www.rocksclusters.org
Huiqun Zhou
----- Original Message -----
Sent: Saturday, April 15, 2006 =
12:33=20
AM
Subject: [Pw_forum] Problem not =
related=20
PWscf
Dear All:
I am sorry to ask a question not related to the =
topics of=20
maillist.
I just want to know which distribution is good for =
building=20
HPC-cluster.My computers are AMD248-64bit cpu.
Recently, i tried to build with RHEL AS4.0, but failed due to =
unstable=20
NIS service. So , Would anyone have experience =
building=20
HPC cluster for AMD x86-64 CPU? Would you please give me some=20
suggestion?
Regards
Hai-Ping
------=_NextPart_000_001F_01C66077.D40AF360--
--__--__--
_______________________________________________
Pw_forum mailing list
Pw_forum at pwscf.org
http://www.democritos.it/mailman/listinfo/pw_forum
End of Pw_forum Digest
**********************************************************************
This email and any files transmitted with it are confidential and
intended solely for the use of the individual or entity to whom they
are addressed. If you have received this email in error please notify
the system manager.
www.imre.a-star.edu.sg
**********************************************************************
From Steven.Kirk at hv.se Tue Apr 18 11:16:47 2006
From: Steven.Kirk at hv.se (Steven Kirk)
Date: Tue, 18 Apr 2006 11:16:47 +0200
Subject: [Pw_forum] Symmetry-finding problems in pw.x
Message-ID: <4444AE7F.9010108@hv.se>
Hello,
I am trying to simulate a multiwalled carbon nanotube in pw.x using the
input file:
&CONTROL
title='MWNT calculation',
calculation='scf',
restart_mode='from_scratch',
prefix='MWNT',
/
&SYSTEM
ibrav = 6,
A = 17.0, B = 17.0, C = 4.26, cosAB = 0, cosAC = 0, cosBC = 0,
nat = 24,
ntyp = 1,
ecutwfc = 73.4986,
/
&ELECTRONS
/
ATOMIC_SPECIES
C 12.0107 C.pbe-rrkjus.UPF
ATOMIC_POSITIONS {crystal}
C 0.18067 0.03594 0.47051
C 0.18067 0.03594 0.80385
C 0.15316 0.10234 0.97051
C 0.15316 0.10234 0.30385
C 0.10234 0.15316 0.47051
C 0.10234 0.15316 0.80385
C 0.03594 0.18067 0.97051
C 0.03594 0.18067 0.30385
C -0.35480 -0.09923 0.18505
C -0.32863 -0.16654 0.35171
C -0.35480 -0.09923 0.85171
C -0.28982 -0.22745 0.18505
C -0.23988 -0.27962 0.35171
C -0.32863 -0.16654 0.68505
C -0.28982 -0.22745 0.85171
C -0.18072 -0.32105 0.18505
C -0.11461 -0.35014 0.35171
C -0.23988 -0.27962 0.68505
C -0.18072 -0.32105 0.85171
C -0.04410 -0.36577 0.18505
C 0.02810 -0.36734 0.35171
C -0.11461 -0.35014 0.68505
C -0.04410 -0.36577 0.85171
C 0.02810 -0.36734 0.68505
K_POINTS gamma
I am trying to make maximum use of symmetry, because the wavefunction
file gets too big when I don't use it.
Unfortunately it seems that pw.x says 'No symmetry !' with the input
data above. How can I make sure that pw.x detects the 4-fold symmetry
and simulates the structure correctly with the minimum number of atoms ?
Many thanks in advance,
Steve Kirk
--
Dr. Steven R. Kirk
Dept. of Technology, Mathematics & Computer Science (P)+46 520 223215
University West (F)+46 520 223299
P.O. Box 957 Trollhattan 461 29 SWEDEN http://taconet.webhop.org
From marzari at MIT.EDU Tue Apr 18 13:38:42 2006
From: marzari at MIT.EDU (Nicola Marzari)
Date: Tue, 18 Apr 2006 13:38:42 +0200
Subject: [Pw_forum] Symmetry-finding problems in pw.x
In-Reply-To: <4444AE7F.9010108@hv.se>
References: <4444AE7F.9010108@hv.se>
Message-ID: <4444CFC2.2080009@mit.edu>
Hi Steve,
here are a few comments (in your case, 2), and then 1) are the likely
culprits).
1) you need to have enough significant digits in your
input for the code to recognize symmetry operations - 5 is probably
borderline, and not enough.
2) you shouldn't force to code to look for fractional translations:
arrange the nanotube such that your coordinates show
symmetry: x in -x, or such (that is to say, place the nanotube so that
it is bisected by symmetry planes such as x=0).
You should also start with a SWNT, and experiment a bit to see when
operations are found.
3) for an ultrasoft carbon, you need between 25 and 30 Ry. your cutoff
is much much larger than needed or reasonable. But you need a dual of
8 to 12 for the charge density cutoff. The tests are actually published
in the pseudopotential page, for this pseudo.
Advice: try to reproduce diamond and graphene results, and then
a simple swnt, before moving to mwnt. Extensive results on
carbon, with this pseudo, are e.g. on a Mounet/Marzari PRB of 2005.
A question for everyone else: could there be an input mode in which
point-group symmetry operations are specified by hand ? Once could have
the code then 1) check that those are symmetry operations, or 2)
multiply atoms using the point group symmetries that have been read.
nicola
Steven Kirk wrote:
> Hello,
>
> I am trying to simulate a multiwalled carbon nanotube in pw.x using the
> input file:
>
> &CONTROL
> title='MWNT calculation',
> calculation='scf',
> restart_mode='from_scratch',
> prefix='MWNT',
> /
> &SYSTEM
> ibrav = 6,
> A = 17.0, B = 17.0, C = 4.26, cosAB = 0, cosAC = 0, cosBC = 0,
> nat = 24,
> ntyp = 1,
> ecutwfc = 73.4986,
> /
> &ELECTRONS
> /
> ATOMIC_SPECIES
> C 12.0107 C.pbe-rrkjus.UPF
> ATOMIC_POSITIONS {crystal}
> C 0.18067 0.03594 0.47051
> C 0.18067 0.03594 0.80385
> C 0.15316 0.10234 0.97051
> C 0.15316 0.10234 0.30385
> C 0.10234 0.15316 0.47051
> C 0.10234 0.15316 0.80385
> C 0.03594 0.18067 0.97051
> C 0.03594 0.18067 0.30385
> C -0.35480 -0.09923 0.18505
> C -0.32863 -0.16654 0.35171
> C -0.35480 -0.09923 0.85171
> C -0.28982 -0.22745 0.18505
> C -0.23988 -0.27962 0.35171
> C -0.32863 -0.16654 0.68505
> C -0.28982 -0.22745 0.85171
> C -0.18072 -0.32105 0.18505
> C -0.11461 -0.35014 0.35171
> C -0.23988 -0.27962 0.68505
> C -0.18072 -0.32105 0.85171
> C -0.04410 -0.36577 0.18505
> C 0.02810 -0.36734 0.35171
> C -0.11461 -0.35014 0.68505
> C -0.04410 -0.36577 0.85171
> C 0.02810 -0.36734 0.68505
> K_POINTS gamma
>
> I am trying to make maximum use of symmetry, because the wavefunction
> file gets too big when I don't use it.
>
> Unfortunately it seems that pw.x says 'No symmetry !' with the input
> data above. How can I make sure that pw.x detects the 4-fold symmetry
> and simulates the structure correctly with the minimum number of atoms ?
>
> Many thanks in advance,
> Steve Kirk
--
---------------------------------------------------------------------
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 ejl7240 at chemail.tamu.edu Tue Apr 18 15:09:31 2006
From: ejl7240 at chemail.tamu.edu (Eduardo J. Lamas)
Date: Tue, 18 Apr 2006 08:09:31 -0500
Subject: [Pw_forum] pw.x always crashes at the end of relax calculations
Message-ID: <001a01c662e9$54d27ae0$0ee35ba5@ch0316>
Hi, pw.x always crashes to me during the last check of the relax
calculations
it happened in versions 2.1.3 and 3.0 and with different compilers.
In order to avoid this I commented the following lines in move_ions.f90
(then the program proceed
normally and ends normally but the check without electronic history is not
longer really done):
CALL hinit0()
CALL potinit()
CALL newd()
CALL wfcinit()
When I want to check that the forces are indeed low I take the coordinates
without the verifications
and run a scf calculation and print the forces.
I am wondering if it is something wrong in those subroutines and if somebody
else was having the same problem.
Maybe during the reinitialization a variable is deleted twice or something
or is accessed after it is deleted?.
Thanks,
Eduardo
This run for example crashes (but nearly all runs were having this problem):
&CONTROL
title='H20',
calculation='relax',
restart_mode='from_scratch',
tprnfor=.true.,
pseudo_dir='/home/lamas/espresso/pseudo/',
outdir='./tmp/',
max_seconds=90000,
/
&SYSTEM
ibrav = 1,
celldm(1)= 15,
nat = 3,
ntyp = 2,
nbnd = 5,
ecutwfc =50.0,
ecutrho =500,
occupations='from_input',
nspin=2,
starting_magnetization=0.0
/
&ELECTRONS
mixing_beta = 0.2,
conv_thr = 1.0d-6
/
&IONS
ion_dynamics='bfgs',
/
ATOMIC_SPECIES
O 16.00 O_PBE_v1.vdb
H 1.00 H_PBE_v1.vdb
ATOMIC_POSITIONS {angstrom}
O 2.021843520 1.760434290 0.28003039
H 2.286102819 0.815021260 0.08933101
H 2.811655420 2.359926539 0.15037213
OCCUPATIONS
1.0 1.0 1.0 1.0 0.0
1.0 1.0 1.0 1.0 0.0
From proffess at yandex.ru Tue Apr 18 15:33:50 2006
From: proffess at yandex.ru (Sergey Lisenkov)
Date: Tue, 18 Apr 2006 17:33:50 +0400 (MSD)
Subject: [Pw_forum] metallic or not
Message-ID: <4444EABE.000003.05743@ariel.yandex.ru>
Dear all,
I have a problem. I am stdying the silicon clathrates. One of them has became metallic after full relaxation using pwscf (2.1.5 version). It is well known that silicon clathrates have band gap wider than tetrahedral silicon.
What can be the problem?
Thanks,
Sergey
From dalcorso at sissa.it Tue Apr 18 15:44:13 2006
From: dalcorso at sissa.it (Andrea Dal Corso)
Date: Tue, 18 Apr 2006 15:44:13 +0200
Subject: [Pw_forum] Non-colinear magnetism
In-Reply-To:
References:
Message-ID: <1145367853.4626.24.camel@dhpc-5-26.sissa.it>
On Mon, 2006-04-10 at 16:09 +0100, H.S.Domingos wrote:
> Dear all,
>
> I would like to ask if the non-colinear magnetism part of the code is
> fairly stable. I would like to do magnetism in permalloys and I dont have
> a feel for how hard that might be with this code.
> Is it advised to use the experimental features for this or is it
> half-baked and unstable ?
As far as I know, the non-collinear and the spin-orbit parts of pw.x are
working. This part of the code has been used less extensively than the
other parts so bugs could still be present. Please help us to find
them.
Best wishes,
Andrea
> Thank you very much once again,
>
> Helder
>
>
> =======================================================================
> | Dr. Helder S. Domingos |
> | |
> | INESC Microsyst & Nanotechnol, Lisbon, P-1000 Portugal |
> | and |
> | R&D unit for Molecular Chemical Physics |
> | Chemistry Department, University of Coimbra |
> =======================================================================
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
--
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 emenendez at macul.ciencias.uchile.cl Tue Apr 18 17:48:12 2006
From: emenendez at macul.ciencias.uchile.cl (Eduardo Ariel Menendez P)
Date: Tue, 18 Apr 2006 11:48:12 -0400 (CLT)
Subject: [Pw_forum] Re: SOC
In-Reply-To: <20060414053744.17757.95638.Mailman@democritos.sissa.it>
References: <20060414053744.17757.95638.Mailman@democritos.sissa.it>
Message-ID:
Does it means that one can obtain bands with spin-orbit splitting using a
fully relativistic pseudopotentials?
Thanks
Eduardo
>
> Message: 1
> Subject: Re: [Pw_forum] SOC
> From: Andrea Dal Corso
> To: pw_forum at pwscf.org
> Organization: SISSA
> Date: Thu, 13 Apr 2006 09:11:32 +0200
> Reply-To: pw_forum at pwscf.org
>
> Yes the SO term is only in the pseudo-potential but valence states
> interact with a fully relativistic pseudo-potential at each scf cycle.
> We make an approximation because we neglect the spin-orbit coupling
> outside the core radius where however the effect is expected to be
> small.
>
> Andrea
>
>
> On Wed, 2006-04-12 at 12:43 -0500, Bhagawan Sahu wrote:
> > Dear pwscf users,
> >
> > Is it true that
> >
> > the bulk spin-orbit coupling (SOC) calculation is done using a
> > relativistic
> > PP for the atom/ion (RRKJ scheme and Andrea Dal Carso's PRB paper) and the
> > SOC effect is not included self-consistently on the valence states during
> > the scf cycle?
> >
> > Sahu
> >
> > _______________________________________________
> > Pw_forum mailing list
> > Pw_forum at pwscf.org
> > http://www.democritos.it/mailman/listinfo/pw_forum
> --
> 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 dalcorso at sissa.it Tue Apr 18 17:54:55 2006
From: dalcorso at sissa.it (Andrea Dal Corso)
Date: Tue, 18 Apr 2006 17:54:55 +0200
Subject: [Pw_forum] Re: SOC
In-Reply-To:
References: <20060414053744.17757.95638.Mailman@democritos.sissa.it>
Message-ID: <1145375695.4626.30.camel@dhpc-5-26.sissa.it>
Yes, the bands published in Phys. Rev. B 71, 115106 (2005) can be
obtained with pw.x. See example22.
Andrea
On Tue, 2006-04-18 at 11:48 -0400, Eduardo Ariel Menendez P wrote:
> Does it means that one can obtain bands with spin-orbit splitting using a
> fully relativistic pseudopotentials?
> Thanks
> Eduardo
>
> >
> > Message: 1
> > Subject: Re: [Pw_forum] SOC
> > From: Andrea Dal Corso
> > To: pw_forum at pwscf.org
> > Organization: SISSA
> > Date: Thu, 13 Apr 2006 09:11:32 +0200
> > Reply-To: pw_forum at pwscf.org
> >
> > Yes the SO term is only in the pseudo-potential but valence states
> > interact with a fully relativistic pseudo-potential at each scf cycle.
> > We make an approximation because we neglect the spin-orbit coupling
> > outside the core radius where however the effect is expected to be
> > small.
> >
> > Andrea
> >
> >
> > On Wed, 2006-04-12 at 12:43 -0500, Bhagawan Sahu wrote:
> > > Dear pwscf users,
> > >
> > > Is it true that
> > >
> > > the bulk spin-orbit coupling (SOC) calculation is done using a
> > > relativistic
> > > PP for the atom/ion (RRKJ scheme and Andrea Dal Carso's PRB paper) and the
> > > SOC effect is not included self-consistently on the valence states during
> > > the scf cycle?
> > >
> > > Sahu
> > >
> > > _______________________________________________
> > > Pw_forum mailing list
> > > Pw_forum at pwscf.org
> > > http://www.democritos.it/mailman/listinfo/pw_forum
> > --
> > 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
> >
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
--
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 eyvaz_isaev at yahoo.com Tue Apr 18 22:07:08 2006
From: eyvaz_isaev at yahoo.com (Eyvaz Isaev)
Date: Tue, 18 Apr 2006 13:07:08 -0700 (PDT)
Subject: [Pw_forum] metallic or not
In-Reply-To: <4444EABE.000003.05743@ariel.yandex.ru>
Message-ID: <20060418200708.31189.qmail@web60315.mail.yahoo.com>
Hi Sergei,
One can suggest that it is due to the LDA which
underestmate the band gap. Did you check with a
pseudopotential you used the band gap in Si, and
unrelaxed clathrate? Is your clathrate direct or
indirect semiconductor?
On the other side relaxation will violate your
initial symmetry that may lead to appearence
additional bands so that a band cross-section might
take place.
Please pay attention that you may have a direct band
gap for a given k-point (even a set of k-points) in
BZ, but indirect gap might be close to zero.
In this case according DOS calculations you can not
conclude whether you have metal or semoconductor. In
this case band structure calculations are more useful.
Hope it helps,
Eyvaz.
--- Sergey Lisenkov wrote:
>
> Dear all,
>
> I have a problem. I am stdying the silicon
> clathrates. One of them has became metallic after
> full relaxation using pwscf (2.1.5 version). It is
> well known that silicon clathrates have band gap
> wider than tetrahedral silicon.
>
> What can be the problem?
>
> Thanks,
> Sergey
> _______________________________________________
> 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 proffess at yandex.ru Tue Apr 18 23:15:58 2006
From: proffess at yandex.ru (Sergey Lisenkov)
Date: Wed, 19 Apr 2006 01:15:58 +0400 (MSD)
Subject: [Pw_forum] metallic or not
In-Reply-To: <20060418200708.31189.qmail@web60315.mail.yahoo.com>
References: <20060418200708.31189.qmail@web60315.mail.yahoo.com>
Message-ID: <4445570E.000001.19710@mfront8.yandex.ru>
Hi Eyvaz,
Thanks a lot for the reply. Yes, the underestimation was the first thing I thought. I use PBE functional and test for silicon diamond gave me a band gap of ~0.8 eV, which is the _standart_ DFT band gap value.
Another thing is that when this clathrate was relaxed by TBMD, the final structure is semiconductor with a band gap around 1.6 eV which is appropriate value. When I relaxed this "relaxed by TBMD" structure it becames a metallic. When I compared two DOS (one from TBMD and another one from PW) I see that are almost identical with one difference. The DOS obtained by PW conatains some additional peaks on Fermi Level.
I think this is a DFT fault. I remember, that in case of carbon clathrates people using DFT found that these structures are metallic, which WAS not confirmed by experiments.
Thanks,
Sergey
From baroni at sissa.it Wed Apr 19 08:18:18 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Wed, 19 Apr 2006 08:18:18 +0200
Subject: [Pw_forum] Fwd: how to input crystal structure of Amm2 SrTiO3
References: <20060419013530.93264.qmail@web15804.mail.cnb.yahoo.com>
Message-ID: <89B2A50A-A6E8-447C-BABA-2F9D55205003@sissa.it>
Dear All: I am forwarding here this message that was sent directly to
me. A reply will follow shortly.
Dear Zhuzhenye: please, send this kind of messages to this forum (to
which you can subscribe from the "users' forum section" of the pwscf
home page" www.pwscf.org). Also, do not forget to browse the mailing
list's archive before starting a new thread in the list. The archive
already contains many answers to many of the questions that you may
want to ask. Thaank you for using PWscf and for writing to us.
Stefano B.
Begin forwarded message:
> From: ?? ?
> Date: April 19, 2006 3:35:30 AM GMT+02:00
> To: stefano.baroni at democritos.it
> Subject: how to input crystal structure of Amm2 SrTiO3
>
> Dear stefano Baroni:
> i want to caculate the properties of SrTiO3 under tensile strain. i
> have known the crystal structure of SrTiO3 under tensile strain is
> Amm2 space group, and atomic cartesian cooridinate : Sr(0,0,0),Ti
> ((0.5+?x,0,0.5),O1(0.5+?x,0,0) O2(0.25+?x,0.25+?y,0.5).there is
> several question about atomic question.
> first, in ATOMIC_POSITION{alat}, is atomic cooridinates along
> cartesian cooridinate? second in ATOMIC_POSITION{crystal}, is
> atomic cooridinates along crystal axis of primitive unit cell
> cooridinate?for hobr and angstrom,the case is same to the alat?
> from symmestry, The SrTiO3 cartesian cooridinate:
> Sr 0.000000 0.000000 0.000000
> Ti 0.500000 0.000000 0.500000
> O 0.500000 0.000000 0.000000
> O 0.250000 -0.250000 0.000000
> O 0.750000 -0.250000 0.500000
> O 0.750000 -0.250000 0.000000
> O 0.250000 -0.250000 0.500000
> but i found that the ion position in still after optimization.
> please tell me how to input structure of AMM2 SrTiO3?
> There is my inputfile of
> SrTIO3
> &CONTROL
> title='st'
> calculation='relax'
> restart_mode = 'from_scratch' ,
> outdir = 'tmp' ,
> pseudo_dir = '/home/zzy/pwscf/pseudo/' ,
> prefix = 'pst' ,
> forc_conv_thr=1.0D-4
> tstress = .true. ,
> tprnfor = .true. ,
> /
> &SYSTEM
> ibrav=9,
> celldm(1)=7.5,
> celldm(2)=1
> celldm(3)=0.955
> nat=7,
> ntyp=3,
> ecutwfc=50,
> ecutrho=450
> /
> &ELECTRONS
> conv_thr=1.0d-6
> /
> &IONS
> ion_dynamics = 'bfgs' ,
> /
> ATOMIC_SPECIES
> Sr 87.62000 038-Sr-ca-sp-vgrp.uspp.UPF
> Ti 47.88000 022-Ti-ca-sp-vgrp.uspp.UPF
> O 15.99960 008-O-ca--vgrp.uspp.UPF
> ATOMIC_POSITIONS {crystal}
> Sr 0 0 0 0 0 0
> Ti 0.5 0 0.5 1 0 0
> O 0.5 0 0 1 0 0
> O 0.25 -0.25 0.0 1 1 0
> O 0.75 -0.25 0.5 1 1 0
> O 0.25 -0.25 0.5 1 1 0
> O 0.75 -0.25 0 1 1 0
> K_POINTS {automatic}
> 6 6 6 1 1 1
>
> Regards
>
> zhuzhenye
> __________________________________________________
> ???????????????
> http://cn.mail.yahoo.com
---
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/20060419/5ecbcc7d/attachment.htm
From baroni at sissa.it Wed Apr 19 09:19:27 2006
From: baroni at sissa.it (Stefano Baroni)
Date: Wed, 19 Apr 2006 09:19:27 +0200
Subject: [Pw_forum] Re: how to input crystal structure of Amm2 SrTiO3
In-Reply-To: <20060419013530.93264.qmail@web15804.mail.cnb.yahoo.com>
References: <20060419013530.93264.qmail@web15804.mail.cnb.yahoo.com>
Message-ID: <512F5B15-BC10-4D90-8AF3-CAC20AAFE9E2@sissa.it>
On Apr 19, 2006, at 3:35 AM, ?? ? wrote:
> Dear stefano Baroni:
> i want to caculate the properties of SrTiO3 under tensile strain. i
> have known the crystal structure of SrTiO3 under tensile strain is
> Amm2 space group, and atomic cartesian cooridinate : Sr(0,0,0),Ti
> ((0.5+?x,0,0.5),O1(0.5+?x,0,0) O2(0.25+?x,0.25+?y,0.5).there is
> several question about atomic question.
> first, in ATOMIC_POSITION{alat}, is atomic cooridinates along
> cartesian cooridinate?
YES atomic coordinates are cartesian for every option of the
ATOMIC_POSITION control card, but "crystal"
> second in ATOMIC_POSITION{crystal}, is atomic cooridinates along
> crystal axis of primitive unit cell cooridinate?for hobr and
> angstrom,the case is same to the alat?
I do not quite understand the question, in any case the "crystal"
atomic coordinates are defined by the convention:
x(:) = \sum_j AT(:,j)*c(j),
where x(:) are the atomic (cartesian) coordinates, and AT(:,1), AT(:,
2), AT(:,3) are the the basis vectors of the Bravais lattice with
their own units (you usually specify AT giving "ibrav" and "celldm
(:)", the latter in Bohr radii). A list of conventional basis vectors
used by PWscf (as well as much other useful information) is found in
the documentation files distributed with the code (INPUT_PW, in the
present case). Beware, though, that you hardly need to pass atomic
positions in crystal units. It is usually far easier to pass them in
cartesian coordinates.
> from symmestry, The SrTiO3 cartesian cooridinate:
> Sr 0.000000 0.000000 0.000000
> Ti 0.500000 0.000000 0.500000
> O 0.500000 0.000000 0.000000
> O 0.250000 -0.250000 0.000000
> O 0.750000 -0.250000 0.500000
> O 0.750000 -0.250000 0.000000
> O 0.250000 -0.250000 0.500000
> but i found that the ion position in still after optimization.
> please tell me how to input structure of AMM2 SrTiO3?
This is a more subtle question which I would invite you to think
about carefully by yourself.
Suppose you want to find the (stable) equilibrium position of a
particle in a double potential well, say V(x)=(x^2-1)^2, and that you
start your numerical search from x=0. The force acting on the
particle at this position is zero by symmetry. So, any algorithm
aiming at finding the zeroes of the force would keep the particle
fixed at x=0. Does this imply that this is an equilibrim position?
Does this imply that this a _stable_ equilibrium position? How would
you use an algorithm which finds the zeroes of the force to find a
stable equilibrium configuration in this case? What has this all to
do with you problem?
Please, give a thought at these questions and revert to us with or
without a solution to your problem.
Stefano B.
> There is my inputfile of
> SrTIO3
> &CONTROL
> title='st'
> calculation='relax'
> restart_mode = 'from_scratch' ,
> outdir = 'tmp' ,
> pseudo_dir = '/home/zzy/pwscf/pseudo/' ,
> prefix = 'pst' ,
> forc_conv_thr=1.0D-4
> tstress = .true. ,
> tprnfor = .true. ,
> /
> &SYSTEM
> ibrav=9,
> celldm(1)=7.5,
> celldm(2)=1
> celldm(3)=0.955
> nat=7,
> ntyp=3,
> ecutwfc=50,
> ecutrho=450
> /
> &ELECTRONS
> conv_thr=1.0d-6
> /
> &IONS
> ion_dynamics = 'bfgs' ,
> /
> ATOMIC_SPECIES
> Sr 87.62000 038-Sr-ca-sp-vgrp.uspp.UPF
> Ti 47.88000 022-Ti-ca-sp-vgrp.uspp.UPF
> O 15.99960 008-O-ca--vgrp.uspp.UPF
> ATOMIC_POSITIONS {crystal}
> Sr 0 0 0 0 0 0
> Ti 0.5 0 0.5 1 0 0
> O 0.5 0 0 1 0 0
> O 0.25 -0.25 0.0 1 1 0
> O 0.75 -0.25 0.5 1 1 0
> O 0.25 -0.25 0.5 1 1 0
> O 0.75 -0.25 0 1 1 0
> K_POINTS {automatic}
> 6 6 6 1 1 1
>
> Regards
>
> zhuzhenye
> __________________________________________________
> ???????????????
> http://cn.mail.yahoo.com
---
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/20060419/c2cc1698/attachment.htm
From sir_puding at tut.by Thu Apr 20 16:01:20 2006
From: sir_puding at tut.by (sir_puding at tut.by)
Date: Thu, 20 Apr 2006 17:01:20 +0300
Subject: [Pw_forum] PW compilation with Threaded and MPI FFTW
Message-ID: <157275544.20060420170120@tut.by>
Hi, All.
Is it possible to use threaded and MPI version of FFTW with espresso?
Configure script produces only -lfftw link option and i think it means
that nonparallel version of FFTW is used. So, how RG space separation
(on several CPU/nodes in pool) is possible in this case?
I'm not a good programmer, so excuse me for dumb questions.
Sergey
--
Please avoid sending me Word or PowerPoint attachments.
See http://www.gnu.org/philosophy/no-word-attachments.html
From hsd22 at hermes.cam.ac.uk Thu Apr 20 16:21:34 2006
From: hsd22 at hermes.cam.ac.uk (H.S.Domingos)
Date: Thu, 20 Apr 2006 15:21:34 +0100 (BST)
Subject: [Pw_forum] PW compilation with Threaded and MPI FFTW
In-Reply-To: <157275544.20060420170120@tut.by>
References: <157275544.20060420170120@tut.by>
Message-ID:
I would also be interested to know this.
For example in the makefile e have :
DFLAGS = ...-D__FFTW -D__USE_INTERNAL_FFTW ....
and that seems to suggest that you can use threaded FFTW by changing the
flags and giving the FFTW path.
Could I then ask if you would kindly tell me what the right flags are ?
best regards,
Helder
=======================================================================
| Dr. Helder S. Domingos |
| |
| INESC Microsyst & Nanotechnol, Lisbon, P-1000 Portugal |
| and |
| R&D unit for Molecular Chemical Physics |
| Chemistry Department, University of Coimbra |
=======================================================================
On Thu, 20 Apr 2006, sir_puding at tut.by wrote:
> Hi, All.
>
> Is it possible to use threaded and MPI version of FFTW with espresso?
> Configure script produces only -lfftw link option and i think it means
> that nonparallel version of FFTW is used. So, how RG space separation
> (on several CPU/nodes in pool) is possible in this case?
>
> I'm not a good programmer, so excuse me for dumb questions.
>
>
> Sergey
>
> --
> Please avoid sending me Word or PowerPoint attachments.
> See http://www.gnu.org/philosophy/no-word-attachments.html
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
From ceresoli at sissa.it Thu Apr 20 16:45:46 2006
From: ceresoli at sissa.it (Davide Ceresoli)
Date: Thu, 20 Apr 2006 16:45:46 +0200
Subject: [Pw_forum] PW compilation with Threaded and MPI FFTW
In-Reply-To: <157275544.20060420170120@tut.by>
References: <157275544.20060420170120@tut.by>
Message-ID: <44479E9A.9080909@sissa.it>
sir_puding at tut.by wrote:
> Hi, All.
>
> Is it possible to use threaded and MPI version of FFTW with espresso?
Is not that simple as just linking a different library (libfftw_threads.a)!
The MPI and threaded version of FFTW require different subroutine calls.
See e.g.: http://www.fftw.org/fftw2_doc/fftw_4.html#SEC47
> Configure script produces only -lfftw link option and i think it means
> that nonparallel version of FFTW is used. So, how RG space separation
> (on several CPU/nodes in pool) is possible in this case?
Parallel FFT is implemented in espresso. See Modules/fft_base.f90.
Davide
From konstantin_kudin at yahoo.com Thu Apr 20 17:36:30 2006
From: konstantin_kudin at yahoo.com (Konstantin Kudin)
Date: Thu, 20 Apr 2006 08:36:30 -0700 (PDT)
Subject: [Pw_forum] question on potential energy in constant pressure simulations
Message-ID: <20060420153630.11356.qmail@web52011.mail.yahoo.com>
Hi all,
I am doing the CP dynamics with one of the lattice vectors being
optimized as well.
I am using a very recent CVS (yesterday's), which should have the
correct stress (knocking on wood).
The parameters for the modified kinetic energy functional are:
ecutwfc = 30.0, ecfixed = 25.0, qcutz = 25.0, q2sigma = 3.0
This is for what used to be just ecutwfc=25.0 in constant volume
simulations.
So what I am seeing is that while the stress becomes smaller, the
potential energy of the system goes up at the same time with the
decreasing stress.
Should not the decrease in stress correspond to the *decreasing*
potential energy? Or is there some funny business with this modified
kinetic energy functional such that I should disregard the potential
energy values and trust the stress instead ?
Thanks!
Kostya
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
From marzari at MIT.EDU Thu Apr 20 17:47:50 2006
From: marzari at MIT.EDU (Nicola Marzari)
Date: Thu, 20 Apr 2006 17:47:50 +0200
Subject: [Pw_forum] question on potential energy in constant pressure
simulations
In-Reply-To: <20060420153630.11356.qmail@web52011.mail.yahoo.com>
References: <20060420153630.11356.qmail@web52011.mail.yahoo.com>
Message-ID: <4447AD26.2050007@mit.edu>
Dear Kostya,
great that your are testing the constant-pressure cp.
There is still an unresolved issue (bug...) for ultrasoft psp, in
the diagonal terms only - that could explain your results. On the bright
side, a lot of other arcane issues (bugs) that had crept in the CVS
have been recently found and fixed by Carlo C. and Paolo G.
The safest first step would be to look at the numerical derivative of
the ground-state energy with respect to your variable lattice
vector - should be equal with the calculated stress
(again, if you are using psp, it will not...).
nicola
Konstantin Kudin wrote:
> Hi all,
>
> I am doing the CP dynamics with one of the lattice vectors being
> optimized as well.
>
> I am using a very recent CVS (yesterday's), which should have the
> correct stress (knocking on wood).
>
> The parameters for the modified kinetic energy functional are:
> ecutwfc = 30.0, ecfixed = 25.0, qcutz = 25.0, q2sigma = 3.0
> This is for what used to be just ecutwfc=25.0 in constant volume
> simulations.
>
> So what I am seeing is that while the stress becomes smaller, the
> potential energy of the system goes up at the same time with the
> decreasing stress.
>
> Should not the decrease in stress correspond to the *decreasing*
> potential energy? Or is there some funny business with this modified
> kinetic energy functional such that I should disregard the potential
> energy values and trust the stress instead ?
>
> Thanks!
> 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
--
---------------------------------------------------------------------
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 konstantin_kudin at yahoo.com Thu Apr 20 18:20:40 2006
From: konstantin_kudin at yahoo.com (Konstantin Kudin)
Date: Thu, 20 Apr 2006 09:20:40 -0700 (PDT)
Subject: [Pw_forum] question on potential energy in constant pressure simulations
In-Reply-To: <4447AD26.2050007@mit.edu>
Message-ID: <20060420162040.77487.qmail@web52002.mail.yahoo.com>
Hi Nicola,
Thanks for the info! I was under the impression that ALL bugs were
fixed by now, but apparently - not. I do in fact use ultra-soft PSPs,
so that would explain my results.
An important consideration for me is that I did not intend just to
test the stress, but rather to use it for some production jobs. So
unfortunately it looks like this has to wait.
Kostya
--- Nicola Marzari wrote:
>
>
> Dear Kostya,
>
> great that your are testing the constant-pressure cp.
>
> There is still an unresolved issue (bug...) for ultrasoft psp, in
> the diagonal terms only - that could explain your results. On the
> bright
> side, a lot of other arcane issues (bugs) that had crept in the CVS
> have been recently found and fixed by Carlo C. and Paolo G.
>
> The safest first step would be to look at the numerical derivative of
> the ground-state energy with respect to your variable lattice
> vector - should be equal with the calculated stress
> (again, if you are using psp, it will not...).
>
>
>
> nicola
>
>
> Konstantin Kudin wrote:
> > Hi all,
> >
> > I am doing the CP dynamics with one of the lattice vectors being
> > optimized as well.
> >
> > I am using a very recent CVS (yesterday's), which should have the
> > correct stress (knocking on wood).
> >
> > The parameters for the modified kinetic energy functional are:
> > ecutwfc = 30.0, ecfixed = 25.0, qcutz = 25.0, q2sigma = 3.0
> > This is for what used to be just ecutwfc=25.0 in constant volume
> > simulations.
> >
> > So what I am seeing is that while the stress becomes smaller, the
> > potential energy of the system goes up at the same time with the
> > decreasing stress.
> >
> > Should not the decrease in stress correspond to the *decreasing*
> > potential energy? Or is there some funny business with this
> modified
> > kinetic energy functional such that I should disregard the
> potential
> > energy values and trust the stress instead ?
> >
> > Thanks!
> > 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
>
> --
> ---------------------------------------------------------------------
> 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
>
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
From cesar at cuca.mda.cinvestav.mx Thu Apr 20 20:14:20 2006
From: cesar at cuca.mda.cinvestav.mx (Cesar Alberto Cab Cauich)
Date: Thu, 20 Apr 2006 13:14:20 -0500 (CDT)
Subject: [Pw_forum] Problem with examples
Message-ID:
Hi espresso users, I am a new user for this code, I already Work with
siesta, and I want Know about the abilities for cp and pwscf. I compiled
the code without errors in a xeon box 3.2 GHz with intel fortran 9.0 and
mkl 8.0 using "configure" utility.
When I try to run the examples, I obtain this error for al.scf.cg.out:
G cutoff = 85.4897 ( 869 G-vectors) FFT grid: ( 15, 15, 15)
nbndx = 6 nbnd = 6 natomwfc = 9 npwx = 113
nelec = 3.00 nkb = 4 ngl = 31
Initial potential from superposition of free atoms
starting charge 2.99794, renormalised to 3.00000
Starting wfc are atomic
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
from cfts_3 : error # 1
routine called by wrong architecture
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
stopping ...
and for al.band.cg.out:
G cutoff = 85.4897 ( 869 G-vectors) FFT grid: ( 15, 15, 15)
nbndx = 8 nbnd = 8 natomwfc = 9 npwx = 113
nelec = 3.00 nkb = 4 ngl = 31
Cannot read rho : file not found
Initial potential from superposition of free atoms
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
from potinit : error # 1
starting and expected charges differ
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
stopping ...
In resume, there are errors for all .out files .....
Some idea?
Greetings
Cesar Alberto Cab Cauich
PhD Student
Departamento de Fisica Aplicada
Centro de Investigacion y Estudios Avanzados
Merida, Mexico.
From nhviet at sissa.it Thu Apr 20 20:12:17 2006
From: nhviet at sissa.it (Nguyen Huy Viet)
Date: Thu, 20 Apr 2006 20:12:17 +0200 (CEST)
Subject: [Pw_forum] Problem with examples
In-Reply-To:
References:
Message-ID:
Hi,
>
> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
> from cfts_3 : error # 1
> routine called by wrong architecture
>
> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>
> stopping ...
It seems that there are problems with using ifc 9.0 (possibly
with precompilation, but I am not very sure) to compile ESPRESSO.
I am using FEDORA CORE 4 and exactly the same error messages appeared.
I solve this problem by using ifc 8.1.
Hope this helps.
Regards,
Huy Viet
--
Huy-Viet Nguyen
PhD student in Condensed Matter Theory Sector,
International School for Advanced Studies,
via Beirut 2-4, I-34014, Trieste, Italy
Office: R 205 New Building
Phone: +39 040 3787 492
Fax: +39 040 3787 528
--
From stewart at cnf.cornell.edu Thu Apr 20 20:59:33 2006
From: stewart at cnf.cornell.edu (stewart at cnf.cornell.edu)
Date: Thu, 20 Apr 2006 14:59:33 -0400
Subject: [Pw_forum] Re: Problem with examples
In-Reply-To:
References:
Message-ID: <20060420185933.22444.qmail@xuxa.iecc.com>
Hi Cesar,
Can you post the Make.sys file generated by configure. This will help
determine where the problem is. I have been able to get the espresso
package 3.0 to compile with ifort 9.0 and mkl 8.0 on a xeon box.
Best regards,
Derek
Cesar Alberto Cab Cauich writes:
> Hi espresso users, I am a new user for this code, I already Work with
> siesta, and I want Know about the abilities for cp and pwscf. I compiled
> the code without errors in a xeon box 3.2 GHz with intel fortran 9.0 and
> mkl 8.0 using "configure" utility.
> When I try to run the examples, I obtain this error for al.scf.cg.out:
>
>
>
>
>
>
> G cutoff = 85.4897 ( 869 G-vectors) FFT grid: ( 15, 15, 15)
>
> nbndx = 6 nbnd = 6 natomwfc = 9 npwx = 113
> nelec = 3.00 nkb = 4 ngl = 31
>
> Initial potential from superposition of free atoms
>
> starting charge 2.99794, renormalised to 3.00000
> Starting wfc are atomic
>
>
> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
> from cfts_3 : error # 1
> routine called by wrong architecture
>
> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>
> stopping ...
>
>
>
>
> and for al.band.cg.out:
>
>
>
>
> G cutoff = 85.4897 ( 869 G-vectors) FFT grid: ( 15, 15, 15)
>
> nbndx = 8 nbnd = 8 natomwfc = 9 npwx = 113
> nelec = 3.00 nkb = 4 ngl = 31
> Cannot read rho : file not found
>
> Initial potential from superposition of free atoms
>
>
> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
> from potinit : error # 1
> starting and expected charges differ
>
> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>
> stopping ...
>
>
>
>
>
>
> In resume, there are errors for all .out files .....
> Some idea?
>
>
> Greetings
>
>
>
> Cesar Alberto Cab Cauich
> PhD Student
> Departamento de Fisica Aplicada
> Centro de Investigacion y Estudios Avanzados
> Merida, Mexico.
>
>
>
> _______________________________________________
> 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 konstantin_kudin at yahoo.com Fri Apr 21 02:42:19 2006
From: konstantin_kudin at yahoo.com (Konstantin Kudin)
Date: Thu, 20 Apr 2006 17:42:19 -0700 (PDT)
Subject: [Pw_forum] Problem with Wannier function centers
In-Reply-To: <20060420185933.22444.qmail@xuxa.iecc.com>
Message-ID: <20060421004219.57870.qmail@web52014.mail.yahoo.com>
Hi all,
It appears that the Wannier code in CP (recent CVS) outputs the
Wannier function centers incorrectly for cells that are specified by
ibrav. The input producing strange centers for a water molecule is
here:
https://kiev.princeton.edu/failing_example_wannier.j
With these coordinates:
ATOMIC_POSITIONS angstrom
O 0.99999987E+00 0.99999761E+00 0.11192157E+01
H 0.10000039E+01 0.17655545E+01 0.52313723E+00
H 0.99999715E+00 0.23446441E+00 0.52313869E+00
wannier centers (fort.26) come out as :
13.042997 11.149277 10.237439
1.702475 -0.381710 -1.063049
1.309338 -0.584178 -0.629999
2.165164 -0.547665 -1.062867
This does not make much sense ...
Kostya
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
From liyanpcl at yahoo.com.cn Fri Apr 21 07:58:01 2006
From: liyanpcl at yahoo.com.cn (li yan)
Date: Fri, 21 Apr 2006 13:58:01 +0800 (CST)
Subject: [Pw_forum] problems about the structure
Message-ID: <20060421055801.34307.qmail@web15608.mail.cnb.yahoo.com>
dear all,
I'm doing the calculation about the scf CuI (P21/M). But the Sym.Ops. in the output file is not constent with that in Materials studio or Wien2k. My input file for pwscf is
&control
calculation='scf',
restart_mode='from_scratch',
prefix='cui'
pseudo_dir = '$PSEUDO_DIR/',
outdir='$TMP_DIR/'
tstress=.t.,
tprnfor=.t.
/
&system
ibrav = 14, celldm(1) =celldm(1) =6.779395,celldm(2) =1.11291986,celldm(3)=1.47949825,celldm(4)=-0.146169358, nat= 4, ntyp= 2,
ecutwfc=100,
/
&electrons
mixing_beta = 0.7
conv_thr = 1.0d-8
/
ATOMIC_SPECIES
Cu 107.87 cu.cpi.UPF
I 126.90 i.cpi.UPF
ATOMIC_POSITIONS {crystal}
Cu 0.815700000 0.75000000 0.77970000
Cu 0.184300000 0.25000000 0.22030000
I 0.300400000 0.25000000 0.71340000
I 0.699600000 0.75000000 0.28660000
K_POINTS {automatic}
8 8 6 0 0 0
EOF
and the Sym.Ops. in the output file cui.scf.out is 2 (with inversion), while it is 4 in Materials studio or Wien2k.
could you please tell me the reason about this difference?
best regards.
---------------------------------
???????????????????????
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060421/84265cc7/attachment.htm
From luch001 at 126.com Fri Apr 21 08:36:16 2006
From: luch001 at 126.com (luch001)
Date: Fri, 21 Apr 2006 14:36:16 +0800 (CST)
Subject: [Pw_forum] A question about elelectron-phonon coupli
ng
Message-ID: <44487D60.00003A.17008@m54.126.com>
Dear all,
I use the version 2.1.5 of pwscf. When I calculate the electron-phonon coupling constant of the LiH. I calculate 47 k points in the IBZ.The output file lambda.dat is like this:
# degauss lambda int alpha2F N(Ef)
0.010 0.519654 0.000000 94.413 0.949231
0.020 0.481122 0.000000 94.412 0.886274
0.030 0.470335 0.000000 94.412 0.880906
0.040 0.458617 0.000000 94.413 0.857064
0.050 0.430279 0.000000 94.415 0.810680
0.060 0.431049 0.000000 94.416 0.811986
0.070 0.475063 0.000000 94.415 0.877232
0.080 0.526288 0.000000 94.412 0.958069
0.090 0.563168 0.000000 94.410 1.022181
0.100 0.581296 0.000000 94.409 1.062365
We kown that . My question is why the lambda has some value while the int alpha2F are totally zero.
Any comments will be appreciated.
Best regards.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060421/a2fe22a3/attachment.htm
From cbarreteau at cea.fr Fri Apr 21 09:29:46 2006
From: cbarreteau at cea.fr (Cyrille Barreteau)
Date: Fri, 21 Apr 2006 09:29:46 +0200 (CEST)
Subject: [Pw_forum] SOC on surfaces: charge is wrong...
In-Reply-To:
References: <20060414053744.17757.95638.Mailman@democritos.sissa.it>
Message-ID: <60281.132.166.17.128.1145604586.squirrel@dsm-mail.saclay.cea.fr>
Dear PWscf users,
I am trying to run pw.x inlcuding spin-orbit coupling
for a Febcc(100) surface (11 layers) and I am facing some problems.
After the first iteration the code stops with the following
error:
WARNING: integrated charge= 87.50803650, expected= 88.00000000
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
from electrons : error # 1
charge is wrong
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Any idea for the origin of this mistake.
I have included below the corresponding input and output files.
thanks in advance
Cyrille
--
==================================================================
Cyrille Barreteau | phone : +33 (0)1 69 08 29 51
CEA Saclay | fax : +33 (0)1 69 08 84 46
DSM/DRECAM/SPCSI | email : cyrille.barreteau at cea.fr
Batiment 462 |
91191 Gif sur Yvette Cedex FRANCE
~~~~~~~~~~~~~~~~~~~~~~~~
http://www-drecam.cea.fr/spcsi/index.php
http://www-drecam.cea.fr/Images/Pisp/cbarreteau/cbarreteau_fr.html
http://www-drecam.cea.fr/Phocea/Membres/Cours/index.php
==================================================================
-------------- next part --------------
A non-text attachment was scrubbed...
Name: fe_surf_001_so_00.in
Type: application/octet-stream
Size: 1436 bytes
Desc: not available
Url : /pipermail/attachments/20060421/c91fd465/attachment.obj
-------------- next part --------------
A non-text attachment was scrubbed...
Name: fe_surf_001_so_00.out
Type: application/octet-stream
Size: 20915 bytes
Desc: not available
Url : /pipermail/attachments/20060421/c91fd465/attachment-0001.obj
From ceresoli at sissa.it Fri Apr 21 09:53:03 2006
From: ceresoli at sissa.it (Davide Ceresoli)
Date: Fri, 21 Apr 2006 09:53:03 +0200
Subject: [Pw_forum] Re: Problem with examples
In-Reply-To: <20060420185933.22444.qmail@xuxa.iecc.com>
References: <20060420185933.22444.qmail@xuxa.iecc.com>
Message-ID: <44488F5F.9080208@sissa.it>
stewart at cnf.cornell.edu wrote:
> Hi Cesar,
> Can you post the Make.sys file generated by configure. This will help
> determine where the problem is. I have been able to get the espresso
> package 3.0 to compile with ifort 9.0 and mkl 8.0 on a xeon box.
> Best regards,
> Derek
Dear all,
I run into this problem several months ago.
Even though -DFFTW is defined in the make.sys, there is a (documented) bug
in the fortran preprocessor of the 9.0 compiler, at least in some
version of it.
Look at this piece of code PW/cft3s.f90:
#if defined (__FFT_MODULE_DRV) && ( defined __AIX || defined __FFTW )
!
CALL cfft3ds( f, n1, n2, n3, nx1, nx2, nx3, 1, dffts%isind, dffts%iplw )
!
#elif defined (__FFT_MODULE_DRV)
!
CALL cfft3d( f, n1, n2, n3, nx1, nx2, nx3, 1 )
!
#elif defined (NOPENCILS)
!
CALL cft_3( f, n1, n2, n3, nx1, nx2, nx3, 2, 1 )
!
#else
!
CALL cfts_3( f, n1, n2, n3, nx1, nx2, nx3, 2, 1, dffts%isind, dffts%iplw )
!
#endif
It happens that last case (#else) always passes through the fortran
preprocessor (fpp).
Possible solutions:
1) upgrade you compiler
2) substitute /opt/intel/blahblah/bin/fpp with that from version 8.1
3) comment all lines such as CALL cfts_3(...)
HTH.
Davide
From scandolo at ictp.it Fri Apr 21 09:52:56 2006
From: scandolo at ictp.it (Sandro Scandolo)
Date: Fri, 21 Apr 2006 09:52:56 +0200
Subject: [Pw_forum] question on potential energy in constant pressure
simulations
In-Reply-To: <20060420162040.77487.qmail@web52002.mail.yahoo.com>
References: <20060420162040.77487.qmail@web52002.mail.yahoo.com>
Message-ID: <44488F58.6050303@ictp.it>
Dear Kostya,
an additional note on the stress tensor, perhaps trivial for you, but
hopefully useful for other less experienced users:
the stress tensor calculated in the Q-E codes is based on the expression
derived by Nielsen and Martin (PRL 50, 697, 1983). Nielsen and Martin's
derivation is based on the Hellmann-Feynman theorem, and therefore
assumes that the basis set is complete. It can be shown that the
derivation is valid also when the basis set is not complete, as long as
the derivative is taken in such a way that the basis set does not
change, to first order, upon infinitesimal changes of the strain tensor.
As a consequence, you should expect the calculated stress tensor to
coincide with the derivative of the energy with respect to the strain,
as calculated by small finite strains, *only* if you are able to keep
the same number of plane waves in the calculations with and without
strain. For isotropic strains (i.e. volume changes, where V->V+dV) this
is easily done by changing the energy cut-off by [(V+dV)/V]^2/3 . For
anisotropic strains (i.e. for the off-diagonal elements of the stress
tensor) it may be more tricky. This is the correct procedure for testing
if the stress tensor is calculated correctly by the code. However, as
you probably know, when the basis set is not complete, a better (in
physical terms) estimate of the stress tensor is obtained by
differentiating the total energy (or potential energy in your notation)
calculated by keeping the wfc cut-off (ecutwfc) fixed in the finite
difference calculation. Unfortunately this leads to discontinous
pressure-volume (or stress-strain) curves, as the basis set size
changes discontinuosly upon strain changes, if the cut-off is fixed (NB:
to be more precise, it leads to a continuous curve only if the k-point
sampling is infinitely dense, which is never the case in practice, and
certainly not with Gamma-point only).
The "qcutz" trick has been introduced as a partial relief of the above
problem. By introducing a (smooth) penalty function for plane waves
whose kinetic energy exceeds a given *fixed* cut off (ecfixed), you can
forget about the need to keep the cut-off fixed in the construction of
the basis set (ecutwfc), as long as ecutwfc is above the tail of the
smooth penalty function. All this is described in M. Bernasconi et al,
J. Phys. Chem. Solids 56, 501 (1995). As a strightforward consequence,
the stress tensor calculated with this trick should be much less
sensitive to whether it is calculated at constant cut-off or at constant
number of plane waves. Hence, Nielsen and Martin's expression, with this
modified functional, gives you a stress tensor in much better agreement
with the physically sounder expression obtained by keeping the cut-off
fixed. However, you should not expect it to coincide exactly with the
derivative of the potential energy when the derivative is calculated by
keeping the cut-off fixed. You should however expect it to coincide
exactly with the derivative taken at constant number of plane waves
(like for the stress tensor without "qcutz").
Regards,
Sandro
Konstantin Kudin wrote:
> Hi Nicola,
>
> Thanks for the info! I was under the impression that ALL bugs were
>fixed by now, but apparently - not. I do in fact use ultra-soft PSPs,
>so that would explain my results.
>
> An important consideration for me is that I did not intend just to
>test the stress, but rather to use it for some production jobs. So
>unfortunately it looks like this has to wait.
>
> Kostya
>
>
>--- Nicola Marzari wrote:
>
>
>
>>Dear Kostya,
>>
>>great that your are testing the constant-pressure cp.
>>
>>There is still an unresolved issue (bug...) for ultrasoft psp, in
>>the diagonal terms only - that could explain your results. On the
>>bright
>>side, a lot of other arcane issues (bugs) that had crept in the CVS
>>have been recently found and fixed by Carlo C. and Paolo G.
>>
>>The safest first step would be to look at the numerical derivative of
>>the ground-state energy with respect to your variable lattice
>>vector - should be equal with the calculated stress
>>(again, if you are using psp, it will not...).
>>
>>
>>
>> nicola
>>
>>
>>Konstantin Kudin wrote:
>>
>>
>>> Hi all,
>>>
>>> I am doing the CP dynamics with one of the lattice vectors being
>>>optimized as well.
>>>
>>> I am using a very recent CVS (yesterday's), which should have the
>>>correct stress (knocking on wood).
>>>
>>> The parameters for the modified kinetic energy functional are:
>>>ecutwfc = 30.0, ecfixed = 25.0, qcutz = 25.0, q2sigma = 3.0
>>> This is for what used to be just ecutwfc=25.0 in constant volume
>>>simulations.
>>>
>>> So what I am seeing is that while the stress becomes smaller, the
>>>potential energy of the system goes up at the same time with the
>>>decreasing stress.
>>>
>>> Should not the decrease in stress correspond to the *decreasing*
>>>potential energy? Or is there some funny business with this
>>>
>>>
>>modified
>>
>>
>>>kinetic energy functional such that I should disregard the
>>>
>>>
>>potential
>>
>>
>>>energy values and trust the stress instead ?
>>>
>>> Thanks!
>>> 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
>>>
>>>
>>--
>>---------------------------------------------------------------------
>>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
>>
>>
>>
>
>
>__________________________________________________
>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 marzari at MIT.EDU Fri Apr 21 17:39:46 2006
From: marzari at MIT.EDU (Nicola Marzari)
Date: Fri, 21 Apr 2006 17:39:46 +0200
Subject: [Pw_forum] Re: Problem with Wannier function centers
In-Reply-To: <20060421150915.84822.qmail@web52014.mail.yahoo.com>
References: <20060421150915.84822.qmail@web52014.mail.yahoo.com>
Message-ID: <4448FCC2.9090709@mit.edu>
Hi Kostya,
it could be a bug, but it could also be an intrinsic issue.
The localization functional, in general, has multiple minima, of
which only one is meaningful in the chemical sense we have grown used to
(sometimes there can be more than one meaningful minimum, but that's
a longer story).
In fact, in the general wannier code (www.wannier.org) we use
projections onto trial wavefunctions to start our localization from
the basin of attraction of the minimum we want. The existence of
"false" minima is ultimately related to the fact that Im ln is a multiply-valued function, and the arbitrary phases of the u_nk
that the wannier code tries to optimize can be such that there is
a 2 pi discontinuity, somewhere in the BZ, between u_nk and its
neigbours.
If you use CP/Gamma sampling (only real wavefunctions), in an orthogonal
cell, this problem has never appeared.
In CP, if the cell is not orthogonal, and you are using more than
3 G vectors to calculate the WF center (it's a sum on a shell of
neighboring G vectors, of Im ln ), you
can still converge to a local minimum.
Yudong Wu had tested the code for all the ibrav, so the Wannier
algebra should be correct - maybe you are converging to a local
minimum.
1) avoid non-orthogonal cells in CP
2) use www.wannier.org, in combination with pwscf, as soon as it is
released (end of April).
3) use projections in CP. We'll add this in the future, but probably
not right away.
nicola
Konstantin Kudin wrote:
> Hi Manu & Nicola,
>
> The centers produced for an orthogonal cell seems to be fine (so the
> code works in generic sense), however, when the cell vectors are
> general, things do not work.
>
> I have peeked briefly at the part of the code that computes the
> centers [wf.f90, near line that has "write( 26" ], it appears that the
> calculation is clearly wrong if the cells are not orthogonal.
>
> Kostya
>
>
> --- Nicola Marzari wrote:
>
>
>>Dear Manu,
>>
>>do you have any sense if the Wannier part still works ?
>>
>>This is your code in the CP, correct ?
>>
>> nicola
>>
>>
>>Konstantin Kudin wrote:
>>
>>
>>> Hi all,
>>>
>>> It appears that the Wannier code in CP (recent CVS) outputs the
>>>Wannier function centers incorrectly for cells that are specified
>>
>>by
>>
>>>ibrav. The input producing strange centers for a water molecule is
>>>here:
>>> https://kiev.princeton.edu/failing_example_wannier.j
>>>
>>> With these coordinates:
>>>ATOMIC_POSITIONS angstrom
>>> O 0.99999987E+00 0.99999761E+00 0.11192157E+01
>>> H 0.10000039E+01 0.17655545E+01 0.52313723E+00
>>> H 0.99999715E+00 0.23446441E+00 0.52313869E+00
>>>
>>> wannier centers (fort.26) come out as :
>>> 13.042997 11.149277 10.237439
>>> 1.702475 -0.381710 -1.063049
>>> 1.309338 -0.584178 -0.629999
>>> 2.165164 -0.547665 -1.062867
>>>
>>> This does not make much sense ...
>>>
>>> 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
>>
>>--
>>---------------------------------------------------------------------
>>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
>>
>
>
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam? Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
--
---------------------------------------------------------------------
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 cesar at cuca.mda.cinvestav.mx Fri Apr 21 20:48:42 2006
From: cesar at cuca.mda.cinvestav.mx (Cesar Alberto Cab Cauich)
Date: Fri, 21 Apr 2006 13:48:42 -0500 (CDT)
Subject: [Pw_forum] Re: Problem with examples
Message-ID:
Thank you very much for your help, Derek, here is my make.sys:
# make.sys. Generated from make.sys.in by configure.
# compilation rules
.SUFFIXES :
.SUFFIXES : .o .c .f .f90
.f90.o:
$(MPIF90) $(F90FLAGS) -c $<
.f.o:
$(F77) $(FFLAGS) -c $<
.c.o:
$(CC) $(CFLAGS) -c $<
CC = icc
MPICC = icc
CFLAGS = -O3 $(DFLAGS) $(IFLAGS)
CPP = icc -E
CPPFLAGS = $(DFLAGS) $(IFLAGS)
F90 = ifort
MPIF90 = ifort
F90FLAGS = $(FFLAGS) -nomodule -fpp $(FDFLAGS) $(IFLAGS) $(MODFLAGS)
F90FLAGS_NOOPT = $(FFLAGS_NOOPT) -nomodule -fpp $(FDFLAGS) $(IFLAGS)$(MODFLAGS)
F77 = ifort
MPIF77 = ifort
FFLAGS = -O2 -tpp6 -assume byterecl
FFLAGS_NOOPT = -O0 -assume byterecl
LD = ifort
LDFLAGS =
AR = ar
ARFLAGS = ruv
RANLIB = echo
BLAS_LIBS = -L/opt/intel/mkl/8.0.1/lib/32 -lmkl_ia32 -lguide
-lpthread
LAPACK_LIBS = -lmkl_lapack
FFT_LIBS =
MPI_LIBS =
MASS_LIBS =
# -----------------------------
# application-specific settings
# See include/defs.h.README for a list of precompilation options
# (possible arguments to -D or -U) and their meaning
DFLAGS = -D__LINUX -D__INTEL -D__FFTW -D__USE_INTERNAL_FFTW
FDFLAGS = $(DFLAGS)
IFLAGS = -I../include
MODFLAGS = -I. -I../Modules -I../PW -I../PH -I../iotk/src
LIBOBJS = ../flib/ptools.a ../flib/flib.a ../clib/clib.a
../iotk/src/libiotk.a
# LIBS must contain the location of all needed external libraries
LIBS = $(LAPACK_LIBS) $(BLAS_LIBS) $(FFT_LIBS) $(MPI_LIBS)
$(MASS_LIBS)
# MYLIB can be one of the following (depending on LIBS):
# blas : compile the local copy of blas routines
# lapack : compile the local copy of lapack routines
# blas_and_lapack : all of the above - use this for a quick test
# or if you don't have an optimized blas/lapack library
# lapack_ibm : compile only lapack routines not present in IBM ESSL
# use this together with IBM ESSL
# lapack_t3e : compile only lapack routines not present in T3E scilib
# use this together with T3E scilib
# lapack_mkl : compile only lapack routines not present in Intel MKL
# use this together with Intel MKL
MYLIB = lapack_mkl
From stewart at cnf.cornell.edu Fri Apr 21 20:37:35 2006
From: stewart at cnf.cornell.edu (stewart at cnf.cornell.edu)
Date: Fri, 21 Apr 2006 14:37:35 -0400
Subject: [Pw_forum] Re: Problem with examples
In-Reply-To:
References:
Message-ID: <20060421183735.18803.qmail@xuxa.iecc.com>
Hi Cesar,
That looks almost identical to the one that I am using to compile things.
The only difference is that I am using the external fftw library for the FFT
routines. This could be the difference since I believe your crash was in
some fft calls. Try downloading the fftw library, compile it with the intel
compiler, and link it in. It should be a little faster anyway :).
Regards,
Derek
Cesar Alberto Cab Cauich writes:
>
> Thank you very much for your help, Derek, here is my make.sys:
>
>
> # make.sys. Generated from make.sys.in by configure.
>
> # compilation rules
>
> .SUFFIXES :
> .SUFFIXES : .o .c .f .f90
>
> .f90.o:
> $(MPIF90) $(F90FLAGS) -c $<
>
> .f.o:
> $(F77) $(FFLAGS) -c $<
>
> .c.o:
> $(CC) $(CFLAGS) -c $<
>
>
> CC = icc
> MPICC = icc
> CFLAGS = -O3 $(DFLAGS) $(IFLAGS)
> CPP = icc -E
> CPPFLAGS = $(DFLAGS) $(IFLAGS)
> F90 = ifort
> MPIF90 = ifort
> F90FLAGS = $(FFLAGS) -nomodule -fpp $(FDFLAGS) $(IFLAGS) $(MODFLAGS)
> F90FLAGS_NOOPT = $(FFLAGS_NOOPT) -nomodule -fpp $(FDFLAGS) $(IFLAGS)$(MODFLAGS)
> F77 = ifort
> MPIF77 = ifort
> FFLAGS = -O2 -tpp6 -assume byterecl
> FFLAGS_NOOPT = -O0 -assume byterecl
> LD = ifort
> LDFLAGS =
> AR = ar
> ARFLAGS = ruv
> RANLIB = echo
> BLAS_LIBS = -L/opt/intel/mkl/8.0.1/lib/32 -lmkl_ia32 -lguide
> -lpthread
> LAPACK_LIBS = -lmkl_lapack
> FFT_LIBS =
> MPI_LIBS =
> MASS_LIBS =
>
> # -----------------------------
> # application-specific settings
>
> # See include/defs.h.README for a list of precompilation options
> # (possible arguments to -D or -U) and their meaning
> DFLAGS = -D__LINUX -D__INTEL -D__FFTW -D__USE_INTERNAL_FFTW
> FDFLAGS = $(DFLAGS)
> IFLAGS = -I../include
> MODFLAGS = -I. -I../Modules -I../PW -I../PH -I../iotk/src
>
> LIBOBJS = ../flib/ptools.a ../flib/flib.a ../clib/clib.a
> ../iotk/src/libiotk.a
> # LIBS must contain the location of all needed external libraries
> LIBS = $(LAPACK_LIBS) $(BLAS_LIBS) $(FFT_LIBS) $(MPI_LIBS)
> $(MASS_LIBS)
> # MYLIB can be one of the following (depending on LIBS):
> # blas : compile the local copy of blas routines
> # lapack : compile the local copy of lapack routines
> # blas_and_lapack : all of the above - use this for a quick test
> # or if you don't have an optimized blas/lapack library
> # lapack_ibm : compile only lapack routines not present in IBM ESSL
> # use this together with IBM ESSL
> # lapack_t3e : compile only lapack routines not present in T3E scilib
> # use this together with T3E scilib
> # lapack_mkl : compile only lapack routines not present in Intel MKL
> # use this together with Intel MKL
> MYLIB = lapack_mkl
>
>
> _______________________________________________
> 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 giannozz at nest.sns.it Sun Apr 23 22:50:35 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Sun, 23 Apr 2006 22:50:35 +0200
Subject: [Pw_forum] building espresso-3.0 with g95?
In-Reply-To:
References:
Message-ID: <200604232250.35947.giannozz@nest.sns.it>
On Sunday 16 April 2006 00:14, Andrey Khavryuchenko wrote:
> g95 -O3 -cpp -D__LINUX -D__G95 -D__FFTW -I../include -I. -I../Modules
> -I../PW -I../PH -I../iotk/src -c ptoolkit.f90 ptoolkit.f90:0: internal
> compiler error: Segmentation fault
> Please submit a full bug report, with preprocessed source if appropriate.
> See http://www.g95.org or mail andyv at firstinter.net for instructions.
> Are there any success stories with g95?
I tried several different versions of g95 in the past (for instance:
G95 (GCC 4.0.1 (g95!) Dec 14 2005) with success. Please do what
the message says: submit a bug report to the author of g95
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 konstantin_kudin at yahoo.com Mon Apr 24 02:34:37 2006
From: konstantin_kudin at yahoo.com (Konstantin Kudin)
Date: Sun, 23 Apr 2006 17:34:37 -0700 (PDT)
Subject: [Pw_forum] still wrong stress in CP [was: question on potential energy in constant pressure simulations]
In-Reply-To: <4447AD26.2050007@mit.edu>
Message-ID: <20060424003437.25438.qmail@web52004.mail.yahoo.com>
--- Nicola Marzari wrote:
> There is still an unresolved issue (bug...) for ultrasoft psp, in
> the diagonal terms only - that could explain your results. On the
> bright
> side, a lot of other arcane issues (bugs) that had crept in the CVS
> have been recently found and fixed by Carlo C. and Paolo G.
I have tried the recent CVS with CP and PWSCF and ultrasoft psp's, and
indeed, while the forces and the off-diagonal stress terms seem to
match quite well, the diagonal stress elements are wrong. CG's stress
evaluation goes through the standard CP stuff, correct?
Another interesting thing is that trajectory files *.cel and *.str
have the 3x3 components transposed compared to the input, such each
lattice vector goes vertically, not horizontally. Is there any reason
for this?
Kostya
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
From caohui1209 at yahoo.com.cn Tue Apr 25 04:59:34 2006
From: caohui1209 at yahoo.com.cn (=?gb2312?q?=BB=D4=20=B2=DC?=)
Date: Tue, 25 Apr 2006 10:59:34 +0800 (CST)
Subject: [Pw_forum] problems with running example
Message-ID: <20060425025934.28499.qmail@web15702.mail.cnb.yahoo.com>
Dear all:
I've installed espresso-2.1.5 and the whole processure seemed to be successful, including configuration and "make pwall". When I ran example01,
it also seemed to be normal. However, the .out files in the result directory were unexpected. The .out files end with " from cfts_3: error # 1; routine called by wrong architecture". It stopped with the k-point list and did not give out the band-energy.
I chose icc and ifort as c- and fortran- compiler. My libs are lapack and blas. I installed espresso in a parallel system but I did not use mpic and mpif90.
Thank you.
---------------------------------
??1G???????????
????-?????????
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060425/3337c289/attachment.htm
From akhavr at gmail.com Tue Apr 25 20:36:42 2006
From: akhavr at gmail.com (Andrey Khavryuchenko)
Date: Tue, 25 Apr 2006 21:36:42 +0300
Subject: [Pw_forum] Re: building espresso-3.0 with g95?
In-Reply-To: <200604232250.35947.giannozz@nest.sns.it> (Paolo Giannozzi's
message of "Sun, 23 Apr 2006 22:50:35 +0200")
References:
<200604232250.35947.giannozz@nest.sns.it>
Message-ID:
Paolo,
Thanks for your email.
"PG" == Paolo Giannozzi wrote:
PG> On Sunday 16 April 2006 00:14, Andrey Khavryuchenko wrote:
>> g95 -O3 -cpp -D__LINUX -D__G95 -D__FFTW -I../include -I. -I../Modules
>> -I../PW -I../PH -I../iotk/src -c ptoolkit.f90 ptoolkit.f90:0: internal
>> compiler error: Segmentation fault
>> Please submit a full bug report, with preprocessed source if appropriate.
>> See http://www.g95.org or mail andyv at firstinter.net for instructions.
>> Are there any success stories with g95?
PG> I tried several different versions of g95 in the past (for instance:
PG> G95 (GCC 4.0.1 (g95!) Dec 14 2005) with success. Please do what
PG> the message says: submit a bug report to the author of g95
Before reporting the bug, I've updated my g95 installation to the most
recent binary available:
akhavr at t30 ~/src/espresso-3.0 $ g95 --version
G95 (GCC 4.0.3 (g95!) Apr 25 2006)
Copyright (C) 2002-2005 Free Software Foundation, Inc.
Now I fail at cell_base.f90 compilation:
g95 -O3 -cpp -D__LINUX -D__G95 -D__FFTW -I../include -I. -I../Modules -I../PW -I../PH -I../iotk/src -c cell_base.f90
In file cell_base.f90:75
REAL(DP) :: at(3,3) = RESHAPE( (/ 0.0d0 /), (/ 3, 3 /), (/ 0.0d0 /) )
1
Error: Array assignment at (1) has different shape on dimension 1 (3/9)
In file cell_base.f90:76
REAL(DP) :: bg(3,3) = RESHAPE( (/ 0.0d0 /), (/ 3, 3 /), (/ 0.0d0 /) )
1
Error: Array assignment at (1) has different shape on dimension 1 (3/9)
In file cell_base.f90:521
celldm(1) = SQRT( at(1,1)**2 + at(1,2)**2 + at(1,3)**2 )
1
Error: Unexpected array reference at (1)
...
(so on for other 'at' references)
--
Andrey V Khavryuchenko http://www.kds.com.ua/
Silver Bullet Software Solutions http://www.livejournal.com/~akhavr
From giannozz at nest.sns.it Tue Apr 25 21:15:53 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Tue, 25 Apr 2006 21:15:53 +0200
Subject: [Pw_forum] problems with running example
In-Reply-To: <20060425025934.28499.qmail@web15702.mail.cnb.yahoo.com>
References: <20060425025934.28499.qmail@web15702.mail.cnb.yahoo.com>
Message-ID: <200604252115.53012.giannozz@nest.sns.it>
On Tuesday 25 April 2006 04:59, ? ? wrote:
> " from cfts_3: error # 1; routine called by wrong architecture".
this is a known problem of some compilers. Please see the message
by Davide Ceresoli (just 4 day old):
http://www.democritos.it/pipermail/pw_forum/2006-April/004011.html
--
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 Apr 25 22:01:38 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Tue, 25 Apr 2006 22:01:38 +0200
Subject: [Pw_forum] problems about the structure
In-Reply-To: <20060421055801.34307.qmail@web15608.mail.cnb.yahoo.com>
References: <20060421055801.34307.qmail@web15608.mail.cnb.yahoo.com>
Message-ID: <200604252201.38594.giannozz@nest.sns.it>
On Friday 21 April 2006 07:58, li yan wrote:
> I'm doing the calculation about the scf CuI (P21/M). But the Sym.Ops.
> in the output file is not constent with that in Materials studio or Wien2k.
> [...] the Sym.Ops. in the output file cui.scf.out is 2 (with inversion),
> while it is 4 in Materials studio or Wien2k. could you please tell me the
> reason about this difference?
this question has been asked no less than 1001 times and it is
answered in detail in the documentation. See for instance
http://www.pwscf.org/guide/3.0/html-node/node60.html
--
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 konstantin_kudin at yahoo.com Wed Apr 26 05:34:42 2006
From: konstantin_kudin at yahoo.com (Konstantin Kudin)
Date: Tue, 25 Apr 2006 20:34:42 -0700 (PDT)
Subject: [Pw_forum] still wrong stress in CP - update
In-Reply-To: <20060424003437.25438.qmail@web52004.mail.yahoo.com>
Message-ID: <20060426033442.21066.qmail@web52010.mail.yahoo.com>
OK, I went through some of the older CVS versions of CP with an H2O
that had ultra-soft potentials.
The stress appears to be correct in the CP version from Feb. 11, 2003
(V1), I base this conclusion on a loose comparison with recent PW
results.
The other version I tried is from Apr. 3, 2004 (V2), and there the
stress is different. This version appears to get the same results as
the recent CVS.
Comparing the components of the stress from V1 and V2 I get the main
differences in de(ps). The difference is in the 3rd significant digit,
and seems to come from drhotmp. This is the main contribution to the
differences in de(tot).
Kostya
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
From giannozz at nest.sns.it Wed Apr 26 12:24:06 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Wed, 26 Apr 2006 12:24:06 +0200
Subject: [Pw_forum] Error in polarization calculations
In-Reply-To: <443E680A.000003.05003@mfront8.yandex.ru>
References: <443E680A.000003.05003@mfront8.yandex.ru>
Message-ID: <200604261224.06205.giannozz@nest.sns.it>
On Thursday 13 April 2006 17:02, Sergey Lisenkov wrote:
> During Berry Phase calculations (3.0 version), I got the error:
> from ylm_q : error # 12915072
> not programmed for L
> I use RRKJUS pseudopotentials and one pseudo from opium.
> What is wrong?
hard to say with so little information ...
--
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 Apr 26 13:41:10 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Wed, 26 Apr 2006 13:41:10 +0200
Subject: [Pw_forum] A question about elelectron-phonon coupli ng
In-Reply-To: <44487D60.00003A.17008@m54.126.com>
References: <44487D60.00003A.17008@m54.126.com>
Message-ID: <200604261341.10095.giannozz@nest.sns.it>
On Friday 21 April 2006 08:36, luch001 wrote:
> I use the version 2.1.5 of pwscf. When I calculate the electron-phonon
> coupling constant of the LiH. I calculate 47 k points in the IBZ.The
> output file lambda.dat is like this:
> [...]
> My question is why the lambda has some value while the
> int alpha2F are totally zero. Any comments will be appreciated.
with so little information available it is hard to offer any meaningful
comment. Please look inside the lambda.f90 code: it is trivial and
to check what is happening there.
--
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 ezadshojaee at hotmail.com Wed Apr 26 14:52:21 2006
From: ezadshojaee at hotmail.com (Ezad Shojaee)
Date: Wed, 26 Apr 2006 12:52:21 +0000
Subject: [Pw_forum] a question about relaxing
Message-ID:
hi
i want to relax a structure like TiO2 and when i set the parameter " dt "
greater than about " 1 " , i get this message :
from checkallsym : error # 5
not orthogonal operation
i wander if anyone knows the exact role of the " dt " parameter & the
meaning of the error ?
thanks
_________________________________________________________________
Don't just search. Find. Check out the new MSN Search!
http://search.msn.click-url.com/go/onm00200636ave/direct/01/
From degironc at sissa.it Wed Apr 26 14:54:49 2006
From: degironc at sissa.it (degironc)
Date: Wed, 26 Apr 2006 14:54:49 +0200
Subject: [Pw_forum] a question about relaxing
In-Reply-To:
References:
Message-ID: <444F6D99.205@sissa.it>
What kind of relaxation are you performing ?
Notice that Parrinello-Rahman damped dynamics in general does not
preserve symmetry while
Wentzovitch damped dynamics does.
stefano
Ezad Shojaee wrote:
> hi
> i want to relax a structure like TiO2 and when i set the parameter "
> dt " greater than about " 1 " , i get this message :
>
> from checkallsym : error # 5
> not orthogonal operation
>
> i wander if anyone knows the exact role of the " dt " parameter & the
> meaning of the error ?
> thanks
>
> _________________________________________________________________
> Don't just search. Find. Check out the new MSN Search!
> http://search.msn.click-url.com/go/onm00200636ave/direct/01/
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
From ezadshojaee at hotmail.com Wed Apr 26 15:00:38 2006
From: ezadshojaee at hotmail.com (Ezad Shojaee)
Date: Wed, 26 Apr 2006 13:00:38 +0000
Subject: [Pw_forum] about vc-relax
Message-ID:
hi
i want to do a vc-relax for TiO2 and when i set the parameter " dt " greater
than about " 1 " , i get this message :
from checkallsym : error # 5
not orthogonal operation
i wander if anyone knows the exact role of the " dt " parameter & the
meaning of the error ?
thanks
_________________________________________________________________
Don't just search. Find. Check out the new MSN Search!
http://search.msn.click-url.com/go/onm00200636ave/direct/01/
From ezadshojaee at hotmail.com Wed Apr 26 15:10:50 2006
From: ezadshojaee at hotmail.com (Ezad Shojaee)
Date: Wed, 26 Apr 2006 13:10:50 +0000
Subject: [Pw_forum] (no subject)
Message-ID:
i am doing vc-relax with these characters:
cell_dynamics='damp_w'
ion_dynamics='damp'
_________________________________________________________________
FREE pop-up blocking with the new MSN Toolbar - get it now!
http://toolbar.msn.click-url.com/go/onm00200415ave/direct/01/
From giannozz at nest.sns.it Wed Apr 26 17:15:21 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Wed, 26 Apr 2006 17:15:21 +0200
Subject: [Pw_forum] Re: building espresso-3.0 with g95?
In-Reply-To:
References: <200604232250.35947.giannozz@nest.sns.it>
Message-ID: <200604261715.21972.giannozz@nest.sns.it>
On Tuesday 25 April 2006 20:36, Andrey Khavryuchenko wrote:
> I've updated my g95 installation to the most recent binary available:
>
> akhavr at t30 ~/src/espresso-3.0 $ g95 --version
> G95 (GCC 4.0.3 (g95!) Apr 25 2006)
> Copyright (C) 2002-2005 Free Software Foundation, Inc.
>
> Now I fail at cell_base.f90 compilation:
I did the same and it works for me
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 Apr 26 17:18:57 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Wed, 26 Apr 2006 17:18:57 +0200
Subject: [Pw_forum] pw.x always crashes at the end of relax calculations
In-Reply-To: <001a01c662e9$54d27ae0$0ee35ba5@ch0316>
References: <001a01c662e9$54d27ae0$0ee35ba5@ch0316>
Message-ID: <200604261718.57613.giannozz@nest.sns.it>
On Tuesday 18 April 2006 15:09, Eduardo J. Lamas wrote:
> Hi, pw.x always crashes to me during the last check of the relax
> calculations. it happened in versions 2.1.3 and 3.0 and with different
> compilers. [..]
> I am wondering [..] if somebody else was having the same problem.
never heard before. I tried your job (with a smaller cutoff)
and it works for me. If the problem is as reproducible as you
claim it is, you should be able to locate where it crashes and 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 Apr 26 17:31:49 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Wed, 26 Apr 2006 17:31:49 +0200
Subject: [Pw_forum] Re: Re: error while parallel running
In-Reply-To: <2b80d0af0604152347l79575a50h86264cfae33b6dd3@mail.gmail.com>
References: <2b80d0af0604152347l79575a50h86264cfae33b6dd3@mail.gmail.com>
Message-ID: <200604261731.49923.giannozz@nest.sns.it>
On Sunday 16 April 2006 08:47, Lin Zhuang wrote:
> I had succesfully run the input file with serial version pw.x and
> parallel version pw.x if only one CPU had been specified, i.e.,
> $ mpirun c0 pw.x -in inputfile > outputfile
> But when I tried to run the same input file with more than one
> CPU [...] I got error like
> from read_namelists : error # 17
> reading namelist control
mis-configured MPI libraries may have strange ideas of what the
current directory is. Try to provide an explicit path to the file
containing the input data.
--
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 Apr 26 18:04:14 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Wed, 26 Apr 2006 18:04:14 +0200
Subject: [Pw_forum] Phonon frequencies different for equivalent q points
In-Reply-To: <20060411184542.11682.qmail@xuxa.iecc.com>
References: <443BD1B3.6090200@ictp.it> <20060411184542.11682.qmail@xuxa.iecc.com>
Message-ID: <200604261804.14603.giannozz@nest.sns.it>
On Tuesday 11 April 2006 20:45, stewart at cnf.cornell.edu wrote:
> I am looking at the high symmetry lines, Gamma-K-M along the [x,x,0]
> direction going from [0,0,0] to [1,1,0]*(2pi/a) and the Gamma-M line
> along the [x,0,0] going from [0,0,0] to [1/sqrt(3),0,0]*(2pi/a).
are you sure these are the correct high-symmetry lines? please note
that phonon wavevectors must be supplied in cartesian coordinates
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 naromero at gmail.com Wed Apr 26 18:41:39 2006
From: naromero at gmail.com (Nichols A. Romero)
Date: Wed, 26 Apr 2006 12:41:39 -0400
Subject: [Pw_forum] parallel scaling in PWSCF
Message-ID: <6ac064b60604260941u15471f47oe6d45fee2ff6ce8@mail.gmail.com>
Hi,
Has anyone performed any very-large scale DFT calculations on PWSCF
using over 64 processors and over 1000 atoms? Does anyone know what
its current limits (system sizes and processors) are on parallel
computing environments with fast interconnects?
Thanks,
--
Nichols A. Romero, Ph.D.
1613 Denise Dr. Apt. D
Forest Hill, MD 21050
443-567-8328 (C)
410-306-0709 (O)
From akohlmey at cmm.upenn.edu Wed Apr 26 19:38:31 2006
From: akohlmey at cmm.upenn.edu (Axel Kohlmeyer)
Date: Wed, 26 Apr 2006 13:38:31 -0400
Subject: [Pw_forum] parallel scaling in PWSCF
In-Reply-To: <6ac064b60604260941u15471f47oe6d45fee2ff6ce8@mail.gmail.com>
References: <6ac064b60604260941u15471f47oe6d45fee2ff6ce8@mail.gmail.com>
Message-ID: <7b6913e90604261038t25d0bb43le5c8a740e793d981@mail.gmail.com>
On 4/26/06, Nichols A. Romero wrote:
> Hi,
hi.
> Has anyone performed any very-large scale DFT calculations on PWSCF
> using over 64 processors and over 1000 atoms? Does anyone know what
> its current limits (system sizes and processors) are on parallel
> computing environments with fast interconnects?
i've recently run a comparatively large job (272 atoms, 560 electrons,
4x4x4 k-points) across 768 processors on a Cray XT3. i had to hack
some constants to make it work. however the scaling is not (yet)
so good and depending on what kind of atoms you want to put into
your system, i.e. if it is metallic, you may be better of with gamma
only and car-parrinello dynamics. i ran the same system as above
with a car-parrinello code (one that is not in quantum espresso) and
would scale out at 512 nodes for gamma point only on an IBM BG/L.
even though i had to use a much smaller time step, i did get much
more trajectory that way.
judging from the CVS commit messages, efforts to optimize the
quantum espresso codes for that kind of machine with large numbers
of nodes are underway. note, that for systems as large as that,
you might run into the scaling limitations of DFT with plane waves,
so for a system that big using one of those 'linear scaling' DFT codes
could be a more promising alternative.
best regards,
axel.
>
> Thanks,
> --
> Nichols A. Romero, Ph.D.
> 1613 Denise Dr. Apt. D
> Forest Hill, MD 21050
> 443-567-8328 (C)
> 410-306-0709 (O)
> _______________________________________________
> 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 afloris at physik.fu-berlin.de Wed Apr 26 19:51:48 2006
From: afloris at physik.fu-berlin.de (Andrea Floris)
Date: Wed, 26 Apr 2006 19:51:48 +0200 (CEST)
Subject: [Pw_forum] fhi98PP lloc
Message-ID:
Hi everybody,
I generated a NCPP with fhi98PP for Ca with semicore states in the
valence, with lmax=lloc=2 (variables values in fhi98PP).
In the UPF conversion it results (correctly, because of the
Kleinman-Bylander form) that:
Max angular momentum component = 1
Number of Projectors = 2
Wavefunctions nl l occ
3S 0 2.00
3P 1 6.00
3D 2 0.00
Moreover the field is the correct conversion of the local
pseudo that I've chosen in the generation.
Question: in the output of pwscf for Ca results lloc=0. I do not
understand this, I was expecting lloc=2...
Does it have any wrong consequence? I
plotted my bands
and they seems reasonable compared with all electron bands.
thanks in advance for any explanation,
Andrea
>> In the .UPF pseudopotential my "Max angular momentum component"
>> in <> is 1 instead of 2.
>correct: the UPF file contains PPs in separable (Kleinman-Bylander)
>form, so you have l=0 and l=1 projectors, l=2 as the local potential.
>P.
From giannozz at nest.sns.it Wed Apr 26 20:41:20 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Wed, 26 Apr 2006 20:41:20 +0200
Subject: [Pw_forum] fhi98PP lloc
In-Reply-To:
References:
Message-ID: <200604262041.20651.giannozz@nest.sns.it>
On Wednesday 26 April 2006 19:51, Andrea Floris wrote:
> Question: in the output of pwscf for Ca results lloc=0.
> I do not understand this, I was expecting lloc=2...
I am afraid that the printed value for lloc does not correspond
to anything sensible, except for obsolete pseudopotential
formats ...
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 Apr 26 20:40:00 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Wed, 26 Apr 2006 20:40:00 +0200
Subject: [Pw_forum] fhi98PP lloc
In-Reply-To:
References:
Message-ID: <200604262040.00067.giannozz@nest.sns.it>
On Wednesday 26 April 2006 19:51, Andrea Floris wrote:
> Question: in the output of pwscf for Ca results lloc=0.
> I do not understand this, I was expecting lloc=2...
I am afraid that the printed value for lloc does not correspond
to anything sensible, except for obsolete pseudopotential
formats ...
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 reboredofa at ornl.gov Wed Apr 26 21:16:16 2006
From: reboredofa at ornl.gov (Fernando A Reboredo)
Date: Wed, 26 Apr 2006 15:16:16 -0400
Subject: [Pw_forum] parallel scaling in PWSCF
References: <6ac064b60604260941u15471f47oe6d45fee2ff6ce8@mail.gmail.com>
<7b6913e90604261038t25d0bb43le5c8a740e793d981@mail.gmail.com>
Message-ID: <003b01c66965$e4178d80$fcc2db80@ornl.gov>
Nichols
The largest things I have run with pwscf are cobalt clusters with stuff on
the surface
55 Co atoms 12 CO molecules and 12au of vacuum around, 45 Ry ecut.
~560 electron including spin and PBE. Those calculation run in 48 nodes 90
processors. I did not study the scaling carefully. The scaling was
reasonably good provided the number of proccessors was a submultiple of the
FFT grid dimension.
Fernando Reboredo
----- Original Message -----
From: "Axel Kohlmeyer"
To:
Sent: Wednesday, April 26, 2006 1:38 PM
Subject: Re: [Pw_forum] parallel scaling in PWSCF
On 4/26/06, Nichols A. Romero wrote:
> Hi,
hi.
> Has anyone performed any very-large scale DFT calculations on PWSCF
> using over 64 processors and over 1000 atoms? Does anyone know what
> its current limits (system sizes and processors) are on parallel
> computing environments with fast interconnects?
i've recently run a comparatively large job (272 atoms, 560 electrons,
4x4x4 k-points) across 768 processors on a Cray XT3. i had to hack
some constants to make it work. however the scaling is not (yet)
so good and depending on what kind of atoms you want to put into
your system, i.e. if it is metallic, you may be better of with gamma
only and car-parrinello dynamics. i ran the same system as above
with a car-parrinello code (one that is not in quantum espresso) and
would scale out at 512 nodes for gamma point only on an IBM BG/L.
even though i had to use a much smaller time step, i did get much
more trajectory that way.
judging from the CVS commit messages, efforts to optimize the
quantum espresso codes for that kind of machine with large numbers
of nodes are underway. note, that for systems as large as that,
you might run into the scaling limitations of DFT with plane waves,
so for a system that big using one of those 'linear scaling' DFT codes
could be a more promising alternative.
best regards,
axel.
>
> Thanks,
> --
> Nichols A. Romero, Ph.D.
> 1613 Denise Dr. Apt. D
> Forest Hill, MD 21050
> 443-567-8328 (C)
> 410-306-0709 (O)
> _______________________________________________
> 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 ferretti.andrea at unimore.it Wed Apr 26 22:28:21 2006
From: ferretti.andrea at unimore.it (Andrea Ferretti)
Date: Wed, 26 Apr 2006 22:28:21 +0200 (CEST)
Subject: [Pw_forum] parallel scaling in PWSCF
In-Reply-To: <003b01c66965$e4178d80$fcc2db80@ornl.gov>
References: <6ac064b60604260941u15471f47oe6d45fee2ff6ce8@mail.gmail.com>
<7b6913e90604261038t25d0bb43le5c8a740e793d981@mail.gmail.com>
<003b01c66965$e4178d80$fcc2db80@ornl.gov>
Message-ID:
Hi everybody,
I am currently running a Copper surface with 140 Cu atoms + a molecule...
the system has 1642 electrons and (due to metallicity) the calculation is
performed for 985 bands (few kpt, like 4)...
due to the 11 electrons for each Cu atom, I have a huge number of bands in
a (relatively) small cell, and so a (relatively) low number of PWs respect
to nbnd.
taking a look at the dimension of wfc, no problem with memory in
principle, even if, due to the weird
dimensions of the system, non-scalable memory is quite large, around 1Gb.
on a IBM Sp5 machine I observed a severe limit in the scaling passing from
32 to 64 procs using both espresso 2.1.x and espresso 3.0...
( anyway, I succeeded in performing a "relax" calculation for the system
!!!! )
as far as I know, this problem might be connected to a serial part in the
diagonalization which has been parallelized in the current CVS version
(as already pointed out by Axel)...
At the moment I am testing this CVS version against my system, I will let
you know the results as soon as possible...
cheers
andrea
--
Andrea Ferretti
Dipartimento di Fisica, Universita' di Modena e Reggio Emilia
Natl. Res. Center S3 INFM-CNR ( http://s3.infm.it )
Via Campi 213/A I-41100 Modena, Italy
Tel: +39 059 2055283
Fax: +39 059 374794
E-mail: ferretti.andrea at unimore.it
URL: http://www.nanoscience.unimo.it
Please, if possible, don't send me MS Word or PowerPoint attachments
Why? See: http://www.gnu.org/philosophy/no-word-attachments.html
On Wed, 26 Apr 2006, Fernando A Reboredo wrote:
> Nichols
> The largest things I have run with pwscf are cobalt clusters with stuff on
> the surface
> 55 Co atoms 12 CO molecules and 12au of vacuum around, 45 Ry ecut.
> ~560 electron including spin and PBE. Those calculation run in 48 nodes 90
> processors. I did not study the scaling carefully. The scaling was
> reasonably good provided the number of proccessors was a submultiple of the
> FFT grid dimension.
> Fernando Reboredo
>
> ----- Original Message -----
> From: "Axel Kohlmeyer"
> To:
> Sent: Wednesday, April 26, 2006 1:38 PM
> Subject: Re: [Pw_forum] parallel scaling in PWSCF
>
>
> On 4/26/06, Nichols A. Romero wrote:
> > Hi,
>
> hi.
>
> > Has anyone performed any very-large scale DFT calculations on PWSCF
> > using over 64 processors and over 1000 atoms? Does anyone know what
> > its current limits (system sizes and processors) are on parallel
> > computing environments with fast interconnects?
>
> i've recently run a comparatively large job (272 atoms, 560 electrons,
> 4x4x4 k-points) across 768 processors on a Cray XT3. i had to hack
> some constants to make it work. however the scaling is not (yet)
> so good and depending on what kind of atoms you want to put into
> your system, i.e. if it is metallic, you may be better of with gamma
> only and car-parrinello dynamics. i ran the same system as above
> with a car-parrinello code (one that is not in quantum espresso) and
> would scale out at 512 nodes for gamma point only on an IBM BG/L.
> even though i had to use a much smaller time step, i did get much
> more trajectory that way.
>
> judging from the CVS commit messages, efforts to optimize the
> quantum espresso codes for that kind of machine with large numbers
> of nodes are underway. note, that for systems as large as that,
> you might run into the scaling limitations of DFT with plane waves,
> so for a system that big using one of those 'linear scaling' DFT codes
> could be a more promising alternative.
>
> best regards,
> axel.
>
>
> >
> > Thanks,
> > --
> > Nichols A. Romero, Ph.D.
> > 1613 Denise Dr. Apt. D
> > Forest Hill, MD 21050
> > 443-567-8328 (C)
> > 410-306-0709 (O)
> > _______________________________________________
> > 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
>
From sbraccia at sissa.it Wed Apr 26 22:40:51 2006
From: sbraccia at sissa.it (carlo sbraccia)
Date: Wed, 26 Apr 2006 16:40:51 -0400
Subject: [Pw_forum] parallel scaling in PWSCF
In-Reply-To:
References: <6ac064b60604260941u15471f47oe6d45fee2ff6ce8@mail.gmail.com> <7b6913e90604261038t25d0bb43le5c8a740e793d981@mail.gmail.com> <003b01c66965$e4178d80$fcc2db80@ornl.gov>
Message-ID: <444FDAD3.4050002@sissa.it>
Hi,
beware using the CVS version of PWSCF: the parallel Davidson has not yet
been fully tested and seems to be buggy.
For this reason we are going to disable it until we shall be able to
solve the problem. To avoid the serial bottleneck in the diagonalization
you can use conjugate-gradient.
carlo
Andrea Ferretti wrote:
>Hi everybody,
>
>I am currently running a Copper surface with 140 Cu atoms + a molecule...
>the system has 1642 electrons and (due to metallicity) the calculation is
>performed for 985 bands (few kpt, like 4)...
>due to the 11 electrons for each Cu atom, I have a huge number of bands in
>a (relatively) small cell, and so a (relatively) low number of PWs respect
>to nbnd.
>taking a look at the dimension of wfc, no problem with memory in
>principle, even if, due to the weird
>dimensions of the system, non-scalable memory is quite large, around 1Gb.
>
>on a IBM Sp5 machine I observed a severe limit in the scaling passing from
>32 to 64 procs using both espresso 2.1.x and espresso 3.0...
>( anyway, I succeeded in performing a "relax" calculation for the system
>!!!! )
>
>as far as I know, this problem might be connected to a serial part in the
>diagonalization which has been parallelized in the current CVS version
>(as already pointed out by Axel)...
>At the moment I am testing this CVS version against my system, I will let
>you know the results as soon as possible...
>
>cheers
>andrea
>
>
>
>
From naromero at gmail.com Wed Apr 26 23:32:32 2006
From: naromero at gmail.com (Nichols A. Romero)
Date: Wed, 26 Apr 2006 17:32:32 -0400
Subject: [Pw_forum] parallel scaling in PWSCF
In-Reply-To: <444FDAD3.4050002@sissa.it>
References: <6ac064b60604260941u15471f47oe6d45fee2ff6ce8@mail.gmail.com>
<7b6913e90604261038t25d0bb43le5c8a740e793d981@mail.gmail.com>
<003b01c66965$e4178d80$fcc2db80@ornl.gov>
<444FDAD3.4050002@sissa.it>
Message-ID: <6ac064b60604261432y26b9a5d5x166562fb6870da19@mail.gmail.com>
Carlo,
I am using the CVS version of PWSCF which I checked out on Monday.
I did 216 carbon atoms (bulk diamond) with USPP from the PWSCF
website. Using E_c = 40 Ry and only the gamma point:
Davidson took 14 min 33 sec. CG took 22 min 46 sec. That is the CPU
time as reported by PWSCF on an SGI Altix running on 8 processors.
I tried a 1728 atom calculation using the same parameters as above and
I ran into some trouble when trying to run it in parallel on 128
processors.
PWSCF seems to seg. fault with an error *possibly* originating in a
setup.f90 routine located here
check_para_diag_efficiency_
I was using the 'Davidson' algorithm in the 1728 atom calculation. I
have *not* tried the larger calculation with 'cg.'
I hope this is helpful.
Bests,
On 4/26/06, carlo sbraccia wrote:
> Hi,
>
> beware using the CVS version of PWSCF: the parallel Davidson has not yet
> been fully tested and seems to be buggy.
> For this reason we are going to disable it until we shall be able to
> solve the problem. To avoid the serial bottleneck in the diagonalization
> you can use conjugate-gradient.
>
> carlo
>
> Andrea Ferretti wrote:
>
> >Hi everybody,
> >
> >I am currently running a Copper surface with 140 Cu atoms + a molecule...
> >the system has 1642 electrons and (due to metallicity) the calculation is
> >performed for 985 bands (few kpt, like 4)...
> >due to the 11 electrons for each Cu atom, I have a huge number of bands in
> >a (relatively) small cell, and so a (relatively) low number of PWs respect
> >to nbnd.
> >taking a look at the dimension of wfc, no problem with memory in
> >principle, even if, due to the weird
> >dimensions of the system, non-scalable memory is quite large, around 1Gb.
> >
> >on a IBM Sp5 machine I observed a severe limit in the scaling passing from
> >32 to 64 procs using both espresso 2.1.x and espresso 3.0...
> >( anyway, I succeeded in performing a "relax" calculation for the system
> >!!!! )
> >
> >as far as I know, this problem might be connected to a serial part in the
> >diagonalization which has been parallelized in the current CVS version
> >(as already pointed out by Axel)...
> >At the moment I am testing this CVS version against my system, I will let
> >you know the results as soon as possible...
> >
> >cheers
> >andrea
> >
> >
> >
> >
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
--
Nichols A. Romero, Ph.D.
1613 Denise Dr. Apt. D
Forest Hill, MD 21050
443-567-8328 (C)
410-306-0709 (O)
From lyw1983 at 163.com Thu Apr 27 14:33:22 2006
From: lyw1983 at 163.com (lyw1983)
Date: Thu, 27 Apr 2006 20:33:22 +0800 (CST)
Subject: [Pw_forum] How should I input the positions of atoms
for such material?
Message-ID: <4450BA12.000098.08570@bj163app60.163.com>
Dear all:
The structure of the Ga2o3 is Monoclinic base-centered, and with
a spacegroup of number12, so the primitive cell of it is not the
unit cell, for this material, we should input celldm(1)=a,
celldm(2)=b/a,celldm(3)=c/a and celldm(4)=cos(ab),but the angle between a and b is 90o, and the angle between a and c is 103.7o,
so we should change the axis as follow: a changes to b, b change
to c and c change to a, now the angle between a and b is 103.7o,
but when I input the positions of atoms, how should I change the
positions?
the iformation of Ga2O3 as below:a=12.23Å, b=3.04Å, c=5.80Å, ?=90o,
?=103.7o, ?=90o, the positions of atoms in primitive cell are:
Ga 0.09040 0.09040 0.79480
Ga 0.34140 0.34140 0.68570
O 0.16740 0.16740 0.10110
O 0.49570 0.49570 0.25530
O 0.82790 0.82790 0.43650
Ga 0.90960 0.90960 0.20520
Ga 0.65860 0.65860 0.31430
O 0.83260 0.83260 0.89890
O 0.50430 0.50430 0.74470
O 0.17210 0.17210 0.56350
Perhaps this question is very stupid, but it really puzzled me for a
long time, so I really need your help!
Thanks very much!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20060427/de9ba2c4/attachment.htm
From lukethulin at netscape.net Thu Apr 27 14:41:55 2006
From: lukethulin at netscape.net (Luke Thulin)
Date: Thu, 27 Apr 2006 08:41:55 -0400
Subject: [Pw_forum] Can't run pwgui and missing chden.x
Message-ID: <4450BC13.7010402@netscape.net>
Hello,
I'm just trying Quantum Espresso for the first time. I've installed it
and successfully run the first few examples. However, I see a couple
problems. If I try to run pwgui I get the following error:
Please define the PWGUI environment variable !!!
PWGUI should point to the PWgui package root directory.
Also, when I looked at the packages that should be available after doing
the make all, chdens.x is not in my espresso-3.0/bin folder.
Thanks,
Luke Thulin
From Giovanni.Cantele at na.infn.it Thu Apr 27 14:53:06 2006
From: Giovanni.Cantele at na.infn.it (Giovanni Cantele)
Date: Thu, 27 Apr 2006 14:53:06 +0200
Subject: [Pw_forum] Can't run pwgui and missing chden.x
In-Reply-To: <4450BC13.7010402@netscape.net>
References: <4450BC13.7010402@netscape.net>
Message-ID: <4450BEB2.4010508@na.infn.it>
>
>
> Please define the PWGUI environment variable !!!
> PWGUI should point to the PWgui package root directory.
You should define the PWGUI environment variable!!
Depending on which shell you are using:
setenv PWGUI something/espresso-3.0/PWgui-3.0
export PWGUI=something/espresso-3.0/PWgui-3.0
>
> Also, when I looked at the packages that should be available after
> doing the make all, chdens.x is not in my espresso-3.0/bin folder.
>
As far as I know, chdens.x has been merged into pp.x. From Changelog in
the CVS version:
2005-09-13 22:30 giannozz
* Doc/INPUT_CHDENS, Doc/INPUT_PP, PP/Makefile, PP/chdens.f90,
PP/postproc.f90, PW/clean_pw.f90, examples/example05/README,
examples/example05/run_example, examples/example16/README,
examples/example16/run_example:
chdens.x merged into pp.x - all functionalities are still there
and it is still possible to do the two steps independently.
The output is basically the sum of the two outputs with minor
differences. Documentation and examples updated, GUI not yet.
You find how to use the "new" pp.x for both post-processing and plotting
(old chdens.x) in espresso3.0/Doc/INPUT_PP
Hope this helps,
Giovanni
--
Dr. Giovanni Cantele
Coherentia CNR-INFM and Dipartimento di Scienze Fisiche
Universita' di Napoli "Federico II"
Complesso Universitario di Monte S. Angelo - Ed. G
Via Cintia, I-80126, Napoli, Italy
Phone: +39 081 676910
Fax: +39 081 676346
E-mail: Giovanni.Cantele at na.infn.it
Web: http://people.na.infn.it/~cantele
From lukethulin at netscape.net Thu Apr 27 15:52:04 2006
From: lukethulin at netscape.net (Luke Thulin)
Date: Thu, 27 Apr 2006 09:52:04 -0400
Subject: [Pw_forum] Can't run pwgui and missing chden.x
In-Reply-To: <4450BEB2.4010508@na.infn.it>
References: <4450BC13.7010402@netscape.net> <4450BEB2.4010508@na.infn.it>
Message-ID: <4450CC84.5040309@netscape.net>
Thanks,
I was confused because I was expecting to see the chdens.x. Also, pwgui
is working after the export command.
Luke
Giovanni.Cantele at na.infn.it wrote:
>
>>
>>
>> Please define the PWGUI environment variable !!!
>> PWGUI should point to the PWgui package root directory.
>
>
> You should define the PWGUI environment variable!!
> Depending on which shell you are using:
> setenv PWGUI something/espresso-3.0/PWgui-3.0
> export PWGUI=something/espresso-3.0/PWgui-3.0
>
>>
>> Also, when I looked at the packages that should be available after
>> doing the make all, chdens.x is not in my espresso-3.0/bin folder.
>>
> As far as I know, chdens.x has been merged into pp.x. From Changelog
> in the CVS version:
> 2005-09-13 22:30 giannozz
>
> * Doc/INPUT_CHDENS, Doc/INPUT_PP, PP/Makefile, PP/chdens.f90,
> PP/postproc.f90, PW/clean_pw.f90, examples/example05/README,
> examples/example05/run_example, examples/example16/README,
> examples/example16/run_example:
>
> chdens.x merged into pp.x - all functionalities are still there
> and it is still possible to do the two steps independently.
> The output is basically the sum of the two outputs with minor
> differences. Documentation and examples updated, GUI not yet.
>
> You find how to use the "new" pp.x for both post-processing and
> plotting (old chdens.x) in espresso3.0/Doc/INPUT_PP
>
> Hope this helps,
> Giovanni
>
From lukethulin at netscape.net Thu Apr 27 16:01:03 2006
From: lukethulin at netscape.net (Luke Thulin)
Date: Thu, 27 Apr 2006 10:01:03 -0400
Subject: [Pw_forum] Can't run pwgui and missing chden.x
In-Reply-To: <4450CC84.5040309@netscape.net>
References: <4450BC13.7010402@netscape.net> <4450BEB2.4010508@na.infn.it> <4450CC84.5040309@netscape.net>
Message-ID: <4450CE9F.1030607@netscape.net>
I take that back, PWgui 3.0 isn't working. If I go to the website and
get version 2.1.3, that works, but 3.0 aborts and because the wish
command is not found.
lukethulin at netscape.net wrote:
> Thanks,
>
> I was confused because I was expecting to see the chdens.x. Also,
> pwgui is working after the export command.
> Luke
>
> Giovanni.Cantele at na.infn.it wrote:
>
>>
>>>
>>>
>>> Please define the PWGUI environment variable !!!
>>> PWGUI should point to the PWgui package root directory.
>>
>>
>>
>> You should define the PWGUI environment variable!!
>> Depending on which shell you are using:
>> setenv PWGUI something/espresso-3.0/PWgui-3.0
>> export PWGUI=something/espresso-3.0/PWgui-3.0
>>
>>>
>>> Also, when I looked at the packages that should be available after
>>> doing the make all, chdens.x is not in my espresso-3.0/bin folder.
>>>
>> As far as I know, chdens.x has been merged into pp.x. From Changelog
>> in the CVS version:
>> 2005-09-13 22:30 giannozz
>>
>> * Doc/INPUT_CHDENS, Doc/INPUT_PP, PP/Makefile, PP/chdens.f90,
>> PP/postproc.f90, PW/clean_pw.f90, examples/example05/README,
>> examples/example05/run_example, examples/example16/README,
>> examples/example16/run_example:
>>
>> chdens.x merged into pp.x - all functionalities are still there
>> and it is still possible to do the two steps independently.
>> The output is basically the sum of the two outputs with minor
>> differences. Documentation and examples updated, GUI not yet.
>>
>> You find how to use the "new" pp.x for both post-processing and
>> plotting (old chdens.x) in espresso3.0/Doc/INPUT_PP
>>
>> Hope this helps,
>> Giovanni
>>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
From giannozz at nest.sns.it Thu Apr 27 17:20:50 2006
From: giannozz at nest.sns.it (Paolo Giannozzi)
Date: Thu, 27 Apr 2006 17:20:50 +0200
Subject: [Pw_forum] Can't run pwgui and missing chden.x
In-Reply-To: <4450CE9F.1030607@netscape.net>
References: <4450BC13.7010402@netscape.net> <4450CC84.5040309@netscape.net> <4450CE9F.1030607@netscape.net>
Message-ID: <200604271720.50195.giannozz@nest.sns.it>
On Thursday 27 April 2006 16:01, Luke Thulin wrote:
> I take that back, PWgui 3.0 isn't working. If I go to the website and
> get version 2.1.3, that works, but 3.0 aborts and because the wish
> command is not found.
if I remember correctly version 3.0 of the GUI is available only as
source code and requires some external libraries to work. For version
2.1.3 stand-alone binaries are available that do not require anything.
--
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 tone.kokalj at ijs.si Thu Apr 27 17:33:13 2006
From: tone.kokalj at ijs.si (Tone Kokalj)
Date: Thu, 27 Apr 2006 17:33:13 +0200
Subject: [Pw_forum] Can't run pwgui and missing chden.x
In-Reply-To: <4450CE9F.1030607@netscape.net>
References: <4450BC13.7010402@netscape.net> <4450BEB2.4010508@na.infn.it>
<4450CC84.5040309@netscape.net> <4450CE9F.1030607@netscape.net>
Message-ID: <1146151993.6568.29.camel@localhost.localdomain>
On Thu, 2006-04-27 at 10:01 -0400, Luke Thulin wrote:
> I take that back, PWgui 3.0 isn't working. If I go to the website and
> get version 2.1.3, that works, but 3.0 aborts and because the wish
> command is not found.
You need the Tcl/Tk software to run the PWgui's source version (see the
README file in the top PWgui's directory).
For the version 2.1.3 you very likely using the binary version (that's
not yet available for 3.0, my fault).
Regards, Tone
From tone.kokalj at ijs.si Fri Apr 28 11:43:28 2006
From: tone.kokalj at ijs.si (Tone Kokalj)
Date: Fri, 28 Apr 2006 11:43:28 +0200
Subject: [Pw_forum] Can't run pwgui and missing chden.x
In-Reply-To: <1146151993.6568.29.camel@localhost.localdomain>
References: <4450BC13.7010402@netscape.net> <4450BEB2.4010508@na.infn.it>
<4450CC84.5040309@netscape.net> <4450CE9F.1030607@netscape.net>
<1146151993.6568.29.camel@localhost.localdomain>
Message-ID: <1146217408.6098.21.camel@localhost.localdomain>
The binaries for PWgui-3.0 have just been created. See:
http://www-k3.ijs.si/kokalj/pwgui/
Regards, Tone
From lukethulin at netscape.net Fri Apr 28 15:24:18 2006
From: lukethulin at netscape.net (Luke Thulin)
Date: Fri, 28 Apr 2006 09:24:18 -0400
Subject: [Pw_forum] Can't run pwgui and missing chden.x
In-Reply-To: <1146217408.6098.21.camel@localhost.localdomain>
References: <4450BC13.7010402@netscape.net> <4450BEB2.4010508@na.infn.it> <4450CC84.5040309@netscape.net> <4450CE9F.1030607@netscape.net> <1146151993.6568.29.camel@localhost.localdomain> <1146217408.6098.21.camel@localhost.localdomain>
Message-ID: <44521782.7090607@netscape.net>
I downloaded the new version and I think it's working. However, when I
execute the pwgui command from my shell, it says it's starting up PWgui
version 2.1.3. I don't see anything in the actual program that tells me
which version I'm running. So is there any way to be sure that I'm
actually running 3.0? I don't see how it could still be the 2.1.3, but
just to be sure.
Thanks,
Luke
tone.kokalj at ijs.si wrote:
>The binaries for PWgui-3.0 have just been created. See:
>http://www-k3.ijs.si/kokalj/pwgui/
>
>Regards, Tone
>
>
>_______________________________________________
>Pw_forum mailing list
>Pw_forum at pwscf.org
>http://www.democritos.it/mailman/listinfo/pw_forum
>
>
From tone.kokalj at ijs.si Fri Apr 28 15:30:13 2006
From: tone.kokalj at ijs.si (Tone Kokalj)
Date: Fri, 28 Apr 2006 15:30:13 +0200
Subject: [Pw_forum] Can't run pwgui and missing chden.x
In-Reply-To: <44521782.7090607@netscape.net>
References: <4450BC13.7010402@netscape.net> <4450BEB2.4010508@na.infn.it>
<4450CC84.5040309@netscape.net> <4450CE9F.1030607@netscape.net>
<1146151993.6568.29.camel@localhost.localdomain>
<1146217408.6098.21.camel@localhost.localdomain>
<44521782.7090607@netscape.net>
Message-ID: <1146231013.6098.28.camel@localhost.localdomain>
On Fri, 2006-04-28 at 09:24 -0400, Luke Thulin wrote:
> I downloaded the new version and I think it's working.
When you launch it, it should say:
==================================================
This is PWgui version: 3.0
--------------------------------------------------
Moreover, the help documents (menu Help) also reveal the version
number. It should be 3.0. If it is not, then probably you are not
running the right program?
I just downloaded from the WEB page, and it is really the 3.0 !
Ciao, Tone
From lukethulin at netscape.net Fri Apr 28 15:43:37 2006
From: lukethulin at netscape.net (Luke Thulin)
Date: Fri, 28 Apr 2006 09:43:37 -0400
Subject: [Pw_forum] Can't run pwgui and missing chden.x
In-Reply-To: <1146231013.6098.28.camel@localhost.localdomain>
References: <4450BC13.7010402@netscape.net> <4450BEB2.4010508@na.infn.it> <4450CC84.5040309@netscape.net> <4450CE9F.1030607@netscape.net> <1146151993.6568.29.camel@localhost.localdomain> <1146217408.6098.21.camel@localhost.localdomain> <44521782.7090607@netscape.net> <1146231013.6098.28.camel@localhost.localdomain>
Message-ID: <44521C09.70106@netscape.net>
The .zip version that I just downloaded now says The is PWgui version: 3.0.
The tar-gzip version that I downloaded initially was the one that said
2.1.3.
tone.kokalj at ijs.si wrote:
>On Fri, 2006-04-28 at 09:24 -0400, Luke Thulin wrote:
>
>
>>I downloaded the new version and I think it's working.
>>
>>
>
>When you launch it, it should say:
> ==================================================
> This is PWgui version: 3.0
> --------------------------------------------------
>
>Moreover, the help documents (menu Help) also reveal the version
>number. It should be 3.0. If it is not, then probably you are not
>running the right program?
>
>I just downloaded from the WEB page, and it is really the 3.0 !
>
>Ciao, Tone
>
>_______________________________________________
>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/20060428/9330cb5c/attachment.htm
From stewart at cnf.cornell.edu Fri Apr 28 20:17:04 2006
From: stewart at cnf.cornell.edu (stewart at cnf.cornell.edu)
Date: Fri, 28 Apr 2006 14:17:04 -0400
Subject: [Pw_forum] Re: Phonon frequencies different for equivalent q points
In-Reply-To: <200604261804.14603.giannozz@nest.sns.it>
References: <443BD1B3.6090200@ictp.it>
<20060411184542.11682.qmail@xuxa.iecc.com>
<200604261804.14603.giannozz@nest.sns.it>
Message-ID: <20060428181704.24228.qmail@xuxa.iecc.com>
Hi Paolo,
Thanks for the tip on the cartesian coordinates. I had assumed the
coordinates were in terms of reciprocal lattice vectors. The results for M
along the two directions now match.
Best regards,
Derek
Paolo Giannozzi writes:
> On Tuesday 11 April 2006 20:45, stewart at cnf.cornell.edu wrote:
>
>> I am looking at the high symmetry lines, Gamma-K-M along the [x,x,0]
>> direction going from [0,0,0] to [1,1,0]*(2pi/a) and the Gamma-M line
>> along the [x,0,0] going from [0,0,0] to [1/sqrt(3),0,0]*(2pi/a).
>
> are you sure these are the correct high-symmetry lines? please note
> that phonon wavevectors must be supplied in cartesian coordinates
>
> 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
> _______________________________________________
> 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 emenendez at macul.ciencias.uchile.cl Fri Apr 28 23:25:57 2006
From: emenendez at macul.ciencias.uchile.cl (Eduardo Ariel Menendez P)
Date: Fri, 28 Apr 2006 17:25:57 -0400 (CLT)
Subject: [Pw_forum] nodes halt problem
Message-ID:
Dear friends
I apologize for a question of sys admin and not of espresso. However, this
is important for me to be able to has either espresso or siesta.
I would like to know if someone has had problem with this hardware
CPU: Xeom 3.06 GHz EMT64 (configures as 32 bits)
Chipset Intel E7320
Network adapter Intel 825401GI integrated in the mother board.
SATA hard disk in each node.
SATA controller 6300 ESB
My Linux is 2.4-27-2-686-smp Debian 1:3.3.6-10
When running large parallel jobs, some nodes halt without any message.
We have made tests to determine if due it is hardware or software, but
we are not sure yet.
The system halts happen with at least two codes and with several versions
of LAM-MPI and MPICH, so it is no the problem.
One job using two CPU in the same node runs up to the end.
The same job using 2CPU in two nodes aborts due to one node halt.
The moment of halt seems random, also the node.
Nodes are connected via a Gbit switch. This make the switch and the
network cards suspects.
One job using 2 nodes and 4 CPU, linked using crossover cables, not the
switch, also abort due to halt. Then it seems that the switch is not
guilty.
Then it seems a problem of the kernel, or the network
controller, or the network adapter. How can I know who is guilty?
Is the problem is software, is there a solution other than install other
operative system? Can I discard software problems?
Thanks
Eduardo
From ezadshojaee at hotmail.com Sat Apr 29 10:38:39 2006
From: ezadshojaee at hotmail.com (Ezad Shojaee)
Date: Sat, 29 Apr 2006 08:38:39 +0000
Subject: [Pw_forum] symmetry problem in vc-relax
Message-ID:
hi
i am doing vc-relax for TiO2 using symmetry-conserving algorithm (
cell_dynamics='damp-w' ) but still i have this message:
from checkallsym : error # 5
not orthogonal operation
does anyone have any suggestion ?
this is the input file for pw.x i am using :
&CONTROL
calculation = "vc-relax",
restart_mode='from_scratch',
prefix = "TiO2",
tstress = .true.
tprnfor = .true.
dt =1,
pseudo_dir = "/root/pseudo",
outdir = "/root/tmp",
/
&SYSTEM
ibrav = 7,
celldm(1) = 3.785,
celldm(3) = 2.5136,
nat = 6,
ntyp = 2,
ecutwfc =40.0,
/
&ELECTRONS
mixing_mode = 'plain'
conv_thr = 1.D-9,
mixing_beta = 0.7,
/
&IONS
ion_dynamics ='damp',
pot_extrapolation='second-order',
wfc_extrapolation='second-order',
/
&CELL
cell_dynamics='damp-w'
/
ATOMIC_SPECIES
Ti 1.00 Ti.vdb.UPF
O 1.00 O.vdb.UPF
ATOMIC_POSITIONS { bohr }
Ti 0.00000000 -0.94625000 -1.18925000
Ti 0.00000000 0.94625000 1.18925000
O 0.00000000 0.94625000 -0.77634000
O 0.00000000 -0.94625000 0.77634000
O 1.89250000 0.94625000 1.60216000
O -1.89250000 -0.94625000 -1.60216000
K_POINTS { automatic }
1 1 1 0 0 0
_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today it's FREE!
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/