lanhaiping at gmail.com> <lanhaiping at gmail.com> wrote:
>
>
>
> Dear Eyvaz,
> Thank you.
> I checked the strcuture again using XCRYSDEN,
> and found no problems related to the  lattices.
>
> Regards,
>
> Hai-Ping
>
>
>
> On 8/20/06, Eyvaz Isaev <eyvaz_isaev at yahoo.com> <eyvaz_isaev at yahoo.com>
> wrote:
>
>
> Hi,
>
> It seems your lattice choice is not correct.
>
>
> Please
>
>
> have a look athttp://cst-www.nrl.navy.mil/lattice/struk/c5.html
>
> Bests,
> Eyvaz.
>
> --- lan haiping <lanhaiping at gmail.com> <lanhaiping at gmail.com> wrote:
>
>
>
> Dear All,
> I also have calculated gamma phonon frequencies
>
>
> of
>
>
>  anatase TiO2 .
> I run ph.x caculation after relaxing the
>
>
> structure,
>
>
>  but  9 of 18 frequencies
> are minus.
>
> Would you please give me some hints and help?
>
> input settings are below, thank you in advance !
>
> *phonons of TiO2 at Gamma
>  &inputph
>   tr2_ph=1.0d-14,
>   recover=.true.,
>   epsil=.true.,
>   prefix='tio2',
>   fildyn='TiO2-gamma.dyn',
>   outdir='/home/haiping/tmp/'
>  /0.0 0.0 0.0*
> **
> *Relaxing calculation input :*
> *&CONTROL
>              title = 'TiO2' ,
>        calculation = 'relax',
>          verbosity = 'default',
>            tprnfor = .true.,
>             outdir = '/home/haiping/tmp/',
>             prefix = 'tio2',
>            disk_io = 'default',
>              nstep = 200,
>         pseudo_dir =
> '/home/haiping/espresso/pseudo/',
> /
> &SYSTEM
>                 ibrav= 14,
>
> celldm(1)=10.2856,celldm(2)=1.0,celldm(3)=1.0,
>
> celldm(4)=-0.759357,celldm(5)=-0.759357,
>                celldm(6)=0.518714
>                nat = 6 ,
>               ntyp = 2 ,
>               nbnd = 30 ,
>            ecutwfc = 30.0000000000,
>            ecutrho = 160.000,
>        occupations = 'fixed' ,
> /
>
>
> &ELECTRONS
>   electron_maxstep = 100,
>           conv_thr = 1.0e-07,
>        mixing_mode = 'plain' ,
>        startingpot = 'atomic' ,
>        startingwfc = 'atomic' ,
>        mixing_beta = 0.300000000,
>        mixing_ndim = 8,
>    diagonalization = 'david_overlap' ,
>   diago_david_ndim = 8,
> /
>
> &IONS
>              ion_dynamics = 'bfgs',
>              pot_extrapolation = 'first_order',
>              wfc_extrapolation = 'first_order',
>
> /
>
> ATOMIC_SPECIES
> O  15.999 O.pw91-van_ak.UPF
> Ti 47.867  Ti.pw91-nsp-van.UPF
> ATOMIC_POSITIONS crystal
> Ti    0.00000   0.00000   0.00000
> Ti    0.75000   0.25000   0.50000
> O     0.20800   0.20800  -0.00000
> O     0.95800   0.45800   0.50000
> O     0.54200   0.04200   0.50000
> O     0.79200   0.79200  -0.00000
> K_POINTS   automatic
> 6 6 7   0 0 0
>
> *
>
>
> On 8/19/06, Ezad Shojaee
>
>
> <ezadshojaee at hotmail.com> <ezadshojaee at hotmail.com>
>
>  wrote:
>
>
> hi
> i am trying to calculate phonon frequencies at
>
>
> gamma point of
>
>
> TiO2(anatase)
> and i have relaxed the structure enough (i
>
>
> think
>
>
>  )but performing PH.Xafter
>
> scf, 2of 18 frequencies  are minus!
> i want to know that what technical problem do
>
>
> i
>
>
>  have? does anyone
>
>
> calculate
> these beore?
> any suggestion will be appreciated
>
>
>
>
>
> --- lan haiping <lanhaiping at gmail.com> <lanhaiping at gmail.com> wrote:
>
>
>
> Dear Eyvaz,
> Thank you.
> I checked the strcuture again using XCRYSDEN,
> and found no problems related to the  lattices.
>
> Regards,
>
> Hai-Ping
>
>
>
> On 8/20/06, Eyvaz Isaev <eyvaz_isaev at yahoo.com> <eyvaz_isaev at yahoo.com>
> wrote:
>
>
> Hi,
>
> It seems your lattice choice is not correct.
>
>
> Please
>
>
> have a look athttp://cst-www.nrl.navy.mil/lattice/struk/c5.html
>
> Bests,
> Eyvaz.
>
> --- lan haiping <lanhaiping at gmail.com> <lanhaiping at gmail.com> wrote:
>
>
>
> Dear All,
> I also have calculated gamma phonon frequencies
>
>
> of
>
>
>  anatase TiO2 .
> I run ph.x caculation after relaxing the
>
>
> structure,
>
>
>  but  9 of 18 frequencies
> are minus.
>
> Would you please give me some hints and help?
>
> input settings are below, thank you in advance !
>
> *phonons of TiO2 at Gamma
>  &inputph
>   tr2_ph=1.0d-14,
>   recover=.true.,
>   epsil=.true.,
>   prefix='tio2',
>   fildyn='TiO2-gamma.dyn',
>   outdir='/home/haiping/tmp/'
>  /0.0 0.0 0.0*
> **
> *Relaxing calculation input :*
> *&CONTROL
>              title = 'TiO2' ,
>        calculation = 'relax',
>          verbosity = 'default',
>            tprnfor = .true.,
>             outdir = '/home/haiping/tmp/',
>             prefix = 'tio2',
>            disk_io = 'default',
>              nstep = 200,
>         pseudo_dir =
> '/home/haiping/espresso/pseudo/',
> /
> &SYSTEM
>                 ibrav= 14,
>
> celldm(1)=10.2856,celldm(2)=1.0,celldm(3)=1.0,
>
> celldm(4)=-0.759357,celldm(5)=-0.759357,
>                celldm(6)=0.518714
>                nat = 6 ,
>               ntyp = 2 ,
>               nbnd = 30 ,
>            ecutwfc = 30.0000000000,
>            ecutrho = 160.000,
>        occupations = 'fixed' ,
> /
>
>
> &ELECTRONS
>   electron_maxstep = 100,
>           conv_thr = 1.0e-07,
>        mixing_mode = 'plain' ,
>        startingpot = 'atomic' ,
>        startingwfc = 'atomic' ,
>        mixing_beta = 0.300000000,
>        mixing_ndim = 8,
>    diagonalization = 'david_overlap' ,
>   diago_david_ndim = 8,
> /
>
> &IONS
>              ion_dynamics = 'bfgs',
>              pot_extrapolation = 'first_order',
>              wfc_extrapolation = 'first_order',
>
> /
>
> ATOMIC_SPECIES
> O  15.999 O.pw91-van_ak.UPF
> Ti 47.867  Ti.pw91-nsp-van.UPF
> ATOMIC_POSITIONS crystal
> Ti    0.00000   0.00000   0.00000
> Ti    0.75000   0.25000   0.50000
> O     0.20800   0.20800  -0.00000
> O     0.95800   0.45800   0.50000
> O     0.54200   0.04200   0.50000
> O     0.79200   0.79200  -0.00000
> K_POINTS   automatic
> 6 6 7   0 0 0
>
> *
>
>
> On 8/19/06, Ezad Shojaee
>
>
> <ezadshojaee at hotmail.com> <ezadshojaee at hotmail.com>
>
>  wrote:
>
>
> hi
> i am trying to calculate phonon frequencies at
>
>
> gamma point of
>
>
> TiO2(anatase)
> and i have relaxed the structure enough (i
>
>
> think
>
>
>  )but performing PH.Xafter
>
> scf, 2of 18 frequencies  are minus!
> i want to know that what technical problem do
>
>
> i
>
>
>  have? does anyone
>
>
> calculate
> these beore?
> any suggestion will be appreciated
>
>
>
>
> _________________________________________________________________
>
>
>   Express yourself instantly with MSN Messenger!
>
>
> Download today it's FREE!
>
>
> http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
>
>             _______________________________________________
>
>
>  Pw_forum mailing listPw_forum at pwscf.org
>
>           http://www.democritos.it/mailman/listinfo/pw_forum
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam
>
>
> protection around
>
>
> http://mail.yahoo.com
> _______________________________________________
> Pw_forum mailing listPw_forum at pwscf.orghttp://www.democritos.it/mailman/listinfo/pw_forum
>
>       __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around http://mail.yahoo.com
> _______________________________________________
> Pw_forum mailing listPw_forum at pwscf.orghttp://www.democritos.it/mailman/listinfo/pw_forum
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060821/d996f3a3/attachment.htm 

From eyvaz_isaev at yahoo.com  Mon Aug 21 15:36:28 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Mon, 21 Aug 2006 06:36:28 -0700 (PDT)
Subject: [Pw_forum] phonon at Gamma point
In-Reply-To: <c92002fa0608210548tce33296jbbca656e666a23e9@mail.gmail.com>
Message-ID: <20060821133628.31977.qmail@web60323.mail.yahoo.com>

Hi,


From emenendez at macul.ciencias.uchile.cl  Mon Aug 21 23:34:48 2006
From: emenendez at macul.ciencias.uchile.cl (Eduardo Ariel Menendez P)
Date: Mon, 21 Aug 2006 17:34:48 -0400 (CLT)
Subject: [Pw_forum] Re: doing QHA calculations (contd)
Message-ID: <Pine.LNX.4.58.0608211720120.10830@macul.ciencias.uchile.cl>

Hi Tuhina,
Maybe you can find useful this personal modification of fqha.f90. The
program is small, and it is easy to modify (if you know fortran).

Comments:
I think the format of the phonon DOS file is not exactly what
is read by fqha.x (see the commented line "!  read (1,*) ndiv, emax, de".
There is no first line like "ndiv, emax, de" in the DOS file).

Moreover, the first frequency (read as nu(1)) is negative in the few DOS
files that I have generated, then the logarithm log(1.0d0-exp(-a3*nu(i)/T)
is NaN (as Eyvaz pointed). I just jumped and began the loop by
i=2, and I hope it will not harm, as DOS(0)=0.

Regards
Eduardo

!
! Copyright (C) 2001-2003 PWSCF group
! This file is distributed under the terms of the
! GNU General Public License. See the file `License'
! in the root directory of the present distribution,
! or http://www.gnu.org/copyleft/gpl.txt .
!
! Modified by E Menendez, 21 Ago 2006
!     Calculate Free Energy F
!     Given phonon DOS, calculate F at various temperatures
!
program fqha
  !
  implicit none
  integer, parameter:: ndivx=10000
  real(8) :: dos(ndivx),nu(ndivx),T,a1,a2,a3,Ftot,norm,F0
  real(8) :: de, emax
  integer :: i,ndiv
  character(len=256) :: filename
  !
  a1=0.5d0/13.6058d0/8065.5d0
  a2=8.617d-5/13.6058d0   ! Boltzmann constant kB in Ry/K
  a3=1.0d0/8065.5d0/8.617d-5 ! Boltzmann constant kB in cm^-1/K
  !
  write (*,'(a,$)') ' file containing the dos >>>'
  read(*,'(a)') filename
  open(unit=1,file=filename,status='old')
!  read (1,*) ndiv, emax, de
  ndiv=0
  i=1
120 continue
   read(1,*,err=110,end=110) nu(i),dos(i)
!     ndiv=ndiv+1
!     write(*,*) nu(i),dos(i),i
     i=i+1
!     if (abs(nu(i)-de*(i-0.5d0)).gt.1.0d-6) stop ' wrong grid'
  goto 120
110 continue
  close(1)
  ndiv=i-1
  if (ndiv.gt.ndivx) stop ' ndivx too small'
  write (*,*) 'ndiv=',ndiv

  emax=nu(ndiv)
  de=(emax-nu(1))/(ndiv-1)

  write (*,'(a,$)') ' file for the Free energy >>>'
  read(*,'(a)') filename
  open(unit=1,file=filename,status='new')
  !
1 continue
  write (*,'(a,$)') ' temperature >>>'
  read (*,*,err=10,end=10) T
  Ftot=0.0d0
  F0=0.0d0
  norm=0.d0
  do i=2,ndiv
!       I found nu(1) in the dos file  was negative,then I changed
!       the initial  'i' from 1 to 2.
     F0=F0+dos(i)*a1*nu(i)
     if (T.gt.0.d0) Ftot=Ftot+dos(i)*a2*T*log(1.0d0-exp(-a3*nu(i)/T))
     norm=norm+dos(i)
  enddo
  Ftot=(F0+Ftot)*de
  norm=norm*de
  write(1,*) T,Ftot
  write(*,*) T,Ftot
  write(*,*) norm,F0
  !
  go to 1
10 close(1)
  !
  stop
end program fqha
!


From kchan at civet.berkeley.edu  Tue Aug 22 02:00:58 2006
From: kchan at civet.berkeley.edu (Kevin Chan)
Date: Mon, 21 Aug 2006 17:00:58 -0700 (PDT)
Subject: [Pw_forum] example25 and LDA+U forces
Message-ID: <Pine.LNX.4.63.0608211631570.22989@civet.berkeley.edu>

Dear Stefano,

This is a long-delayed reply to your message on 21 Jul 2006.  Thanks very 
much for your explanation.  I updated to espresso-3.1.1 and got the same 
results as yours and the reference files for example25.  So things make 
sense.

I was previously using espresso-3.1, for which my results did match the 
reference files in that version.  But it looks like the 
starting_ns_eigenvalue was set differently from version 3.1.1:
3,1
<     starting_ns_eigenvalue(5,2,2) = 1.d0
<     starting_ns_eigenvalue(5,1,3) = 1.d0
3.1.1
>     starting_ns_eigenvalue(3,2,2) = 1.d0
>     starting_ns_eigenvalue(3,1,3) = 1.d0

This gives totally different results, for example in 3.1:
feo_user_ns.out, gamma
spin-up
     -8.7970  -7.5699   1.9772   3.8697   3.8697   5.8105   5.8105   6.4562
      7.7598   7.7779   7.7779   8.5005   8.5005  10.5601  10.5601  11.5452
     12.6406  13.4449  13.4449  19.8736
      Fermi energy 10.5885 ev
      correction for metals     =    -0.00404010 ryd

This was the source of my confusion.

Regards,
Kevin



From eyvaz_isaev at yahoo.com  Tue Aug 22 02:12:06 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Mon, 21 Aug 2006 17:12:06 -0700 (PDT)
Subject: [Pw_forum] charge is wrong
In-Reply-To: <57363.152.14.74.125.1156099173.squirrel@webmail.ncsu.edu>
Message-ID: <20060822001206.58399.qmail@web60316.mail.yahoo.com>

Dear all,

I would like to add more for the question where there
was mentioned a disagreement between integrated and
expected number of electrons in the system with odd
number of electrons. Before I noticed that in this
case one has to specify occupations='smearing' and
relevant 2 keywords to let the program to deal with a
metallic system. 
But, sure, occupations='tetahedra' (without additional
keywords) works fine for both metals and
semiconductors.  

Bests,
Eyvaz.  

--- Zheng Xiaohong <xzheng2 at ncsu.edu> wrote:

> Thank you very much!! My input file is as follows:
> 
>
===========================================================
> &CONTROL
>   calculation  = "relax",
>   prefix       = "Au17",
>   pseudo_dir   =
> "/spin/home/zheng/espresso-3.1.1/pseudo",
>   outdir       = "/tmp/work/zheng/pwscf/tmp",
> /
> &SYSTEM
>   ibrav     = 1,
>   celldm(1) = 60.D0,
>   nat       = 17,
>   ntyp      = 1,
>   ecutwfc   = 24.D0,
>   ecutrho   = 144.D0,
> /
> &ELECTRONS
>   conv_thr    = 1.D-5,
>   mixing_beta = 0.7D0,
> /
> &IONS
>   pot_extrapolation = "second_order",
>   wfc_extrapolation = "second_order",
> /
> ATOMIC_SPECIES
> Au 197.0  Au.pbe-nd-van.UPF
> ATOMIC_POSITIONS {angstrom}
> Au     4.23819    3.15126    3.98519
> Au     2.79854    4.30837    6.75063
> Au     1.54105    3.26386    4.27488
> Au     4.19307    1.13230    6.09571
> Au     0.48371    0.92178    4.84325
> Au     2.91535    0.57625    3.82944
> Au     2.77156    5.50267    4.15187
> Au     5.54670    0.59657    3.84844
> Au     6.66741    1.86328    7.07452
> Au     4.20412    6.53544    6.69729
> ......
> 
> K_POINTS {Gamma}
> 
>
======================================================
> 
> Do you think this is a bug of this program. Why will
> it stop for odd
> number of electrons? This will not happen in some
> other software, e.g.
> WIEN2K or VASP.
> 
> If I do spin-polarized calculation, will it go on?
> 
> Xiaohong
> 
> 
> > On 8/20/06, xzheng2 at ncsu.edu <xzheng2 at ncsu.edu>
> wrote:
> >> Dear all,
> >>
> >> In my calculation,the following error happened:
> >
> > hi,
> >
> > these kind of things usually happen, when you have
> an
> > error in your input (e.g. you didn't do a spin
> polarized calculation
> > with an odd number of electrons). other than that
> there is little
> > else to help you without seeing your input file.
> >
> >>
>
===========================================================
> >>      WARNING: integrated charge=   186.30000000,
> expected=
> >> 187.00000000
> 
> >>
> >> 
>
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
> >>      from electrons : error #         1
> >>      charge is wrong
> >> 
>
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
> >>
> >>      stopping ...
> >>
>
===========================================================
> >>
> >> It stopped here. I am a new user of PWscf. I do
> not know what to do now.
> >> Which parameter should I modify so that the
> calculation will continue,
> >
> >
> > you not only want to get your calculation to
> continue, you want the
> > calculation to produce a _correct result_. note
> that your calculation
> > can finish and still produce a totally bogus
> result. with DFT quite a
> > few things are mathematically correct, so can be
> computed, but still
> > have no physical meaning. so you always have to
> make sure you
> > know what you are doing and double check
> everything.
> >
> > cheers,
> >     axel.
> >
> >> please? Thank you very much!
> >>
> >> _______________________________________________
> >> Pw_forum mailing list
> >> Pw_forum at pwscf.org
> >>
> http://www.democritos.it/mailman/listinfo/pw_forum
> >>
> >>
> >
> >
> > --
> >
>
=======================================================================
> > Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu  
> http://www.cmm.upenn.edu
> >   Center for Molecular Modeling   --   University
> of Pennsylvania
> > Department of Chemistry, 231 S.34th Street,
> Philadelphia, PA 19104-6323
> > tel: 1-215-898-1582,  fax: 1-215-573-6233, 
> office-tel: 1-215-898-5425
> >
>
=======================================================================
> > If you make something idiot-proof, the universe
> creates a better idiot.
> > _______________________________________________
> > Pw_forum mailing list
> > Pw_forum at pwscf.org
> > http://www.democritos.it/mailman/listinfo/pw_forum
> >
> 
> 
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From kchan at civet.berkeley.edu  Tue Aug 22 03:32:46 2006
From: kchan at civet.berkeley.edu (Kevin Chan)
Date: Mon, 21 Aug 2006 18:32:46 -0700 (PDT)
Subject: [Pw_forum] phonon, gk_sort error
Message-ID: <Pine.LNX.4.63.0608211802550.9242@civet.berkeley.edu>

Hello,

I am doing a simple test to calculate Si gamma phonons.

I run
pw.x -npool 4 < si3.scf.in > si3.scf.out

then
ph.x -npool 4 < si3.phG.in > si3.phG.out

which produces the following error:
      from gk_sort : error #         1
      array gk out-of-bounds

The error does not occur when I set -npool 1.

The error occurs for espresso-3.1 and 3.1.1, but not 2.1.5.

My input files are below.  Please let me know if I should provide other 
information.  Thanks.

Regards,
Kevin Chan

----------
si3.scf.in
----------

  &control
     calculation='scf',
     restart_mode='from_scratch',
     prefix='si3'
     pseudo_dir = '/users/harvard/kchan/espresso-3.1.1/pseudo/',
     outdir='/gpfs/kchan/sitest3/'
  /
  &system
     ibrav = 2, celldm(1) =10.20, nat=  2, ntyp= 1,
     ecutwfc = 18.0
  /
  &electrons
     mixing_beta = 0.7
     conv_thr =  1.0d-8
  /
ATOMIC_SPECIES
  Si  28.086  Si.vbc.UPF
ATOMIC_POSITIONS
  Si 0.00 0.00 0.00
  Si 0.25 0.25 0.25
K_POINTS automatic
  4 4 4 0 0 0


----------
si3.phG.in
----------

phonons of Si at Gamma
  &inputph
   tr2_ph=1.0d-14,
   prefix='si3',
   epsil=.true.,
   amass(1)=28.08,
   outdir='/gpfs/kchan/sitest3/',
   fildyn='si3.dynG',
  /
0.0 0.0 0.0



From yuhailin_79 at msn.com  Tue Aug 22 05:12:40 2006
From: yuhailin_79 at msn.com (hailin yu)
Date: Tue, 22 Aug 2006 11:12:40 +0800
Subject: [Pw_forum] help me: how to calculate free energy
Message-ID: <BAY106-F3EE958CD24876DA0ABB5D93460@phx.gbl>

Dear all:
       In recently, I have calculated the phonon dispersion and phonon 
density of states of alpha-Sn using PWSCF. And our results are very 
agreement with previously work(PRB 57,10421(1998), phonon dispersion and 
phonon density of states). But when I calucalte the free energy, our results 
are very different from the previously work, and our results are too large. 
So I want to konw there have any programme to calculate the free energy F(T) 
when attainted the phonon density of states.

Thanks a lot!

yuhailin

_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today it's FREE! 
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/



From yuhailin_79 at msn.com  Tue Aug 22 05:14:18 2006
From: yuhailin_79 at msn.com (hailin yu)
Date: Tue, 22 Aug 2006 11:14:18 +0800
Subject: [Pw_forum] help me: how to calculate free energy
Message-ID: <BAY106-F186996E5B01D5753594C0E93460@phx.gbl>

Dear all:
       In recently, I have calculated the phonon dispersion and phonon 
density of states of alpha-Sn using PWSCF. And our results are very 
agreement with previously work(PRB 57,10421(1998), phonon dispersion and 
phonon density of states). But when I calucalte the free energy, our results 
are very different from the previously work, and our results are too large. 
So I want to konw there have any programme to calculate the free energy F(T) 
when attainted the phonon density of states.

Thanks a lot!

yuhailin

_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today it's FREE! 
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/



From baroni at sissa.it  Tue Aug 22 13:09:32 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Tue, 22 Aug 2006 13:09:32 +0200
Subject: [Pw_forum] help me: how to calculate free energy
In-Reply-To: <BAY106-F186996E5B01D5753594C0E93460@phx.gbl>
References: <BAY106-F186996E5B01D5753594C0E93460@phx.gbl>
Message-ID: <427F666F-AA83-40AE-A0BA-96E413D5858F@sissa.it>


On Aug 22, 2006, at 5:14 AM, hailin yu wrote:

> Dear all:
>       In recently, I have calculated the phonon dispersion and  
> phonon density of states of alpha-Sn using PWSCF. And our results  
> are very agreement with previously work(PRB 57,10421(1998), phonon  
> dispersion and phonon density of states).

very good

> But when I calucalte the free energy, our results are very  
> different from the previously work, and our results are too large.  
> So I want to konw there have any programme to calculate the free  
> energy F(T) when attainted the phonon density of states.

The vibrational free energy can ben calculated from the vibrational  
DOS using Eq. (1) of the above paper and replacing  the sum over  
normal modes with an \int n(\omega) d\omega. This results in a simple  
one-dimensional integral over a finite range. Are you sure you want  
us to send you a 10-lines piece of code doing just this?

If you have already written down your own 10-lines code and still  
find wrong results, you may want to:

1) Check that your 10-lines code give correct results when used to  
integrate functions that you can integrate analitically. Beware that,  
due to Van Hove singularities, the DOS may have to be sampled at a  
rather large number of points in order to obtain sensible numerical  
results.

2) Check the normalization of the DOS. In order for the \int n 
(\omega) d\omega to give the same result as in Eq. (1), The integral  
of the DOS must be equal to the number of normal modes per unit cell,  
i.e. 3N, N being the number of atoms. With this normalization, Eq.  
(1) will give the free energy per unit cell.

Hoper this helps.

Stefano B.


>
> Thanks a lot!
>
> yuhailin
>
> _________________________________________________________________
> Express yourself instantly with MSN Messenger! Download today it's  
> FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060822/ebca7932/attachment.htm 

From hushujun at mail.sdu.edu.cn  Tue Aug 22 14:39:21 2006
From: hushujun at mail.sdu.edu.cn (Shujun Hu)
Date: Tue, 22 Aug 2006 20:39:21 +0800
Subject: [Pw_forum] Code starting trouble on the PC Cluster
Message-ID: <356250361.19202@mail.sdu.edu.cn>

Dear Drs,

When starting the parallel code pw.x on the cluster (Intel Pentium-D and MPICH2),
usually i need run the command several times. Nothing appear on the screen for the
early commands but the last one seems good. I have checked the process queue and
found that lots of process named pw.x under python were sleeping. I donnt know
why.


                                         Shujun Hu








From giannozz at nest.sns.it  Tue Aug 22 14:16:59 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Tue, 22 Aug 2006 14:16:59 +0200
Subject: [Pw_forum] insufficient virtual memory! What to do?
In-Reply-To: <20060818214704.48130.qmail@web60323.mail.yahoo.com>
References: <20060818214704.48130.qmail@web60323.mail.yahoo.com>
Message-ID: <200608221416.59816.giannozz@nest.sns.it>

On Friday 18 August 2006 23:47, Eyvaz Isaev wrote:

> Try reduce number of k-points or ecutrho. 

reducing the number of k-points will not help. Reducing
ecutwfc and/or ecutrho will. Also using the strict minimum
of states (nbnd) will help.

There are a few more options affecting memory usage: 
"diagonalization" ('cg' uses less memory), "startingwfc"
('atomic' may use a lot of memory),  "mixing_ndim" (the
largest, the more memory is used)

Paolo
-- 
Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From giannozz at nest.sns.it  Tue Aug 22 14:33:05 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Tue, 22 Aug 2006 14:33:05 +0200
Subject: [Pw_forum] phonon at Gamma point
In-Reply-To: <1156026985.44e7926954412@webmail.usask.ca>
References: <BAY124-F26D6306CF7801FFB9DCFA9AA430@phx.gbl> <c92002fa0608190538y24a9f4devbaadcee0ecefd55e@mail.gmail.com> <1156026985.44e7926954412@webmail.usask.ca>
Message-ID: <200608221433.05596.giannozz@nest.sns.it>

On Sunday 20 August 2006 00:36, Yansun Yao wrote:

>       Other than the internal coordinates, did you optimize the unit cell?

while a good thing in general, optimization of the unit cell is not needed
to have positive phonon frequencies at q=0, as long as atomic positions
are in equilibrium

Paolo
-- 
Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From giannozz at nest.sns.it  Tue Aug 22 15:26:48 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Tue, 22 Aug 2006 15:26:48 +0200
Subject: [Pw_forum] doing QHA calculations
In-Reply-To: <28022696.1155892892132.JavaMail.root@webmail.ncl.res.in>
References: <28022696.1155892892132.JavaMail.root@webmail.ncl.res.in>
Message-ID: <200608221526.48911.giannozz@nest.sns.it>

On Friday 18 August 2006 11:21, Tuhina Kelkar wrote:

> when I run fqha.x I get the following message everytime "wrong grid" 
> and the program stops

unfortunately fqha.x has not been updated for a long time and it doesn't 
read any longer the output from phonon dos calculation. It is a very simple
piece of code that can be easily fixed ans modified. Notice that the
contribution  of acoustic modes at q=0 must be evaluated analytically, 
not numerically.

A better code can be found in the software used at VLAB, U. Minnesota.
See the tutorial: http://www.vlab.msi.umn.edu/events/tutorials.shtml

Paolo
-- 
Paolo Giannozzi             e-mail:  p.giannozzi at sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From giannozz at nest.sns.it  Tue Aug 22 15:36:31 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Tue, 22 Aug 2006 15:36:31 +0200
Subject: [Pw_forum] Code starting trouble on the PC Cluster
In-Reply-To: <356250361.19202@mail.sdu.edu.cn>
References: <356250361.19202@mail.sdu.edu.cn>
Message-ID: <200608221536.31160.giannozz@nest.sns.it>

On Tuesday 22 August 2006 14:39, Shujun Hu wrote:

> When starting the parallel code pw.x on the cluster (Intel Pentium-D and
> MPICH2), usually i need run the command several times. Nothing appear on
> the screen for the early commands but the last one seems good. I have
> checked the process queue and found that lots of process named pw.x under
> python were sleeping. I don't know why.

nor do I, but it seems to me very unlikely that the reason for such 
behavior is related to the fortran code itself. It looks more like a
configuration problem of your PC cluster

Paolo
-- 
Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From swblelia at ehu.es  Tue Aug 22 16:10:20 2006
From: swblelia at ehu.es (Aritz Leonardo Liceranzu)
Date: Tue, 22 Aug 2006 16:10:20 +0200
Subject: [Pw_forum] about e1,e2 and e3
Message-ID: <200608221610.20924.swblelia@ehu.es>

Hello all:

I am having more difficulties than expected to plot |psi|^2

one question concerning the plotting plane in the PP_INPUT:

e1 and e2 refer to REAL or RECIPROCAL space?

thanks


From eyvaz_isaev at yahoo.com  Tue Aug 22 16:10:36 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Tue, 22 Aug 2006 07:10:36 -0700 (PDT)
Subject: [Pw_forum] doing QHA calculations
In-Reply-To: <200608221526.48911.giannozz@nest.sns.it>
Message-ID: <20060822141036.92498.qmail@web60325.mail.yahoo.com>

Hi Paolo,

Thanks for your additions concerning memory reducing.

What concerns staffs related to the QHA, actually I
have written a set of programs which can calculate the
next properties:
- free energy,
- specific heat,
- entropy
- internal energy
- projected DOS and atom contribution to above
mentioned properties
- Debye temperature
- Gruneisen parameter

They use only the force constants matrix (X.fc file)
and BZ information as input.

Hopefully I will be able to add more, say correction 
bulk modulus, and the linear thermal expansion. 

Bests,
Eyvaz.

--- Paolo Giannozzi <giannozz at nest.sns.it> wrote:

> On Friday 18 August 2006 11:21, Tuhina Kelkar wrote:
> 
> > when I run fqha.x I get the following message
> everytime "wrong grid" 
> > and the program stops
> 
> unfortunately fqha.x has not been updated for a long
> time and it doesn't 
> read any longer the output from phonon dos
> calculation. It is a very simple
> piece of code that can be easily fixed ans modified.
> Notice that the
> contribution  of acoustic modes at q=0 must be
> evaluated analytically, 
> not numerically.
> 
> A better code can be found in the software used at
> VLAB, U. Minnesota.
> See the tutorial:
> http://www.vlab.msi.umn.edu/events/tutorials.shtml
> 
> Paolo
> -- 
> Paolo Giannozzi             e-mail: 
> p.giannozzi at sns.it
> Scuola Normale Superiore    Phone:   +39/050-509876,
> Fax:-563513 
> Piazza dei Cavalieri 7      I-56126 Pisa, Italy
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From yuhailin_79 at msn.com  Tue Aug 22 16:39:32 2006
From: yuhailin_79 at msn.com (hailin yu)
Date: Tue, 22 Aug 2006 22:39:32 +0800
Subject: [Pw_forum] help me: how to calculate free energy
In-Reply-To: <427F666F-AA83-40AE-A0BA-96E413D5858F@sissa.it>
Message-ID: <BAY106-F10444BCD755FF071A5FE2F93460@phx.gbl>

Dear Stefano B:
      Thank you for your advice.
I have writen a programme to compute the free energy according to 
Equ(1)(PRB,57,10421(1998)). And I have checked the normalization of the DOS 
and found the integral of the DOS  equal to 5.932386, agreement with 3*2 (2 
atomic per unit cell, is unit cell or primitive cell?? ). I also have using 
Equ(3)( Chem. Phys. Lett. 417,272-276(2005)) to compute free energy and 
attainted the same results with using Equ(1)(PRB,57,10421(1998)). but our 
results is still different with the previous work (PRB,57,10421(1998)). At 
the range of 0~400 K, our results ( free energy) decrease nearly 6 eV,  
while the previous work is about 0.26 eV. Where are wrong?

Thank you very much!
     yuhailin
the follow is my pragramme
                integer i
	double precision hi,k,T,Z,F,dos
	double precision aSn(197),aSndos(197)

	open(1,file="aSndos.dat")
	open(2,file="aSn-FreeEnergy.dat")

	hi= 1.055  !
	k=1.38	   ! the Boltzman's constant

	do i=1,197
	read(1,*)aSn(i),aSndos(i) ! the unit of omega is CM-1
	dos=dos+aSndos(i)
    enddo
	print *,dos

	do T=1.0,400.0,1.0
	F=0.0
    do i=2,197
	Z=(hi*aSn(i)*0.029979*10.0)/(2*k*T) ! the value of (h*omega)/(2*K*T); 
0.029979 is the coefficient of CM-1 ~THz
    F=F+k*6.24151*(10.0E-5)*T*log(exp(z)-exp(-z))*aSndos(i) ! the sum of 
k*T*ln[2*sinh(h*omega/2*k*T)]
	                                         !the unit of F is eV; 
1J=6.24151*10E18 eV
    enddo
	write(2,*)T,F
	enddo

    close(1)
	close(2)

    end


>From: Stefano Baroni <baroni at sissa.it>
>Reply-To: pw_forum at pwscf.org
>To: pw_forum at pwscf.org
>Subject: Re: [Pw_forum] help me: how to calculate free energy
>Date: Tue, 22 Aug 2006 13:09:32 +0200
>
>
>On Aug 22, 2006, at 5:14 AM, hailin yu wrote:
>
>>Dear all:
>>       In recently, I have calculated the phonon dispersion and  phonon 
>>density of states of alpha-Sn using PWSCF. And our results  are very 
>>agreement with previously work(PRB 57,10421(1998), phonon  dispersion and 
>>phonon density of states).
>
>very good
>
>>But when I calucalte the free energy, our results are very  different from 
>>the previously work, and our results are too large.  So I want to konw 
>>there have any programme to calculate the free  energy F(T) when attainted 
>>the phonon density of states.
>
>The vibrational free energy can ben calculated from the vibrational  DOS 
>using Eq. (1) of the above paper and replacing  the sum over  normal modes 
>with an \int n(\omega) d\omega. This results in a simple  one-dimensional 
>integral over a finite range. Are you sure you want  us to send you a 
>10-lines piece of code doing just this?
>
>If you have already written down your own 10-lines code and still  find 
>wrong results, you may want to:
>
>1) Check that your 10-lines code give correct results when used to  
>integrate functions that you can integrate analitically. Beware that,  due 
>to Van Hove singularities, the DOS may have to be sampled at a  rather 
>large number of points in order to obtain sensible numerical  results.
>
>2) Check the normalization of the DOS. In order for the \int n (\omega) 
>d\omega to give the same result as in Eq. (1), The integral  of the DOS 
>must be equal to the number of normal modes per unit cell,  i.e. 3N, N 
>being the number of atoms. With this normalization, Eq.  (1) will give the 
>free energy per unit cell.
>
>Hoper this helps.
>
>Stefano B.
>
>
>>
>>Thanks a lot!
>>
>>yuhailin
>>
>>_________________________________________________________________
>>Express yourself instantly with MSN Messenger! Download today it's  FREE! 
>>http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
>>
>>_______________________________________________
>>Pw_forum mailing list
>>Pw_forum at pwscf.org
>>http://www.democritos.it/mailman/listinfo/pw_forum
>
>---
>Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  Trieste
>[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)
>
>Please, if possible, don't  send me MS Word or PowerPoint attachments
>Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html
>
>
>

_________________________________________________________________
Check the weather nationwide with MSN Search: Try it now!  
http://search.msn.com/results.aspx?q=weather&FORM=WLMTAG



From giannozz at nest.sns.it  Tue Aug 22 16:49:48 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Tue, 22 Aug 2006 16:49:48 +0200
Subject: [Pw_forum] about e1,e2 and e3
In-Reply-To: <200608221610.20924.swblelia@ehu.es>
References: <200608221610.20924.swblelia@ehu.es>
Message-ID: <200608221649.48962.giannozz@nest.sns.it>

On Tuesday 22 August 2006 16:10, Aritz Leonardo Liceranzu wrote:

> I am having more difficulties than expected to plot |psi|^2
>
> one question concerning the plotting plane in the PP_INPUT:
>
> e1 and e2 refer to REAL or RECIPROCAL space?

real space, in alat=celldm(1) units

Paolo
-- 
Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From giannozz at nest.sns.it  Tue Aug 22 16:59:34 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Tue, 22 Aug 2006 16:59:34 +0200
Subject: [Pw_forum] help me: how to calculate free energy
In-Reply-To: <BAY106-F10444BCD755FF071A5FE2F93460@phx.gbl>
References: <BAY106-F10444BCD755FF071A5FE2F93460@phx.gbl>
Message-ID: <200608221659.34622.giannozz@nest.sns.it>

On Tuesday 22 August 2006 16:39, hailin yu wrote:

> at the range of 0~400 K, our results ( free energy) decrease nearly 6 eV,
> while the previous work is about 0.26 eV. Where are wrong?

you carefully checked units, didn't you?

Paolo
-- 
Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From akohlmey at cmm.upenn.edu  Tue Aug 22 17:33:53 2006
From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer)
Date: Tue, 22 Aug 2006 11:33:53 -0400
Subject: [Pw_forum] Code starting trouble on the PC Cluster
In-Reply-To: <356250361.19202@mail.sdu.edu.cn>
References: <356250361.19202@mail.sdu.edu.cn>
Message-ID: <7b6913e90608220833o1f0c5546i1a6ff8daaee1d5ea@mail.gmail.com>

On 8/22/06, Shujun Hu <hushujun at mail.sdu.edu.cn> wrote:
>
> Dear Drs,

hi,

> When starting the parallel code pw.x on the cluster (Intel Pentium-D and MPICH2),
> usually i need run the command several times. Nothing appear on the screen for the

why??

> early commands but the last one seems good. I have checked the process queue and
> found that lots of process named pw.x under python were sleeping. I donnt know
> why.

without any detail information, it is almost impossible to give
specific advice. you have to help us to help you. none of us
is a psychic and can read your mind (or the one of your computer).

so you should:
a) try out and test the test programs, that come with the
MPICH2 software. if you cannot make them run, you
have to fix your setup, since you won't be able to run
_any_ parallel MPI program.

b) let is know the command line you were using. please check
the manual about running parallel jobs. many parallel environments
do not support reading input files from stdin, so you may need to
use the -input flag. the fact that you have many hung processes
is an indication that something like this might be happening.


hope that helps,
    axel.
>
>
>                                          Shujun Hu
>
>
>
>
>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
>


-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
  Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.


From ajoshi at mines.edu  Tue Aug 22 18:25:34 2006
From: ajoshi at mines.edu (Ameya Joshi)
Date: Tue, 22 Aug 2006 10:25:34 -0600
Subject: [Pw_forum] Nickel surface calculations
References: <BAY106-F10444BCD755FF071A5FE2F93460@phx.gbl>
Message-ID: <005301c6c607$996a9e90$3b20438a@brucite>

Hello all,

Attached is my input file for relaxing the (111) surface for Nickel. The run ended with an error message, that I fail to understand. Heres the last part of the output:

*******************************************************************************
number of scf cycles = 23

number of bfgs steps = 11

energy old = -257.0883841977 ryd

energy new = -257.0883839659 ryd

CASE: energy_new > energy_old

new trust radius = 0.0000499952 bohr

trust_radius < trust_radius_min

resetting bfgs history

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

from bfgs : error # 1

bfgs history already reset at previous step

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

stopping ...

*******************************************************************************

I checked the previous step in the output and found that the bfgs history was reset at the previous step too (as the error message says). Can anyone please explain what this means and a way to remedy this ? Just for information, the geometry and energy was almost converged by the end of the run. 

Thanks,

Ameya Joshi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060822/6979f640/attachment.htm 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ni_111_3layer_nbands_20.in
Type: application/octet-stream
Size: 1288 bytes
Desc: not available
Url : /pipermail/attachments/20060822/6979f640/attachment.obj 

From akohlmey at cmm.upenn.edu  Tue Aug 22 18:51:39 2006
From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer)
Date: Tue, 22 Aug 2006 12:51:39 -0400
Subject: [Pw_forum] Nickel surface calculations
In-Reply-To: <005301c6c607$996a9e90$3b20438a@brucite>
References: <BAY106-F10444BCD755FF071A5FE2F93460@phx.gbl>
	 <005301c6c607$996a9e90$3b20438a@brucite>
Message-ID: <7b6913e90608220951l3c150418w6b96467f55fe7b58@mail.gmail.com>

ameya,

please check the mailing list archives first. similar reports
have been sent in (and replied to) before.

it looks as if your geometry optimization has problems
to converge. your convergence threshold for the SCF
is pretty large. it should to be smaller by about two
orders of magnitude than the convergence parameters
for the geometry.

cheers,
    axel.

On 8/22/06, Ameya Joshi <ajoshi at mines.edu> wrote:
>
>
> Hello all,
>
> Attached is my input file for relaxing the (111) surface for Nickel. The run
> ended with an error message, that I fail to understand. Heres the last part
> of the output:
>
> *******************************************************************************
>
>
> number of scf cycles = 23
>
> number of bfgs steps = 11
>
> energy old = -257.0883841977 ryd
>
> energy new = -257.0883839659 ryd
>
> CASE: energy_new > energy_old
>
> new trust radius = 0.0000499952 bohr
>
> trust_radius < trust_radius_min
>
> resetting bfgs history
>
> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>
> from bfgs : error # 1
>
> bfgs history already reset at previous step
>
> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>
> stopping ...
>
> *******************************************************************************
>
> I checked the previous step in the output and found that the bfgs history
> was reset at the previous step too (as the error message says). Can anyone
> please explain what this means and a way to remedy this ? Just for
> information, the geometry and energy was almost converged by the end of the
> run.
>
> Thanks,
>
> Ameya Joshi
>


-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
  Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.


From baroni at sissa.it  Tue Aug 22 21:43:15 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Tue, 22 Aug 2006 21:43:15 +0200
Subject: [Pw_forum] help me: how to calculate free energy
In-Reply-To: <BAY106-F10444BCD755FF071A5FE2F93460@phx.gbl>
References: <BAY106-F10444BCD755FF071A5FE2F93460@phx.gbl>
Message-ID: <046F1505-440F-4D87-9309-4650C1CCB4ED@sissa.it>


On Aug 22, 2006, at 4:39 PM, hailin yu wrote:

> Dear Stefano B:
>      Thank you for your advice.
> I have writen a programme to compute the free energy according to  
> Equ(1)(PRB,57,10421(1998)).

"eq 1"

> And I have checked the normalization of the DOS and found the  
> integral of the DOS  equal to 5.932386, agreement with 3*2 (2  
> atomic per unit cell, is unit cell or primitive cell?? ). I also  
> have using Equ(3)( Chem. Phys. Lett. 417,272-276(2005))

"eq 2"

> to compute free energy and attainted the same results with using Equ 
> (1)(PRB,57,10421(1998)).

I am confused. The equations that you cite are the same, but for the  
fact that in the first one calculates the free energies directly by  
sampling phonon frequencies in the BZ, whereas in the second one  
makes explicit use of the density of states. It thus seems that,  
while using the first equation, you do *NOT* use any DOS, but just  
the phonon frequencies calculated by PHONON. But then you should have  
*TWO* ten-lines fortran codes, don't you? If you really use the first  
eq as is and the second as is, the fact that you obtain the same  
result is good news: your two 10lines codes are likely to be almost  
correct.

> but our results is still different with the previous work (PRB, 
> 57,10421(1998)). At the range of 0~400 K, our results ( free  
> energy) decrease nearly 6 eV,  while the previous work is about  
> 0.26 eV. Where are wrong?

difficult to say. may be units, but you may want to post the  
calculated DOS in a sensible format (e.g., two columns: first  
frequency (specify units, second DOS)

Also, see a recent post by Eyvaz Isaev who seems eager to share some  
of the codes he has written to calculate thermodynamic properties  
using QHA/DFPT.

Cheers - SB


---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060822/6a48c069/attachment.htm 

From eyvaz_isaev at yahoo.com  Tue Aug 22 22:55:44 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Tue, 22 Aug 2006 13:55:44 -0700 (PDT)
Subject: [Pw_forum] help me: how to calculate free energy
In-Reply-To: <046F1505-440F-4D87-9309-4650C1CCB4ED@sissa.it>
Message-ID: <20060822205544.78275.qmail@web60320.mail.yahoo.com>

Hi Stefano,
 
> Also, see a recent post by Eyvaz Isaev who seems
> eager to share some  of the codes he has written to
calculate thermodynamic properties  using QHA/DFPT.
> 
Yes, you are right. 

Very bests,
Eyvaz.


> Cheers - SB
> 
> 
> ---
> Stefano Baroni - SISSA  &  DEMOCRITOS National
> Simulation Center -  
> Trieste
> [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni
> (skype)
> 
> Please, if possible, don't  send me MS Word or
> PowerPoint attachments
> Why? See: 
>
http://www.gnu.org/philosophy/no-word-attachments.html
> 
> 
> 
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From emenendez at macul.ciencias.uchile.cl  Tue Aug 22 22:04:05 2006
From: emenendez at macul.ciencias.uchile.cl (Eduardo Ariel Menendez P)
Date: Tue, 22 Aug 2006 16:04:05 -0400 (CLT)
Subject: [Pw_forum] help me: how to calculate free energy
In-Reply-To: <20060822205601.32395.36202.Mailman@democritos.sissa.it>
References: <20060822205601.32395.36202.Mailman@democritos.sissa.it>
Message-ID: <Pine.LNX.4.58.0608221601520.31094@macul.ciencias.uchile.cl>

Well, I am calculating free energies now, and I would like to try your
programs.
Eduardo
>
> Hi Stefano,
>
> > Also, see a recent post by Eyvaz Isaev who seems
> > eager to share some  of the codes he has written to
> calculate thermodynamic properties  using QHA/DFPT.
> >
> Yes, you are right.
>
> Very bests,
> Eyvaz.
>
>
> > Cheers - SB
> >
> >
> > ---
> > Stefano Baroni - SISSA  &  DEMOCRITOS National
> > Simulation Center -
> > Trieste
> > [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni
> > (skype)
> >
> > Please, if possible, don't  send me MS Word or
> > PowerPoint attachments
> > Why? See:
> >
> http://www.gnu.org/philosophy/no-word-attachments.html
> >
> >
> >
> >
>
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
>
>
> --__--__--
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
>
> End of Pw_forum Digest
>


From duanxm at ts.infn.it  Wed Aug 23 06:19:46 2006
From: duanxm at ts.infn.it (Xiangmei Duan)
Date: Wed, 23 Aug 2006 06:19:46 +0200 (CEST)
Subject: [Pw_forum] heat of formation for Mg3N2
Message-ID: <Pine.LNX.4.61.0608230617070.20125@dawn.ts.infn.it>

Dear all,

I've opotimized \alpha-Mg3N2 with cubic anti-bixbyite structure (J. Mater. 
Chem., 12, 2475, 2002)using espresso-2.1.5,
and got the reasonable lattice constant 9.896 Ang. (LDA) (exp: 9.953 Ang).
When I calculated the heat of formation (in terms of the bulk Mg and N2),
I was expected to get an absolute value larger than the experimental one,
for the well known overbinding of LDA. In fact, I got -2.91 eV (exp. -4.80 
eV).
(we tested with LDA calculation using other code like Dmol^3, and got 
-5.38 eV with the lattice parameter 9.942 A)

For N2, the calculated bond length is 1.103 AA, and the binding energy is 
11.11 eV, and the exp. values 1.098 A and  9.91 eV

For Mg bulk, the lattice constants are 3.1317 A (a) and 5.21 (c). 
(comparing with 3.21 and 5.21)

Thanks in advance for your comments.

Best wishes,
Xiangmei

Here is the input for Mg3N2:
-----------------------------------

  &control
     calculation = 'relax'
     restart_mode='from_scratch',
     prefix='Mg3N2',
     tstress = .true.
     tprnfor = .true.
     pseudo_dir = '/home/562/xxd562/pw_potential/',
     outdir='/short/g46/Mg3N2/'
  /
  &system
     ibrav=  0, celldm(1) =0.0,
     nat=  80, ntyp= 2, ecutwfc =60.0,
  /
  &electrons
     mixing_beta = 0.5
     conv_thr =  1.0d-6
  /
  &ions
  /
ATOMIC_SPECIES
  N   14.0067   N.pz-rrkjus.UPF  (!!! I tried N.pz-vbc.UPF as well)
  Mg  24.3050   Mg.pz-n-vbc.UPF
ATOMIC_POSITIONS {bohr}
Mg      -2.859307108   2.203853843   7.281662824
Mg       2.854200083   7.144760834   7.281662591
Mg      -2.857214509  -2.207201820   2.065435928
Mg       2.852106059  -7.148768053   2.064960719
Mg       7.281662824  -2.859307108   2.203853843
Mg      -7.284950160  -6.496093980   2.204067017
Mg       7.281662591   2.854200083   7.144760834
Mg      -7.284813079   6.493789879   7.144043305
Mg       2.203853843   7.281662824  -2.859307108
Mg       2.204067017  -7.284950160  -6.496093980
Mg      -2.206521406  -2.069113782  -6.497311905
Mg       7.144043305  -7.284813079   6.493789879
Mg       2.854906875  -2.208126838  -7.284612512
Mg      -2.857632220  -7.147335003  -7.285149136
Mg       2.853458071   2.203185880  -2.069725394
Mg      -2.856132703   7.144985498  -2.069587582
Mg      -7.284612512   2.854906875  -2.208126838
Mg       7.281876795   6.491954164  -2.206999931
Mg      -7.285149136  -2.857632220  -7.147335003
Mg       7.282657619  -6.496621551  -7.148989844
Mg      -2.208126838  -7.284612512   2.854906875
Mg      -2.206999931   7.281876795   6.491954164
Mg       2.203039138   2.065404974   6.494363489
Mg      -7.148989844   7.282657619  -6.496621551
Mg       6.495220533  -7.149251987  -2.068929708
Mg      -6.497311905  -2.206521406  -2.069113782
Mg       6.493789879   7.144043305  -7.284813079
Mg      -6.496093980   2.204067017  -7.284950160
Mg      -2.068929708   6.495220533  -7.149251987
Mg       2.064960719   2.852106059  -7.148768053
Mg      -2.069113782  -6.497311905  -2.206521406
Mg       2.065435928  -2.857214509  -2.207201820
Mg      -7.149251987  -2.068929708   6.495220533
Mg      -7.148768053   2.064960719   2.852106059
Mg       7.144760834   7.281662591   2.854200083
Mg      -2.207201820   2.065435928  -2.857214509
Mg      -6.498845599   7.145349533   2.064874914
Mg       6.494363489   2.203039138   2.065404974
Mg      -6.496621551  -7.148989844   7.282657619
Mg       6.491954164  -2.206999931   7.281876795
Mg       2.064874914  -6.498845599   7.145349533
Mg      -2.069587582  -2.856132703   7.144985498
Mg       2.065404974   6.494363489   2.203039138
Mg      -2.069725394   2.853458071   2.203185880
Mg       7.145349533   2.064874914  -6.498845599
Mg       7.144985498  -2.069587582  -2.856132703
Mg      -7.147335003  -7.285149136  -2.857632220
Mg       2.203185880  -2.069725394   2.853458071
N        4.674971919   4.674971919   4.674971919
N       -4.679097645   4.673994784   4.673250945
N        4.673250945  -4.679097645   4.673994784
N       -4.677243850  -4.676660785   4.675574921
N        4.675574921  -4.677243850  -4.676660785
N       -4.678385311  -4.678385311  -4.678385311
N       -4.676660785   4.675574921  -4.677243850
N        4.673994784   4.673250945  -4.679097645
N        9.349058284  -4.678339930   8.762003402
N        9.349380983   4.674933400   0.581867352
N        8.762003402   9.349058284  -4.678339930
N       -8.765957447  -0.002109422  -4.677660355
N       -4.678339930   8.762003402   9.349058284
N       -4.677660355  -8.765957447  -0.002109422
N        4.674355140  -0.586753626  -0.002368775
N        9.348593427   4.673979601  -8.766153991
N        9.348905103  -4.677360759  -0.586128762
N       -8.766153991   9.348593427   4.673979601
N        8.761436635  -0.001647603   4.674640396
N        4.673979601  -8.766153991   9.348593427
N        4.674640396   8.761436635  -0.001647603
N       -4.677968914   0.582078512  -0.001873839
N       -0.002368775   4.674355140  -0.586753626
N       -0.002109422  -4.677660355  -8.765957447
N       -0.586753626  -0.002368775   4.674355140
N        0.581867352   9.349380983   4.674933400
N        4.674933400   0.581867352   9.349380983
N       -0.001873839  -4.677968914   0.582078512
N       -0.001647603   4.674640396   8.761436635
N        0.582078512  -0.001873839  -4.677968914
N       -0.586128762   9.348905103  -4.677360759
N       -4.677360759  -0.586128762   9.348905103
K_POINTS {automatic}
2 2 2 1 1 1  (!!! with the opotimized coordinates, I used the k-mesh 4 4 4 
0 0 0 to do 'scf')
CELL_PARAMETERS
            18.70000000000000    0.00000000000000    0.00000000000000
               0.00000000000000  18.70000000000000    0.00000000000000
               0.00000000000000    0.00000000000000  18.70000000000000
EOF
--------------------


From baroni at sissa.it  Wed Aug 23 08:32:15 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Wed, 23 Aug 2006 08:32:15 +0200
Subject: [Pw_forum] heat of formation for Mg3N2
In-Reply-To: <Pine.LNX.4.61.0608230617070.20125@dawn.ts.infn.it>
References: <Pine.LNX.4.61.0608230617070.20125@dawn.ts.infn.it>
Message-ID: <9F608030-A2A2-4DD1-A5BF-8EE0A42D57C1@sissa.it>

Dear Xiangmei:

On Aug 23, 2006, at 6:19 AM, Xiangmei Duan wrote:

>
> Dear all,
>
> I've opotimized \alpha-Mg3N2 with cubic anti-bixbyite structure (J.  
> Mater. Chem., 12, 2475, 2002)using espresso-2.1.5,
> and got the reasonable lattice constant 9.896 Ang. (LDA) (exp:  
> 9.953 Ang).
> When I calculated the heat of formation (in terms of the bulk Mg  
> and N2),
> I was expected to get an absolute value larger than the  
> experimental one,
> for the well known overbinding of LDA.

this I do not quite understand. it seems to me that you are comparing  
*TWO* binding energies: that of \alpha-Mg3N2 with the appriopriate  
average of those of Mg and N2. When the two terms of a difference are  
(supposedly) both too large, why do you expect the difference to be  
too large too? For N2, you mean N2 in the gase phase, I guess (and  
not "bulk")?

> In fact, I got -2.91 eV (exp. -4.80 eV).
> (we tested with LDA calculation using other code like Dmol^3, and  
> got -5.38 eV with the lattice parameter 9.942 A)

This seems to me too large a difference: same structure, same  
functional ???

> For N2, the calculated bond length is 1.103 AA, and the binding  
> energy is 11.11 eV, and the exp. values 1.098 A and  9.91 eV

binding of what with respect to what? molecular binding wrt isolated  
atoms?

> For Mg bulk, the lattice constants are 3.1317 A (a) and 5.21 (c).  
> (comparing with 3.21 and 5.21)

how about the binding energy?

Cheers - Stefano

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060823/ff39b54a/attachment.htm 

From yuhailin_79 at msn.com  Wed Aug 23 10:06:53 2006
From: yuhailin_79 at msn.com (hailin yu)
Date: Wed, 23 Aug 2006 16:06:53 +0800
Subject: [Pw_forum] help: free energy
Message-ID: <BAY106-F288F2AD4E904650DF5229793470@phx.gbl>

Dear Eyvaz:
      Dear Eyvaz, can you give me the codes to calculate thermodynamic 
properties? I did't find the post of Stefano refered.
    Thanks!

Dear all:
     I give the phonon density of states of alpha-Sn which I attained using 
PWSCF ( and I sure it's correctly) in the attach( The frequency unit is 
cm-1). I have used the Equ(3)( Chem. Phys. Lett. 417,272-276(2005)) to 
calculate the free energy and I am sure the units is right. But, as 
previously mentioned, my results is too large compared with previously work. 
So I hope someone can compute the free energy and tell what's wrong.
Thanks a lot!
yuhailin

_________________________________________________________________
Don't just search. Find. Check out the new MSN Search! 
http://search.msn.click-url.com/go/onm00200636ave/direct/01/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: aSndos.dat
Type: application/octet-stream
Size: 2432 bytes
Desc: not available
Url : /pipermail/attachments/20060823/c8f9d559/attachment.obj 

From duan at physics.usyd.edu.au  Wed Aug 23 10:07:42 2006
From: duan at physics.usyd.edu.au (Xiangmei Duan)
Date: Wed, 23 Aug 2006 18:07:42 +1000 (EST)
Subject: [Pw_forum] heat of formation for Mg3N2
In-Reply-To: <9F608030-A2A2-4DD1-A5BF-8EE0A42D57C1@sissa.it>
References: <Pine.LNX.4.61.0608230617070.20125@dawn.ts.infn.it>
 <9F608030-A2A2-4DD1-A5BF-8EE0A42D57C1@sissa.it>
Message-ID: <Pine.LNX.4.62.0608231744350.9491@cmt.physics.usyd.edu.au>

Dear stefano,

Many thanks for your reply.

On Wed, 23 Aug 2006, Stefano Baroni wrote:

>> and got the reasonable lattice constant 9.896 Ang. (LDA) (exp: 9.953 Ang).
>> When I calculated the heat of formation (in terms of the bulk Mg and N2),
>> I was expected to get an absolute value larger than the experimental one,
>> for the well known overbinding of LDA.
>
> this I do not quite understand. it seems to me that you are comparing *TWO* 
> binding energies: that of \alpha-Mg3N2 with the appriopriate average of those 
> of Mg and N2. When the two terms of a difference are (supposedly) both too 
> large, why do you expect the difference to be too large too? For N2, you mean 
> N2 in the gase phase, I guess (and not "bulk")?

For the formation energy (H), I used the fomula
H = E_tot(Mg3N2) - 3E_tot(Mg:bulk) - E_tot(N2) (N2 from the pas phase)
The lattice constant is smaller than the experimental value, which 
indicates the overbinding. Am I right ?

>> In fact, I got -2.91 eV (exp. -4.80 eV).
>> (we tested with LDA calculation using other code like Dmol^3, and got -5.38 
>> eV with the lattice parameter 9.942 A)
> This seems to me too large a difference: same structure, same functional ???

me too.
SAME structure, both LDA, DMol^3 use all-electrons for the basis set.

>> For N2, the calculated bond length is 1.103 AA, and the binding energy is 
>> 11.11 eV, and the exp. values 1.098 A and  9.91 eV
>
> binding of what with respect to what? molecular binding wrt isolated atoms?
E_b = -[E_tot (N2)- 2E_tot (N-atom)]  both in gas phase

>> For Mg bulk, the lattice constants are 3.1317 A (a) and 5.21 (c). 
>> (comparing with 3.21 and 5.21)
>
> how about the binding energy?
E_b = -[E_tot(Mg_bulk)- 2 E_tot(Mg-at)(gas)] = 4.95 eV
The total energies for free atoms (N and Mg) include the spin-polarization 
energies.

Best wishes,
Xiangmei


From sareh_zaree at yahoo.com  Wed Aug 23 10:33:43 2006
From: sareh_zaree at yahoo.com (sareh zarei)
Date: Wed, 23 Aug 2006 09:33:43 +0100 (BST)
Subject: [Pw_forum] superposition
Message-ID: <20060823083343.14051.qmail@web31012.mail.mud.yahoo.com>

  Dear all 
  I am trying to plot difference electron density (difference between electron density of crystal and superposition of atomic density) but I do not know how can I calculate superposition of atomic density .
   
   can you suggest me some way for doing it.
    regards
   sareh
   
   Iran,Isfahan
   Isfahan University of Technology
 Send instant messages to your online friends http://uk.messenger.yahoo.com 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060823/6bf90b3c/attachment.htm 

From baroni at sissa.it  Wed Aug 23 16:05:29 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Wed, 23 Aug 2006 16:05:29 +0200
Subject: [Pw_forum] help: free energy
In-Reply-To: <BAY106-F288F2AD4E904650DF5229793470@phx.gbl>
References: <BAY106-F288F2AD4E904650DF5229793470@phx.gbl>
Message-ID: <89F7EE8F-F787-4F3F-B5F7-9AF421920798@sissa.it>

Dear Yu Hailin,

I was rather confused by the units of your 10-lines code, so I took  
the liberty to re-write it in a way which, I hope, will allow you to  
check more easily the units. Here it is. When run onto your data, it  
reproduces the free energy of Fig. 3 of our paper.

Regards - Stefano B.


On Aug 23, 2006, at 10:06 AM, hailin yu wrote:

> Dear Eyvaz:
>      Dear Eyvaz, can you give me the codes to calculate  
> thermodynamic properties? I did't find the post of Stefano refered.
>    Thanks!
>
> Dear all:
>     I give the phonon density of states of alpha-Sn which I  
> attained using PWSCF ( and I sure it's correctly) in the attach 
> ( The frequency unit is cm-1). I have used the Equ(3)( Chem. Phys.  
> Lett. 417,272-276(2005)) to calculate the free energy and I am sure  
> the units is right. But, as previously mentioned, my results is too  
> large compared with previously work. So I hope someone can compute  
> the free energy and tell what's wrong.
> Thanks a lot!
> yuhailin
>
> _________________________________________________________________
> Don't just search. Find. Check out the new MSN Search! http:// 
> search.msn.click-url.com/go/onm00200636ave/direct/01/
> <aSndos.dat>

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html
?

-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060823/8971874f/attachment.htm 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: free-energy.f90
Type: application/octet-stream
Size: 2110 bytes
Desc: not available
Url : /pipermail/attachments/20060823/8971874f/attachment.obj 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060823/8971874f/attachment-0001.htm 

From baroni at sissa.it  Wed Aug 23 16:20:11 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Wed, 23 Aug 2006 16:20:11 +0200
Subject: [Pw_forum] heat of formation for Mg3N2
In-Reply-To: <Pine.LNX.4.62.0608231744350.9491@cmt.physics.usyd.edu.au>
References: <Pine.LNX.4.61.0608230617070.20125@dawn.ts.infn.it> <9F608030-A2A2-4DD1-A5BF-8EE0A42D57C1@sissa.it> <Pine.LNX.4.62.0608231744350.9491@cmt.physics.usyd.edu.au>
Message-ID: <0B0B2C91-BED1-41E1-BA61-4D5B352BE928@sissa.it>


On Aug 23, 2006, at 10:07 AM, Xiangmei Duan wrote:

>
> Dear stefano,
>
> Many thanks for your reply.
>
> On Wed, 23 Aug 2006, Stefano Baroni wrote:
>
>>> and got the reasonable lattice constant 9.896 Ang. (LDA) (exp:  
>>> 9.953 Ang).
>>> When I calculated the heat of formation (in terms of the bulk Mg  
>>> and N2),
>>> I was expected to get an absolute value larger than the  
>>> experimental one,
>>> for the well known overbinding of LDA.
>>
>> this I do not quite understand. it seems to me that you are  
>> comparing *TWO* binding energies: that of \alpha-Mg3N2 with the  
>> appriopriate average of those of Mg and N2. When the two terms of  
>> a difference are (supposedly) both too large, why do you expect  
>> the difference to be too large too? For N2, you mean N2 in the  
>> gase phase, I guess (and not "bulk")?
>
> For the formation energy (H), I used the fomula
> H = E_tot(Mg3N2) - 3E_tot(Mg:bulk) - E_tot(N2) (N2 from the pas phase)

Mg is hexagonal, two atoms/cell, so shouldn't you divide E_tot 
(Mg:bulk) by 2 ??? Below, you do compare the energy of bulk Mg with  
that of 2 isolated atoms ...

> The lattice constant is smaller than the experimental value, which  
> indicates the overbinding. Am I right ?

I think so. My point was that, strictly speaking, the fact that you  
obtain too large a formation energy as defined above cannot be  
ascribed to any overbinding effects, because overbinding affects all  
the terms of the formula which have different signs ... do different  
PP's give the same results? how about the kin ene cutoff? (60 Ry  
should not be too bad).


>>> In fact, I got -2.91 eV (exp. -4.80 eV).
>>> (we tested with LDA calculation using other code like Dmol^3, and  
>>> got -5.38 eV with the lattice parameter 9.942 A)
>> This seems to me too large a difference: same structure, same  
>> functional ???
>
> me too.
> SAME structure, both LDA, DMol^3 use all-electrons for the basis set.

that cannot justify a difference of 2.5 eV between PP and AE  
calculations ...

If nothing else works, you may want to consider non-linear charge  
corrections on Mg?

Cheers - Stefano

>
>>> For N2, the calculated bond length is 1.103 AA, and the binding  
>>> energy is 11.11 eV, and the exp. values 1.098 A and  9.91 eV
>>
>> binding of what with respect to what? molecular binding wrt  
>> isolated atoms?
> E_b = -[E_tot (N2)- 2E_tot (N-atom)]  both in gas phase
>
>>> For Mg bulk, the lattice constants are 3.1317 A (a) and 5.21 (c).  
>>> (comparing with 3.21 and 5.21)
>>
>> how about the binding energy?
> E_b = -[E_tot(Mg_bulk)- 2 E_tot(Mg-at)(gas)] = 4.95 eV
> The total energies for free atoms (N and Mg) include the spin- 
> polarization energies.
>
> Best wishes,
> Xiangmei
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060823/aed4f921/attachment.htm 

From timtro at gmail.com  Wed Aug 23 17:59:50 2006
From: timtro at gmail.com (Tim Teatro)
Date: Wed, 23 Aug 2006 11:59:50 -0400
Subject: [Pw_forum] pw.x not able to specity initial ion velocities
Message-ID: <560bcca0608230859j3e419233q3322964d893e2d36@mail.gmail.com>

Hello all.

This is probably an easy one. I'm sorry to bother you with it.

When I try to specify starting velocities for ions in a pw.x input file
(obviously for dynamics), I set ion_velocities = 'from_input' and specify an
ATOMIC_VELOCITIES card. When it runs, I get a warning that a card is
ignored. The ignored card is (for whatever reason) the last entry in my
ATOMIC_VELOCITIES card ( Species(n)   Vx(n)  Vy(n)   Vz(n) ), and the
temperature calculated at the end of the first step certainly does not
reflect the temperature that the system should have given the velocities I
specified.

I looked in the source code, and according to read_cards.f90, the program
should hit the velocity card and read of the next nat lines, so I have no
idea why it would try to read off my last velocity entry as a card. There
also doesn't seem to be any comments that this bit of code is restricted to
cp.x or something. But is it?

Thanks!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060823/7c0650cd/attachment.htm 

From akohlmey at cmm.upenn.edu  Wed Aug 23 18:03:49 2006
From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer)
Date: Wed, 23 Aug 2006 12:03:49 -0400
Subject: [Pw_forum] pw.x not able to specity initial ion velocities
In-Reply-To: <560bcca0608230859j3e419233q3322964d893e2d36@mail.gmail.com>
References: <560bcca0608230859j3e419233q3322964d893e2d36@mail.gmail.com>
Message-ID: <7b6913e90608230903h46ab20cdx6ee12a3767898459@mail.gmail.com>

tim,

please check your input and check whether the number of
lines for the atoms and velocities both match the 'nat' variable.
if they do please post your input file, so we can examine it.

thanks,
    axel.

On 8/23/06, Tim Teatro <timtro at gmail.com> wrote:
> Hello all.
>
> This is probably an easy one. I'm sorry to bother you with it.
>
> When I try to specify starting velocities for ions in a pw.x input file
> (obviously for dynamics), I set ion_velocities = 'from_input' and specify an
> ATOMIC_VELOCITIES card. When it runs, I get a warning that a card is
> ignored. The ignored card is (for whatever reason) the last entry in my
> ATOMIC_VELOCITIES card ( Species(n)   Vx(n)  Vy(n)   Vz(n) ), and the
> temperature calculated at the end of the first step certainly does not
> reflect the temperature that the system should have given the velocities I
> specified.
>
> I looked in the source code, and according to read_cards.f90, the program
> should hit the velocity card and read of the next nat lines, so I have no
> idea why it would try to read off my last velocity entry as a card. There
> also doesn't seem to be any comments that this bit of code is restricted to
> cp.x or something. But is it?
>
> Thanks!
>


-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
  Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.


From emenendez at macul.ciencias.uchile.cl  Wed Aug 23 17:14:50 2006
From: emenendez at macul.ciencias.uchile.cl (Eduardo Ariel Menendez P)
Date: Wed, 23 Aug 2006 11:14:50 -0400 (CLT)
Subject: [Pw_forum] Re: phonon, gk_sort error
Message-ID: <Pine.LNX.4.58.0608231055550.15997@macul.ciencias.uchile.cl>

Hi,
I have the same runtime error that Kevin reported on August 21 with
Version 3.1.1. This
happens with example02. -npool 1 works ok, but -npool >1 produces errors.
For example
mpirun -np 2 ~/ChemUtils/Espresso/espresso-3.1.1/bin/pw.x -npool 2 < c.scf.in
mpirun -np 2 ~/ChemUtils/Espresso/espresso-3.1.1/bin/ph.x -npool 2 < c.phG.in
ends  with
     nbndx  =     4  nbnd   =     4  natomwfc =     8  npwx   =     184
     nelec  =   8.00  nkb   =    16  ngl    =     118

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

     from gk_sort : error #         1
     array gk out-of-bounds

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

The error depends on the number of CPUs, e.g.

mpirun -np 4 ~/ChemUtils/Espresso/espresso-3.1.1/bin/pw.x -npool 2 < c.scf.in
mpirun -np 4 ~/ChemUtils/Espresso/espresso-3.1.1/bin/ph.x -npool 2 < c.phG.in
ends without an explicit message, such as,
     nbndx  =     4  nbnd   =     4  natomwfc =     8  npwx   =      97
     nelec  =   8.00  nkb   =    16  ngl    =     118
-----------------------------------------------------------------------------
One of the processes started by mpirun has exited with a nonzero exit
code.  This typically indicates that the process finished in error.
If your process did not finish in error, be sure to include a "return
0" or "exit(0)" in your C code before exiting the application.

In everycase, npool 1 works OK .

Moreover
dos.x give the same error message "array gk out-of-bounds" when -npool >1

regards

Eduardo A. Menendez Proupin
Department of Physics
Faculty of Science
University of Chile
Las Palmeras 3425
?u?oa, Santiago
Chile
Phone: 56+2+978 74 11
http://fisica.ciencias.uchile.cl/~emenendez/

Hello,

I am doing a simple test to calculate Si gamma phonons.

I run
pw.x -npool 4 < si3.scf.in > si3.scf.out

then
ph.x -npool 4 < si3.phG.in > si3.phG.out

which produces the following error:
      from gk_sort : error #         1
      array gk out-of-bounds

The error does not occur when I set -npool 1.

The error occurs for espresso-3.1 and 3.1.1, but not 2.1.5.

My input files are below.  Please let me know if I should provide other
information.  Thanks.

Regards,
Kevin Chan


From akara at ksu.edu  Wed Aug 23 18:50:29 2006
From: akara at ksu.edu (akara at ksu.edu)
Date: Wed, 23 Aug 2006 11:50:29 -0500
Subject: [Pw_forum] magnetic anisotropy energy (MAE)
Message-ID: <1156351829.44ec87556a5e9@webmail.ksu.edu>


Dear all,

I would like to know if espresso has the capability to calculate
the MAE of a given system (surfaces preferably) and if the answer
is yes, can I have some details on how to do that?

Thank you very much.

A. Kara


From konstantin_kudin at yahoo.com  Wed Aug 23 18:53:28 2006
From: konstantin_kudin at yahoo.com (Konstantin Kudin)
Date: Wed, 23 Aug 2006 09:53:28 -0700 (PDT)
Subject: [Pw_forum] pw.x not able to specity initial ion velocities
In-Reply-To: <7b6913e90608230903h46ab20cdx6ee12a3767898459@mail.gmail.com>
Message-ID: <20060823165328.68184.qmail@web52001.mail.yahoo.com>

 Actually, the velocity input appears to be almost purely unsupported.
Consider these lines in read_cards.f90 :

       ELSE IF ( TRIM(card) == 'ATOMIC_VELOCITIES' ) THEN
          !
          CALL card_ion_velocities( input_line )
          IF ( ( prog == 'PW' .OR. prog == 'CP' ) .AND. ionode ) &
             WRITE( stdout,'(A)') 'Warning: card '//trim(input_line)//'
ignored'

 What this piece of code is saying is that for either PW or CP the
ATOMIC_VELOCITIES card is ignored (when is it supported then?). While
this of course does not guarantee that this message is reliable, my own
experience with trying to input velocities into CP seems to confirm the
lack of such functionality.

 Kostya

 
 

--- Axel Kohlmeyer <akohlmey at cmm.upenn.edu> wrote:

> tim,
> 
> please check your input and check whether the number of
> lines for the atoms and velocities both match the 'nat' variable.
> if they do please post your input file, so we can examine it.
> 
> thanks,
>     axel.
> 
> On 8/23/06, Tim Teatro <timtro at gmail.com> wrote:
> > Hello all.
> >
> > This is probably an easy one. I'm sorry to bother you with it.
> >
> > When I try to specify starting velocities for ions in a pw.x input
> file
> > (obviously for dynamics), I set ion_velocities = 'from_input' and
> specify an
> > ATOMIC_VELOCITIES card. When it runs, I get a warning that a card
> is
> > ignored. The ignored card is (for whatever reason) the last entry
> in my
> > ATOMIC_VELOCITIES card ( Species(n)   Vx(n)  Vy(n)   Vz(n) ), and
> the
> > temperature calculated at the end of the first step certainly does
> not
> > reflect the temperature that the system should have given the
> velocities I
> > specified.
> >
> > I looked in the source code, and according to read_cards.f90, the
> program
> > should hit the velocity card and read of the next nat lines, so I
> have no
> > idea why it would try to read off my last velocity entry as a card.
> There
> > also doesn't seem to be any comments that this bit of code is
> restricted to
> > cp.x or something. But is it?
> >
> > Thanks!
> >
> 
> 
> -- 
>
=======================================================================
> Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu  
> http://www.cmm.upenn.edu
>   Center for Molecular Modeling   --   University of Pennsylvania
> Department of Chemistry, 231 S.34th Street, Philadelphia, PA
> 19104-6323
> tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel:
> 1-215-898-5425
>
=======================================================================
> If you make something idiot-proof, the universe creates a better
> idiot.
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From cyrille.barreteau at cea.fr  Wed Aug 23 19:29:38 2006
From: cyrille.barreteau at cea.fr (Cyrille Barreteau)
Date: Wed, 23 Aug 2006 19:29:38 +0200 (CEST)
Subject: [Pw_forum] magnetic anisotropy energy (MAE)
In-Reply-To: <1156351829.44ec87556a5e9@webmail.ksu.edu>
References: <1156351829.44ec87556a5e9@webmail.ksu.edu>
Message-ID: <2026.81.56.70.96.1156354178.squirrel@dsm-mail.saclay.cea.fr>

Yes you can calculate the MAE of a given system.
You first need a relativistic pseudo-potential
(including spin-orbit coupling)
You will have to set
    lspinorb=.true.,
    noncolin=.true.,
Then you just need to define the theta and phi.
    angle1(1) =  theta
    angle2(1) =  phi

If theta and phi define a high symmetry direction the
system will stay in the initial theta and phi.
You can therefore easily get the MAE of a surface
by taking theta and phi in or out of the plane.

If ever you need the whole E(theta,phi) curve you
will need to use a penalization...

Just a comment: MAE are usually very small numbers (below meV)
and you need an excellent convergence with respect to k-points
cut-off etc...


good luck
   Cyrille

ps You can have a look at my paper:
Magnetism of iron: from the bulk to the monatomic wire  Gabriel Aut?s,
Cyrille Barreteau, Daniel Spanjaard and Marie-Catherine Desjonqu?res  J.
Phys.: Condens. Matter 18 (2006) 6785-6813.


Le Mer 23 ao?t 2006 18:50, akara at ksu.edu a ?crit :
>
>
> Dear all,
>
> I would like to know if espresso has the capability to calculate
> the MAE of a given system (surfaces preferably) and if the answer
> is yes, can I have some details on how to do that?
>
> Thank you very much.
>
> A. Kara
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>



From emenendez at macul.ciencias.uchile.cl  Wed Aug 23 18:32:24 2006
From: emenendez at macul.ciencias.uchile.cl (Eduardo Ariel Menendez P)
Date: Wed, 23 Aug 2006 12:32:24 -0400 (CLT)
Subject: [Pw_forum] serial algorithm in parallel phonon calculation
Message-ID: <Pine.LNX.4.58.0608231224300.15997@macul.ciencias.uchile.cl>

Hi,
Looking at the ph.x output, when running in parallel with four CPUs
(-npool 1) I found the following intriguing report
***********************
     Calculation of q =   0.2500  0.2500 -0.2500

     looking for the optimal diagonalization algorithm ...

     a serial algorithm will be used
***********************

Is it important or is there a parallel algorithm? I also see that top
reports
  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
11020 eduardo   16   0  107m  20m  50m R 63.9  1.0  32:49.09 ph.x
11019 eduardo   16   0  107m  20m  50m S 21.6  1.0  11:22.56 ph.x
and the same in the other dual node. Is it related to the algorithm?

Thanks
Eduardo

Eduardo A. Menendez Proupin
Department of Physics
Faculty of Science
University of Chile
Las Palmeras 3425
?u?oa, Santiago
Chile
Phone: 56+2+978 74 11
http://fisica.ciencias.uchile.cl/~emenendez/


From akara at ksu.edu  Wed Aug 23 19:48:24 2006
From: akara at ksu.edu (akara at ksu.edu)
Date: Wed, 23 Aug 2006 12:48:24 -0500
Subject: [Pw_forum] magnetic anisotropy energy (MAE)
In-Reply-To: <2026.81.56.70.96.1156354178.squirrel@dsm-mail.saclay.cea.fr>
References: <1156351829.44ec87556a5e9@webmail.ksu.edu> <2026.81.56.70.96.1156354178.squirrel@dsm-mail.saclay.cea.fr>
Message-ID: <1156355304.44ec94e828278@webmail.ksu.edu>

THANK you Cyrille. I ll have a look at your paper.

Kader

Quoting Cyrille Barreteau <cyrille.barreteau at cea.fr>:

> Yes you can calculate the MAE of a given system.
> You first need a relativistic pseudo-potential
> (including spin-orbit coupling)
> You will have to set
>     lspinorb=.true.,
>     noncolin=.true.,
> Then you just need to define the theta and phi.
>     angle1(1) =  theta
>     angle2(1) =  phi
>
> If theta and phi define a high symmetry direction the
> system will stay in the initial theta and phi.
> You can therefore easily get the MAE of a surface
> by taking theta and phi in or out of the plane.
>
> If ever you need the whole E(theta,phi) curve you
> will need to use a penalization...
>
> Just a comment: MAE are usually very small numbers (below meV)
> and you need an excellent convergence with respect to k-points
> cut-off etc...
>
>
> good luck
>    Cyrille
>
> ps You can have a look at my paper:
> Magnetism of iron: from the bulk to the monatomic wire  Gabriel
> Aut??s,
> Cyrille Barreteau, Daniel Spanjaard and Marie-Catherine Desjonqu??res
>  J.
> Phys.: Condens. Matter 18 (2006) 6785-6813.
>
>
> Le Mer 23 ao??t 2006 18:50, akara at ksu.edu a ??crit :
> >
> >
> > Dear all,
> >
> > I would like to know if espresso has the capability to calculate
> > the MAE of a given system (surfaces preferably) and if the answer
> > is yes, can I have some details on how to do that?
> >
> > Thank you very much.
> >
> > A. Kara
> > _______________________________________________
> > Pw_forum mailing list
> > Pw_forum at pwscf.org
> > http://www.democritos.it/mailman/listinfo/pw_forum
> >
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
>




From giannozz at nest.sns.it  Wed Aug 23 20:12:52 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Wed, 23 Aug 2006 20:12:52 +0200
Subject: [Pw_forum] phonon, gk_sort error
In-Reply-To: <Pine.LNX.4.63.0608211802550.9242@civet.berkeley.edu>
References: <Pine.LNX.4.63.0608211802550.9242@civet.berkeley.edu>
Message-ID: <200608232012.52204.giannozz@nest.sns.it>

On Tuesday 22 August 2006 03:32, Kevin Chan wrote:

> I am doing a simple test to calculate Si gamma phonons.
> [..]
> which produces the following error: 
>       from gk_sort : error #         1
>       array gk out-of-bounds
>
> The error does not occur when I set -npool 1.
>
> The error occurs for espresso-3.1 and 3.1.1, but not 2.1.5.
> [...]
> My input files are below.  Please let me know if I should provide other
> information.

no more information is needed. There is (rather stupid) bug that under 
some circumstances leads to this error. If no error appears, the
calculations should be reliable. It will be fixed in the CVS version as 
soon as possible. No warranty that there will be a quick fix for stable
versions, though.

Paolo
-- 
Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From giannozz at nest.sns.it  Wed Aug 23 20:19:56 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Wed, 23 Aug 2006 20:19:56 +0200
Subject: [Pw_forum] serial algorithm in parallel phonon calculation
In-Reply-To: <Pine.LNX.4.58.0608231224300.15997@macul.ciencias.uchile.cl>
References: <Pine.LNX.4.58.0608231224300.15997@macul.ciencias.uchile.cl>
Message-ID: <200608232019.56476.giannozz@nest.sns.it>

On Wednesday 23 August 2006 18:32, Eduardo Ariel Menendez P wrote:

> Looking at the ph.x output, when running in parallel with four CPUs
> (-npool 1) I found the following intriguing report
> ***********************
>      Calculation of q =   0.2500  0.2500 -0.2500
>
>      looking for the optimal diagonalization algorithm ...
>
>      a serial algorithm will be used
> ***********************

this refers to the "conventional" diagonalization, used in the
Davidson iterative diagonalization to diagonalize over a subspace.
It is an additional level of parallelization that may or may not be
convenient to have. 

>   PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
> 11020 eduardo   16   0  107m  20m  50m R 63.9  1.0  32:49.09 ph.x
> 11019 eduardo   16   0  107m  20m  50m S 21.6  1.0  11:22.56 ph.x
> and the same in the other dual node. Is it related to the algorithm? 

no, it isn't. Are the two processing running on different CPU's?
it looks very lousy.

Paolo
-- 
Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From giannozz at nest.sns.it  Wed Aug 23 20:39:39 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Wed, 23 Aug 2006 20:39:39 +0200
Subject: [Pw_forum] pw.x not able to specity initial ion velocities
In-Reply-To: <20060823165328.68184.qmail@web52001.mail.yahoo.com>
References: <20060823165328.68184.qmail@web52001.mail.yahoo.com>
Message-ID: <200608232039.39216.giannozz@nest.sns.it>

On Wednesday 23 August 2006 18:53, Konstantin Kudin wrote:

> Consider these lines in read_cards.f90 :
>
>        ELSE IF ( TRIM(card) == 'ATOMIC_VELOCITIES' ) THEN
>           !
>           CALL card_ion_velocities( input_line )
>           IF ( ( prog == 'PW' .OR. prog == 'CP' ) .AND. ionode ) &
>              WRITE( stdout,'(A)') 'Warning: card '//trim(input_line)//' 
ignored'
>
>  What this piece of code is saying is that for either PW or CP the
> ATOMIC_VELOCITIES card is ignored (when is it supported then?).

with FPMD, which still runs if required (I think) and performs a few
things that either PW or CP do not yet suppport (I think). This is one :-)

Paolo
--
Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From marzari at MIT.EDU  Wed Aug 23 22:34:07 2006
From: marzari at MIT.EDU (Nicola Marzari)
Date: Wed, 23 Aug 2006 16:34:07 -0400
Subject: [Pw_forum] pw.x not able to specity initial ion velocities
In-Reply-To: <20060823165328.68184.qmail@web52001.mail.yahoo.com>
References: <20060823165328.68184.qmail@web52001.mail.yahoo.com>
Message-ID: <44ECBBBF.4010005@mit.edu>

A side comment: initializing velocities makes sense for a
born-oppenheimer simulation, but not for a car-parrinello one.

In the latter case, the electronic wavefunctions must acquire a
dynamics that follows quasi-adiabatically the ionic one. So, one wants
to avoid an abrupt start (i.e. ionic velocities different from zero)
since this would excite a number of frequencies in the
electronic evolution (that we do not want, would lower the quality of
the simulation, and that would take a lot of time to thermalize out).
	
			nicola


Konstantin Kudin wrote:
>  Actually, the velocity input appears to be almost purely unsupported.
> Consider these lines in read_cards.f90 :
> 
>        ELSE IF ( TRIM(card) == 'ATOMIC_VELOCITIES' ) THEN
>           !
>           CALL card_ion_velocities( input_line )
>           IF ( ( prog == 'PW' .OR. prog == 'CP' ) .AND. ionode ) &
>              WRITE( stdout,'(A)') 'Warning: card '//trim(input_line)//'
> ignored'
> 
>  What this piece of code is saying is that for either PW or CP the
> ATOMIC_VELOCITIES card is ignored (when is it supported then?). While
> this of course does not guarantee that this message is reliable, my own
> experience with trying to input velocities into CP seems to confirm the
> lack of such functionality.
> 
>  Kostya


-- 
---------------------------------------------------------------------
Prof Nicola Marzari   Department of Materials Science and Engineering
13-5066   MIT   77 Massachusetts Avenue   Cambridge MA 02139-4307 USA
tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu


From niuyingli at yahoo.com.cn  Wed Aug 23 22:56:40 2006
From: niuyingli at yahoo.com.cn (=?gb2312?q?=D3=A2=C0=FB=20=C5=A3?=)
Date: Thu, 24 Aug 2006 04:56:40 +0800 (CST)
Subject: [Pw_forum] About electron-phonon coupling constants calculation!
Message-ID: <20060823205640.4039.qmail@web15607.mail.cnb.yahoo.com>

Dear all,
  I have a question about electron-phonon coupling.
  Pwscf calculates electron-phonon coupling constants
for each q point one by one.
  But another way had been proposed in the article:
PRL 72,372(1994),

Linear-Response Calculations of Electron-Phonon
Interactions

  The authors selected 10 q-points to get the
IFCs(Interatomic Force constants) in the real space.
Then they used the IFCs to calculated 47 phonon
frequencies and egenvectors to get phonon line widths.
But when they calculated the electron-phonon
interaction matrix elements, this is no
self-consistent screening ptential \deltaVscf. So they
should "using a small setup" to deal with the problem!
  I don't know what they actually did. And I just want
to know can I calculate the Electron-Phonon
Interaction constants by this manner. If so, I can do
more dense grids on q space to calculate line widths
and spectral function.


	

	
		
___________________________________________________________ 
Mp3???-???????
http://music.yahoo.com.cn/?source=mail_mailbox_footer


From giannozz at nest.sns.it  Wed Aug 23 23:03:01 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Wed, 23 Aug 2006 23:03:01 +0200
Subject: [Pw_forum] segmentation fault
In-Reply-To: <52613.152.14.74.117.1155767643.squirrel@webmail.ncsu.edu>
References: <52613.152.14.74.117.1155767643.squirrel@webmail.ncsu.edu>
Message-ID: <200608232303.01951.giannozz@nest.sns.it>

On Thursday 17 August 2006 00:34, Vivek Ranjan wrote:

> I am running a job. The programme stops in the first scf iteration without
> any error message and writes the following on the stdout :
>
> Traceback for process 387481(msp mode, ssp 2) apid 387481.0 on node 0
>      cegterg_para at cegterg_+0x1578 (0x1059118) at cegterg.f90
>                   cegterg_+0x0488 (0x10657A8) at :127
>         c_bands_k at c_bands_+0x235C (0x103F61C) at :707
>                   c_bands_+0x04B4 (0x10413D4) at :105
>                 electrons_+0x1814 (0x112E0F4) at :269
>                     pwscf_+0x074C (0x1003A2C) at pwscf.f90:105
> Fault: Attempt to dereference null pointer: 0x0
> Segmentation fault
>
> Can someone please suggest what could be the problem ?

since you have such a nice traceback, please locate the line where
the error occurs. It looks like a deallocation of something that is not
allocated. It shouldn't be difficult to understand why

Paolo
-- 
Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From giannozz at nest.sns.it  Wed Aug 23 23:08:22 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Wed, 23 Aug 2006 23:08:22 +0200
Subject: [Pw_forum] electron-phonon coupling at a certain k point
In-Reply-To: <59012.202.41.111.151.1155529519.squirrel@202.41.111.151>
References: <59012.202.41.111.151.1155529519.squirrel@202.41.111.151>
Message-ID: <200608232308.22163.giannozz@nest.sns.it>

On Monday 14 August 2006 06:25, Tiju Thomas wrote:

>[...] I don't need to calculate the electron-phonon coupling for all the
> phonons. Is there a prescription that you can suggest to do this in the
> fastest manner (without doing the dense q point calculation suggested in
> example 7 for phonons)

you can do phonons on a grid of q vectors, or one q vector at the time.
The same applies to electron-phonon coupling coefficients. Calculations
at different q vectors are independent

Paolo
-- 
Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From eyvaz_isaev at yahoo.com  Thu Aug 24 01:06:13 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Wed, 23 Aug 2006 16:06:13 -0700 (PDT)
Subject: [Pw_forum] help: free energy
In-Reply-To: <BAY106-F288F2AD4E904650DF5229793470@phx.gbl>
Message-ID: <20060823230613.9599.qmail@web60325.mail.yahoo.com>

Dear Hailin, and Eduardo,

I can send you a version of fqha.f90 (completely
different of fqha.f90 in /pwtools, but I used unit
conversions) to calculate the free energy, C_v and S.
The idea behind is, sure, the same.
Please be a little bit patient because I would like be
sure that it can read (I am sure that it is OK, but
prefer to check) phonon DOS file provided by matdyn.x.
Usually I use phonon DOS calculated by my own
integration (by real tetrahedra method) program.

Bests,
Eyvaz.



--- hailin yu <yuhailin_79 at msn.com> wrote:

> Dear Eyvaz:
>       Dear Eyvaz, can you give me the codes to
> calculate thermodynamic 
> properties? I did't find the post of Stefano
> refered.
>     Thanks!
> 
> Dear all:
>      I give the phonon density of states of alpha-Sn
> which I attained using 
> PWSCF ( and I sure it's correctly) in the attach(
> The frequency unit is 
> cm-1). I have used the Equ(3)( Chem. Phys. Lett.
> 417,272-276(2005)) to 
> calculate the free energy and I am sure the units is
> right. But, as 
> previously mentioned, my results is too large
> compared with previously work. 
> So I hope someone can compute the free energy and
> tell what's wrong.
> Thanks a lot!
> yuhailin
> 
>
_________________________________________________________________
> Don't just search. Find. Check out the new MSN
> Search! 
>
http://search.msn.click-url.com/go/onm00200636ave/direct/01/
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From konstantin_kudin at yahoo.com  Thu Aug 24 01:27:22 2006
From: konstantin_kudin at yahoo.com (Konstantin Kudin)
Date: Wed, 23 Aug 2006 16:27:22 -0700 (PDT)
Subject: [Pw_forum] pw.x not able to specity initial ion velocities
In-Reply-To: <44ECBBBF.4010005@mit.edu>
Message-ID: <20060823232722.79129.qmail@web52008.mail.yahoo.com>

--- Nicola Marzari <marzari at MIT.EDU> wrote:

> 
> A side comment: initializing velocities makes sense for a
> born-oppenheimer simulation, but not for a car-parrinello one.

 Actually, it can make sense for CP as well (obviously, with some help
from the operator!).

> In the latter case, the electronic wavefunctions must acquire a
> dynamics that follows quasi-adiabatically the ionic one. So, one
> wants
> to avoid an abrupt start (i.e. ionic velocities different from zero)
> since this would excite a number of frequencies in the
> electronic evolution (that we do not want, would lower the quality of
> the simulation, and that would take a lot of time to thermalize out).

 This is true, however, thanks to the conjugate gradient (CG) [ which I
think actually originated from Nicola's group :-) ] one can do a couple
of CG steps with proper CP timestep and whatever initial velocities,
and then restart the CP from that. The electronic velocities in this
case are fully consistent with the moving ions.

 Also note that sometime ago I added a CP option to change the timestep
as well, which makes the code rescale the ionic velocities
proportionally such that the temperature is preserved. Again, here one
goes CP(dt1)->CG(dt2, some rescaling)->CP(dt2), and everything is nice,
consistent, and at the same temperature as before :-)

 I use both these features all the time.

 Kostya

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From duan at physics.usyd.edu.au  Thu Aug 24 07:00:58 2006
From: duan at physics.usyd.edu.au (Xiangmei Duan)
Date: Thu, 24 Aug 2006 15:00:58 +1000 (EST)
Subject: [Pw_forum] heat of formation for Mg3N2
In-Reply-To: <0B0B2C91-BED1-41E1-BA61-4D5B352BE928@sissa.it>
References: <Pine.LNX.4.61.0608230617070.20125@dawn.ts.infn.it>
 <9F608030-A2A2-4DD1-A5BF-8EE0A42D57C1@sissa.it>
 <Pine.LNX.4.62.0608231744350.9491@cmt.physics.usyd.edu.au>
 <0B0B2C91-BED1-41E1-BA61-4D5B352BE928@sissa.it>
Message-ID: <Pine.LNX.4.62.0608241454510.5021@cmt.physics.usyd.edu.au>

Dear Stefano,

I found the problem about the heat of formation for Mg3N2, which came from 
the bulk Mg. The binding energy I got 4.95/2 eV is too large comparing to 
other LDA calc. (~1.76-1.78 eV) and expt. vealus (1.51 eV).
My lattice parameter c is too large !!! - not fully optimized !

Now the binding energy for Mg bulk is 1.79 eV, and the heat of formation 
for Mg3N2 is -4.97 eV.

Sorry for my carelessness,
Best wishes,
Xiangmei




From baroni at sissa.it  Thu Aug 24 08:49:37 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Thu, 24 Aug 2006 08:49:37 +0200
Subject: [Pw_forum] heat of formation for Mg3N2
In-Reply-To: <Pine.LNX.4.62.0608241454510.5021@cmt.physics.usyd.edu.au>
References: <Pine.LNX.4.61.0608230617070.20125@dawn.ts.infn.it> <9F608030-A2A2-4DD1-A5BF-8EE0A42D57C1@sissa.it> <Pine.LNX.4.62.0608231744350.9491@cmt.physics.usyd.edu.au> <0B0B2C91-BED1-41E1-BA61-4D5B352BE928@sissa.it> <Pine.LNX.4.62.0608241454510.5021@cmt.physics.usyd.edu.au>
Message-ID: <7D656219-B196-4CE8-9072-4EEAB66F6888@sissa.it>

All's well that ends well. Still, the discrepancy between your calcs  
and Dmol3's seems still a bit too large ...
Stefano

On Aug 24, 2006, at 7:00 AM, Xiangmei Duan wrote:

>
> Dear Stefano,
>
> I found the problem about the heat of formation for Mg3N2, which  
> came from the bulk Mg. The binding energy I got 4.95/2 eV is too  
> large comparing to other LDA calc. (~1.76-1.78 eV) and expt. vealus  
> (1.51 eV).
> My lattice parameter c is too large !!! - not fully optimized !
>
> Now the binding energy for Mg bulk is 1.79 eV, and the heat of  
> formation for Mg3N2 is -4.97 eV.
>
> Sorry for my carelessness,
> Best wishes,
> Xiangmei
>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060824/f1db5162/attachment.htm 

From cyrille.barreteau at cea.fr  Thu Aug 24 10:09:57 2006
From: cyrille.barreteau at cea.fr (Cyrille Barreteau)
Date: Thu, 24 Aug 2006 10:09:57 +0200
Subject: [Pw_forum] Permanent Research Fellow position
Message-ID: <44ED5ED5.80902@cea.fr>

A permanent Research Fellow position is available in the theoretical 
group of /Modelisation of Surfaces, Interfaces and Nanostructures/ 
(MSIN) of the laboratory /Service de Physique et Chimie des Surfaces et 
Interfaces /(SPCSI), at CEA-Saclay (http://www-drecam.cea.fr/spcsi/), 
France.


*Profile:*
We are looking for an expert in ab-initio calculations (DFT-based 
methods), having a strong expertise in coding and development of codes 
and willing to apply such numerical techniques to the study  of 
realistic systems in relation to electronic transport and/or excitations 
in nano-objects (atomic contacts, single molecules, clusters).
The successful candidate will join the MSIN group and will be expected 
to strongly interact with the members of the group, for example by
(i) implementing new theoretical approaches developed in our group,
(ii) using ab-initio calculations to extract relevant parameters for 
model Hamiltonians,
(iii) developing codes for nanoscale electronic transport.
The successful candidate will then strengthen the existing research 
activities of our group and will also be expected to develop 
collaborations with experimental groups of SPCSI on the following topics:
- quantum transport in nanostructures and molecular nanojunctions,
- electronic and optical excitation spectroscopies in nanoscale objects,
- interaction and organization at the molecule/surface interface.

*Background: *
The applicants should hold a PhD, have postdoctoral research experience 
and should have a proven track-record of cutting-edge research (in 
theoretical condensed-matter physics / chemistry, or a related field), 
including an ability to interact effectively within a research team 
involving experimentalists as well as other theoreticians.

Interested candidates are expected to send:
    (1)   Full CV
    (2)   List of publications.

To Dr C. Barreteau (cyrille.barreteau at cea.fr) and Dr H. Ness 
(herve.ness at cea.fr).
Informal enquiries are welcome and may be addressed to C. Barreteau and 
H. Ness.

-- 
==================================================================

Cyrille Barreteau

CEA Saclay DSM/DRECAM/SPCSI

Batiment 462

91191 Gif sur Yvette Cedex FRANCE
~~~~~~~~~~~~~~~~~~~~~~~~
phone : +33 (0)1 69 08 29 51
fax : +33 (0)1 69 08 84 46
email : cyrille.barreteau at cea.fr

~~~~~~~~~~~~~~~~~~~~~~~~

http://www-drecam.cea.fr/Images/Pisp/cbarreteau/cbarreteau_fr.html

==================================================================



From yuhailin_79 at msn.com  Thu Aug 24 10:52:03 2006
From: yuhailin_79 at msn.com (hailin yu)
Date: Thu, 24 Aug 2006 16:52:03 +0800
Subject: [Pw_forum] help: free energy
In-Reply-To: <89F7EE8F-F787-4F3F-B5F7-9AF421920798@sissa.it>
Message-ID: <BAY106-F2182A73C28E22E5CFA61F893440@phx.gbl>

Dear Stefano B.,
       Thank you for you reply. I have run the code (which you give me  in 
the attach) onto my date and the results are well good agreement with your 
paper (PRB,57,10421(1998)).
       But I am still confused about the code. In the 
Equ(1)(PRB,57,10421(1998)), there is "h" in the
ln[2sinh(h*omega/kB*T)] part. But I did't find "h" appear in the code. So 
can you tell me why "h"
did't appear in the code.
     Thank you very much!
regards!
     yuhailin

_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today it's FREE! 
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/



From lanhaiping at gmail.com  Thu Aug 24 14:33:18 2006
From: lanhaiping at gmail.com (lan haiping)
Date: Thu, 24 Aug 2006 20:33:18 +0800
Subject: [Pw_forum] phonon at Gamma point
In-Reply-To: <20060820145956.23780.qmail@web60312.mail.yahoo.com>
References: <c92002fa0608191743h458f45f5ubf539f93e99ed772@mail.gmail.com>
	 <20060820145956.23780.qmail@web60312.mail.yahoo.com>
Message-ID: <c92002fa0608240533w2f74a466kecce4697ac0ec0f2@mail.gmail.com>

Dear Eyvaz :

I just relaxed Anatase TiO2  with *ibrav= 7 , *
then output gave that Only 8 symmetry Operations are found.
According  to your mail, this structure should have 16 Symmetry Operations,
so
how can i obtain correct  symmetry operations ?

Input file is below , thank you in advance !


*&CONTROL
             title = 'TiO2' ,
       calculation = 'relax',
         verbosity = 'default',
           tprnfor = .true.,
            outdir = '/home/haiping/tmp/',
            prefix = 'tio2-1',
           disk_io = 'default',
             nstep = 200,
        pseudo_dir = '/home/haiping/espresso/pseudo/',
/*

*&SYSTEM
                ibrav= 7,
               celldm(1)=7.1356,celldm(3)=2.51218,
               nat = 6 ,
              ntyp = 2 ,
              nbnd = 30 ,
           ecutwfc = 30.0000000000,
       occupations = 'fixed' ,
*

*&ELECTRONS
  electron_maxstep = 100,
          conv_thr = 1.0e-09,
       mixing_mode = 'plain' ,
       startingpot = 'atomic' ,
       startingwfc = 'atomic' ,
       mixing_beta = 0.300000000,
       mixing_ndim = 8,
   diagonalization = 'david_overlap' ,
  diago_david_ndim = 8,
/*

*&IONS
             ion_dynamics = 'bfgs',
             pot_extrapolation = 'first_order',
             wfc_extrapolation = 'first_order',*

*/*

*O  15.999  O.pw91-van_ak.UPF
Ti 47.867  Ti.pw91-nsp-van.UPF
ATOMIC_POSITIONS angstrom
Ti       0.000000000   0.000000000   0.000000000
Ti       0.000000000   1.888000000   2.371500000
O        0.000000000   0.000000000  -1.983097630
O        0.000000000   0.000000000   1.983097630
O        0.000000000   1.888000000   0.387796645
O        0.000000000   1.888000000   4.355203355
K_POINTS   automatic
6 6  7   0 0 0
*
*

* On 8/20/06, Eyvaz Isaev <eyvaz_isaev at yahoo.com> wrote:
>
> OK, let me hint more.
>
> Anatase has a Pearson symbol tI12 or space group
> I4_1/amd. So, ibrav should be 7, but you have chosen a
> monoclinic lattice, i.e. ibrav=14. The space group has
> 16 symmetry elements, but your choice after correcting
> ecutrho gives only 4 symmetry elements .
>
> Imagine  FCC lattice and its conventional basis
> vectors now. According to you now we have to put
> ibrav=4 or 5. But ibrav=2 for FCC lattice.
>
> In addition, I also checked your input file via
> XCrysDen before I replied previously. Yes, there is no
> problem with it, but a figure does not look like
> anatase though  you used  nonzero ibrav. XcrysDen
> usually replicates atomic positions  if  ibrav.neq.0.
> For example, if you put ibrav=0  for  FCC cell and
> specify conventional CELL_PARAMETERS you will find  a
> rhombohedron, but using ibrav=2 and  default basis
> vectors you find a cubic cell.
>
> Bests,
> Eyvaz.
>
>
> --- lan haiping <lanhaiping at gmail.com> wrote:
>
> > Dear Eyvaz,
> > Thank you.
> > I checked the strcuture again using XCRYSDEN,
> > and found no problems related to the  lattices.
> >
> > Regards,
> >
> > Hai-Ping
> >
> >
> >
> > On 8/20/06, Eyvaz Isaev <eyvaz_isaev at yahoo.com>
> > wrote:
> > >
> > > Hi,
> > >
> > > It seems your lattice choice is not correct.
> > Please
> > > have a look at
> > > http://cst-www.nrl.navy.mil/lattice/struk/c5.html
> > >
> > > Bests,
> > > Eyvaz.
> > >
> > > --- lan haiping <lanhaiping at gmail.com> wrote:
> > >
> > > > Dear All,
> > > > I also have calculated gamma phonon frequencies
> > of
> > > > anatase TiO2 .
> > > > I run ph.x caculation after relaxing the
> > structure,
> > > > but  9 of 18 frequencies
> > > > are minus.
> > > >
> > > > Would you please give me some hints and help?
> > > >
> > > > input settings are below, thank you in advance !
> > > >
> > > > *phonons of TiO2 at Gamma
> > > >  &inputph
> > > >   tr2_ph=1.0d-14,
> > > >   recover=.true.,
> > > >   epsil=.true.,
> > > >   prefix='tio2',
> > > >   fildyn='TiO2-gamma.dyn',
> > > >   outdir='/home/haiping/tmp/'
> > > >  /
> > > > 0.0 0.0 0.0*
> > > > **
> > > > *Relaxing calculation input :*
> > > > *&CONTROL
> > > >              title = 'TiO2' ,
> > > >        calculation = 'relax',
> > > >          verbosity = 'default',
> > > >            tprnfor = .true.,
> > > >             outdir = '/home/haiping/tmp/',
> > > >             prefix = 'tio2',
> > > >            disk_io = 'default',
> > > >              nstep = 200,
> > > >         pseudo_dir =
> > > > '/home/haiping/espresso/pseudo/',
> > > > /
> > > > &SYSTEM
> > > >                 ibrav= 14,
> > > >
> > > > celldm(1)=10.2856,celldm(2)=1.0,celldm(3)=1.0,
> > > >
> > > > celldm(4)=-0.759357,celldm(5)=-0.759357,
> > > >                celldm(6)=0.518714
> > > >                nat = 6 ,
> > > >               ntyp = 2 ,
> > > >               nbnd = 30 ,
> > > >            ecutwfc = 30.0000000000,
> > > >            ecutrho = 160.000,
> > > >        occupations = 'fixed' ,
> > > > /
> > > >
> > > >
> > > > &ELECTRONS
> > > >   electron_maxstep = 100,
> > > >           conv_thr = 1.0e-07,
> > > >        mixing_mode = 'plain' ,
> > > >        startingpot = 'atomic' ,
> > > >        startingwfc = 'atomic' ,
> > > >        mixing_beta = 0.300000000,
> > > >        mixing_ndim = 8,
> > > >    diagonalization = 'david_overlap' ,
> > > >   diago_david_ndim = 8,
> > > > /
> > > >
> > > > &IONS
> > > >              ion_dynamics = 'bfgs',
> > > >              pot_extrapolation = 'first_order',
> > > >              wfc_extrapolation = 'first_order',
> > > >
> > > > /
> > > >
> > > > ATOMIC_SPECIES
> > > > O  15.999 O.pw91-van_ak.UPF
> > > > Ti 47.867  Ti.pw91-nsp-van.UPF
> > > > ATOMIC_POSITIONS crystal
> > > > Ti    0.00000   0.00000   0.00000
> > > > Ti    0.75000   0.25000   0.50000
> > > > O     0.20800   0.20800  -0.00000
> > > > O     0.95800   0.45800   0.50000
> > > > O     0.54200   0.04200   0.50000
> > > > O     0.79200   0.79200  -0.00000
> > > > K_POINTS   automatic
> > > > 6 6 7   0 0 0
> > > >
> > > > *
> > > >
> > > >
> > > > On 8/19/06, Ezad Shojaee
> > <ezadshojaee at hotmail.com>
> > > > wrote:
> > > > >
> > > > > hi
> > > > > i am trying to calculate phonon frequencies at
> > > > gamma point of
> > > > > TiO2(anatase)
> > > > > and i have relaxed the structure enough (i
> > think
> > > > )but performing PH.Xafter
> > > > > scf, 2of 18 frequencies  are minus!
> > > > > i want to know that what technical problem do
> > i
> > > > have? does anyone
> > > > > calculate
> > > > > these beore?
> > > > > any suggestion will be appreciated
> > > > >
> > > > >
> > > >
> > >
>
> --- lan haiping <lanhaiping at gmail.com> wrote:
>
> > Dear Eyvaz,
> > Thank you.
> > I checked the strcuture again using XCRYSDEN,
> > and found no problems related to the  lattices.
> >
> > Regards,
> >
> > Hai-Ping
> >
> >
> >
> > On 8/20/06, Eyvaz Isaev <eyvaz_isaev at yahoo.com>
> > wrote:
> > >
> > > Hi,
> > >
> > > It seems your lattice choice is not correct.
> > Please
> > > have a look at
> > > http://cst-www.nrl.navy.mil/lattice/struk/c5.html
> > >
> > > Bests,
> > > Eyvaz.
> > >
> > > --- lan haiping <lanhaiping at gmail.com> wrote:
> > >
> > > > Dear All,
> > > > I also have calculated gamma phonon frequencies
> > of
> > > > anatase TiO2 .
> > > > I run ph.x caculation after relaxing the
> > structure,
> > > > but  9 of 18 frequencies
> > > > are minus.
> > > >
> > > > Would you please give me some hints and help?
> > > >
> > > > input settings are below, thank you in advance !
> > > >
> > > > *phonons of TiO2 at Gamma
> > > >  &inputph
> > > >   tr2_ph=1.0d-14,
> > > >   recover=.true.,
> > > >   epsil=.true.,
> > > >   prefix='tio2',
> > > >   fildyn='TiO2-gamma.dyn',
> > > >   outdir='/home/haiping/tmp/'
> > > >  /
> > > > 0.0 0.0 0.0*
> > > > **
> > > > *Relaxing calculation input :*
> > > > *&CONTROL
> > > >              title = 'TiO2' ,
> > > >        calculation = 'relax',
> > > >          verbosity = 'default',
> > > >            tprnfor = .true.,
> > > >             outdir = '/home/haiping/tmp/',
> > > >             prefix = 'tio2',
> > > >            disk_io = 'default',
> > > >              nstep = 200,
> > > >         pseudo_dir =
> > > > '/home/haiping/espresso/pseudo/',
> > > > /
> > > > &SYSTEM
> > > >                 ibrav= 14,
> > > >
> > > > celldm(1)=10.2856,celldm(2)=1.0,celldm(3)=1.0,
> > > >
> > > > celldm(4)=-0.759357,celldm(5)=-0.759357,
> > > >                celldm(6)=0.518714
> > > >                nat = 6 ,
> > > >               ntyp = 2 ,
> > > >               nbnd = 30 ,
> > > >            ecutwfc = 30.0000000000,
> > > >            ecutrho = 160.000,
> > > >        occupations = 'fixed' ,
> > > > /
> > > >
> > > >
> > > > &ELECTRONS
> > > >   electron_maxstep = 100,
> > > >           conv_thr = 1.0e-07,
> > > >        mixing_mode = 'plain' ,
> > > >        startingpot = 'atomic' ,
> > > >        startingwfc = 'atomic' ,
> > > >        mixing_beta = 0.300000000,
> > > >        mixing_ndim = 8,
> > > >    diagonalization = 'david_overlap' ,
> > > >   diago_david_ndim = 8,
> > > > /
> > > >
> > > > &IONS
> > > >              ion_dynamics = 'bfgs',
> > > >              pot_extrapolation = 'first_order',
> > > >              wfc_extrapolation = 'first_order',
> > > >
> > > > /
> > > >
> > > > ATOMIC_SPECIES
> > > > O  15.999 O.pw91-van_ak.UPF
> > > > Ti 47.867  Ti.pw91-nsp-van.UPF
> > > > ATOMIC_POSITIONS crystal
> > > > Ti    0.00000   0.00000   0.00000
> > > > Ti    0.75000   0.25000   0.50000
> > > > O     0.20800   0.20800  -0.00000
> > > > O     0.95800   0.45800   0.50000
> > > > O     0.54200   0.04200   0.50000
> > > > O     0.79200   0.79200  -0.00000
> > > > K_POINTS   automatic
> > > > 6 6 7   0 0 0
> > > >
> > > > *
> > > >
> > > >
> > > > On 8/19/06, Ezad Shojaee
> > <ezadshojaee at hotmail.com>
> > > > wrote:
> > > > >
> > > > > hi
> > > > > i am trying to calculate phonon frequencies at
> > > > gamma point of
> > > > > TiO2(anatase)
> > > > > and i have relaxed the structure enough (i
> > think
> > > > )but performing PH.Xafter
> > > > > scf, 2of 18 frequencies  are minus!
> > > > > i want to know that what technical problem do
> > i
> > > > have? does anyone
> > > > > calculate
> > > > > these beore?
> > > > > any suggestion will be appreciated
> > > > >
> > > > >
> > > >
> > >
> >
> _________________________________________________________________
> > > > > Express yourself instantly with MSN Messenger!
> > > > Download today it's FREE!
> > > > >
> > > >
> > >
> >
> http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
> > > > >
> > > > >
> > _______________________________________________
> > > > > Pw_forum mailing list
> > > > > Pw_forum at pwscf.org
> > > > >
> > http://www.democritos.it/mailman/listinfo/pw_forum
> > > > >
> > > >
> > >
> > >
> > > __________________________________________________
> > > Do You Yahoo!?
> > > Tired of spam?  Yahoo! Mail has the best spam
> > protection around
> > > http://mail.yahoo.com
> > > _______________________________________________
> > > Pw_forum mailing list
> > > Pw_forum at pwscf.org
> > > http://www.democritos.it/mailman/listinfo/pw_forum
> > >
> >
>
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060824/1510893b/attachment.htm 

From baroni at sissa.it  Thu Aug 24 14:46:12 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Thu, 24 Aug 2006 14:46:12 +0200
Subject: [Pw_forum] help: free energy
In-Reply-To: <BAY106-F2182A73C28E22E5CFA61F893440@phx.gbl>
References: <BAY106-F2182A73C28E22E5CFA61F893440@phx.gbl>
Message-ID: <52357D86-EFE1-406A-8490-65E6956E9515@sissa.it>

Because it is included in the various conversion constants (cm^-1->eV- 
 >cal, etc.)  If it was not, it would be weird to measure energy (or  
even frequencies!) in cm^-1.

Please, take the time to think by yourself to the small code and make  
sure you understand each and every line, particularly the only non  
obvious (but trivial nevertheless) ones concering units. It took me  
some time to figure out the value of all the constants, so do not  
feel too bad if it will aslo take some time to you!

Take care - Stefano

On Aug 24, 2006, at 10:52 AM, hailin yu wrote:

> Dear Stefano B.,
>       Thank you for you reply. I have run the code (which you give  
> me  in the attach) onto my date and the results are well good  
> agreement with your paper (PRB,57,10421(1998)).
>       But I am still confused about the code. In the Equ(1)(PRB, 
> 57,10421(1998)), there is "h" in the
> ln[2sinh(h*omega/kB*T)] part. But I did't find "h" appear in the  
> code. So can you tell me why "h"
> did't appear in the code.
>     Thank you very much!
> regards!
>     yuhailin
>
> _________________________________________________________________
> Express yourself instantly with MSN Messenger! Download today it's  
> FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060824/d3fc6a98/attachment.htm 

From baroni at sissa.it  Thu Aug 24 14:56:15 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Thu, 24 Aug 2006 14:56:15 +0200
Subject: [Pw_forum] phonon at Gamma point
In-Reply-To: <c92002fa0608240533w2f74a466kecce4697ac0ec0f2@mail.gmail.com>
References: <c92002fa0608191743h458f45f5ubf539f93e99ed772@mail.gmail.com> <20060820145956.23780.qmail@web60312.mail.yahoo.com> <c92002fa0608240533w2f74a466kecce4697ac0ec0f2@mail.gmail.com>
Message-ID: <375B31B1-211F-4533-A347-0CA12CBD4699@sissa.it>

I have an exercise for you. Try to solve yourself your problem by  
browsing the archive of this mailing list, where this very same   
question has been asked and answered several times.

Also, I would like to urge everybody here to make a serious effort to  
browse the archive _before_ posting a request. I think it is sign of  
respect for people who have already taken the time to help others,  
and will also help us being helpful to those who have problems not  
previously met by anybody.

Just as a hint: you problem has to do with non-symmorphic symmetry  
groups and with the way symmetry operations are found out in the  
code ...

Hope this helps

Stefano B.

On Aug 24, 2006, at 2:33 PM, lan haiping wrote:

>
> Dear Eyvaz :
>
> I just relaxed Anatase TiO2  with ibrav= 7 ,
> then output gave that Only 8 symmetry Operations are found.
> According  to your mail, this structure should have 16 Symmetry  
> Operations, so
> how can i obtain correct  symmetry operations ?
>
> Input file is below , thank you in advance !
>
> &CONTROL
>              title = 'TiO2' ,
>        calculation = 'relax',
>          verbosity = 'default',
>            tprnfor = .true.,
>             outdir = '/home/haiping/tmp/',
>             prefix = 'tio2-1',
>            disk_io = 'default',
>              nstep = 200,
>         pseudo_dir = '/home/haiping/espresso/pseudo/',
> /
>
> &SYSTEM
>                 ibrav= 7,
>                celldm(1)=7.1356,celldm(3)=2.51218,
>                nat = 6 ,
>               ntyp = 2 ,
>               nbnd = 30 ,
>            ecutwfc = 30.0000000000 ,
>        occupations = 'fixed' ,
>
> &ELECTRONS
>   electron_maxstep = 100,
>           conv_thr = 1.0e-09,
>        mixing_mode = 'plain' ,
>        startingpot = 'atomic' ,
>        startingwfc = 'atomic' ,
>        mixing_beta = 0.300000000 ,
>        mixing_ndim = 8,
>    diagonalization = 'david_overlap' ,
>   diago_david_ndim = 8,
> /
>
> &IONS
>              ion_dynamics = 'bfgs',
>              pot_extrapolation = 'first_order',
>              wfc_extrapolation = 'first_order',
>
> /
>
> O  15.999  O.pw91-van_ak.UPF
> Ti 47.867  Ti.pw91-nsp-van.UPF
> ATOMIC_POSITIONS angstrom
> Ti       0.000000000   0.000000000   0.000000000
> Ti       0.000000000   1.888000000   2.371500000
> O        0.000000000    0.000000000  -1.983097630
> O        0.000000000   0.000000000   1.983097630
> O        0.000000000   1.888000000   0.387796645
> O        0.000000000   1.888000000   4.355203355
> K_POINTS   automatic
> 6 6  7   0 0 0
>
>
>
> On 8/20/06, Eyvaz Isaev <eyvaz_isaev at yahoo.com> wrote:
> OK, let me hint more.
>
> Anatase has a Pearson symbol tI12 or space group
> I4_1/amd. So, ibrav should be 7, but you have chosen a
> monoclinic lattice, i.e. ibrav=14. The space group has
> 16 symmetry elements, but your choice after correcting
> ecutrho gives only 4 symmetry elements .
>
> Imagine  FCC lattice and its conventional basis
> vectors now. According to you now we have to put
> ibrav=4 or 5. But ibrav=2 for FCC lattice.
>
> In addition, I also checked your input file via
> XCrysDen before I replied previously. Yes, there is no
> problem with it, but a figure does not look like
> anatase though  you used  nonzero ibrav. XcrysDen
> usually replicates atomic positions  if  ibrav.neq.0.
> For example, if you put ibrav=0  for  FCC cell and
> specify conventional CELL_PARAMETERS you will find  a
> rhombohedron, but using ibrav=2 and  default basis
> vectors you find a cubic cell.
>
> Bests,
> Eyvaz.
>
>
> --- lan haiping <lanhaiping at gmail.com> wrote:
>
> > Dear Eyvaz,
> > Thank you.
> > I checked the strcuture again using XCRYSDEN,
> > and found no problems related to the  lattices.
> >
> > Regards,
> >
> > Hai-Ping
> >
> >
> >
> > On 8/20/06, Eyvaz Isaev <eyvaz_isaev at yahoo.com >
> > wrote:
> > >
> > > Hi,
> > >
> > > It seems your lattice choice is not correct.
> > Please
> > > have a look at
> > > http://cst-www.nrl.navy.mil/lattice/struk/c5.html
> > >
> > > Bests,
> > > Eyvaz.
> > >
> > > --- lan haiping <lanhaiping at gmail.com> wrote:
> > >
> > > > Dear All,
> > > > I also have calculated gamma phonon frequencies
> > of
> > > > anatase TiO2 .
> > > > I run ph.x caculation after relaxing the
> > structure,
> > > > but  9 of 18 frequencies
> > > > are minus.
> > > >
> > > > Would you please give me some hints and help?
> > > >
> > > > input settings are below, thank you in advance !
> > > >
> > > > *phonons of TiO2 at Gamma
> > > >  &inputph
> > > >   tr2_ph=1.0d-14,
> > > >   recover=.true.,
> > > >   epsil=.true.,
> > > >   prefix='tio2',
> > > >   fildyn='TiO2-gamma.dyn',
> > > >   outdir='/home/haiping/tmp/'
> > > >  /
> > > > 0.0 0.0 0.0*
> > > > **
> > > > *Relaxing calculation input :*
> > > > *&CONTROL
> > > >              title = 'TiO2' ,
> > > >        calculation = 'relax',
> > > >          verbosity = 'default',
> > > >            tprnfor = .true.,
> > > >             outdir = '/home/haiping/tmp/',
> > > >             prefix = 'tio2',
> > > >            disk_io = 'default',
> > > >              nstep = 200,
> > > >         pseudo_dir =
> > > > '/home/haiping/espresso/pseudo/',
> > > > /
> > > > &SYSTEM
> > > >                 ibrav= 14,
> > > >
> > > > celldm(1)=10.2856,celldm(2)=1.0,celldm(3)=1.0,
> > > >
> > > > celldm(4)=- 0.759357,celldm(5)=-0.759357,
> > > >                celldm(6)=0.518714
> > > >                nat = 6 ,
> > > >               ntyp = 2 ,
> > > >               nbnd = 30 ,
> > > >            ecutwfc = 30.0000000000,
> > > >            ecutrho = 160.000,
> > > >        occupations = 'fixed' ,
> > > > /
> > > >
> > > >
> > > > &ELECTRONS
> > > >   electron_maxstep = 100,
> > > >           conv_thr = 1.0e-07,
> > > >        mixing_mode = 'plain' ,
> > > >        startingpot = 'atomic' ,
> > > >        startingwfc = 'atomic' ,
> > > >        mixing_beta = 0.300000000,
> > > >        mixing_ndim = 8,
> > > >    diagonalization = 'david_overlap' ,
> > > >   diago_david_ndim = 8,
> > > > /
> > > >
> > > > &IONS
> > > >              ion_dynamics = 'bfgs',
> > > >              pot_extrapolation = 'first_order',
> > > >              wfc_extrapolation = 'first_order',
> > > >
> > > > /
> > > >
> > > > ATOMIC_SPECIES
> > > > O  15.999 O.pw91-van_ak.UPF
> > > > Ti 47.867  Ti.pw91-nsp-van.UPF
> > > > ATOMIC_POSITIONS crystal
> > > > Ti    0.00000   0.00000    0.00000
> > > > Ti    0.75000   0.25000   0.50000
> > > > O     0.20800   0.20800  -0.00000
> > > > O     0.95800   0.45800   0.50000
> > > > O     0.54200   0.04200   0.50000
> > > > O     0.79200   0.79200  -0.00000
> > > > K_POINTS   automatic
> > > > 6 6 7   0 0 0
> > > >
> > > > *
> > > >
> > > >
> > > > On 8/19/06, Ezad Shojaee
> > <ezadshojaee at hotmail.com>
> > > > wrote:
> > > > >
> > > > > hi
> > > > > i am trying to calculate phonon frequencies at
> > > > gamma point of
> > > > > TiO2(anatase)
> > > > > and i have relaxed the structure enough (i
> > think
> > > > )but performing PH.Xafter
> > > > > scf, 2of 18 frequencies  are minus!
> > > > > i want to know that what technical problem do
> > i
> > > > have? does anyone
> > > > > calculate
> > > > > these beore?
> > > > > any suggestion will be appreciated
> > > > >
> > > > >
> > > >
> > >
>
> --- lan haiping <lanhaiping at gmail.com> wrote:
>
> > Dear Eyvaz,
> > Thank you.
> > I checked the strcuture again using XCRYSDEN,
> > and found no problems related to the  lattices.
> >
> > Regards,
> >
> > Hai-Ping
> >
> >
> >
> > On 8/20/06, Eyvaz Isaev < eyvaz_isaev at yahoo.com>
> > wrote:
> > >
> > > Hi,
> > >
> > > It seems your lattice choice is not correct.
> > Please
> > > have a look at
> > > http://cst-www.nrl.navy.mil/lattice/struk/c5.html
> > >
> > > Bests,
> > > Eyvaz.
> > >
> > > --- lan haiping < lanhaiping at gmail.com> wrote:
> > >
> > > > Dear All,
> > > > I also have calculated gamma phonon frequencies
> > of
> > > > anatase TiO2 .
> > > > I run ph.x caculation after relaxing the
> > structure,
> > > > but  9 of 18 frequencies
> > > > are minus.
> > > >
> > > > Would you please give me some hints and help?
> > > >
> > > > input settings are below, thank you in advance !
> > > >
> > > > *phonons of TiO2 at Gamma
> > > >  &inputph
> > > >   tr2_ph=1.0d-14,
> > > >   recover=.true.,
> > > >   epsil=.true.,
> > > >   prefix='tio2',
> > > >   fildyn='TiO2-gamma.dyn',
> > > >   outdir='/home/haiping/tmp/'
> > > >  /
> > > > 0.0 0.0 0.0*
> > > > **
> > > > *Relaxing calculation input :*
> > > > *&CONTROL
> > > >              title = 'TiO2' ,
> > > >        calculation = 'relax',
> > > >          verbosity = 'default',
> > > >            tprnfor = .true.,
> > > >             outdir = '/home/haiping/tmp/',
> > > >             prefix = 'tio2',
> > > >            disk_io = 'default',
> > > >              nstep = 200,
> > > >         pseudo_dir =
> > > > '/home/haiping/espresso/pseudo/',
> > > > /
> > > > &SYSTEM
> > > >                 ibrav= 14,
> > > >
> > > > celldm(1)= 10.2856,celldm(2)=1.0,celldm(3)=1.0,
> > > >
> > > > celldm(4)=-0.759357,celldm(5)=-0.759357,
> > > >                celldm(6)=0.518714
> > > >                nat = 6 ,
> > > >               ntyp = 2 ,
> > > >               nbnd = 30 ,
> > > >            ecutwfc = 30.0000000000,
> > > >            ecutrho = 160.000,
> > > >        occupations = 'fixed' ,
> > > > /
> > > >
> > > >
> > > > &ELECTRONS
> > > >   electron_maxstep = 100,
> > > >           conv_thr = 1.0e-07,
> > > >        mixing_mode = 'plain' ,
> > > >        startingpot = 'atomic' ,
> > > >        startingwfc = 'atomic' ,
> > > >        mixing_beta = 0.300000000,
> > > >        mixing_ndim = 8,
> > > >    diagonalization = 'david_overlap' ,
> > > >   diago_david_ndim = 8,
> > > > /
> > > >
> > > > &IONS
> > > >              ion_dynamics = 'bfgs',
> > > >              pot_extrapolation = 'first_order',
> > > >              wfc_extrapolation = 'first_order',
> > > >
> > > > /
> > > >
> > > > ATOMIC_SPECIES
> > > > O  15.999 O.pw91-van_ak.UPF
> > > > Ti 47.867  Ti.pw91-nsp-van.UPF
> > > > ATOMIC_POSITIONS crystal
> > > > Ti    0.00000   0.00000   0.00000
> > > > Ti    0.75000   0.25000   0.50000
> > > > O     0.20800   0.20800  -0.00000
> > > > O     0.95800   0.45800   0.50000
> > > > O     0.54200   0.04200   0.50000
> > > > O     0.79200   0.79200  -0.00000
> > > > K_POINTS   automatic
> > > > 6 6 7   0 0 0
> > > >
> > > > *
> > > >
> > > >
> > > > On 8/19/06, Ezad Shojaee
> > <ezadshojaee at hotmail.com>
> > > > wrote:
> > > > >
> > > > > hi
> > > > > i am trying to calculate phonon frequencies at
> > > > gamma point of
> > > > > TiO2(anatase)
> > > > > and i have relaxed the structure enough (i
> > think
> > > > )but performing PH.Xafter
> > > > > scf, 2of 18 frequencies  are minus!
> > > > > i want to know that what technical problem do
> > i
> > > > have? does anyone
> > > > > calculate
> > > > > these beore?
> > > > > any suggestion will be appreciated
> > > > >
> > > > >
> > > >
> > >
> >
> _________________________________________________________________
> > > > > Express yourself instantly with MSN Messenger!
> > > > Download today it's FREE!
> > > > >
> > > >
> > >
> >
> http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
> > > > >
> > > > >
> > _______________________________________________
> > > > > Pw_forum mailing list
> > > > > Pw_forum at pwscf.org
> > > > >
> > http://www.democritos.it/mailman/listinfo/pw_forum
> > > > >
> > > >
> > >
> > >
> > > __________________________________________________
> > > Do You Yahoo!?
> > > Tired of spam?  Yahoo! Mail has the best spam
> > protection around
> > > http://mail.yahoo.com
> > > _______________________________________________
> > > Pw_forum mailing list
> > > Pw_forum at pwscf.org
> > > http://www.democritos.it/mailman/listinfo/pw_forum
> > >
> >
>
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060824/8e228622/attachment.htm 

From yi.kong at gmail.com  Thu Aug 24 15:29:36 2006
From: yi.kong at gmail.com (Yi Kong)
Date: Thu, 24 Aug 2006 21:29:36 +0800
Subject: [Pw_forum] why the phonon eigendisplacement are so much different?
Message-ID: <8c9dc66c0608240629m4b2e482ev13ae2324d29d0e96@mail.gmail.com>

Dear all,
     I want to know the atomic displacement and vibrational amplitude under
different pressures in alloys, for example AB3. At the gamma point, the file
AB3.dynG contains the calculated results, part of the acoustic branches are
shown as following, and I know from the mailing list that after the omega
line is the eigendisplacement of each atom.

**********************************AB3.dynG*********************
.............
     omega( 1) =       1.321104 [THz] =      44.067563 [cm-1]
 (  0.054100  0.000000  0.104646  0.000000 -0.485673  0.000000 )
 (  0.054107  0.000000  0.104658  0.000000 -0.486512  0.000000 )
 (  0.054107  0.000000  0.104827  0.000000 -0.485728  0.000000 )
 (  0.054194  0.000000  0.104658  0.000000 -0.485728  0.000000 )
     omega( 2) =       1.321104 [THz] =      44.067563 [cm-1]
 ( -0.230668  0.000000 -0.427404  0.000000 -0.117786  0.000000 )
 ( -0.230694  0.000000 -0.427453  0.000000 -0.117990  0.000000 )
 ( -0.230694  0.000000 -0.428142  0.000000 -0.117800  0.000000 )
 ( -0.231067  0.000000 -0.427453  0.000000 -0.117800  0.000000 )
     omega( 3) =       1.321104 [THz] =      44.067563 [cm-1]
 ( -0.440024  0.000000  0.236919  0.000000  0.002033  0.000000 )
 ( -0.440074  0.000000  0.236946  0.000000  0.002036  0.000000 )
 ( -0.440074  0.000000  0.237328  0.000000  0.002033  0.000000 )
 ( -0.440784  0.000000  0.236946  0.000000  0.002033  0.000000 )
........
************************************************************************************
While when we perform Fourier transformation with q2r.x, we can get the
interatomic force constants in real space. Then with program matdyn.x, we
can recalculate phonon modes at any q. The results of acoustic branches at
gamma point, are also shown as following (with the acoustic sum rule
applied):

*****************************matdyn.modes***********************
.......................
     omega( 1) =       0.000000 [THz] =      -0.000013 [cm-1]
 ( -0.008792   0.000000     0.001394   0.000000     0.499921   0.000000   )
 ( -0.008792   0.000000     0.001394   0.000000     0.499921   0.000000   )
 ( -0.008792   0.000000     0.001394   0.000000     0.499921   0.000000   )
 ( -0.008792   0.000000     0.001394   0.000000     0.499921   0.000000   )
     omega( 2) =       0.000000 [THz] =       0.000016 [cm-1]
 ( -0.027902   0.000000     0.499217   0.000000    -0.001883   0.000000   )
 ( -0.027902   0.000000     0.499217   0.000000    -0.001883   0.000000   )
 ( -0.027902   0.000000     0.499217   0.000000    -0.001883   0.000000   )
 ( -0.027902   0.000000     0.499217   0.000000    -0.001883   0.000000   )
     omega( 3) =       0.000001 [THz] =       0.000023 [cm-1]
 ( -0.499143   0.000000    -0.027930   0.000000    -0.008701   0.000000   )
 ( -0.499143   0.000000    -0.027930   0.000000    -0.008701   0.000000   )
 ( -0.499143   0.000000    -0.027930   0.000000    -0.008701   0.000000   )
 ( -0.499143   0.000000    -0.027930   0.000000    -0.008701   0.000000   )
***************************************************************************************

My question is that between these two results of eigendisplacement, why they
are so much different and which one is reasonable. Beside, another question
is that, could I square these eigendisplacement, respectively and plus them
together to obtain the vibrational amplitude (that is, SQRT[x^2 + y^2 + z^2]
) .

Thanks in advance!

Sincerely

Kong

-- 
=============================
Yi Kong
Building 23, Room 240
Department of MS&E
Tsinghua University
PR China, 100084

Phone:86-10-6278-1255
Fax:    86-10-6278-1255
Email: kongy03 at mails.tsinghua.edu.cn
or       yi.kong at gmail.com
=============================
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060824/d0b1d17e/attachment.htm 

From stewart at cnf.cornell.edu  Thu Aug 24 16:03:14 2006
From: stewart at cnf.cornell.edu (stewart at cnf.cornell.edu)
Date: Thu, 24 Aug 2006 10:03:14 -0400
Subject: [Pw_forum] Publications resulting from PWscf/Quantum Espresso
In-Reply-To: <8c9dc66c0608240629m4b2e482ev13ae2324d29d0e96@mail.gmail.com>
References: <8c9dc66c0608240629m4b2e482ev13ae2324d29d0e96@mail.gmail.com>
Message-ID: <20060824140315.9721.qmail@xuxa.iecc.com>

Hi all, 

  I was wondering if it would be possible to set up a page on the PWscf or 
Quantum Espresso site where people could post references to papers done with 
the package.  I think this would provide a great resource for people 
starting out with the code and also highlight the strengths of the approach. 
This could probably be done in a wiki format with some level of control 
implemented. 

Best regards, 

Derek 

################################
Derek Stewart, Ph. D.
Scientific Computation Associate
250 Duffield Hall
Cornell Nanoscale Facility (CNF)
Ithaca, NY 14853
stewart (at) cnf.cornell.edu
(607) 255-2856 



From eyvaz_isaev at yahoo.com  Thu Aug 24 16:05:48 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Thu, 24 Aug 2006 07:05:48 -0700 (PDT)
Subject: [Pw_forum] phonon at Gamma point
In-Reply-To: <c92002fa0608240533w2f74a466kecce4697ac0ec0f2@mail.gmail.com>
Message-ID: <20060824140548.46812.qmail@web60323.mail.yahoo.com>

Hi,
 
> I just relaxed Anatase TiO2  with *ibrav= 7 , *
> then output gave that Only 8 symmetry Operations are
> found.
Because of that your setup is not consistent. 
Your lattice parameter is
7.1356*0.529177=3.7759954012, but 2 Oxygen atoms have
X-component (in units of alat) of 0.5. 
You put 1.888, but, 1.888 is not equal 
3.7759954012/2=1.8879977. So, they could not be
considered the same within 10^{-6} which is a default
mismatch (please correct me if I am wrong). For sure
you have to use 7  digits (after point) in such a kind
specific cases, as c/a, b/a (and cosines, presumably).
Besides, your atomic positions are completely
different  of these specified in 
http://cst-www.nrl.navy.mil/lattice/struk.xmol/c5.pos
Please use lattice parameters and atomic coordinates
(in Angstrom) as they are in the link, and report what
happens. 

I am not sure that it is a good idea to relax a well
known structure if you have a problem with its
symmetry.

Bests,
Eyvaz.

> According  to your mail, this structure should have
> 16 Symmetry Operations,
> so
> how can i obtain correct  symmetry operations ?
> 
> Input file is below , thank you in advance !
> 
> 
> *&CONTROL
>              title = 'TiO2' ,
>        calculation = 'relax',
>          verbosity = 'default',
>            tprnfor = .true.,
>             outdir = '/home/haiping/tmp/',
>             prefix = 'tio2-1',
>            disk_io = 'default',
>              nstep = 200,
>         pseudo_dir =
> '/home/haiping/espresso/pseudo/',
> /*
> 
> *&SYSTEM
>                 ibrav= 7,
>                celldm(1)=7.1356,celldm(3)=2.51218,
>                nat = 6 ,
>               ntyp = 2 ,
>               nbnd = 30 ,
>            ecutwfc = 30.0000000000,
>        occupations = 'fixed' ,
> *
> 
> *&ELECTRONS
>   electron_maxstep = 100,
>           conv_thr = 1.0e-09,
>        mixing_mode = 'plain' ,
>        startingpot = 'atomic' ,
>        startingwfc = 'atomic' ,
>        mixing_beta = 0.300000000,
>        mixing_ndim = 8,
>    diagonalization = 'david_overlap' ,
>   diago_david_ndim = 8,
> /*
> 
> *&IONS
>              ion_dynamics = 'bfgs',
>              pot_extrapolation = 'first_order',
>              wfc_extrapolation = 'first_order',*
> 
> */*
> 
> *O  15.999  O.pw91-van_ak.UPF
> Ti 47.867  Ti.pw91-nsp-van.UPF
> ATOMIC_POSITIONS angstrom
> Ti       0.000000000   0.000000000   0.000000000
> Ti       0.000000000   1.888000000   2.371500000
> O        0.000000000   0.000000000  -1.983097630
> O        0.000000000   0.000000000   1.983097630
> O        0.000000000   1.888000000   0.387796645
> O        0.000000000   1.888000000   4.355203355
> K_POINTS   automatic
> 6 6  7   0 0 0
> *
> *
> 
> * On 8/20/06, Eyvaz Isaev <eyvaz_isaev at yahoo.com>
> wrote:
> >
> > OK, let me hint more.
> >
> > Anatase has a Pearson symbol tI12 or space group
> > I4_1/amd. So, ibrav should be 7, but you have
> chosen a
> > monoclinic lattice, i.e. ibrav=14. The space group
> has
> > 16 symmetry elements, but your choice after
> correcting
> > ecutrho gives only 4 symmetry elements .
> >
> > Imagine  FCC lattice and its conventional basis
> > vectors now. According to you now we have to put
> > ibrav=4 or 5. But ibrav=2 for FCC lattice.
> >
> > In addition, I also checked your input file via
> > XCrysDen before I replied previously. Yes, there
> is no
> > problem with it, but a figure does not look like
> > anatase though  you used  nonzero ibrav. XcrysDen
> > usually replicates atomic positions  if 
> ibrav.neq.0.
> > For example, if you put ibrav=0  for  FCC cell and
> > specify conventional CELL_PARAMETERS you will find
>  a
> > rhombohedron, but using ibrav=2 and  default basis
> > vectors you find a cubic cell.
> >
> > Bests,
> > Eyvaz.
> >
> >
> > --- lan haiping <lanhaiping at gmail.com> wrote:
> >
> > > Dear Eyvaz,
> > > Thank you.
> > > I checked the strcuture again using XCRYSDEN,
> > > and found no problems related to the  lattices.
> > >
> > > Regards,
> > >
> > > Hai-Ping
> > >
> > >
> > >
> > > On 8/20/06, Eyvaz Isaev <eyvaz_isaev at yahoo.com>
> > > wrote:
> > > >
> > > > Hi,
> > > >
> > > > It seems your lattice choice is not correct.
> > > Please
> > > > have a look at
> > > >
> http://cst-www.nrl.navy.mil/lattice/struk/c5.html
> > > >
> > > > Bests,
> > > > Eyvaz.
> > > >
> > > > --- lan haiping <lanhaiping at gmail.com> wrote:
> > > >
> > > > > Dear All,
> > > > > I also have calculated gamma phonon
> frequencies
> > > of
> > > > > anatase TiO2 .
> > > > > I run ph.x caculation after relaxing the
> > > structure,
> > > > > but  9 of 18 frequencies
> > > > > are minus.
> > > > >
> > > > > Would you please give me some hints and
> help?
> > > > >
> > > > > input settings are below, thank you in
> advance !
> > > > >
> > > > > *phonons of TiO2 at Gamma
> > > > >  &inputph
> > > > >   tr2_ph=1.0d-14,
> > > > >   recover=.true.,
> > > > >   epsil=.true.,
> > > > >   prefix='tio2',
> > > > >   fildyn='TiO2-gamma.dyn',
> > > > >   outdir='/home/haiping/tmp/'
> > > > >  /
> > > > > 0.0 0.0 0.0*
> > > > > **
> > > > > *Relaxing calculation input :*
> > > > > *&CONTROL
> > > > >              title = 'TiO2' ,
> > > > >        calculation = 'relax',
> > > > >          verbosity = 'default',
> > > > >            tprnfor = .true.,
> > > > >             outdir = '/home/haiping/tmp/',
> > > > >             prefix = 'tio2',
> > > > >            disk_io = 'default',
> > > > >              nstep = 200,
> > > > >         pseudo_dir =
> > > > > '/home/haiping/espresso/pseudo/',
> > > > > /
> > > > > &SYSTEM
> > > > >                 ibrav= 14,
> > > > >
> > > > >
> celldm(1)=10.2856,celldm(2)=1.0,celldm(3)=1.0,
> > > > >
> > > > > celldm(4)=-0.759357,celldm(5)=-0.759357,
> > > > >                celldm(6)=0.518714
> > > > >                nat = 6 ,
> > > > >               ntyp = 2 ,
> > > > >               nbnd = 30 ,
> > > > >            ecutwfc = 30.0000000000,
> > > > >            ecutrho = 160.000,
> > > > >        occupations = 'fixed' ,
> > > > > /
> > > > >
> > > > >
> > > > > &ELECTRONS
> > > > >   electron_maxstep = 100,
> > > > >           conv_thr = 1.0e-07,
> > > > >        mixing_mode = 'plain' ,
> > > > >        startingpot = 'atomic' ,
> > > > >        startingwfc = 'atomic' ,
> > > > >        mixing_beta = 0.300000000,
> > > > >        mixing_ndim = 8,
> > > > >    diagonalization = 'david_overlap' ,
> > > > >   diago_david_ndim = 8,
> > > > > /
> 
=== message truncated ===


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From giannozz at nest.sns.it  Thu Aug 24 16:10:48 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Thu, 24 Aug 2006 16:10:48 +0200
Subject: [Pw_forum] Using bands.x to extra band structure info
In-Reply-To: <20060814160733.15723.qmail@xuxa.iecc.com>
References: <44da25f7.363.3f1a.23677@ic.sunysb.edu> <00a701c6bcb6$34006a30$b7305ba0@ornl.gov> <20060814160733.15723.qmail@xuxa.iecc.com>
Message-ID: <200608241610.49802.giannozz@nest.sns.it>

On Monday 14 August 2006 18:07, stewart at cnf.cornell.edu wrote:

>    I would like to use bands.x to extract the band structure data from a
> nscf calculation.  However, when I try to run it, it states that it needs
> the potential file.

I don't think it really needs it, but it uses routines that read the data
files, including the potential, and complain if it is not there

> How do I insure this is generated in the nscf calculation.

it is not generated in the non-scf calculation: it should be there
from the scf calculation (more exactly, it used to be there, since
now only the charge density is written and used)

Paolo
-- 
Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From janietwen at gmail.com  Fri Aug 25 00:13:04 2006
From: janietwen at gmail.com (Janiet Wen)
Date: Fri, 25 Aug 2006 06:13:04 +0800
Subject: [Pw_forum] About WanT calculation for large system
In-Reply-To: <392e95900608241508w6ad8cd55y62217afa18d57bb7@mail.gmail.com>
References: <392e95900608241508w6ad8cd55y62217afa18d57bb7@mail.gmail.com>
Message-ID: <392e95900608241513v7d20481cn8efd4db55a857c7b@mail.gmail.com>

 Dear Prof Marzari and developers :

        Recently, I read  two papers about quantum conductance calculation
based on MLWFs schem for real-space hamiltonian ,i.e. cond-mat/0608146
&PRL95 076804. I did find these two simulations gave quantum conductances
for large systems, .i.e . longest CNT is about 40 nm, which is of thousands
atoms.
     As far as i know, the number of  atoms  that  most  PP-PW codes  can
deal with is ~ 200. So , I think  the hamiltonian for obtaining quantum
conductance of these larger system is not likely to be calculated with PP-PW
method directly , since MLWFs are just post-processing tool for  electronic
structure according to the mannual. There must be some tricks for building
these large systems' real space hamiltonian, though i donot know how to.
Then, I tried to read most original papers about Green functions method and
MLWFs , Below is my understanding of this two simualtions , would you please
give me some comments ?
     For small systems, based on PW calculations ,  we can get full
hamiltonian of the super cell.  Just after distentanglement and localization
procedures, MLWFs are obtained for this supercell.  Then, We can obtain the
real-space hamiltonian using MLWFs as minimal basis set.  With  this
hamiltonian with overlap matrix , we can calculated Quantum conductance of
this system.
    While for lage systems, I think most calculations are similar to the
small system , except the procedure of  real-space hamiltonian buliding .
Since i donot catch the ideas of this step, I just guess that we can bulid
the real-space hamiltonian of large system with  the real-space hamiltonian
of  smaller systems . But i donot know much about this procedure. Would you
please help me out ? Maybe it is trival , but it is really a big barrier
for me to "tunnel ".


  Best wishes,

 Janiet
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060825/8ed09b8a/attachment.htm 

From marzari at MIT.EDU  Fri Aug 25 00:35:57 2006
From: marzari at MIT.EDU (Nicola Marzari)
Date: Thu, 24 Aug 2006 18:35:57 -0400
Subject: [Pw_forum] About WanT calculation for large system
In-Reply-To: <392e95900608241513v7d20481cn8efd4db55a857c7b@mail.gmail.com>
References: <392e95900608241508w6ad8cd55y62217afa18d57bb7@mail.gmail.com> <392e95900608241513v7d20481cn8efd4db55a857c7b@mail.gmail.com>
Message-ID: <44EE29CD.5050409@mit.edu>



Dear Janiet,


your understanding is absolutely correct - we will email you a
copy of Young-Su Lee PhD, where these steps are spelt out
in greater detail (and you should look at the 1999 PRB of Marco 
Buongiorno Nardelli, and at the 2004 PRB of Arrigo Calzolari et al).

Note that the procedure we use is slightly different from the WanT
approach (2004 PRB cited above), in that it deals with the
some subtleties of Gamma sampling, both in the disentanglement (the
disentangled space is complex, even if the kohn-sham orbitals are real),
and in the calculation of the matrix elements (you need to calculate 
overlaps in real space, truncating the tails). This Gamma-specific
code is not yet in www.wannier.org, although it is making its way there.

> Since i donot catch the ideas of this step, I just guess that we can 
> bulid the real-space hamiltonian of large system with  the real-space 
> hamiltonian of  smaller systems . But i donot know much about this 
> procedure. Would you please help me out ? Maybe it is trival , but it is 
> really a big barrier  for me to "tunnel ".

First of all, your systems in PP-PW are always enormous - infinite.
Their unit cells can be large or small - you want unit cells, for this
problem, large enough so that the matrix elements of the Hamiltonian
in between MLWFs that sit more than a unit cell apart are zero (for
the nanotubes, unit cells of 100-200 atoms).

Now, think about the Hamiltonian of the very large system (maybe not
even periodic), expressed in a basis of Wannier functions. Now, most
matrix elements will be zero - whenever the distance of the centers is
large. So, the only non-zero elements will be between MLWFs that are
reasonably close to each other. What are those matrix elements ? Well, 
you can just calculate them in a system whose unit cell is 100-200 atoms.

This is much better explained with a drwaing - that's why I'm sending
Y-S thesis - but I'm sure the general idea (that is a tight-binding 
idea, if you think at it) is in Marco Nardelli paper (and many others).


				nicola



-- 
---------------------------------------------------------------------
Prof Nicola Marzari   Department of Materials Science and Engineering
13-5066   MIT   77 Massachusetts Avenue   Cambridge MA 02139-4307 USA
tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu


From marzari at MIT.EDU  Fri Aug 25 00:41:56 2006
From: marzari at MIT.EDU (Nicola Marzari)
Date: Thu, 24 Aug 2006 18:41:56 -0400
Subject: [Pw_forum] PWSCF Netiquette: Please use some form of name / affiliation
In-Reply-To: <392e95900608241513v7d20481cn8efd4db55a857c7b@mail.gmail.com>
References: <392e95900608241508w6ad8cd55y62217afa18d57bb7@mail.gmail.com> <392e95900608241513v7d20481cn8efd4db55a857c7b@mail.gmail.com>
Message-ID: <44EE2B34.9060501@mit.edu>


Dear All,


a modest proposal - it would be great if people were to identify
themselves more explictly in these postings. This helps a lot in
understanding the user base, the environment, and just in getting to
know you personally.

Also, we all feel better in discussing with a person, than with an
email alias. Once I discovered that a poster was working in my same 
building...

We all have concerns about privacy, and e-mail spam. To avoid that,
you can concoct simple rules so that your email address cannot
be harvested (e.g.  marzari at mit.edu could be transformed in
mit at marzari.edu, or anything like this), use a spamscreen, or
still use an email from a generic account, but at least put
your name and affiliation.
	
			nicola

---------------------------------------------------------------------
Prof Nicola Marzari   Department of Materials Science and Engineering
13-5066   MIT   77 Massachusetts Avenue   Cambridge MA 02139-4307 USA
tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu


From akohlmey at cmm.upenn.edu  Fri Aug 25 01:09:01 2006
From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer)
Date: Thu, 24 Aug 2006 19:09:01 -0400
Subject: [Pw_forum] PWSCF Netiquette: Please use some form of name / affiliation
In-Reply-To: <44EE2B34.9060501@mit.edu>
References: <392e95900608241508w6ad8cd55y62217afa18d57bb7@mail.gmail.com>
	 <392e95900608241513v7d20481cn8efd4db55a857c7b@mail.gmail.com>
	 <44EE2B34.9060501@mit.edu>
Message-ID: <7b6913e90608241609v78666de8ub7f25c853e188647@mail.gmail.com>

On 8/24/06, Nicola Marzari <marzari at mit.edu> wrote:
>
>
> Dear All,
>
>
> a modest proposal - it would be great if people were to identify
> themselves more explictly in these postings. This helps a lot in
> understanding the user base, the environment, and just in getting to
> know you personally.

seconded.

> Also, we all feel better in discussing with a person, than with an
> email alias. Once I discovered that a poster was working in my same
> building...

furthermore, using a real name increases the chances to get a
serious reply. it is just human nature that one cares more about
somebody who behaves like a real person than some abstract.

> We all have concerns about privacy, and e-mail spam. To avoid that,
> you can concoct simple rules so that your email address cannot
> be harvested (e.g.  marzari at mit.edu could be transformed in
> mit at marzari.edu, or anything like this), use a spamscreen, or
> still use an email from a generic account, but at least put
> your name and affiliation.

absolutely. i consider using the real name and affiliation part of
the 'contract' of working with free (scientific) software. after all,
if people publish a paper with their work, they don't use an
alias for _that_, right?

finally, this correlates well, with derek's idea of having a collection
of papers published that use the codes in the QE package.
apart from the usefulness for newcomers, this list could be
extremely helpful in securing funding for people working on
QE as developers, as more usage and publications demonstrate
the usefulness of the work to improve the code.

this way everybody benefits.

best regards,
    axel

>
>                         nicola
>
> ---------------------------------------------------------------------
> Prof Nicola Marzari   Department of Materials Science and Engineering
> 13-5066   MIT   77 Massachusetts Avenue   Cambridge MA 02139-4307 USA
> tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
>


-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
  Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.


From marzari at MIT.EDU  Fri Aug 25 01:43:35 2006
From: marzari at MIT.EDU (Nicola Marzari)
Date: Thu, 24 Aug 2006 19:43:35 -0400
Subject: [Pw_forum] Publications resulting from PWscf/Quantum Espresso
In-Reply-To: <20060824140315.9721.qmail@xuxa.iecc.com>
References: <8c9dc66c0608240629m4b2e482ev13ae2324d29d0e96@mail.gmail.com> <20060824140315.9721.qmail@xuxa.iecc.com>
Message-ID: <44EE39A7.4090102@mit.edu>

Dear Derek,

brilliant idea.

I am working in this very days to the wiki of Espresso - I will announce
in more detail project et al to everyone soon. But it's there, and will
soon take shape:

http://www.quantum-espresso.org/wiki/index.php/Main_Page

(comments are welcome, to the list, not personal).

I've created a Bibliography section, and inside it there is a Quantum
Espresso Application menu, divided year by year.

As soon as the wiki is operative, people will be able to add
their contribution into this (we need to have some instructions
on a common format, and the like).

			nicola



stewart at cnf.cornell.edu wrote:
> Hi all,
>  I was wondering if it would be possible to set up a page on the PWscf 
> or Quantum Espresso site where people could post references to papers 
> done with the package.  I think this would provide a great resource for 
> people starting out with the code and also highlight the strengths of 
> the approach. This could probably be done in a wiki format with some 
> level of control implemented.
> Best regards,
> Derek
> ################################
> Derek Stewart, Ph. D.
> Scientific Computation Associate
> 250 Duffield Hall
> Cornell Nanoscale Facility (CNF)
> Ithaca, NY 14853
> stewart (at) cnf.cornell.edu
> (607) 255-2856
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum


-- 
---------------------------------------------------------------------
Prof Nicola Marzari   Department of Materials Science and Engineering
13-5066   MIT   77 Massachusetts Avenue   Cambridge MA 02139-4307 USA
tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu


From yuhailin_79 at msn.com  Fri Aug 25 03:57:28 2006
From: yuhailin_79 at msn.com (hailin yu)
Date: Fri, 25 Aug 2006 09:57:28 +0800
Subject: [Pw_forum] help: free energy
In-Reply-To: <52357D86-EFE1-406A-8490-65E6956E9515@sissa.it>
Message-ID: <BAY106-F204677FDB2C3EC4DA312BB93450@phx.gbl>

Dear Stefano,
      Thank you very much!
regards!
yuhailin

_________________________________________________________________
Don't just search. Find. Check out the new MSN Search! 
http://search.msn.click-url.com/go/onm00200636ave/direct/01/



From hushujun at mail.sdu.edu.cn  Fri Aug 25 07:45:15 2006
From: hushujun at mail.sdu.edu.cn (Shujun Hu)
Date: Fri, 25 Aug 2006 13:45:15 +0800
Subject: [Pw_forum] Code starting trouble on the PC Cluster
Message-ID: <356484715.09115@mail.sdu.edu.cn>

Hi Axel,
It is my pleasure to receive your advise and I am sorry for the ambiguous
description of the problem. 
  Neither the input file nor the configuration process of the code is iffy. My
trouble is to start the program, not running it. When i try to start the program
on 4 nodes of the cluster (each node contains 2 core as in Pentium-D, so 8 process
will be started) and type the command:
# mpiexec -n 8 pw.x -npool 4<job.inp
nothing appears on the screen even waiting for one night! Then I found it should
be typed again and again:

# mpiexec -n 8 pw.x -npool 4<job.inp
nothing on the screen. Then type:

# mpiexec -n 8 pw.x -npool 4<job.inp
nothing on the screen. Then type:

......

# mpiexec -n 8 pw.x -npool 4<job.inp

     Program PWSCF     v.3.0    starts ...
     Today is 14May2006 at 13: 0:51 
     Parallel version (MPI)
     Number of processors in use:       8
......

Surprisingly it is OK!!! The following process is good and it can achieve the
convergency. I have checked the process queue and found that now lots of process
named pw.x under python were sleeping. I donnt know exactly what python is. Maybe
it is part of the mpich programm. So i guess the problem comes from the connecting
of the nodes. 
  By the way,  another problem also puzzles me. There are 6 nodes in my cluster
platform. So the most efficient way is to run 12 processes. However, it seems that
the command:

# mpiexec -n 12 pw.x -npool 6<job.inp              #command-1

will sleep forever. But the following one:

# mpiexec -n 16 pw.x -npool 8<job.inp              #command-2

runs well. The FFT grid is the default setting of 75*75*50 and K-points is set to
be 4*4*6. Considering the symmetry the number of the K-points is 60 (given by the
output file). In the user's guide the "Parallelization issues" part says the
number of processors Np=Npk*Npr. The pools (Npk, 6 or 8 in the upper commands)
should be a divisor of the number of k-points. Npr, here set to be 2, certainly
meet the requirement that it should be a divisor of the FFT grid of z axis. But
Npk is strange. It seems that command-1 is more correct, at least better, than
command-2. However, the fact is opposite. I donnt know why.
  Any suggestion will be appreciated. Thanks.

                                       Shujun




From baroni at sissa.it  Fri Aug 25 10:43:30 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Fri, 25 Aug 2006 10:43:30 +0200
Subject: [Pw_forum] Publications resulting from PWscf/Quantum Espresso
In-Reply-To: <44EE39A7.4090102@mit.edu>
References: <8c9dc66c0608240629m4b2e482ev13ae2324d29d0e96@mail.gmail.com> <20060824140315.9721.qmail@xuxa.iecc.com> <44EE39A7.4090102@mit.edu>
Message-ID: <62C003E3-8A56-4330-A3C3-E95BCCE2396A@sissa.it>

Dear Derek and Nicola: GREAT!
I was hesitating to reply to Derek and to support his idea, waiting  
for the first occasion to talk to Nicola.
Nic has been swifter than me!
Very good, and thanks a lot.
Stefano

On Aug 25, 2006, at 1:43 AM, Nicola Marzari wrote:

>
> Dear Derek,
>
> brilliant idea.
>
> I am working in this very days to the wiki of Espresso - I will  
> announce
> in more detail project et al to everyone soon. But it's there, and  
> will
> soon take shape:
>
> http://www.quantum-espresso.org/wiki/index.php/Main_Page
>
> (comments are welcome, to the list, not personal).
>
> I've created a Bibliography section, and inside it there is a Quantum
> Espresso Application menu, divided year by year.
>
> As soon as the wiki is operative, people will be able to add
> their contribution into this (we need to have some instructions
> on a common format, and the like).
>
> 			nicola
>
>
>
> stewart at cnf.cornell.edu wrote:
>> Hi all,
>>  I was wondering if it would be possible to set up a page on the  
>> PWscf or Quantum Espresso site where people could post references  
>> to papers done with the package.  I think this would provide a  
>> great resource for people starting out with the code and also  
>> highlight the strengths of the approach. This could probably be  
>> done in a wiki format with some level of control implemented.
>> Best regards,
>> Derek
>> ################################
>> Derek Stewart, Ph. D.
>> Scientific Computation Associate
>> 250 Duffield Hall
>> Cornell Nanoscale Facility (CNF)
>> Ithaca, NY 14853
>> stewart (at) cnf.cornell.edu
>> (607) 255-2856
>> _______________________________________________
>> Pw_forum mailing list
>> Pw_forum at pwscf.org
>> http://www.democritos.it/mailman/listinfo/pw_forum
>
>
> -- 
> ---------------------------------------------------------------------
> Prof Nicola Marzari   Department of Materials Science and Engineering
> 13-5066   MIT   77 Massachusetts Avenue   Cambridge MA 02139-4307 USA
> tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060825/1720be5d/attachment.htm 

From konstantin_kudin at yahoo.com  Fri Aug 25 18:11:31 2006
From: konstantin_kudin at yahoo.com (Konstantin Kudin)
Date: Fri, 25 Aug 2006 09:11:31 -0700 (PDT)
Subject: [Pw_forum] PWSCF Netiquette: Please use some form of name / affiliation
In-Reply-To: <7b6913e90608241609v78666de8ub7f25c853e188647@mail.gmail.com>
Message-ID: <20060825161131.35441.qmail@web52009.mail.yahoo.com>

--- Axel Kohlmeyer <akohlmey at cmm.upenn.edu> wrote:

> On 8/24/06, Nicola Marzari <marzari at mit.edu> wrote:
> >
> >
> > Dear All,
> >
> >
> > a modest proposal - it would be great if people were to identify
> > themselves more explictly in these postings. This helps a lot in
> > understanding the user base, the environment, and just in getting
> to
> > know you personally.
> 
> seconded.

 I disagree with this proposal. It seems that the choice of whether to
reveal one's affiliation or not should definitely be with the poster,
as it has been so far.

 In my view, if the points raised in a given posting are valid, then
those can be considered on their own merit without any knowledge of the
poster's affiliation (or even real name). After all, we all gain from
the improved understanding of the different aspects of our
software/methods.

 Alternatively, if a posting shows that there is little background in
the field, then again affiliation is not really necessary, since
generic pointers (such as "read the Solid State by xyz") do not require
any idea about user's background.

 Kostya

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From d.schmicker at physik.de  Fri Aug 25 18:38:34 2006
From: d.schmicker at physik.de (Detlef Schmicker)
Date: Fri, 25 Aug 2006 18:38:34 +0200
Subject: [Pw_forum] Units in plots
Message-ID: <1156523914.5201.4.camel@localhost>

Hello,

I just started using pwscf. Which unit is the 

total potential plotted if I use pp.x

I expect Ryd but I can not find any information in mail archive and
documentation.

Thanks a lot

Detlef Schmicker


&INPUTPP
                      prefix = 'al' ,
                      outdir = '/home/detlef/espresso-3.1.1/GUI/TMP/' ,
                     filplot = 'rho' ,
                    plot_num = 2,
 /
 &PLOT
                       nfile = 1 ,
                   filepp(1) = '/home/detlef/espresso-3.1.1/GUI/rho',
                   weight(1) = 1.0,
                     fileout = 'al.plot.dat' ,
                       iflag = 1 ,
               output_format = 0 ,
                       e1(1) = 1.0,
                       e1(2) = 1.0,
                       e1(3) = 0.0,
                       x0(1) = 0.0,
                       x0(2) = 0.0,
                       x0(3) = 0.0,
                          nx = 1000 ,
 /



-- 
Detlef Schmicker <d.schmicker at physik.de>



From baroni at sissa.it  Fri Aug 25 18:45:26 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Fri, 25 Aug 2006 18:45:26 +0200
Subject: [Pw_forum] PWSCF Netiquette: Please use some form of name / affiliation
In-Reply-To: <20060825161131.35441.qmail@web52009.mail.yahoo.com>
References: <20060825161131.35441.qmail@web52009.mail.yahoo.com>
Message-ID: <296EEE4F-2FDC-4F5F-A03B-835251DC5473@sissa.it>


On Aug 25, 2006, at 6:11 PM, Konstantin Kudin wrote:

> --- Axel Kohlmeyer <akohlmey at cmm.upenn.edu> wrote:
>
>> On 8/24/06, Nicola Marzari <marzari at mit.edu> wrote:
>>>
>>>
>>> Dear All,
>>>
>>>
>>> a modest proposal - it would be great if people were to identify
>>> themselves more explictly in these postings. This helps a lot in
>>> understanding the user base, the environment, and just in getting
>> to
>>> know you personally.
>>
>> seconded.
>
>  I disagree with this proposal. It seems that the choice of whether to
> reveal one's affiliation or not should definitely be with the poster,
> as it has been so far.

I disagree with your disagreement. I think that the point raised by  
Nicola is mainly a point of politeness that, as such, must not (nor  
could it) be enforced. There may well be very good reasons not be  
polite in some circumstances (even personal sensitivity: what sounds  
polite to me may not sound so to others!). Personally, I will  
continue to reply to posts that I find interesting whether or not  
they are "properly signed". I am sure that Nicola will do the same.  
Nevertheless, there is nothing wrong in letting other people know  
that for some of us signed correspondence is far more pleasant than  
anonymous one. Nothing more, nothing less than this.

>  In my view, if the points raised in a given posting are valid, then
> those can be considered on their own merit without any knowledge of  
> the
> poster's affiliation (or even real name). After all, we all gain from
> the improved understanding of the different aspects of our
> software/methods.

Sure. For this reason nobody thinks to enforce a formal signature.  
But I think that most of the subscribers to this mailing list will  
not object to sign their correspondence. If some of them ("of us")  
will prefer to continue to use nicknames, well ... please do!


Stefano

>
>  Alternatively, if a posting shows that there is little background in
> the field, then again affiliation is not really necessary, since
> generic pointers (such as "read the Solid State by xyz") do not  
> require
> any idea about user's background.
>
>  Kostya
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060825/c38288ea/attachment.htm 

From konstantin_kudin at yahoo.com  Fri Aug 25 20:03:53 2006
From: konstantin_kudin at yahoo.com (Konstantin Kudin)
Date: Fri, 25 Aug 2006 11:03:53 -0700 (PDT)
Subject: [Pw_forum] Publications resulting from PWscf/Quantum Espresso
In-Reply-To: <62C003E3-8A56-4330-A3C3-E95BCCE2396A@sissa.it>
Message-ID: <20060825180353.76227.qmail@web52005.mail.yahoo.com>

 Hi all,

 While were are on the subject I think it is also worthwhile to
consider setting up a list of references on the methods included into
QE (as opposed just to the applications alone). 

 The purpose would be 2-fold. First, to give a precise pointer for the
detailed description of a given method. And second, to facilitate
proper citations of the QE developers' papers, such as any time a given
approach is used for applications, the corresponding work by a QE
person is cited.

 Kostya

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From marzari at MIT.EDU  Fri Aug 25 20:38:06 2006
From: marzari at MIT.EDU (Nicola Marzari)
Date: Fri, 25 Aug 2006 14:38:06 -0400
Subject: [Pw_forum] Publications resulting from PWscf/Quantum Espresso
In-Reply-To: <20060825180353.76227.qmail@web52005.mail.yahoo.com>
References: <20060825180353.76227.qmail@web52005.mail.yahoo.com>
Message-ID: <44EF438E.9030803@mit.edu>

Yes ! I had put a "Reference papers" section yesterday, and
somehow, automagically, they are already appearing (Derek ?).

Nicola Bonini (bonini at mit.edu) is in charge of opening wiki
accounts - please email him.

As mentioned, I'm still working on some planning and homework for
all, so there should be a more detailed mail on the wiki and how to 
contorbute to it in the next few days.

			nicola



Konstantin Kudin wrote:
>  Hi all,
> 
>  While were are on the subject I think it is also worthwhile to
> consider setting up a list of references on the methods included into
> QE (as opposed just to the applications alone). 
> 
>  The purpose would be 2-fold. First, to give a precise pointer for the
> detailed description of a given method. And second, to facilitate
> proper citations of the QE developers' papers, such as any time a given
> approach is used for applications, the corresponding work by a QE
> person is cited.
> 
>  Kostya
> 
-- 
---------------------------------------------------------------------
Prof Nicola Marzari   Department of Materials Science and Engineering
13-5066   MIT   77 Massachusetts Avenue   Cambridge MA 02139-4307 USA
tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu


From stewart at cnf.cornell.edu  Fri Aug 25 21:20:50 2006
From: stewart at cnf.cornell.edu (stewart at cnf.cornell.edu)
Date: Fri, 25 Aug 2006 15:20:50 -0400
Subject: [Pw_forum] Re: Publications resulting from PWscf/Quantum Espresso
In-Reply-To: <44EF438E.9030803@mit.edu>
References: <20060825180353.76227.qmail@web52005.mail.yahoo.com>
            <44EF438E.9030803@mit.edu>
Message-ID: <20060825192050.12346.qmail@xuxa.iecc.com>

Hi all, 

I agree that a reference section is important.  I've had a chance to put a 
few papers in on the wiki page including the review on DFPT.  For now, I am 
going with the following format for papers, 

Paper title (hyperlinked to the doi identifier for the paper), Authors names 
(first letter, last name), Full title of journal, volume, page number, 
(year). 

Most journals now provide a doi or digital object identifier near the title 
and abstract for a paper.  For example, Baroni et al's review on DFPT in 
Reviews of Modern Physics has the DOI: 10.1103/RevModPhys.73.515.  To 
provide a link to that paper all you have to do is put this after the url 

http://dx.doi.org/ 

so you get 

http://dx.doi.org/10.1103/RevModPhys.73.515 

The nice thing about this approach is that it should preserve the link even 
if the journal switches websites. 

Best regards, 

Derek 

 

Nicola Marzari writes: 

> 
> Yes ! I had put a "Reference papers" section yesterday, and
> somehow, automagically, they are already appearing (Derek ?). 
> 
> Nicola Bonini (bonini at mit.edu) is in charge of opening wiki
> accounts - please email him. 
> 
> As mentioned, I'm still working on some planning and homework for
> all, so there should be a more detailed mail on the wiki and how to 
> contorbute to it in the next few days. 
> 
> 			nicola 
> 
>  
> 
> Konstantin Kudin wrote:
>>  Hi all, 
>> 
>>  While were are on the subject I think it is also worthwhile to
>> consider setting up a list of references on the methods included into
>> QE (as opposed just to the applications alone).  
>> 
>>  The purpose would be 2-fold. First, to give a precise pointer for the
>> detailed description of a given method. And second, to facilitate
>> proper citations of the QE developers' papers, such as any time a given
>> approach is used for applications, the corresponding work by a QE
>> person is cited. 
>> 
>>  Kostya 
>> 
> -- 
> ---------------------------------------------------------------------
> Prof Nicola Marzari   Department of Materials Science and Engineering
> 13-5066   MIT   77 Massachusetts Avenue   Cambridge MA 02139-4307 USA
> tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
 


################################
Derek Stewart, Ph. D.
Scientific Computation Associate
250 Duffield Hall
Cornell Nanoscale Facility (CNF)
Ithaca, NY 14853
stewart (at) cnf.cornell.edu
(607) 255-2856 



From akohlmey at cmm.upenn.edu  Fri Aug 25 21:36:40 2006
From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer)
Date: Fri, 25 Aug 2006 15:36:40 -0400
Subject: [Pw_forum] Publications resulting from PWscf/Quantum Espresso
In-Reply-To: <20060825180353.76227.qmail@web52005.mail.yahoo.com>
References: <62C003E3-8A56-4330-A3C3-E95BCCE2396A@sissa.it>
	 <20060825180353.76227.qmail@web52005.mail.yahoo.com>
Message-ID: <7b6913e90608251236m4f31ecb7t973a07541cf888a4@mail.gmail.com>

On 8/25/06, Konstantin Kudin <konstantin_kudin at yahoo.com> wrote:
>  Hi all,
>
>  While were are on the subject I think it is also worthwhile to
> consider setting up a list of references on the methods included into
> QE (as opposed just to the applications alone).

as long as it does not make the output excessively long
(e.g. uses abbreviated names and journal and a DOI instead of the full title)
in principle you could borrow the 'please_cite' subroutine from gromacs.

a.

>  The purpose would be 2-fold. First, to give a precise pointer for the
> detailed description of a given method. And second, to facilitate
> proper citations of the QE developers' papers, such as any time a given
> approach is used for applications, the corresponding work by a QE
> person is cited.
>
>  Kostya
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
>


-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
  Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.


From konstantin_kudin at yahoo.com  Fri Aug 25 21:54:53 2006
From: konstantin_kudin at yahoo.com (Konstantin Kudin)
Date: Fri, 25 Aug 2006 12:54:53 -0700 (PDT)
Subject: [Pw_forum] Re: Publications resulting from PWscf/Quantum Espresso
In-Reply-To: <20060825192050.12346.qmail@xuxa.iecc.com>
Message-ID: <20060825195453.37246.qmail@web52009.mail.yahoo.com>

--- stewart at cnf.cornell.edu wrote:

> Hi all, 
> 
> I agree that a reference section is important.  I've had a chance to
> put a 
> few papers in on the wiki page including the review on DFPT.  For
> now, I am 
> going with the following format for papers, 
> 
> Paper title (hyperlinked to the doi identifier for the paper),
> Authors names 
> (first letter, last name), Full title of journal, volume, page
> number, 
> (year). 
>

 Good setup. I would also add the DOI identifier in the explicit text
form (not just in the hyperlink itself), in case anybody wants to copy
it without going through "paste link" stuff.

 Kostya

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From tr.kelkar at ncl.res.in  Sat Aug 26 08:12:04 2006
From: tr.kelkar at ncl.res.in (Tuhina Kelkar)
Date: Sat, 26 Aug 2006 11:42:04 +0530 (IST)
Subject: [Pw_forum] regarding energies obtained on running fqha.x
Message-ID: <14756106.1156572724974.JavaMail.root@webmail.ncl.res.in>

Hi everyone!
I have some very simple questions to ask. 
(a) Are the energies obtained on running fqha.x at different temperatures simply the phonon free energies? 
(b) Are the energy units Ryd/cell? And by "cell" does it mean the supercell that we had defined for calculating the phonon DOS or the primitive unit cell?
(c) Why do the imaginary frequencies die as we go for supercells?
Thanks in advance.
Regards,
Tuhina
---------------------------------------------------------------------------------------------------------------
Tuhina Kelkar
Junior Research Fellow
Physical Chemistry Division,
National Chemical Laboratory,
Pune - 411 008
INDIA
Contact: 020-25902298
alternative email: tuhina at gmail.com
---------------------------------------------------------------------------------------------------------------
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060826/5c59a508/attachment.htm 

From eyvaz_isaev at yahoo.com  Sat Aug 26 14:14:52 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Sat, 26 Aug 2006 05:14:52 -0700 (PDT)
Subject: [Pw_forum] regarding energies obtained on running fqha.x
In-Reply-To: <14756106.1156572724974.JavaMail.root@webmail.ncl.res.in>
Message-ID: <20060826121452.92363.qmail@web60312.mail.yahoo.com>

Hi, 
 
> (a) Are the energies obtained on running fqha.x at
> different temperatures simply the phonon free
> energies? 
More precisely, it is the phonon contribution to the
free  energy.

> (b) Are the energy units Ryd/cell? And by "cell"
> does it mean the supercell that we had defined for
> calculating the phonon DOS or the primitive unit
> cell?
Ry per cell you used to calculate phonons, but not
primitive unit cell. 

> (c) Why do the imaginary frequencies die as we go
> for supercells?
That is strange. It might be that there is a mistake
in your supercell setup. If not the case, your phonon
setup is not careful. 

Bests,
Eyvaz.

> Thanks in advance.
> Regards,
> Tuhina
>
---------------------------------------------------------------------------------------------------------------
> Tuhina Kelkar
> Junior Research Fellow
> Physical Chemistry Division,
> National Chemical Laboratory,
> Pune - 411 008
> INDIA
> Contact: 020-25902298
> alternative email: tuhina at gmail.com
>
---------------------------------------------------------------------------------------------------------------
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From lanhaiping at gmail.com  Sat Aug 26 18:57:30 2006
From: lanhaiping at gmail.com (lan haiping)
Date: Sun, 27 Aug 2006 00:57:30 +0800
Subject: [Pw_forum] phonon at Gamma point
In-Reply-To: <20060824140548.46812.qmail@web60323.mail.yahoo.com>
References: <c92002fa0608240533w2f74a466kecce4697ac0ec0f2@mail.gmail.com>
	 <20060824140548.46812.qmail@web60323.mail.yahoo.com>
Message-ID: <c92002fa0608260957te6776afh8b688b283cbe455@mail.gmail.com>

Dear Eyvaz,

Thank you for this instructions!

I just tried the coordinates from NRL-weblink, and didnot find any problems
as before.

But i have another problem want to ask you  , which  is just a result that i
wanna finish
Stefano's  exercise assigned last mail.

I just found that inversion symmetry would not found by pw.x code when the
FFT grds are not compatable with fractional translation. Then, I tried to
mannually set fft grid by nr1,nr2,nr3 .
Take$(espresso) /examples/example02/si.scf.in for example, when *ecutwfc* is
18.0Ry, the pw.x would found all Sym Ops, with smallest FFT grids
(20,20,20). But when *ecutwfc* is set to 30.0Ry with (30,30,30) fft grids
default, the code gave ""warning: symmetry operation # N not allowed".
According to  mannual, this warning is mostly related to fractional
translation incompatable with the fft grids. So, i  tried to reset fft grid
respectly with (36,36,36), (40,40,40),(60,60,60) and (54,54,54) , no such
warning messages are given for the first 3 situation but same occurs for the
last fft grids.  I cannot understand this test results, since the userguide
states that this warning is due to"*FFT grid dimension along that direction
is not divisible respectively by 2 or by 3, the symmetry operation will not
transform the FFT grid into itself*" . Obviously, all grids dimension along
x/y/z direction can be divided by 2 or 3.
   My understanding is correct or not ?



Regards,
Hai-Ping








On 8/24/06, Eyvaz Isaev <eyvaz_isaev at yahoo.com> wrote:
>
> Hi,
>
> > I just relaxed Anatase TiO2  with *ibrav= 7 , *
> > then output gave that Only 8 symmetry Operations are
> > found.
> Because of that your setup is not consistent.
> Your lattice parameter is
> 7.1356*0.529177=3.7759954012, but 2 Oxygen atoms have
> X-component (in units of alat) of 0.5.
> You put 1.888, but, 1.888 is not equal
> 3.7759954012/2=1.8879977. So, they could not be
> considered the same within 10^{-6} which is a default
> mismatch (please correct me if I am wrong). For sure
> you have to use 7  digits (after point) in such a kind
> specific cases, as c/a, b/a (and cosines, presumably).
> Besides, your atomic positions are completely
> different  of these specified in
> http://cst-www.nrl.navy.mil/lattice/struk.xmol/c5.pos
> Please use lattice parameters and atomic coordinates
> (in Angstrom) as they are in the link, and report what
> happens.
>
> I am not sure that it is a good idea to relax a well
> known structure if you have a problem with its
> symmetry.
>
> Bests,
> Eyvaz.
>
> > According  to your mail, this structure should have
> > 16 Symmetry Operations,
> > so
> > how can i obtain correct  symmetry operations ?
> >
> > Input file is below , thank you in advance !
> >
> >
> > *&CONTROL
> >              title = 'TiO2' ,
> >        calculation = 'relax',
> >          verbosity = 'default',
> >            tprnfor = .true.,
> >             outdir = '/home/haiping/tmp/',
> >             prefix = 'tio2-1',
> >            disk_io = 'default',
> >              nstep = 200,
> >         pseudo_dir =
> > '/home/haiping/espresso/pseudo/',
> > /*
> >
> > *&SYSTEM
> >                 ibrav= 7,
> >                celldm(1)=7.1356,celldm(3)=2.51218,
> >                nat = 6 ,
> >               ntyp = 2 ,
> >               nbnd = 30 ,
> >            ecutwfc = 30.0000000000,
> >        occupations = 'fixed' ,
> > *
> >
> > *&ELECTRONS
> >   electron_maxstep = 100,
> >           conv_thr = 1.0e-09,
> >        mixing_mode = 'plain' ,
> >        startingpot = 'atomic' ,
> >        startingwfc = 'atomic' ,
> >        mixing_beta = 0.300000000,
> >        mixing_ndim = 8,
> >    diagonalization = 'david_overlap' ,
> >   diago_david_ndim = 8,
> > /*
> >
> > *&IONS
> >              ion_dynamics = 'bfgs',
> >              pot_extrapolation = 'first_order',
> >              wfc_extrapolation = 'first_order',*
> >
> > */*
> >
> > *O  15.999  O.pw91-van_ak.UPF
> > Ti 47.867  Ti.pw91-nsp-van.UPF
> > ATOMIC_POSITIONS angstrom
> > Ti       0.000000000   0.000000000   0.000000000
> > Ti       0.000000000   1.888000000   2.371500000
> > O        0.000000000   0.000000000   -1.983097630
> > O        0.000000000   0.000000000   1.983097630
> > O        0.000000000   1.888000000   0.387796645
> > O        0.000000000   1.888000000   4.355203355
> > K_POINTS   automatic
> > 6 6  7   0 0 0
> > *
> > *
> >
> > * On 8/20/06, Eyvaz Isaev <eyvaz_isaev at yahoo.com>
> > wrote:
> > >
> > > OK, let me hint more.
> > >
> > > Anatase has a Pearson symbol tI12 or space group
> > > I4_1/amd. So, ibrav should be 7, but you have
> > chosen a
> > > monoclinic lattice, i.e. ibrav=14. The space group
> > has
> > > 16 symmetry elements, but your choice after
> > correcting
> > > ecutrho gives only 4 symmetry elements .
> > >
> > > Imagine  FCC lattice and its conventional basis
> > > vectors now. According to you now we have to put
> > > ibrav=4 or 5. But ibrav=2 for FCC lattice.
> > >
> > > In addition, I also checked your input file via
> > > XCrysDen before I replied previously. Yes, there
> > is no
> > > problem with it, but a figure does not look like
> > > anatase though  you used  nonzero ibrav. XcrysDen
> > > usually replicates atomic positions  if
> > ibrav.neq.0.
> > > For example, if you put ibrav=0  for  FCC cell and
> > > specify conventional CELL_PARAMETERS you will find
> >  a
> > > rhombohedron, but using ibrav=2 and  default basis
> > > vectors you find a cubic cell.
> > >
> > > Bests,
> > > Eyvaz.
> > >
> > >
> > > --- lan haiping <lanhaiping at gmail.com> wrote:
> > >
> > > > Dear Eyvaz,
> > > > Thank you.
> > > > I checked the strcuture again using XCRYSDEN,
> > > > and found no problems related to the  lattices.
> > > >
> > > > Regards,
> > > >
> > > > Hai-Ping
> > > >
> > > >
> > > >
> > > > On 8/20/06, Eyvaz Isaev <eyvaz_isaev at yahoo.com >
> > > > wrote:
> > > > >
> > > > > Hi,
> > > > >
> > > > > It seems your lattice choice is not correct.
> > > > Please
> > > > > have a look at
> > > > >
> > http://cst-www.nrl.navy.mil/lattice/struk/c5.html
> > > > >
> > > > > Bests,
> > > > > Eyvaz.
> > > > >
> > > > > --- lan haiping <lanhaiping at gmail.com> wrote:
> > > > >
> > > > > > Dear All,
> > > > > > I also have calculated gamma phonon
> > frequencies
> > > > of
> > > > > > anatase TiO2 .
> > > > > > I run ph.x caculation after relaxing the
> > > > structure,
> > > > > > but  9 of 18 frequencies
> > > > > > are minus.
> > > > > >
> > > > > > Would you please give me some hints and
> > help?
> > > > > >
> > > > > > input settings are below, thank you in
> > advance !
> > > > > >
> > > > > > *phonons of TiO2 at Gamma
> > > > > >  &inputph
> > > > > >   tr2_ph=1.0d-14,
> > > > > >   recover=.true.,
> > > > > >   epsil=.true.,
> > > > > >   prefix='tio2',
> > > > > >   fildyn='TiO2-gamma.dyn',
> > > > > >   outdir='/home/haiping/tmp/'
> > > > > >  /
> > > > > > 0.0 0.0 0.0*
> > > > > > **
> > > > > > *Relaxing calculation input :*
> > > > > > *&CONTROL
> > > > > >              title = 'TiO2' ,
> > > > > >        calculation = 'relax',
> > > > > >          verbosity = 'default',
> > > > > >            tprnfor = .true.,
> > > > > >             outdir = '/home/haiping/tmp/',
> > > > > >             prefix = 'tio2',
> > > > > >            disk_io = 'default',
> > > > > >              nstep = 200,
> > > > > >         pseudo_dir =
> > > > > > '/home/haiping/espresso/pseudo/',
> > > > > > /
> > > > > > &SYSTEM
> > > > > >                 ibrav= 14,
> > > > > >
> > > > > >
> > celldm(1)=10.2856,celldm(2)=1.0,celldm(3)=1.0,
> > > > > >
> > > > > > celldm(4)=-0.759357,celldm(5)=-0.759357,
> > > > > >                celldm(6)=0.518714
> > > > > >                nat = 6 ,
> > > > > >               ntyp = 2 ,
> > > > > >               nbnd = 30 ,
> > > > > >            ecutwfc = 30.0000000000,
> > > > > >            ecutrho = 160.000,
> > > > > >        occupations = 'fixed' ,
> > > > > > /
> > > > > >
> > > > > >
> > > > > > &ELECTRONS
> > > > > >   electron_maxstep = 100,
> > > > > >           conv_thr = 1.0e-07,
> > > > > >        mixing_mode = 'plain' ,
> > > > > >        startingpot = 'atomic' ,
> > > > > >        startingwfc = 'atomic' ,
> > > > > >        mixing_beta = 0.300000000,
> > > > > >        mixing_ndim = 8,
> > > > > >    diagonalization = 'david_overlap' ,
> > > > > >   diago_david_ndim = 8,
> > > > > > /
> >
> === message truncated ===
>
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060827/38a03d4c/attachment.htm 

From baroni at sissa.it  Sat Aug 26 19:02:36 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Sat, 26 Aug 2006 19:02:36 +0200
Subject: [Pw_forum] phonon at Gamma point
In-Reply-To: <c92002fa0608260957te6776afh8b688b283cbe455@mail.gmail.com>
References: <c92002fa0608240533w2f74a466kecce4697ac0ec0f2@mail.gmail.com> <20060824140548.46812.qmail@web60323.mail.yahoo.com> <c92002fa0608260957te6776afh8b688b283cbe455@mail.gmail.com>
Message-ID: <ABEA4F76-4FB3-4689-A8B1-81A3CE605DE6@sissa.it>

Hai-Ping: forgive my poor understanding of you progresses, but let me  
just you a very simple hint. In order for the program to be able to  
detect a symmetry operation, it is necessary that the fractional  
translation associated with it can be represented on the discrete FFT  
grid. As simple as that. Best wishes - Stefano

On Aug 26, 2006, at 6:57 PM, lan haiping wrote:

> Dear Eyvaz,
>
> Thank you for this instructions!
>
> I just tried the coordinates from NRL-weblink, and didnot find any  
> problems as before.
>
> But i have another problem want to ask you  , which  is just a  
> result that i wanna finish
> Stefano's  exercise assigned last mail.
>
> I just found that inversion symmetry would not found by pw.x code  
> when the FFT grds are not compatable with fractional translation.  
> Then, I tried to mannually set fft grid by nr1,nr2,nr3 .
> Take$(espresso) /examples/example02/si.scf.in for example, when  
> ecutwfc is 18.0Ry, the pw.x would found all Sym Ops, with smallest  
> FFT grids (20,20,20). But when ecutwfc is set to 30.0Ry with  
> (30,30,30) fft grids default, the code gave ""warning: symmetry  
> operation # N not allowed". According to  mannual, this warning is  
> mostly related to fractional translation incompatable with the fft  
> grids. So, i  tried to reset fft grid respectly with (36,36,36),  
> (40,40,40),(60,60,60) and (54,54,54) , no such warning messages are  
> given for the first 3 situation but same occurs for the last fft  
> grids.  I cannot understand this test results, since the userguide  
> states that this warning is due to" FFT grid dimension along that  
> direction is not divisible respectively by 2 or by 3, the symmetry  
> operation will not transform the FFT grid into itself" . Obviously,  
> all grids dimension along x/y/z direction can be divided by 2 or 3.
>    My understanding is correct or not ?
>
>
>
> Regards,
> Hai-Ping
>
>
>
>
>
>
>
>
> On 8/24/06, Eyvaz Isaev <eyvaz_isaev at yahoo.com > wrote:
> Hi,
>
> > I just relaxed Anatase TiO2  with *ibrav= 7 , *
> > then output gave that Only 8 symmetry Operations are
> > found.
> Because of that your setup is not consistent.
> Your lattice parameter is
> 7.1356*0.529177=3.7759954012, but 2 Oxygen atoms have
> X-component (in units of alat) of 0.5.
> You put 1.888, but, 1.888 is not equal
> 3.7759954012/2=1.8879977. So, they could not be
> considered the same within 10^{-6} which is a default
> mismatch (please correct me if I am wrong). For sure
> you have to use 7  digits (after point) in such a kind
> specific cases, as c/a, b/a (and cosines, presumably).
> Besides, your atomic positions are completely
> different  of these specified in
> http://cst-www.nrl.navy.mil/lattice/struk.xmol/c5.pos
> Please use lattice parameters and atomic coordinates
> (in Angstrom) as they are in the link, and report what
> happens.
>
> I am not sure that it is a good idea to relax a well
> known structure if you have a problem with its
> symmetry.
>
> Bests,
> Eyvaz.
>
> > According  to your mail, this structure should have
> > 16 Symmetry Operations,
> > so
> > how can i obtain correct  symmetry operations ?
> >
> > Input file is below , thank you in advance !
> >
> >
> > *&CONTROL
> >              title = 'TiO2' ,
> >        calculation = 'relax',
> >          verbosity = 'default',
> >            tprnfor = .true.,
> >             outdir = '/home/haiping/tmp/',
> >             prefix = 'tio2-1',
> >            disk_io = 'default',
> >              nstep = 200,
> >         pseudo_dir =
> > '/home/haiping/espresso/pseudo/',
> > /*
> >
> > *&SYSTEM
> >                 ibrav= 7,
> >                celldm(1)=7.1356,celldm(3)=2.51218,
> >                nat = 6 ,
> >               ntyp = 2 ,
> >               nbnd = 30 ,
> >            ecutwfc = 30.0000000000,
> >        occupations = 'fixed' ,
> > *
> >
> > *&ELECTRONS
> >   electron_maxstep = 100,
> >           conv_thr = 1.0e-09,
> >        mixing_mode = 'plain' ,
> >        startingpot = 'atomic' ,
> >        startingwfc = 'atomic' ,
> >        mixing_beta = 0.300000000,
> >        mixing_ndim = 8,
> >    diagonalization = 'david_overlap' ,
> >   diago_david_ndim = 8,
> > /*
> >
> > *&IONS
> >              ion_dynamics = 'bfgs',
> >              pot_extrapolation = 'first_order',
> >              wfc_extrapolation = 'first_order',*
> >
> > */*
> >
> > *O  15.999  O.pw91-van_ak.UPF
> > Ti 47.867  Ti.pw91-nsp-van.UPF
> > ATOMIC_POSITIONS angstrom
> > Ti       0.000000000   0.000000000   0.000000000
> > Ti       0.000000000   1.888000000   2.371500000
> > O        0.000000000   0.000000000   -1.983097630
> > O        0.000000000   0.000000000   1.983097630
> > O        0.000000000   1.888000000   0.387796645
> > O        0.000000000   1.888000000   4.355203355
> > K_POINTS   automatic
> > 6 6  7   0 0 0
> > *
> > *
> >
> > * On 8/20/06, Eyvaz Isaev < eyvaz_isaev at yahoo.com>
> > wrote:
> > >
> > > OK, let me hint more.
> > >
> > > Anatase has a Pearson symbol tI12 or space group
> > > I4_1/amd. So, ibrav should be 7, but you have
> > chosen a
> > > monoclinic lattice, i.e. ibrav=14. The space group
> > has
> > > 16 symmetry elements, but your choice after
> > correcting
> > > ecutrho gives only 4 symmetry elements .
> > >
> > > Imagine  FCC lattice and its conventional basis
> > > vectors now. According to you now we have to put
> > > ibrav=4 or 5. But ibrav=2 for FCC lattice.
> > >
> > > In addition, I also checked your input file via
> > > XCrysDen before I replied previously. Yes, there
> > is no
> > > problem with it, but a figure does not look like
> > > anatase though  you used  nonzero ibrav. XcrysDen
> > > usually replicates atomic positions  if
> > ibrav.neq.0.
> > > For example, if you put ibrav=0  for  FCC cell and
> > > specify conventional CELL_PARAMETERS you will find
> >  a
> > > rhombohedron, but using ibrav=2 and  default basis
> > > vectors you find a cubic cell.
> > >
> > > Bests,
> > > Eyvaz.
> > >
> > >
> > > --- lan haiping < lanhaiping at gmail.com> wrote:
> > >
> > > > Dear Eyvaz,
> > > > Thank you.
> > > > I checked the strcuture again using XCRYSDEN,
> > > > and found no problems related to the  lattices.
> > > >
> > > > Regards,
> > > >
> > > > Hai-Ping
> > > >
> > > >
> > > >
> > > > On 8/20/06, Eyvaz Isaev < eyvaz_isaev at yahoo.com >
> > > > wrote:
> > > > >
> > > > > Hi,
> > > > >
> > > > > It seems your lattice choice is not correct.
> > > > Please
> > > > > have a look at
> > > > >
> > http://cst-www.nrl.navy.mil/lattice/struk/c5.html
> > > > >
> > > > > Bests,
> > > > > Eyvaz.
> > > > >
> > > > > --- lan haiping < lanhaiping at gmail.com> wrote:
> > > > >
> > > > > > Dear All,
> > > > > > I also have calculated gamma phonon
> > frequencies
> > > > of
> > > > > > anatase TiO2 .
> > > > > > I run ph.x caculation after relaxing the
> > > > structure,
> > > > > > but  9 of 18 frequencies
> > > > > > are minus.
> > > > > >
> > > > > > Would you please give me some hints and
> > help?
> > > > > >
> > > > > > input settings are below, thank you in
> > advance !
> > > > > >
> > > > > > *phonons of TiO2 at Gamma
> > > > > >  &inputph
> > > > > >   tr2_ph=1.0d-14,
> > > > > >   recover=.true.,
> > > > > >   epsil=.true.,
> > > > > >   prefix='tio2',
> > > > > >   fildyn='TiO2-gamma.dyn',
> > > > > >   outdir='/home/haiping/tmp/'
> > > > > >  /
> > > > > > 0.0 0.0 0.0*
> > > > > > **
> > > > > > *Relaxing calculation input :*
> > > > > > *&CONTROL
> > > > > >              title = 'TiO2' ,
> > > > > >        calculation = 'relax',
> > > > > >          verbosity = 'default',
> > > > > >            tprnfor = .true.,
> > > > > >             outdir = '/home/haiping/tmp/',
> > > > > >             prefix = 'tio2',
> > > > > >            disk_io = 'default',
> > > > > >              nstep = 200,
> > > > > >         pseudo_dir =
> > > > > > '/home/haiping/espresso/pseudo/',
> > > > > > /
> > > > > > &SYSTEM
> > > > > >                 ibrav= 14,
> > > > > >
> > > > > >
> > celldm(1)=10.2856,celldm(2)=1.0 ,celldm(3)=1.0,
> > > > > >
> > > > > > celldm(4)=-0.759357,celldm(5)=-0.759357,
> > > > > >                celldm(6)=0.518714
> > > > > >                nat = 6 ,
> > > > > >               ntyp = 2 ,
> > > > > >               nbnd = 30 ,
> > > > > >            ecutwfc = 30.0000000000,
> > > > > >            ecutrho = 160.000,
> > > > > >        occupations = 'fixed' ,
> > > > > > /
> > > > > >
> > > > > >
> > > > > > &ELECTRONS
> > > > > >   electron_maxstep = 100,
> > > > > >           conv_thr = 1.0e-07,
> > > > > >        mixing_mode = 'plain' ,
> > > > > >        startingpot = 'atomic' ,
> > > > > >        startingwfc = 'atomic' ,
> > > > > >        mixing_beta = 0.300000000,
> > > > > >        mixing_ndim = 8,
> > > > > >    diagonalization = 'david_overlap' ,
> > > > > >   diago_david_ndim = 8,
> > > > > > /
> >
> === message truncated ===
>
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060826/fee50e2c/attachment.htm 

From emenendez at macul.ciencias.uchile.cl  Sat Aug 26 19:36:55 2006
From: emenendez at macul.ciencias.uchile.cl (Eduardo Ariel Menendez P)
Date: Sat, 26 Aug 2006 13:36:55 -0400 (CLT)
Subject: [Pw_forum] ld1.x normalization and overlap integrals
Message-ID: <Pine.LNX.4.58.0608261326470.7890@macul.ciencias.uchile.cl>

Hello,
Doing an all-electron calculation with atomic code (ld1.x) I can read
from the output that the overlap integrals are not zero

    normalization and overlap integrals

     s(1S/1S) =  1.000000  <r> =   0.0172  <r2> =    0.0004  r(max) =
0.0106
     s(1S/2S) = -0.036214
     s(1S/3S) = -0.015876
     s(1S/4S) = -0.007834
     s(1S/5S) = -0.003464
     s(1S/6S) = -0.001003
     s(2S/2S) =  1.000000  <r> =   0.0724  <r2> =    0.0062  r(max) =
0.0612
     s(2S/3S) = -0.010197

I think that the overlap integrals should 0. They are small, but should be
zero or at least, decrease if the radial mesh is made finer. I have seen
this behaviour in the exmaples, e.g.
espresso-3.1/atomic_doc/all-electron/reference/cu.out

In my case, the overlap integrals are 0.000000 if rel=0 or rel=2,
and give the avobe numbers if rel=1.

This is the input file.
 &input
        title='Pt',
        zed=78.0,
        rel=1,
        iswitch=1,
        xmin=-6.0,
        dx=0.005,
        lsd=0,
        config='[Xe] 4f14 5d8 6s2 6p0',
        dft='PZ',
        prefix='pt.ae.pz'
 /


Thanks
Eduardo


From lanhaiping at gmail.com  Sat Aug 26 22:02:32 2006
From: lanhaiping at gmail.com (lan haiping)
Date: Sun, 27 Aug 2006 04:02:32 +0800
Subject: [Pw_forum] phonon at Gamma point
In-Reply-To: <ABEA4F76-4FB3-4689-A8B1-81A3CE605DE6@sissa.it>
References: <c92002fa0608240533w2f74a466kecce4697ac0ec0f2@mail.gmail.com>
	 <20060824140548.46812.qmail@web60323.mail.yahoo.com>
	 <c92002fa0608260957te6776afh8b688b283cbe455@mail.gmail.com>
	 <ABEA4F76-4FB3-4689-A8B1-81A3CE605DE6@sissa.it>
Message-ID: <c92002fa0608261302t4adabfe5o46215933b05ba1ed@mail.gmail.com>

Dear stefano,

Thank you. I think i catch this indication..

regards,
Hai-Ping
On 8/27/06, Stefano Baroni <baroni at sissa.it> wrote:
>
>  Hai-Ping: forgive my poor understanding of you progresses, but let me
> just you a very simple hint. In order for the program to be able to detect a
> symmetry operation, it is necessary that the fractional translation
> associated with it can be represented on the discrete FFT grid. As simple as
> that. Best wishes - Stefano
>
>  On Aug 26, 2006, at 6:57 PM, lan haiping wrote:
>
>  Dear Eyvaz,
>
> Thank you for this instructions!
>
> I just tried the coordinates from NRL-weblink, and didnot find any
> problems as before.
>
> But i have another problem want to ask you  , which  is just a result that
> i wanna finish
> Stefano's  exercise assigned last mail.
>
> I just found that inversion symmetry would not found by pw.x code when the
> FFT grds are not compatable with fractional translation. Then, I tried to
> mannually set fft grid by nr1,nr2,nr3 .
> Take$(espresso) /examples/example02/si.scf.in for example, when *ecutwfc*is
> 18.0Ry, the pw.x would found all Sym Ops, with smallest FFT grids
> (20,20,20). But when *ecutwfc* is set to 30.0Ry with (30,30,30) fft grids
> default, the code gave ""warning: symmetry operation # N not allowed".
> According to  mannual, this warning is mostly related to fractional
> translation incompatable with the fft grids. So, i  tried to reset fft grid
> respectly with (36,36,36), (40,40,40),(60,60,60) and (54,54,54) , no such
> warning messages are given for the first 3 situation but same occurs for the
> last fft grids.  I cannot understand this test results, since the userguide
> states that this warning is due to" *FFT grid dimension along that
> direction is not divisible respectively by 2 or by 3, the symmetry operation
> will not transform the FFT grid into itself*" . Obviously, all grids
> dimension along x/y/z direction can be divided by 2 or 3.
>    My understanding is correct or not ?
>
>
>
> Regards,
> Hai-Ping
>
>
>
>
>
>
>
>
> On 8/24/06, Eyvaz Isaev <eyvaz_isaev at yahoo.com > wrote:
> >
> > Hi,
> >
> > > I just relaxed Anatase TiO2  with *ibrav= 7 , *
> > > then output gave that Only 8 symmetry Operations are
> > > found.
> > Because of that your setup is not consistent.
> > Your lattice parameter is
> > 7.1356*0.529177=3.7759954012, but 2 Oxygen atoms have
> > X-component (in units of alat) of 0.5.
> > You put 1.888, but, 1.888 is not equal
> > 3.7759954012/2=1.8879977. So, they could not be
> > considered the same within 10^{-6} which is a default
> > mismatch (please correct me if I am wrong). For sure
> > you have to use 7  digits (after point) in such a kind
> > specific cases, as c/a, b/a (and cosines, presumably).
> > Besides, your atomic positions are completely
> > different  of these specified in
> > http://cst-www.nrl.navy.mil/lattice/struk.xmol/c5.pos
> > Please use lattice parameters and atomic coordinates
> > (in Angstrom) as they are in the link, and report what
> > happens.
> >
> > I am not sure that it is a good idea to relax a well
> > known structure if you have a problem with its
> > symmetry.
> >
> > Bests,
> > Eyvaz.
> >
> > > According  to your mail, this structure should have
> > > 16 Symmetry Operations,
> > > so
> > > how can i obtain correct  symmetry operations ?
> > >
> > > Input file is below , thank you in advance !
> > >
> > >
> > > *&CONTROL
> > >              title = 'TiO2' ,
> > >        calculation = 'relax',
> > >          verbosity = 'default',
> > >            tprnfor = .true.,
> > >             outdir = '/home/haiping/tmp/',
> > >             prefix = 'tio2-1',
> > >            disk_io = 'default',
> > >              nstep = 200,
> > >         pseudo_dir =
> > > '/home/haiping/espresso/pseudo/',
> > > /*
> > >
> > > *&SYSTEM
> > >                 ibrav= 7,
> > >                celldm(1)=7.1356,celldm(3)=2.51218,
> > >                nat = 6 ,
> > >               ntyp = 2 ,
> > >               nbnd = 30 ,
> > >            ecutwfc = 30.0000000000,
> > >        occupations = 'fixed' ,
> > > *
> > >
> > > *&ELECTRONS
> > >   electron_maxstep = 100,
> > >           conv_thr = 1.0e-09,
> > >        mixing_mode = 'plain' ,
> > >        startingpot = 'atomic' ,
> > >        startingwfc = 'atomic' ,
> > >        mixing_beta = 0.300000000,
> > >        mixing_ndim = 8,
> > >    diagonalization = 'david_overlap' ,
> > >   diago_david_ndim = 8,
> > > /*
> > >
> > > *&IONS
> > >              ion_dynamics = 'bfgs',
> > >              pot_extrapolation = 'first_order',
> > >              wfc_extrapolation = 'first_order',*
> > >
> > > */*
> > >
> > > *O  15.999  O.pw91-van_ak.UPF
> > > Ti 47.867  Ti.pw91-nsp-van.UPF
> > > ATOMIC_POSITIONS angstrom
> > > Ti       0.000000000   0.000000000   0.000000000
> > > Ti       0.000000000   1.888000000   2.371500000
> > > O        0.000000000   0.000000000   -1.983097630
> > > O        0.000000000   0.000000000   1.983097630
> > > O        0.000000000   1.888000000   0.387796645
> > > O        0.000000000   1.888000000   4.355203355
> > > K_POINTS   automatic
> > > 6 6  7   0 0 0
> > > *
> > > *
> > >
> > > * On 8/20/06, Eyvaz Isaev < eyvaz_isaev at yahoo.com>
> > > wrote:
> > > >
> > > > OK, let me hint more.
> > > >
> > > > Anatase has a Pearson symbol tI12 or space group
> > > > I4_1/amd. So, ibrav should be 7, but you have
> > > chosen a
> > > > monoclinic lattice, i.e. ibrav=14. The space group
> > > has
> > > > 16 symmetry elements, but your choice after
> > > correcting
> > > > ecutrho gives only 4 symmetry elements .
> > > >
> > > > Imagine  FCC lattice and its conventional basis
> > > > vectors now. According to you now we have to put
> > > > ibrav=4 or 5. But ibrav=2 for FCC lattice.
> > > >
> > > > In addition, I also checked your input file via
> > > > XCrysDen before I replied previously. Yes, there
> > > is no
> > > > problem with it, but a figure does not look like
> > > > anatase though  you used  nonzero ibrav. XcrysDen
> > > > usually replicates atomic positions  if
> > > ibrav.neq.0.
> > > > For example, if you put ibrav=0  for  FCC cell and
> > > > specify conventional CELL_PARAMETERS you will find
> > >  a
> > > > rhombohedron, but using ibrav=2 and  default basis
> > > > vectors you find a cubic cell.
> > > >
> > > > Bests,
> > > > Eyvaz.
> > > >
> > > >
> > > > --- lan haiping < lanhaiping at gmail.com> wrote:
> > > >
> > > > > Dear Eyvaz,
> > > > > Thank you.
> > > > > I checked the strcuture again using XCRYSDEN,
> > > > > and found no problems related to the  lattices.
> > > > >
> > > > > Regards,
> > > > >
> > > > > Hai-Ping
> > > > >
> > > > >
> > > > >
> > > > > On 8/20/06, Eyvaz Isaev < eyvaz_isaev at yahoo.com >
> > > > > wrote:
> > > > > >
> > > > > > Hi,
> > > > > >
> > > > > > It seems your lattice choice is not correct.
> > > > > Please
> > > > > > have a look at
> > > > > >
> > > http://cst-www.nrl.navy.mil/lattice/struk/c5.html
> > > > > >
> > > > > > Bests,
> > > > > > Eyvaz.
> > > > > >
> > > > > > --- lan haiping < lanhaiping at gmail.com> wrote:
> > > > > >
> > > > > > > Dear All,
> > > > > > > I also have calculated gamma phonon
> > > frequencies
> > > > > of
> > > > > > > anatase TiO2 .
> > > > > > > I run ph.x caculation after relaxing the
> > > > > structure,
> > > > > > > but  9 of 18 frequencies
> > > > > > > are minus.
> > > > > > >
> > > > > > > Would you please give me some hints and
> > > help?
> > > > > > >
> > > > > > > input settings are below, thank you in
> > > advance !
> > > > > > >
> > > > > > > *phonons of TiO2 at Gamma
> > > > > > >  &inputph
> > > > > > >   tr2_ph=1.0d-14,
> > > > > > >   recover=.true.,
> > > > > > >   epsil=.true.,
> > > > > > >   prefix='tio2',
> > > > > > >   fildyn='TiO2-gamma.dyn',
> > > > > > >   outdir='/home/haiping/tmp/'
> > > > > > >  /
> > > > > > > 0.0 0.0 0.0*
> > > > > > > **
> > > > > > > *Relaxing calculation input :*
> > > > > > > *&CONTROL
> > > > > > >              title = 'TiO2' ,
> > > > > > >        calculation = 'relax',
> > > > > > >          verbosity = 'default',
> > > > > > >            tprnfor = .true.,
> > > > > > >             outdir = '/home/haiping/tmp/',
> > > > > > >             prefix = 'tio2',
> > > > > > >            disk_io = 'default',
> > > > > > >              nstep = 200,
> > > > > > >         pseudo_dir =
> > > > > > > '/home/haiping/espresso/pseudo/',
> > > > > > > /
> > > > > > > &SYSTEM
> > > > > > >                 ibrav= 14,
> > > > > > >
> > > > > > >
> > > celldm(1)=10.2856,celldm(2)=1.0 ,celldm(3)=1.0,
> > > > > > >
> > > > > > > celldm(4)=-0.759357,celldm(5)=-0.759357,
> > > > > > >                celldm(6)=0.518714
> > > > > > >                nat = 6 ,
> > > > > > >               ntyp = 2 ,
> > > > > > >               nbnd = 30 ,
> > > > > > >            ecutwfc = 30.0000000000,
> > > > > > >            ecutrho = 160.000,
> > > > > > >        occupations = 'fixed' ,
> > > > > > > /
> > > > > > >
> > > > > > >
> > > > > > > &ELECTRONS
> > > > > > >   electron_maxstep = 100,
> > > > > > >           conv_thr = 1.0e-07,
> > > > > > >        mixing_mode = 'plain' ,
> > > > > > >        startingpot = 'atomic' ,
> > > > > > >        startingwfc = 'atomic' ,
> > > > > > >        mixing_beta = 0.300000000,
> > > > > > >        mixing_ndim = 8,
> > > > > > >    diagonalization = 'david_overlap' ,
> > > > > > >   diago_david_ndim = 8,
> > > > > > > /
> > >
> > === message truncated ===
> >
> >
> > __________________________________________________
> > Do You Yahoo!?
> > Tired of spam?  Yahoo! Mail has the best spam protection around
> > http://mail.yahoo.com
> > _______________________________________________
> > Pw_forum mailing list
> > Pw_forum at pwscf.org
> > http://www.democritos.it/mailman/listinfo/pw_forum
> >
>
>
>   ---
> Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center - Trieste
> [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)
>
> Please, if possible, don't  send me MS Word or PowerPoint attachments
> Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html
>
>
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060827/891faee0/attachment.htm 

From prasenjit at jncasr.ac.in  Sun Aug 27 08:18:03 2006
From: prasenjit at jncasr.ac.in (PRASENJIT GHOSH)
Date: Sun, 27 Aug 2006 11:48:03 +0530 (IST)
Subject: [Pw_forum] BFGS convergence
Message-ID: <44416.202.41.111.151.1156659483.squirrel@202.41.111.151>

Dear all,

I have 2 doubts:
1) specific to my cal.
I'm trying to relax 4layer MgO(100) surface with IML Rh atom & NO on it.
I'm keeping the bottom 2 layers fixed & relaxing rest of the layers along
all directions. I have set forc_conv_thr as 5.0D-4. When the BFGS
optimization is complete the total force printed out is 0.001184 Ry/au.,
which is higher than the threshold.

The force on individual atoms are as given below:

     atom   1 type  4   force =     0.00000000    0.00019439    0.00035388
     atom   2 type  2   force =     0.00000000    0.00003994    0.00030581
     atom   3 type  3   force =     0.00004990    0.00006395    0.00026932
     atom   4 type  3   force =    -0.00004990    0.00006395    0.00026932
     atom   5 type  3   force =     0.00000510   -0.00010716    0.00017645
     atom   6 type  3   force =    -0.00000510   -0.00010716    0.00017645
     atom   7 type  3   force =    -0.00015740    0.00021421    0.00025840
     atom   8 type  3   force =     0.00015740    0.00021421    0.00025840
     atom   9 type  1   force =     0.00000000    0.00017531    0.00012835
     atom  10 type  1   force =     0.00000000    0.00012586    0.00009343
     atom  11 type  2   force =    -0.00003854   -0.00009570   -0.00010196
     atom  12 type  2   force =     0.00003854   -0.00009570   -0.00010196
     atom  13 type  1   force =     0.00000000    0.00002684    0.00020808
     atom  14 type  1   force =     0.00000000   -0.00008463    0.00035348
     atom  15 type  2   force =    -0.00002782   -0.00001293    0.00001952
     atom  16 type  2   force =     0.00002782   -0.00001293    0.00001952
     atom  17 type  1   force =     0.00000000    0.00001629    0.00009910
     atom  18 type  1   force =     0.00000000   -0.00003038    0.00014986
     atom  19 type  2   force =     0.00000763    0.00008195    0.00007784
     atom  20 type  2   force =    -0.00000763    0.00008195    0.00007784
     atom  21 type  2   force =     0.00000000    0.00003195    0.00010070
     atom  22 type  2   force =     0.00000000    0.00003854    0.00016380
     atom  23 type  1   force =    -0.00010405   -0.00006846    0.00012527
     atom  24 type  1   force =     0.00010405   -0.00006846    0.00012527
     atom  25 type  2   force =     0.00000000   -0.00002769    0.00001703
     atom  26 type  2   force =     0.00000000   -0.00007519    0.00008055
     atom  27 type  1   force =    -0.00004277    0.00010317   -0.00017790
     atom  28 type  1   force =     0.00004277    0.00010317   -0.00017790
     atom  29 type  2   force =     0.00000000    0.00001178    0.00015564
     atom  30 type  2   force =     0.00000000    0.00009848    0.00005032
     atom  31 type  1   force =    -0.00009405   -0.00007765    0.00004686
     atom  32 type  1   force =     0.00009405   -0.00007765    0.00004686
     atom  33 type  1   force =     0.00000000   -0.00027438   -0.00258973
     atom  34 type  1   force =     0.00000000    0.00085927   -0.00230357
     atom  35 type  2   force =    -0.00002762    0.00106477    0.00266372
     atom  36 type  2   force =     0.00002762    0.00106477    0.00266372
     atom  37 type  1   force =     0.00000000   -0.00127835   -0.00262410
     atom  38 type  1   force =     0.00000000   -0.00187757   -0.00264360
     atom  39 type  2   force =     0.00033189   -0.00088088    0.00323964
     atom  40 type  2   force =    -0.00033189   -0.00088088    0.00323964
     atom  41 type  1   force =     0.00000000    0.00132099   -0.00267980
     atom  42 type  1   force =     0.00000000    0.00088146   -0.00307944
     atom  43 type  2   force =     0.00043540   -0.00046399    0.00208586
     atom  44 type  2   force =    -0.00043540   -0.00046399    0.00208586
     atom  45 type  2   force =     0.00000000   -0.00001967   -0.01458796
     atom  46 type  2   force =     0.00000000    0.00001272   -0.01458917
     atom  47 type  1   force =    -0.00001307   -0.00005348    0.01390984
     atom  48 type  1   force =     0.00001307   -0.00005348    0.01390984
     atom  49 type  2   force =     0.00000000   -0.00068276   -0.01433952
     atom  50 type  2   force =     0.00000000   -0.00070884   -0.01425615
     atom  51 type  1   force =     0.00001464   -0.00004340    0.01348247
     atom  52 type  1   force =    -0.00001464   -0.00004340    0.01348247
     atom  53 type  2   force =     0.00000000    0.00075245   -0.01418097
     atom  54 type  2   force =     0.00000000    0.00074367   -0.01430795
     atom  55 type  1   force =     0.00003199    0.00014034    0.01388566
     atom  56 type  1   force =    -0.00003199    0.00014034    0.01388566

     Total force =     0.001184     Total SCF correction =     0.000147


Atoms 33-56 are kept fixed. Hence the forces on them is high whereas the
forces on each of the individual components of the relaxed atoms are below
the threshold.

My question is when the code calculates the force (grad) & their
differencec between 2 consecutive BFGS iter. (grad_error) & is it only for
the atoms which are allowed to relax or for the full system?

If its the second case then can anybody suggest where I'm going wrong?

With regards,
Prasenjit.

PRASENJIT GHOSH,
Ph. D STUDENT,
THEORETICAL SCIENCES UNIT,
JAWAHARLAL NEHRU CENTRE,
JAKKUR P. O.,
BANGALORE - 560064,
INDIA.

PHONE:+91-80-22082835 (OFFICE)
            +91 9880519401 (MOBILE)


From ding at sissa.it  Sun Aug 27 09:39:26 2006
From: ding at sissa.it (Xunlei Ding)
Date: Sun, 27 Aug 2006 09:39:26 +0200
Subject: [Pw_forum] What to do when NEB calculation is difficult to converge?
Message-ID: <44F14C2E.4080503@sissa.it>

Dear all,
I am now doing a neb calculation on a surface adsorption system with 
pwscf3.0
The default values are used in the calculation as below:
     calculation                   =  neb
     restart_mode                  =  restart
     opt_scheme                    =  quick-min
     num_of_images                 =  7
     nstep                         =  500
     first_last_opt                =  F
     coarse-grained phase-space    =  F
     use_freezing                  =  F
     ds                            =  1.0000 a.u.
     CI_scheme                     =  highest-TS
     k_max                         =  0.6169 a.u.
     k_min                         =  0.6169 a.u.
     path_thr                      =  0.1000 eV / A
     initial path length           = 12.2601 bohr
     initial inter-image distance  =  2.0434 bohr

It has more than 400 iterations and has not converged. Should I change 
some settings, such as ds, or k_max and k_min?
Below are the activation energies for the last few iterations, and they 
are not monotone decreasing but oscillating.
Would anybody give me some suggestions?

     activation energy (->) =   0.273443 eV
     activation energy (->) =   0.273431 eV
     activation energy (->) =   0.273719 eV
     activation energy (->) =   0.273765 eV
     activation energy (->) =   0.273207 eV
     activation energy (->) =   0.273148 eV
     activation energy (->) =   0.273311 eV
     activation energy (->) =   0.273501 eV
     activation energy (->) =   0.272957 eV
     activation energy (->) =   0.272818 eV
     activation energy (->) =   0.272865 eV
     activation energy (->) =   0.273155 eV
     activation energy (->) =   0.272750 eV
     activation energy (->) =   0.272463 eV
     activation energy (->) =   0.272443 eV
     activation energy (->) =   0.272777 eV
     activation energy (->) =   0.272574 eV
     activation energy (->) =   0.272118 eV
     activation energy (->) =   0.272080 eV
     activation energy (->) =   0.272367 eV
     activation energy (->) =   0.273228 eV
     activation energy (->) =   0.272663 eV
     activation energy (->) =   0.272621 eV
     activation energy (->) =   0.272935 eV
     activation energy (->) =   0.273033 eV
     activation energy (->) =   0.272427 eV
     activation energy (->) =   0.272386 eV
     activation energy (->) =   0.272558 eV
     activation energy (->) =   0.272823 eV
     activation energy (->) =   0.272391 eV

Yours,
Ding


From ding at sissa.it  Sun Aug 27 09:51:36 2006
From: ding at sissa.it (Xunlei Ding)
Date: Sun, 27 Aug 2006 09:51:36 +0200
Subject: [Pw_forum] What to do when NEB calculation is difficult to converge?
In-Reply-To: <44F14C2E.4080503@sissa.it>
References: <44F14C2E.4080503@sissa.it>
Message-ID: <44F14F08.8040301@sissa.it>

I forgot to give the errors for the last iteration.

     image        energy (eV)        error (eV/A)        frozen

         1     -37090.9593488            0.017636            T
         2     -37090.9338352            0.063849            F
         3     -37090.8813092            0.184154            F
         4     -37090.8113275            0.294601            F
         5     -37090.6869583            0.182737            F
         6     -37090.8612055            0.071489            F
         7     -37091.1264731            0.013874            T

Yours,
Ding

Xunlei Ding wrote:

> Dear all,
> I am now doing a neb calculation on a surface adsorption system with 
> pwscf3.0
> The default values are used in the calculation as below:
>     calculation                   =  neb
>     restart_mode                  =  restart
>     opt_scheme                    =  quick-min
>     num_of_images                 =  7
>     nstep                         =  500
>     first_last_opt                =  F
>     coarse-grained phase-space    =  F
>     use_freezing                  =  F
>     ds                            =  1.0000 a.u.
>     CI_scheme                     =  highest-TS
>     k_max                         =  0.6169 a.u.
>     k_min                         =  0.6169 a.u.
>     path_thr                      =  0.1000 eV / A
>     initial path length           = 12.2601 bohr
>     initial inter-image distance  =  2.0434 bohr
>
> It has more than 400 iterations and has not converged. Should I change 
> some settings, such as ds, or k_max and k_min?
> Below are the activation energies for the last few iterations, and 
> they are not monotone decreasing but oscillating.
> Would anybody give me some suggestions?
>
>     activation energy (->) =   0.273443 eV
>     activation energy (->) =   0.273431 eV
>     activation energy (->) =   0.273719 eV
>     activation energy (->) =   0.273765 eV
>     activation energy (->) =   0.273207 eV
>     activation energy (->) =   0.273148 eV
>     activation energy (->) =   0.273311 eV
>     activation energy (->) =   0.273501 eV
>     activation energy (->) =   0.272957 eV
>     activation energy (->) =   0.272818 eV
>     activation energy (->) =   0.272865 eV
>     activation energy (->) =   0.273155 eV
>     activation energy (->) =   0.272750 eV
>     activation energy (->) =   0.272463 eV
>     activation energy (->) =   0.272443 eV
>     activation energy (->) =   0.272777 eV
>     activation energy (->) =   0.272574 eV
>     activation energy (->) =   0.272118 eV
>     activation energy (->) =   0.272080 eV
>     activation energy (->) =   0.272367 eV
>     activation energy (->) =   0.273228 eV
>     activation energy (->) =   0.272663 eV
>     activation energy (->) =   0.272621 eV
>     activation energy (->) =   0.272935 eV
>     activation energy (->) =   0.273033 eV
>     activation energy (->) =   0.272427 eV
>     activation energy (->) =   0.272386 eV
>     activation energy (->) =   0.272558 eV
>     activation energy (->) =   0.272823 eV
>     activation energy (->) =   0.272391 eV
>
> Yours,
> Ding
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>


From lanhaiping at gmail.com  Sun Aug 27 13:27:17 2006
From: lanhaiping at gmail.com (lan haiping)
Date: Sun, 27 Aug 2006 19:27:17 +0800
Subject: [Pw_forum] CRASH of ph.x calculation in example06
Message-ID: <c92002fa0608270427k6584016fw4215d82a242eac47@mail.gmail.com>

 Dear all,

   I wanna obtain phonon spectrum of the irreducible BZ, then i tried to run
$(espresso)/examples/example06.
  But the code came to crash for matrix diagonalization error,  and just
gave  alas.dyn0, alas.dyn1, alas.dyn2 files .
  I then just changed the k_points in scf calculation  from 2 points to
mkgrid (4,4,4, 0,0,0),  and the calculation can be finished sucessfully.

 so, i would like to know whether this  error is due to math-lib or k-point
setting in scf calculation ?

Regards,

Hai-Ping
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060827/b3f1a612/attachment.htm 

From lanhaiping at gmail.com  Sun Aug 27 14:21:36 2006
From: lanhaiping at gmail.com (lan haiping)
Date: Sun, 27 Aug 2006 20:21:36 +0800
Subject: [Pw_forum] Re: CRASH of ph.x calculation in example06
In-Reply-To: <c92002fa0608270427k6584016fw4215d82a242eac47@mail.gmail.com>
References: <c92002fa0608270427k6584016fw4215d82a242eac47@mail.gmail.com>
Message-ID: <c92002fa0608270521h1a756d12n989328dcfc0ea605@mail.gmail.com>

Sorry ,
i then checked  the calculation of version 3.1.1, and didnot come to such
error.
Maybe this error is only in  Version 3.0.

I should move to the latest version.

regards,

hai-ping



On 8/27/06, lan haiping <lanhaiping at gmail.com> wrote:
>
>   Dear all,
>
>    I wanna obtain phonon spectrum of the irreducible BZ, then i tried to
> run $(espresso)/examples/example06.
>   But the code came to crash for matrix diagonalization error,  and just
> gave  alas.dyn0, alas.dyn1, alas.dyn2 files .
>   I then just changed the k_points in scf calculation  from 2 points to
> mkgrid (4,4,4, 0,0,0),  and the calculation can be finished sucessfully.
>
>  so, i would like to know whether this  error is due to math-lib or
> k-point setting in scf calculation ?
>
> Regards,
>
> Hai-Ping
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060827/2629ed27/attachment.htm 

From emenendez at macul.ciencias.uchile.cl  Sun Aug 27 14:16:14 2006
From: emenendez at macul.ciencias.uchile.cl (Eduardo Ariel Menendez P)
Date: Sun, 27 Aug 2006 08:16:14 -0400 (CLT)
Subject: [Pw_forum] how was generated Pt.pz-nd-rrkjus.UPF
Message-ID: <Pine.LNX.4.58.0608270757350.17031@macul.ciencias.uchile.cl>

Hello,
I need to generate pseudopotentials similar to Pt.pz-nd-rrkjus.UPF
that is available in the espresso  web site
The info is
<PP_INFO>
Generated using "atomic" code by A. Dal Corso  (espresso distribution)
Author: anonymous   Generation date: 15Mar2005
Pt
    1        The Pseudo was generated with a Scalar-Relativistic Calculation
    0 2.6000000E+00    L component and cutoff radius for Local Potential
nl pn  l   occ               Rcut            Rcut US             E pseu
6P  2  1  0.00      2.60000000000      2.60000000000     -0.46872226149
5D  3  2  9.00      2.10000000000      2.40000000000     -0.43874809514
5D  3  2  0.00      2.10000000000      2.40000000000     -0.20000000000
6S  1  0  1.00      2.60000000000      2.60000000000     -0.43874662441
</PP_INFO>

However, when I try to generate a similar pseudo using the following
input

 &input
        title='Pt',
        zed=78.0,
        rel=1,
        iswitch=3,
        rlderiv=2.50,
        eminld=-4.0,
        emaxld=4.0,
        deld=0.02,
        nld=4,
        config='[Xe] 4f14 5d8 6s2 6p0',
        dft='PZ',
        prefix='pt.pz'
 /
 &inputp
        pseudotype=3,
        lloc=0,
        file_pseudopw='Pt.pz-rkkj3-us2.UPF',
        nlcc=.true.,
        rcore=1.2,
        rcloc=2.5
 /
4
6P  2  1  0.00 -0.00  2.20  2.60
5D  3  2  8.00  0.00  2.15  2.40
5D  3  2  0.00 -0.20  2.15  2.40
6S  1  0  2.00  0.00  2.15  2.60

I get the output

 ld1.x< pt.pp-us.in

     program ld1 starts. version 3.1
     today is 27Aug2006 at  9: 5:41



%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     from ld1_readin : error #         1
     US requires at least two energies per channel

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

     stopping ...

The meaning of the message is clear, but the existing ppseudopotential was
generated with some version of ld1.x that allowed 1 projector per channel.

What version?

Is the input file available?

What is the problem using less than two energies per channel?

Thanks


Eduardo A. Menendez Proupin
Department of Physics
Faculty of Science
University of Chile
Las Palmeras 3425
?u?oa, Santiago
Chile


From stewart at cnf.cornell.edu  Sun Aug 27 18:37:20 2006
From: stewart at cnf.cornell.edu (stewart at cnf.cornell.edu)
Date: Sun, 27 Aug 2006 12:37:20 -0400
Subject: [Pw_forum] Problems running pwcond.x : error =?utf-8?B?InByb2JsZW1z?= with
 the linear =?utf-8?B?c3lzdGVtIg==?=
In-Reply-To: <44EF438E.9030803@mit.edu>
References: <20060825180353.76227.qmail@web52005.mail.yahoo.com>
            <44EF438E.9030803@mit.edu>
Message-ID: <20060827163720.22256.qmail@xuxa.iecc.com>

Hi everyone, 

   I am running into the following error when I run pwcond.x with both 
version 3.0 and version 3.1.1 

   from transmit : error #       891
    problems with the linear system 

   I am trying to run some transmission calculations on an infinite coaxial 
nanowire system using pwcond.x  Based on example 12 that came with the 
distribution, I assumed that I could take the pw calculation for the base 
unit and then use this for the calculation for the open system using 
ikind=1.  The nanowire consists of 14 atoms (two six atoms rings separated 
by single layers with a central atom).  Any suggestions from people who have 
used pwcond.x in the past would be greatly appreciated! 

Thanks, 

Derek 

The input file for pwcond is posted below: 

&inputcond
 outdir='/scratch/derek/pwscf/pd_nanowire/coax_6_1/7.435_larger_unit_cell/ve 
rsion_3.1.1_run/'
   prefixl='pd',
   prefixs='pd',
   tran_file='trans.coax61pd'
   ikind=1
   iofspin=1
   energy0=0.6d0
   denergy=-0.1d0
   ewind=1.d0
   epsproj=1.d-3
   nz1=4
   cutplot = 1.d0
/
   1
   0. 0. 1.0
   20 

 

################################
Derek Stewart, Ph. D.
Scientific Computation Associate
250 Duffield Hall
Cornell Nanoscale Facility (CNF)
Ithaca, NY 14853
stewart (at) cnf.cornell.edu
(607) 255-2856 



From giannozz at nest.sns.it  Sun Aug 27 21:58:57 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Sun, 27 Aug 2006 21:58:57 +0200
Subject: [Pw_forum] ld1.x normalization and overlap integrals
In-Reply-To: <Pine.LNX.4.58.0608261326470.7890@macul.ciencias.uchile.cl>
References: <Pine.LNX.4.58.0608261326470.7890@macul.ciencias.uchile.cl>
Message-ID: <200608272158.57782.giannozz@nest.sns.it>

On Saturday 26 August 2006 19:36, Eduardo Ariel Menendez P wrote:

> Doing an all-electron calculation with atomic code (ld1.x) I can read
> from the output that the overlap integrals are not zero
> [...] 
> In my case, the overlap integrals are 0.000000 if rel=0 or rel=2,
> and give the above numbers if rel=1.

if rel=0 or rel=2 the overlap integrals are respectively calculated 
between atomic orbitals (non-relativistic case: Schroedinger 
Kohn-Sham equation) or between spinors ((fully relativistic case:
Dirac-like equation from the relativistic extension of DFT).

In the rel=1 (the so-called scalar relativistic case) only the main
relativistic corrections are evaluated and only the major component 
of the spinor is actually calculated. In this case the "overlap" integrals
are missing the contribution of the minor components.

Paolo
-- 
Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From giannozz at nest.sns.it  Sun Aug 27 22:02:08 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Sun, 27 Aug 2006 22:02:08 +0200
Subject: [Pw_forum] Publications resulting from PWscf/Quantum Espresso
In-Reply-To: <20060825180353.76227.qmail@web52005.mail.yahoo.com>
References: <20060825180353.76227.qmail@web52005.mail.yahoo.com>
Message-ID: <200608272202.08264.giannozz@nest.sns.it>

On Friday 25 August 2006 20:03, Konstantin Kudin wrote:

>  While were are on the subject I think it is also worthwhile to
> consider setting up a list of references on the methods included into
> QE (as opposed just to the applications alone).

there is a partial list of such references in Doc/refman.tex

Paolo
-- 
Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From yuhailin_79 at msn.com  Mon Aug 28 04:23:25 2006
From: yuhailin_79 at msn.com (hailin yu)
Date: Mon, 28 Aug 2006 10:23:25 +0800
Subject: [Pw_forum] help: the phonon calculation
Message-ID: <BAY106-F750F541C0653D2FDACCAF93380@phx.gbl>

Dear all:
     I have some quesitons want to know. Can you tell me ? Thanks a lot!
1).Recently, I had calculated Ag3N which has the anti-ReO3 strucuture (i.e, 
has the same structure with Cu3N,PRB 53,12648) by using CASTEP and found 
it's mechnical stabel  ( I find the elastic siffness coefficients of Ag3N 
satisfy the generalized elastic stability criteria). But when I calculated 
the phonon dispersion of Ag3N using PWSCF, I found there was a soft mode 
along G-X direcition (i.e the phonon frequency is negative). For a negative 
frequency appear, it means that the structure is not stable? If so, the 
results of CASTEP and PWSCF is confict? This quesiton is still confuzed me 
and I am eager to receive your reply. Thanks!

2). For f-electronic materials (such as Pt), for their lattice dynamics 
properties, it is reasonable to calculate using PWSCF ?

Thanks!

     yuhailin

_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today it's FREE! 
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/



From prasenjit at jncasr.ac.in  Mon Aug 28 05:21:48 2006
From: prasenjit at jncasr.ac.in (PRASENJIT GHOSH)
Date: Mon, 28 Aug 2006 08:51:48 +0530 (IST)
Subject: [Pw_forum] BFGS convergence
Message-ID: <39171.202.41.111.151.1156735308.squirrel@202.41.111.151>

Dear all,

I'm trying to relax 4layer MgO(100) surface with IML Rh atom & NO on it.
I'm keeping the bottom 2 layers fixed & relaxing rest of the layers along
all directions. I have set forc_conv_thr as 5.0D-4. When the BFGS
optimization is complete the total force printed out is 0.001184 Ry/au.,
which is higher than the threshold.

The force on individual atoms are as given below:

     atom   1 type  4   force =     0.00000000    0.00019439    0.00035388
     atom   2 type  2   force =     0.00000000    0.00003994    0.00030581
     atom   3 type  3   force =     0.00004990    0.00006395    0.00026932
     atom   4 type  3   force =    -0.00004990    0.00006395    0.00026932
     atom   5 type  3   force =     0.00000510   -0.00010716    0.00017645
     atom   6 type  3   force =    -0.00000510   -0.00010716    0.00017645
     atom   7 type  3   force =    -0.00015740    0.00021421    0.00025840
     atom   8 type  3   force =     0.00015740    0.00021421    0.00025840
     atom   9 type  1   force =     0.00000000    0.00017531    0.00012835
     atom  10 type  1   force =     0.00000000    0.00012586    0.00009343
     atom  11 type  2   force =    -0.00003854   -0.00009570   -0.00010196
     atom  12 type  2   force =     0.00003854   -0.00009570   -0.00010196
     atom  13 type  1   force =     0.00000000    0.00002684    0.00020808
     atom  14 type  1   force =     0.00000000   -0.00008463    0.00035348
     atom  15 type  2   force =    -0.00002782   -0.00001293    0.00001952
     atom  16 type  2   force =     0.00002782   -0.00001293    0.00001952
     atom  17 type  1   force =     0.00000000    0.00001629    0.00009910
     atom  18 type  1   force =     0.00000000   -0.00003038    0.00014986
     atom  19 type  2   force =     0.00000763    0.00008195    0.00007784
     atom  20 type  2   force =    -0.00000763    0.00008195    0.00007784
     atom  21 type  2   force =     0.00000000    0.00003195    0.00010070
     atom  22 type  2   force =     0.00000000    0.00003854    0.00016380
     atom  23 type  1   force =    -0.00010405   -0.00006846    0.00012527
     atom  24 type  1   force =     0.00010405   -0.00006846    0.00012527
     atom  25 type  2   force =     0.00000000   -0.00002769    0.00001703
     atom  26 type  2   force =     0.00000000   -0.00007519    0.00008055
     atom  27 type  1   force =    -0.00004277    0.00010317   -0.00017790
     atom  28 type  1   force =     0.00004277    0.00010317   -0.00017790
     atom  29 type  2   force =     0.00000000    0.00001178    0.00015564
     atom  30 type  2   force =     0.00000000    0.00009848    0.00005032
     atom  31 type  1   force =    -0.00009405   -0.00007765    0.00004686
     atom  32 type  1   force =     0.00009405   -0.00007765    0.00004686
     atom  33 type  1   force =     0.00000000   -0.00027438   -0.00258973
     atom  34 type  1   force =     0.00000000    0.00085927   -0.00230357
     atom  35 type  2   force =    -0.00002762    0.00106477    0.00266372
     atom  36 type  2   force =     0.00002762    0.00106477    0.00266372
     atom  37 type  1   force =     0.00000000   -0.00127835   -0.00262410
     atom  38 type  1   force =     0.00000000   -0.00187757   -0.00264360
     atom  39 type  2   force =     0.00033189   -0.00088088    0.00323964
     atom  40 type  2   force =    -0.00033189   -0.00088088    0.00323964
     atom  41 type  1   force =     0.00000000    0.00132099   -0.00267980
     atom  42 type  1   force =     0.00000000    0.00088146   -0.00307944
     atom  43 type  2   force =     0.00043540   -0.00046399    0.00208586
     atom  44 type  2   force =    -0.00043540   -0.00046399    0.00208586
     atom  45 type  2   force =     0.00000000   -0.00001967   -0.01458796
     atom  46 type  2   force =     0.00000000    0.00001272   -0.01458917
     atom  47 type  1   force =    -0.00001307   -0.00005348    0.01390984
     atom  48 type  1   force =     0.00001307   -0.00005348    0.01390984
     atom  49 type  2   force =     0.00000000   -0.00068276   -0.01433952
     atom  50 type  2   force =     0.00000000   -0.00070884   -0.01425615
     atom  51 type  1   force =     0.00001464   -0.00004340    0.01348247
     atom  52 type  1   force =    -0.00001464   -0.00004340    0.01348247
     atom  53 type  2   force =     0.00000000    0.00075245   -0.01418097
     atom  54 type  2   force =     0.00000000    0.00074367   -0.01430795
     atom  55 type  1   force =     0.00003199    0.00014034    0.01388566
     atom  56 type  1   force =    -0.00003199    0.00014034    0.01388566

     Total force =     0.001184     Total SCF correction =     0.000147


Atoms 33-56 are kept fixed. Hence the forces on them is high whereas the
forces on each of the individual components of the relaxed atoms are below
the threshold.

My question is when the code calculates the force (grad) & their
differencec between 2 consecutive BFGS iter. (grad_error) & is it only for
the atoms which are allowed to relax or for the full system or something
is going wrong in the calculation? I'm using version 3.0.

With regards,
Prasenjit.

PRASENJIT GHOSH,
Ph. D STUDENT,
THEORETICAL SCIENCES UNIT,
JAWAHARLAL NEHRU CENTRE,
JAKKUR P. O.,
BANGALORE - 560064,
INDIA.

PHONE:+91-80-22082835 (OFFICE)
            +91 9880519401 (MOBILE)


From guobaozeng at yahoo.com.cn  Mon Aug 28 05:48:58 2006
From: guobaozeng at yahoo.com.cn (=?gb2312?q?=B9=F9=B1=A6=D4=F6?=)
Date: Mon, 28 Aug 2006 11:48:58 +0800 (CST)
Subject: [Pw_forum] =?gb2312?q?re=A3=BA=20Welcome=20to=20the=20"Pw=5Fforum"=20mailing=20list?=
 =?gb2312?q?=20(Digest=20mode)?=
In-Reply-To: <20060828034701.21699.85036.Mailman@democritos.sissa.it>
Message-ID: <20060828034858.23710.qmail@web15905.mail.cnb.yahoo.com>



pw_forum-request at pwscf.org ???  Welcome to the Pw_forum at pwscf.org mailing list!

To post to this list, send your email to:

pw_forum at pwscf.org

General information about the mailing list is at:

http://www.democritos.it/mailman/listinfo/pw_forum

If you ever want to unsubscribe or change your options (eg, switch to
or from digest mode, change your password, etc.), visit your
subscription page at:

http://www.democritos.it/mailman/options/pw_forum/guobaozeng%40yahoo.com.cn


You can also make such adjustments via email by sending a message to:

Pw_forum-request at pwscf.org

with the word `help' in the subject or body (don't include the
quotes), and you will get back a message with instructions.

You must know your password to change your options (including changing
the password, itself) or to unsubscribe. It is:

rocket5

If you forget your password, don't worry, you will receive a monthly
reminder telling you what all your pwscf.org mailing list passwords
are, and how to unsubscribe or change your options. There is also a
button on your options page that will email your current password to
you.

You may also have your password mailed to you automatically from the
Web page noted above.


 		
---------------------------------
????????-3.5G???20M??? 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060828/60c59158/attachment.htm 

From lanhaiping at gmail.com  Mon Aug 28 07:43:21 2006
From: lanhaiping at gmail.com (lan haiping)
Date: Mon, 28 Aug 2006 13:43:21 +0800
Subject: [Pw_forum] help: the phonon calculation
In-Reply-To: <BAY106-F750F541C0653D2FDACCAF93380@phx.gbl>
References: <BAY106-F750F541C0653D2FDACCAF93380@phx.gbl>
Message-ID: <c92002fa0608272243w33a3f6e0h4d5a1e4665ed7847@mail.gmail.com>

On 8/28/06, hailin yu <yuhailin_79 at msn.com> wrote:
>
> Dear all:
>     I have some quesitons want to know. Can you tell me ? Thanks a lot!
> 1).Recently, I had calculated Ag3N which has the anti-ReO3 strucuture (i.e
> ,
> has the same structure with Cu3N,PRB 53,12648) by using CASTEP and found
> it's mechnical stabel  ( I find the elastic siffness coefficients of Ag3N
> satisfy the generalized elastic stability criteria). But when I calculated
> the phonon dispersion of Ag3N using PWSCF, I found there was a soft mode
> along G-X direcition (i.e the phonon frequency is negative). For a
> negative
> frequency appear, it means that the structure is not stable? If so, the
> results of CASTEP and PWSCF is confict? This quesiton is still confuzed me
> and I am eager to receive your reply. Thanks!



 you should give your input setting of calculation

2). For f-electronic materials (such as Pt), for their lattice dynamics
> properties, it is reasonable to calculate using PWSCF ?


as far as i know, there is no problem for pwscf to calculate metal's lattice
dynamics ,
since it is  Pseudopotential -plane wave method..


regards,

hai-ping

Thanks!
>
>     yuhailin
>
> _________________________________________________________________
> Express yourself instantly with MSN Messenger! Download today it's FREE!
> http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060828/85191569/attachment.htm 

From stevendixonmail at yahoo.com  Mon Aug 28 08:06:34 2006
From: stevendixonmail at yahoo.com (Steven Dixon)
Date: Sun, 27 Aug 2006 23:06:34 -0700 (PDT)
Subject: [Pw_forum] whether PWSCF can calculte the transport spectral function
Message-ID: <20060828060634.17357.qmail@web58301.mail.re3.yahoo.com>

Dear all:
 I am a new user of PWSCF. I write my questions as follows:
 In the PHYSICAL REVIEW B, VOLUME 57, NUMBER 18  11276, it introduce the
 electron-phonon intreaction spectral function of superconductor and the transport
 spectral function, and it said the ef&#64257;ciency factor is the only difference between
 these two functions. I know that PWSCF can calculte the electron-phonon intreaction
 spectral function, I want to know whether PWSCF can calculte the transport
 spectral function.
 
 Steven Dixon
 
 
 			
---------------------------------
Get your own web address for just $1.99/1st yr. We'll help. Yahoo! Small Business.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060827/552f97ac/attachment.htm 

From lanhaiping at gmail.com  Mon Aug 28 08:16:33 2006
From: lanhaiping at gmail.com (lan haiping)
Date: Mon, 28 Aug 2006 14:16:33 +0800
Subject: [Pw_forum] whether PWSCF can calculte the transport spectral function
In-Reply-To: <20060828060634.17357.qmail@web58301.mail.re3.yahoo.com>
References: <20060828060634.17357.qmail@web58301.mail.re3.yahoo.com>
Message-ID: <c92002fa0608272316s4f62efd6t704f49321347df02@mail.gmail.com>

you means electron-phonon interaction coefficient ?

as far as i know, pwscf can do it for metal systems ..

regards,

hai-ping


On 8/28/06, Steven Dixon <stevendixonmail at yahoo.com> wrote:
>
> Dear all:
> I am a new user of PWSCF. I write my questions as follows:
> In the PHYSICAL REVIEW B, VOLUME 57, NUMBER 18  11276, it introduce the
> electron-phonon intreaction spectral function of superconductor and the
> transport
> spectral function, and it said the ef?ciency factor is the only difference
> between
> these two functions. I know that PWSCF can calculte the electron-phonon
> intreaction
> spectral function, I want to know whether PWSCF can calculte the transport
> spectral function.
>
>
> Steven Dixon
>
>  ------------------------------
> Get your own web address for just $1.99/1st yr. We'll help. Yahoo! Small
> Business<http://us.rd.yahoo.com/evt=41244/*http:/smallbusiness.yahoo.com/>.
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060828/f8e94e1b/attachment.htm 

From dalcorso at sissa.it  Mon Aug 28 09:25:58 2006
From: dalcorso at sissa.it (Andrea Dal Corso)
Date: Mon, 28 Aug 2006 09:25:58 +0200
Subject: [Pw_forum] Problems running pwcond.x : error "problems with
	the linear system"
In-Reply-To: <20060827163720.22256.qmail@xuxa.iecc.com>
References: <20060825180353.76227.qmail@web52005.mail.yahoo.com>
	 <44EF438E.9030803@mit.edu>  <20060827163720.22256.qmail@xuxa.iecc.com>
Message-ID: <1156749959.4009.34.camel@dhpc-5-50.sissa.it>

I suggest that you first try to get the complex band structure correctly
and then the transmission. You need to converge the parameters ewind,
epsproj and nz1 in order to have correct band structures before
computing transmission. Ewind and epsproj control the number of basis
functions used in the calculation (more details in the original paper of
Choi and Ihm PRB 59, 2267 (1999)). 
A very accurate calculation is done with a large ewind (4.0 or 5.0) and
a small epsproj (10^-8, 10^-9).
The actual basis set reduction is written in the output. For instance in
the example of the alwire:

 ngper, n2d =          177          37

where ngper gives the number of G_perpendicular vectors (that is the
basis size for a fully converged calculation) and the actual number n2d
of basis functions used.
You have to find a reasonable compromise between accuracy and cpu time
decreasing ewind and increasing epsproj in such a way to reduce the
basis until the complex bands remain accurate.

Also the number of points used to make the integrals in each slab nz1
could be small in your case. The default for nz1 is 11.  

Finally sometimes the routine for solving the generalized eigenvalue
problem does not find all the complex eigenvalues and this reflects in
the impossibility to solve the linear system in the transmission
calculation.
If this is the case you can try to improve the situation adding a very
small number on the diagonal of the A and B matrices of the generalized
eigenvalue problem. This is done with the variable delgep. A value
10^-9, 10^-8 usually solves the problem without changing the
eigenvalues.

Hope this helps,

Andrea


On Sun, 2006-08-27 at 12:37 -0400, stewart at cnf.cornell.edu wrote:
>    ewind=1.d0
>    epsproj=1.d-3
>    nz1=4
-- 
Andrea Dal Corso                    Tel. 0039-040-3787428
SISSA, Via Beirut 2/4               Fax. 0039-040-3787528
34014 Trieste (Italy)               e-mail: dalcorso at sissa.it



From baroni at sissa.it  Mon Aug 28 13:44:57 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Mon, 28 Aug 2006 13:44:57 +0200
Subject: [Pw_forum] help: the phonon calculation
In-Reply-To: <c92002fa0608272243w33a3f6e0h4d5a1e4665ed7847@mail.gmail.com>
References: <BAY106-F750F541C0653D2FDACCAF93380@phx.gbl> <c92002fa0608272243w33a3f6e0h4d5a1e4665ed7847@mail.gmail.com>
Message-ID: <FC95DE8B-E093-48E6-8CC1-E64853E57833@sissa.it>


On Aug 28, 2006, at 7:43 AM, lan haiping wrote:

>
>
> On 8/28/06, hailin yu <yuhailin_79 at msn.com> wrote:
> Dear all:
>     I have some quesitons want to know. Can you tell me ? Thanks a  
> lot!
> 1).Recently, I had calculated Ag3N which has the anti-ReO3  
> strucuture ( i.e,
> has the same structure with Cu3N,PRB 53,12648) by using CASTEP and  
> found
> it's mechnical stabel  ( I find the elastic siffness coefficients  
> of Ag3N
> satisfy the generalized elastic stability criteria). But when I  
> calculated
> the phonon dispersion of Ag3N using PWSCF, I found there was a soft  
> mode
> along G-X direcition (i.e the phonon frequency is negative). For a  
> negative
> frequency appear, it means that the structure is not stable? If so,  
> the
> results of CASTEP and PWSCF is confict? This quesiton is still  
> confuzed me
> and I am eager to receive your reply. Thanks!
>
>
>  you should give your input setting of calculation
>

This question deserves a more general answer that does not depend on  
any specific input of any specific calculation. The answer is kind of  
trivial, once one fully understands the question, but I am afraid  
that Halin Yu has some problems with the question itself. Suppose you  
are at the saddle between two valleys in a mountain landscape. Forces  
acting on you are zero. Is the equilibrium stable or unstable? Well,  
it depends on the direction: if you keep climbing up to the top of  
one mountain nearby, you will say that the equilibrium is stable. If  
you roll downhill, you would say instead that the equilibrium is  
unstable. Suppose now that your map only indicates the path climbing  
up. You would positively conclude that the euilibrium is stable. This  
is the conclusion that you drew from your CASTEP calculation that  
state that the equilibrium is stable with respect to distortions  
which change the shape of the unit cell (elastic stability criteria).  
These criteria do not say anything about the stability with respect  
to a phonon distortion which, by definition, does not change the  
shape of the unit cell.

> 2). For f-electronic materials (such as Pt), for their lattice  
> dynamics
> properties, it is reasonable to calculate using PWSCF ?
>
> as far as i know, there is no problem for pwscf to calculate  
> metal's lattice dynamics ,
> since it is  Pseudopotential -plane wave method..

as long as you do things properly ...

Stefano B.

>
>
> regards,
>
> hai-ping
>
> Thanks!
>
>     yuhailin
>
> _________________________________________________________________
> Express yourself instantly with MSN Messenger! Download today it's  
> FREE!
> http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060828/3ba9fd73/attachment.htm 

From wierzbom at ts.infn.it  Mon Aug 28 14:58:57 2006
From: wierzbom at ts.infn.it (Malgorzata Wierzbowska)
Date: Mon, 28 Aug 2006 14:58:57 +0200 (CEST)
Subject: [Pw_forum] whether PWSCF can calculte the transport spectral
 function
In-Reply-To: <20060828060634.17357.qmail@web58301.mail.re3.yahoo.com>
References: <20060828060634.17357.qmail@web58301.mail.re3.yahoo.com>
Message-ID: <Pine.LNX.4.61.0608281441490.9973@dawn.ts.infn.it>

On Sun, 27 Aug 2006, Steven Dixon wrote:

> Dear all:
> I am a new user of PWSCF. I write my questions as follows:
> In the PHYSICAL REVIEW B, VOLUME 57, NUMBER 18  11276, it introduce the
> electron-phonon intreaction spectral function of superconductor and the transport
> spectral function, and it said the ef&#64257;ciency factor is the only difference between
> these two functions. I know that PWSCF can calculte the electron-phonon intreaction
> spectral function, I want to know whether PWSCF can calculte the transport
> spectral function.
>
> Steven Dixon
>

   No pwscf does not compute transport a2F. In order to do this you may
compare equations 4,5, 8 and 9 from the paper you cite. Then you can modify
the code in file PH/elphon.f90, subroutine elphsum.
You grep on "la2F" and you go to the place where the phonon
linewidth "gamma" is calculated (around line 600).
Gamma is given as a function of the mode "nu" and the broadening "isig".

Since the efficiency factor depends both on mode and on k-point,
you have to start your changes where gamma depends
on k, in objects: gf, ctemp, noint, g3.

Do you use option la2F=.true., or you do it in older fashion?

Please let me know whether you see what needs to be done.
If not I will change this part.

  Have fun,
   Gosia



>
>
> ---------------------------------
> Get your own web address for just $1.99/1st yr. We'll help. Yahoo! Small Business.


From giannozz at nest.sns.it  Mon Aug 28 15:39:05 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Mon, 28 Aug 2006 15:39:05 +0200
Subject: [Pw_forum] phonon, gk_sort error
In-Reply-To: <Pine.LNX.4.63.0608211802550.9242@civet.berkeley.edu>
References: <Pine.LNX.4.63.0608211802550.9242@civet.berkeley.edu>
Message-ID: <200608281539.05141.giannozz@nest.sns.it>

On Tuesday 22 August 2006 03:32, Kevin Chan wrote:

>       from gk_sort : error #         1
>       array gk out-of-bounds
>
> The error does not occur when I set -npool 1.

I think the following should fix the problem:
change line 2171 of PW/pw_restart.f90 from
       DO ik = 1, nkstot
to
       DO ik = 1, nks
remove all *.x, recompile

Paolo
-- 
Paolo Giannozzi             Phone:   +39/050-509876
DEMOCRITOS and SNS          Fax:     +39/050-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From emenendez at macul.ciencias.uchile.cl  Mon Aug 28 15:56:45 2006
From: emenendez at macul.ciencias.uchile.cl (Eduardo Ariel Menendez P)
Date: Mon, 28 Aug 2006 09:56:45 -0400 (CLT)
Subject: [Pw_forum] how to saturate dangling bonds in surface
Message-ID: <Pine.LNX.4.58.0608280946090.8659@macul.ciencias.uchile.cl>

Hello,
I have a problem of surface defect in MgO. This is ionic with rocksalt
structure. In a supercell calculation one has a real surface and an
artificial surface that should mimic the bulk material. ?Are there
dangling bonds in the bulk-like surface? Must they be saturated with a
pseudo hidrogen (with charge one, at wich distance) ?
Thanks

Eduardo A. Menendez Proupin
Department of Physics
Faculty of Science
University of Chile
Las Palmeras 3425
?u?oa, Santiago
Chile
Phone: 56+2+978 74 11
http://fisica.ciencias.uchile.cl/~emenendez/


From ongphuongvu at yahoo.com  Tue Aug 29 03:54:46 2006
From: ongphuongvu at yahoo.com (vu ongphuong)
Date: Mon, 28 Aug 2006 18:54:46 -0700 (PDT)
Subject: [Pw_forum] Problem with q2r.x
Message-ID: <20060829015446.21651.qmail@web50615.mail.yahoo.com>

Dear all users
   
  I have done a self-consistent calculation with pw.x, then a phonon calculation at Gamma pwith ph.x. =>gaas.dynG
  After that I perform Fourier transformation with q2r.x  and run matdyn.x to recalculate phonon modes.
  This is my q2r.in
   
  &input
   fildyn='gaas.dyn', zasr='simple', flfrc='gaas1.fc'
 /
   
  I followed the example06 of PWSCF (for a grid of q-points, not only at Gamma as my case), but q2r.x give error message:
   
  forrtl: severe (19): invalid reference to variable in NAMELIST input, unit 5, file stdin, line 2, position 9
Image              PC        Routine            Line        Source             
q2r.x              082BD88C  Unknown               Unknown  Unknown
 
Stack trace terminated abnormally.
   
  Do the q2r.x run only in case of grid of q-points? and not in 1 point (Gamma)? 
  Please tell me how to set up the input-files of q2r.x in case of phonon calculation only at Gamma. 
  Thank you very much.

 		
---------------------------------
 All-new Yahoo! Mail - Fire up a more powerful email and get things done faster.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060828/cb44e8aa/attachment.htm 

From giannozz at nest.sns.it  Tue Aug 29 08:29:27 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Tue, 29 Aug 2006 08:29:27 +0200
Subject: [Pw_forum] Problem with q2r.x
In-Reply-To: <20060829015446.21651.qmail@web50615.mail.yahoo.com>
References: <20060829015446.21651.qmail@web50615.mail.yahoo.com>
Message-ID: <200608290829.27269.giannozz@nest.sns.it>

On Tuesday 29 August 2006 03:54, vu ongphuong wrote:

>   I followed the example06 of PWSCF (for a grid of q-points, not only at
> Gamma as my case), but q2r.x give error message:
>
>   forrtl: severe (19): invalid reference to variable in NAMELIST input,
> unit 5, file stdin, line 2, position 9 

this error message is very clear: there is a problem in your input data

>   Do the q2r.x run only in case of grid of q-points? and not in 1 point
> (Gamma)? 

there is no point in performing the Fourier interpolation if you have
just Gamma

Paolo
-- 
Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From giannozz at nest.sns.it  Tue Aug 29 11:50:53 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Tue, 29 Aug 2006 11:50:53 +0200
Subject: [Pw_forum] how to saturate dangling bonds in surface
In-Reply-To: <Pine.LNX.4.58.0608280946090.8659@macul.ciencias.uchile.cl>
References: <Pine.LNX.4.58.0608280946090.8659@macul.ciencias.uchile.cl>
Message-ID: <200608291150.54031.giannozz@nest.sns.it>

On Monday 28 August 2006 15:56, Eduardo Ariel Menendez P wrote:

> I have a problem of surface defect in MgO. This is ionic with rocksalt
> structure. In a supercell calculation one has a real surface and an
> artificial surface that should mimic the bulk material. ?Are there
> dangling bonds in the bulk-like surface? Must they be saturated with 
> a pseudo hidrogen (with charge one, at wich distance) ?

I am not sure what people actually does when calculating 
such an object, but I think it might be convenient to have 
two real surfaces. Surface relaxation in MgO shouldn't be 
significant beyond a few layers, so it should be feasible
with reasonably-sized supercells

Paolo
-- 
Paolo Giannozzi             Phone:   +39/050-509876
DEMOCRITOS and SNS          Fax:     +39/050-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From giannozz at nest.sns.it  Tue Aug 29 12:21:20 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Tue, 29 Aug 2006 12:21:20 +0200
Subject: [Pw_forum] BFGS convergence
In-Reply-To: <39171.202.41.111.151.1156735308.squirrel@202.41.111.151>
References: <39171.202.41.111.151.1156735308.squirrel@202.41.111.151>
Message-ID: <200608291221.20270.giannozz@nest.sns.it>

On Monday 28 August 2006 05:21, PRASENJIT GHOSH wrote:

> I'm keeping the bottom 2 layers fixed & relaxing rest of the layers along
> all directions. I have set forc_conv_thr as 5.0D-4. When the BFGS
> optimization is complete the total force printed out is 0.001184 Ry/au.,
> which is higher than the threshold.

this is not the quantity that is checked for convergence, though:
---
forc_conv_thr  REAL ( default = 1.0D-3 )
               convergence threshold on forces (a.u) for ionic
               minimization: the convergence criterion is satisfied
               when all components of all forces are smaller than
               forc_conv_thr.
---

> My question is when the code calculates the force (grad) & their
> differencec between 2 consecutive BFGS iter. (grad_error) & is it only for
> the atoms which are allowed to relax or for the full system or something
> is going wrong in the calculation? I'm using version 3.0.

forces on atoms that are kept fixed are not considered in
the convergence criterion, or else you will never achieve
convergence in presence of a constraint

Paolo
-- 
Paolo Giannozzi             Phone:   +39/050-509876
DEMOCRITOS and SNS          Fax:     +39/050-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From giannozz at nest.sns.it  Tue Aug 29 12:24:01 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Tue, 29 Aug 2006 12:24:01 +0200
Subject: [Pw_forum] Units in plots
In-Reply-To: <1156523914.5201.4.camel@localhost>
References: <1156523914.5201.4.camel@localhost>
Message-ID: <200608291224.01988.giannozz@nest.sns.it>

On Friday 25 August 2006 18:38, Detlef Schmicker wrote:

> Which unit is the total potential plotted if I use pp.x
>
> I expect Ryd

good guess. Potentials are actually e*V, thus they are in energy units.

> but I can not find any information in mail archive and documentation.

this kind of information is typically buried in the most 
obscure places in the code ...

Paolo
-- 
Paolo Giannozzi             Phone:   +39/050-509876
DEMOCRITOS and SNS          Fax:     +39/050-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From giannozz at nest.sns.it  Tue Aug 29 12:26:05 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Tue, 29 Aug 2006 12:26:05 +0200
Subject: [Pw_forum] help: the phonon calculation
In-Reply-To: <BAY106-F750F541C0653D2FDACCAF93380@phx.gbl>
References: <BAY106-F750F541C0653D2FDACCAF93380@phx.gbl>
Message-ID: <200608291226.05546.giannozz@nest.sns.it>

On Monday 28 August 2006 04:23, hailin yu wrote:

> 2). For f-electronic materials (such as Pt), for their lattice dynamics
> properties, it is reasonable to calculate using PWSCF ?

it is, as long as they are correctly described by DFT, and that they
do not need a disproportionate large cutoff; i.e., often it is not.

Paolo
-- 
Paolo Giannozzi             Phone:   +39/050-509876
DEMOCRITOS and SNS          Fax:     +39/050-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From prasenjit at jncasr.ac.in  Tue Aug 29 13:00:42 2006
From: prasenjit at jncasr.ac.in (PRASENJIT GHOSH)
Date: Tue, 29 Aug 2006 16:30:42 +0530 (IST)
Subject: [Pw_forum] BFGS convergence
In-Reply-To: <200608291221.20270.giannozz@nest.sns.it>
References: <39171.202.41.111.151.1156735308.squirrel@202.41.111.151>
    <200608291221.20270.giannozz@nest.sns.it>
Message-ID: <50541.202.41.111.151.1156849242.squirrel@202.41.111.151>

Dear Paolo,

Thanks for the clarification.


>> I'm keeping the bottom 2 layers fixed & relaxing rest of the layers
>> along
>> all directions. I have set forc_conv_thr as 5.0D-4. When the BFGS
>> optimization is complete the total force printed out is 0.001184 Ry/au.,
>> which is higher than the threshold.
>
> this is not the quantity that is checked for convergence, though:
> ---
> forc_conv_thr  REAL ( default = 1.0D-3 )
>                convergence threshold on forces (a.u) for ionic
>                minimization: the convergence criterion is satisfied
>            javascript:sqspellIgnoreAll()    when all components of all
forces are smaller than
>                forc_conv_thr.
> ---

That means grad_thr and  forc_conv_thr are two different quantities. Am I
correct? what is then grad_thr?

Moreover how important is it to satisfy Wolfe's condition? What I
understood is that Wolfe's condition ensures that one always goes
downhill. But under quadratic approx. grad.step < 0.0 already ensures that
one is going downhill
.

With regards,
Prasenjit.




PRASENJIT GHOSH,
Ph. D STUDENT,
THEORETICAL SCIENCES UNIT,
JAWAHARLAL NEHRU CENTRE,
JAKKUR P. O.,
BANGALORE - 560064,
INDIA.

PHONE:+91-80-22082835 (OFFICE)
            +91 9880519401 (MOBILE)


From d.schmicker at physik.de  Tue Aug 29 14:31:25 2006
From: d.schmicker at physik.de (Detlef Schmicker)
Date: Tue, 29 Aug 2006 14:31:25 +0200
Subject: [Pw_forum] Units in plots
In-Reply-To: <200608291224.01988.giannozz@nest.sns.it>
References: <1156523914.5201.4.camel@localhost>
	 <200608291224.01988.giannozz@nest.sns.it>
Message-ID: <1156854685.16794.2.camel@localhost>

Am Dienstag, den 29.08.2006, 12:24 +0200 schrieb Paolo Giannozzi:
> On Friday 25 August 2006 18:38, Detlef Schmicker wrote:
> 
> > Which unit is the total potential plotted if I use pp.x
> >
> > I expect Ryd
> 
> good guess. Potentials are actually e*V, thus they are in energy units.

Thanks a lot: Hartee and eV could have been possible too. 

> 
> > but I can not find any information in mail archive and documentation.
> 
> this kind of information is typically buried in the most 
> obscure places in the code ...

I hoped so, but I checked all files for Ryd and Hartree, and could not
find a comment explaining it.....

> 
> Paolo
-- 
Detlef Schmicker <d.schmicker at physik.de>



From stevendixonmail at yahoo.com  Tue Aug 29 15:40:01 2006
From: stevendixonmail at yahoo.com (Steven Dixon)
Date: Tue, 29 Aug 2006 06:40:01 -0700 (PDT)
Subject: [Pw_forum] Re: whether PWSCF can calculte the transport spectral
In-Reply-To: <20060829053743.4492.90735.Mailman@democritos.sissa.it>
Message-ID: <20060829134001.33829.qmail@web58309.mail.re3.yahoo.com>

Dear Gosia:
Thank you for your reply.
I know that group velocity can be calculated from the
derivative of the
band energy respect to kx, ky, kz. Then the efficiency
factor can be
calculated, then a2F. I have read some codes about
calculation of the
spectral function. But  I still don't know how to
calculate the efficiency
properly and add it to the transport spectral
function. Can you help me
to modify that codes, and sent it to me?

Another question: If the structure is anisotropy.
Whether the electrical
conductivity tensor can be calculate use this
theory/PWcodes?

Thanks in advance and best wishes

Steven Dixon

> > Dear all:
> > I am a new user of PWSCF. I write my questions as
> follows:
> > In the PHYSICAL REVIEW B, VOLUME 57, NUMBER 18 
> 11276, it introduce the
> > electron-phonon intreaction spectral function of
> superconductor and the transport
> > spectral function, and it said the
> ef&#64257;ciency factor is the only difference
> between
> > these two functions. I know that PWSCF can
> calculte the electron-phonon intreaction
> > spectral function, I want to know whether PWSCF
> can calculte the transport
> > spectral function.
> >
> > Steven Dixon
> >
> 
>    No pwscf does not compute transport a2F. In order
> to do this you may
> compare equations 4,5, 8 and 9 from the paper you
> cite. Then you can modify
> the code in file PH/elphon.f90, subroutine elphsum.
> You grep on "la2F" and you go to the place where the
> phonon
> linewidth "gamma" is calculated (around line 600).
> Gamma is given as a function of the mode "nu" and
> the broadening "isig".
> 
> Since the efficiency factor depends both on mode and
> on k-point,
> you have to start your changes where gamma depends
> on k, in objects: gf, ctemp, noint, g3.
> 
> Do you use option la2F=.true., or you do it in older
> fashion?
> 
> Please let me know whether you see what needs to be
> done.
> If not I will change this part.
> 
>   Have fun,
>    Gosia

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From wierzbom at ts.infn.it  Tue Aug 29 16:07:41 2006
From: wierzbom at ts.infn.it (Malgorzata Wierzbowska)
Date: Tue, 29 Aug 2006 16:07:41 +0200 (CEST)
Subject: [Pw_forum] Re: whether PWSCF can calculte the transport spectral
In-Reply-To: <20060829134001.33829.qmail@web58309.mail.re3.yahoo.com>
References: <20060829134001.33829.qmail@web58309.mail.re3.yahoo.com>
Message-ID: <Pine.LNX.4.61.0608291604570.22591@dawn.ts.infn.it>

On Tue, 29 Aug 2006, Steven Dixon wrote:

> Dear Gosia:
> Thank you for your reply.
> I know that group velocity can be calculated from the
> derivative of the
> band energy respect to kx, ky, kz. Then the efficiency
> factor can be
> calculated, then a2F. I have read some codes about
> calculation of the
> spectral function. But  I still don't know how to
> calculate the efficiency
> properly and add it to the transport spectral
> function. Can you help me
> to modify that codes, and sent it to me?
>

   Let us talk about the code between us and we will
   make it public at the end.

> Another question: If the structure is anisotropy.
> Whether the electrical
> conductivity tensor can be calculate use this
> theory/PWcodes?
>

   Yes. There is a  vector k-dependence.

   Gosia



> Thanks in advance and best wishes
>
> Steven Dixon
>
>>> Dear all:
>>> I am a new user of PWSCF. I write my questions as
>> follows:
>>> In the PHYSICAL REVIEW B, VOLUME 57, NUMBER 18
>> 11276, it introduce the
>>> electron-phonon intreaction spectral function of
>> superconductor and the transport
>>> spectral function, and it said the
>> ef&#64257;ciency factor is the only difference
>> between
>>> these two functions. I know that PWSCF can
>> calculte the electron-phonon intreaction
>>> spectral function, I want to know whether PWSCF
>> can calculte the transport
>>> spectral function.
>>>
>>> Steven Dixon
>>>
>>
>>    No pwscf does not compute transport a2F. In order
>> to do this you may
>> compare equations 4,5, 8 and 9 from the paper you
>> cite. Then you can modify
>> the code in file PH/elphon.f90, subroutine elphsum.
>> You grep on "la2F" and you go to the place where the
>> phonon
>> linewidth "gamma" is calculated (around line 600).
>> Gamma is given as a function of the mode "nu" and
>> the broadening "isig".
>>
>> Since the efficiency factor depends both on mode and
>> on k-point,
>> you have to start your changes where gamma depends
>> on k, in objects: gf, ctemp, noint, g3.
>>
>> Do you use option la2F=.true., or you do it in older
>> fashion?
>>
>> Please let me know whether you see what needs to be
>> done.
>> If not I will change this part.
>>
>>   Have fun,
>>    Gosia
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>


From niuyingli at yahoo.com.cn  Tue Aug 29 18:16:29 2006
From: niuyingli at yahoo.com.cn (=?gb2312?q?=D3=A2=C0=FB=20=C5=A3?=)
Date: Wed, 30 Aug 2006 00:16:29 +0800 (CST)
Subject: [Pw_forum] PWscf can not find out symmetry operations!!!
Message-ID: <20060829161630.89611.qmail@web15609.mail.cnb.yahoo.com>

Dear all,
  I am calculate properties of diamond
surface(001),the lattice parameters of the slab is :
2.5221A,5.0442A,16.0253A,90,90,90
after atoms relax ,the atom positions(crystal) are:
C     0.24978  -0.11238   0.21129 

C     0.24980   0.60988   0.21129 

C     0.74978  -0.01422   0.17583 

C     0.74980   0.51175   0.17582 

C     0.74981   0.24879   0.13029 

C     0.74981   0.74881   0.11624 

C     0.24982   0.24883   0.07780 

C     0.24982   0.74885   0.07061 

C     0.24984  -0.00014   0.02461 

C     0.24984   0.49787   0.02461 

C     0.74985  -0.00211  -0.02461 

C     0.74986   0.49988  -0.02461 

C     0.74987   0.24889  -0.07061 

C     0.74987   0.74888  -0.07780 

C     0.24987   0.24889  -0.11624 

C     0.24988   0.74888  -0.13029 

C     0.24990   0.01186  -0.17582 

C     0.24987   0.48589  -0.17583 

C     0.74990   0.11000  -0.21129 

C     0.74987   0.38773  -0.21129 


but PWscf can not find any symmetry operations of the
system.
I setup the struct and use Material Studio find 8
space group operations include inversion! The
tolerance is 0.01A.

Then I read PW manual:

the number of significant figures in the atomic
positions is not large enough. In file PW/eqvect.f90,
the variable accep is used to decide whether a
rotation is a symmetry operation. Its current value
(10-5 ) is quite strict: a rotated atom must coincide
with another atom to 5 significant digits. You may
change the value of accep and recompile.

I change accep from 1.0d-5 to 1.0d-2 then to 1.0d-1
,but PWscf still find no symmetry. 

What should I do? 
Why pwscf choose 1.0d-5,so strict?


	

	
		
___________________________________________________________ 
Mp3???-???????
http://music.yahoo.com.cn/?source=mail_mailbox_footer


From baroni at sissa.it  Tue Aug 29 18:24:05 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Tue, 29 Aug 2006 18:24:05 +0200
Subject: [Pw_forum] PWscf can not find out symmetry operations!!!
In-Reply-To: <20060829161630.89611.qmail@web15609.mail.cnb.yahoo.com>
References: <20060829161630.89611.qmail@web15609.mail.cnb.yahoo.com>
Message-ID: <0BF8D5DC-A855-44B6-8A18-1622DA2CA69A@sissa.it>

Dear Niu Ying Li,

this very same question has been discussed many times in this forum,  
the last of which only a few days ago. Would you be so kind as to  
browse the archive of the mailing list and revert us after finding  
and reading the some of the relevant posts, in case the situation is  
not yet clear to you?

Many thanks for your interest in Quantum ESPRESSO

Stefano Baroni

On Aug 29, 2006, at 6:16 PM, ?? ? wrote:

> Dear all,
>   I am calculate properties of diamond
> surface(001),the lattice parameters of the slab is :
> 2.5221A,5.0442A,16.0253A,90,90,90
> after atoms relax ,the atom positions(crystal) are:
> C     0.24978  -0.11238   0.21129
>
> C     0.24980   0.60988   0.21129
>
> C     0.74978  -0.01422   0.17583
>
> C     0.74980   0.51175   0.17582
>
> C     0.74981   0.24879   0.13029
>
> C     0.74981   0.74881   0.11624
>
> C     0.24982   0.24883   0.07780
>
> C     0.24982   0.74885   0.07061
>
> C     0.24984  -0.00014   0.02461
>
> C     0.24984   0.49787   0.02461
>
> C     0.74985  -0.00211  -0.02461
>
> C     0.74986   0.49988  -0.02461
>
> C     0.74987   0.24889  -0.07061
>
> C     0.74987   0.74888  -0.07780
>
> C     0.24987   0.24889  -0.11624
>
> C     0.24988   0.74888  -0.13029
>
> C     0.24990   0.01186  -0.17582
>
> C     0.24987   0.48589  -0.17583
>
> C     0.74990   0.11000  -0.21129
>
> C     0.74987   0.38773  -0.21129
>
>
> but PWscf can not find any symmetry operations of the
> system.
> I setup the struct and use Material Studio find 8
> space group operations include inversion! The
> tolerance is 0.01A.
>
> Then I read PW manual:
>
> the number of significant figures in the atomic
> positions is not large enough. In file PW/eqvect.f90,
> the variable accep is used to decide whether a
> rotation is a symmetry operation. Its current value
> (10-5 ) is quite strict: a rotated atom must coincide
> with another atom to 5 significant digits. You may
> change the value of accep and recompile.
>
> I change accep from 1.0d-5 to 1.0d-2 then to 1.0d-1
> ,but PWscf still find no symmetry.
>
> What should I do?
> Why pwscf choose 1.0d-5,so strict?
>
>
> 	
>
> 	
> 		
> ___________________________________________________________
> Mp3???-???????
> http://music.yahoo.com.cn/?source=mail_mailbox_footer
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060829/7e532f06/attachment.htm 

From sridevi at uprrp.edu  Wed Aug 30 00:06:03 2006
From: sridevi at uprrp.edu (Krishnan  Sridevi)
Date: Tue, 29 Aug 2006 18:06:03 -0400 (AST)
Subject: [Pw_forum] NEB calcualtion - Problem
Message-ID: <27151.136.145.186.69.1156889163.squirrel@wmail.uprrp.edu>

Dear all,

I try to run NEB calculation, but I have a problem.
I have compiled the serial version on an P4 system. i could run the
examples. I did a check example and was right till example 4.

This is the input file:
&CONTROL
                 calculation = 'neb' ,
                restart_mode = 'restart' ,
                      outdir =
'/home/Sridevi/Downloads/espresso-3.1.1/juanma/temp/'
,
                  pseudo_dir =
'/home/Sridevi/Downloads/espresso-3.1.1/pseudo/' ,
                      prefix = 'CH4O2_CH4OOH' ,
                       nstep = 0 ,
 /
 &SYSTEM
                       ibrav = 0,
                   celldm(1) = 1.D0,
                         nat = 7,
                        ntyp = 3,
                  tot_charge = 1,
                     ecutwfc = 20.0D0 ,
                     ecutrho = 100.0D0 ,
                 occupations = 'smearing' ,
                     degauss = 0.03D0 ,
                       nspin = 2 ,
   starting_magnetization(1) = 0.5D0,
 /
 &ELECTRONS
                    conv_thr = 1.D-8 ,
                 mixing_beta = 0.3D0 ,
 /
 &IONS
           pot_extrapolation = 'second_order' ,
           wfc_extrapolation = 'second_order' ,
               num_of_images = 8 ,
                  opt_scheme = 'broyden' ,
                          ds = 2.D0 ,
                    path_thr = 0.1D0 ,
                   CI_scheme = 'manual' ,
                       k_max = 0.3D0 ,
                       k_min = 0.2D0 ,
 /
CELL_PARAMETERS { cubic }
    12.000000000    0.000000000    0.000000000
     0.000000000    5.000000000    0.000000000
     0.000000000    0.000000000    5.000000000
ATOMIC_SPECIES
    H    1.00794   HUSPBE.RRKJ3
    C   12.005852  C.pbe-rrkjus.UPF
    O   16.002714  O.pbe-rrkjus.UPF
ATOMIC_POSITIONS { bohr }
first_image
    C       1.954897    -0.064367     0.000191  0 0 0
    H       3.028255    -0.250143    -0.002686  1 1 1
    H       1.802200     1.016517    -0.048979  1 1 1
    H       1.533716    -0.563048    -0.877534  1 1 1
    H       1.551082    -0.478320     0.928585  1 1 1
    O      -0.820010     0.362853    -0.000173  1 1 1
    O      -1.635569    -0.280204     0.000107  1 1 1
last_image
    C      -1.161432    0.181714     0.000017
    H      -0.969529    1.258642    -0.003615
    H      -1.675000   -0.151640    -0.902206
    H      -1.672119   -0.146187     0.905885
    H       0.960800    1.122493    -0.000061
    O       0.114715   -0.546742    -0.000042
    O       1.175840    0.150042     0.000028

K_POINTS gamma
CLIMBING_IMAGES
5

and the output file is the following:

 Program PWSCF     v.3.1.1  starts ...
     Today is 29Aug2006 at 18: 1:23

     Ultrasoft (Vanderbilt) Pseudopotentials

     Current dimensions of program pwscf are:

     ntypx = 10   npk = 40000  lmax =  3
     nchix =  6  ndmx =  2000  nbrx = 14  nqfx =  8

     RRKJ3 Ultrasoft PP for H


     gamma-point specific algorithms are used


     initial path length           =  8.3312 bohr
     initial inter-image distance  =  1.1902 bohr

     calculation                   =  neb
     restart_mode                  =  from_scratch
     opt_scheme                    =  broyden
     num_of_images                 =  8
     nstep                         =  0
     CI_scheme                     =  manual
    first_last_opt                =  F
     coarse-grained phase-space    =  F
     use_freezing                  =  F
     ds                            =  2.0000 a.u.
     k_max                         =  0.3000 a.u.
     k_min                         =  0.2000 a.u.
     suggested k_max               =  0.1542 a.u.
     suggested k_min               =  0.1028 a.u.
     path_thr                      =  0.1000 eV / A

     list of climbing images :   5,

     activation energy (->) =   0.000000 eV
     activation energy (<-) =   0.000000 eV

     image        energy (eV)        error (eV/A)        frozen

         1          0.0000000            0.000000            F
         2          0.0000000            0.000000            F
         3          0.0000000            0.000000            F
         4          0.0000000            0.000000            F
         5          0.0000000            0.000000            F
         6          0.0000000            0.000000            F
         7          0.0000000            0.000000            F
         8          0.0000000            0.000000            F

  path length          =  8.331 bohr
     inter-image distance =  1.190 bohr

     PWSCF        :     0.08s CPU time







     General routines

Please help.




From prasenjit at jncasr.ac.in  Wed Aug 30 05:33:03 2006
From: prasenjit at jncasr.ac.in (PRASENJIT GHOSH)
Date: Wed, 30 Aug 2006 09:03:03 +0530 (IST)
Subject: [Pw_forum] NEB calcualtion - Problem
In-Reply-To: <27151.136.145.186.69.1156889163.squirrel@wmail.uprrp.edu>
References: <27151.136.145.186.69.1156889163.squirrel@wmail.uprrp.edu>
Message-ID: <54277.202.41.111.151.1156908783.squirrel@202.41.111.151>

What is your problem?
If ur problem is that the forces are zero then that's fine. You have set
nstep=0 and hence the code doesn't do any iterations. It will only
generate the path. This is generally done to check whether the initial
guess of the path generated by the code is reasonable or not.

Prasenjit.

On Wed, August 30, 2006 3:36, Krishnan  Sridevi said:
> Dear all,
>
> I try to run NEB calculation, but I have a problem.
> I have compiled the serial version on an P4 system. i could run the
> examples. I did a check example and was right till example 4.
>
> This is the input file:
> &CONTROL
>                  calculation = 'neb' ,
>                 restart_mode = 'restart' ,
>                       outdir =
> '/home/Sridevi/Downloads/espresso-3.1.1/juanma/temp/'
> ,
>                   pseudo_dir =
> '/home/Sridevi/Downloads/espresso-3.1.1/pseudo/' ,
>                       prefix = 'CH4O2_CH4OOH' ,
>                        nstep = 0 ,
>  /
>  &SYSTEM
>                        ibrav = 0,
>                    celldm(1) = 1.D0,
>                          nat = 7,
>                         ntyp = 3,
>                   tot_charge = 1,
>                      ecutwfc = 20.0D0 ,
>                      ecutrho = 100.0D0 ,
>                  occupations = 'smearing' ,
>                      degauss = 0.03D0 ,
>                        nspin = 2 ,
>    starting_magnetization(1) = 0.5D0,
>  /
>  &ELECTRONS
>                     conv_thr = 1.D-8 ,
>                  mixing_beta = 0.3D0 ,
>  /
>  &IONS
>            pot_extrapolation = 'second_order' ,
>            wfc_extrapolation = 'second_order' ,
>                num_of_images = 8 ,
>                   opt_scheme = 'broyden' ,
>                           ds = 2.D0 ,
>                     path_thr = 0.1D0 ,
>                    CI_scheme = 'manual' ,
>                        k_max = 0.3D0 ,
>                        k_min = 0.2D0 ,
>  /
> CELL_PARAMETERS { cubic }
>     12.000000000    0.000000000    0.000000000
>      0.000000000    5.000000000    0.000000000
>      0.000000000    0.000000000    5.000000000
> ATOMIC_SPECIES
>     H    1.00794   HUSPBE.RRKJ3
>     C   12.005852  C.pbe-rrkjus.UPF
>     O   16.002714  O.pbe-rrkjus.UPF
> ATOMIC_POSITIONS { bohr }
> first_image
>     C       1.954897    -0.064367     0.000191  0 0 0
>     H       3.028255    -0.250143    -0.002686  1 1 1
>     H       1.802200     1.016517    -0.048979  1 1 1
>     H       1.533716    -0.563048    -0.877534  1 1 1
>     H       1.551082    -0.478320     0.928585  1 1 1
>     O      -0.820010     0.362853    -0.000173  1 1 1
>     O      -1.635569    -0.280204     0.000107  1 1 1
> last_image
>     C      -1.161432    0.181714     0.000017
>     H      -0.969529    1.258642    -0.003615
>     H      -1.675000   -0.151640    -0.902206
>     H      -1.672119   -0.146187     0.905885
>     H       0.960800    1.122493    -0.000061
>     O       0.114715   -0.546742    -0.000042
>     O       1.175840    0.150042     0.000028
>
> K_POINTS gamma
> CLIMBING_IMAGES
> 5
>
> and the output file is the following:
>
>  Program PWSCF     v.3.1.1  starts ...
>      Today is 29Aug2006 at 18: 1:23
>
>      Ultrasoft (Vanderbilt) Pseudopotentials
>
>      Current dimensions of program pwscf are:
>
>      ntypx = 10   npk = 40000  lmax =  3
>      nchix =  6  ndmx =  2000  nbrx = 14  nqfx =  8
>
>      RRKJ3 Ultrasoft PP for H
>
>
>      gamma-point specific algorithms are used
>
>
>      initial path length           =  8.3312 bohr
>      initial inter-image distance  =  1.1902 bohr
>
>      calculation                   =  neb
>      restart_mode                  =  from_scratch
>      opt_scheme                    =  broyden
>      num_of_images                 =  8
>      nstep                         =  0
>      CI_scheme                     =  manual
>     first_last_opt                =  F
>      coarse-grained phase-space    =  F
>      use_freezing                  =  F
>      ds                            =  2.0000 a.u.
>      k_max                         =  0.3000 a.u.
>      k_min                         =  0.2000 a.u.
>      suggested k_max               =  0.1542 a.u.
>      suggested k_min               =  0.1028 a.u.
>      path_thr                      =  0.1000 eV / A
>
>      list of climbing images :   5,
>
>      activation energy (->) =   0.000000 eV
>      activation energy (<-) =   0.000000 eV
>
>      image        energy (eV)        error (eV/A)        frozen
>
>          1          0.0000000            0.000000            F
>          2          0.0000000            0.000000            F
>          3          0.0000000            0.000000            F
>          4          0.0000000            0.000000            F
>          5          0.0000000            0.000000            F
>          6          0.0000000            0.000000            F
>          7          0.0000000            0.000000            F
>          8          0.0000000            0.000000            F
>
>   path length          =  8.331 bohr
>      inter-image distance =  1.190 bohr
>
>      PWSCF        :     0.08s CPU time
>
>
>
>
>
>
>
>      General routines
>
> Please help.
>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>


PRASENJIT GHOSH,
Ph. D STUDENT,
THEORETICAL SCIENCES UNIT,
JAWAHARLAL NEHRU CENTRE,
JAKKUR P. O.,
BANGALORE - 560064,
INDIA.

PHONE:+91-80-22082835 (OFFICE)
            +91 9880519401 (MOBILE)


From nguyenhalvt at yahoo.com  Wed Aug 30 08:36:45 2006
From: nguyenhalvt at yahoo.com (Nguyen Ngoc Ha)
Date: Tue, 29 Aug 2006 23:36:45 -0700 (PDT)
Subject: [Pw_forum] PseudoPotential for Ce?
Message-ID: <20060830063645.77334.qmail@web31003.mail.mud.yahoo.com>

   Dear PWscf Users!
   I have a plan to study Ce and its compound, but I
don't have any PP of Ce, I don't know about PP
generation or convert PP.
   Could you send me a copy of PP for Ce?
   Thank you very much for your helps! 

Nguyen, Ngoc Ha
Hanoi University of Education
Faculty of Chemistry
Department of Physical Chemistry
Tel: Office: 04/8330842
Home: 04/7681083
Mobile: 0912129517

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From tone.kokalj at ijs.si  Wed Aug 30 09:24:04 2006
From: tone.kokalj at ijs.si (Tone Kokalj)
Date: Wed, 30 Aug 2006 09:24:04 +0200
Subject: [Pw_forum] What to do when NEB calculation is difficult to
	converge?
In-Reply-To: <44F14C2E.4080503@sissa.it>
References: <44F14C2E.4080503@sissa.it>
Message-ID: <1156922644.5953.21.camel@localhost.localdomain>

On Sun, 2006-08-27 at 09:39 +0200, Xunlei Ding wrote:
> Dear all,
> I am now doing a neb calculation on a surface adsorption system with 
> pwscf3.0

> It has more than 400 iterations and has not converged. 

This is a very large number of iterations, and the error on the 3rd,
4th, and 5th image is still large. It is difficult to guess why it does
not converge from your data. The time step ds=1.0 seems reasonable (as
far as my experience go). 

One thing that you may try is to increase the number of images!

How to do that? You need to do a new calculation with larger number of
images, but you can use the currently calculated path as the initial
path for this new calculation.

You may either try with the path_int.sh script that is located in
pwtools/ directory (edit it to suit your needs). This will generate a
new path file will requested number of images, and in the pw.x input
file you specify then new number of images and restart_mode='restart'.
The other possibility is to insert the current path (for example, the
axsf file the pw.x has produced) into the input file of pw.x (specify
new number of images and restart_mode='from_scratch'). You need to do
some hand-editing then:

ATOMIC_POSITIONS angstroms
first_image
   (atomic position of image-1)
intermediate_image
   (atomic position of image-1)
intermediate_image
....
last_image
   (atomic positions of last image)

Regards, Tone




From ding at sissa.it  Wed Aug 30 09:49:03 2006
From: ding at sissa.it (Xunlei Ding)
Date: Wed, 30 Aug 2006 09:49:03 +0200
Subject: [Pw_forum] What to do when NEB calculation is difficult to	converge?
In-Reply-To: <1156922644.5953.21.camel@localhost.localdomain>
References: <44F14C2E.4080503@sissa.it> <1156922644.5953.21.camel@localhost.localdomain>
Message-ID: <44F542EF.9070406@sissa.it>

Thank you very much for your detailed instruction.
I'll try it.
Best regards,
Lei

Tone Kokalj wrote:

>On Sun, 2006-08-27 at 09:39 +0200, Xunlei Ding wrote:
>  
>
>>Dear all,
>>I am now doing a neb calculation on a surface adsorption system with 
>>pwscf3.0
>>    
>>
>
>  
>
>>It has more than 400 iterations and has not converged. 
>>    
>>
>
>This is a very large number of iterations, and the error on the 3rd,
>4th, and 5th image is still large. It is difficult to guess why it does
>not converge from your data. The time step ds=1.0 seems reasonable (as
>far as my experience go). 
>
>One thing that you may try is to increase the number of images!
>
>How to do that? You need to do a new calculation with larger number of
>images, but you can use the currently calculated path as the initial
>path for this new calculation.
>
>You may either try with the path_int.sh script that is located in
>pwtools/ directory (edit it to suit your needs). This will generate a
>new path file will requested number of images, and in the pw.x input
>file you specify then new number of images and restart_mode='restart'.
>The other possibility is to insert the current path (for example, the
>axsf file the pw.x has produced) into the input file of pw.x (specify
>new number of images and restart_mode='from_scratch'). You need to do
>some hand-editing then:
>
>ATOMIC_POSITIONS angstroms
>first_image
>   (atomic position of image-1)
>intermediate_image
>   (atomic position of image-1)
>intermediate_image
>....
>last_image
>   (atomic positions of last image)
>
>Regards, Tone
>
>
>_______________________________________________
>Pw_forum mailing list
>Pw_forum at pwscf.org
>http://www.democritos.it/mailman/listinfo/pw_forum
>
>  
>


From giannozz at nest.sns.it  Wed Aug 30 10:15:36 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Wed, 30 Aug 2006 10:15:36 +0200
Subject: [Pw_forum] superposition
In-Reply-To: <20060823083343.14051.qmail@web31012.mail.mud.yahoo.com>
References: <20060823083343.14051.qmail@web31012.mail.mud.yahoo.com>
Message-ID: <200608301015.36727.giannozz@nest.sns.it>

On Wednesday 23 August 2006 10:33, sareh zarei wrote:

> I am trying to plot difference electron density (difference between
> electron density of crystal and superposition of atomic density) but
> I do not know how can I calculate superposition of atomic density .

the pw.x code calculates the superposition of atomic charge densities
(from data in the pseudopotential files) in order to estimate the starting
potential. I think you can perform a non-scf calculation starting from 
atomic charge densities. I am not sure whetrher in this case the charge
density is saved to a file or not. If it isn't, you may need to modify the
code

Paolo
-- 
Paolo Giannozzi             Phone:   +39/050-509876
DEMOCRITOS and SNS          Fax:     +39/050-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy


From fabris at democritos.it  Wed Aug 30 11:07:42 2006
From: fabris at democritos.it (stefano fabris)
Date: Wed, 30 Aug 2006 11:07:42 +0200
Subject: [Pw_forum] how to saturate dangling bonds in surface
In-Reply-To: <Pine.LNX.4.58.0608280946090.8659@macul.ciencias.uchile.cl>
References: <Pine.LNX.4.58.0608280946090.8659@macul.ciencias.uchile.cl>
Message-ID: <1BA87CBF-1D4C-4839-88D3-91A3E04393C0@democritos.it>

> Hello,
> I have a problem of surface defect in MgO. This is ionic with rocksalt
> structure. In a supercell calculation one has a real surface and an
> artificial surface that should mimic the bulk material. ?Are there
> dangling bonds in the bulk-like surface? Must they be saturated with a
> pseudo hidrogen (with charge one, at wich distance) ?
> Thanks
>

Dear Eduardo,

I am not sure to completely understand your problem, but here are  
some thoughts, possibly relevant. The method used to model ionic  
surfaces (and their defects) with periodic approaches depends on the  
specific surface you are interested in, specifically if it is polar  
or non polar.

The case of non polar surfaces, as the MgO (001) or the (110), is  
simpler and saturation is not really necessary. The obvious method  
used in this case is to build a slab model of the surface  
sufficiently thick so that the atoms in the middle of the slab are  
effectively bulk-like. You can check this condition in various ways,  
for example by comparing the PDOS of the central atoms in the slab  
with the PDOS calculated in a bulk supercell. In this case, you can  
often get away with a reasonable small number of layers, particularly  
with simple oxides like MgO.

The case of polar surfaces, as the MgO (111) is more delicate, since  
you have to pay attention to artificial dipole moments that, if not  
properly handled, increase monotonically with the slab thickness and  
lead to divergency of the surface energy. Also, they influence  
surface relaxations and structural relaxations around defects.  
Saturation of the dangling bonds is one possibility to cope with this  
systems, but other methods have been proposed, as for example the use  
of slabs in which the dipole moment is canceled by symmetry or by an  
external field. Here is some relevant bibliography: E. Kaxiras et  
al., PRB 35, 9625 (1987); B. Meyer and D. Vanderbilt, PRB 63, 205426  
(2001); L. Bengtsson, PRB 59, 12301 (1999).

Hope it helps. All the best,
			Stefano


> Eduardo A. Menendez Proupin
> Department of Physics
> Faculty of Science
> University of Chile
> Las Palmeras 3425
> ?u?oa, Santiago
> Chile
> Phone: 56+2+978 74 11
> http://fisica.ciencias.uchile.cl/~emenendez/
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum



From wierzbom at ts.infn.it  Wed Aug 30 14:22:57 2006
From: wierzbom at ts.infn.it (Malgorzata Wierzbowska)
Date: Wed, 30 Aug 2006 14:22:57 +0200 (CEST)
Subject: [Pw_forum] how to saturate dangling bonds in surface
In-Reply-To: <1BA87CBF-1D4C-4839-88D3-91A3E04393C0@democritos.it>
References: <Pine.LNX.4.58.0608280946090.8659@macul.ciencias.uchile.cl>
 <1BA87CBF-1D4C-4839-88D3-91A3E04393C0@democritos.it>
Message-ID: <Pine.LNX.4.61.0608301349120.25626@dawn.ts.infn.it>

On Wed, 30 Aug 2006, stefano fabris wrote:


  Since pwscf is a periodic program, Eduardo has obviously a problem
with z-direction. Broken Mg-O bonds are not saturated.
For the hydrogen saturation method one may put O-H bond about 0.95Ang.
Franc Rittner used that method for rutile to saturate clusters
in the embeded cluster method, but for Eduardo it applies to the surface.
The method should work more/less for the adsorption of neutral
molecules on the slab.

For the adsorption of charged species, however, the H saturation method
is far inaccurate.
In the old Evjen paper there is a proof that the calcelation
of moments up to the quadrupole is necessary for the convergence of
the potential above the surface. Dipol moment is not enough!
HM Evjen Phys Rev vol. 39, p. 675 (1932).
I would recommend to read this paper first, because many good scientists
recently forgot it and publish wrong results.

For the methods of cancellation of electrostatic moments one can have a
look to paper by Claudine Noguera and to her book, and to
L. Stolarczyk and L. Piela Int. J. of Quant. Chem. vol. 22 p. 911 (1982).

   Regards,
   Gosia

PS. Eduardo, if you like I may send you my thesis about these things.


>
>> Hello,
>> I have a problem of surface defect in MgO. This is ionic with rocksalt
>> structure. In a supercell calculation one has a real surface and an
>> artificial surface that should mimic the bulk material. ?Are there
>> dangling bonds in the bulk-like surface? Must they be saturated with a
>> pseudo hidrogen (with charge one, at wich distance) ?
>> Thanks
>> 
>
> Dear Eduardo,
>
> I am not sure to completely understand your problem, but here are some 
> thoughts, possibly relevant. The method used to model ionic surfaces (and 
> their defects) with periodic approaches depends on the specific surface you 
> are interested in, specifically if it is polar or non polar.
>
> The case of non polar surfaces, as the MgO (001) or the (110), is simpler and 
> saturation is not really necessary. The obvious method used in this case is 
> to build a slab model of the surface sufficiently thick so that the atoms in 
> the middle of the slab are effectively bulk-like. You can check this 
> condition in various ways, for example by comparing the PDOS of the central 
> atoms in the slab with the PDOS calculated in a bulk supercell. In this case, 
> you can often get away with a reasonable small number of layers, particularly 
> with simple oxides like MgO.
>
> The case of polar surfaces, as the MgO (111) is more delicate, since you have 
> to pay attention to artificial dipole moments that, if not properly handled, 
> increase monotonically with the slab thickness and lead to divergency of the 
> surface energy. Also, they influence surface relaxations and structural 
> relaxations around defects. Saturation of the dangling bonds is one 
> possibility to cope with this systems, but other methods have been proposed, 
> as for example the use of slabs in which the dipole moment is canceled by 
> symmetry or by an external field. Here is some relevant bibliography: E. 
> Kaxiras et al., PRB 35, 9625 (1987); B. Meyer and D. Vanderbilt, PRB 63, 
> 205426 (2001); L. Bengtsson, PRB 59, 12301 (1999).
>
> Hope it helps. All the best,
> 			Stefano
>
>
>> Eduardo A. Menendez Proupin
>> Department of Physics
>> Faculty of Science
>> University of Chile
>> Las Palmeras 3425
>> ?u?oa, Santiago
>> Chile
>> Phone: 56+2+978 74 11
>> http://fisica.ciencias.uchile.cl/~emenendez/
>> _______________________________________________
>> Pw_forum mailing list
>> Pw_forum at pwscf.org
>> http://www.democritos.it/mailman/listinfo/pw_forum
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>

From gaoguoying1981 at yahoo.com.cn  Wed Aug 30 16:09:17 2006
From: gaoguoying1981 at yahoo.com.cn (=?gb2312?q?=D3=A2=B9=FA=20=B8=DF?=)
Date: Wed, 30 Aug 2006 22:09:17 +0800 (CST)
Subject: [Pw_forum] lambda convergence
Message-ID: <20060830140917.64925.qmail@web15213.mail.cnb.yahoo.com>

Dear all PWSCF users,
     Before I calcualte the electron-phonon coupling constant, I should judge the convergence of lambda with k-mesh and Gaussian broadening. But my results
  show that lambda at a choosed q point do not close to be a constant with increasing Gaussian broadening at different k-meshs, it continues to decrease. I think the largest k-mesh has been already very large. So what is the reason for this and how should I choose the gaussian broadening  in this case.
  Thanks for any suggestion.
                                                                                              user
                                                                                            guoying  

 		
---------------------------------
 ??????-3.5G???20M??
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060830/092ea28b/attachment.htm 

From baroni at sissa.it  Wed Aug 30 19:57:20 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Wed, 30 Aug 2006 19:57:20 +0200
Subject: [Pw_forum] lambda convergence
In-Reply-To: <20060830140917.64925.qmail@web15213.mail.cnb.yahoo.com>
References: <20060830140917.64925.qmail@web15213.mail.cnb.yahoo.com>
Message-ID: <DF8A37ED-1D2D-46E2-9E2B-654A4B304CA7@sissa.it>


On Aug 30, 2006, at 4:09 PM, ?? ? wrote:

> Dear all PWSCF users,
>    Before I calcualte the electron-phonon coupling constant, I  
> should judge the convergence of lambda with k-mesh and Gaussian  
> broadening. But my results
> show that lambda at a choosed q point do not close to be a constant  
> with increasing Gaussian broadening at different k-meshs,

you mean _decreasing_ don't you?

SB

> it continues to decrease. I think the largest k-mesh has been  
> already very large. So what is the reason for this and how should I  
> choose the gaussian broadening  in this case.
> Thanks for any suggestion.
>                                                                        
>                       user
>                                                                        
>                     guoying
>
> ??????-3.5G???20M??

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060830/0ac4dba0/attachment.htm 

From niuyingli at yahoo.com.cn  Thu Aug 31 03:28:17 2006
From: niuyingli at yahoo.com.cn (Ó¢Àû Å£)
Date: Wed, 30 Aug 2006 18:28:17 -0700 (PDT)
Subject: [Pw_forum] PWscf can not find out symmetry operations!!!
In-Reply-To: <20060829162501.28354.91341.Mailman@democritos.sissa.it>
Message-ID: <20060831012817.71656.qmail@web15601.mail.cnb.yahoo.com>

Dear Pro. Baroni
Thanks a lot for your help!
  I have said that It was no use to set accep=1.0D-2
and recompiled pw.x. The program still could not find
any symmetry operation.
  Then I selected 7 inequivalent atoms, then used 8
symmetry operations:
  x,y,z
  -x,-y,z
  -x+1/2,y+1/2,-z
  x+1/2,-y+1/2,-z
  -x+1/2,-y+1/2,-z
  x+1/2,y+1/2,-z
  x,-y,z
  -x,y,z
to transform the 7 inequivalent atoms to 20 atoms in
the primary unit cell.
  Then I ran scf. I use Norm-Conserving PP and chose
ecutwfc=70,ecutrho=600.
  There appeared message "warning:symmetry operation #
 3/4/5/6 not allowed  fractional translation:" in the
out file as discussed before. Only 4 symmetry
operations without inversion could be found!
I set ecutrho = 500 and calculated it again. Then 
all 8 symmetry operations was found!
  The stress tensor did not change too much!
  stress tensor(kbar) from original atom positions:
    49.20      0.01      0.07
     0.01    -16.72      0.16
     0.07      0.16      2.61
  stress tensor(kbar) from changed atom positions:
    49.98      0.00      0.00
     0.00    -16.13      0.00
     0.00      0.00      7.30
  All the nondiagonal elements changed to 0.00.
  I think changed accpe from 1.0D-5 to 1.0D-2 will be
harmful for the calculations. So I reset it to 1.0D-5
and recompiled. 
  I have no other method to make the program find out
all 8 symmetry operations than this method.

  Another question:
  I use Yahoo mail box. How can I post the reply
message next to the message which is replied in the
mailing list. I alway started a new discussion line
with reply message! Sorry for that!


On Aug 29, 2006, at 18:24:05 +0200, Stefano Baroni
wrote:
>Dear Niu Ying Li,
>
>this very same question has been discussed many
>times in this forum, 
>=20
>the last of which only a few days ago. Would you be
>so kind as to =20
>browse the archive of the mailing list and revert us
>after finding =20
>and reading the some of the relevant posts, in case
>the situation is 
>=20
>not yet clear to you?
>
>Many thanks for your interest in Quantum ESPRESSO
>
>Stefano Baroni
>
>On Aug 29, 2006, at 6:16 PM, =D3=A2=C0=FB =C5=A3
>wrote:
>
>> Dear all,
>>   I am calculate properties of diamond
>> surface(001),the lattice parameters of the slab
>>is :
>> 2.5221A,5.0442A,16.0253A,90,90,90
>> after atoms relax ,the atom positions(crystal) are:
>> C     0.24978  -0.11238   0.21129
>>
>> C     0.24980   0.60988   0.21129
>>
>> C     0.74978  -0.01422   0.17583
>>
>> C     0.74980   0.51175   0.17582
>>
>> C     0.74981   0.24879   0.13029
>>
>> C     0.74981   0.74881   0.11624
>>
>> C     0.24982   0.24883   0.07780
>>
>> C     0.24982   0.74885   0.07061
>>
>> C     0.24984  -0.00014   0.02461
>>
>> C     0.24984   0.49787   0.02461
>>
>> C     0.74985  -0.00211  -0.02461
>>
>> C     0.74986   0.49988  -0.02461
>>
>> C     0.74987   0.24889  -0.07061
>>
>> C     0.74987   0.74888  -0.07780
>>
>> C     0.24987   0.24889  -0.11624
>>
>> C     0.24988   0.74888  -0.13029
>>
>> C     0.24990   0.01186  -0.17582
>>
>> C     0.24987   0.48589  -0.17583
>>
>> C     0.74990   0.11000  -0.21129
>>
>> C     0.74987   0.38773  -0.21129
>>
>>
>> but PWscf can not find any symmetry operations of
>>the
>> system.
>> I setup the struct and use Material Studio find 8
>> space group operations include inversion! The
>> tolerance is 0.01A.
>>
>> Then I read PW manual:
>>
>> the number of significant figures in the atomic
>> positions is not large enough. In file
>>PW/eqvect.f90,
>> the variable accep is used to decide whether a
>> rotation is a symmetry operation. Its current value
>> (10-5 ) is quite strict: a rotated atom must
>>coincide
>> with another atom to 5 significant digits. You may
>> change the value of accep and recompile.
>>
>> I change accep from 1.0d-5 to 1.0d-2 then to 1.0d-1
>> ,but PWscf still find no symmetry.
>>
>> What should I do?
>> Why pwscf choose 1.0d-5,so strict?


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From baroni at sissa.it  Thu Aug 31 04:01:40 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Thu, 31 Aug 2006 04:01:40 +0200
Subject: [Pw_forum] PWscf can not find out symmetry operations!!!
In-Reply-To: <20060831012817.71656.qmail@web15601.mail.cnb.yahoo.com>
References: <20060831012817.71656.qmail@web15601.mail.cnb.yahoo.com>
Message-ID: <9658FD56-9B8E-45D8-AF8C-3912BA80B16C@sissa.it>


On Aug 31, 2006, at 3:28 AM, ???? ?? wrote:

> Dear Pro. Baroni
> Thanks a lot for your help!
>   I have said that It was no use to set accep=1.0D-2
> and recompiled pw.x. The program still could not find
> any symmetry operation.
>   Then I selected 7 inequivalent atoms, then used 8
> symmetry operations:
>   x,y,z
>   -x,-y,z
>   -x+1/2,y+1/2,-z
>   x+1/2,-y+1/2,-z
>   -x+1/2,-y+1/2,-z
>   x+1/2,y+1/2,-z
>   x,-y,z
>   -x,y,z
> to transform the 7 inequivalent atoms to 20 atoms in
> the primary unit cell.

I can hardly understand what you did. If by applying some symmetry  
operations to a bunch of positions ("atoms") you obtain a larger  
number of atoms, doesn't this mean that those operations are not  
"symmetrry operations"???

>   Then I ran scf. I use Norm-Conserving PP and chose
> ecutwfc=70,ecutrho=600.

there is ABSOLUTELY no point in using ecutrho>ecutwfc with norm  
conserving PP's

>   There appeared message "warning:symmetry operation #
>  3/4/5/6 not allowed  fractional translation:" in the
> out file as discussed before. Only 4 symmetry
> operations without inversion could be found!
> I set ecutrho = 500 and calculated it again. Then
> all 8 symmetry operations was found!
>   The stress tensor did not change too much!
>   stress tensor(kbar) from original atom positions:
>     49.20      0.01      0.07
>      0.01    -16.72      0.16
>      0.07      0.16      2.61
>   stress tensor(kbar) from changed atom positions:
>     49.98      0.00      0.00
>      0.00    -16.13      0.00
>      0.00      0.00      7.30
>   All the nondiagonal elements changed to 0.00.
>   I think changed accpe from 1.0D-5 to 1.0D-2 will be
> harmful for the calculations. So I reset it to 1.0D-5
> and recompiled.
>   I have no other method to make the program find out
> all 8 symmetry operations than this method.

Have you browsed in the mailing list's archive, as I suggested?  
Please, let me NOT rediscuss this point that I and others discussed  
the last time a few days ago

>
>   Another question:
>   I use Yahoo mail box. How can I post the reply
> message next to the message which is replied in the
> mailing list. I alway started a new discussion line
> with reply message! Sorry for that!

Please, ask Terry Semel. He certainly knows better than me.

SB

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060831/81650522/attachment.htm 

From etabel at hotmail.com  Thu Aug 31 06:07:51 2006
From: etabel at hotmail.com (Eric Abel)
Date: Thu, 31 Aug 2006 00:07:51 -0400
Subject: [Pw_forum] phonon eigenvectors
Message-ID: <BAY105-F22012372B52FC6219ED85D43F0@phx.gbl>

Hello PWSCF users,

I have a question which is sort of a follow-up to a similar question which I 
asked previously.  Anyway, I was wondering about the imaginary part of the 
phonon eigenvectors, which at the time, Stefano had informed me are more or 
less arbitrary.   At the time that made sense.  But now I am trying to make 
sense of the following eigenvectors:

q =       0.0000      0.0563      0.0000
**************************************************************************
     omega( 1) =      -0.308036 [THz] =     -10.275037 [cm-1]
(  0.000000   0.000000     0.000000  -0.142674    -0.338669   0.000000   )
(  0.000000   0.000000     0.031918  -0.201463    -0.388896  -0.061613   )
(  0.000000   0.000000     0.019612  -0.123787    -0.367551  -0.058231   )
(  0.000000   0.000000     0.000000  -0.226888    -0.354765   0.000000   )
(  0.000000   0.000000     0.036445  -0.230039    -0.328252  -0.052005   )
(  0.000000   0.000000     0.000000  -0.096682    -0.403396   0.000000   )
     omega( 2) =       0.617574 [THz] =      20.600196 [cm-1]
( -0.403925  -0.012145     0.000000   0.000000     0.000000   0.000000   )
( -0.397050  -0.075201     0.000000   0.000000     0.000000   0.000000   )
( -0.397199  -0.075229     0.000000   0.000000     0.000000   0.000000   )
( -0.404073  -0.012149     0.000000   0.000000     0.000000   0.000000   )
( -0.408949  -0.077455     0.000000   0.000000     0.000000   0.000000   )
( -0.416112  -0.012511     0.000000   0.000000     0.000000   0.000000   )
     omega( 3) =       0.788199 [THz] =      26.291668 [cm-1]
(  0.000000   0.000000     0.000000   0.389819    -0.233273   0.000000   )
(  0.000000   0.000000    -0.055122   0.347926    -0.110454  -0.017499   )
(  0.000000   0.000000    -0.061881   0.390589    -0.148121  -0.023467   )
(  0.000000   0.000000     0.000000   0.342277    -0.187111   0.000000   )
(  0.000000   0.000000    -0.050913   0.321357    -0.249835  -0.039581   )
(  0.000000   0.000000     0.000000   0.384275    -0.092029   0.000000   )

These are the first 3 (accoustic) modes of the spectrum.  If the imaginary 
part doesn't matter, then it would appear that the accoustic displacement 
along the b-direction is practically zero.  However, the imaginary part is 
the same magnitude as the displacements in the other orthogonal directions.  
Is there any physical meaning to this, or is it just an artifact of the 
calculation?

Eric,
Ph.D. Student, M.I.T.




From baroni at sissa.it  Thu Aug 31 10:05:04 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Thu, 31 Aug 2006 10:05:04 +0200
Subject: [Pw_forum] phonon eigenvectors
In-Reply-To: <BAY105-F22012372B52FC6219ED85D43F0@phx.gbl>
References: <BAY105-F22012372B52FC6219ED85D43F0@phx.gbl>
Message-ID: <2EF90765-3545-4B6C-B93D-2F3E5DFB40B9@sissa.it>

Eric: do not be satisfied by meaningful answer to your meaningful and  
legitimate questions. You have to *understand* them (i.e. do not stop  
thinking and striving to understand, until answers given by others   
sound to you as YOUR answers. Keep asking yourself questions until  
you are sure that you can answer all of them).

Signed: Grandad

Coming now to your present doubt. You have to understand here that  
what is arbitrary here is the overall _phase_ of the eigenvectors.  
This obviously implies (in some sense) that the imaginary part is  
arbitrary, but *NOT* that it can (always) be neglected, as this  
specic case  shows. What has to be nonzero is NOT the real part of  
the eigenvector, but its modulus.

Is it clearer now? Could you answer similar questions by yourself in  
the future?

Signed: Stefano B.


On Aug 31, 2006, at 6:07 AM, Eric Abel wrote:

> Hello PWSCF users,
>
> I have a question which is sort of a follow-up to a similar  
> question which I asked previously.  Anyway, I was wondering about  
> the imaginary part of the phonon eigenvectors, which at the time,  
> Stefano had informed me are more or less arbitrary.   At the time  
> that made sense.  But now I am trying to make sense of the  
> following eigenvectors:
>
> q =       0.0000      0.0563      0.0000
> ********************************************************************** 
> ****
>     omega( 1) =      -0.308036 [THz] =     -10.275037 [cm-1]
> (  0.000000   0.000000     0.000000  -0.142674    -0.338669    
> 0.000000   )
> (  0.000000   0.000000     0.031918  -0.201463    -0.388896   
> -0.061613   )
> (  0.000000   0.000000     0.019612  -0.123787    -0.367551   
> -0.058231   )
> (  0.000000   0.000000     0.000000  -0.226888    -0.354765    
> 0.000000   )
> (  0.000000   0.000000     0.036445  -0.230039    -0.328252   
> -0.052005   )
> (  0.000000   0.000000     0.000000  -0.096682    -0.403396    
> 0.000000   )
>     omega( 2) =       0.617574 [THz] =      20.600196 [cm-1]
> ( -0.403925  -0.012145     0.000000   0.000000     0.000000    
> 0.000000   )
> ( -0.397050  -0.075201     0.000000   0.000000     0.000000    
> 0.000000   )
> ( -0.397199  -0.075229     0.000000   0.000000     0.000000    
> 0.000000   )
> ( -0.404073  -0.012149     0.000000   0.000000     0.000000    
> 0.000000   )
> ( -0.408949  -0.077455     0.000000   0.000000     0.000000    
> 0.000000   )
> ( -0.416112  -0.012511     0.000000   0.000000     0.000000    
> 0.000000   )
>     omega( 3) =       0.788199 [THz] =      26.291668 [cm-1]
> (  0.000000   0.000000     0.000000   0.389819    -0.233273    
> 0.000000   )
> (  0.000000   0.000000    -0.055122   0.347926    -0.110454   
> -0.017499   )
> (  0.000000   0.000000    -0.061881   0.390589    -0.148121   
> -0.023467   )
> (  0.000000   0.000000     0.000000   0.342277    -0.187111    
> 0.000000   )
> (  0.000000   0.000000    -0.050913   0.321357    -0.249835   
> -0.039581   )
> (  0.000000   0.000000     0.000000   0.384275    -0.092029    
> 0.000000   )
>
> These are the first 3 (accoustic) modes of the spectrum.  If the  
> imaginary part doesn't matter, then it would appear that the  
> accoustic displacement along the b-direction is practically zero.   
> However, the imaginary part is the same magnitude as the  
> displacements in the other orthogonal directions.  Is there any  
> physical meaning to this, or is it just an artifact of the  
> calculation?
>
> Eric,
> Ph.D. Student, M.I.T.
>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060831/672a3818/attachment.htm 

From etabel at hotmail.com  Thu Aug 31 15:22:21 2006
From: etabel at hotmail.com (Eric Abel)
Date: Thu, 31 Aug 2006 09:22:21 -0400
Subject: [Pw_forum] phonon eigenvectors
In-Reply-To: <2EF90765-3545-4B6C-B93D-2F3E5DFB40B9@sissa.it>
Message-ID: <BAY105-F346E6638E0EDD398988F4FD43F0@phx.gbl>

>From: Stefano Baroni <baroni at sissa.it>
>Reply-To: pw_forum at pwscf.org
>To: pw_forum at pwscf.org
>Subject: Re: [Pw_forum] phonon eigenvectors
>Date: Thu, 31 Aug 2006 10:05:04 +0200
>
>Eric: do not be satisfied by meaningful answer to your meaningful and  
>legitimate questions. You have to *understand* them (i.e. do not stop  
>thinking and striving to understand, until answers given by others   sound 
>to you as YOUR answers. Keep asking yourself questions until  you are sure 
>that you can answer all of them).
>
>Signed: Grandad
>
>Coming now to your present doubt. You have to understand here that  what is 
>arbitrary here is the overall _phase_ of the eigenvectors.  This obviously 
>implies (in some sense) that the imaginary part is  arbitrary, but *NOT* 
>that it can (always) be neglected, as this  specic case  shows. What has to 
>be nonzero is NOT the real part of  the eigenvector, but its modulus.

Stefano, I apologize if it sounded as though I was trying to trivialize your 
(previous) answer.  I read most of the postings to this users forum, and 
understand there are a great many who prefer to simply ask questions rather 
than seek out the answers themselves.  As for me, I am too proud to ask for 
help, so when I do, it's because I feel I have no where else to turn.  That 
said, I probably didn't ask the question in the proper way.  My bad.  My gut 
instinct was that the modulus, rather than the real part, was the important 
value...I just wanted confirmation.   The questions which I asked myself 
before posting to the forum were the following:

Question: What is the meaning of an eigenvector?
Answer:  An eigenvector tells how the atoms are displaced in the vibration.

Question: What is the imaginary part?
Answer: The imaginary part is a phase factor.  If one atom has a significant 
imaginary part with respect to the rest, then then it's displacement in the 
vibration will be phase shifted with respect to the others.  In the case of 
a completely imaginary eigenvector, the displacement will be completely out 
of phase.

Question:  If the imaginary part is a phase factor, then what does it mean 
if all of the components are imaginary?
Answer:  First answer: nothing.  If all of the atoms are "out of phase", 
then they are "in phase" with respect to eachother, therefore, having a 
completely imaginary or completely real eigenvector should be equivalent.

Question:  Great.  If this is the case, then why does this eigenvector come 
up imaginary when all of the other vectors come up real?
Answer:  Hmmm.  That's a good one.  There must be some reason that the code 
chooses an imaginary eigenvector for this mode...time to get help on this 
one.  We are fresh out of answers.

And with this I come first to the pwscf discussion archive.  I didn't find 
any discussion of imaginary eigenvectors with real eigenvalues.  I looked to 
the literature, with the same result.  There comes a time when one asks 
himself questions to which he doesn't have the answers, then it comes time 
to discuss with his peers.  With that I come to you with my question:

Why did the code "choose" to make this displacement imaginary.  Is this 
simply an artifact of the matrix diagonalization, or is there some physical 
implications to this?


Thank you for your time and patience.

Eric
>
>Is it clearer now? Could you answer similar questions by yourself in  the 
>future?
>
>Signed: Stefano B.
>
>
>On Aug 31, 2006, at 6:07 AM, Eric Abel wrote:
>
>>Hello PWSCF users,
>>
>>I have a question which is sort of a follow-up to a similar  question 
>>which I asked previously.  Anyway, I was wondering about  the imaginary 
>>part of the phonon eigenvectors, which at the time,  Stefano had informed 
>>me are more or less arbitrary.   At the time  that made sense.  But now I 
>>am trying to make sense of the  following eigenvectors:
>>
>>q =       0.0000      0.0563      0.0000
>>********************************************************************** 
>>****
>>     omega( 1) =      -0.308036 [THz] =     -10.275037 [cm-1]
>>(  0.000000   0.000000     0.000000  -0.142674    -0.338669    0.000000   
>>)
>>(  0.000000   0.000000     0.031918  -0.201463    -0.388896   -0.061613   
>>)
>>(  0.000000   0.000000     0.019612  -0.123787    -0.367551   -0.058231   
>>)
>>(  0.000000   0.000000     0.000000  -0.226888    -0.354765    0.000000   
>>)
>>(  0.000000   0.000000     0.036445  -0.230039    -0.328252   -0.052005   
>>)
>>(  0.000000   0.000000     0.000000  -0.096682    -0.403396    0.000000   
>>)
>>     omega( 2) =       0.617574 [THz] =      20.600196 [cm-1]
>>( -0.403925  -0.012145     0.000000   0.000000     0.000000    0.000000   
>>)
>>( -0.397050  -0.075201     0.000000   0.000000     0.000000    0.000000   
>>)
>>( -0.397199  -0.075229     0.000000   0.000000     0.000000    0.000000   
>>)
>>( -0.404073  -0.012149     0.000000   0.000000     0.000000    0.000000   
>>)
>>( -0.408949  -0.077455     0.000000   0.000000     0.000000    0.000000   
>>)
>>( -0.416112  -0.012511     0.000000   0.000000     0.000000    0.000000   
>>)
>>     omega( 3) =       0.788199 [THz] =      26.291668 [cm-1]
>>(  0.000000   0.000000     0.000000   0.389819    -0.233273    0.000000   
>>)
>>(  0.000000   0.000000    -0.055122   0.347926    -0.110454   -0.017499   
>>)
>>(  0.000000   0.000000    -0.061881   0.390589    -0.148121   -0.023467   
>>)
>>(  0.000000   0.000000     0.000000   0.342277    -0.187111    0.000000   
>>)
>>(  0.000000   0.000000    -0.050913   0.321357    -0.249835   -0.039581   
>>)
>>(  0.000000   0.000000     0.000000   0.384275    -0.092029    0.000000   
>>)
>>
>>These are the first 3 (accoustic) modes of the spectrum.  If the  
>>imaginary part doesn't matter, then it would appear that the  accoustic 
>>displacement along the b-direction is practically zero.   However, the 
>>imaginary part is the same magnitude as the  displacements in the other 
>>orthogonal directions.  Is there any  physical meaning to this, or is it 
>>just an artifact of the  calculation?
>>
>>Eric,
>>Ph.D. Student, M.I.T.
>>
>>
>>_______________________________________________
>>Pw_forum mailing list
>>Pw_forum at pwscf.org
>>http://www.democritos.it/mailman/listinfo/pw_forum
>
>---
>Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  Trieste
>[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)
>
>Please, if possible, don't  send me MS Word or PowerPoint attachments
>Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html
>
>
>




From reboredofa at ornl.gov  Thu Aug 31 15:57:12 2006
From: reboredofa at ornl.gov (Fernando A Reboredo)
Date: Thu, 31 Aug 2006 09:57:12 -0400
Subject: [Pw_forum] Answers and Questions
References: <BAY105-F22012372B52FC6219ED85D43F0@phx.gbl>
 <2EF90765-3545-4B6C-B93D-2F3E5DFB40B9@sissa.it>
Message-ID: <05c101c6cd05$5bbb7420$b7305ba0@ornl.gov>

Dear All,

I feel compelled to write because I am always amazed by the degree of patience that the developers of these ESPRESO package show towards the questions in this mailing list. I also confess that I feel embarrassed to ask my own questions since I know that I will be stealing research time of colleagues.  I thank the developers for making this package available. 

 

However, there are two types of questions 1) the ones that refer to the inner reason of the error messages (that I am not qualified to answer) 2) the ones that refer to a chapter of "Introduction to Solid State Theory" or "Quantum Mechanics". Since I feel I am not better teacher that Aschoft or Kohen I also remain quiet there. 

 

I interrupt my silence to suggest to askers to think whether a question is type 1 or 2. 

 

Thanks again for the hard work and patience.

 

I am Fernando A. Reboredo ORNL (and I approve this message)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060831/8e155eee/attachment.htm 

From akohlmey at cmm.upenn.edu  Thu Aug 31 16:33:14 2006
From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer)
Date: Thu, 31 Aug 2006 10:33:14 -0400
Subject: [Pw_forum] Answers and Questions
In-Reply-To: <05c101c6cd05$5bbb7420$b7305ba0@ornl.gov>
References: <BAY105-F22012372B52FC6219ED85D43F0@phx.gbl>
	 <2EF90765-3545-4B6C-B93D-2F3E5DFB40B9@sissa.it>
	 <05c101c6cd05$5bbb7420$b7305ba0@ornl.gov>
Message-ID: <7b6913e90608310733j20c8d0f4le74e54959713af2a@mail.gmail.com>

On 8/31/06, Fernando A Reboredo <reboredofa at ornl.gov> wrote:
> Dear All,

dear fernando,

thanks for sharing your thoughts with us.

> I feel compelled to write because I am always amazed by the degree of
> patience that the developers of these ESPRESO package show towards the
> questions in this mailing list. I also confess that I feel embarrassed to
> ask my own questions since I know that I will be stealing research time of
> colleagues.  I thank the developers for making this package available.

you may be 'stealing' some research time, but you also have to realize
that even trivial beginner's questions occasionally point out flaws in either
the implementation or the documentation. furthermore, even if you may
not be the most qualified person to respond, you may still have encountered
a similar problem, so you would actually be giving back some of the 'stolen'
time by responding and reporting your own experiences. this is how communities
do work: if everybody contributes, everybody benefits.

> However, there are two types of questions 1) the ones that refer to the
> inner reason of the error messages (that I am not qualified to answer) 2)

well, in my personal experience, trying to figure out what error messages
in a code mean (by looking at the sources), has helped me a lot to understand
how program packages work and trying to figure out whether this is a
legitimate problem, a bug in the code or just a flaw in the input file,
is a good training to avoid problems for the next input you may need to
write and may make you more qualified.

> the ones that refer to a chapter of "Introduction to Solid State Theory" or
> "Quantum Mechanics". Since I feel I am not better teacher that Aschoft or
> Kohen I also remain quiet there.

see above. even when you may not be the better teacher right now,
responding to the best of your knowledge may actually _make_ you
a better teacher. a lot of it is just a matter of practice. you can always
start small and only provide answers to problems where you feel confident
and refer to the literature for the rest.

on top of that, a good way to contribute (and that applies to everybody)
would be to collect frequently asked questions and their answers from the
mailing list archives and integrate them into the quantum-espresso wiki
pages. i've done something similar for a different project a couple of years
back and it was _extremely_ helpful in getting a better understanding,
while at the same time the works was more that of an editor, i.e., you
didn't have to be an expert, but just take the available answers and edit
them into one more consistent text. this is somewhat time consuming,
but given the large number of people in this forum, it should not be
so much if this is shared amongst them. as i wrote before: everybody
contributes, everybody benefits.

> I interrupt my silence to suggest to askers to think whether a question is
> type 1 or 2.

part of the problem of a beginner in using tools like quantum espresso
is, that you frequently cannot tell, where this problem originates from.
many people answering here realize this fact (everybody has been through
that in some way at some point in time) and are willing to give people
some leeway at the beginning (there is no real gain from being rude
over e-mail regardless) and only get increasingly irritated when people
start taking advantage of that.

ok. i guess this is enough 'preaching' for me for today. please everybody
give this some thought and (hopefully) help us to make the QE project
even better and even more fun than it is already right now.

thanks for reading and ciao,
   axel.
>
>
> Thanks again for the hard work and patience.
>
>
>
> I am Fernando A. Reboredo ORNL (and I approve this message)


-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
  Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.


From reboredofa at ornl.gov  Thu Aug 31 18:35:10 2006
From: reboredofa at ornl.gov (Fernando A Reboredo)
Date: Thu, 31 Aug 2006 12:35:10 -0400
Subject: [Pw_forum] Answers and Questions
References: <BAY105-F22012372B52FC6219ED85D43F0@phx.gbl>
 <2EF90765-3545-4B6C-B93D-2F3E5DFB40B9@sissa.it>
 <05c101c6cd05$5bbb7420$b7305ba0@ornl.gov>
 <7b6913e90608310733j20c8d0f4le74e54959713af2a@mail.gmail.com>
Message-ID: <061801c6cd1b$6d3d41e0$b7305ba0@ornl.gov>

Axel
You convinced me.  I will try my best.

The energy of a set of atoms near the minimum can be frequently approximated 
as second order expansion in the atomic displacements with respect to the 
minimum. Within this approximation, the classical evolution in time of the 
atomic positions leads to a set of coupled second order equations in the 
displacements that can be decoupled by a unitary transformation (canonical) 
acting on the vector of displacements (see Ashcrof/Mermin Solid State 
Physics Chapter 22). After this transformation [that digitalizes the 
dynamical matrix in Eq (22.54)] one obtains a set of decoupled equations 
equivalent to independent harmonic oscillators, which are called normal 
modes, in general, and phonons in a solid state context.

The classical evolution in time of an arbitrary set of small displacement 
{u(r,t)} can be decomposed as a superposition of the evolution of the normal 
modes (as explained in the same text) where the coefficients in the 
expansion are chosen to satisfy the initial conditions of the ions.

The differential equation of the harmonic oscillator
x''(t) = -\omega^2 x(t) being a second order equation has two independent 
solutions of the form A \exp( I \omega t) + B (I - \omega t) where t is the 
time I is \sqrt(-1) and \omega is associated  2 \pi times the oscillator 
frequency. A and B are arbitrary constants set to satisfy the initial 
conditions (position and velocity at time zero).

Prove that a change in A  by and arbitrary complex number of modulus 1 is 
equivalent to change the origin of  t.

Prove that if the eigenvectors of the dynamical matrix  are multiplied by an 
arbitrary complex number of modulus one, a) the result is also a normalized 
eigenvector of the same dynamical matrix b) the coefficients of the 
expansion describing a given atomic motion will change only by  a 
multiplicative factor equal to the complex conjugate of the that number.

Prove that if two normal modes have the same frequency (which is often the 
case of a system with symmetries), then an arbitrary linear combination of 
these modes (with arbitrary complex coefficients) is also a normal mode.

Prove that the same is valid in the quantum mechanical case.

Fernando A. Reboredo


----- Original Message ----- 
From: "Axel Kohlmeyer" <akohlmey at cmm.upenn.edu>
To: <pw_forum at pwscf.org>
Sent: Thursday, August 31, 2006 10:33 AM
Subject: Re: [Pw_forum] Answers and Questions


> On 8/31/06, Fernando A Reboredo <reboredofa at ornl.gov> wrote:
>> Dear All,
>
> dear fernando,
>
> thanks for sharing your thoughts with us.
>
>> I feel compelled to write because I am always amazed by the degree of
>> patience that the developers of these ESPRESO package show towards the
>> questions in this mailing list. I also confess that I feel embarrassed to
>> ask my own questions since I know that I will be stealing research time 
>> of
>> colleagues.  I thank the developers for making this package available.
>
> you may be 'stealing' some research time, but you also have to realize
> that even trivial beginner's questions occasionally point out flaws in 
> either
> the implementation or the documentation. furthermore, even if you may
> not be the most qualified person to respond, you may still have 
> encountered
> a similar problem, so you would actually be giving back some of the 
> 'stolen'
> time by responding and reporting your own experiences. this is how 
> communities
> do work: if everybody contributes, everybody benefits.
>
>> However, there are two types of questions 1) the ones that refer to the
>> inner reason of the error messages (that I am not qualified to answer) 2)
>
> well, in my personal experience, trying to figure out what error messages
> in a code mean (by looking at the sources), has helped me a lot to 
> understand
> how program packages work and trying to figure out whether this is a
> legitimate problem, a bug in the code or just a flaw in the input file,
> is a good training to avoid problems for the next input you may need to
> write and may make you more qualified.
>
>> the ones that refer to a chapter of "Introduction to Solid State Theory" 
>> or
>> "Quantum Mechanics". Since I feel I am not better teacher that Aschoft or
>> Kohen I also remain quiet there.
>
> see above. even when you may not be the better teacher right now,
> responding to the best of your knowledge may actually _make_ you
> a better teacher. a lot of it is just a matter of practice. you can always
> start small and only provide answers to problems where you feel confident
> and refer to the literature for the rest.
>
> on top of that, a good way to contribute (and that applies to everybody)
> would be to collect frequently asked questions and their answers from the
> mailing list archives and integrate them into the quantum-espresso wiki
> pages. i've done something similar for a different project a couple of 
> years
> back and it was _extremely_ helpful in getting a better understanding,
> while at the same time the works was more that of an editor, i.e., you
> didn't have to be an expert, but just take the available answers and edit
> them into one more consistent text. this is somewhat time consuming,
> but given the large number of people in this forum, it should not be
> so much if this is shared amongst them. as i wrote before: everybody
> contributes, everybody benefits.
>
>> I interrupt my silence to suggest to askers to think whether a question 
>> is
>> type 1 or 2.
>
> part of the problem of a beginner in using tools like quantum espresso
> is, that you frequently cannot tell, where this problem originates from.
> many people answering here realize this fact (everybody has been through
> that in some way at some point in time) and are willing to give people
> some leeway at the beginning (there is no real gain from being rude
> over e-mail regardless) and only get increasingly irritated when people
> start taking advantage of that.
>
> ok. i guess this is enough 'preaching' for me for today. please everybody
> give this some thought and (hopefully) help us to make the QE project
> even better and even more fun than it is already right now.
>
> thanks for reading and ciao,
>   axel.
>>
>>
>> Thanks again for the hard work and patience.
>>
>>
>>
>> I am Fernando A. Reboredo ORNL (and I approve this message)
>
>
> -- 
> =======================================================================
> Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
>  Center for Molecular Modeling   --   University of Pennsylvania
> Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
> tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
> =======================================================================
> If you make something idiot-proof, the universe creates a better idiot.
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum 



From niuyingli at gmail.com  Thu Aug 31 19:48:40 2006
From: niuyingli at gmail.com (Yingli Niu)
Date: Fri, 1 Sep 2006 01:48:40 +0800
Subject: [Pw_forum] PWscf can not find out symmetry operations!!!
Message-ID: <6398ac910608311048t6c622f2dt3e82e9b288962e63@mail.gmail.com>

Respected Prof. Baroni
    Sorry for that I didn't browsed in the mailing list's archive carefully!
    I think you didn't understand me due to my english language problems.
1. I did a relax calculation for a slab with 10 layers, 20 C atoms. Than I
got 20 relaxed positions.
2. I did a scf calculation using the 20 positions. And pw.x found "No
symmetry".
3. I inputed the structure into Material Studio, and it find 8 symmetry
operations and 7 inequivalent atoms. The tolerance is 0.01A.
4. I used 8 symmetry operations to act on the 7 positions, then got 56
positions(some of them are superpositions).
5  I selected 20 positions in the primary unit cell(there were only 20
positions in one primary unit cell) and input them into the PWscf's run_scf
script. Then pw.x find 8 symmetry operations.
    This method is equivalent to modify the atom positions manually in order
to make the pw.x to find symmetry operations!
    The manual say that :

>the number of significant figures in the atomic positions is not large
enough. In file
>PW/eqvect.f90, the variable accep is used to decide whether a rotation is a
symmetry operation.
>Its current value (10-5 ) is quite strict: a rotated atom must coincide
with another atom to
>5 significant digits. You may change the value of accep and recompile.

    In the last email I said that I had changed the value of accep and
recompiled. But it made no use to find any symmetry operations. So I have no
method other than to modify the atom positions by symmetry manually.
  Now I think the problem has been solved!
  Many thanks again and forgive me to bring so many troubles!
Best wishes!
Yingli Niu


On Aug 31, 2006, at 04:01:40 +0200, Prof. Baroni wrote:

>> Dear Pro. Baroni

>>* Thanks a lot for your help!
*>>*   I have said that It was no use to set accep=3D1.0D-2
*>>* and recompiled pw.x. The program still could not find
*>>* any symmetry operation.
*>>*   Then I selected 7 inequivalent atoms, then used 8
*>>* symmetry operations:
*>>*   x,y,z
*>>*   -x,-y,z
*>>*   -x+1/2,y+1/2,-z
*>>*   x+1/2,-y+1/2,-z
*>>*   -x+1/2,-y+1/2,-z
*>>*   x+1/2,y+1/2,-z
*>>*   x,-y,z
*>>*   -x,y,z
*>>* to transform the 7 inequivalent atoms to 20 atoms in
*>>* the primary unit cell.
*>
>I can hardly understand what you did. If by applying some symmetry =20
>operations to a bunch of positions ("atoms") you obtain a larger =20
>number of atoms, doesn't this mean that those operations are not =20
>"symmetrry operations"???
>>*   Then I ran scf. I use Norm-Conserving PP and chose
*>>* ecutwfc=3D70,ecutrho=3D600.
*>
>there is ABSOLUTELY no point in using ecutrho>ecutwfc with norm =20
>conserving PP's
>
>>*   There appeared message "warning:symmetry operation #
*>>*  3/4/5/6 not allowed  fractional translation:" in the
*>>* out file as discussed before. Only 4 symmetry
*>>* operations without inversion could be found!
*>>* I set ecutrho =3D 500 and calculated it again. Then
*>>* all 8 symmetry operations was found!
*>>*   The stress tensor did not change too much!
*>>*   stress tensor(kbar) from original atom positions:
*>>*     49.20      0.01      0.07
*>>*      0.01    -16.72      0.16
*>>*      0.07      0.16      2.61
*>>*   stress tensor(kbar) from changed atom positions:
*>>*     49.98      0.00      0.00
*>>*      0.00    -16.13      0.00
*>>*      0.00      0.00      7.30
*>>*   All the nondiagonal elements changed to 0.00.
*>>*   I think changed accpe from 1.0D-5 to 1.0D-2 will be
*>>* harmful for the calculations. So I reset it to 1.0D-5
*>>* and recompiled.
*>>*   I have no other method to make the program find out
*>>* all 8 symmetry operations than this method.
*
>Have you browsed in the mailing list's archive, as I suggested? =20
>Please, let me NOT rediscuss this point that I and others discussed =20
>the last time a few days ago
>
>>*
*>>*   Another question:
*>>*   I use Yahoo mail box. How can I post the reply
*>>* message next to the message which is replied in the
*>>* mailing list. I alway started a new discussion line
*>>* with reply message! Sorry for that!
*>>
>Please, ask Terry Semel. He certainly knows better than me.
>
>SB
>
>---
>Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center - =20
>Trieste
>[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060901/6b1d6746/attachment.htm 

From etabel at hotmail.com  Thu Aug 31 20:06:55 2006
From: etabel at hotmail.com (Eric Abel)
Date: Thu, 31 Aug 2006 14:06:55 -0400
Subject: [Pw_forum] Answers and Questions
In-Reply-To: <7b6913e90608310733j20c8d0f4le74e54959713af2a@mail.gmail.com>
Message-ID: <BAY105-F8F7EC86C836BB523B9ED4D43F0@phx.gbl>

I have to agree with Axel here...but I would like to add my two cents.  In 
the end, this is a discussion forum...meaning anyone who is a subscriber can 
answer the questions...it doesn't necessarily have to be the developers.  By 
definition it is an open discussion.

I would consider myself an advance beginner...I feel comfortable dealing 
with the input files and various error messages, most of which I was able to 
figure out on my own using the forum archive and the QE documentation.  As 
far as the physics behind everything, I won't pretend to be an expert, but I 
have learned a lot from my time dealing with these sorts of calculations.


From niuyingli at gmail.com  Thu Aug 31 20:33:05 2006
From: niuyingli at gmail.com (Yingli Niu)
Date: Fri, 1 Sep 2006 02:33:05 +0800
Subject: [Pw_forum] PWscf can not find out symmetry operations!!!
Message-ID: <6398ac910608311133v3078c138i9eef72e60684ef25@mail.gmail.com>

>>*   Then I ran scf. I use Norm-Conserving PP and chose
*>*> ecutwfc=3D70,ecutrho=3D600.
*
>there is ABSOLUTELY no point in using ecutrho>ecutwfc with norm
>conserving PP's

I found that sometimes increasing ecutrho will make the acoustic
phonons at (0,0,0)
close to 0. So, I didn't use the default value ecutrho = 4* ecutwfc.

Best Wishes!

Yingli Niu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060901/e0b357fd/attachment.htm 

From akohlmey at cmm.upenn.edu  Thu Aug 31 20:41:32 2006
From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer)
Date: Thu, 31 Aug 2006 14:41:32 -0400
Subject: [Pw_forum] Answers and Questions
In-Reply-To: <BAY105-F8F7EC86C836BB523B9ED4D43F0@phx.gbl>
References: <7b6913e90608310733j20c8d0f4le74e54959713af2a@mail.gmail.com>
	 <BAY105-F8F7EC86C836BB523B9ED4D43F0@phx.gbl>
Message-ID: <7b6913e90608311141r1c050f4ama7af37889c1593ef@mail.gmail.com>

On 8/31/06, Eric Abel <etabel at hotmail.com> wrote:

[...]

> From my perspective, I would say that some people are too eager to ask
> questions...the answers they seek are clearly present in the current
> available resources.   It just takes a little elbow grease to find them.
> It's much easier to simply post a question to the forum.  I don't support
> this...researching one's own problem is far more educational than having
> someone answer it for you.  We all learned this in the 8th grade.  That

yep. but you also have to understand, that people can get carried away.
my strategy towards this is to gradually provide less direct information
and move to point out ways to how to obtain the information yourself.
some people need to be helped across the first big barrier before they
realize that investing effort into finding the right answers
themselves pays off.
the whole thing is a matter of finding the right balance and understanding
the background.

> said, I feel that these wiki ideas are great, as they would make it easier
> to find answers.  I don't know if this is planned but the wikis should also
> contain the reference papers which disuss the thoery that goes into pwscf.

<grin>you should have looked into the mailing list archives, this has
been discussed recently</grin>
please head over to:
http://www.quantum-espresso.org/wiki/index.php/Main_Page
and ask to get an account (well actually you can probably
request it in person. ;-) ). i have made the experience, that
people who have just learned a method or how to use a tool
(i.e. advanced beginners) write the best documentation.

cheers,
  axel.

> As far as the developers are concerned, I am truly impressed with their
> patience and competence.  It is intimidating enough for a graduate student
> and pwscf beginner to post to the forum, it would be much worse if those
> answering the questions were condescending.  Instead, they treat every
> question more or less equally, and for that I applaud them.


>
> Thank you for all of your hard work and patience.
>
> Eric Abel
> Ph.D. Student, M.I.T.
>
>



-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
  Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.


From baroni at sissa.it  Thu Aug 31 21:56:37 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Thu, 31 Aug 2006 21:56:37 +0200
Subject: [Pw_forum] PWscf can not find out symmetry operations!!!
In-Reply-To: <6398ac910608311133v3078c138i9eef72e60684ef25@mail.gmail.com>
References: <6398ac910608311133v3078c138i9eef72e60684ef25@mail.gmail.com>
Message-ID: <01804BB1-96F8-47A9-B828-1C4A96287C74@sissa.it>


On Aug 31, 2006, at 8:33 PM, Yingli Niu wrote:

> >>   Then I ran scf. I use Norm-Conserving PP and chose
> >> ecutwfc=3D70,ecutrho=3D600.
>
> >there is ABSOLUTELY no point in using ecutrho>ecutwfc with norm
>
> >conserving PP's
>
> I found that sometimes increasing ecutrho will make the acoustic  
> phonons at (0,0,0)
> close to 0. So, I didn't use the default value ecutrho = 4* ecutwfc.

I see! Good point! However: do other modes improve too?
If so, please report.

If not: why not? Is it worth to spend more computer time to get a  
good estimate of a number (0.0) that you know in advance?

These are important points to understand, report, and discuss.

Stefano

>
> Best Wishes!
>
> Yingli Niu
>

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060831/020eeac1/attachment.htm 

From baroni at sissa.it  Thu Aug 31 22:28:04 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Thu, 31 Aug 2006 22:28:04 +0200
Subject: [Pw_forum] phonon eigenvectors
In-Reply-To: <BAY105-F346E6638E0EDD398988F4FD43F0@phx.gbl>
References: <BAY105-F346E6638E0EDD398988F4FD43F0@phx.gbl>
Message-ID: <56DF6181-2682-471B-AC4B-ECCFB8613513@sissa.it>

Dear Eric: I feel flattered (and a little guilty) for your taking me  
so seriously. Also, I may have had a role in starting the discussion  
taken up by Axel and Fernando which, to tell the truth, I do not  
fully understand. All in all, forgive my being (involuntarily!)  
paternalistic, and let's come to the real stuff!

> Question: What is the meaning of an eigenvector?
> Answer:  An eigenvector tells how the atoms are displaced in the  
> vibration.

OK

> Question: What is the imaginary part?
> Answer: The imaginary part is a phase factor.

Forgive my fussiness: its is *not* a phase factor. A non zero  
imaginary part is *due* to a non zero phase.

> If one atom has a significant imaginary part with respect to the  
> rest, then then it's displacement in the vibration will be phase  
> shifted with respect to the others.

OK - More precisely: a phase dfference between two different atomic  
components of a same eigenvectors imply that the motions of the two  
atoms are out of phase (i.e. the velocities of the two may vanish at  
different times)

> In the case of a completely imaginary eigenvector, the displacement  
> will be completely out of phase.

This statement, instead, is meaningless. Ou of phase WITH RESPECT TO  
WHAT? The only meaningful thing is the phase difference between the  
eigenvector components of two different atoms. An overall phase equal  
for all the atoms is simply equivalent to a shift of the origin of  
time. CAN YOU SEE THIS?

Can you see the analogies with the well know statement that quantum  
mechanical wavefunctions (a problem which is conceptually TOTALLY  
different) are defined to within an overall phase factor?

> Question:  If the imaginary part is a phase factor, then what does  
> it mean if all of the components are imaginary?
> Answer:  First answer: nothing.  If all of the atoms are "out of  
> phase", then they are "in phase" with respect to eachother,  
> therefore, having a completely imaginary or completely real  
> eigenvector should be equivalent.

VERY GOOD. YOU GOT IT!

> Question:  Great.  If this is the case, then why does this  
> eigenvector come up imaginary when all of the other vectors come up  
> real?
> Answer:  Hmmm.  That's a good one.  There must be some reason that  
> the code chooses an imaginary eigenvector for this mode...time to  
> get help on this one.  We are fresh out of answers.

Very good, Erich. You came to the point. The answer is deceivingly  
simple. If the the phase is *physically* irrelevant, how could a  
physically sound mathematical algorithm choose it? Answer (deeper  
than it may sound at first): AT RANDOM! Actual algorithms may not be  
really random, but it wouldn't harm if they were!

> And with this I come first to the pwscf discussion archive.  I  
> didn't find any discussion of imaginary eigenvectors with real  
> eigenvalues.

Hey, hey, hey! Slow down! Real eigenvalues are a consequence of the  
Hermiticity of the dynamical matrix and have nothing to do with the  
eigenvector being or not real. A Hermitean matrix may or may not have  
complex eigenvectors. If the matrix is real, the phases can alwayes  
(*not* "must": you see the difference?) in such a way that the  
eigenvectors are real. If the matrix is not real (still being  
Hermitean), I do not know: eigenvectors are in general complex (i.e.  
no overall phase can be chosen  so as to make all of their components  
real), but they may also be in some particular instance real ...

> I looked to the literature, with the same result.  There comes a  
> time when one asks himself questions to which he doesn't have the  
> answers, then it comes time to discuss with his peers.  With that I  
> come to you with my question:
>
> Why did the code "choose" to make this displacement imaginary.  Is  
> this simply an artifact of the matrix diagonalization,

YES

> or is there some physical implications to this?

NO

> Thank you for your time and patience.

Thank you for challenging them! ;-)
(I am not kidding: it's refreshing and instructive for us to help, no  
less that it is for you to be assisted - hopefully ;-)

Stefano

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060831/c45cc2a8/attachment.htm 

From niuyingli at gmail.com  Thu Aug 31 23:54:06 2006
From: niuyingli at gmail.com (Yingli Niu)
Date: Fri, 1 Sep 2006 05:54:06 +0800
Subject: [Pw_forum] PWscf can not find out symmetry operations!!!
Message-ID: <6398ac910608311454t4dfe1d70me517142c1318d4d5@mail.gmail.com>

I only found this feature at Gamma point, for other points I will test!

> On Aug 31, 2006, at 8:33 PM, Yingli Niu wrote:
>
> > >>   Then I ran scf. I use Norm-Conserving PP and chose
> > >> ecutwfc=3D70,ecutrho=3D600.
> >
> > >there is ABSOLUTELY no point in using ecutrho>ecutwfc with norm
> >
> > >conserving PP's
> >
> > I found that sometimes increasing ecutrho will make the acoustic
> > phonons at (0,0,0)
> > close to 0. So, I didn't use the default value ecutrho = 4* ecutwfc.
>
> I see! Good point! However: do other modes improve too?
> If so, please report.
>
> If not: why not? Is it worth to spend more computer time to get a
> good estimate of a number (0.0) that you know in advance?
>
> These are important points to understand, report, and discuss.
>
> Stefano
>
> >
> > Best Wishes!
> >
> > Yingli Niu
> >
>
> ---
> Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -
> Trieste
> [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)


From emenendez at macul.ciencias.uchile.cl  Thu Aug 31 23:07:11 2006
From: emenendez at macul.ciencias.uchile.cl (Eduardo Ariel Menendez P)
Date: Thu, 31 Aug 2006 17:07:11 -0400 (CLT)
Subject: [Pw_forum] superposition
In-Reply-To: <20060830175802.19943.73669.Mailman@democritos.sissa.it>
References: <20060830175802.19943.73669.Mailman@democritos.sissa.it>
Message-ID: <Pine.LNX.4.58.0608311559430.32344@macul.ciencias.uchile.cl>

Hello
I think that this combination of keywords does the job of generate the
superposition of atomic densities

               restart_mode = 'from_scratch' ,
           electron_maxstep = 1,
                   conv_thr = 1.0D+10 ,  ! Yes, it is D+10
                startingpot = 'atomic' ,
            	mixing_beta = 0.0,
                startingwfc = 'atomic'
at least it runs up to the end and writes a density file.
Regards
Eduardo
>
> On Wednesday 23 August 2006 10:33, sareh zarei wrote:
>
> > I am trying to plot difference electron density (difference between
> > electron density of crystal and superposition of atomic density) but
> > I do not know how can I calculate superposition of atomic density .
>
> the pw.x code calculates the superposition of atomic charge densities
> (from data in the pseudopotential files) in order to estimate the starting
> potential. I think you can perform a non-scf calculation starting from
> atomic charge densities. I am not sure whetrher in this case the charge
> density is saved to a file or not. If it isn't, you may need to modify the
> code
>
> Paolo
> --