From srijan.india at gmail.com Wed Nov 1 06:26:56 2006 From: srijan.india at gmail.com (Srijan Kumar Saha) Date: Wed, 1 Nov 2006 10:56:56 +0530 Subject: [Pw_forum] Non cubic structures In-Reply-To: <45462078.1040804@mit.edu> References: <4545CB63.5010409@lg.ehu.es> <3fcc7e670610300445l2a29ebecvc9d414627bc6599e@mail.gmail.com> <4545F5F3.8040208@mit.edu> <3fcc7e670610300547v2497f153s1ee99068bd8c890c@mail.gmail.com> <45462078.1040804@mit.edu> Message-ID: <3fcc7e670610312126n77ada714xfdb7ea88f8534c45@mail.gmail.com> Dear Users, I thought I would try to keep a high informational content for the postings. But I'm seeing a minor mismatch in the directory Example No.3. I don't know whether this forum is the correct place to point out this type of error. But I think this type of error may confuse other students (Specially them who enjoy to understand every details instead of just runnnig ./run_example03 blindly). Here is the error :::::: It's a part of README file of EXAMPLE No. 03. This example illustrates how to use pw.x to compute the equilibrium geometry of a simple molecule, CO, and of an Al (001) slab. The calculation proceeds as follows (for the meaning of the cited input variables see the appropriate INPUT_* file) 1) make a geometry relaxation for CO molecule performing a series of self-consistent calculations and computing the forces on atoms (input=co.rx.in, output=co.rx.out). The molecule is put in a cubic box of side 12 Bohr. Note that ibrav=0 therefore the Bravais lattice fundamental vectors are read after cards 'CELL_PARAMETERS' (where we also specify the type of symmetry, cubic or hexagonal). The cell parameter is not specified in celldm(1), but deduced from Bravais lattice vectors. But the input file (run_example) is like &SYSTEM ibrav = 1, celldm(1) = 12.D0, Here ibrav is not 0. If you think this error is not so important please ignore my mail. With best regards, Srijan -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061101/bfea6df5/attachment.htm From rudrabnrj at gmail.com Wed Nov 1 07:37:11 2006 From: rudrabnrj at gmail.com (rudra banerjee) Date: Wed, 1 Nov 2006 12:07:11 +0530 Subject: [Pw_forum] error list Message-ID: <2e36f8e50610312237v61d3b2d8p85ea93561305dd07@mail.gmail.com> hello friends, can i check anyway what exactly the error list means? to be more precious, if iam gettin a command like %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from read_namelists : error # 19 reading namelist system %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... 2 where should i look to see where the fault is??? -- rudra have a good time! -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061101/b86bf150/attachment.htm From ding at sissa.it Wed Nov 1 08:31:07 2006 From: ding at sissa.it (Xunlei Ding) Date: Wed, 01 Nov 2006 08:31:07 +0100 Subject: [Pw_forum] error list In-Reply-To: <2e36f8e50610312237v61d3b2d8p85ea93561305dd07@mail.gmail.com> References: <2e36f8e50610312237v61d3b2d8p85ea93561305dd07@mail.gmail.com> Message-ID: <45484D3B.3090907@sissa.it> Dear Rudra, It means there is an error when "reading namelist system". So you should check the "&system" section in you input file. Hope this be helpful. Yours, Ding rudra banerjee wrote: > hello friends, > can i check anyway what exactly the error list means? to be more > precious, if iam gettin a command like > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > from read_namelists : error # 19 > reading namelist system > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > stopping ... > 2 > > > where should i look to see where the fault is??? > > -- > rudra > > have a good time! From degironc at sissa.it Wed Nov 1 17:35:13 2006 From: degironc at sissa.it (degironc) Date: Wed, 01 Nov 2006 17:35:13 +0100 Subject: [Pw_forum] error list In-Reply-To: <45484D3B.3090907@sissa.it> References: <2e36f8e50610312237v61d3b2d8p85ea93561305dd07@mail.gmail.com> <45484D3B.3090907@sissa.it> Message-ID: <4548CCC1.3000609@sissa.it> The QE distribution has a graphic user interface that can be used to create well formatted input files or check already created input files.. Now I don't remember exactly (I never use it) but you have the possibility to open an existing input file with the GUI and then to save it and at that point the GUI makes a number of checks and spot mispelled / misplaced / non-existent variables ... which are the most common errors in input files. hope this helps, stefano Xunlei Ding wrote: > Dear Rudra, > > It means there is an error when "reading namelist system". > So you should check the "&system" section in you input file. > Hope this be helpful. > > Yours, > Ding > > rudra banerjee wrote: > >> hello friends, >> can i check anyway what exactly the error list means? to be more >> precious, if iam gettin a command like >> >> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% >> >> from read_namelists : error # 19 >> reading namelist system >> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% >> >> >> stopping ... >> 2 >> >> >> where should i look to see where the fault is??? >> >> -- >> rudra >> >> have a good time! > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From degironc at sissa.it Wed Nov 1 19:49:11 2006 From: degironc at sissa.it (degironc) Date: Wed, 01 Nov 2006 19:49:11 +0100 Subject: [Pw_forum] Non cubic structures In-Reply-To: <3fcc7e670610312126n77ada714xfdb7ea88f8534c45@mail.gmail.com> References: <4545CB63.5010409@lg.ehu.es> <3fcc7e670610300445l2a29ebecvc9d414627bc6599e@mail.gmail.com> <4545F5F3.8040208@mit.edu> <3fcc7e670610300547v2497f153s1ee99068bd8c890c@mail.gmail.com> <45462078.1040804@mit.edu> <3fcc7e670610312126n77ada714xfdb7ea88f8534c45@mail.gmail.com> Message-ID: <4548EC27.1030603@sissa.it> Dear Srijan, thank you for pointing out the inconsistency between the README file and the actual script in example03, it is fixed now in the cvs version and will propagate in future releases of the package. A problem with documentation is that it is usually outdated unless constantly re-checked. thanks, Stefano de Gironcoli Srijan Kumar Saha wrote: > > Dear Users, > > I thought I would try to keep a high informational content > for the postings. But I'm seeing a minor mismatch in the directory > Example No.3. > I don't know whether this forum is the correct place to point out > this type of error. > But I think this type of error may confuse other students > (Specially them who enjoy > to understand every details instead of just runnnig > ./run_example03 blindly). > > Here is the error :::::: > > It's a part of README file of EXAMPLE No. 03. > > This example illustrates how to use pw.x to compute the equilibrium > geometry of a simple molecule, CO, and of an Al (001) slab. > > The calculation proceeds as follows (for the meaning of the cited input > variables see the appropriate INPUT_* file) > > 1) make a geometry relaxation for CO molecule performing a series of > self-consistent calculations and computing the forces on atoms > (input=co.rx.in , output=co.rx.out). > The molecule is put in a cubic box of side 12 Bohr. > Note that ibrav=0 therefore the Bravais lattice fundamental vectors > are read after cards 'CELL_PARAMETERS' (where we also specify the > type of symmetry, cubic or hexagonal). The cell parameter is not > specified in celldm(1), but deduced from Bravais lattice vectors. > > > But the input file (run_example) is like > > &SYSTEM > ibrav = 1, > celldm(1) = 12.D0, > > Here ibrav is not 0. > > If you think this error is not so important please ignore my mail. > > With best regards, > Srijan > From rudrabnrj at gmail.com Fri Nov 3 12:37:27 2006 From: rudrabnrj at gmail.com (rudra banerjee) Date: Fri, 3 Nov 2006 17:07:27 +0530 Subject: [Pw_forum] compilation issue Message-ID: <2e36f8e50611030337k1c81cb9dgd9335a9e9144478b@mail.gmail.com> dear friends, being new, i am in trouble again! i have to shoot pwscf on a simd machine. the point is can i check if pwscf is statically compiled or not???which file should i look at??? lookin for your help thanking you -- rudra have a good time! -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061103/f2a52605/attachment.htm From lanhaiping at gmail.com Fri Nov 3 13:30:19 2006 From: lanhaiping at gmail.com (lan haiping) Date: Fri, 3 Nov 2006 20:30:19 +0800 Subject: [Pw_forum] compilation issue In-Reply-To: <2e36f8e50611030337k1c81cb9dgd9335a9e9144478b@mail.gmail.com> References: <2e36f8e50611030337k1c81cb9dgd9335a9e9144478b@mail.gmail.com> Message-ID: hi, you could find some run.sh script in $espresso/Examples. if quantum-espresso is properly compiled, you can succesfully finish examples there with run.sh scripts hope help. regards, hai-ping On 11/3/06, rudra banerjee wrote: > > dear friends, > being new, i am in trouble again! i have to shoot pwscf on a simd > machine. the point is can i check if pwscf is statically compiled or > not???which file should i look at??? > lookin for your help > thanking you > > -- > rudra > > have a good time! -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061103/5bc677b6/attachment.htm From ceresoli at sissa.it Fri Nov 3 13:47:21 2006 From: ceresoli at sissa.it (Davide Ceresoli) Date: Fri, 03 Nov 2006 13:47:21 +0100 Subject: [Pw_forum] compilation issue In-Reply-To: <2e36f8e50611030337k1c81cb9dgd9335a9e9144478b@mail.gmail.com> References: <2e36f8e50611030337k1c81cb9dgd9335a9e9144478b@mail.gmail.com> Message-ID: <454B3A59.7030000@sissa.it> rudra banerjee wrote: > dear friends, > being new, i am in trouble again! i have to shoot pwscf on a simd > machine. the point is can i check if pwscf is statically compiled or > not???which file should i look at??? Try the ldd command: # ldd pw.x not a dynamic executable -- +----------------------------------------------------------+ Davide Ceresoli Scuola Internazionale Superiore di Studi Avanzati (SISSA) via Beirut 2-4 I-34014 Trieste, Italy Phone: +39-040-3787-448 Fax: +39-040-3787-528 Mobile: +39-347-1001570 Homepage: http://www.sissa.it/~ceresoli +----------------------------------------------------------+ From rudrabnrj at gmail.com Fri Nov 3 16:54:26 2006 From: rudrabnrj at gmail.com (rudra banerjee) Date: Fri, 3 Nov 2006 21:24:26 +0530 Subject: [Pw_forum] compilation issue In-Reply-To: <454B3A59.7030000@sissa.it> References: <2e36f8e50611030337k1c81cb9dgd9335a9e9144478b@mail.gmail.com> <454B3A59.7030000@sissa.it> Message-ID: <2e36f8e50611030754g5d261983j90d05ddd1715954@mail.gmail.com> thanks!!! but the problem is even deeper! i need to compile it statically. can there is possibly any change that i can make in the make file to compile it statically??? rudra On 11/3/06, Davide Ceresoli wrote: > > rudra banerjee wrote: > > dear friends, > > being new, i am in trouble again! i have to shoot pwscf on a simd > > machine. the point is can i check if pwscf is statically compiled or > > not???which file should i look at??? > Try the ldd command: > # ldd pw.x > not a dynamic executable > > > -- > +----------------------------------------------------------+ > Davide Ceresoli > Scuola Internazionale Superiore di Studi Avanzati (SISSA) > via Beirut 2-4 > I-34014 Trieste, Italy > Phone: +39-040-3787-448 Fax: +39-040-3787-528 > Mobile: +39-347-1001570 > Homepage: http://www.sissa.it/~ceresoli > +----------------------------------------------------------+ > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- rudra have a good time! -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061103/26d1f2f8/attachment.htm From wmbmacam at lg.ehu.es Fri Nov 3 16:56:47 2006 From: wmbmacam at lg.ehu.es (=?ISO-8859-1?Q?Miguel_Mart=EDnez_Canales?=) Date: Fri, 03 Nov 2006 16:56:47 +0100 Subject: [Pw_forum] compilation issue In-Reply-To: <2e36f8e50611030754g5d261983j90d05ddd1715954@mail.gmail.com> References: <2e36f8e50611030337k1c81cb9dgd9335a9e9144478b@mail.gmail.com> <454B3A59.7030000@sissa.it> <2e36f8e50611030754g5d261983j90d05ddd1715954@mail.gmail.com> Message-ID: <454B66BF.4000703@lg.ehu.es> rudra banerjee wrote: > thanks!!! > but the problem is even deeper! > i need to compile it statically. can there is possibly any change that > i can make in the make file to compile it statically??? > > rudra You can parse the configure option that is explained in the README, --disable-shared -- ---------------------------------------- Miguel Mart?nez Canales Dto. F?sica de la Materia Condensada UPV/EHU Facultad de Ciencia y Tecnolog?a Apdo. 644 48080 Bilbao (Spain) Fax: +34 94 601 3500 Tlf: +34 94 601 5437 ---------------------------------------- "If you have an apple and I have an apple and we exchange these apples then you and I will still each have one apple. But if you have an idea and I have an idea and we exchange these ideas, then each of us will have two ideas." George Bernard Shaw From lanhaiping at gmail.com Fri Nov 3 18:05:55 2006 From: lanhaiping at gmail.com (lan haiping) Date: Sat, 4 Nov 2006 01:05:55 +0800 Subject: [Pw_forum] compilation issue In-Reply-To: <2e36f8e50611030754g5d261983j90d05ddd1715954@mail.gmail.com> References: <2e36f8e50611030337k1c81cb9dgd9335a9e9144478b@mail.gmail.com> <454B3A59.7030000@sissa.it> <2e36f8e50611030754g5d261983j90d05ddd1715954@mail.gmail.com> Message-ID: Do you mean link libs required statically ? you should remove dynamical link in make.sys found by configure scripts, and define proper paths for statical libs. hope help.. regards, hai-ping On 11/3/06, rudra banerjee wrote: > thanks!!! > but the problem is even deeper! > i need to compile it statically. can there is possibly any change that i > can make in the make file to compile it statically??? > > rudra > > On 11/3/06, Davide Ceresoli wrote: > > > > rudra banerjee wrote: > > > dear friends, > > > being new, i am in trouble again! i have to shoot pwscf on a simd > > > machine. the point is can i check if pwscf is statically compiled or > > > not???which file should i look at??? > > Try the ldd command: > > # ldd pw.x > > not a dynamic executable > > > > > > -- > > +----------------------------------------------------------+ > > Davide Ceresoli > > Scuola Internazionale Superiore di Studi Avanzati (SISSA) > > via Beirut 2-4 > > I-34014 Trieste, Italy > > Phone: +39-040-3787-448 Fax: +39-040-3787-528 > > Mobile: +39-347-1001570 > > Homepage: http://www.sissa.it/~ceresoli > > +----------------------------------------------------------+ > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > -- > rudra > > have a good time! > -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061104/b23672a9/attachment.htm From Andrea.Marini at roma2.infn.it Fri Nov 3 18:38:53 2006 From: Andrea.Marini at roma2.infn.it (Andrea Marini) Date: Fri, 3 Nov 2006 18:38:53 +0100 (CET) Subject: [Pw_forum] Electron Phonon interaction: simple-minded questions Message-ID: Dear Forum users, I am trying to calculate the el-ph matrix elements of bulk Silicon for many bands (let's say 200). The aim is to perform finite temperature self-energy calculations. I have some doubts about my runs, surely due to something I do not fully understand. 1. If I do the phonon calculation using a 2x2x2 MP grid PW performs a band structure calculation (for each q-point) on the k-points obtained reducing by symmetry the 2x2x2 grid + the points linked by a translation of q. For example at q = ( 0.0000 -1.0000 0.0000 ) cart. coord. in units 2pi/a_0 k( 1) = ( 0.0000000 0.0000000 0.0000000), wk = 0.2500000 k( 2) = ( 0.0000000 -1.0000000 0.0000000), wk = 0.0000000 k( 3) = ( 0.5000000 -0.5000000 0.5000000), wk = 1.000000 k( 4) = ( 0.5000000 -1.5000000 0.5000000), wk = 0.0000000 k( 5) = ( 0.0000000 -1.0000000 0.0000000), wk = 0.2500000 k( 6) = ( 0.0000000 -2.0000000 0.0000000), wk = 0.0000000 k( 7) = ( -1.0000000 0.0000000 0.0000000), wk = 0.5000000 k( 8) = ( -1.0000000 -1.0000000 0.0000000), wk = 0.0000000 My question is: as I am using a MP grid the shifted points (the ones with zero weight) are still points of the 2x2x2 grid (like k(2) and k(5)) . Why does PW calculates the states for these points twice ? Is there any way to avoid it ? 2. Why does PW performs a band structure calculation for every q-point of a subset of the 2x2x2 grid. Can I calculate the states of the whole grid once and use it for each q-point phonon calculation ? 3. It seems that my el-ph matrix elements are "strange" in the high energy regime, so that I am not able to converge the phononic self-energy as a function of the bands used. The point is that, while I get marvelous Eliashberg functions for a metal like Al when I look at the el-ph matrix elements between the fermi surface and high energy bands I'd expect them to go to zero as a function of the upper band energy, while it is not. Does any of you have experience of the high energy behavior of the el-ph matrix elements ? 4. When I try to calculate the matrix elements for many states I get this error message already in the sc run %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% task # 1 from cgramg1_k : error # 1 negative or zero norm in S %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% This error disappears when I increase the energy cutoff. The input file is &control calculation = 'scf',prefix='Si', pseudo_dir = '/home/marini/Sources/espresso/espresso-3.1//pseudo', wf_collect=.true. / &system ibrav= 2, celldm(1) =10.183, nat= 2, ntyp= 1,ecutwfc = 15, nbnd = 100 la2f =.true. / &electrons diago_full_acc=.TRUE. diagonalization='david' conv_thr = 1.0d-13 diago_cg_maxiter = 40 / ATOMIC_SPECIES Si 28.086 Si.vbc.UPF ATOMIC_POSITIONS Si 0.00 0.00 0.00 Si 0.25 0.25 0.25 K_POINTS {automatic} 2 2 2 0 0 0 THANKS FOR ANY HELP Andrea -------------------------------------------------------------------------- Andrea MARINI Physics Department, University of Rome "Tor Vergata" (Italy) - phone: +39-0672594894 - fax: +39-062023507 - -- andrea.marini at roma2.infn.it - http://www.fisica.uniroma2.it/~marini/ -- From s305igun at mail.chem.itb.ac.id Fri Nov 3 19:37:40 2006 From: s305igun at mail.chem.itb.ac.id (Rahmat Gunawan) Date: Sat, 4 Nov 2006 01:37:40 +0700 (WIT) Subject: [Pw_forum] Thanks and I need your help Message-ID: <32940.167.205.72.7.1162579060.squirrel@webmail.chem.itb.ac.id> Dear PWscf Forum... Halo, my name is Rahmat Gunawan, I am a first year of Postgraduate Student of Chemistry Dept. Institut Teknologi Bandung, Indonesia. My disertation research use Quantum-Espresso and XCrySDen Packet. I am beginner in this packet, but I always have problem in "input PWscf". My research area is adsorption diatomic gas molecules on graphite. The diatomic gas molecules likes: H2, F2, N2, O2 and Cl2 too. Would You like help me how entry input graphite cell in PWscf Packet? and I have another problem that in my Quantum-Espresso don't have chdens.x in /PP, so my file in PWgui can not run my input.pw Thanks very much for PW_Forum. Success for all Quantum-Espresso user... Sincerely Yours Rahmat Gunawan From akohlmey at cmm.chem.upenn.edu Fri Nov 3 21:42:50 2006 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Fri, 3 Nov 2006 15:42:50 -0500 (EST) Subject: [Pw_forum] Thanks and I need your help In-Reply-To: <32940.167.205.72.7.1162579060.squirrel@webmail.chem.itb.ac.id> Message-ID: On Sat, 4 Nov 2006, Rahmat Gunawan wrote: RG> Dear PWscf Forum... RG> RG> Halo, my name is Rahmat Gunawan, I am a first year of Postgraduate Student RG> of Chemistry Dept. Institut Teknologi Bandung, Indonesia. dear rahmat, RG> My disertation research use Quantum-Espresso and XCrySDen Packet. I am RG> beginner in this packet, but I always have problem in "input PWscf". My RG> research area is adsorption diatomic gas molecules on graphite. The RG> diatomic gas molecules likes: H2, F2, N2, O2 and Cl2 too. first i would suggest you do some survey of the literature on how accurate DFT can describe those interactions. you may be surprised unless your project is to work on improving functionals... RG> Would You like help me how entry input graphite cell in PWscf Packet? and RG> I have another problem that in my Quantum-Espresso don't have chdens.x in RG> /PP, so my file in PWgui can not run my input.pw please search the mailing list archives. the following could be helpful: http://www.democritos.it/pipermail/pw_forum/2005-November/003312.html cheers, axel. RG> RG> Thanks very much for PW_Forum. Success for all Quantum-Espresso user... RG> RG> Sincerely Yours RG> RG> RG> Rahmat Gunawan RG> _______________________________________________ RG> Pw_forum mailing list RG> Pw_forum at pwscf.org RG> http://www.democritos.it/mailman/listinfo/pw_forum RG> -- ======================================================================= 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 s305igun at mail.chem.itb.ac.id Sat Nov 4 05:57:31 2006 From: s305igun at mail.chem.itb.ac.id (Rahmat Gunawan) Date: Sat, 4 Nov 2006 11:57:31 +0700 (WIT) Subject: [Pw_forum] How to make chdens.x? In-Reply-To: References: <32940.167.205.72.7.1162579060.squirrel@webmail.chem.itb.ac.id> Message-ID: <2418.167.205.72.36.1162616251.squirrel@webmail.chem.itb.ac.id> Dear PW_Forum I would like to thanks for your attention to solve my problem. I follow solving problem from Axel and Aaron and I was running my file.pw: ibrav=4, celldm(1)=a (in atomic units), celldm(3)=c/a, nat=4, ntyp=1 ATOMIC_POSITIONS (alat) C 0.0 0.000000000 0.0 C 0.0 0.577350270 0.0 C 0.0 0.000000000 c/(2a) C 0.5 0.288675135 c/(2a) but, my PWgui need chdens.x and My PWgui can't run In my Espresso/PP have files: chdens.f90 and chdens.o so Would You like give me some method to make chdens.x? Oke Thanks very much for PW_Forum Good luch for Your Research... Sincerely Yours Rahmat Gunawan From ceresoli at sissa.it Sat Nov 4 10:14:16 2006 From: ceresoli at sissa.it (Davide Ceresoli) Date: Sat, 04 Nov 2006 10:14:16 +0100 Subject: [Pw_forum] compilation issue In-Reply-To: References: <2e36f8e50611030337k1c81cb9dgd9335a9e9144478b@mail.gmail.com> <454B3A59.7030000@sissa.it> <2e36f8e50611030754g5d261983j90d05ddd1715954@mail.gmail.com> Message-ID: <454C59E8.7000803@sissa.it> lan haiping wrote: > Do you mean link libs required statically ? > you should remove dynamical link in make.sys found by configure scripts, > and define proper paths for statical libs. > > hope help.. > If you want to enforce static linking, make sure the following line is present in your make.sys: LDFLAGS = -static On some linux distributions, however, the Intel compiler fails during the linking phase. The solution is: LDFLAGS = -i-static HTH. Davide From rudrabnrj at gmail.com Sun Nov 5 04:55:12 2006 From: rudrabnrj at gmail.com (rudra banerjee) Date: Sun, 5 Nov 2006 09:25:12 +0530 Subject: [Pw_forum] compilation issue In-Reply-To: <454C59E8.7000803@sissa.it> References: <2e36f8e50611030337k1c81cb9dgd9335a9e9144478b@mail.gmail.com> <454B3A59.7030000@sissa.it> <2e36f8e50611030754g5d261983j90d05ddd1715954@mail.gmail.com> <454C59E8.7000803@sissa.it> Message-ID: <2e36f8e50611041955oc8386e0rfcc978b6dd084659@mail.gmail.com> thank you david it is solved On 11/4/06, Davide Ceresoli wrote: > > lan haiping wrote: > > Do you mean link libs required statically ? > > you should remove dynamical link in make.sys found by configure scripts, > > and define proper paths for statical libs. > > > > hope help.. > > > > If you want to enforce static linking, make sure the > following line is present in your make.sys: > LDFLAGS = -static > > On some linux distributions, however, the Intel compiler > fails during the linking phase. The solution is: > LDFLAGS = -i-static > > HTH. > > Davide > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- rudra have a good time! -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061105/867a1943/attachment.htm From akohlmey at cmm.chem.upenn.edu Sun Nov 5 17:22:33 2006 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sun, 5 Nov 2006 11:22:33 -0500 (EST) Subject: [Pw_forum] How to make chdens.x? In-Reply-To: <2418.167.205.72.36.1162616251.squirrel@webmail.chem.itb.ac.id> Message-ID: On Sat, 4 Nov 2006, Rahmat Gunawan wrote: RG> Dear PW_Forum RG> RG> I would like to thanks for your attention to solve my problem. RG> RG> I follow solving problem from Axel and Aaron and I was running my file.pw: RG> RG> ibrav=4, celldm(1)=a (in atomic units), celldm(3)=c/a, nat=4, ntyp=1 RG> RG> ATOMIC_POSITIONS (alat) RG> C 0.0 0.000000000 0.0 RG> C 0.0 0.577350270 0.0 RG> C 0.0 0.000000000 c/(2a) RG> C 0.5 0.288675135 c/(2a) please note, this is not a valid input file, but merely a shortcut notation for a geometry. please have a look at the material at: http://www.vlab.msi.umn.edu/events/lecture.shtml especially from friday (05/26) to sunday (05/28). RG> but, my PWgui need chdens.x and My PWgui can't run why don't you just run pw.x directly? RG> In my Espresso/PP have files: chdens.f90 and chdens.o RG> so Would You like give me some method to make chdens.x? the chdens functinonality has been folded into pp.x about a year ago. there still seems to be some problems with updating pwgui accordingly. from the cvs changelog 2005-12-28 10:12 giannozz * GUI/PWgui/src/: run.itcl, settings.itcl: References to chdens.x removed - there is still one left in run.itcl I am not sure what should be done in that case cheers, axel. RG> RG> Oke Thanks very much for PW_Forum RG> RG> Good luch for Your Research... RG> RG> Sincerely Yours RG> RG> RG> Rahmat Gunawan RG> _______________________________________________ RG> Pw_forum mailing list RG> Pw_forum at pwscf.org RG> http://www.democritos.it/mailman/listinfo/pw_forum RG> -- ======================================================================= 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 s305igun at mail.chem.itb.ac.id Mon Nov 6 04:31:18 2006 From: s305igun at mail.chem.itb.ac.id (Rahmat Gunawan) Date: Mon, 6 Nov 2006 10:31:18 +0700 (WIT) Subject: [Pw_forum] Yes, PWscf do not need chdens.x In-Reply-To: References: <2418.167.205.72.36.1162616251.squirrel@webmail.chem.itb.ac.id> Message-ID: <3840.167.205.72.36.1162783878.squirrel@webmail.chem.itb.ac.id> Dear PW_Forum... Thanks to Axel to make it clear to me. That in PWscf-3.1.1 we do not need chdens.x as it was merged in pp.x And probably PWgui still has problems related to this change As Axel suggests, What we did id to use pw.x straightly, and to copy all *.x file to /usr/bin/ And it works! Now we want calculate band structure of graphite Sincerely Yours Thanks again Rahmat Gunawan > On Sat, 4 Nov 2006, Rahmat Gunawan wrote: > > RG> Dear PW_Forum > RG> > RG> I would like to thanks for your attention to solve my problem. > RG> > RG> I follow solving problem from Axel and Aaron and I was running my > file.pw: > RG> > RG> ibrav=4, celldm(1)=a (in atomic units), celldm(3)=c/a, nat=4, > ntyp=1 > RG> > RG> ATOMIC_POSITIONS (alat) > RG> C 0.0 0.000000000 0.0 > RG> C 0.0 0.577350270 0.0 > RG> C 0.0 0.000000000 c/(2a) > RG> C 0.5 0.288675135 c/(2a) > > please note, this is not a valid input file, but > merely a shortcut notation for a geometry. > > please have a look at the material at: > http://www.vlab.msi.umn.edu/events/lecture.shtml > especially from friday (05/26) to sunday (05/28). > > RG> but, my PWgui need chdens.x and My PWgui can't run > > why don't you just run pw.x directly? > > RG> In my Espresso/PP have files: chdens.f90 and chdens.o > RG> so Would You like give me some method to make chdens.x? > > the chdens functinonality has been folded into pp.x > about a year ago. there still seems to be some problems > with updating pwgui accordingly. from the cvs changelog > > 2005-12-28 10:12 giannozz > > * GUI/PWgui/src/: run.itcl, settings.itcl: > > References to chdens.x removed - there is still one left in > run.itcl > I am not sure what should be done in that case > > > cheers, > axel. > > RG> > RG> Oke Thanks very much for PW_Forum > RG> > RG> Good luch for Your Research... > RG> > RG> Sincerely Yours > RG> > RG> > RG> Rahmat Gunawan > RG> _______________________________________________ > RG> Pw_forum mailing list > RG> Pw_forum at pwscf.org > RG> http://www.democritos.it/mailman/listinfo/pw_forum > RG> > > -- > ======================================================================= > 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 giannozz at nest.sns.it Mon Nov 6 16:53:17 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 6 Nov 2006 16:53:17 +0100 Subject: [Pw_forum] timer on XT3 In-Reply-To: <45464284.000003.23349@webmail9.yandex.ru> References: <45464284.000003.23349@webmail9.yandex.ru> Message-ID: <296FC5FB-FF9D-4093-816E-D6A178C34713@nest.sns.it> On Oct 30, 2006, at 19:20 , Sergey Lisenkov wrote: > I just wanted to say that in CVS version of pwscf timer is not > working: thanks to Axel, it should work now --- Paolo Giannozzi, Democritos and University of Udine, Italy From stewart at cnf.cornell.edu Mon Nov 6 21:00:53 2006 From: stewart at cnf.cornell.edu (stewart at cnf.cornell.edu) Date: Mon, 06 Nov 2006 15:00:53 -0500 Subject: [Pw_forum] Running nscf without altering wfc and charge density files In-Reply-To: References: Message-ID: <20061106200053.25972.qmail@xuxa.iecc.com> Hi everyone, I am setting up a process where I will need to run several nscf calculations for band energy values for arbitrary k-points using the wfc and charge density from an initial scf calculation. However, I want to insure that the nscf runs do not write over the original wfc and charge density files so I can do subsequent nscf calculations. Is there a way to do this with pwscf? Or do I need to keep a storage directory with the original results and copy them over after every run? Thanks, Derek ################################ Derek Stewart, Ph. D. Scientific Computation Associate 250 Duffield Hall Cornell Nanoscale Facility (CNF) Ithaca, NY 14853 stewart (at) cnf.cornell.edu (607) 255-2856 From akohlmey at cmm.chem.upenn.edu Mon Nov 6 21:52:55 2006 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Mon, 6 Nov 2006 15:52:55 -0500 (EST) Subject: [Pw_forum] Running nscf without altering wfc and charge density files In-Reply-To: <20061106200053.25972.qmail@xuxa.iecc.com> Message-ID: On Mon, 6 Nov 2006 stewart at cnf.cornell.edu wrote: DS> Hi everyone, DS> DS> I am setting up a process where I will need to run several nscf calculations DS> for band energy values for arbitrary k-points using the wfc and charge DS> density from an initial scf calculation. However, I want to insure that the DS> nscf runs do not write over the original wfc and charge density files so I DS> can do subsequent nscf calculations. Is there a way to do this with pwscf? hi derek, it might work, but there is no special flag. you could try making the restart directory read-only with chmod and then see if pw.x still works. i made an audit of file accesses recently (to optimize i/o buffering on a disk-less machine) and it looks as if there are less writes now. i didn't try an nscf run, though. DS> Or do I need to keep a storage directory with the original results and copy DS> them over after every run? that is the recommended way. cheers, axel. DS> DS> Thanks, DS> DS> Derek DS> DS> DS> ################################ DS> Derek Stewart, Ph. D. DS> Scientific Computation Associate DS> 250 Duffield Hall DS> Cornell Nanoscale Facility (CNF) DS> Ithaca, NY 14853 DS> stewart (at) cnf.cornell.edu DS> (607) 255-2856 DS> DS> _______________________________________________ DS> Pw_forum mailing list DS> Pw_forum at pwscf.org DS> http://www.democritos.it/mailman/listinfo/pw_forum DS> -- ======================================================================= 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 giannozz at nest.sns.it Tue Nov 7 10:26:21 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 7 Nov 2006 10:26:21 +0100 Subject: [Pw_forum] Running nscf without altering wfc and charge density files In-Reply-To: <20061106200053.25972.qmail@xuxa.iecc.com> References: <20061106200053.25972.qmail@xuxa.iecc.com> Message-ID: <016CD2AC-C65E-4722-8636-A9C5D16DDBCD@nest.sns.it> On Nov 6, 2006, at 21:00 , stewart at cnf.cornell.edu wrote: > I am setting up a process where I will need to run several nscf > calculations > for band energy values for arbitrary k-points using the wfc and > charge density > from an initial scf calculation. However, I want to insure that > the nscf runs do > not write over the original wfc and charge density files so I can > do subsequent > nscf calculations. Is there a way to do this with pwscf? Or do I > need to keep a > storage directory with the original results and copy them over > after every run? the latter. At least in the CVS version, the self-consistent charge density is never overwritten, but the xml data file and the wavefunctions are. Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From giannozz at nest.sns.it Tue Nov 7 10:35:01 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 7 Nov 2006 10:35:01 +0100 Subject: [Pw_forum] Electron Phonon interaction: simple-minded questions In-Reply-To: References: Message-ID: <6CADF027-D7A1-45B9-9F5E-BD9B7A182771@nest.sns.it> On Nov 3, 2006, at 18:38 , Andrea Marini wrote: > My question is: as I am using a MP grid the shifted points (the > ones with zero weight) are still points of the 2x2x2 grid (like k > (2) and k(5)) . Why does PW calculates the states for these points > twice ? because in this particular case you have k(1)+q=k(5) > Is there any way to avoid it ? for sure there is, but it requires a programming effort that is not justified by a small performance gain in special cases only > 2. Why does PW performs a band structure calculation for every q- > point of a subset of the 2x2x2 grid. Can I calculate the states of > the whole grid once and use it for each q-point phonon calculation ? I think this is the way it is usually done, at least in recent versions > The point is that, while I get marvelous Eliashberg functions for a > metal like Al when I look at the el-ph matrix elements between the > fermi surface and high energy bands I'd expect them to go to zero > as a function of the upper band energy, while it is not. Does any > of you have experience of the high energy behavior of the el-ph > matrix elements ? I haven't any experience. High-energy bands are likely to be less accurate. > 4. When I try to calculate the matrix elements for many states I > get this error message already in the sc run > > task # 1 > from cgramg1_k : error # 1 > negative or zero norm in S > > This error disappears when I increase the energy cutoff. the entire machinery is based on the assumption that the number of states is much smaller than the number of plane waves, so when this is no longer true it may break down Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From niuyingli at gmail.com Wed Nov 8 03:03:40 2006 From: niuyingli at gmail.com (Yingli Niu) Date: Wed, 8 Nov 2006 10:03:40 +0800 Subject: [Pw_forum] About electron-phonon calculation on one point with version 3.1.1 Message-ID: <6398ac910611071803y2c563be2rcf479a7b7b723dfd@mail.gmail.com> Dear all I have read example07/run_example in espresso-3.1.1. It is convenient to calculate the q-points for a full dispersion with ldisp=.true. . But before el-ph calculation, it is better to make a convergence test on one or several q-points. Then I have to use the script example07/run_example in espresso-3.0. Because there is no example showing how to calculate el-ph on one q-point in espresso-3.1.1. I try to use example07/run_example in espresso-3.1.1 to calculate el-ph on one q-point: 1. make a scf calculation on dense k-mesh for el-ph calculation (case.scf.fit.in with la2F = .true.) 2. make a scf calculation for phonon calculation(case.scf.in) 3. make el-ph calculation: cat > li.elph.in_ecut << EOF Electron-phonon coefficients for Li &inputph reduce_io = .true. tr2_ph = 1.0d-14 prefix = 'li' fildyn = 'li.dyn' fildvscf = 'li.dv' amass(1) = 6.941 outdir = '$TMP_DIR/' elph = .true. trans = .true. lnscf = .true. ldisp = .false. / 0.00 -0.25 0.25 EOF I thought that it would calculate phonons on q-point(0.00 -0.25 0.25)because of "trans=.true." and "lnscf=.true.". But I failed. I make a calculation as in espresso-3.0 1. make a scf calculation on dense k-mesh for el-ph calculation (case.scf.in without la2F = .true.) 2. make a nscf calculation for phonon calculation (case.nscf1.in, calculation='phonon') 3. make a phonon calculation ( with elph=.flase. , trans=.true. , ldisp=.false.) 4. make a nscf calculation for elph calculation (case.nscf1.in, calculation='phonon') 5. make a el-ph calculation (with elph=.true. , trans=.true. , ldisp=.false.) I got the results! My question is: 1. Must we calculate one q-point el-ph using run_example in espresso-3.0? That is to say the method which I test (scf.fit, scf, el-ph) is wrong! 2. If we got error result for el-ph calculation on some q-point with ldisp=.true., when should we set restart_mode = 'restart'? In case.scf.fit.in or case.scf.in in espresso-3.1.1? 3. What's wrong with my script ? It should have calculated phonon and el-ph because it set trans=.true. ,elph=.true. to calculate on 0.00 -0.25 0.25? From chaohao2002 at 163.com Thu Nov 9 02:17:17 2006 From: chaohao2002 at 163.com (Chaohao Hu) Date: Thu, 9 Nov 2006 09:17:17 +0800 Subject: [Pw_forum] How to calculate charge density difference Message-ID: <4552853D.00C368.30936> Dear pwscf users, I have known that charge density difference (CDD) can be used to explain charge transfer and bonding characteristic in many references. But how to calculate it in pwscf code. I know pwscf can calculate charge density (CD) directly. For the interface including surface I and II, can we solve this problem as the following way: ================================== CDD(interface) = CD(interface) - CD(surface I) - CD(surface II) ================================== Certainly, the interface and suface I and II all maintain the same cell shape. Is it right? Can someone give more details or hints on it? I would appreciate your great help. Thanks in advance. Best regards, Chaohao ============================================================ Dr.Ch.H.Hu Institute of Metal Research,Chinese Academy of Sciences--IMR Wenhua Road, 72 110016, Shenyang, Liaoning (China) Phone: +086-024-23971641 Email: chhu at imr.ac.cn From duan at physics.usyd.edu.au Thu Nov 9 03:30:01 2006 From: duan at physics.usyd.edu.au (Xiangmei Duan) Date: Thu, 9 Nov 2006 13:30:01 +1100 (EST) Subject: [Pw_forum] How to calculate charge density difference In-Reply-To: <4552853D.00C368.30936> References: <4552853D.00C368.30936> Message-ID: for details see INPUT_CHDENS !----FOR i = 1, nfile: ! ! filepp(i) file containing the 3D charge (produced by pp.x) ! (AT LEAST filepp(1) REQUIRED) ! weight(i) weight - The quantity to be plotted will be ! weight(1)*rho(1) + weight(2)*rho(2) + weight(3)*rho(3)+... ! (OPTIONAL: default weight(1)=1.0) ! ! BEWARE: atomic coordinates are read from the first file; ! if their number is different for different files, ! the first file must have the largest number of atoms For your case, you have 3 files, with the weights 1.0 dor infterface, and -1.0 for surface I and II. Hope it helps, Xiangmei On Thu, 9 Nov 2006, Chaohao Hu wrote: > Dear pwscf users, > > I have known that charge density difference (CDD) can be used to > explain charge transfer and bonding characteristic in many > references. But how to calculate it in pwscf code. I know > pwscf can calculate charge density (CD) directly. For the interface > including surface I and II, can we solve this problem as the > following way: > ================================== > CDD(interface) = CD(interface) - CD(surface I) - CD(surface II) > ================================== > Certainly, the interface and suface I and II all maintain the same cell > shape. Is it right? > > Can someone give more details or hints on it? I would appreciate your > great help. > > Thanks in advance. > > Best regards, > Chaohao > > ============================================================ > Dr.Ch.H.Hu > Institute of Metal Research,Chinese Academy of Sciences--IMR > Wenhua Road, 72 > 110016, Shenyang, Liaoning (China) > Phone: +086-024-23971641 > Email: chhu at imr.ac.cn > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- ___________________________ School of Physics, A28 The University of sydney NSW, Sydney 2006 Tel: 02 9036 6039 Fax: 02 9351 7726 _____________________________________ From munima at iitg.ernet.in Thu Nov 9 04:49:46 2006 From: munima at iitg.ernet.in (Munima Bora Sahariah) Date: Thu, 9 Nov 2006 09:19:46 +0530 (IST) Subject: [Pw_forum] problem in running PWgui Message-ID: <34974.202.141.80.80.1163044186.squirrel@webmail.iitg.ernet.in> Hi, I am trying to run PWgui in my Linux 2.4 system and is having the following problem 1. it says ================================================== This is PWgui version: 3.1.1 -------------------------------------------------- PWgui: using the system default "wish" interpreter ./pwgui: line 54: wish: command not found 2. I tried replacing "wish" with "itkwish". It does not work. 3. I installed ActiveTCL. Still having the same message. Could anyone please help me out? Thanks Munima ----------------------------------------------------------------------------------- This email was sent from IIT Guwahati Webmail. If you are not the intended recipient, please contact the sender by email and delete all copies; your cooperation in this regard is appreciated. http://www.iitg.ernet.in From chaohao2002 at 163.com Thu Nov 9 04:53:41 2006 From: chaohao2002 at 163.com (Chaohao Hu) Date: Thu, 9 Nov 2006 11:53:41 +0800 Subject: [Pw_forum] Another question about building a spinel structure Message-ID: <4552A9DB.00BD2D.24542> Dear pwscf users, Now I want to visualize a spinel structure Al2MgO4 (space group: F-3dm; 227) using XCrySDen. But I can not visualize it rightly. My input files are enclosed here. Please you help me anlysis it. ======================================= cat > spinel.scf.in << EOF &control calculation='scf' restart_mode='from_scratch', tstress = .true. tprnfor = .true. prefix='spinel', pseudo_dir = '$PSEUDO_DIR/', outdir='$TMP_DIR/' / &system ibrav= 2, celldm(1) =15.2595, nat= 14, ntyp= 3, ecutwfc =36.0 / &electrons conv_thr = 1.0d-8 mixing_beta = 0.7 / ATOMIC_SPECIES Mg 24.305 Mg.pw91-np-van.UPF Al 26.982 Al.pw91-n-van.UPF O 15.9994 O.pbe-van_ak.UPF ATOMIC_POSITIONS Mg 0.125 0.125 0.125 Mg -0.125 -0.125 -0.125 Al 0.5 0.5 0.5 Al 0 0.5 0.5 Al 0.5 0 0.5 Al 0.5 0.5 0 O 0.236 0.236 0.236 O -0.208 0.236 0.236 O 0.236 -0.208 0.236 O 0.236 0.236 -0.208 O -0.236 -0.236 -0.236 O 0.208 -0.236 -0.236 O -0.236 0.208 -0.236 O -0.236 -0.236 0.208 K_POINTS {automatic} 5 5 5 1 1 1 EOF ========================================== However, when I change "ibrav" from 2 to 0, some error information always appears as following: ======================== invalid number: incomprehensible list input apparent state: unit 5 (unnamed) last format: list io lately reading sequential formatted external IO ======================== cat > spinel.scf.in << EOF &control calculation='scf' restart_mode='from_scratch', tstress = .true. tprnfor = .true. prefix='spinel', pseudo_dir = '$PSEUDO_DIR/', outdir='$TMP_DIR/' / &system ibrav= 0, celldm(1) =15.259, nat= 14, ntyp= 3, ecutwfc =36.0 / &electrons conv_thr = 1.0d-8 mixing_beta = 0.7 / ATOMIC_SPECIES Mg 24.305 Mg.pw91-np-van.UPF Al 26.982 Al.pw91-n-van.UPF O 15.9994 O.pbe-van_ak.UPF ATOMIC_POSITIONS Mg 0.125 0.125 0.125 Mg -0.125 -0.125 -0.125 Al 0.5 0.5 0.5 Al 0 0.5 0.5 Al 0.5 0 0.5 Al 0.5 0.5 0 O 0.236 0.236 0.236 O -0.208 0.236 0.236 O 0.236 -0.208 0.236 O 0.236 0.236 -0.208 O -0.236 -0.236 -0.236 O 0.208 -0.236 -0.236 O -0.236 0.208 -0.236 O -0.236 -0.236 0.208 K_POINTS {automatic} 5 5 5 1 1 1 CELL_PARAMETERS 0.0000 0.5000 0.5000 0.5000 0.0000 0.5000 0.5000 0.5000 0.0000 EOF =========================================== How to solve this problem? Best regards, Chaohao ============================================================ Dr. Chaohao Hu Institute of Metal Research, Chinese Academy of Sciences--IMR Wenhua Road, 72 110016, Shenyang, Liaoning (China) Phone: +86-024-23971641 Email: chhu at imr.ac.cn From chaohao2002 at 163.com Thu Nov 9 05:34:00 2006 From: chaohao2002 at 163.com (Chaohao Hu) Date: Thu, 9 Nov 2006 12:34:00 +0800 Subject: [Pw_forum] =?gb2312?B?UmU6ICjI8NDHzOHKvi20y9PKvP6/ycTcysfArLv408q8/ilSZTogW1B3X2ZvcnVtXSBIb3cgdG8gY2FsY3VsYXRlIGNoYXJnZSBkZW5zaXR5IGRpZmZlcmVuY2U=?= Message-ID: <4552AFBF.00C01E.28403> Thank you for your useful hints! >for details see INPUT_CHDENS > > !----FOR i = 1, nfile: > ! > ! filepp(i) file containing the 3D charge (produced by pp.x) > ! (AT LEAST filepp(1) REQUIRED) > ! weight(i) weight - The quantity to be plotted will be > ! weight(1)*rho(1) + weight(2)*rho(2) + weight(3)*rho(3)+... > ! (OPTIONAL: default weight(1)=1.0) > ! > ! BEWARE: atomic coordinates are read from the first >file; > ! if their number is different for different files, > ! the first file must have the largest number of atoms >For your case, you have 3 files, with the weights 1.0 dor infterface, and >-1.0 for surface I and II. > Best regards, Chaohao ============================================================ Dr. Chaohao Hu Institute of Metal Research, Chinese Academy of Sciences--IMR Wenhua Road, 72 110016, Shenyang, Liaoning (China) Phone: +86-024-23971641 Email: chhu at imr.ac.cn From chaohao2002 at 163.com Thu Nov 9 07:15:14 2006 From: chaohao2002 at 163.com (Chaohao Hu) Date: Thu, 9 Nov 2006 14:15:14 +0800 Subject: [Pw_forum] Another question about building a spinel structure Message-ID: <4552CB08.00E728.15796> I am sorry for my careless mistake. "ATOMIC_POSITIONS" should be set to "crystal" option. Now this structure is OK. However, for the case of "ibrav= 0", I still can not solve it. >Dear pwscf users, >Now I want to visualize a spinel structure Al2MgO4 (space group: F-3dm; 227) >using XCrySDen. But I can not visualize it rightly. My input files are enclosed >here. Please you help me anlysis it. >======================================= >cat > spinel.scf.in << EOF > &control > calculation='scf' > restart_mode='from_scratch', > tstress = .true. > tprnfor = .true. > prefix='spinel', > pseudo_dir = '$PSEUDO_DIR/', > outdir='$TMP_DIR/' > / > &system > ibrav= 2, celldm(1) =15.2595, nat= 14, ntyp= 3, > ecutwfc =36.0 > / > &electrons > conv_thr = 1.0d-8 > mixing_beta = 0.7 > / >ATOMIC_SPECIES > Mg 24.305 Mg.pw91-np-van.UPF > Al 26.982 Al.pw91-n-van.UPF > O 15.9994 O.pbe-van_ak.UPF >ATOMIC_POSITIONS >Mg 0.125 0.125 0.125 >Mg -0.125 -0.125 -0.125 >Al 0.5 0.5 0.5 >Al 0 0.5 0.5 >Al 0.5 0 0.5 >Al 0.5 0.5 0 >O 0.236 0.236 0.236 >O -0.208 0.236 0.236 >O 0.236 -0.208 0.236 >O 0.236 0.236 -0.208 >O -0.236 -0.236 -0.236 >O 0.208 -0.236 -0.236 >O -0.236 0.208 -0.236 >O -0.236 -0.236 0.208 >K_POINTS {automatic} > 5 5 5 1 1 1 >EOF >========================================== > >However, when I change "ibrav" from 2 to 0, some error information always >appears as following: >======================== >invalid number: incomprehensible list input >apparent state: unit 5 (unnamed) >last format: list io >lately reading sequential formatted external IO >======================== > >cat > spinel.scf.in << EOF > &control > calculation='scf' > restart_mode='from_scratch', > tstress = .true. > tprnfor = .true. > prefix='spinel', > pseudo_dir = '$PSEUDO_DIR/', > outdir='$TMP_DIR/' > / > &system > ibrav= 0, celldm(1) =15.259, nat= 14, ntyp= 3, > ecutwfc =36.0 > / > &electrons > conv_thr = 1.0d-8 > mixing_beta = 0.7 > / >ATOMIC_SPECIES > Mg 24.305 Mg.pw91-np-van.UPF > Al 26.982 Al.pw91-n-van.UPF > O 15.9994 O.pbe-van_ak.UPF >ATOMIC_POSITIONS >Mg 0.125 0.125 0.125 >Mg -0.125 -0.125 -0.125 >Al 0.5 0.5 0.5 >Al 0 0.5 0.5 >Al 0.5 0 0.5 >Al 0.5 0.5 0 >O 0.236 0.236 0.236 >O -0.208 0.236 0.236 >O 0.236 -0.208 0.236 >O 0.236 0.236 -0.208 >O -0.236 -0.236 -0.236 >O 0.208 -0.236 -0.236 >O -0.236 0.208 -0.236 >O -0.236 -0.236 0.208 >K_POINTS {automatic} > 5 5 5 1 1 1 >CELL_PARAMETERS > 0.0000 0.5000 0.5000 > 0.5000 0.0000 0.5000 > 0.5000 0.5000 0.0000 >EOF >=========================================== >How to solve this problem? > >Best regards, >Chaohao > >============================================================ >Dr. Chaohao Hu >Institute of Metal Research, Chinese Academy of Sciences--IMR >Wenhua Road, 72 >110016, Shenyang, Liaoning (China) >Phone: +86-024-23971641 >Email: chhu at imr.ac.cn > > > >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum > Thanks in advance. Best regards, Chaohao ============================================================ Dr. Chaohao Hu Institute of Metal Research, Chinese Academy of Sciences--IMR Wenhua Road, 72 110016, Shenyang, Liaoning (China) Phone: +86-024-23971641 Email: chhu at imr.ac.cn From tone.kokalj at ijs.si Thu Nov 9 10:55:27 2006 From: tone.kokalj at ijs.si (Tone Kokalj) Date: Thu, 09 Nov 2006 10:55:27 +0100 Subject: [Pw_forum] problem in running PWgui In-Reply-To: <34974.202.141.80.80.1163044186.squirrel@webmail.iitg.ernet.in> References: <34974.202.141.80.80.1163044186.squirrel@webmail.iitg.ernet.in> Message-ID: <1163066127.2465.3.camel@localhost.localdomain> On Thu, 2006-11-09 at 09:19 +0530, Munima Bora Sahariah wrote: > Hi, > > I am trying to run PWgui in my Linux 2.4 system and is having the following > problem > > 1. it says > > ================================================== > This is PWgui version: 3.1.1 > -------------------------------------------------- > > > PWgui: using the system default "wish" interpreter > > ./pwgui: line 54: wish: command not found This is because you installed the source version of PWgui, which requires Tcl/Tk + extensions. You can try with binary version instead. > 3. I installed ActiveTCL. Still having the same message. In this case, the source PWgui should work. You have to either add the directory of ActiveTcl binaries to the PATH, or in pwgui file change the line: wish $PWGUI/pwgui.tcl to /path/to/my/ActiveTclBinaries/wish $PWGUI/pwgui.tcl Regards, Tone From jibiaoli at gmail.com Fri Nov 10 05:44:00 2006 From: jibiaoli at gmail.com (Clark Lee) Date: Fri, 10 Nov 2006 12:44:00 +0800 Subject: [Pw_forum] How to calculate charge density difference In-Reply-To: References: Message-ID: A difference plot can be obtained by subtracting from a total system(e.g. an adsorbate on surface) the densities of each seperated system (e.g., the adsorbate, clean surface), keeping atomic position the same.This is very similar to the method of calculating an adsorption energy On 11/9/06, Xiangmei Duan wrote: > > > for details see INPUT_CHDENS > > !----FOR i = 1, nfile: > ! > ! filepp(i) file containing the 3D charge (produced by pp.x) > ! (AT LEAST filepp(1) REQUIRED) > ! weight(i) weight - The quantity to be plotted will be > ! weight(1)*rho(1) + weight(2)*rho(2) + > weight(3)*rho(3)+... > ! (OPTIONAL: default weight(1)=1.0) > ! > ! BEWARE: atomic coordinates are read from the first > file; > ! if their number is different for different files, > ! the first file must have the largest number of atoms > For your case, you have 3 files, with the weights 1.0 dor infterface, and > -1.0 for surface I and II. > > Hope it helps, > Xiangmei > > On Thu, 9 Nov 2006, Chaohao Hu wrote: > > > Dear pwscf users, > > > > I have known that charge density difference (CDD) can be used to > > explain charge transfer and bonding characteristic in many > > references. But how to calculate it in pwscf code. I know > > pwscf can calculate charge density (CD) directly. For the interface > > including surface I and II, can we solve this problem as the > > following way: > > ================================== > > CDD(interface) = CD(interface) - CD(surface I) - CD(surface II) > > ================================== > > Certainly, the interface and suface I and II all maintain the same cell > > shape. Is it right? > > > > Can someone give more details or hints on it? I would appreciate your > > great help. > > > > Thanks in advance. > > > > Best regards, > > Chaohao > > > > ============================================================ > > Dr.Ch.H.Hu > > Institute of Metal Research,Chinese Academy of Sciences--IMR > > Wenhua Road, 72 > > 110016, Shenyang, Liaoning (China) > > Phone: +086-024-23971641 > > Email: chhu at imr.ac.cn > > > > > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > -- > ___________________________ > School of Physics, A28 > The University of sydney > NSW, Sydney 2006 > Tel: 02 9036 6039 > Fax: 02 9351 7726 > _____________________________________ > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Standing on the shoulders of Newton -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061110/7d5db4e8/attachment.htm From jibiaoli at gmail.com Fri Nov 10 06:14:33 2006 From: jibiaoli at gmail.com (Clark Lee) Date: Fri, 10 Nov 2006 13:14:33 +0800 Subject: [Pw_forum] Another question about building a spinel structure In-Reply-To: <-4974276132272767522@unknownmsgid> References: <-4974276132272767522@unknownmsgid> Message-ID: Since the number of atomc specified is 14 for spinel structure, I supposethe case of ibrav= 2 could probably remove your trouble. Institute of Metal Research, Chinese Academy of Sciences--IMR Wenhua Road, 72 110016, Shenyang, Liaoning (China) Clark On 11/9/06, Chaohao Hu wrote: > > I am sorry for my careless mistake. "ATOMIC_POSITIONS" should be set to > "crystal" > option. Now this structure is OK. However, for the case of "ibrav= 0", I > still > can not solve it. > -- Standing on the shoulders of Newton -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061110/159c5be3/attachment.htm From jibiaoli at gmail.com Fri Nov 10 06:15:43 2006 From: jibiaoli at gmail.com (Clark Lee) Date: Fri, 10 Nov 2006 13:15:43 +0800 Subject: [Pw_forum] Another question about building a spinel structure In-Reply-To: References: <-4974276132272767522@unknownmsgid> Message-ID: Sorry, ibrav=1 On 11/10/06, Clark Lee wrote: > > Since the number of atomc specified is 14 for spinel structure, I supposethe case of ibrav= 2 > could probably remove your trouble. > > Institute of Metal Research, Chinese Academy of Sciences--IMR > Wenhua Road, 72 > 110016, Shenyang, Liaoning (China) > Clark > > On 11/9/06, Chaohao Hu wrote: > > > > I am sorry for my careless mistake. "ATOMIC_POSITIONS" should be set to > > "crystal" > > option. Now this structure is OK. However, for the case of "ibrav= 0", I > > still > > can not solve it. > > > > > > -- > Standing on the shoulders of Newton -- Standing on the shoulders of Newton -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061110/2ffe53cf/attachment.htm From chaohao2002 at 163.com Fri Nov 10 06:37:32 2006 From: chaohao2002 at 163.com (Chaohao Hu) Date: Fri, 10 Nov 2006 13:37:32 +0800 Subject: [Pw_forum] A question about relaxing structure Message-ID: <455413B4.018285.12007> Dear pwscf users, When I perform relaxing calculation (calculation ='relax' and ion_dynamics ='bfgs'), I encounter a problem related to symmetry. The error information is as following. But the scf calculation is OK. My system is a spinel structure (F-3dm). %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from checkallsym error # 2 symmetry operation not satisfied %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% How to solve this problem? I would appreciate your help. Best regards, Chaohao ============================================================ Dr. Chaohao Hu Institute of Metal Research, Chinese Academy of Sciences--IMR Wenhua Road, 72 110016, Shenyang, Liaoning (China) Phone: +86-024-23971641 Email: chhu at imr.ac.cn From chaohao2002 at 163.com Fri Nov 10 07:24:10 2006 From: chaohao2002 at 163.com (Chaohao Hu) Date: Fri, 10 Nov 2006 14:24:10 +0800 Subject: [Pw_forum] =?gb2312?B?UmU6ICjI8NDHzOHKvi20y9PKvP6/ycTcysfArLv408q8/ilSZTogW1B3X2ZvcnVtXSBBbm90aGVyIHF1ZXN0aW9uIGFib3V0IGJ1aWxkaW5nIGEgc3BpbmVsIHN0cnVjdHVyZQ==?= Message-ID: <45541B13.018768.17295> Sorry, The set of "ibrav=1" can not visualize this structure rightly. The conventional cell for this spinel structure should contain 56 atoms. The choice of "ibrav=2" and "ATOMIC_POSITIONS {crystal}" should be rational. Sorry, ibrav=1 On 11/10/06, Clark Lee wrote: Since the number of atomc specified is 14 for spinel structure, I suppose the case of ibrav= 2 could probably remove your trouble. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061110/851da45a/attachment.htm From munima at iitg.ernet.in Fri Nov 10 09:09:33 2006 From: munima at iitg.ernet.in (Munima Bora Sahariah) Date: Fri, 10 Nov 2006 13:39:33 +0530 (IST) Subject: [Pw_forum] problem in running calculations from PWgui. Message-ID: <47673.202.141.80.30.1163146173.squirrel@webmail.iitg.ernet.in> Hi, I can now get the PWgui window. But this time I am having problem in running a scf calculation (from the examples) from the menubar. An application error window appears which says - can't read "::pwscf::settings (CHDENS)":no such element in array. Please help me out. Thanks Munima ----------------------------------------------------------------------------------- This email was sent from IIT Guwahati Webmail. If you are not the intended recipient, please contact the sender by email and delete all copies; your cooperation in this regard is appreciated. http://www.iitg.ernet.in From cyrille.barreteau at cea.fr Fri Nov 10 10:27:12 2006 From: cyrille.barreteau at cea.fr (Cyrille Barreteau) Date: Fri, 10 Nov 2006 10:27:12 +0100 Subject: [Pw_forum] Iron cluster relaxation In-Reply-To: <4B7044B9-7DF9-4943-B854-1256BF837684@sissa.it> References: <55639d30610250444p7c71146eq8ae0faa309f92c1a@mail.gmail.com> <1161777636.3995.13.camel@dhpc-5-32.sissa.it> <4B7044B9-7DF9-4943-B854-1256BF837684@sissa.it> Message-ID: <455445F0.2060503@cea.fr> Dear Pwscf users I am trying to relax an iron cluster (cubocathedron) and pw.x systematicaly ends with the following error: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from bfgs : error # 1 bfgs history already reset at previous step %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... 2 You can find the input and output files on the following web site: http://www-drecam.cea.fr/Images/Pisp/cbarreteau/temporaire/cuboctahedron_relax.in http://www-drecam.cea.fr/Images/Pisp/cbarreteau/temporaire/cuboctahedron_relax.out thanks in advance for any constructive comment :-) cyrille -- ================================================================== Cyrille Barreteau CEA Saclay DSM/DRECAM/SPCSI Batiment 462 91191 Gif sur Yvette Cedex FRANCE ~~~~~~~~~~~~~~~~~~~~~~~~ phone : +33 (0)1 69 08 29 51 fax : +33 (0)1 69 08 84 46 email : cyrille.barreteau at cea.fr ~~~~~~~~~~~~~~~~~~~~~~~~ http://www-drecam.cea.fr/Images/Pisp/cbarreteau/cbarreteau_fr.html ================================================================== From s305igun at mail.chem.itb.ac.id Fri Nov 10 13:13:18 2006 From: s305igun at mail.chem.itb.ac.id (Rahmat Gunawan) Date: Fri, 10 Nov 2006 19:13:18 +0700 (WIT) Subject: [Pw_forum] problem in running calculations from PWgui. In-Reply-To: <47673.202.141.80.30.1163146173.squirrel@webmail.iitg.ernet.in> References: <47673.202.141.80.30.1163146173.squirrel@webmail.iitg.ernet.in> Message-ID: <1618.167.205.72.36.1163160798.squirrel@webmail.chem.itb.ac.id> Hi Munima, I try help you out of this problem. Please konfirm to Axel first... problem of CHDENS konfirm us that our input is wrong, my be we write pseudopotential is incorect. so try again and choose pseudopotential must be right. for example: atom Cu, the pseudopotential must Cu.UPF And write too ecutwfc and ecutrho Oke try again... Tahnks Rahmat Gunawan > Hi, > > I can now get the PWgui window. But this time I am having problem in > running a scf calculation (from the examples) from the menubar. An > application error window appears which says - can't read > "::pwscf::settings (CHDENS)":no such element in array. > > Please help me out. > > Thanks > Munima > > > ----------------------------------------------------------------------------------- > This email was sent from IIT Guwahati Webmail. If you are not the > intended recipient, please contact the sender by email and delete all > copies; your cooperation in this regard is appreciated. > http://www.iitg.ernet.in > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From akohlmey at cmm.chem.upenn.edu Fri Nov 10 17:55:52 2006 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Fri, 10 Nov 2006 11:55:52 -0500 (EST) Subject: [Pw_forum] Iron cluster relaxation In-Reply-To: <455445F0.2060503@cea.fr> Message-ID: On Fri, 10 Nov 2006, Cyrille Barreteau wrote: CB> Dear Pwscf users dear cyrille, did you check the mailing list archives? IIRC, this error seems to happen if you try to optimize the geometry too tightly for the given convergence of the wavefunction or when you suffer from 'ripples' due to a too small density cutoff. salut, axel. CB> I am trying to relax an iron cluster (cubocathedron) and CB> pw.x systematicaly ends with the following error: CB> CB> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% CB> from bfgs : error # 1 CB> bfgs history already reset at previous step CB> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% CB> CB> stopping ... CB> 2 CB> CB> CB> You can find the input and output files on the following web site: CB> CB> http://www-drecam.cea.fr/Images/Pisp/cbarreteau/temporaire/cuboctahedron_relax.in CB> http://www-drecam.cea.fr/Images/Pisp/cbarreteau/temporaire/cuboctahedron_relax.out CB> CB> CB> thanks in advance for any constructive comment :-) CB> CB> cyrille CB> CB> -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From konstantin_kudin at yahoo.com Fri Nov 10 20:57:13 2006 From: konstantin_kudin at yahoo.com (Konstantin Kudin) Date: Fri, 10 Nov 2006 11:57:13 -0800 (PST) Subject: [Pw_forum] Iron cluster relaxation In-Reply-To: Message-ID: <20061110195714.27392.qmail@web57714.mail.re3.yahoo.com> Axel, While your explanation is technically correct, it does not really address the underlying issue :-) It is not clear why the job should have a hard crash like this instead of either terminating gracefully, recovering from the condition in some way, or at least printing the very thing that you say, i.e. that "optimization terminated due to noise in the forces, increase cutoff", or something else similar. Actually, in my experience just restarting the job with the last coordinates has a very good chance to advance further, and even find a lower energy structure since the BFGS will be started from scratch. Kostya --- Axel Kohlmeyer wrote: > On Fri, 10 Nov 2006, Cyrille Barreteau wrote: > > CB> Dear Pwscf users > > dear cyrille, > > did you check the mailing list archives? > > IIRC, this error seems to happen if you try to > optimize the geometry too tightly for the > given convergence of the wavefunction or > when you suffer from 'ripples' due to a > too small density cutoff. > > salut, > axel. > > CB> I am trying to relax an iron cluster (cubocathedron) and > CB> pw.x systematicaly ends with the following error: > CB> > CB> > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > CB> from bfgs : error # 1 > CB> bfgs history already reset at previous step > CB> > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > CB> > CB> stopping ... > CB> 2 > CB> > CB> > CB> You can find the input and output files on the following web > site: > CB> > CB> > http://www-drecam.cea.fr/Images/Pisp/cbarreteau/temporaire/cuboctahedron_relax.in > CB> > http://www-drecam.cea.fr/Images/Pisp/cbarreteau/temporaire/cuboctahedron_relax.out > CB> > CB> > CB> thanks in advance for any constructive comment :-) > CB> > CB> cyrille > CB> > CB> > > -- > ======================================================================= > 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 > ____________________________________________________________________________________ Cheap talk? Check out Yahoo! Messenger's low PC-to-Phone call rates. http://voice.yahoo.com From akohlmey at cmm.chem.upenn.edu Fri Nov 10 21:22:24 2006 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Fri, 10 Nov 2006 15:22:24 -0500 (EST) Subject: [Pw_forum] Iron cluster relaxation In-Reply-To: <20061110195714.27392.qmail@web57714.mail.re3.yahoo.com> Message-ID: On Fri, 10 Nov 2006, Konstantin Kudin wrote: kostya, KK> It is not clear why the job should have a hard crash like this instead KK> of either terminating gracefully, recovering from the condition in some KK> way, or at least printing the very thing that you say, i.e. that KK> "optimization terminated due to noise in the forces, increase cutoff", KK> or something else similar. i agree. after this issue has come up quite a few times in the last months, it may be worth putting in the effort and changing the code, so that it a) exits gracefully, b) has a few more options on how to handle the resets, e.g. force a minimum number of steps between two resets, or have a small(!) randomization of coordinates. that just leaves the usual problem: who is going to do it? anybody in the market for making a lot of people happy? KK> Actually, in my experience just restarting the job with the last KK> coordinates has a very good chance to advance further, and even find a KK> lower energy structure since the BFGS will be started from scratch. hmmm. if you see the latter, then perhaps the reset is not a complete reset and there may be a bug (unless you have a 'randomization' of the coordinates due to limited accuracy when reading in the so far optimized coordinates). cheers, axel. KK> KK> Kostya KK> KK> KK> KK> --- Axel Kohlmeyer wrote: KK> KK> > On Fri, 10 Nov 2006, Cyrille Barreteau wrote: KK> > KK> > CB> Dear Pwscf users KK> > KK> > dear cyrille, KK> > KK> > did you check the mailing list archives? KK> > KK> > IIRC, this error seems to happen if you try to KK> > optimize the geometry too tightly for the KK> > given convergence of the wavefunction or KK> > when you suffer from 'ripples' due to a KK> > too small density cutoff. KK> > KK> > salut, KK> > axel. KK> > KK> > CB> I am trying to relax an iron cluster (cubocathedron) and KK> > CB> pw.x systematicaly ends with the following error: KK> > CB> KK> > CB> KK> > KK> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% KK> > CB> from bfgs : error # 1 KK> > CB> bfgs history already reset at previous step KK> > CB> KK> > KK> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% KK> > CB> KK> > CB> stopping ... KK> > CB> 2 KK> > CB> KK> > CB> KK> > CB> You can find the input and output files on the following web KK> > site: KK> > CB> KK> > CB> KK> > KK> http://www-drecam.cea.fr/Images/Pisp/cbarreteau/temporaire/cuboctahedron_relax.in KK> > CB> KK> > KK> http://www-drecam.cea.fr/Images/Pisp/cbarreteau/temporaire/cuboctahedron_relax.out KK> > CB> KK> > CB> KK> > CB> thanks in advance for any constructive comment :-) KK> > CB> KK> > CB> cyrille KK> > CB> KK> > CB> KK> > KK> > -- KK> > KK> ======================================================================= KK> > Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu KK> > http://www.cmm.upenn.edu KK> > Center for Molecular Modeling -- University of Pennsylvania KK> > Department of Chemistry, 231 S.34th Street, Philadelphia, PA KK> > 19104-6323 KK> > tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: KK> > 1-215-898-5425 KK> > KK> ======================================================================= KK> > If you make something idiot-proof, the universe creates a better KK> > idiot. KK> > KK> > _______________________________________________ KK> > Pw_forum mailing list KK> > Pw_forum at pwscf.org KK> > http://www.democritos.it/mailman/listinfo/pw_forum KK> > KK> KK> KK> KK> KK> ____________________________________________________________________________________ KK> Cheap talk? KK> Check out Yahoo! Messenger's low PC-to-Phone call rates. KK> http://voice.yahoo.com KK> _______________________________________________ KK> Pw_forum mailing list KK> Pw_forum at pwscf.org KK> http://www.democritos.it/mailman/listinfo/pw_forum KK> -- ======================================================================= 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 lanhaiping at gmail.com Sat Nov 11 06:34:43 2006 From: lanhaiping at gmail.com (lan haiping) Date: Sat, 11 Nov 2006 13:34:43 +0800 Subject: [Pw_forum] nscf calculation failure~ Message-ID: Hi,All. I came to a CRASH with this error information in nscf calculation : "%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% task # 1 from cgramg1_k : error # 1 negative or zero norm in S %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% " I browsed the mail list, it seems this error is related to ultrasoft pseudopotential's bad behavior, as pointing out by paolo "S is the overlap matrix in the Ultrasoft pseudopotential formalism. It has to be positive definite. Likely one (or more than one) of the pseudopotentials is bad " But no CRASH is occured in 'scf' or 'relax' calculation, it's too strange to belevie the problem of pseudopotential, I also try to use 'cg' method for diagonalization instead of 'david ' method, and no such CRASH would be occured. Regards, hai-ping -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061111/aa7e1a11/attachment.htm From joydeep at jncasr.ac.in Sun Nov 12 06:21:36 2006 From: joydeep at jncasr.ac.in (Joydeep Bhattacharjee) Date: Sun, 12 Nov 2006 10:51:36 +0530 Subject: [Pw_forum] Charged system relaxation Message-ID: <4556AF60.3050709@jncasr.ac.in> Hi I want to do a structural relaxation of a charged cluster using pwscf. I guess the relevant input variable is nelec. I am slightly confused about what happens when I use the metallic options: occupations='smearing', smearing='methfessel-paxton', degauss=0.004, for example. With the metallic options and nelec= nbnd*2 + #, where nbnd = (number of valence bands) and # is the number of extra electrons I want to include, the Fermi energy seems to indicate that the number electrons corresponding to the occupied states is still the same as in the neutral system, ie nbnd*2. So how does the code account for the extra charge while calculating the ground state density? And, if I want to infer about binding energies from such calculations, what kind of corrections I should look for to account for the compensating jellium background. Also, I will be glad to learn some extra tricks related to fast scf convergence in such scenarios. Thanks in advance. Joydeep JNCASR, Bangalore From hqzhou at nju.edu.cn Mon Nov 13 03:08:55 2006 From: hqzhou at nju.edu.cn (Huiqun Zhou) Date: Mon, 13 Nov 2006 10:08:55 +0800 Subject: [Pw_forum] Another question about building a spinel structure References: <4552CB08.00E728.15796> Message-ID: <004401c706c8$ac845cd0$1d00a8c0@solarflare> ibrav = 2 is correct, but you need to confirm if your atomic positions are really those that in a PRIMITIVE cell. Huiqun ----- Original Message ----- From: "Chaohao Hu" To: Sent: Thursday, November 09, 2006 2:15 PM Subject: Re: [Pw_forum] Another question about building a spinel structure >I am sorry for my careless mistake. "ATOMIC_POSITIONS" should be set to >"crystal" > option. Now this structure is OK. However, for the case of "ibrav= 0", I > still > can not solve it. > >>Dear pwscf users, >>Now I want to visualize a spinel structure Al2MgO4 (space group: F-3dm; >>227) >>using XCrySDen. But I can not visualize it rightly. My input files are >>enclosed >>here. Please you help me anlysis it. >>======================================= >>cat > spinel.scf.in << EOF >> &control >> calculation='scf' >> restart_mode='from_scratch', >> tstress = .true. >> tprnfor = .true. >> prefix='spinel', >> pseudo_dir = '$PSEUDO_DIR/', >> outdir='$TMP_DIR/' >> / >> &system >> ibrav= 2, celldm(1) =15.2595, nat= 14, ntyp= 3, >> ecutwfc =36.0 >> / >> &electrons >> conv_thr = 1.0d-8 >> mixing_beta = 0.7 >> / >>ATOMIC_SPECIES >> Mg 24.305 Mg.pw91-np-van.UPF >> Al 26.982 Al.pw91-n-van.UPF >> O 15.9994 O.pbe-van_ak.UPF >>ATOMIC_POSITIONS >>Mg 0.125 0.125 0.125 >>Mg -0.125 -0.125 -0.125 >>Al 0.5 0.5 0.5 >>Al 0 0.5 0.5 >>Al 0.5 0 0.5 >>Al 0.5 0.5 0 >>O 0.236 0.236 0.236 >>O -0.208 0.236 0.236 >>O 0.236 -0.208 0.236 >>O 0.236 0.236 -0.208 >>O -0.236 -0.236 -0.236 >>O 0.208 -0.236 -0.236 >>O -0.236 0.208 -0.236 >>O -0.236 -0.236 0.208 >>K_POINTS {automatic} >> 5 5 5 1 1 1 >>EOF >>========================================== >> >>However, when I change "ibrav" from 2 to 0, some error information always >>appears as following: >>======================== >>invalid number: incomprehensible list input >>apparent state: unit 5 (unnamed) >>last format: list io >>lately reading sequential formatted external IO >>======================== >> >>cat > spinel.scf.in << EOF >> &control >> calculation='scf' >> restart_mode='from_scratch', >> tstress = .true. >> tprnfor = .true. >> prefix='spinel', >> pseudo_dir = '$PSEUDO_DIR/', >> outdir='$TMP_DIR/' >> / >> &system >> ibrav= 0, celldm(1) =15.259, nat= 14, ntyp= 3, >> ecutwfc =36.0 >> / >> &electrons >> conv_thr = 1.0d-8 >> mixing_beta = 0.7 >> / >>ATOMIC_SPECIES >> Mg 24.305 Mg.pw91-np-van.UPF >> Al 26.982 Al.pw91-n-van.UPF >> O 15.9994 O.pbe-van_ak.UPF >>ATOMIC_POSITIONS >>Mg 0.125 0.125 0.125 >>Mg -0.125 -0.125 -0.125 >>Al 0.5 0.5 0.5 >>Al 0 0.5 0.5 >>Al 0.5 0 0.5 >>Al 0.5 0.5 0 >>O 0.236 0.236 0.236 >>O -0.208 0.236 0.236 >>O 0.236 -0.208 0.236 >>O 0.236 0.236 -0.208 >>O -0.236 -0.236 -0.236 >>O 0.208 -0.236 -0.236 >>O -0.236 0.208 -0.236 >>O -0.236 -0.236 0.208 >>K_POINTS {automatic} >> 5 5 5 1 1 1 >>CELL_PARAMETERS >> 0.0000 0.5000 0.5000 >> 0.5000 0.0000 0.5000 >> 0.5000 0.5000 0.0000 >>EOF >>=========================================== >>How to solve this problem? >> >>Best regards, >>Chaohao >> >>============================================================ >>Dr. Chaohao Hu >>Institute of Metal Research, Chinese Academy of Sciences--IMR >>Wenhua Road, 72 >>110016, Shenyang, Liaoning (China) >>Phone: +86-024-23971641 >>Email: chhu at imr.ac.cn >> >> >> >>_______________________________________________ >>Pw_forum mailing list >>Pw_forum at pwscf.org >>http://www.democritos.it/mailman/listinfo/pw_forum >> > > Thanks in advance. > Best regards, > Chaohao > > ============================================================ > Dr. Chaohao Hu > Institute of Metal Research, Chinese Academy of Sciences--IMR > Wenhua Road, 72 > 110016, Shenyang, Liaoning (China) > Phone: +86-024-23971641 > Email: chhu at imr.ac.cn > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From cyrille.barreteau at cea.fr Mon Nov 13 08:37:02 2006 From: cyrille.barreteau at cea.fr (Cyrille Barreteau) Date: Mon, 13 Nov 2006 08:37:02 +0100 Subject: [Pw_forum] Iron cluster relaxation In-Reply-To: References: Message-ID: <4558209E.9050706@cea.fr> Dear Kostya and Axel I have restarted my job with the last coordinates and the geometry optimization converged in just one scf cycle!! ========================================================== BFGS Geometry Optimization bfgs converged in 1 scf cycles and 0 bfgs steps End of BFGS Geometry Optimization ============================================= So I guess the previous "crashed run" did converge correctely toward the right optimal atomic configuration!! thanks cyrille >On Fri, 10 Nov 2006, Konstantin Kudin wrote: > >kostya, > >KK> It is not clear why the job should have a hard crash like this instead >KK> of either terminating gracefully, recovering from the condition in some >KK> way, or at least printing the very thing that you say, i.e. that >KK> "optimization terminated due to noise in the forces, increase cutoff", >KK> or something else similar. > >i agree. after this issue has come up quite a few times >in the last months, it may be worth putting in the effort >and changing the code, so that it a) exits gracefully, >b) has a few more options on how to handle the resets, >e.g. force a minimum number of steps between two resets, >or have a small(!) randomization of coordinates. > >that just leaves the usual problem: who is going to do it? >anybody in the market for making a lot of people happy? > >KK> Actually, in my experience just restarting the job with the last >KK> coordinates has a very good chance to advance further, and even find a >KK> lower energy structure since the BFGS will be started from scratch. > >hmmm. if you see the latter, then perhaps the reset is not >a complete reset and there may be a bug (unless you have >a 'randomization' of the coordinates due to limited >accuracy when reading in the so far optimized coordinates). > >cheers, > axel. > >KK> >KK> Kostya >KK> >KK> >KK> >KK> --- Axel Kohlmeyer wrote: >KK> >KK> > On Fri, 10 Nov 2006, Cyrille Barreteau wrote: >KK> > >KK> > CB> Dear Pwscf users >KK> > >KK> > dear cyrille, >KK> > >KK> > did you check the mailing list archives? >KK> > >KK> > IIRC, this error seems to happen if you try to >KK> > optimize the geometry too tightly for the >KK> > given convergence of the wavefunction or >KK> > when you suffer from 'ripples' due to a >KK> > too small density cutoff. >KK> > >KK> > salut, >KK> > axel. >KK> > >KK> > CB> I am trying to relax an iron cluster (cubocathedron) and >KK> > CB> pw.x systematicaly ends with the following error: >KK> > CB> >KK> > CB> >KK> > >KK> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% >KK> > CB> from bfgs : error # 1 >KK> > CB> bfgs history already reset at previous step >KK> > CB> >KK> > >KK> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% >KK> > CB> >KK> > CB> stopping ... >KK> > CB> 2 >KK> > CB> >KK> > CB> >KK> > CB> You can find the input and output files on the following web >KK> > site: >KK> > CB> >KK> > CB> >KK> > >KK> http://www-drecam.cea.fr/Images/Pisp/cbarreteau/temporaire/cuboctahedron_relax.in >KK> > CB> >KK> > >KK> http://www-drecam.cea.fr/Images/Pisp/cbarreteau/temporaire/cuboctahedron_relax.out >KK> > CB> >KK> > CB> >KK> > CB> thanks in advance for any constructive comment :-) >KK> > CB> >KK> > CB> cyrille >KK> > CB> >KK> > CB> >KK> > >KK> > -- >KK> > >KK> ======================================================================= >KK> > Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu >KK> > http://www.cmm.upenn.edu >KK> > Center for Molecular Modeling -- University of Pennsylvania >KK> > Department of Chemistry, 231 S.34th Street, Philadelphia, PA >KK> > 19104-6323 >KK> > tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: >KK> > 1-215-898-5425 >KK> > >KK> ======================================================================= >KK> > If you make something idiot-proof, the universe creates a better >KK> > idiot. >KK> > >KK> > _______________________________________________ >KK> > Pw_forum mailing list >KK> > Pw_forum at pwscf.org >KK> > http://www.democritos.it/mailman/listinfo/pw_forum >KK> > >KK> >KK> >KK> >KK> >KK> ____________________________________________________________________________________ >KK> Cheap talk? >KK> Check out Yahoo! Messenger's low PC-to-Phone call rates. >KK> http://voice.yahoo.com >KK> _______________________________________________ >KK> Pw_forum mailing list >KK> Pw_forum at pwscf.org >KK> http://www.democritos.it/mailman/listinfo/pw_forum >KK> > > > -- ================================================================== Cyrille Barreteau CEA Saclay DSM/DRECAM/SPCSI Batiment 462 91191 Gif sur Yvette Cedex FRANCE ~~~~~~~~~~~~~~~~~~~~~~~~ phone : +33 (0)1 69 08 29 51 fax : +33 (0)1 69 08 84 46 email : cyrille.barreteau at cea.fr ~~~~~~~~~~~~~~~~~~~~~~~~ http://www-drecam.cea.fr/Images/Pisp/cbarreteau/cbarreteau_fr.html ================================================================== From degironc at sissa.it Mon Nov 13 08:58:06 2006 From: degironc at sissa.it (degironc) Date: Mon, 13 Nov 2006 08:58:06 +0100 Subject: [Pw_forum] Charged system relaxation In-Reply-To: <4556AF60.3050709@jncasr.ac.in> References: <4556AF60.3050709@jncasr.ac.in> Message-ID: <4558258E.1000008@sissa.it> Dear Joydeep, nelec is indeed the right variable to set if you want a charged system. It should be set to the total number of electrons of the charged system. nbnd should be chosen large enough that (taking into account possible magnetization of the system) all electrons can be accomodated in the computed bands. If you run pw.x with verbosity='high' when convergence is reached it should write the bands energies and their occupations... It should be easy to verify if the results is what you expect. stefano Joydeep Bhattacharjee wrote: > Hi > > I want to do a structural relaxation of a charged cluster using pwscf. > I guess the relevant input variable is nelec. > I am slightly confused about what happens when I use the metallic > options: > occupations='smearing', smearing='methfessel-paxton', degauss=0.004, > for example. > With the metallic options and nelec= nbnd*2 + #, where nbnd = (number > of valence bands) > and # is the number of extra electrons I want to include, the Fermi > energy seems to indicate > that the number electrons corresponding to the occupied states is > still the same as in the > neutral system, ie nbnd*2. > So how does the code account for the extra charge while calculating > the ground state density? > And, if I want to infer about binding energies from such calculations, > what kind of corrections > I should look for to account for the compensating jellium background. > Also, I will be glad to learn some extra tricks related to fast scf > convergence in such scenarios. > Thanks in advance. > > Joydeep > JNCASR, Bangalore > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From cyrille.barreteau at cea.fr Mon Nov 13 09:31:26 2006 From: cyrille.barreteau at cea.fr (Cyrille Barreteau) Date: Mon, 13 Nov 2006 09:31:26 +0100 Subject: [Pw_forum] Iron cluster relaxation In-Reply-To: <4558209E.9050706@cea.fr> References: <4558209E.9050706@cea.fr> Message-ID: <45582D5E.7070409@cea.fr> In fact I have looked in more details at the results of pw.x calculations and there is a funny behaviour. The restarted calculations converges towards a totally different scf solution with a total magnetization 1mu_B larger than the previous "crashed" calculation. I guess there are several magnetic solutions and I "jumped" from one solution to the other.. Funny.. cyrille > Dear Kostya and Axel > > I have restarted my job with the last coordinates and > the geometry optimization converged in just one scf cycle!! > > ========================================================== > BFGS Geometry Optimization > > bfgs converged in 1 scf cycles and 0 bfgs steps > > End of BFGS Geometry Optimization > ============================================= > So I guess the previous "crashed run" did converge correctely > toward the right optimal atomic configuration!! > > thanks > > cyrille > >> On Fri, 10 Nov 2006, Konstantin Kudin wrote: >> >> kostya, >> >> KK> It is not clear why the job should have a hard crash like this >> instead >> KK> of either terminating gracefully, recovering from the condition >> in some >> KK> way, or at least printing the very thing that you say, i.e. that >> KK> "optimization terminated due to noise in the forces, increase >> cutoff", >> KK> or something else similar. >> >> i agree. after this issue has come up quite a few times in the last >> months, it may be worth putting in the effort >> and changing the code, so that it a) exits gracefully, >> b) has a few more options on how to handle the resets, >> e.g. force a minimum number of steps between two resets, >> or have a small(!) randomization of coordinates. >> >> that just leaves the usual problem: who is going to do it? >> anybody in the market for making a lot of people happy? >> >> KK> Actually, in my experience just restarting the job with the last >> KK> coordinates has a very good chance to advance further, and even >> find a >> KK> lower energy structure since the BFGS will be started from scratch. >> >> hmmm. if you see the latter, then perhaps the reset is not >> a complete reset and there may be a bug (unless you have >> a 'randomization' of the coordinates due to limited >> accuracy when reading in the so far optimized coordinates). >> >> cheers, >> axel. >> >> KK> KK> Kostya >> KK> KK> KK> KK> --- Axel Kohlmeyer wrote: >> KK> KK> > On Fri, 10 Nov 2006, Cyrille Barreteau wrote: >> KK> > KK> > CB> Dear Pwscf users >> KK> > KK> > dear cyrille, >> KK> > KK> > did you check the mailing list archives? >> KK> > KK> > IIRC, this error seems to happen if you try to >> KK> > optimize the geometry too tightly for the >> KK> > given convergence of the wavefunction or >> KK> > when you suffer from 'ripples' due to a >> KK> > too small density cutoff. >> KK> > KK> > salut, >> KK> > axel. >> KK> > KK> > CB> I am trying to relax an iron cluster (cubocathedron) and >> KK> > CB> pw.x systematicaly ends with the following error: >> KK> > CB> KK> > CB> KK> > >> KK> >> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% >> >> KK> > CB> from bfgs : error # 1 >> KK> > CB> bfgs history already reset at previous step >> KK> > CB> KK> > >> KK> >> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% >> >> KK> > CB> KK> > CB> stopping ... >> KK> > CB> 2 >> KK> > CB> KK> > CB> KK> > CB> You can find the input and output files >> on the following web >> KK> > site: >> KK> > CB> KK> > CB> >> KK> > >> KK> >> http://www-drecam.cea.fr/Images/Pisp/cbarreteau/temporaire/cuboctahedron_relax.in >> >> KK> > CB> >> KK> > >> KK> >> http://www-drecam.cea.fr/Images/Pisp/cbarreteau/temporaire/cuboctahedron_relax.out >> >> KK> > CB> KK> > CB> KK> > CB> thanks in advance for any constructive >> comment :-) >> KK> > CB> KK> > CB> cyrille >> KK> > CB> KK> > CB> KK> > KK> > -- KK> > >> KK> >> ======================================================================= >> KK> > Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu KK> > >> http://www.cmm.upenn.edu >> KK> > Center for Molecular Modeling -- University of Pennsylvania >> KK> > Department of Chemistry, 231 S.34th Street, Philadelphia, PA >> KK> > 19104-6323 >> KK> > tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: >> KK> > 1-215-898-5425 >> KK> > >> KK> >> ======================================================================= >> KK> > If you make something idiot-proof, the universe creates a better >> KK> > idiot. >> KK> > KK> > _______________________________________________ >> KK> > Pw_forum mailing list >> KK> > Pw_forum at pwscf.org >> KK> > http://www.democritos.it/mailman/listinfo/pw_forum >> KK> > KK> KK> KK> KK> KK> >> ____________________________________________________________________________________ >> >> KK> Cheap talk? >> KK> Check out Yahoo! Messenger's low PC-to-Phone call rates. >> KK> http://voice.yahoo.com >> KK> _______________________________________________ >> KK> Pw_forum mailing list >> KK> Pw_forum at pwscf.org >> KK> http://www.democritos.it/mailman/listinfo/pw_forum >> KK> >> >> > > -- ================================================================== Cyrille Barreteau CEA Saclay DSM/DRECAM/SPCSI Batiment 462 91191 Gif sur Yvette Cedex FRANCE ~~~~~~~~~~~~~~~~~~~~~~~~ phone : +33 (0)1 69 08 29 51 fax : +33 (0)1 69 08 84 46 email : cyrille.barreteau at cea.fr ~~~~~~~~~~~~~~~~~~~~~~~~ http://www-drecam.cea.fr/Images/Pisp/cbarreteau/cbarreteau_fr.html ================================================================== From akohlmey at cmm.chem.upenn.edu Mon Nov 13 18:32:02 2006 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Mon, 13 Nov 2006 12:32:02 -0500 (EST) Subject: [Pw_forum] nscf calculation failure~ In-Reply-To: Message-ID: On Sat, 11 Nov 2006, lan haiping wrote: HP> I browsed the mail list, it seems this error is related to ultrasoft HP> pseudopotential's bad behavior, as pointing out by paolo "S is the HP> overlap matrix in the Ultrasoft pseudopotential formalism. It has to HP> be positive definite. Likely one (or more than one) of the HP> pseudopotentials is bad " nevertheless, it may simply be that the pseudopotial is only 'marginally bad' or that different algorithms are less sensitive to the problematic pseudopotential(s). i know from personal experience, that a few of the pseudopotential generated from parameter files shipped with the vanderbilt atomic code have problems when re-creating them with the current version of the code, most likely due to changes in the fitting algorithm after the pseudopotential parameters were added. this seems to mainly affect rather heavy elements and pseudopotentials that were created as part of the dacapo project (or recreated with a different functional but using the same parameters). those potentials were conceived with an older version of the atomic code (7.0.0, see the dacapo homepage), where those problems may not have appeard. the potentials still test ok with the internal tests, so one really has to look through the (not exactly user friendly) log files. :-( one can usually correct for the problems (there are reports of 'negative densities' in the pseudopotential generation log) by adding a few additional constraints to the pseudopotential fits, but of course after that you need to re-verify the pseudopotential. it still is more convenient that having to build a pseudopotential completely from scratch... BTW. if somebody is interested in looking into this in more detail, for example by checking and (if needed) improving the pseudopotential files on www.pwscf.org, i should have a few example inputs with constraints (the explanation in the docs can be confusing) and can give a few (empirical!) hints on where to add those constraints. HP> But no CRASH is occured in 'scf' or 'relax' calculation, it's too strange HP> to belevie the problem of pseudopotential, please check for the indicators listed above, or let us know what pseudopotential files you are using. if you can prove that the pseudopotentials you use do not suffer from the problem i mentioned, it would be a much higher incentive to fix the code. thanks and best regards, axel. HP> I also try to use 'cg' method for diagonalization instead of 'david ' HP> method, and no such CRASH would be occured. HP> HP> Regards, HP> hai-ping HP> -- ======================================================================= 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 dylu at ucdavis.edu Mon Nov 13 19:36:50 2006 From: dylu at ucdavis.edu (Deyu Lu) Date: Mon, 13 Nov 2006 10:36:50 -0800 Subject: [Pw_forum] Postdoctoral and Graduate Student Positions in Angstrom Group Message-ID: <1163443010.5318.3.camel@localhost.localdomain> Postdoctoral (PD) and graduate student positions are available in the Angstrom group let by Prof. Giulia Galli, at the University of California, Davis (http://angstrom.ucdavis.edu/). The group activity is focused on the development and application of electronic structure and quantum simulation methods to the study of materials and nanostructures. In particular, a postdoctoral position is available, starting January 2007, to carry out a two year materials science project, in collaboration with Shell; and both post-doctoral and graduate student positions are available to work in the general area of quantum simulations of materials. PD candidates should have a PhD in Physics, Chemistry or Materials Science; a strong computational background in electronic structure and first principles simulations is preferred. Interested candidates should send their curriculum vitae and list of references to: Prof. Giulia Galli, gagalli at ucdavis.edu -- *************************************************************************** Deyu Lu (Ph.D) 190 Chemistry Building University of California, Davis One Shields Avenue Davis, CA 95616 Office phone: (530) 754-9663 Group Webpage: http://angstrom.ucdavis.edu/ *************************************************************************** From akohlmey at cmm.chem.upenn.edu Mon Nov 13 22:22:44 2006 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Mon, 13 Nov 2006 16:22:44 -0500 (EST) Subject: [Pw_forum] Iron cluster relaxation In-Reply-To: <45582D5E.7070409@cea.fr> Message-ID: On Mon, 13 Nov 2006, Cyrille Barreteau wrote: CB> In fact I have looked in more details at the results of pw.x calculations CB> and there is a funny behaviour. The restarted calculations converges towards CB> a totally different scf solution with a total magnetization 1mu_B larger CB> than the previous "crashed" calculation. I guess there are several magnetic CB> solutions and I "jumped" from one solution to the other.. Funny.. hmmmm.... interesting. that could explain the geometry optimizer getting 'stuck' even if you use a large enough cutoff and tight enough wavefunction convergence. since you always restart from the previous wavefunction, there is a high tendency to stick to a similar state, even if it is not the optimal one anymore. did you restart from the previous wavefunction/density or with a new (atomic or random) wavefunction? sadly, this kind of situation is one of the problems that can happen, sometimes rather unexpectedly (it happened to me in a CP dynamics only after several ps simulation time when the system seemed equilibrated and i switched the thermostats for production run). how about adding a (paranioa) flag to pw.x that allows to do a final wavefunction optimization from an initial wfc (either random or atomic) at the last step of a geometry optimization and then compare the energies/eigenvalues and warn if the differences are large (the flag could be a floating point number with 0.0 meaning no check)? cheers, axel. CB> CB> cyrille CB> CB> CB> > Dear Kostya and Axel CB> > CB> > I have restarted my job with the last coordinates and CB> > the geometry optimization converged in just one scf cycle!! CB> > CB> > ========================================================== CB> > BFGS Geometry Optimization CB> > CB> > bfgs converged in 1 scf cycles and 0 bfgs steps CB> > CB> > End of BFGS Geometry Optimization CB> > ============================================= CB> > So I guess the previous "crashed run" did converge correctely CB> > toward the right optimal atomic configuration!! CB> > CB> > thanks CB> > CB> > cyrille CB> > CB> >> On Fri, 10 Nov 2006, Konstantin Kudin wrote: CB> >> CB> >> kostya, CB> >> CB> >> KK> It is not clear why the job should have a hard crash like this CB> >> instead CB> >> KK> of either terminating gracefully, recovering from the condition CB> >> in some CB> >> KK> way, or at least printing the very thing that you say, i.e. that CB> >> KK> "optimization terminated due to noise in the forces, increase CB> >> cutoff", CB> >> KK> or something else similar. CB> >> CB> >> i agree. after this issue has come up quite a few times in the last CB> >> months, it may be worth putting in the effort CB> >> and changing the code, so that it a) exits gracefully, CB> >> b) has a few more options on how to handle the resets, CB> >> e.g. force a minimum number of steps between two resets, CB> >> or have a small(!) randomization of coordinates. CB> >> CB> >> that just leaves the usual problem: who is going to do it? CB> >> anybody in the market for making a lot of people happy? CB> >> CB> >> KK> Actually, in my experience just restarting the job with the last CB> >> KK> coordinates has a very good chance to advance further, and even CB> >> find a CB> >> KK> lower energy structure since the BFGS will be started from scratch. CB> >> CB> >> hmmm. if you see the latter, then perhaps the reset is not CB> >> a complete reset and there may be a bug (unless you have CB> >> a 'randomization' of the coordinates due to limited CB> >> accuracy when reading in the so far optimized coordinates). CB> >> CB> >> cheers, CB> >> axel. CB> >> CB> >> KK> KK> Kostya CB> >> KK> KK> KK> KK> --- Axel Kohlmeyer wrote: CB> >> KK> KK> > On Fri, 10 Nov 2006, Cyrille Barreteau wrote: CB> >> KK> > KK> > CB> Dear Pwscf users CB> >> KK> > KK> > dear cyrille, CB> >> KK> > KK> > did you check the mailing list archives? CB> >> KK> > KK> > IIRC, this error seems to happen if you try to CB> >> KK> > optimize the geometry too tightly for the CB> >> KK> > given convergence of the wavefunction or CB> >> KK> > when you suffer from 'ripples' due to a CB> >> KK> > too small density cutoff. CB> >> KK> > KK> > salut, CB> >> KK> > axel. CB> >> KK> > KK> > CB> I am trying to relax an iron cluster (cubocathedron) and CB> >> KK> > CB> pw.x systematicaly ends with the following error: CB> >> KK> > CB> KK> > CB> KK> > CB> >> KK> CB> >> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% CB> >> CB> >> KK> > CB> from bfgs : error # 1 CB> >> KK> > CB> bfgs history already reset at previous step CB> >> KK> > CB> KK> > CB> >> KK> CB> >> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% CB> >> CB> >> KK> > CB> KK> > CB> stopping ... CB> >> KK> > CB> 2 CB> >> KK> > CB> KK> > CB> KK> > CB> You can find the input and output files CB> >> on the following web CB> >> KK> > site: CB> >> KK> > CB> KK> > CB> CB> >> KK> > CB> >> KK> CB> >> http://www-drecam.cea.fr/Images/Pisp/cbarreteau/temporaire/cuboctahedron_relax.in CB> >> CB> >> KK> > CB> CB> >> KK> > CB> >> KK> CB> >> http://www-drecam.cea.fr/Images/Pisp/cbarreteau/temporaire/cuboctahedron_relax.out CB> >> CB> >> KK> > CB> KK> > CB> KK> > CB> thanks in advance for any constructive CB> >> comment :-) CB> >> KK> > CB> KK> > CB> cyrille CB> >> KK> > CB> KK> > CB> KK> > KK> > -- KK> > CB> >> KK> CB> >> ======================================================================= CB> >> KK> > Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu KK> > CB> >> http://www.cmm.upenn.edu CB> >> KK> > Center for Molecular Modeling -- University of Pennsylvania CB> >> KK> > Department of Chemistry, 231 S.34th Street, Philadelphia, PA CB> >> KK> > 19104-6323 CB> >> KK> > tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: CB> >> KK> > 1-215-898-5425 CB> >> KK> > CB> >> KK> CB> >> ======================================================================= CB> >> KK> > If you make something idiot-proof, the universe creates a better CB> >> KK> > idiot. CB> >> KK> > KK> > _______________________________________________ CB> >> KK> > Pw_forum mailing list CB> >> KK> > Pw_forum at pwscf.org CB> >> KK> > http://www.democritos.it/mailman/listinfo/pw_forum CB> >> KK> > KK> KK> KK> KK> KK> CB> >> ____________________________________________________________________________________ CB> >> CB> >> KK> Cheap talk? CB> >> KK> Check out Yahoo! Messenger's low PC-to-Phone call rates. CB> >> KK> http://voice.yahoo.com CB> >> KK> _______________________________________________ CB> >> KK> Pw_forum mailing list CB> >> KK> Pw_forum at pwscf.org CB> >> KK> http://www.democritos.it/mailman/listinfo/pw_forum CB> >> KK> CB> >> CB> >> CB> > CB> > CB> CB> CB> -- ======================================================================= 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 lanhaiping at gmail.com Tue Nov 14 12:15:00 2006 From: lanhaiping at gmail.com (lan haiping) Date: Tue, 14 Nov 2006 19:15:00 +0800 Subject: [Pw_forum] nscf calculation failure~ In-Reply-To: References: Message-ID: Thank you, Axel. I just came to this error when i tried to study bandstructure of anatase tio2 , with pbe-sp-van_ak.UPF from pwscf.org . Would you please tell me more details to fix this problem ? I would like to try . Regards, hai-ping On 11/14/06, Axel Kohlmeyer wrote: > > On Sat, 11 Nov 2006, lan haiping wrote: > > > HP> I browsed the mail list, it seems this error is related to ultrasoft > HP> pseudopotential's bad behavior, as pointing out by paolo "S is the > HP> overlap matrix in the Ultrasoft pseudopotential formalism. It has to > HP> be positive definite. Likely one (or more than one) of the > HP> pseudopotentials is bad " > > nevertheless, it may simply be that the pseudopotial is only > 'marginally bad' or that different algorithms are less sensitive > to the problematic pseudopotential(s). i know from personal experience, > that a few of the pseudopotential generated from parameter files shipped > with the vanderbilt atomic code have problems when re-creating them with > the current version of the code, most likely due to changes in the > fitting algorithm after the pseudopotential parameters were added. > > this seems to mainly affect rather heavy elements and pseudopotentials > that were created as part of the dacapo project (or recreated with a > different functional but using the same parameters). those potentials > were conceived with an older version of the atomic code (7.0.0, see the > dacapo homepage), where those problems may not have appeard. the > potentials still test ok with the internal tests, so one really has to > look through the (not exactly user friendly) log files. :-( > > one can usually correct for the problems (there are reports of > 'negative densities' in the pseudopotential generation log) by > adding a few additional constraints to the pseudopotential fits, > but of course after that you need to re-verify the pseudopotential. > it still is more convenient that having to build a pseudopotential > completely from scratch... > > BTW. if somebody is interested in looking into this in more detail, > for example by checking and (if needed) improving the pseudopotential > files on www.pwscf.org, i should have a few example inputs with > constraints (the explanation in the docs can be confusing) and can > give a few (empirical!) hints on where to add those constraints. > > HP> But no CRASH is occured in 'scf' or 'relax' calculation, it's too > strange > HP> to belevie the problem of pseudopotential, > > please check for the indicators listed above, or let us know > what pseudopotential files you are using. if you can prove that > the pseudopotentials you use do not suffer from the problem i > mentioned, it would be a much higher incentive to fix the code. > > thanks and best regards, > axel. > > HP> I also try to use 'cg' method for diagonalization instead of 'david > ' > HP> method, and no such CRASH would be occured. > HP> > HP> Regards, > HP> hai-ping > HP> > > -- > ======================================================================= > 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 > -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061114/511201f7/attachment.htm From joydeep at jncasr.ac.in Wed Nov 15 14:50:24 2006 From: joydeep at jncasr.ac.in (Joydeep Bhattacharjee) Date: Wed, 15 Nov 2006 19:20:24 +0530 Subject: [Pw_forum] Structural relaxation in electric field Message-ID: <455B1B20.4080509@jncasr.ac.in> Hi Thanks Stefano for your reply regarding the charged cluster relaxation. I now have a problem with structural relaxation of a cluster in the presence of electric fields as well. For electric field in one direction the rlaxations takes place nicely, but if I reverse the field the relaxation "converges" prematurely when nothing has actually converged, except the "new trust radius". Below is the relevant excerpt from the output file: - - - - - - - - - - - - ! total energy = -333.72126191 ryd estimated scf accuracy < 0.00000001 ryd band energy sum = -76.44745596 ryd one-electron contribution = -907.39503374 ryd hartree contribution = 445.49262839 ryd xc contribution = -46.19803489 ryd ewald contribution = 171.58360056 ryd electric field correction = 2.79557778 ryd correction for metals = 0.00000000 ryd convergence has been achieved Forces acting on atoms (Ry/au): atom 1 type 1 force = 0.00000000 0.00000000 0.00000000 atom 2 type 1 force = 0.00000000 0.00000000 0.00000000 atom 3 type 1 force = 0.00000000 0.00000000 0.00000000 atom 4 type 2 force = 0.00000000 0.00000000 0.00000000 atom 5 type 2 force = 0.00000000 0.00000000 0.00000000 atom 6 type 3 force = -0.23800154 0.00299140 0.00514244 atom 7 type 3 force = -0.23517359 -0.00315217 -0.00557059 Total force = 0.334705 Total SCF correction = 0.000274 number of scf cycles = 3 number of bfgs steps = 1 energy old = -343.2789722890 ryd energy new = -333.7212619063 ryd CASE: energy_new > energy_old new trust radius = 0.0000000000 bohr bfgs converged in 3 scf cycles and 1 bfgs steps Final energy = -343.2789722890 ryd Saving the approximate inverse hessian CELL_PARAMETERS (alat) 1.000000000 0.000000000 0.000000000 0.000000000 1.000000000 0.000000000 0.000000000 0.000000000 1.000000000 ATOMIC_POSITIONS (bohr) - - - - - - - - - - - - - - Writing file WS3.pun for program phonon - - - - - - - - - - - - - - ********************************************************************* The input file for a test efield: &control calculation='relax' restart_mode='from_scratch', prefix='WS3', pseudo_dir = '/export/joydeep/PWSCF/pseudi', outdir='/export/joydeep/PWSCF/Mo2S3_H2onMo_efield/tmp', tefield= .TRUE. / &system ibrav=0, nat=7, ntyp=3, occupations='smearing', smearing='methfessel-paxton', degauss=0.004, ecutwfc = 25.0, ecutrho = 130.0, edir = 1, emaxpos = 0.6, eopreg = 0.05, eamp = 0.02 / &electrons conv_thr = 1.0d-7 mixing_beta = 0.4 / &ions upscale=10 / ATOMIC_SPECIES S 6.0 S.pbe-van_bm.UPF Mo 14.0 Mo.pbe-van_uvw.UPF H 1.0 H.pbe-van_bm.UPF ATOMIC_POSITIONS {bohr} - - - - CELL_PARAMETERS {cubic} 30.0 0.0 0.0 0.0 30.0 0.0 0.0 0.0 30.0 K_POINTS {Gamma} ********************************************************************** Thanks in avance for your advise. Joydeep -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061115/322d3464/attachment.htm From ongphuongvu at yahoo.com Thu Nov 16 14:37:11 2006 From: ongphuongvu at yahoo.com (vu ongphuong) Date: Thu, 16 Nov 2006 05:37:11 -0800 (PST) Subject: [Pw_forum] Need of Perdew-Zunger exch-corr - US-PPs Message-ID: <43631.26993.qm@web50615.mail.yahoo.com> Dear all users I am in need of the Perdew-Zunger (LDA) exch-corr-Vanderbilt ultrasoft pseudopotentials of the following atom: Ti, O, Sr, Ba, Ca, Pb, Zr. Unfortunately, these pseudopotentials are not available at http://www.pwscf.org/pseudo.htm. I am trying to generate these pseudopotentials; and I think it must take me time to generate successfully. I will very grateful if someone who have these pseudopotentials give me the copies of them. Thank you very much. --------------------------------- Check out the all-new Yahoo! Mail beta - Fire up a more powerful email and get things done faster. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061116/028a8b4b/attachment.htm From marzari at MIT.EDU Thu Nov 16 17:31:58 2006 From: marzari at MIT.EDU (Nicola Marzari) Date: Thu, 16 Nov 2006 11:31:58 -0500 Subject: [Pw_forum] Need of Perdew-Zunger exch-corr - US-PPs In-Reply-To: <43631.26993.qm@web50615.mail.yahoo.com> References: <43631.26993.qm@web50615.mail.yahoo.com> Message-ID: <455C927E.704@mit.edu> Dear Vu, you can find all of them at http://www.physics.rutgers.edu/~dhv/uspp/ You'll need to compile David Vanderbilt's pseudopotential code (also available from the same site). I think espresso still reads the .van format - otherwise, just convert them to UPF using the converter from the espresso distribution. Best, nicola vu ongphuong wrote: > Dear all users > > I am in need of the Perdew-Zunger (LDA) exch-corr-Vanderbilt ultrasoft > pseudopotentials of the following atom: Ti, O, Sr, Ba, Ca, Pb, Zr. > Unfortunately, these pseudopotentials are not available at > http://www.pwscf.org/pseudo.htm. I am trying to generate these > pseudopotentials; and I think it must take me time to generate > successfully. I will very grateful if someone who have these > pseudopotentials give me the copies of them. > > Thank you very much. > > > > ------------------------------------------------------------------------ > Check out the all-new Yahoo! Mail beta > > - Fire up a more powerful email and get things done faster. -- --------------------------------------------------------------------- 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 Jeremie.Teyssier at physics.unige.ch Thu Nov 16 21:35:44 2006 From: Jeremie.Teyssier at physics.unige.ch (jeremie) Date: Thu, 16 Nov 2006 21:35:44 +0100 Subject: [Pw_forum] PB with electron-phonon coupling in parallel program Message-ID: <000301c709be$ca6ff6c0$d1e021c0@PORTABLE> Dear all I am running the parallel version of PWscf 3.1.1 for electron phonon calculation on a Linux cluster. This calculation based on the example 7 runs perfectly in the serial mode (or with parallel executable but on one node) but I have a problem when I run the calculation on more than one node. The first q point works fine but for the following I get the error message when calculating the EP coupling: ************************************************************************** omega( 1) = 0.389370 [THz] = 12.988061 [cm-1] omega( 2) = 0.389370 [THz] = 12.988061 [cm-1] %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% from elphsum : error # 1 pools and a2F not implemented %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% stopping ... omega( 3) = 0.389370 [THz] = 12.988061 [cm-1] omega( 4) = 2.653368 [THz] = 88.507417 [cm-1] omega( 5) = 2.653368 [THz] = 88.507417 [cm-1] omega( 6) = 2.653368 [THz] = 88.507417 [cm-1] omega( 7) = 16.187836 [THz] = 539.971685 [cm-1] %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% from elphsum : error # 1 pools and a2F not implemented %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% stopping ... omega( 8) = 16.187836 [THz] = 539.971685 [cm-1] omega( 9) = 16.187836 [THz] = 539.971685 [cm-1] omega(10) = 16.688065 [THz] = 556.657634 [cm-1] %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% from elphsum : error # 1 pools and a2F not implemented %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% stopping ... omega(11) = 16.688065 [THz] = 556.657634 [cm-1] %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% from elphsum : error # 1 pools and a2F not implemented %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% stopping ... omega(12) = 16.688065 [THz] = 556.657634 [cm-1] omega(13) = 24.175463 [THz] = 806.411998 [cm-1] omega(14) = 24.175463 [THz] = 806.411998 [cm-1] omega(15) = 24.175463 [THz] = 806.411998 [cm-1] omega(16) = 28.151777 [THz] = 939.048456 [cm-1] omega(17) = 28.151777 [THz] = 939.048456 [cm-1] %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% from elphsum : error # 1 pools and a2F not implemented %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% stopping ... omega(18) = 28.151777 [THz] = 939.048456 [cm-1] omega(19) = 32.608332 [THz] = 1087.704105 [cm-1] omega(20) = 32.608332 [THz] = 1087.704105 [cm-1] omega(21) = 38.921190 [THz] = 1298.279771 [cm-1] ************************************************************************** electron-phonon interaction ... %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% from elphsum : error # 1 pools and a2F not implemented %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% stopping ... %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% from elphsum : error # 1 pools and a2F not implemented %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% stopping ... %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% from elphsum : error # 1 pools and a2F not implemented %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% stopping ... Has anybody an idea? Jeremie Teyssier DPMC 24 quai Ernest Ensermet CH-1211 Geneve 4 SUISSE tel: (41) 22 379 65 78 fax: (41) 22 379 68 69 web: http://optics.unige.ch/jeremie/home_jeremie.html -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061116/dd59956a/attachment.htm From jzwanzig at dal.ca Thu Nov 16 23:09:16 2006 From: jzwanzig at dal.ca (Josef Zwanziger) Date: Thu, 16 Nov 2006 18:09:16 -0400 Subject: [Pw_forum] Status of NMR? Message-ID: <1163714957.22088.1.camel@jzwanzig-desktop> Dear Colleagues, I am wondering what the current status of the GIPAW NMR project is? In September it sounded like progress was good, but now I notice it has disappeared from the CSV download (no NMR_new directory, etc). Any comments? Thanks, Joe -- Josef W. Zwanziger Prof. of Chemistry and Canada Research Chair in NMR Studies of Materials Director, Atlantic Region Magnetic Resonance Centre Dept. of Chemistry Dalhousie University Halifax, NS B3H 4J3 Canada tel: 902.494.1960 fax: 902.494.1867 email: jzwanzig at dal.ca web: http://jwz.chem.dal.ca From konstantin_kudin at yahoo.com Fri Nov 17 01:15:05 2006 From: konstantin_kudin at yahoo.com (Konstantin Kudin) Date: Thu, 16 Nov 2006 16:15:05 -0800 (PST) Subject: [Pw_forum] Status of NMR? In-Reply-To: <1163714957.22088.1.camel@jzwanzig-desktop> Message-ID: <20061117001505.32039.qmail@web57702.mail.re3.yahoo.com> Dear Joe, I'll comment! At the risk of sounding politically unpopular :-), my suggestion is the following. While you are waiting for NMR in QE, use CPMD for NMR calculations. From what I have seen, it works quite well for that purpose, and parallelization is excellent. Kostya --- Josef Zwanziger wrote: > Dear Colleagues, > > I am wondering what the current status of the GIPAW NMR project is? > In > September it sounded like progress was good, but now I notice it has > disappeared from the CSV download (no NMR_new directory, etc). Any > comments? > > Thanks, > Joe > > > -- > Josef W. Zwanziger > Prof. of Chemistry and Canada Research Chair in NMR Studies of > Materials > Director, Atlantic Region Magnetic Resonance Centre > Dept. of Chemistry > Dalhousie University > Halifax, NS B3H 4J3 Canada > > tel: 902.494.1960 > fax: 902.494.1867 > email: jzwanzig at dal.ca > web: http://jwz.chem.dal.ca > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ____________________________________________________________________________________ Sponsored Link $200,000 mortgage for $660/ mo - 30/15 yr fixed, reduce debt - http://yahoo.ratemarketplace.com From giannozz at nest.sns.it Fri Nov 17 09:06:19 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Fri, 17 Nov 2006 09:06:19 +0100 Subject: [Pw_forum] PB with electron-phonon coupling in parallel program In-Reply-To: <000301c709be$ca6ff6c0$d1e021c0@PORTABLE> References: <000301c709be$ca6ff6c0$d1e021c0@PORTABLE> Message-ID: On Nov 16, 2006, at 21:35 , jeremie wrote: > from elphsum : error # 1 > pools and a2F not implemented > Has anybody an idea? electron-phonon calculations do not work together with k-point parallelization Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From ceresoli at sissa.it Fri Nov 17 11:38:49 2006 From: ceresoli at sissa.it (Davide Ceresoli) Date: Fri, 17 Nov 2006 11:38:49 +0100 Subject: [Pw_forum] Status of NMR? In-Reply-To: <20061117001505.32039.qmail@web57702.mail.re3.yahoo.com> References: <20061117001505.32039.qmail@web57702.mail.re3.yahoo.com> Message-ID: <455D9139.4090401@sissa.it> Konstantin Kudin wrote: > Dear Joe, > > I'll comment! > > At the risk of sounding politically unpopular :-), my suggestion is > the following. While you are waiting for NMR in QE, use CPMD for NMR > calculations. From what I have seen, it works quite well for that > purpose, and parallelization is excellent. > > Kostya Dear Kostya, at the risk of sounding politically unpopular :-), with CPMD you can compute only the relative $^1$H chemical shifts, since there is no reconstruction whatsoever. As shown in PRB 63, 24510, in a pseudopotential approach, the chemical shift is made up of four terms: the core contribution, the bare term, the paramagnetic correction and the diamagnetic correction. - the bare term is the contibution from valence electrons. - the core term can be computed from an atomic calculation and does not depend on the chemical environment. - the diamagnetic correction depends slightly on the chemical environment. - the paramagnetic correction depends a lot on the chemical environment. If you are interested in the *relative* chemical shifts, only the core contributions cancels out, while the paramagnetic and diamagnetic do not! Hydrogen is a very particular case, since the paramagnetic correction vanishes and the diamagnetic one is nearly constant. This said, I would like to remember that Quantum-Espresso is a GPL, community based project, and everyone is welcome to contribute as much one can. I'm not getting paid for this job and I'm working on other projects. There are still unsolved bugs in the PAW part. If anyone would like to help... The code has not disappeared in the CVS version. I guess the NMR_new directory is not tagged for the download using the "anonymous" CVS. Best regards, Davide -- +----------------------------------------------------------+ Davide Ceresoli Scuola Internazionale Superiore di Studi Avanzati (SISSA) via Beirut 2-4 I-34014 Trieste, Italy Phone: +39-040-3787-448 Fax: +39-040-3787-528 Mobile: +39-347-1001570 Homepage: http://www.sissa.it/~ceresoli +----------------------------------------------------------+ From jzwanzig at dal.ca Fri Nov 17 14:25:47 2006 From: jzwanzig at dal.ca (Josef Zwanziger) Date: Fri, 17 Nov 2006 09:25:47 -0400 Subject: [Pw_forum] Status of NMR again Message-ID: <1163769947.7326.22.camel@jzwanzig-desktop> Dear Colleagues and especially Davide Ceresoli, First, thanks for the suggestions about CPMD, I had in fact checked into that and had found as Davide points out too that it is implemented in a way that effectively restricts it to light atoms, which is not the area of the periodic table I'm working in. More importantly, I'm concerned from Davide's reply that he or any of the developers might think that I as a user do not appreciate their contributions and the fact that it's not a commercial venture and they have many other pressing issues. The fact is that I think pwscf is a wonderful tool, beautifully executed and supported, and I am extremely grateful for their efforts. I only wrote because, as a heavy NMR user, this new feature will be especially important for me and my colleagues in the community, and when I saw in the anonymous CVS download that it no longer appeared, I became concerned that the project had been closed. I am very happy to hear that it is going forward and look forward to using it. Davide also points out that everyone can help in this project, and I assume that includes me, so Davide, if you want to suggest specific places I could help, we can discuss what I can do. all the best, and thanks again to the developers for this project, Joe -- Josef W. Zwanziger Prof. of Chemistry and Canada Research Chair in NMR Studies of Materials Director, Atlantic Region Magnetic Resonance Centre Dept. of Chemistry Dalhousie University Halifax, NS B3H 4J3 Canada tel: 902.494.1960 fax: 902.494.1867 email: jzwanzig at dal.ca web: http://jwz.chem.dal.ca From baroni at sissa.it Fri Nov 17 17:01:15 2006 From: baroni at sissa.it (Stefano Baroni) Date: Fri, 17 Nov 2006 17:01:15 +0100 Subject: [Pw_forum] Status of NMR again In-Reply-To: <1163769947.7326.22.camel@jzwanzig-desktop> References: <1163769947.7326.22.camel@jzwanzig-desktop> Message-ID: <6A0B8C5E-B372-4F8A-9332-10ECD69ADE57@sissa.it> Dear Josef, many thanks for your interest in and kind words for our work. I just made a handful of telephone calls and I have been told that NMR functionalities are very close to be available in the Quantum Espresso distribution. There seems to be a nasty bug which is still escaping a close hunt, but I am confident that in a few weeks a first release of "pwnmr" (or whatever ...) should be available. Once again, many thanks for your consideration and patience. Stefano Baroni On Nov 17, 2006, at 2:25 PM, Josef Zwanziger wrote: > Dear Colleagues and especially Davide Ceresoli, > > First, thanks for the suggestions about CPMD, I had in fact checked > into > that and had found as Davide points out too that it is implemented > in a > way that effectively restricts it to light atoms, which is not the > area > of the periodic table I'm working in. > > More importantly, I'm concerned from Davide's reply that he or any of > the developers might think that I as a user do not appreciate their > contributions and the fact that it's not a commercial venture and they > have many other pressing issues. The fact is that I think pwscf is a > wonderful tool, beautifully executed and supported, and I am extremely > grateful for their efforts. I only wrote because, as a heavy NMR user, > this new feature will be especially important for me and my colleagues > in the community, and when I saw in the anonymous CVS download that it > no longer appeared, I became concerned that the project had been > closed. > I am very happy to hear that it is going forward and look forward to > using it. > > Davide also points out that everyone can help in this project, and I > assume that includes me, so Davide, if you want to suggest specific > places I could help, we can discuss what I can do. > > all the best, and thanks again to the developers for this project, > > Joe > -- > Josef W. Zwanziger > Prof. of Chemistry and Canada Research Chair in NMR Studies of > Materials > Director, Atlantic Region Magnetic Resonance Centre > Dept. of Chemistry > Dalhousie University > Halifax, NS B3H 4J3 Canada > > tel: 902.494.1960 > fax: 902.494.1867 > email: jzwanzig at dal.ca > web: http://jwz.chem.dal.ca > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061117/922c11d9/attachment.htm From jin.zhang.pku at gmail.com Sun Nov 19 08:29:15 2006 From: jin.zhang.pku at gmail.com (Jin Zhang) Date: Sun, 19 Nov 2006 15:29:15 +0800 Subject: [Pw_forum] Confused about vibrational modes of isolated molecule from ph.x Message-ID: <5d233b6a0611182329s46725cd4mf772a3a75f786c9a@mail.gmail.com> Dear all, I've tested the supercell approach using ph.x to calculate the vibrational frequency of a benzene molecule. It works indeed. And I got 36(12*3) eigen values. The thing I cannot understand is, there're 12 negative frequencies existing. I know 3 of them are purely translational modes and another 3 are rotational ones. But what about the rest 6? I'm a little confused and hopefully someone can help. The eigenvals are pasted below and thanks in advance! (the result may not be accurate due to a only-15Ry ecut I used, but it should not give out qualitative different result) omega( 1) = -75.772470 [THz] = -2527.514330 [cm-1] omega( 2) = -70.357356 [THz] = -2346.884371 [cm-1] omega( 3) = -69.734785 [THz] = -2326.117499 [cm-1] omega( 4) = -60.745067 [THz] = -2026.250783 [cm-1] omega( 5) = -60.633397 [THz] = -2022.525849 [cm-1] omega( 6) = -54.624392 [THz] = -1822.085685 [cm-1] omega( 7) = -53.629434 [THz] = -1788.897259 [cm-1] omega( 8) = -49.221716 [THz] = -1641.870640 [cm-1] omega( 9) = -48.767076 [THz] = -1626.705379 [cm-1] omega(10) = -45.544107 [THz] = -1519.197979 [cm-1] omega(11) = -40.582868 [THz] = -1353.707753 [cm-1] omega(12) = -29.501111 [THz] = -984.057687 [cm-1] omega(13) = 17.297930 [THz] = 577.000669 [cm-1] omega(14) = 18.430268 [THz] = 614.771652 [cm-1] omega(15) = 18.978064 [THz] = 633.044293 [cm-1] omega(16) = 21.573348 [THz] = 719.614207 [cm-1] omega(17) = 21.661957 [THz] = 722.569926 [cm-1] omega(18) = 22.212850 [THz] = 740.945851 [cm-1] omega(19) = 26.603080 [THz] = 887.389115 [cm-1] omega(20) = 27.163155 [THz] = 906.071359 [cm-1] omega(21) = 28.988845 [THz] = 966.970217 [cm-1] omega(22) = 29.897145 [THz] = 997.268046 [cm-1] omega(23) = 30.039795 [THz] = 1002.026353 [cm-1] omega(24) = 34.221037 [THz] = 1141.498514 [cm-1] omega(25) = 35.556617 [THz] = 1186.048959 [cm-1] omega(26) = 41.365733 [THz] = 1379.821484 [cm-1] omega(27) = 48.259770 [THz] = 1609.783366 [cm-1] omega(28) = 51.789502 [THz] = 1727.523303 [cm-1] omega(29) = 51.871548 [THz] = 1730.260091 [cm-1] omega(30) = 55.413486 [THz] = 1848.407220 [cm-1] omega(31) = 85.179130 [THz] = 2841.288830 [cm-1] omega(32) = 87.544947 [THz] = 2920.204497 [cm-1] omega(33) = 88.051107 [THz] = 2937.088312 [cm-1] omega(34) = 89.398003 [THz] = 2982.016222 [cm-1] omega(35) = 93.664201 [THz] = 3124.322195 [cm-1] omega(36) = 95.571010 [THz] = 3187.926929 [cm-1] -- Jin Zhang Dep. of Physics, Peking Univ. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061119/41ac825b/attachment.htm From sareh_zaree at yahoo.com Sun Nov 19 09:47:09 2006 From: sareh_zaree at yahoo.com (sareh zarei) Date: Sun, 19 Nov 2006 08:47:09 +0000 (GMT) Subject: [Pw_forum] integration volumes Message-ID: <20061119084709.22657.qmail@web31008.mail.mud.yahoo.com> Dear all I have a question about calculateing the atomic or total magnetization in your code. I would like to know how does one gets magnetization ?I mean that what do you select the integration volumes aruond the atoms? do you do the intgral just up to R_C radial or any other choise? your s.zarei Isfahan univrsity of technology Isfahan Iran Send instant messages to your online friends http://uk.messenger.yahoo.com From wmbmacam at lg.ehu.es Sun Nov 19 11:49:54 2006 From: wmbmacam at lg.ehu.es (=?ISO-8859-1?Q?Miguel_Mart=EDnez_Canales?=) Date: Sun, 19 Nov 2006 11:49:54 +0100 Subject: [Pw_forum] Confused about vibrational modes of isolated molecule from ph.x In-Reply-To: <5d233b6a0611182329s46725cd4mf772a3a75f786c9a@mail.gmail.com> References: <5d233b6a0611182329s46725cd4mf772a3a75f786c9a@mail.gmail.com> Message-ID: <456036D2.6030901@lg.ehu.es> Hi Jin, Jin Zhang escribi?: > The thing I cannot understand is, there're 12 negative frequencies > existing. I know 3 of them are purely translational modes and another 3 > are rotational ones. But what about the rest 6? I'm a little confused > and hopefully someone can help. > > The eigenvals are pasted below and thanks in advance! > > (the result may not be accurate due to a only-15Ry ecut I used, but it > should not give out qualitative different result) For a start, I wouldn't think 15 rydbergs is enough. Have you tested the pseudopotentials for ecut convergence? 25 or 30 would probably be a better choice. Anyway, what you probably are missing most in your calculations is k-point sampling convergence. Translational and rotational frequencies are a bit too large. What happens if you increase your k-point grid? Anyway, don?t forget to relax your molecule with the new parametres. BTW: What kind of cell are you using? I've never done molecular normal mode calculations but I guess that in the case of bencene using an hexagonal supercell might make sense. -- ---------------------------------------- Miguel Mart?nez Canales Dto. F?sica de la Materia Condensada UPV/EHU Facultad de Ciencia y Tecnolog?a Apdo. 644 48080 Bilbao (Spain) Fax: +34 94 601 3500 Tlf: +34 94 601 5437 ---------------------------------------- "The problem with Renault is that they dont have the skills to convince FIA that the mass damper is just a brake cooler." Annonymous From ezadshojaee at hotmail.com Sun Nov 19 12:48:08 2006 From: ezadshojaee at hotmail.com (Ezad Shojaee) Date: Sun, 19 Nov 2006 11:48:08 +0000 Subject: [Pw_forum] phonon at Gamma Message-ID: hi calculating phonons at gamma for my case, i have some frequencies minus and far from experimental data. there is no doubt in the structure and these are my input parameters ecutwfc=200 ecutrho=1600 K_points {automatic} 12 12 6 0 0 0 and the calculation is in the GGA approximation what have i missed? thanx for any help _________________________________________________________________ Express yourself instantly with MSN Messenger! Download today it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ From wmbmacam at lg.ehu.es Sun Nov 19 13:01:44 2006 From: wmbmacam at lg.ehu.es (=?ISO-8859-1?Q?Miguel_Mart=EDnez_Canales?=) Date: Sun, 19 Nov 2006 13:01:44 +0100 Subject: [Pw_forum] phonon at Gamma In-Reply-To: References: Message-ID: <456047A8.8030908@lg.ehu.es> Hi Ezad, Ezad Shojaee escribi?: > hi > calculating phonons at gamma for my case, i have some frequencies minus > and far from experimental data. there is no doubt in the structure and > these are my input parameters > > ecutwfc=200 > ecutrho=1600 > K_points {automatic} > 12 12 6 0 0 0 > > and the calculation is in the GGA approximation > what have i missed? > thanx for any help You might increase the k-point sampling to see if you have hit convergence. When q->gamma, acoustic modes need an increasingly large k-point grid to get good results. When one calculates the phonons via C(q) and C(r), this is usually overcome by forcing the acoustic sum rule. -- ---------------------------------------- Miguel Mart?nez Canales Dto. F?sica de la Materia Condensada UPV/EHU Facultad de Ciencia y Tecnolog?a Apdo. 644 48080 Bilbao (Spain) Fax: +34 94 601 3500 Tlf: +34 94 601 5437 ---------------------------------------- "The problem with Renault is that they dont have the skills to convince FIA that the mass damper is just a brake cooler." Annonymous From jin.zhang.pku at gmail.com Sun Nov 19 14:42:03 2006 From: jin.zhang.pku at gmail.com (Jin Zhang) Date: Sun, 19 Nov 2006 21:42:03 +0800 Subject: [Pw_forum] Confused about vibrational modes of isolated molecule from ph.x In-Reply-To: <456036D2.6030901@lg.ehu.es> References: <5d233b6a0611182329s46725cd4mf772a3a75f786c9a@mail.gmail.com> <456036D2.6030901@lg.ehu.es> Message-ID: <5d233b6a0611190542q19bcd0edq94b8c17917357c8@mail.gmail.com> Dear MMC, The supercell I used is a simple cubic one (let ibrav=8 celldim(1,2,3)=20bohr). Because the molecule is isolated, I think the kpoint sampling should be irrelevant. And I used gamma point for the calculation. I increased ecut to 25Ry, and celldim to 30bohr) the number of negative modes increase by 9 (to 21modes)! OMG, there must be sth wrong! I'll attach my input and see if someone can give some comments. !-----------------------omega------------------- omega( 1) = -123.660534 [THz] = -4124.898815 [cm-1] omega( 2) = -122.070631 [THz] = -4071.865015 [cm-1] omega( 3) = -109.408034 [THz] = -3649.483445 [cm-1] omega( 4) = -103.611190 [THz] = -3456.120267 [cm-1] omega( 5) = -100.288151 [THz] = -3345.274863 [cm-1] omega( 6) = -86.921490 [THz] = -2899.408087 [cm-1] omega( 7) = -86.648147 [THz] = -2890.290289 [cm-1] omega( 8) = -78.536226 [THz] = -2619.703943 [cm-1] omega( 9) = -63.552993 [THz] = -2119.913758 [cm-1] omega(10) = -60.437624 [THz] = -2015.995524 [cm-1] omega(11) = -59.290658 [THz] = -1977.736621 [cm-1] omega(12) = -59.022664 [THz] = -1968.797246 [cm-1] omega(13) = -58.953578 [THz] = -1966.492750 [cm-1] omega(14) = -46.326804 [THz] = -1545.306104 [cm-1] omega(15) = -31.263942 [THz] = -1042.859776 [cm-1] omega(16) = -31.093207 [THz] = -1037.164629 [cm-1] omega(17) = -27.758798 [THz] = -925.939988 [cm-1] omega(18) = -27.694407 [THz] = -923.792135 [cm-1] omega(19) = -24.097859 [THz] = -803.823410 [cm-1] omega(20) = -20.675736 [THz] = -689.672916 [cm-1] omega(21) = -20.064562 [THz] = -669.286187 [cm-1] omega(22) = 15.218497 [THz] = 507.637805 [cm-1] omega(23) = 16.277849 [THz] = 542.974191 [cm-1] omega(24) = 20.249558 [THz] = 675.457041 [cm-1] omega(25) = 21.320279 [THz] = 711.172668 [cm-1] omega(26) = 22.104349 [THz] = 737.326626 [cm-1] omega(27) = 23.427206 [THz] = 781.452687 [cm-1] omega(28) = 30.392570 [THz] = 1013.793763 [cm-1] omega(29) = 43.644315 [THz] = 1455.827328 [cm-1] omega(30) = 55.006552 [THz] = 1834.833267 [cm-1] omega(31) = 57.997854 [THz] = 1934.613035 [cm-1] omega(32) = 61.898591 [THz] = 2064.728462 [cm-1] omega(33) = 68.862483 [THz] = 2297.020463 [cm-1] omega(34) = 68.979827 [THz] = 2300.934652 [cm-1] omega(35) = 69.303015 [THz] = 2311.715098 [cm-1] omega(36) = 69.697807 [THz] = 2324.884051 [cm-1] On 11/19/06, Miguel Mart?nez Canales < wmbmacam at lg.ehu.es> wrote: > > Hi Jin, > > Jin Zhang escribi?: > > The thing I cannot understand is, there're 12 negative frequencies > > existing. I know 3 of them are purely translational modes and another 3 > > are rotational ones. But what about the rest 6? I'm a little confused > > and hopefully someone can help. > > > > The eigenvals are pasted below and thanks in advance! > > > > (the result may not be accurate due to a only-15Ry ecut I used, but it > > should not give out qualitative different result) > > For a start, I wouldn't think 15 rydbergs is enough. Have you tested the > pseudopotentials for ecut convergence? 25 or 30 would probably be a better > choice. Anyway, what you probably are missing most in your calculations is > k-point sampling convergence. Translational and rotational frequencies are > > a bit too large. What happens if you increase your k-point grid? Anyway, > don?t forget to relax your molecule with the new parametres. > > BTW: What kind of cell are you using? I've never done molecular normal > mode > calculations but I guess that in the case of bencene using an hexagonal > supercell might make sense. > > -- > ---------------------------------------- > Miguel Mart?nez Canales > Dto. F?sica de la Materia Condensada > UPV/EHU > Facultad de Ciencia y Tecnolog?a > Apdo. 644 > 48080 Bilbao (Spain) > Fax: +34 94 601 3500 > Tlf: +34 94 601 5437 > ---------------------------------------- > > "The problem with Renault is that they dont have > the skills to convince FIA that the mass damper > is just a brake cooler." > > Annonymous > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Jin Zhang Dep. of Physics, Peking Univ. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061119/e7f3d11c/attachment.htm -------------- next part -------------- A non-text attachment was scrubbed... Name: al.scf.fit.in Type: application/octet-stream Size: 933 bytes Desc: not available Url : /pipermail/attachments/20061119/e7f3d11c/attachment.obj -------------- next part -------------- A non-text attachment was scrubbed... Name: al.scf.in Type: application/octet-stream Size: 916 bytes Desc: not available Url : /pipermail/attachments/20061119/e7f3d11c/attachment-0001.obj -------------- next part -------------- A non-text attachment was scrubbed... Name: al.elph.in Type: application/octet-stream Size: 249 bytes Desc: not available Url : /pipermail/attachments/20061119/e7f3d11c/attachment-0002.obj From degironc at sissa.it Mon Nov 20 11:25:07 2006 From: degironc at sissa.it (degironc) Date: Mon, 20 Nov 2006 11:25:07 +0100 Subject: [Pw_forum] Confused about vibrational modes of isolated molecule from ph.x In-Reply-To: <5d233b6a0611190542q19bcd0edq94b8c17917357c8@mail.gmail.com> References: <5d233b6a0611182329s46725cd4mf772a3a75f786c9a@mail.gmail.com> <456036D2.6030901@lg.ehu.es> <5d233b6a0611190542q19bcd0edq94b8c17917357c8@mail.gmail.com> Message-ID: <45618283.1030403@sissa.it> did you relaxed the structure ? These are the forces I get with your input [a typical threshold for a structural relaxation is forc_conv_thr REAL ( default = 1.0D-3 ) ]. stefano Forces acting on atoms (Ry/au): atom 1 type 2 force = -0.06870568 0.00011605 0.00070580 atom 2 type 1 force = 0.00057171 -0.00133033 0.00026104 atom 3 type 2 force = -0.03470840 0.06008037 0.00060235 atom 4 type 1 force = -0.00200170 -0.00052894 -0.00023669 atom 5 type 2 force = 0.03477692 0.05989867 -0.00004091 atom 6 type 1 force = 0.00029793 0.00027534 -0.00032982 atom 7 type 2 force = 0.06862846 -0.00002844 -0.00054894 atom 8 type 1 force = -0.00256825 0.00128897 -0.00025706 atom 9 type 2 force = 0.03495275 -0.06050673 -0.00040942 atom 10 type 1 force = -0.00020204 0.00131722 0.00025463 atom 11 type 2 force = -0.03450037 -0.05925204 0.00020053 atom 12 type 1 force = 0.00345864 -0.00133014 -0.00020152 Total force = 0.169286 Total SCF correction = 0.000107 stefano Jin Zhang wrote: > Dear MMC, > > The supercell I used is a simple cubic one (let ibrav=8 > celldim(1,2,3)=20bohr). Because the molecule is isolated, I think the > kpoint sampling should be irrelevant. And I used gamma point for the > calculation. > > I increased ecut to 25Ry, and celldim to 30bohr) the number of > negative modes increase by 9 (to 21modes)! > OMG, there must be sth wrong! > I'll attach my input and see if someone can give some comments. > !-----------------------omega------------------- > omega( 1) = -123.660534 [THz] = -4124.898815 [cm-1] > omega( 2) = -122.070631 [THz] = -4071.865015 [cm-1] > omega( 3) = -109.408034 [THz] = -3649.483445 [cm-1] > omega( 4) = -103.611190 [THz] = -3456.120267 [cm-1] > omega( 5) = -100.288151 [THz] = -3345.274863 [cm-1] > omega( 6) = -86.921490 [THz] = -2899.408087 [cm-1] > omega( 7) = -86.648147 [THz] = -2890.290289 [cm-1] > omega( 8) = -78.536226 [THz] = -2619.703943 [cm-1] > omega( 9) = -63.552993 [THz] = -2119.913758 [cm-1] > omega(10) = -60.437624 [THz] = -2015.995524 [cm-1] > omega(11) = - 59.290658 [THz] = -1977.736621 [cm-1] > omega(12) = -59.022664 [THz] = -1968.797246 [cm-1] > omega(13) = -58.953578 [THz] = -1966.492750 [cm-1] > omega(14) = -46.326804 [THz] = -1545.306104 [cm-1] > omega(15) = -31.263942 [THz] = -1042.859776 [cm-1] > omega(16) = -31.093207 [THz] = -1037.164629 [cm-1] > omega(17) = -27.758798 [THz] = -925.939988 [cm-1] > omega(18) = - 27.694407 [THz] = -923.792135 [cm-1] > omega(19) = -24.097859 [THz] = -803.823410 [cm-1] > omega(20) = -20.675736 [THz] = -689.672916 [cm-1] > omega(21) = -20.064562 [THz] = -669.286187 [cm-1] > omega(22) = 15.218497 [THz] = 507.637805 [cm-1] > omega(23) = 16.277849 [THz] = 542.974191 [cm-1] > omega(24) = 20.249558 [THz] = 675.457041 [cm-1] > omega(25) = 21.320279 [THz] = 711.172668 [cm-1] > omega(26) = 22.104349 [THz] = 737.326626 [cm-1] > omega(27) = 23.427206 [THz] = 781.452687 [cm-1] > omega(28) = 30.392570 [THz] = 1013.793763 [cm-1] > omega(29) = 43.644315 [THz] = 1455.827328 [cm-1] > omega(30) = 55.006552 [THz] = 1834.833267 [cm-1] > omega(31) = 57.997854 [THz] = 1934.613035 [cm-1] > omega(32) = 61.898591 [THz] = 2064.728462 [cm-1] > omega(33) = 68.862483 [THz] = 2297.020463 [cm-1] > omega(34) = 68.979827 [THz] = 2300.934652 [cm-1] > omega(35) = 69.303015 [THz] = 2311.715098 [cm-1] > omega(36) = 69.697807 [THz] = 2324.884051 [cm-1] > > On 11/19/06, *Miguel Mart?nez Canales* < wmbmacam at lg.ehu.es > > wrote: > > Hi Jin, > > Jin Zhang escribi?: > > The thing I cannot understand is, there're 12 negative frequencies > > existing. I know 3 of them are purely translational modes and > another 3 > > are rotational ones. But what about the rest 6? I'm a little > confused > > and hopefully someone can help. > > > > The eigenvals are pasted below and thanks in advance! > > > > (the result may not be accurate due to a only-15Ry ecut I used, > but it > > should not give out qualitative different result) > > For a start, I wouldn't think 15 rydbergs is enough. Have you > tested the > pseudopotentials for ecut convergence? 25 or 30 would probably be > a better > choice. Anyway, what you probably are missing most in your > calculations is > k-point sampling convergence. Translational and rotational > frequencies are > a bit too large. What happens if you increase your k-point grid? > Anyway, > don?t forget to relax your molecule with the new parametres. > > BTW: What kind of cell are you using? I've never done molecular > normal mode > calculations but I guess that in the case of bencene using an > hexagonal > supercell might make sense. > > -- > ---------------------------------------- > Miguel Mart?nez Canales > Dto. F?sica de la Materia Condensada > UPV/EHU > Facultad de Ciencia y Tecnolog?a > Apdo. 644 > 48080 Bilbao (Spain) > Fax: +34 94 601 3500 > Tlf: +34 94 601 5437 > ---------------------------------------- > > "The problem with Renault is that they dont have > the skills to convince FIA that the mass damper > is just a brake cooler." > > Annonymous > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > -- > Jin Zhang > Dep. of Physics, > Peking Univ. From jin.zhang.pku at gmail.com Mon Nov 20 13:27:55 2006 From: jin.zhang.pku at gmail.com (Jin Zhang) Date: Mon, 20 Nov 2006 20:27:55 +0800 Subject: [Pw_forum] Confused about vibrational modes of isolated molecule from ph.x In-Reply-To: <45618283.1030403@sissa.it> References: <5d233b6a0611182329s46725cd4mf772a3a75f786c9a@mail.gmail.com> <456036D2.6030901@lg.ehu.es> <5d233b6a0611190542q19bcd0edq94b8c17917357c8@mail.gmail.com> <45618283.1030403@sissa.it> Message-ID: <5d233b6a0611200427w27ad0305pc63196f73eb8ea44@mail.gmail.com> No. I didn't relax the molecule. As a matter of fact, I got this structure from want code's self test, and thought it was already relaxed-.- I'll relax it later and see if good things happen. Best, -- Jin Zhang Dep. of Physics, Peking Univ. On 11/20/06, degironc wrote: > > did you relaxed the structure ? > These are the forces I get with your input [a typical threshold for a > structural relaxation is > forc_conv_thr REAL ( default = 1.0D-3 ) ]. > stefano > > Forces acting on atoms (Ry/au): > > atom 1 type 2 force = -0.06870568 0.00011605 0.00070580 > atom 2 type 1 force = 0.00057171 -0.00133033 0.00026104 > atom 3 type 2 force = -0.03470840 0.06008037 > 0.00060235 > atom 4 type 1 force = -0.00200170 -0.00052894 -0.00023669 > atom 5 type 2 force = 0.03477692 0.05989867 -0.00004091 > atom 6 type 1 force = 0.00029793 0.00027534 - > 0.00032982 > atom 7 type 2 force = 0.06862846 -0.00002844 -0.00054894 > atom 8 type 1 force = -0.00256825 0.00128897 -0.00025706 > atom 9 type 2 force = 0.03495275 -0.06050673 - > 0.00040942 > atom 10 type 1 force = -0.00020204 0.00131722 0.00025463 > atom 11 type 2 force = -0.03450037 -0.05925204 0.00020053 > atom 12 type 1 force = 0.00345864 -0.00133014 -0.00020152 > > Total force = 0.169286 Total SCF correction = 0.000107 > > stefano > > > Jin Zhang wrote: > > > Dear MMC, > > > > The supercell I used is a simple cubic one (let ibrav=8 > > celldim(1,2,3)=20bohr). Because the molecule is isolated, I think the > > kpoint sampling should be irrelevant. And I used gamma point for the > > calculation. > > > > I increased ecut to 25Ry, and celldim to 30bohr) the number of > > negative modes increase by 9 (to 21modes)! > > OMG, there must be sth wrong! > > I'll attach my input and see if someone can give some comments. > > !-----------------------omega------------------- > > omega( 1) = -123.660534 [THz] = -4124.898815 [cm-1] > > omega( 2) = -122.070631 [THz] = -4071.865015 [cm-1] > > omega( 3) = -109.408034 [THz] = -3649.483445 [cm-1] > > omega( 4) = - 103.611190 [THz] = -3456.120267 [cm-1] > > omega( 5) = -100.288151 [THz] = -3345.274863 [cm-1] > > omega( 6) = -86.921490 [THz] = -2899.408087 [cm-1] > > omega( 7) = -86.648147 [THz] = - 2890.290289 [cm-1] > > omega( 8) = -78.536226 [THz] = -2619.703943 [cm-1] > > omega( 9) = -63.552993 [THz] = -2119.913758 [cm-1] > > omega(10) = -60.437624 [THz] = -2015.995524 [cm-1] > > omega(11) = - 59.290658 [THz] = -1977.736621 [cm-1] > > omega(12) = -59.022664 [THz] = -1968.797246 [cm-1] > > omega(13) = -58.953578 [THz] = -1966.492750 [cm-1] > > omega(14) = -46.326804 [THz] = -1545.306104 [cm-1] > > omega(15) = -31.263942 [THz] = -1042.859776 [cm-1] > > omega(16) = -31.093207 [THz] = -1037.164629 [cm-1] > > omega(17) = - 27.758798 [THz] = -925.939988 [cm-1] > > omega(18) = - 27.694407 [THz] = -923.792135 [cm-1] > > omega(19) = -24.097859 [THz] = -803.823410 [cm-1] > > omega(20) = -20.675736 [THz] = - 689.672916 [cm-1] > > omega(21) = -20.064562 [THz] = -669.286187 [cm-1] > > omega(22) = 15.218497 [THz] = 507.637805 [cm-1] > > omega(23) = 16.277849 [THz] = 542.974191 [cm-1] > > omega(24) = 20.249558 [THz] = 675.457041 [cm-1] > > omega(25) = 21.320279 [THz] = 711.172668 [cm-1] > > omega(26) = 22.104349 [THz] = 737.326626 [cm-1] > > omega(27) = 23.427206 [THz] = 781.452687 [cm-1] > > omega(28) = 30.392570 [THz] = 1013.793763 [cm-1] > > omega(29) = 43.644315 [THz] = 1455.827328 [cm-1] > > omega(30) = 55.006552 [THz] = 1834.833267 [cm-1] > > omega(31) = 57.997854 [THz] = 1934.613035 [cm-1] > > omega(32) = 61.898591 [THz] = 2064.728462 [cm-1] > > omega(33) = 68.862483 [THz] = 2297.020463 [cm-1] > > omega(34) = 68.979827 [THz] = 2300.934652 [cm-1] > > omega(35) = 69.303015 [THz] = 2311.715098 [cm-1] > > omega(36) = 69.697807 [THz] = 2324.884051 [cm-1] > > > > On 11/19/06, *Miguel Mart?nez Canales* < wmbmacam at lg.ehu.es > > > wrote: > > > > Hi Jin, > > > > Jin Zhang escribi?: > > > The thing I cannot understand is, there're 12 negative frequencies > > > existing. I know 3 of them are purely translational modes and > > another 3 > > > are rotational ones. But what about the rest 6? I'm a little > > confused > > > and hopefully someone can help. > > > > > > The eigenvals are pasted below and thanks in advance! > > > > > > (the result may not be accurate due to a only-15Ry ecut I used, > > but it > > > should not give out qualitative different result) > > > > For a start, I wouldn't think 15 rydbergs is enough. Have you > > tested the > > pseudopotentials for ecut convergence? 25 or 30 would probably be > > a better > > choice. Anyway, what you probably are missing most in your > > calculations is > > k-point sampling convergence. Translational and rotational > > frequencies are > > a bit too large. What happens if you increase your k-point grid? > > Anyway, > > don?t forget to relax your molecule with the new parametres. > > > > BTW: What kind of cell are you using? I've never done molecular > > normal mode > > calculations but I guess that in the case of bencene using an > > hexagonal > > supercell might make sense. > > > > -- > > ---------------------------------------- > > Miguel Mart?nez Canales > > Dto. F?sica de la Materia Condensada > > UPV/EHU > > Facultad de Ciencia y Tecnolog?a > > Apdo. 644 > > 48080 Bilbao (Spain) > > Fax: +34 94 601 3500 > > Tlf: +34 94 601 5437 > > ---------------------------------------- > > > > "The problem with Renault is that they dont have > > the skills to convince FIA that the mass damper > > is just a brake cooler." > > > > Annonymous > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > > > > -- > > Jin Zhang > > Dep. of Physics, > > Peking Univ. > > > _______________________________________________ > 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/20061120/cf5f17ef/attachment.htm From gaoguoying at gmail.com Mon Nov 20 13:24:21 2006 From: gaoguoying at gmail.com (Guoying Gao) Date: Mon, 20 Nov 2006 20:24:21 +0800 Subject: [Pw_forum] some questions about structure optimizing Message-ID: <6234c0c60611200424r557929c9vb81a6db66eab40fb@mail.gmail.com> Dear everyone, When I am doing the structure optimizing, I met with the following questions: First, I do the scf calculation with the experimental crystal constants and atomic positions (the phase a is high-pressure one), the output file gives the right symmetry operations and almost the same pressure as the experiment. Then, I do the structure optimizing under the experimental pressure with calculation = vc-relax, cell_dynamics=damp-pr. After the convergence is achieved, we do scf with the optimized parameters, however, the symmetry operations diminished and the output pressure is much higher than that of the experiment. We also tried finding the optimized results which kept symmetry before the convergence is achieved and reoptimized, do scf when the convergence is achieved again, but the results didn't change a lot. We lso tried cell_dynamics=damp-w , different pseudopotentials and the results changed little. Anyone who knows the reason for this will be welcomed. Thanks in advance! gao From lzhuang at whu.edu.cn Mon Nov 20 13:56:18 2006 From: lzhuang at whu.edu.cn (Lin Zhuang) Date: Mon, 20 Nov 2006 20:56:18 +0800 Subject: [Pw_forum] error in parallel run Message-ID: <2b80d0af0611200456i7c575c90q30c835d29bfe7e17@mail.gmail.com> Dear espresso user: I have compiled espresso 3.1.1 with icc/ifort 9.1 & LAM/MPI 7.1.2. When I run the example01, I encountered the following error: MPI_COMM_WORLD: Call stack withn LAM -MPI_Allreduce() -main() I read through this mailing list, somebody said this is a bug only in Intel compiler 9.1, but I also had this error with icc/ifort 9.0. Somebody suggested to modified the f_def.h, I followed the instruction and found pw.x can run successfully in most examples except for those PDOS calculations, but ph.x failed in those examples. What can I do now? Is there a ultimate solution? All the best, Lin Zhuang From gbrandino at gmail.com Tue Nov 21 00:16:47 2006 From: gbrandino at gmail.com (Giuseppe Piero Brandino) Date: Tue, 21 Nov 2006 00:16:47 +0100 Subject: [Pw_forum] error in parallel run In-Reply-To: <2b80d0af0611200456i7c575c90q30c835d29bfe7e17@mail.gmail.com> References: <2b80d0af0611200456i7c575c90q30c835d29bfe7e17@mail.gmail.com> Message-ID: Dear espresso user, i have some questions about the pp.x program. Asking the code to write the potential, for example plot_num=11=the V_bare + V_H potential, does it print the potential or the potential energy? I know that it prints it in Rydberg, so an energy. How do i get the potential? simply dividing by e or dividing by the total (ionic and electronic) charge density in each point? Something else, is there a simple way to get the total charge density (electronic + ionic) from pp.x? Thanks in advance, Giuseppe Brandino -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061121/7a16e7f3/attachment.htm From gbrandino at gmail.com Tue Nov 21 00:26:59 2006 From: gbrandino at gmail.com (Giuseppe Piero Brandino) Date: Tue, 21 Nov 2006 00:26:59 +0100 Subject: [Pw_forum] About postprocessing Message-ID: Dear espresso user, i have some questions about the pp.x program. Asking the code to write the potential, for example plot_num=11=the V_bare + V_H potential, does it print the potential or the potential energy? I know that it prints it in Rydberg, so an energy. How do i get the potential? simply dividing by e or dividing by the total (ionic and electronic) charge density in each point? Something else, is there a simple way to get the total charge density (electronic + ionic) from pp.x? Thanks in advance, Giuseppe Brandino -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061121/7fcc7084/attachment.htm From jzwanzig at dal.ca Tue Nov 21 03:33:49 2006 From: jzwanzig at dal.ca (Josef Zwanziger) Date: Mon, 20 Nov 2006 22:33:49 -0400 Subject: [Pw_forum] efg.x input: a tiny bug? Message-ID: <1164076429.13724.24.camel@jzwanzig-desktop> Dear Colleagues, On exploring efg.x in detail, both with my own examples and example24, I've found the following issue. INPUT_EFG requests the quadrupole moment for each nucleus be entered in units of 10E-30 C/m^2, while the program output repeats the input listed as units of 10E-30 m^2. The second case makes more sense as quadrupole moments are typically given as areas; however, for the example, oxygen-17 is input as 2.55, while the quadrupole moment of oxygen-17 is 2.55 barns = 2.55E-28 m^2, not 2.55E-30 m^2. Nevertheless, the resulting output of about 6 MHz is about right for Cq for SiO2. In efg.x, Cq is computed from the efg principal element v(1), I presume in atomic units, as Cq = v(1)*Q*rytoev*2*ANGSTROM_AU**2*ELECTRONVOLT_SI*1E18/6.6262. EFG's look like energy/(charge * length^2), so in Cq = eq*eQ/h, the eq part will be, in SI units, v(1) * (Joule/Hartree)*(1/electron charge in C)*(Bohr/0.529E-10 m)**2 Here the first two factors together equal rytoev*2, and the length factor is the same as ANGSTROM_AU**2 * 1E20. The eQ part of the calculation should be Q * (electron charge in C) = Q*ELECTRONVOLT_SI * 1E-28, presuming we input Q directly in Barns, as the input of example24 seems to do. Then on multiplying these factors and dividing by h, I get v(1)*(rytoev*2)*(ANGSTROM_AU**2)*Q*ELECTRONVOLT_SI*(1E20*1E-28)/6.6262E-34, and if I throw in 1E-6 to convert from Hz to MHz, the last line becomes v(1)*(rytoev*2)*(ANGSTROM_AU**2)*Q*ELECTRONVOLT_SI*1E20/6.6262, the same as the code except 1E18 -> 1E20. Now, the code as written is right if I input Q in units of 10E-30 m^2, as it appears to want, but then I would have to enter Q as 255 in example 24, not 2.55, and the computed Cq will be 600 MHz, not 6 MHz as it should be. On the other hand, if I alter the code to change the factor from 1E18 to 1E20 (as it seems like it should be), and enter Q in the right units of 2.55 (E-28 m^2), then of course the output will still be 600 MHz, not 6. Anyone tell me what I'm doing wrong? Sorry for the long post but this is a simple thing that doesn't seem as simple as it should be. Thanks, Joe -- Josef W. Zwanziger Prof. of Chemistry and Canada Research Chair in NMR Studies of Materials Director, Atlantic Region Magnetic Resonance Centre Dept. of Chemistry Dalhousie University Halifax, NS B3H 4J3 Canada tel: 902.494.1960 fax: 902.494.1867 email: jzwanzig at dal.ca web: http://jwz.chem.dal.ca From baroni at sissa.it Tue Nov 21 05:14:01 2006 From: baroni at sissa.it (Stefano Baroni) Date: Tue, 21 Nov 2006 05:14:01 +0100 Subject: [Pw_forum] About postprocessing In-Reply-To: References: Message-ID: On Nov 21, 2006, at 12:26 AM, Giuseppe Piero Brandino wrote: > Dear espresso user, > i have some questions about the pp.x program. > Asking the code to write the potential, for example > > plot_num=11=the V_bare + V_H potential, > > does it print the potential or the potential energy? > I know that it prints it in Rydberg, so an energy. right > How do i get the potential? simply dividing by e right > or dividing by the total (ionic and electronic) charge density > in each point? this is a bit confusing to me. why would you divide by the local charge density? an energy divided by a charge density makes a potential times a volume, which is not what you expect, and furthermore I am not aware of any sensible use of the resulting quantity. > Something else, > is there a simple way to get the total charge density (electronic + > ionic) from pp.x? The ionic *pseudo*charge density is ill-defined when the pseudopotential is non local (as it usually the case). On the other hand, the physical (all-electron) valence+core charge density is not simply the sum of the valence pseudo-charge+ionic charge. What would be needed here is the sum of the all-electron ionic charge density + the valence charge density obtained from the reconstructed wave- functions. This is what is done the PAW formalism (see PE Bloechl, Phys. Rev. B 50, 17953 - 17979 (1994)). Work is in progress to implement PAW in Quantum Espresso. Meanwhile, there might be tools available to obtain a (more or less) quick-and-dirty recostruction of the charge, but others may be more appropriate than me to address this point. Stefano Baroni --- 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/20061121/5e1c178b/attachment.htm From s305igun at mail.chem.itb.ac.id Wed Nov 22 04:38:25 2006 From: s305igun at mail.chem.itb.ac.id (Rahmat Gunawan) Date: Wed, 22 Nov 2006 10:38:25 +0700 (WIT) Subject: [Pw_forum] About: the system is metallic, specify occupations? In-Reply-To: References: Message-ID: <1177.167.205.72.36.1164166705.squirrel@webmail.chem.itb.ac.id> Dear PW_Forum... I have tried some examples to know how graphite system can learned in PWscf. I read some article in journals: J.Phys.Chem. B., 2006, 110, 14832 and J.Chem.Phys., 118(3), 1003. I have problem about input, now I consult to PW_Forum my interaction bond input of "graphite---H" &CONTROL calculation = 'scf' , restart_mode = 'from_scratch' , outdir = '/home/s3/s305igun/tmp/' , pseudo_dir = '/home/s3/s305igun/pseudo/' , prefix = 'C' , tstress = .true. , tprnfor = .true. , / &SYSTEM ibrav = 4, celldm(1) = 4.64117, celldm(3) = 2.7264, nat = 5, ntyp = 2, ecutwfc = 25.0 , ecutrho = 100.0 , / &ELECTRONS / ATOMIC_SPECIES H 1.00794 H.vbc.UPF C 12.01100 C.pz-vbc.UPF ATOMIC_POSITIONS crystal H 0.000000000 0.000000000 3.250000000 C 0.000000000 0.000000000 0.000000000 C 0.333333333 0.666666666 0.000000000 C 0.000000000 0.000000000 0.500000000 K_POINTS 10 0.1250000 0.1250000 0.1250000 1.00 0.1250000 0.1250000 0.3750000 3.00 0.1250000 0.1250000 0.6250000 3.00 0.1250000 0.1250000 0.8750000 3.00 0.1250000 0.3750000 0.3750000 3.00 0.1250000 0.3750000 0.6250000 6.00 0.1250000 0.3750000 0.8750000 6.00 0.1250000 0.6250000 0.6250000 3.00 0.3750000 0.3750000 0.3750000 1.00 0.3750000 0.3750000 0.6250000 3.00 I need your help, is it true input? because at output file write: "the system is metallic, specify occupations?" I do not understand about this statements. Thanks very much, succes for your research Sincerely yours. Rahmat gunawan From degironc at sissa.it Wed Nov 22 09:17:20 2006 From: degironc at sissa.it (degironc) Date: Wed, 22 Nov 2006 09:17:20 +0100 Subject: [Pw_forum] About: the system is metallic, specify occupations? In-Reply-To: <1177.167.205.72.36.1164166705.squirrel@webmail.chem.itb.ac.id> References: <1177.167.205.72.36.1164166705.squirrel@webmail.chem.itb.ac.id> Message-ID: <45640790.2060502@sissa.it> A system with an even number of electron per unit cell is an isulator or a metal according to whether or not these electrons (doubly) occupy completely a set of bands separated from the subsequent ones by an energy gap. A system with an odd number of particles can't possibly do that (if it is not magnetic). In PWscf when dealing with a metal or a magnetic system you need to specify how to describe partial occupation of energy bands. Relevant input variables are: occupation, smearing, degauss in the system namelist best, stefano PS: please note that in crystal coordinates H 0.000000000 0.000000000 3.250000000 is equivalent to H 0.000000000 0.000000000 0.250000000 is this what you want ??? Rahmat Gunawan wrote: >Dear PW_Forum... > >I have tried some examples to know how graphite system can learned in >PWscf. I read some article in journals: J.Phys.Chem. B., 2006, 110, 14832 >and J.Chem.Phys., 118(3), 1003. I have problem about input, now I consult >to PW_Forum my interaction bond input of "graphite---H" > >&CONTROL > calculation = 'scf' , > restart_mode = 'from_scratch' , > outdir = '/home/s3/s305igun/tmp/' , > pseudo_dir = '/home/s3/s305igun/pseudo/' , > prefix = 'C' , > tstress = .true. , > tprnfor = .true. , > / > &SYSTEM > ibrav = 4, > celldm(1) = 4.64117, > celldm(3) = 2.7264, > nat = 5, > ntyp = 2, > ecutwfc = 25.0 , > ecutrho = 100.0 , > / > &ELECTRONS > / >ATOMIC_SPECIES > H 1.00794 H.vbc.UPF > C 12.01100 C.pz-vbc.UPF >ATOMIC_POSITIONS crystal > H 0.000000000 0.000000000 3.250000000 > C 0.000000000 0.000000000 0.000000000 > C 0.333333333 0.666666666 0.000000000 > C 0.000000000 0.000000000 0.500000000 >K_POINTS > 10 > 0.1250000 0.1250000 0.1250000 1.00 > 0.1250000 0.1250000 0.3750000 3.00 > 0.1250000 0.1250000 0.6250000 3.00 > 0.1250000 0.1250000 0.8750000 3.00 > 0.1250000 0.3750000 0.3750000 3.00 > 0.1250000 0.3750000 0.6250000 6.00 > 0.1250000 0.3750000 0.8750000 6.00 > 0.1250000 0.6250000 0.6250000 3.00 > 0.3750000 0.3750000 0.3750000 1.00 > 0.3750000 0.3750000 0.6250000 3.00 > >I need your help, is it true input? because at output file write: "the >system is metallic, specify occupations?" I do not understand about this >statements. > >Thanks very much, succes for your research > >Sincerely yours. > >Rahmat gunawan >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum > > From Giovanni.Cantele at na.infn.it Wed Nov 22 09:11:10 2006 From: Giovanni.Cantele at na.infn.it (Giovanni Cantele) Date: Wed, 22 Nov 2006 09:11:10 +0100 Subject: [Pw_forum] About: the system is metallic, specify occupations? In-Reply-To: <1177.167.205.72.36.1164166705.squirrel@webmail.chem.itb.ac.id> References: <1177.167.205.72.36.1164166705.squirrel@webmail.chem.itb.ac.id> Message-ID: <4564061E.5090409@na.infn.it> Rahmat Gunawan wrote: > I need your help, is it true input? because at output file write: "the > system is metallic, specify occupations?" I do not understand about this > statements. > > Thanks very much, succes for your research > > > You may want to have a look to the user guide, pag. 56 (sec. Troubleshooting) http://www.pwscf.org/guide/3.1.1/other/users-guide-3.1.1.pdf to the Doc/INPUT_PW file, as well as to examples/example01 where a couple of examples on how to deal with metallic systems are given. 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 baettigp at gmx.net Wed Nov 22 10:09:53 2006 From: baettigp at gmx.net (=?iso-8859-1?Q?=22Pio_B=E4ttig=22?=) Date: Wed, 22 Nov 2006 10:09:53 +0100 Subject: [Pw_forum] Compile problem on a Hitachi SR11000, AIX Message-ID: <20061122090953.27770@gmx.net> Dear PWscf-users and -authors, I am trying to compile PWscf 3.1.1 on a Hitachi SR11000 (IBM Power5 1.9GHz). (It compiled without problems on a PC under linux using the intel compilers). Unfortunately, just using the standard configure-script did not work and stopped with the following messages: ---- espresso-3.1.1/> ./configure checking build system type... powerpc-ibm-aix5.2.0.0 checking architecture... aix checking for mpxlf90_r... mpxlf90_r checking for Fortran 77 compiler default output file name... configure: error: Fortran 77 compiler cannot create executables See `config.log' for more details. ---- (the correct fortran-compiler should be mpxlf90) I then set the following environment variables: ---- export FC=mpixf90 export F77=mpixf90 export F90=mpixf90 export CC=mpicc export FFLAGS="-Os -64 -precexp=4 -nompi -noparallel -qfree=f90 -cpp" export LDFLAGS="-64 -noparallel -L/usr/local/lib -lblas_s -llapack_s -noparallel" export BLAS_LIBS="-lblas_s" export LAPACK_LIBS="-llapack_s" ---- which generated the following make.sys-file: ---- espresso-3.1.1/> more make.sys # make.sys. Generated from make.sys.in by configure. # compilation rules .SUFFIXES : .SUFFIXES : .o .c .f .f90 .f90.o: $(CPP) $(CPPFLAGS) $< -o $*.F90 $(MPIF90) $(F90FLAGS) -c $*.F90 -o $*.o .f.o: $(F77) $(FFLAGS) -c $< .c.o: $(CC) $(CFLAGS) -c $< CC = mpicc MPICC = cc CFLAGS = -O $(DFLAGS) $(IFLAGS) CPP = mpicc -E CPPFLAGS = $(DFLAGS) $(IFLAGS) F90 = mpixf90 MPIF90 = mpixf90 F90FLAGS = $(FFLAGS) $(FDFLAGS) $(IFLAGS) $(MODFLAGS) F90FLAGS_NOOPT = $\(FFLAGS_NOOPT) $(FDFLAGS) $(IFLAGS) $(MODFLAGS) F77 = mpixf90 MPIF77 = mpixf90 FFLAGS = -Os -64 -precexp=4 -nompi -noparallel -qfree=f90 -cpp FFLAGS_NOOPT = -O0 LD = mpixf90 LDFLAGS = -64 -noparallel -L/usr/local/lib -lblas_s -llapack_s -noparallel AR = ar ARFLAGS = -X64 ruv RANLIB = echo BLAS_LIBS = -lblas_s LAPACK_LIBS = -llapack_s 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__AIX -D__FFTW -D__USE_INTERNAL_FFTW -D__MPI -D__PARA FDFLAGS = $(DFLAGS) IFLAGS = -I../include MODFLAGS = -I. -I../Modules -I../PW -I../PH -I../iotk/src -I../CPV 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 = ---- After deleting line 9 and changing 10 to the following: ---- $(MPIF90) $(F90FLAGS) -c $< ---- I tried to compile with 'make all'. Unfortunately, I got a lot of ---- f90: compile start : iotk_base.f90 *OFORT90 V01-03-/A entered. KCHF444K 12 55 an operand is missing. KCHF444K 12 56 an operand is missing. ---- for the iotk* files. (the errors were: KCHF278K 12 "The reference to a generic procedure invokes two or more specific procedures with the same interface." - 72 occurences KCHF419K 12 "The number of arguments or their type is invalid in a reference to a generic name." - 58 occurrences KCHF474K 04 "The variable is referenced, but is never defined." - 3 occurrences KCHF475K 00 "The variable is defined, but is never referenced." -no problem - 86 occurrences KCHF476K 00 "The variable is declared, but is never used in an executable statement." -no problem - 251 occurrences KCHF656K "File xxx is referenced from the source statement file library." - 293 occurrences Error-code explanations from http://www.scl.pjwstk.edu.pl/hitachi/manual/6a203311/m0190126.htm ) I tried to add the line #define __IOTK_WORKAROUND5 after the last #endif of iotk/include/iotk_config.h (last line) as has been suggested in http://www.democritos.it/pipermail/pw_forum/2006-September/004901.html and tried 'make clean; make all' but got still exactly the same errors. the only files that have been generated by the compilation are ./iotk/src/iotk_attr.o and ./iotk/src/iotk_dat.o . I would be grateful if anyone could give me a hint on how to get PWscf working. If any more information is needed, or the full compile-log, I will provide it. Thank you very much in advance, Pio Baettig P.S. The machine runs on AIX 5.2.0, -- Der GMX SmartSurfer hilft bis zu 70% Ihrer Onlinekosten zu sparen! Ideal f?r Modem und ISDN: http://www.gmx.net/de/go/smartsurfer From s305igun at mail.chem.itb.ac.id Wed Nov 22 10:17:42 2006 From: s305igun at mail.chem.itb.ac.id (Rahmat Gunawan) Date: Wed, 22 Nov 2006 16:17:42 +0700 (WIT) Subject: [Pw_forum] About: the system is metallic, specify occupations? In-Reply-To: <45640790.2060502@sissa.it> References: <1177.167.205.72.36.1164166705.squirrel@webmail.chem.itb.ac.id> <45640790.2060502@sissa.it> Message-ID: <1461.167.205.72.36.1164187062.squirrel@webmail.chem.itb.ac.id> Dear PW_Forum Thanks to Stefano it is good for me. I try your suggestion to build my input Thanks again Sincerely Yours Rahmat Gunawan > A system with an even number of electron per unit cell is an isulator or > a metal > according to whether or not these electrons (doubly) occupy completely a > set of bands > separated from the subsequent ones by an energy gap. > A system with an odd number of particles can't possibly do that (if it > is not magnetic). > In PWscf when dealing with a metal or a magnetic system you need to > specify how > to describe partial occupation of energy bands. > Relevant input variables are: occupation, smearing, degauss in the > system namelist > best, > stefano > > PS: please note that in crystal coordinates > > H 0.000000000 0.000000000 3.250000000 > > is equivalent to > > H 0.000000000 0.000000000 0.250000000 > > is this what you want ??? > > > > Rahmat Gunawan wrote: > >>Dear PW_Forum... >> >>I have tried some examples to know how graphite system can learned in >>PWscf. I read some article in journals: J.Phys.Chem. B., 2006, 110, 14832 >>and J.Chem.Phys., 118(3), 1003. I have problem about input, now I consult >>to PW_Forum my interaction bond input of "graphite---H" >> >>&CONTROL >> calculation = 'scf' , >> restart_mode = 'from_scratch' , >> outdir = '/home/s3/s305igun/tmp/' , >> pseudo_dir = '/home/s3/s305igun/pseudo/' , >> prefix = 'C' , >> tstress = .true. , >> tprnfor = .true. , >> / >> &SYSTEM >> ibrav = 4, >> celldm(1) = 4.64117, >> celldm(3) = 2.7264, >> nat = 5, >> ntyp = 2, >> ecutwfc = 25.0 , >> ecutrho = 100.0 , >> / >> &ELECTRONS >> / >>ATOMIC_SPECIES >> H 1.00794 H.vbc.UPF >> C 12.01100 C.pz-vbc.UPF >>ATOMIC_POSITIONS crystal >> H 0.000000000 0.000000000 3.250000000 >> C 0.000000000 0.000000000 0.000000000 >> C 0.333333333 0.666666666 0.000000000 >> C 0.000000000 0.000000000 0.500000000 >>K_POINTS >> 10 >> 0.1250000 0.1250000 0.1250000 1.00 >> 0.1250000 0.1250000 0.3750000 3.00 >> 0.1250000 0.1250000 0.6250000 3.00 >> 0.1250000 0.1250000 0.8750000 3.00 >> 0.1250000 0.3750000 0.3750000 3.00 >> 0.1250000 0.3750000 0.6250000 6.00 >> 0.1250000 0.3750000 0.8750000 6.00 >> 0.1250000 0.6250000 0.6250000 3.00 >> 0.3750000 0.3750000 0.3750000 1.00 >> 0.3750000 0.3750000 0.6250000 3.00 >> >>I need your help, is it true input? because at output file write: "the >>system is metallic, specify occupations?" I do not understand about this >>statements. >> >>Thanks very much, succes for your research >> >>Sincerely yours. >> >>Rahmat gunawan >>_______________________________________________ >>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 s305igun at mail.chem.itb.ac.id Wed Nov 22 10:29:19 2006 From: s305igun at mail.chem.itb.ac.id (Rahmat Gunawan) Date: Wed, 22 Nov 2006 16:29:19 +0700 (WIT) Subject: [Pw_forum] About: the system is metallic, specify occupations? In-Reply-To: <4564061E.5090409@na.infn.it> References: <1177.167.205.72.36.1164166705.squirrel@webmail.chem.itb.ac.id> <4564061E.5090409@na.infn.it> Message-ID: <1517.167.205.72.36.1164187759.squirrel@webmail.chem.itb.ac.id> Drea PW_Forum Thanks to Giovanni for your suggestion, it is clear to me. Success for your research Sincerely Yours Rahmat Gunawan > Rahmat Gunawan wrote: >> I need your help, is it true input? because at output file write: "the >> system is metallic, specify occupations?" I do not understand about this >> statements. >> >> Thanks very much, succes for your research >> >> >> > You may want to have a look to the user guide, pag. 56 (sec. > Troubleshooting) > > http://www.pwscf.org/guide/3.1.1/other/users-guide-3.1.1.pdf > > to the Doc/INPUT_PW file, as well as to examples/example01 where a > couple of examples on how to deal with metallic systems are given. > > 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 > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From guido.fratesi at unimi.it Wed Nov 22 09:59:30 2006 From: guido.fratesi at unimi.it (Guido Fratesi) Date: Wed, 22 Nov 2006 09:59:30 +0100 (CET) Subject: [Pw_forum] Re: About: the system is metallic, specify occupations? In-Reply-To: <20061122063754.24267.45272.Mailman@democritos.sissa.it> References: <20061122063754.24267.45272.Mailman@democritos.sissa.it> Message-ID: > ATOMIC_POSITIONS crystal > H 0.000000000 0.000000000 3.250000000 > C 0.000000000 0.000000000 0.000000000 > C 0.333333333 0.666666666 0.000000000 > C 0.000000000 0.000000000 0.500000000 You have an odd number N of electrons per unit cell, so your system IS metallic, and one cannot simply occupy the lowest N/2 bands as would be the default for an insulator. Hence you have to specify how to fill the electron bands. See file Doc/INPUT_PW, variables "occupation", "degauss", "smearing" and references given there for details. [If you want to think more about that, consider an H atom in a large box, suitable for an isolated-atom calculation. Then you have 1 electron per unit cell, which is odd again, so the system made of the repeated H atoms far away the one from the other is metallic, and you will find with PWscf that the lowest energy is given by half-filling a band which is nothing more than the (bonding) combination of the 1s orbitals, as it should be. Moreover, the results will not be much affected by the values you choose for the variables specified above. Can you imagine why?] Hope this helps, Guido Guido Fratesi Dipartimento di Fisica Universita` degli studi di Milano Via Celoria, 16 - 20133 MI (Italy) From gbrandino at gmail.com Wed Nov 22 11:14:58 2006 From: gbrandino at gmail.com (Giuseppe Piero Brandino) Date: Wed, 22 Nov 2006 11:14:58 +0100 Subject: [Pw_forum] About postprocessing In-Reply-To: References: Message-ID: Thank you sir for the fast reply, Giuseppe Brandino On 11/21/06, Stefano Baroni wrote: > > > On Nov 21, 2006, at 12:26 AM, Giuseppe Piero Brandino wrote: > > Dear espresso user, > i have some questions about the pp.x program. > Asking the code to write the potential, for example > > plot_num=11=the V_bare + V_H potential, > > does it print the potential or the potential energy? > I know that it prints it in Rydberg, so an energy. > > > > right > > How do i get the potential? simply dividing by e > > > > > right > > or dividing by the total (ionic and electronic) charge density > in each point? > > > > this is a bit confusing to me. why would you divide by the local charge > density? an energy divided by a charge density makes a potential times a > volume, which is not what you expect, and furthermore I am not aware of any > sensible use of the resulting quantity. > > Something else, > is there a simple way to get the total charge density (electronic + ionic) > from pp.x? > > > > The ionic *pseudo*charge density is ill-defined when the pseudopotential > is non local (as it usually the case). On the other hand, the physical > (all-electron) valence+core charge density is not simply the sum of the > valence pseudo-charge+ionic charge. What would be needed here is the sum of > the all-electron ionic charge density + the valence charge density obtained > from the reconstructed wave-functions. This is what is done the PAW > formalism (see PE Bloechl, Phys. Rev. B 50, 17953 - 17979 (1994)). Work is > in progress to implement PAW in Quantum Espresso. Meanwhile, there might be > tools available to obtain a (more or less) quick-and-dirty recostruction of > the charge, but others may be more appropriate than me to address this > point. > > > Stefano Baroni > > --- > 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/20061122/bada6a89/attachment.htm From katalin.gaal-nagy at physik.uni-regensburg.de Wed Nov 22 12:28:00 2006 From: katalin.gaal-nagy at physik.uni-regensburg.de (Katalin Gaal-Nagy) Date: Wed, 22 Nov 2006 12:28:00 +0100 (CET) Subject: [Pw_forum] subroutine: star_q Message-ID: Dear all, I try to calculate for a given system the q points which are equvalent to a given one. For this purpose I try to use the routine star_q given in the ph directory of the quantum espresso package (espresso-3.1): subroutine star_q (xq, at, bg, ibrav, symm_type, nat, tau, ityp, & nr1, nr2, nr3, nsym, s, invs, irt, rtau, nq, sxq, isq, imq, noinv, & modenum) Most of the arguments are clear for me, I just have some problems in understanding the input parameters noinv and modenum. The parameter noinv sets the inversion symmetry of the system, thus, it should be set to .true. in the case there is no inversion symmetry, am I right? The parameter modenum is the number of modes to be done, thus, in case of the calculation of all phonon modes of the crystal, it should be 3*nat where nat is the number of atoms in the basis. Did I understood it correctly? Another question (just for curiosity): what will change for the q in star(q), if modenum<3*nat, e.g., modenum=0? Of course, it is not nice to set noinv "by hand", therefore it would be nice if someone could give me a hint which subroutine I could utilize for determining this parameter ... if not, I will search in the pw code and the pw directory ... Thanks for your help! Best wishes, Katalin From giannozz at nest.sns.it Wed Nov 22 13:08:33 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Wed, 22 Nov 2006 13:08:33 +0100 Subject: [Pw_forum] Compile problem on a Hitachi SR11000, AIX In-Reply-To: <20061122090953.27770@gmx.net> References: <20061122090953.27770@gmx.net> Message-ID: <805F9E59-8DCB-4F88-BBF4-7A1B45713989@nest.sns.it> On Nov 22, 2006, at 10:09 , Pio B?ttig wrote: > (the correct fortran-compiler should be mpxlf90) "./configure F90=mpxlf90" should do the job > I then set the following environment variables: > > ---- > export FC=mpixf90 is it mpxlf90 or mpixf90? the name of the compiler is important for configure. > FDFLAGS = $(DFLAGS) for the IBM xlf compiler you should have something like > FDFLAGS = -D__AIX,-D__FFTW,-D__USE_INTERNAL_FFTW, -D__MPI,- > D__PARA i.e. with commas instead of spaces. Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From s305igun at mail.chem.itb.ac.id Wed Nov 22 14:09:40 2006 From: s305igun at mail.chem.itb.ac.id (Rahmat Gunawan) Date: Wed, 22 Nov 2006 20:09:40 +0700 (WIT) Subject: [Pw_forum] Re: About: the system is metallic, specify occupations? In-Reply-To: References: <20061122063754.24267.45272.Mailman@democritos.sissa.it> Message-ID: <1652.167.205.72.36.1164200980.squirrel@webmail.chem.itb.ac.id> Dear PW_Forum Thanks to Guido... this is help me. now I try again input and it is work. Thanks again Success for your research Sincerely Yours. Rahmat Gunawan PostGraduate Student Dept. of Chemistry Institut Teknologi Bandung West Java, Indonesia >> ATOMIC_POSITIONS crystal >> H 0.000000000 0.000000000 3.250000000 >> C 0.000000000 0.000000000 0.000000000 >> C 0.333333333 0.666666666 0.000000000 >> C 0.000000000 0.000000000 0.500000000 > > You have an odd number N of electrons per unit cell, so your system IS > metallic, and one cannot simply occupy the lowest N/2 bands as would be > the default for an insulator. > Hence you have to specify how to fill the electron bands. > See file Doc/INPUT_PW, variables "occupation", "degauss", "smearing" and > references given there for details. > > [If you want to think more about that, consider an H atom in a large box, > suitable for an isolated-atom calculation. Then you have 1 electron per > unit cell, which is odd again, so the system made of the repeated H > atoms far away the one from the other is metallic, and you will find > with PWscf that the lowest energy is given by half-filling a band which > is nothing more than the (bonding) combination of the 1s orbitals, as it > should be. Moreover, the results will not be much affected by the values > you choose for the variables specified above. Can you imagine why?] > > Hope this helps, > Guido > > Guido Fratesi > Dipartimento di Fisica > Universita` degli studi di Milano > Via Celoria, 16 - 20133 MI (Italy) > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From luch001 at 126.com Wed Nov 22 14:42:45 2006 From: luch001 at 126.com (luch001) Date: Wed, 22 Nov 2006 21:42:45 +0800 (CST) Subject: [Pw_forum] questions on lattice dynamical principle Message-ID: <456453D5.00007F.11433@bj126app5.126.com> Dear all: I have read the paper titled ??Theory of systematic absence of NaCl-type (??-Sn-type) high pressure phases in covalent (ionic) semiconductors?? (Phys Rev Lett. 82, 767, 1999). The phonon behaviors are dominated by the competition between the attractive electrostatic Madelung energy ?CZ2/R (where Z is the excess charge within the atomic sphere) and the repulsive short-range energy.I can not understand well the formula of the force constant given in the last part of page 769. I thought to present the formula here. Unfortunately, I can not add it in my email text. I am very sorry for that. How does it come from? Does it apply to all the systems or just apply to some specific system? The values of A and ?? are determined by what factors? The authors claimed that the transverse acoustic mode at X point lowers the repulsive energy and raises the Madelung energy, but why? I have pondered hardly and consulted with others. But I have reached no conclusion. Any reply will be appreciated! With all my best regards! Yours sincerely Jingyun Zhang -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061122/75714047/attachment.htm From babazade at gmail.com Wed Nov 22 15:36:04 2006 From: babazade at gmail.com (Amin Babazadeh) Date: Wed, 22 Nov 2006 18:06:04 +0330 Subject: [Pw_forum] use doual core Message-ID: dear users I use a computer with a dual core CPU.but my problem is that when i run a program only one of the CPU work but i like to share the work (solving k points) between cpus. I think that i should install the parallel pwscf but when i run the ./configure only single pwscf installed. would you please tell me or give me a document that solve my problem. -- AMIN -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061122/45a5922c/attachment.htm From akohlmey at cmm.chem.upenn.edu Wed Nov 22 16:34:33 2006 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Wed, 22 Nov 2006 10:34:33 -0500 Subject: [Pw_forum] use doual core In-Reply-To: References: Message-ID: <7b6913e90611220734s15038d2ei36e1e7a28166f146@mail.gmail.com> On 11/22/06, Amin Babazadeh wrote: > dear users > I use a computer with a dual core CPU.but my problem is that when i run a > program only one of the CPU work but i like to share the work (solving k > points) between cpus. > I think that i should install the parallel pwscf but when i run the > ./configure only single pwscf installed. you have to install an MPI package (e.g. LAM/MPI, OpenMPI, MPICH) first and then configure/compile quantum espresso with parallel support as explained in the provided installation instructions. please make sure that your MPI package is compatible with or compiled for the fortran compiler you want to use to compiled quantum espresso with. precompiled packages may be compiled for a different compiler and thus need special flags or environment variables set to use the compiler of your choice. axel. > would you please tell me or give me a document that solve my problem. > > > -- > AMIN -- ======================================================================= 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 degironc at sissa.it Wed Nov 22 16:59:05 2006 From: degironc at sissa.it (degironc) Date: Wed, 22 Nov 2006 16:59:05 +0100 Subject: [Pw_forum] questions on lattice dynamical principle In-Reply-To: <456453D5.00007F.11433@bj126app5.126.com> References: <456453D5.00007F.11433@bj126app5.126.com> Message-ID: <456473C9.6090201@sissa.it> good questions... why don't you ask the authors ? stefano luch001 wrote: > Dear all: > > I have read the paper titled ??Theory of systematic absence of > NaCl-type (??-Sn-type) high pressure phases in covalent (ionic) > semiconductors?? (Phys Rev Lett. *82*, 767, 1999). The phonon behaviors > are dominated by the competition between the attractive electrostatic > Madelung energy ?CZ^2 /R (where Z is the excess charge within the > atomic sphere) and the repulsive short-range energy.I can not > understand well the formula of the force constant given in the last > part of page 769. I thought to present the formula here. > Unfortunately, I can not add it in my email text. I am very sorry for > that. > > How does it come from? Does it apply to all the systems or just apply > to some specific system? The values of A and ?? are determined by what > factors? > > The authors claimed that the transverse acoustic mode at X point > lowers the repulsive energy and raises the Madelung energy, but why? > > I have pondered hardly and consulted with others. But I have reached > no conclusion. > > Any reply will be appreciated! > > With all my best regards! > > Yours sincerely > > Jingyun Zhang > > > > > > > > > K a p p a \ A N D 1 \ R e e b o k ?? ?? ?? 9 9 ?? > ?? ?? \ ?? ?? \ ?? ?? ?? ?? ?? ?? ?? ?? ?? 1 9 9 ?? ?? ?? , ?? ?? ?? ?? > ?? ! > From jzwanzig at dal.ca Wed Nov 22 17:30:53 2006 From: jzwanzig at dal.ca (Josef W. Zwanziger) Date: Wed, 22 Nov 2006 12:30:53 -0400 Subject: [Pw_forum] Pseudopotentials for efg.x Message-ID: <001c01c70e53$975987d0$8311ad81@jwzpc5> Dear Colleagues, Can someone post the input files that were used to make the pseudopotentials used in example24 of the examples distribution? This is the example that uses efg.x to compute the electric field gradients for SiO2. The example uses the pseudopotential files SiPBE_nc.UPF and OPBE_nc.UPF, and the reconstruction files Si.recon and O.recon. EFG.X requires the use of norm-conserving pseudopotentials, and the UPF files seem to suggest that the ones used are single-projector pseudopotentials. Using the supplied files, I get about 6 MHz for O-17 in SiO2, which is reasonable, but I cannot reproduce anything like this using pseudopotentials I make with the ld1 code. The ones I make from ld1 seem in other respects quite good (good agreement between log derivatives for ae wf's and ps wf's, good agreement of eigenvalues, better than 1 meV, and for example using my Si ncpp and PBE, the lattice constant of Si optimizes to 0.5% longer than experiment, which seems reasonable). But the efg's that result from these pseudopotentials are terrible. It would help to see in detail how the ones used in the example were made. While I'm on the subject of efg.x, can anyone comment on the input units for Q, the subject of my last post? The data is requested in 10^-30 m^2, but the example seems to be using barns, which is 10^-28 m^2. What's going on here? Thanks much, Joe Josef W. Zwanziger Prof. of Chemistry Canada Research Chair in NMR Studies of Materials Director, Atlantic Region Magnetic Resonance Centre Dalhousie University Halifax, NS B3H 4J3 Canada tel: 902.494.1960 net: jzwanzig at dal.ca fax: 902.494.1867 web: http://jwz.chem.dal.ca From giannozz at nest.sns.it Wed Nov 22 18:04:25 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Wed, 22 Nov 2006 18:04:25 +0100 Subject: [Pw_forum] Pseudopotentials for efg.x In-Reply-To: <001c01c70e53$975987d0$8311ad81@jwzpc5> References: <001c01c70e53$975987d0$8311ad81@jwzpc5> Message-ID: <6A36632E-F145-4045-BBC1-1BD23CB9F6AF@nest.sns.it> On Nov 22, 2006, at 17:30 , Josef W. Zwanziger wrote: > While I'm on the subject of efg.x, can anyone comment on the > input units for Q, the subject of my last post? The data is > requested in 10^-30 m^2, but the example seems to be using > barns, which is 10^-28 m^2. What's going on here? well, the person who wrote that code left physics some time ago. All the available information is, to the best of my knowledge, contained in the code itself, in the input documentation Doc/INPUT_EFG, and in example 24. If you find something that doesn't look correct, it might actually be incorrect. Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From luch001 at 126.com Thu Nov 23 02:26:26 2006 From: luch001 at 126.com (luch001) Date: Thu, 23 Nov 2006 09:26:26 +0800 (CST) Subject: [Pw_forum] problems with lattice dynamical principle Message-ID: <4564F8C2.00004C.32373@bj126app4.126.com> Dear all, I have read the paper titled ??Theory of systematic absence of NaCl-type (??-Sn-type) high pressure phases in covalent (ionic) semiconductors?? (Phys Rev Lett. 82, 767, 1999). The phonon behaviors are dominated by the competition between the attractive electrostatic Madelung energy ?CZ2/R (where Z is the excess charge within the atomic sphere) and the repulsive short-range energy. I can not understand well the formula of the force constant given in the last part of page 769, which reads: ?????? = nA??????/Rn+2-Z2??????/R3-??R??R??/ R5 How does it come from? Does it apply to all the systems or just apply to some specific system? The values of A and ?? are determined by what factors? The authors claimed that the transverse acoustic mode at X point lowers the repulsive energy and raises the Madelung energy, but why? I have pondered hardly and consulted with others. But I have reached no conclusion. I wrote letter to Alex Zunger, one of the two authors. He did not write back yet. Any reply will be appreciated! With all my best regards! Yours sincerely Jingyun Zhang -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061123/c9c403ed/attachment.htm From hqzhou at nju.edu.cn Thu Nov 23 09:46:22 2006 From: hqzhou at nju.edu.cn (Huiqun Zhou) Date: Thu, 23 Nov 2006 16:46:22 +0800 Subject: [Pw_forum] Comparison of 3.1.1 and 3.2 (cvs) Message-ID: <01d801c70edb$dbce5e30$1d00a8c0@solarflare> Dear list users, 10 days ago, I checked out the cvs version and compared the speed of both versions against the same test case on same machine. pwscf was compiled using Intel Fortran 9.0 for x86_64, MPI was Intel MPI-2 2.0, just the same environment for generating the results of my previous tests on AMD opteron 285, Intel Xeon 5060, 5140, 5150 platforms that I posted to this forum. This test was carried on a Intel 5050 machine (dempsey 3.0GHz/667MHz, 4GB memory). As my previous tests, the test case is still the one to calculate total energy of MgAl2O4 with calcium ferric structure, and 28 atoms in the unit cell. Here are the numbers: (1) V3.1.1 1 core : 8m10s (8m09.52s) 2 cores: 4m57s (4m52.95s) 4 cores: 2m23s (2m17.68s) (2) V3.2 (cvs, about 10 days ago) 1 core : 9m29.95s (9m11.09s) 2 cores: 6m03.24s (5m26.85s) 4 cores: 2m35.26s (2m04.52s) There are two problems here: (1) The differences between wall time and CPU time with version 3.1.1 is small and seems reasonable while that between the two with cvs version is relatively large. Why? (2) Version 3.1.1 is much faster than the cvs version as the former took only 13 scf iterations to reach convergence while the latter needed 16 iterations. Of course, it's just a cvs version and development is in progress. I hope that developers have already noticed the difference. Thanks to all developers for their great effort! Huiqun Zhou From baroni at sissa.it Thu Nov 23 10:07:49 2006 From: baroni at sissa.it (Stefano Baroni) Date: Thu, 23 Nov 2006 10:07:49 +0100 Subject: [Pw_forum] Comparison of 3.1.1 and 3.2 (cvs) In-Reply-To: <01d801c70edb$dbce5e30$1d00a8c0@solarflare> References: <01d801c70edb$dbce5e30$1d00a8c0@solarflare> Message-ID: This shorte note just to say that I do appreciate this kind of reports. Thanks. I do not have any specific comments, but I am sure that this note will be taken in due account in the further development of the code. I wish that others will take the time to report similar problems, if any. Of course, the more specific and thorough the report is, the more useful it will be. Thanks again - Stefano On Nov 23, 2006, at 9:46 AM, Huiqun Zhou wrote: > Dear list users, > > 10 days ago, I checked out the cvs version and compared > the speed of both versions against the same test case on > same machine. > > pwscf was compiled using Intel Fortran 9.0 for x86_64, MPI > was Intel MPI-2 2.0, just the same environment for generating > the results of my previous tests on AMD opteron 285, Intel > Xeon 5060, 5140, 5150 platforms that I posted to this forum. > > This test was carried on a Intel 5050 machine (dempsey 3.0GHz/ > 667MHz, 4GB memory). As my previous tests, the test case is still > the one to calculate total energy of MgAl2O4 with calcium ferric > structure, and 28 atoms in the unit cell. Here are the numbers: > > (1) V3.1.1 > 1 core : 8m10s (8m09.52s) > 2 cores: 4m57s (4m52.95s) > 4 cores: 2m23s (2m17.68s) > > (2) V3.2 (cvs, about 10 days ago) > 1 core : 9m29.95s (9m11.09s) > 2 cores: 6m03.24s (5m26.85s) > 4 cores: 2m35.26s (2m04.52s) > > There are two problems here: > (1) The differences between wall time and CPU time with version > 3.1.1 is small and seems reasonable while that between the two > with cvs version is relatively large. Why? > (2) Version 3.1.1 is much faster than the cvs version as the former > took only 13 scf iterations to reach convergence while the latter > needed 16 iterations. > > Of course, it's just a cvs version and development is in progress. > I hope that developers have already noticed the difference. > > Thanks to all developers for their great effort! > > > Huiqun Zhou > > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061123/1248f337/attachment.htm From giannozz at nest.sns.it Thu Nov 23 10:22:23 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 23 Nov 2006 10:22:23 +0100 Subject: [Pw_forum] subroutine: star_q In-Reply-To: References: Message-ID: <1D70C38E-45F7-4050-812E-1F3E5A179781@nest.sns.it> On Nov 22, 2006, at 12:28 , Katalin Gaal-Nagy wrote: > I try to calculate for a given system the q points which are > equivalent > to a given one. For this purpose I try to use the routine star_q given > in the ph directory of the quantum espresso package [...] the star_q routine is actually quite complicated. There are several other places where this is done in the code; one should be in PW/kpoint_grid.f90, another pwtools/kpoints.f > The parameter noinv sets the inversion symmetry of the system well, not exactly. "noinv" when .true. removes the effects of the symmetry sending z into -z. This was an input variable used for special purposes in cases (i.e. surfaces) but it is never read in the recent versions. You can safely set it to .false. > The parameter modenum is the number of modes to be done no, it is the index of a specific phonon mode, used for calculations of a single mode. Just set it to 0. Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From giannozz at nest.sns.it Thu Nov 23 12:26:00 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 23 Nov 2006 12:26:00 +0100 Subject: [Pw_forum] Comparison of 3.1.1 and 3.2 (cvs) In-Reply-To: <01d801c70edb$dbce5e30$1d00a8c0@solarflare> References: <01d801c70edb$dbce5e30$1d00a8c0@solarflare> Message-ID: <6CAFF00A-7453-453D-AEF0-19F42DED1025@nest.sns.it> On Nov 23, 2006, at 9:46 , Huiqun Zhou wrote: > There are two problems here: > (1) The differences between wall time and CPU time with version > 3.1.1 is small and seems reasonable while that between the two > with cvs version is relatively large. Why? maybe because they are not the same quantity in the two cases. There have been some changes to the CPU and wall time counters in the cvs version with respect to the previous one. Note that printed cpu and wall times may not always be what they are intended to be, depending on the compiler. On at least some versions of the Intel compiler, a call to "standard" routine cpu_time returns the wall time, for instance. > (2) Version 3.1.1 is much faster than the cvs version as the former > took only 13 scf iterations to reach convergence while the latter > needed 16 iterations. the number of scf iteration may occasionally be sensitive to minor numerical changes. There hasn't been any major change to the self-consistency algorithm, so I don't expect any major or generalized change in the overall performances. > Of course, it's just a cvs version and development is in progress. > I hope that developers have already noticed the difference. personally I haven't noticed anything: benchmarking is not my preferred activity (I would even say, like a famous physicist about chemistry: "nothing I am more allergic to". CPU and wall times are irreproducible on modern machines: I have several times noticed variations of 10 to 15% with no apparent reason). > Thanks to all developers for their great effort! thank to you for your report. I hope that somebody will have a look at it and use it to improve performances. Note however that unless a clear reason (if any) is found for the performance degradation you observed, it will be hard to do any better Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From katalin.gaal-nagy at physik.uni-regensburg.de Thu Nov 23 13:05:17 2006 From: katalin.gaal-nagy at physik.uni-regensburg.de (Katalin Gaal-Nagy) Date: Thu, 23 Nov 2006 13:05:17 +0100 (CET) Subject: [Pw_forum] subroutine: star_q In-Reply-To: <1D70C38E-45F7-4050-812E-1F3E5A179781@nest.sns.it> References: <1D70C38E-45F7-4050-812E-1F3E5A179781@nest.sns.it> Message-ID: Dear Paolo, thank you for your reply! >> I try to calculate for a given system the q points which are equivalent >> to a given one. For this purpose I try to use the routine star_q given >> in the ph directory of the quantum espresso package [...] > the star_q routine is actually quite complicated. There are > several other places where this is done in the code; one > should be in PW/kpoint_grid.f90, another pwtools/kpoints.f I had the hope to get the equivalent q points in tha same ordering as in, e.q., the files with the dynamical matrices. That was the reason to take the routine star_q, since the ordering and, sometimes, also the sign of the componets is different for q sets calculated with pwtools/kpoints.f and PW/kpoint_grid.f90. Anyway. >> The parameter noinv sets the inversion symmetry of the system > well, not exactly. "noinv" when .true. removes the effects of the > symmetry sending z into -z. This was an input variable used for > special purposes in cases (i.e. surfaces) but it is never read in > the recent versions. You can safely set it to .false. Bene! I will do it. >> The parameter modenum is the number of modes to be done > no, it is the index of a specific phonon mode, used for > calculations of a single mode. Just set it to 0. I did it. Doing a small test it turned out that I get the same q points like in some dynmat-file, but with a different ordering. I am a little bit wondering, since the procedure is the same as in the ph.x program. But as long as it works for my (bct) system, I am fine. As soon as I have time I will do some checks also for systems with other lattices, and as soon as I have everything in a clean, documented way, I will give it to the pwscf (or better quantum espresso) community. Thank you for your help! Best wishes, Katalin From wyanchao at gmail.com Thu Nov 23 13:45:11 2006 From: wyanchao at gmail.com (wang yanchao) Date: Thu, 23 Nov 2006 20:45:11 +0800 Subject: [Pw_forum] question Message-ID: <55639d30611230445q4f2e9062ufe37afd98c0d230f@mail.gmail.com> Dear users: when I used run_pp to caculate the densety of charge,I found a qustion that the number of points are 55 55 55 in output file.but in fact what I input the nx,ny,nz is100,100,100.then I read the code of PP,I found that 55 55 55 is nr1+1 nr2+1 nr3+1.I think "nr" is the K_point.In fact I ,however,caculated the K_point is gamma.This make me confused.In an addtion ,I found the value of nx,ny nz is limited in code of PP.the max value is 64.I do not know the reson why it choose this value.If you know ,please tell me .Thank you in advance. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061123/d4589782/attachment.htm From degironc at sissa.it Thu Nov 23 14:09:42 2006 From: degironc at sissa.it (degironc) Date: Thu, 23 Nov 2006 14:09:42 +0100 Subject: [Pw_forum] problems with lattice dynamical principle In-Reply-To: <4564F8C2.00004C.32373@bj126app4.126.com> References: <4564F8C2.00004C.32373@bj126app4.126.com> Message-ID: <45659D96.90000@sissa.it> MY explanation of what happens is the following: Just take a simple model of nearest neighbhor interaction... I know coulomb interaction is long range but in TA(X) mode you are moving back and forth laterally (100) planes of NaCl, these planes are neutral on average so let us forget about what happens beyond the first shell. Let's have a coulomb attractive part -Z^2/R between (any) atom of a (100) plane and its nearest neighbhor in the adjacent plane (which happens to be of opposite charge) and a short range repulsive part A/R^n again only between nearest neighbhors. Then calculate the restoring force for this nearest neighbhor interaction: you move atom (1) in the central plane and calculate the negative of the force induced on an atom (2) in the nearby plane: Phi (1,alpha;2,beta) = - Z^2/R^3 (delta_{alpha,beta} - 3 R_alpha R_beta / R^2) + n A/ R^n+2 (delta_{alpha,beta} - (n+2) R_alpha R_beta / R^2) with symmetry arguments and sum rules you get Phi(1,alpha;1beta) = -2 * Phi(1,alpha;2,beta) and similarly for the other cases... At X motion of nearest neighbor planes are out of phase so the total restoring force for such a phonon is -4 * Phi(1,alpha;2,beta) or Phi(X)_{alpha,beta} = +4 [ Z^2/R^3 (delta_{alpha,beta} - 3 R_alpha R_beta / R^2) - n A/ R^n+2 (delta_{alpha,beta} - (n+2) R_alpha R_beta / R^2) ] Now for TA modes and just nearest neighbhor interaction displacement is orthogonal to R so forget about R_alpha R_beta term. You are left with Phi( X, TA) = +4 [ Z^2/R^3 - n A/ R^n+2 ] As you apply pressure, the negative (unstable) term coming from the short range repulsive interaction grows more rapidly (because n>>1) than the positive (stabilizing) Couomb term thus leading to instability. The higher the ionicity (the larger Z) of the system the larger will be the pressure needed to destabilize it (assuming similar short range term) .... At least this is the way I saw the situation when I discussed with Alex Zunger the results of their calculations in Aspen back in 1998... Again, if you want genuine author's explanation you should ask them. stefano luch001 wrote: > Dear all, > > I have read the paper titled ??Theory of systematic absence of > NaCl-type (??-Sn-type) high pressure phases in covalent (ionic) > semiconductors?? (Phys Rev Lett. *82*, 767, 1999). The phonon behaviors > are dominated by the competition between the attractive electrostatic > Madelung energy ?CZ^2 /R (where Z is the excess charge within the > atomic sphere) and the repulsive short-range energy. I can not > understand well the formula of the force constant given in the last > part of page 769, which reads: > > ??_???? = nA??_???? /R^n+2 -Z^2 ??_???? /R^3 -??R_?? R_?? / R^5 > > How does it come from? Does it apply to all the systems or just apply > to some specific system? The values of A and ?? are determined by what > factors? > > The authors claimed that the transverse acoustic mode at X point > lowers the repulsive energy and raises the Madelung energy, but why? > > I have pondered hardly and consulted with others. But I have reached > no conclusion. I wrote letter to Alex Zunger, one of the two authors. > He did not write back yet. > > Any reply will be appreciated! > > With all my best regards! > > Yours sincerely > > Jingyun Zhang > From konstantin_kudin at yahoo.com Thu Nov 23 18:42:30 2006 From: konstantin_kudin at yahoo.com (Konstantin Kudin) Date: Thu, 23 Nov 2006 09:42:30 -0800 (PST) Subject: [Pw_forum] Comparison of 3.1.1 and 3.2 (cvs) In-Reply-To: <01d801c70edb$dbce5e30$1d00a8c0@solarflare> Message-ID: <151818.2436.qm@web57709.mail.re3.yahoo.com> --- Huiqun Zhou wrote: > (2) Version 3.1.1 is much faster than the cvs version as the former > took only 13 scf iterations to reach convergence while the latter > needed 16 iterations. Dear Huiqun, If the number of the SCF iterations is the same regardless of the number of cpus for both versions, then please post some kind of convergence pattern like for total energies. That would give developers a chance to evaluate if any of the recent changes might have caused this. Sometimes small numerical noise is a benign "feature", and sometimes it is a bug, so figuring out what is going on for your test would be useful. Also, perhaps you could try a different benchmark to see how things work there, if there are any differences found there as well. Kostya ____________________________________________________________________________________ Sponsored Link Compare mortgage rates for today. Get up to 5 free quotes. Www2.nextag.com From degironc at sissa.it Thu Nov 23 22:03:45 2006 From: degironc at sissa.it (degironc) Date: Thu, 23 Nov 2006 22:03:45 +0100 Subject: [Pw_forum] question In-Reply-To: <55639d30611230445q4f2e9062ufe37afd98c0d230f@mail.gmail.com> References: <55639d30611230445q4f2e9062ufe37afd98c0d230f@mail.gmail.com> Message-ID: <45660CB0.9080504@sissa.it> dear wang yanchao, sorry I've some difficulty to understand your problem. could you please repeat your question telling more accurately what are you trying to do, which code are you trying to use, with what input, and what confuses you ? stefano wang yanchao wrote: > Dear users: > when I used run_pp to caculate the densety of charge,I found a > qustion that the number of points are 55 55 55 in output file.but in > fact what I input the nx,ny,nz is100,100,100.then I read the code of > PP,I found that 55 55 55 is nr1+1 nr2+1 nr3+1.I think "nr" is the > K_point.In fact I ,however,caculated the K_point is gamma.This make me > confused.In an addtion ,I found the value of nx,ny nz is limited in > code of PP.the max value is 64.I do not know the reson why it choose > this value.If you know ,please tell me .Thank you in advance. From akohlmey at cmm.chem.upenn.edu Thu Nov 23 22:33:52 2006 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Thu, 23 Nov 2006 16:33:52 -0500 (EST) Subject: [Pw_forum] question In-Reply-To: <55639d30611230445q4f2e9062ufe37afd98c0d230f@mail.gmail.com> Message-ID: On Thu, 23 Nov 2006, wang yanchao wrote: WY> Dear users: WY> when I used run_pp to caculate the densety of charge,I found a qustion WY> that the number of points are 55 55 55 in output file.but in fact what I WY> input the nx,ny,nz is100,100,100.then I read the code of PP,I found that 55 WY> 55 55 is nr1+1 nr2+1 nr3+1.I think "nr" is the K_point.In fact I nr1,nr2,nr3 (and its siblings nr1s,nr2s,nr3s and nr1b,nr2b,nr3b) are traditionally the grid dimensions of the real space meshes on which the density is stored (e.g. for evaluating the functionals or computing forces). see, e.g., Modules/input_parameters.f90 this is independent of your k-point selection. WY> ,however,caculated the K_point is gamma.This make me confused.In an addtion WY> ,I found the value of nx,ny nz is limited in code of PP.the max value is WY> 64.I do not know the reson why it choose this value.If you know ,please tell WY> me .Thank you in advance. unless you need a specific real space resolution, you should stick with the grid dimensions used in your pw.x calculations. for tighter (or coarser) grids the code interpolates via fourier transforms (as the real space density is usually generated from the g-space representation when needed). hope that helps, axel. WY> -- ======================================================================= 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 hqzhou at nju.edu.cn Fri Nov 24 08:25:13 2006 From: hqzhou at nju.edu.cn (Huiqun Zhou) Date: Fri, 24 Nov 2006 15:25:13 +0800 Subject: [Pw_forum] Comparison of 3.1.1 and 3.2 (cvs) References: <151818.2436.qm@web57709.mail.re3.yahoo.com> Message-ID: <001b01c70f99$b0e8fae0$1d00a8c0@solarflare> Kostya, The number of SCF iterations is indeed the same regardless of the number of CPU cores used for both versions. Furthermore, the process of SCF convergence is precisely the same whatever we use 1, 2 or 4 CPU cores. This is true for both versions. And the final total energy found by both version is the same. But, the behaviour of convergence of one version compared to another is different if you watch the changes of numbers after decimal point. It seems that the criteria of convergence of each version are different. I have created three graphs for the pattern of convergence on 1, 2, 4 cores (but they are tottally the same as mentioned above). Each graph contains a comparison of patterns of the two versions. Please let me know where I can post the graphs, each of which is about 55KB, as I don't think this forum permit posting a file with such large size. Here I first include the six data files extracted from the results. you can plot the graph yourself, too. Please give me a pointer what kind of benchmarks you want me to run. Do you think it's worth tring the same system with different type of run (relax, ...), different method of diagonalization? Currently, all results I reported made use of 'david' method for diagonalization. Regards, Huiqun ----- Original Message ----- From: "Konstantin Kudin" To: Sent: Friday, November 24, 2006 1:42 AM Subject: Re: [Pw_forum] Comparison of 3.1.1 and 3.2 (cvs) > > --- Huiqun Zhou wrote: >> (2) Version 3.1.1 is much faster than the cvs version as the former >> took only 13 scf iterations to reach convergence while the latter >> needed 16 iterations. > > Dear Huiqun, > > If the number of the SCF iterations is the same regardless of the > number of cpus for both versions, then please post some kind of > convergence pattern like for total energies. That would give developers > a chance to evaluate if any of the recent changes might have caused > this. Sometimes small numerical noise is a benign "feature", and > sometimes it is a bug, so figuring out what is going on for your test > would be useful. > > Also, perhaps you could try a different benchmark to see how things > work there, if there are any differences found there as well. > > Kostya > > > > ____________________________________________________________________________________ > Sponsored Link > > Compare mortgage rates for today. > Get up to 5 free quotes. > Www2.nextag.com > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -------------- next part -------------- A non-text attachment was scrubbed... Name: test.zip Type: application/octet-stream Size: 1465 bytes Desc: not available Url : /pipermail/attachments/20061124/dc7071b9/attachment.obj From degironc at sissa.it Fri Nov 24 09:39:13 2006 From: degironc at sissa.it (degironc) Date: Fri, 24 Nov 2006 09:39:13 +0100 Subject: [Pw_forum] Comparison of 3.1.1 and 3.2 (cvs) In-Reply-To: <001b01c70f99$b0e8fae0$1d00a8c0@solarflare> References: <151818.2436.qm@web57709.mail.re3.yahoo.com> <001b01c70f99$b0e8fae0$1d00a8c0@solarflare> Message-ID: <4566AFB1.2010603@sissa.it> Dear Huiqun Zhou, a couple of changes have been made in the mixing routine (mainly to reduce its huge memory requirement in certain cases) from 3.1.1 to 3.2. These might be the reason for the different behaviors. Could you please post your input so that we can try to fgure out what happened ? Since every realization of an algorithm is the result of a compromise between speed, memory use, work required to implement it, etc and since every compromise will likely be more or less good depending on the particular application considered, a comparison for different "typical" cases would also be useful in order to assess the "average" performance (or loss of it) from 3.1.1 to 3.2 Thank you, stefano Huiqun Zhou wrote: > Kostya, > > The number of SCF iterations is indeed the same regardless of the > number of CPU cores used for both versions. Furthermore, the process > of SCF convergence is precisely the same whatever we use 1, 2 or 4 > CPU cores. This is true for both versions. And the final total energy > found > by both version is the same. But, the behaviour of convergence of one > version compared to another is different if you watch the changes of > numbers after decimal point. It seems that the criteria of convergence of > each version are different. > > I have created three graphs for the pattern of convergence on 1, 2, 4 > cores > (but they are tottally the same as mentioned above). Each graph contains > a comparison of patterns of the two versions. Please let me know where > I can post the graphs, each of which is about 55KB, as I don't think this > forum permit posting a file with such large size. > > Here I first include the six data files extracted from the results. > you can > plot the graph yourself, too. > > Please give me a pointer what kind of benchmarks you want me to run. > Do you think it's worth tring the same system with different type of > run (relax, ...), > different method of diagonalization? Currently, all results I reported > made use > of 'david' method for diagonalization. > > Regards, > > Huiqun > > > ----- Original Message ----- From: "Konstantin Kudin" > > To: > Sent: Friday, November 24, 2006 1:42 AM > Subject: Re: [Pw_forum] Comparison of 3.1.1 and 3.2 (cvs) > > >> >> --- Huiqun Zhou wrote: >> >>> (2) Version 3.1.1 is much faster than the cvs version as the former >>> took only 13 scf iterations to reach convergence while the latter >>> needed 16 iterations. >> >> >> Dear Huiqun, >> >> If the number of the SCF iterations is the same regardless of the >> number of cpus for both versions, then please post some kind of >> convergence pattern like for total energies. That would give developers >> a chance to evaluate if any of the recent changes might have caused >> this. Sometimes small numerical noise is a benign "feature", and >> sometimes it is a bug, so figuring out what is going on for your test >> would be useful. >> >> Also, perhaps you could try a different benchmark to see how things >> work there, if there are any differences found there as well. >> >> Kostya >> >> >> >> ____________________________________________________________________________________ >> >> Sponsored Link >> >> Compare mortgage rates for today. >> Get up to 5 free quotes. >> Www2.nextag.com >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> From hqzhou at nju.edu.cn Fri Nov 24 10:56:01 2006 From: hqzhou at nju.edu.cn (Huiqun Zhou) Date: Fri, 24 Nov 2006 17:56:01 +0800 Subject: [Pw_forum] Comparison of 3.1.1 and 3.2 (cvs) References: <151818.2436.qm@web57709.mail.re3.yahoo.com> <001b01c70f99$b0e8fae0$1d00a8c0@solarflare> <4566AFB1.2010603@sissa.it> Message-ID: <003101c70fae$c01116a0$1d00a8c0@solarflare> Stefano, Here are the input file and a short script to run tests. Huiqun ----- Original Message ----- From: "degironc" To: Sent: Friday, November 24, 2006 4:39 PM Subject: Re: [Pw_forum] Comparison of 3.1.1 and 3.2 (cvs) > Dear Huiqun Zhou, > a couple of changes have been made in the mixing routine (mainly to > reduce its > huge memory requirement in certain cases) from 3.1.1 to 3.2. These might > be the reason for the different behaviors. Could you please post your > input so that we can try to fgure out what happened ? > Since every realization of an algorithm is the result of a compromise > between speed, > memory use, work required to implement it, etc and since every compromise > will likely > be more or less good depending on the particular application considered, a > comparison for > different "typical" cases would also be useful in order to assess the > "average" performance > (or loss of it) from 3.1.1 to 3.2 > Thank you, > stefano > > Huiqun Zhou wrote: > >> Kostya, >> >> The number of SCF iterations is indeed the same regardless of the >> number of CPU cores used for both versions. Furthermore, the process >> of SCF convergence is precisely the same whatever we use 1, 2 or 4 >> CPU cores. This is true for both versions. And the final total energy >> found >> by both version is the same. But, the behaviour of convergence of one >> version compared to another is different if you watch the changes of >> numbers after decimal point. It seems that the criteria of convergence of >> each version are different. >> >> I have created three graphs for the pattern of convergence on 1, 2, 4 >> cores >> (but they are tottally the same as mentioned above). Each graph contains >> a comparison of patterns of the two versions. Please let me know where >> I can post the graphs, each of which is about 55KB, as I don't think this >> forum permit posting a file with such large size. >> >> Here I first include the six data files extracted from the results. you >> can >> plot the graph yourself, too. >> >> Please give me a pointer what kind of benchmarks you want me to run. >> Do you think it's worth tring the same system with different type of run >> (relax, ...), >> different method of diagonalization? Currently, all results I reported >> made use >> of 'david' method for diagonalization. >> >> Regards, >> >> Huiqun >> >> >> ----- Original Message ----- From: "Konstantin Kudin" >> >> To: >> Sent: Friday, November 24, 2006 1:42 AM >> Subject: Re: [Pw_forum] Comparison of 3.1.1 and 3.2 (cvs) >> >> >>> >>> --- Huiqun Zhou wrote: >>> >>>> (2) Version 3.1.1 is much faster than the cvs version as the former >>>> took only 13 scf iterations to reach convergence while the latter >>>> needed 16 iterations. >>> >>> >>> Dear Huiqun, >>> >>> If the number of the SCF iterations is the same regardless of the >>> number of cpus for both versions, then please post some kind of >>> convergence pattern like for total energies. That would give developers >>> a chance to evaluate if any of the recent changes might have caused >>> this. Sometimes small numerical noise is a benign "feature", and >>> sometimes it is a bug, so figuring out what is going on for your test >>> would be useful. >>> >>> Also, perhaps you could try a different benchmark to see how things >>> work there, if there are any differences found there as well. >>> >>> Kostya >>> >>> >>> >>> ____________________________________________________________________________________ >>> Sponsored Link >>> >>> Compare mortgage rates for today. >>> Get up to 5 free quotes. >>> Www2.nextag.com >>> _______________________________________________ >>> Pw_forum mailing list >>> Pw_forum at pwscf.org >>> http://www.democritos.it/mailman/listinfo/pw_forum >>> > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -------------- next part -------------- A non-text attachment was scrubbed... Name: input.zip Type: application/octet-stream Size: 1073 bytes Desc: not available Url : /pipermail/attachments/20061124/a9e20d74/attachment.obj From hqzhou at nju.edu.cn Fri Nov 24 11:22:02 2006 From: hqzhou at nju.edu.cn (Huiqun Zhou) Date: Fri, 24 Nov 2006 18:22:02 +0800 Subject: [Pw_forum] Comparison of 3.1.1 and 3.2 (cvs) References: <151818.2436.qm@web57709.mail.re3.yahoo.com> <001b01c70f99$b0e8fae0$1d00a8c0@solarflare> <4566AFB1.2010603@sissa.it> Message-ID: <00cd01c70fb2$62300ce0$1d00a8c0@solarflare> Hi, Here is the graph for the convergence patterns of v3.1.1 and v3.2. I have reduced the size to about 13KB. Huiqun ----- Original Message ----- From: "degironc" To: Sent: Friday, November 24, 2006 4:39 PM Subject: Re: [Pw_forum] Comparison of 3.1.1 and 3.2 (cvs) > Dear Huiqun Zhou, > a couple of changes have been made in the mixing routine (mainly to > reduce its > huge memory requirement in certain cases) from 3.1.1 to 3.2. These might > be the reason for the different behaviors. Could you please post your > input so that we can try to fgure out what happened ? > Since every realization of an algorithm is the result of a compromise > between speed, > memory use, work required to implement it, etc and since every compromise > will likely > be more or less good depending on the particular application considered, a > comparison for > different "typical" cases would also be useful in order to assess the > "average" performance > (or loss of it) from 3.1.1 to 3.2 > Thank you, > stefano > > Huiqun Zhou wrote: > >> Kostya, >> >> The number of SCF iterations is indeed the same regardless of the >> number of CPU cores used for both versions. Furthermore, the process >> of SCF convergence is precisely the same whatever we use 1, 2 or 4 >> CPU cores. This is true for both versions. And the final total energy >> found >> by both version is the same. But, the behaviour of convergence of one >> version compared to another is different if you watch the changes of >> numbers after decimal point. It seems that the criteria of convergence of >> each version are different. >> >> I have created three graphs for the pattern of convergence on 1, 2, 4 >> cores >> (but they are tottally the same as mentioned above). Each graph contains >> a comparison of patterns of the two versions. Please let me know where >> I can post the graphs, each of which is about 55KB, as I don't think this >> forum permit posting a file with such large size. >> >> Here I first include the six data files extracted from the results. you >> can >> plot the graph yourself, too. >> >> Please give me a pointer what kind of benchmarks you want me to run. >> Do you think it's worth tring the same system with different type of run >> (relax, ...), >> different method of diagonalization? Currently, all results I reported >> made use >> of 'david' method for diagonalization. >> >> Regards, >> >> Huiqun >> >> >> ----- Original Message ----- From: "Konstantin Kudin" >> >> To: >> Sent: Friday, November 24, 2006 1:42 AM >> Subject: Re: [Pw_forum] Comparison of 3.1.1 and 3.2 (cvs) >> >> >>> >>> --- Huiqun Zhou wrote: >>> >>>> (2) Version 3.1.1 is much faster than the cvs version as the former >>>> took only 13 scf iterations to reach convergence while the latter >>>> needed 16 iterations. >>> >>> >>> Dear Huiqun, >>> >>> If the number of the SCF iterations is the same regardless of the >>> number of cpus for both versions, then please post some kind of >>> convergence pattern like for total energies. That would give developers >>> a chance to evaluate if any of the recent changes might have caused >>> this. Sometimes small numerical noise is a benign "feature", and >>> sometimes it is a bug, so figuring out what is going on for your test >>> would be useful. >>> >>> Also, perhaps you could try a different benchmark to see how things >>> work there, if there are any differences found there as well. >>> >>> Kostya >>> >>> >>> >>> ____________________________________________________________________________________ >>> Sponsored Link >>> >>> Compare mortgage rates for today. >>> Get up to 5 free quotes. >>> Www2.nextag.com >>> _______________________________________________ >>> Pw_forum mailing list >>> Pw_forum at pwscf.org >>> http://www.democritos.it/mailman/listinfo/pw_forum >>> > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -------------- next part -------------- A non-text attachment was scrubbed... Name: convergency_4cores-2.png Type: image/png Size: 12763 bytes Desc: not available Url : /pipermail/attachments/20061124/436d4293/attachment.png From emenendez at macul.ciencias.uchile.cl Fri Nov 24 13:22:41 2006 From: emenendez at macul.ciencias.uchile.cl (Eduardo Ariel Menendez P) Date: Fri, 24 Nov 2006 09:22:41 -0300 (CLST) Subject: [Pw_forum] Re: Comparison of 3.1.1 and 3.2 (cvs) Message-ID: Hello, I posted this yesterday, but it seems to have gone to /dev/null. I insist because I think it is important to benchmark with a "large" job. ********************************** I would like to add that the scaling of jobs that take a few minutes of CPU is totally different to jobs of a few hours. For instance, for a small calculation with 1 or 2 atoms using the pw.x compiled and linked to the Intel MKL library makes no difference with compiling the source, at least if the Intel compiler is used. However, for a job of 64 atoms that runs in 2 hours or more, linking either to MKL or to the compiled from source BLAS/LAPCK, the difference can be a factor of 5. This is an example /home/eduardo/Chemutils/Espresso/espresso-cvs/espresso/bin-ifort-serial-fftwmkl/pw.x > cdte.scf10.out real 149m1.400s user 147m39.509s sys 1m21.829s /home/eduardo/Chemutils/Espresso/espresso-cvs/espresso/bin-ifort-serial/pw.x > cdte.scf10.out real 627m25.528s user 626m4.072s sys 1m22.537s Using the fftw of MKL and using the source makes no difference, at least I have not found any improvement, but the BLAS/LAPACK is important. In summary, do benchmark for large systems, because the scaling of a small system is not important by itself, and it cannot be extrapolated (I cannot say why). Eduardo From Giovanni.Cantele at na.infn.it Fri Nov 24 14:58:59 2006 From: Giovanni.Cantele at na.infn.it (Giovanni Cantele) Date: Fri, 24 Nov 2006 14:58:59 +0100 Subject: [Pw_forum] Re: Comparison of 3.1.1 and 3.2 (cvs) In-Reply-To: References: Message-ID: <4566FAA3.2080703@na.infn.it> Eduardo Ariel Menendez P wrote: > Hello, > I posted this yesterday, but it seems to have gone to /dev/null. > I insist because I think it is important to benchmark with a "large" job. > ********************************** > I would like to add that the scaling of jobs that take a few minutes of > CPU is totally different to jobs of a few hours. For instance, for a small > calculation with 1 or 2 atoms using the pw.x compiled and linked to > the Intel MKL library makes no > difference with compiling the source, at least if the Intel compiler is > used. However, for a job of 64 > atoms that runs in 2 hours or more, linking either to MKL or to the > compiled from source BLAS/LAPCK, the difference can be a factor of 5. > This is an example > /home/eduardo/Chemutils/Espresso/espresso-cvs/espresso/bin-ifort-serial-fftwmkl/pw.x > > cdte.scf10.out > real 149m1.400s > user 147m39.509s > sys 1m21.829s > /home/eduardo/Chemutils/Espresso/espresso-cvs/espresso/bin-ifort-serial/pw.x > > cdte.scf10.out > real 627m25.528s > user 626m4.072s > sys 1m22.537s > Using the fftw of MKL and using the source makes no difference, at least I > have not found any improvement, but the BLAS/LAPACK is important. In > summary, do benchmark for large systems, because the scaling of a small > system is not important by itself, and it cannot be extrapolated (I > cannot say why). > Eduardo > Hi, I was just wondering why my experience gave exactly the opposite results as the one posted yesterday, namely CVS (3.2) version faster than 3.1.1. My run was on 8 alpha ev7 CPUs, 1150 MHz. I found CPU time of 2h7m with 3.1.1, 1h23m (1h26m wall time) using cvs. Please note that in 3.1.1 parallel diagonalization was used, whereas in cvs it wasn't! Does it make sense? The runs was relaxation of an Sr-terminated SrTiO3-110 surface, involving 19 atoms (152 electrons): ibrav=8 celldm(1) = 7.37 celldm(2) = 1.41421356 celldm(3) = 6.717514421 ... ecutwfc=30.0 ecutrho=180.0 ... K_POINTS { automatic } 4 4 1 1 1 0 (due to symmetry the calculation reduces to 4 k-points) I've tried to make the same test as the one reported in the forum (mgal2o4-cf.scf.in) and got the following results (CPU time / wall time): #CPU 3.1.1 CVS 1 9m58 s/ 10m12s 11m02s / 11m23s 2 5m34s / 5m46s 7m03s / 7m12s 4 2m54s / 2m60s 3m32s / 3m38s In this case CVS is slower than 3.1.1, indeed. 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 jibiaoli at gmail.com Sat Nov 25 08:12:27 2006 From: jibiaoli at gmail.com (Clark Lee) Date: Sat, 25 Nov 2006 15:12:27 +0800 Subject: [Pw_forum] help needed with projwfc.x In-Reply-To: <57185.202.41.111.151.1157465481.squirrel@202.41.111.151> References: <57185.202.41.111.151.1157465481.squirrel@202.41.111.151> Message-ID: The same question. Anyone else knows what's the sequence of the d-orbitals (dxy,dxz,dyz,dx2-y2,dz2) are assigned in results from projwfc.x On 9/5/06, aarti srirangarajan wrote: > > Dear PWSCF developers, > > Could anyone please explain the output of the program projwfc.x? > My question is how the values for m = 1,2,3,4,5 are assigned ( m should be > -2 to +2) ? And in the output files for the PDOS for d-orbitals, what is > the sequence of the data listed in it.(in terms of dxy,dxz,dyz etc.). > Thanking you in anticipation > Aarti > TSU,JNCASR > -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061125/16164c92/attachment.htm From giannozz at nest.sns.it Sat Nov 25 08:40:29 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Sat, 25 Nov 2006 08:40:29 +0100 Subject: [Pw_forum] help needed with projwfc.x In-Reply-To: References: <57185.202.41.111.151.1157465481.squirrel@202.41.111.151> Message-ID: On Nov 25, 2006, at 8:12 , Clark Lee wrote: > The same question. > Anyone else knows what's the sequence of the d-orbitals > (dxy,dxz,dyz,dx2-y2,dz2) are assigned in results from projwfc.x the same answer, from: http://www.democritos.it/pipermail/pw_forum/2004-July/001148.html --- the order of m components of Y_{lm} is determined by routine "ylmr2": 1, cos(phi), sin(phi), cos(2*phi), sin(2*phi), .., cos(m*phi), sin (m*phi) where phi is the polar angle (x=r cos(theta)cos(phi), y=r cos(theta) sin(phi)) --- Paolo > On 9/5/06, aarti srirangarajan wrote: Dear > PWSCF developers, > > Could anyone please explain the output of the program projwfc.x? > My question is how the values for m = 1,2,3,4,5 are assigned ( m > should be > -2 to +2) ? And in the output files for the PDOS for d-orbitals, > what is > the sequence of the data listed in it.(in terms of dxy,dxz,dyz etc.). > Thanking you in anticipation > Aarti > TSU,JNCASR --- Paolo Giannozzi, Democritos and University of Udine, Italy From degironc at sissa.it Sat Nov 25 08:45:04 2006 From: degironc at sissa.it (Stefano de Gironcoli) Date: Sat, 25 Nov 2006 08:45:04 +0100 (CET) Subject: [Pw_forum] help needed with projwfc.x In-Reply-To: References: <57185.202.41.111.151.1157465481.squirrel@202.41.111.151> Message-ID: the sequence is: 1 dz2 (m=0) 2 dzx (real combination of m=+/-1 with cosine) 3 dzy (real combination of m=+/-1 with sine) 4 dx2-y2 (real combination of m=+/-2 with cosine) 5 dxy (real combination of m=+/-1 with sine) similarly for p orbitals 1 pz (m=0) 2 px (real combination of m=+/-1 with cosine) 3 py (real combination of m=+/-1 with sine) stefano On Sat, 25 Nov 2006, Clark Lee wrote: > The same question. > Anyone else knows what's the sequence of the d-orbitals > (dxy,dxz,dyz,dx2-y2,dz2) are assigned in results from projwfc.x > > > On 9/5/06, aarti srirangarajan wrote: >> >> Dear PWSCF developers, >> >> Could anyone please explain the output of the program projwfc.x? >> My question is how the values for m = 1,2,3,4,5 are assigned ( m should be >> -2 to +2) ? And in the output files for the PDOS for d-orbitals, what is >> the sequence of the data listed in it.(in terms of dxy,dxz,dyz etc.). >> Thanking you in anticipation >> Aarti >> TSU,JNCASR >> > From frank at caltech.edu Sat Nov 25 10:19:09 2006 From: frank at caltech.edu (Francesco Ciucci) Date: Sat, 25 Nov 2006 01:19:09 -0800 Subject: [Pw_forum] intel mac --- compilation Message-ID: <28A3A706-DA30-45BC-BC74-AD1EA2070D00@caltech.edu> Hi all, I am trying to build QuantumEspresso 3.1.1 on my MacBook (the little white one). I have as fortran compilers: g77, f2c and gfortran. I run configure as follows: ./configure F77=g77 MPIF90=gfortran CC=gcc F90=gfortran BLAS_LIBS="- framework accelerate" LAPACK_LIBS=" " CFLAGS=-O3 and I get at the end "configure: success". Although, when I run "sudo make all" I get what I reported in PS (sorry it's lengthy). Could anyone help me out? Thank you, Fra P.S. test -d bin || mkdir bin if test -d iotk ; then \ ( cd iotk ; if test "make" = "" ; then make TLDEPS= libiotk.a ; \ else make TLDEPS= libiotk.a ; fi ) ; fi cd src ; make libiotk.a cpp -P -traditional -D__FFTW -D__USE_INTERNAL_FFTW -I../include iotk_base.f90 -o iotk_base.F90 ! Input/Output Tool Kit (IOTK) ! Copyright (C) 2004,2005 Giovanni Bussi ! ! This library is free software; you can redistribute it and/or ! modify it under the terms of the GNU Lesser General Public ! License as published by the Free Software Foundation; either ! version 2.1 of the License, or (at your option) any later version. ! ! This library is distributed in the hope that it will be useful, ! but WITHOUT ANY WARRANTY; without even the implied warranty of ! MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU ! Lesser General Public License for more details. ! ! You should have received a copy of the GNU Lesser General Public ! License along with this library; if not, write to the Free Software ! Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA !----------------------------------------------------------------------- -------! ! Inclusion of configuration file ! Input/Output Tool Kit (IOTK) ! Copyright (C) 2004 Giovanni Bussi ! ! This library is free software; you can redistribute it and/or ! modify it under the terms of the GNU Lesser General Public ! License as published by the Free Software Foundation; either ! version 2.1 of the License, or (at your option) any later version. ! ! This library is distributed in the hope that it will be useful, ! but WITHOUT ANY WARRANTY; without even the implied warranty of ! MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU ! Lesser General Public License for more details. ! ! You should have received a copy of the GNU Lesser General Public ! License along with this library; if not, write to the Free Software ! Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA ! !----------------------------------------------------------------------- -------! ! CONFIGURATION FILE FOR IOTK 1.0.1 !----------------------------------------------------------------------- -------! ! The following lines map some commonly defined system macro to the internal ! iotk macros. ! Iotk macros which are not defined take their default values. ! See the manual for a list of iotk macros. ! Generic options valid for quantum-espresso ! QE uses ranks up to four and default integer/logicals only ! some compilers do not like the following ! #define __IOTK_REAL1 selected_real_kind(6,30) ! #define __IOTK_REAL2 selected_real_kind(14,200) ! so we use explicit kinds ! Machine-dependent options ! Only for compilers that require some special tricks !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! Inclusion of the auxiliary macros ! The macros are defined with -D option or inside iotk_config.h ! The default values are set here ! Maximum rank of an array ! Minimum value used in iotk_free_unit ! Maximum value used in iotk_free_unit ! Unit for errors ! Kind for header in binary files ! Character (or eventually string) for newline ! It may be adjusted for particular systems ! Unix achar(10) ! Mac-OS achar(13) ! Windows ? (now it should be a single byte) ! Character for EOS ! These are the default kinds, which depend on the options used ! during the library compilation ! Only default characters are implemented ! For logical, integer and real types, the c precompiler ! looks for defined kinds. If no kind is found, the default ! is used as __IOTK_type1 ! Some useful check follow ! Complex are treated indentically to reals ! These lines map the definitions. !----------------------------------------------------------------------- -------! module iotk_base implicit none save !----------------------------------------------------------------------- -------! ! In this module, all names are public ! For this reason, it should not be used directly by the end user. !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! Version strings and integer constants character(5), parameter :: iotk_version = "1.0.1" integer, parameter :: iotk_version_major = 1 integer, parameter :: iotk_version_minor = 0 integer, parameter :: iotk_version_patch = 1 character(3), parameter :: iotk_file_version = "1.0" integer, parameter :: iotk_file_version_major = 1 integer, parameter :: iotk_file_version_minor = 0 !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! Special characters character, parameter :: iotk_newline = achar(10) character, parameter :: iotk_eos = achar(0) !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! Max number of controls integer, parameter :: iotk_ncontrol = 255 ! (2**8-1) !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! Max lengths for strings integer, parameter :: iotk_taglenx = 65535 ! (2**16-1) integer, parameter :: iotk_namlenx = 256 integer, parameter :: iotk_attlenx = iotk_taglenx - iotk_namlenx - 1 ! for space integer, parameter :: iotk_vallenx = 32768 integer, parameter :: iotk_linlenx = 4096 integer, parameter :: iotk_fillenx = 1024 integer, parameter :: iotk_linlen = 128 integer, parameter :: iotk_indent = 2 integer, parameter :: iotk_maxindent = 12 !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! Kind for the header integer (number of digits in (iotk_ncontrol+1)* (iotk_taglenx+1)) integer, parameter :: iotk_header_kind = selected_int_kind(8) !----------------------------------------------------------------------- -------! ! The following options can be modified runtime !----------------------------------------------------------------------- -------! ! Margins for unit search integer :: iotk_unitmin = 90000 integer :: iotk_unitmax = 99999 integer :: iotk_error_unit = 0 !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! Size of the buffer for iotk_getline ! (it is intended for efficiency; the total length of a line should be <= iotk_linlenx) integer :: iotk_getline_buffer = 1024 !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! If true, exhausting the error pool causes an overflow warning logical :: iotk_error_warn_overflow = .false. !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! Map of controls into XML tags ! control = 1 < > ! control = 2 ! control = 3 < /> ! control = 4 ! control = 5 ! control = 128 is a special tag for binary files (continuation tag) !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! Alphabet character(26), parameter :: lowalphabet = "abcdefghijklmnopqrstuvwxyz" character(26), parameter :: upalphabet = "ABCDEFGHIJKLMNOPQRSTUVWXYZ" character(52), parameter :: alphabet = lowalphabet//upalphabet character(53), parameter :: alphabet_ = alphabet//"_" character(10), parameter :: numbers = "0123456789" !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! List of characters which are not separators in a dat or attribute array character(66), parameter :: not_separator = alphabet_//numbers//"+-." !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! Rules for names character(54), parameter :: iotk_namcharfirst = alphabet//"_:" character(66), parameter :: iotk_namchar = iotk_namcharfirst// numbers//".-" !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! Default kinds, depending on compilers and compilation options for the library source integer, parameter :: iotk_defkind_character = kind("a") integer, parameter :: iotk_defkind_logical = kind(.true.) integer, parameter :: iotk_defkind_integer = kind(1) integer, parameter :: iotk_defkind_real = kind(1.0) integer, parameter :: iotk_defkind_complex = kind(1.0) !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! Maximum allowed rank integer, parameter :: iotk_maxrank = 4 ! Controlled by cpp integer, parameter :: iotk_maxrank_hard = 4 ! Controlled by sprep preprocessing !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! Internal type dealing with io units type iotk_unit integer :: unit ! fortran unit character(iotk_namlenx) :: root ! name of the root tag logical :: skip_root ! if true, root tag is not written automatically logical :: raw ! if true, the file is raw data integer :: level ! the hierarchical level inside the file logical :: close_at_end ! if true, the file has to be fortran-closed when iotk_close_* is called type (iotk_unit), pointer :: son ! a pointer to the son in the multi-file model type (iotk_unit), pointer :: parent ! a pointer to the parent in the multi-file model type (iotk_unit), pointer :: next ! a pointer to the next unit in the linked list end type iotk_unit !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! Special type used to force optional argument labelling. type iotk_dummytype integer :: dummy end type iotk_dummytype !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! Linked list of iotk_unit objects logical :: iotk_units_init = .false. type (iotk_unit), pointer :: iotk_units !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! Internal type dealing with error messages type iotk_error character, pointer :: str(:) end type iotk_error !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! Max length of a line in the error message. Any longer line will be cut integer, parameter :: iotk_error_linelength = 120 !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! Maximum number of errors which can be handled at the same time integer, parameter :: iotk_error_pool_size = 100 !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! Static pool of errors type(iotk_error) :: iotk_error_pool (iotk_error_pool_size) !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! Flags concerning the error pool: ! If true, that element of the pool is in usage logical :: iotk_error_pool_used (iotk_error_pool_size) = .false. !----------------------------------------------------------------------- -------! !----------------------------------------------------------------------- -------! ! These integers are set in increasing order to trace the order errors are raised ! They are then used to eliminate old errors if the user forgets to do that integer :: iotk_error_pool_order (iotk_error_pool_size) = 0 !----------------------------------------------------------------------- -------! end module iotk_base i686-apple-darwin8-gcc-4.0.1: c: No such file or directory make[2]: *** [iotk_base.o] Error 1 make[1]: *** [libiotk.a] Error 2 make: *** [libiotk] Error 2 From giannozz at nest.sns.it Sat Nov 25 11:15:26 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Sat, 25 Nov 2006 11:15:26 +0100 Subject: [Pw_forum] intel mac --- compilation In-Reply-To: <28A3A706-DA30-45BC-BC74-AD1EA2070D00@caltech.edu> References: <28A3A706-DA30-45BC-BC74-AD1EA2070D00@caltech.edu> Message-ID: <0583F0C3-603D-4776-8A67-89AC5E7CA393@nest.sns.it> On Nov 25, 2006, at 10:19 , Francesco Ciucci wrote: > I am trying to build QuantumEspresso 3.1.1 on my MacBook > (the little white one). > > I have as fortran compilers: g77, f2c and gfortran. g77 and f2c are not fortran-90 compilers; gfortran is, but the last time I tried it didn't work. Use g95, it works. In the latest version of QE (3.2, available in the next few days), intel macs should be recognized by 'configure' (mine is). Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From Gianni.Profeta at aquila.infn.it Sat Nov 25 17:07:10 2006 From: Gianni.Profeta at aquila.infn.it (Gianni Profeta) Date: Sat, 25 Nov 2006 17:07:10 +0100 (CET) Subject: [Pw_forum] AIX + XML files In-Reply-To: <44ED5ED5.80902@cea.fr> References: <44ED5ED5.80902@cea.fr> Message-ID: Dear users, I find the following problem on IBM AIX Version 5.1 running parallel version of pw.x from versions > 3.1 (including the CVS version). 1) Configure and compilation proceed without problems (as usual for this type of machine and for all the versions of the code) 2) running a test input (one k-point) the code stops with the following error message just before the first scf iteration: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from write_rho_xml : error # 1 cannot opentmp/Al.save/charge-density.xml file for writing %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 3) increasing the k-point mesh, the code does not stop but remains running (even if it does not produce outputs) I think the problem is related to the iotk library, indeed I found a post of Paolo regarding the incompatibility of iotk library in AIX enviroment. http://www.democritos.it/pipermail/pw_forum/2006-September/004912.html But the proposed solution does not work, in my case. Any suggestions? Thank you. Gianni From konstantin_kudin at yahoo.com Sat Nov 25 18:34:12 2006 From: konstantin_kudin at yahoo.com (Konstantin Kudin) Date: Sat, 25 Nov 2006 09:34:12 -0800 (PST) Subject: [Pw_forum] AIX + XML files In-Reply-To: Message-ID: <522495.49493.qm@web57707.mail.re3.yahoo.com> Sorry, this is not a solution post, just another data point. On an ancient IBM (aix 4.2 and xlf 6.1.1 or something, 32-bit compile) I tried QE-3.1 and the CP code would always hang on the 2nd write to the xml restart files (1st time the restart directory was saved fine). The program kept consuming the resources though. With the version QE-3.1.1 things started working fine, and I had not tried anything since. So perhaps someone could patch all these clues for IBM problems together ... Kostya --- Gianni Profeta wrote: > Dear users, > > I find the following problem on IBM AIX Version 5.1 running > parallel version of pw.x from versions > 3.1 (including the CVS > version). > > 1) Configure and compilation proceed without problems (as usual for > this > type of machine and for all the versions of the code) > > 2) running a test input (one k-point) the code stops with the > following > error message just before the first scf iteration: > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > from write_rho_xml : error # 1 > cannot opentmp/Al.save/charge-density.xml file for writing > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > 3) increasing the k-point mesh, the code does not stop but remains > running (even if it does not produce outputs) > > I think the problem is related to the iotk library, indeed > I found a post of Paolo regarding the incompatibility of iotk library > in > AIX enviroment. > http://www.democritos.it/pipermail/pw_forum/2006-September/004912.html > But the proposed solution does not work, in my case. > > Any suggestions? > > Thank you. > > Gianni > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ____________________________________________________________________________________ Do you Yahoo!? Everyone is raving about the all-new Yahoo! Mail beta. http://new.mail.yahoo.com From eyvaz_isaev at yahoo.com Sat Nov 25 21:17:17 2006 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Sat, 25 Nov 2006 12:17:17 -0800 (PST) Subject: [Pw_forum] QHA code Message-ID: <20061125201717.67119.qmail@web60315.mail.yahoo.com> Dear users, Below is a sketch from a README file for quasiharmonic calculations with provided force constants matrix. Now all calculations can be done using just a script which requires only a few input paramters. 1) phonon_DOS.x - phonon DOS and atom projected phonon DOS with g_x, g_y, and g_z components. 2) Atom_Integrated_DOS.x - integrates phonon DOS, and atom projected DOS to find out atom-specific contribution to the zero point vibration energy, phonon DOS, vibration energy, specific heat, entropy, internal energy NB!!! The limitation for the program is the presence of only one frequency gap in the phonon spectrum, i.e. the program assumes that there are well separated low-lying "acoustic" modes and optical branches. 3) Debye.x - calculates the Debye temperature 4) F_QHA.x - Total vibrational energy, the specific heat, entropy and internal energy in the quasiharmonic approximation If you find these features interesting then I can make the code available for the community , so that it can be distributed together with our lovely code. Bests, Eyvaz. ____________________________________________________________________________________ Want to start your own business? Learn how on Yahoo! Small Business. http://smallbusiness.yahoo.com/r-index From akohlmey at cmm.chem.upenn.edu Sun Nov 26 02:50:12 2006 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sat, 25 Nov 2006 20:50:12 -0500 (EST) Subject: [Pw_forum] intel mac --- compilation In-Reply-To: <28A3A706-DA30-45BC-BC74-AD1EA2070D00@caltech.edu> Message-ID: On Sat, 25 Nov 2006, Francesco Ciucci wrote: FC> Hi all, francesco, FC> I am trying to build QuantumEspresso 3.1.1 on my MacBook (the little FC> white one). FC> FC> I have as fortran compilers: g77, f2c and gfortran. please do not mix and match different fortran compilers. g77 and f2c use a very similar, but not identical runtime, gfortran has its own and different runtime. while with some crazy hacks one could make the compilation succeed, the result would be desastrous. FC> FC> I run configure as follows: FC> FC> ./configure F77=g77 MPIF90=gfortran CC=gcc F90=gfortran BLAS_LIBS="- FC> framework accelerate" LAPACK_LIBS=" " CFLAGS=-O3 FC> FC> and I get at the end "configure: success". FC> FC> Although, when I run "sudo make all" I get what I reported in PS FC> (sorry it's lengthy). there is _no_ need to compile with root priviledge (in fact, i strongly advise against doing _anything_ as root that can be done without, which is more than what most people would expect). this way you can easily ruin your whole installation! no regular application program should need this, especially for compilation. i just tried the latest cvs version with the gfortran shipped with fedora core 6 linux and it fails on Modules/input_parameters.f90 due to an internal compiler error, i.e. a bug of the compiler. FC> FC> Could anyone help me out? FC> FC> Thank you, FC> Fra FC> FC> P.S. FC> test -d bin || mkdir bin FC> if test -d iotk ; then \ FC> ( cd iotk ; if test "make" = "" ; then make TLDEPS= libiotk.a ; \ FC> else make TLDEPS= libiotk.a ; fi ) ; fi FC> cd src ; make libiotk.a FC> cpp -P -traditional -D__FFTW -D__USE_INTERNAL_FFTW -I../include FC> iotk_base.f90 -o iotk_base.F90 FC> -------! FC> FC> !----------------------------------------------------------------------- FC> -------! FC> ! Maximum number of errors which can be handled at the same time FC> integer, parameter :: iotk_error_pool_size = 100 FC> !----------------------------------------------------------------------- FC> -------! this is because you seem to have a (modified) cpp on your machine, that does not support the '-o' flag. for this you can change the .f90 compilation rule as follows .f90.o: $(CPP) $(CPPFLAGS) $< -o $*.F90 $(MPIF90) $(F90FLAGS) -c $*.F90 -o $*.o please note, that the makefile syntax demands a 'Tab' and not spaces in front of $(CPP) and $(MPIF90). cheers, axel. -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From akohlmey at cmm.chem.upenn.edu Sun Nov 26 02:52:45 2006 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sat, 25 Nov 2006 20:52:45 -0500 (EST) Subject: [Pw_forum] AIX + XML files In-Reply-To: Message-ID: On Sat, 25 Nov 2006, Gianni Profeta wrote: CP> Dear users, CP> CP> I find the following problem on IBM AIX Version 5.1 running CP> parallel version of pw.x from versions > 3.1 (including the CVS version). CP> CP> 1) Configure and compilation proceed without problems (as usual for this CP> type of machine and for all the versions of the code) CP> CP> 2) running a test input (one k-point) the code stops with the following CP> error message just before the first scf iteration: CP> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% CP> from write_rho_xml : error # 1 CP> cannot opentmp/Al.save/charge-density.xml file for writing CP> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% please check your input file. it looks as if you used 'tmp' instead of '/tmp'. all QE codes will fail if you try to write to a non-existing directory... axel. CP> CP> 3) increasing the k-point mesh, the code does not stop but remains CP> running (even if it does not produce outputs) CP> CP> I think the problem is related to the iotk library, indeed CP> I found a post of Paolo regarding the incompatibility of iotk library in CP> AIX enviroment. CP> http://www.democritos.it/pipermail/pw_forum/2006-September/004912.html CP> But the proposed solution does not work, in my case. CP> CP> Any suggestions? CP> CP> Thank you. CP> CP> Gianni CP> _______________________________________________ CP> Pw_forum mailing list CP> Pw_forum at pwscf.org CP> http://www.democritos.it/mailman/listinfo/pw_forum CP> -- ======================================================================= 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 suriano at sissa.it Sun Nov 26 03:14:54 2006 From: suriano at sissa.it (Antonio Suriano) Date: Sun, 26 Nov 2006 03:14:54 +0100 Subject: [Pw_forum] If some geek wants some fun: pwscf on cell. Message-ID: <4568F89E.5080608@sissa.it> IBM has developed a new fast parallel processor called Cell with 8 cores. This processor is specialized in operations on vectors and matrices and it is piloted by an evolution of PPC. Some people can run Linux on this kind of machines (Yellow DOG Linux). I think it will be very interesting if some computer geek would find an hack to run PWSCF on this kind of machines. Actually there is this supercomputer that has a fast evolution of G5 as core unit + an 8 core CELL for fast vectorial operations + 2 very fast floating points unit id est 11 overall processor It has 20 gigabytes hard disk and 512 mega ram (256 for the core and 256 for the floating point vectorial units) that runs at the same speed of the processors+ a dvd player. This supercomputer has an overall performance in floating point of 1.8 teraflops. This supercomputer costs only 599 euros, it is developed by Sony and its name is Playstation 3. It is not a joke, some people are doing protein stuff only using the free processor time of machines running videogames meanwhile: http://en.wikipedia.org/wiki/Folding%40home Chiacchiera con i tuoi amici in tempo reale! http://it.yahoo.com/mail_it/foot/*http://it.messenger.yahoo.com From akohlmey at cmm.chem.upenn.edu Sun Nov 26 03:44:20 2006 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sat, 25 Nov 2006 21:44:20 -0500 (EST) Subject: [Pw_forum] If some geek wants some fun: pwscf on cell. In-Reply-To: <4568F89E.5080608@sissa.it> Message-ID: On Sun, 26 Nov 2006, Antonio Suriano wrote: AS> IBM has developed a new fast parallel processor called Cell with 8 cores. AS> This processor is specialized in operations on vectors and matrices and AS> it is piloted by an evolution of PPC. AS> AS> AS> Some people can run Linux on this kind of machines (Yellow DOG Linux). AS> I think it will be very interesting if some computer geek would find an AS> hack to run PWSCF on this kind of machines. note, that in its current form, cell has only single precision floating point. that does not hurt (much) with classical MD codes like gromacs (which is what folding at home uses) if they are carefully programmed, and you don't need accurate dynamics, but only reasonable statistics. AS> Actually there is this supercomputer that has a fast evolution of G5 as AS> core unit + an 8 core AS> CELL for fast vectorial operations + 2 very fast floating points unit id AS> est 11 overall processor AS> It has 20 gigabytes hard disk and 512 mega ram (256 for the core and 256 AS> for the floating point vectorial units) that runs at the same speed of AS> the processors+ a dvd player. AS> AS> This supercomputer has an overall performance in floating point of 1.8 AS> teraflops. please be careful with those numbers, this is most likely peak performance, i.e. pointless for any practical applications... AS> AS> This supercomputer costs only 599 euros, it is developed by Sony and AS> its name is Playstation 3. AS> AS> It is not a joke, some people are doing protein stuff only using the AS> free processor time of machines running videogames meanwhile: AS> http://en.wikipedia.org/wiki/Folding%40home you also need a very special compiler for that kind of hardware. last time a heard. a cell development kit (including hardware) was around $20k... this was about half a year ago, though. for applications like quantum espresso, the most promising route to make this kind of hardware useful, would be to a) make sure the code uses BLAS/LAPACK whereever possible and b) help people like jack dongarra to build a BLAS/LAPACK that works with a mix of single and double precision, yet provides full double precision accuracy. btw: if you go and look at the folding at home statistics, you'd see that you can actually make a supercomputer from your desktop by adding a 3d graphics card. current GPUs have even more power than cell (it is just a PITA to program them and get the data off them). ciao, axel. AS> AS> AS> Chiacchiera con i tuoi amici in tempo reale! AS> http://it.yahoo.com/mail_it/foot/*http://it.messenger.yahoo.com AS> _______________________________________________ AS> Pw_forum mailing list AS> Pw_forum at pwscf.org AS> http://www.democritos.it/mailman/listinfo/pw_forum AS> -- ======================================================================= 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 jibiaoli at gmail.com Sun Nov 26 05:34:40 2006 From: jibiaoli at gmail.com (Clark Lee) Date: Sun, 26 Nov 2006 12:34:40 +0800 Subject: [Pw_forum] help needed with projwfc.x In-Reply-To: References: <57185.202.41.111.151.1157465481.squirrel@202.41.111.151> Message-ID: Thank you Stefano and Paolo for the detailed and definitive answers! On 11/25/06, Stefano de Gironcoli wrote: > > > the sequence is: > > 1 dz2 (m=0) > 2 dzx (real combination of m=+/-1 with cosine) > 3 dzy (real combination of m=+/-1 with sine) > 4 dx2-y2 (real combination of m=+/-2 with cosine) > 5 dxy (real combination of m=+/-1 with sine) > > similarly for p orbitals > > 1 pz (m=0) > 2 px (real combination of m=+/-1 with cosine) > 3 py (real combination of m=+/-1 with sine) > > stefano > > On Sat, 25 Nov 2006, Clark Lee wrote: > > > The same question. > > Anyone else knows what's the sequence of the d-orbitals > > (dxy,dxz,dyz,dx2-y2,dz2) are assigned in results from projwfc.x > > > > > > On 9/5/06, aarti srirangarajan wrote: > >> > >> Dear PWSCF developers, > >> > >> Could anyone please explain the output of the program projwfc.x? > >> My question is how the values for m = 1,2,3,4,5 are assigned ( m should > be > >> -2 to +2) ? And in the output files for the PDOS for d-orbitals, what is > >> the sequence of the data listed in it.(in terms of dxy,dxz,dyz etc.). > >> Thanking you in anticipation > >> Aarti > >> TSU,JNCASR > >> > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Standing on the shoulders of Newton From babazade at gmail.com Sun Nov 26 11:11:51 2006 From: babazade at gmail.com (Amin Babazadeh) Date: Sun, 26 Nov 2006 13:41:51 +0330 Subject: [Pw_forum] use doual core In-Reply-To: <7b6913e90611220734s15038d2ei36e1e7a28166f146@mail.gmail.com> References: <7b6913e90611220734s15038d2ei36e1e7a28166f146@mail.gmail.com> Message-ID: Dear Dr.Kohlmeyer Thanks a lot for your help.I did what you said.i installed fftw-3.1.2.tar lam-7.1.2.tar mpich2-1.0.4p1.tar openmpi-1.1.2.tar on fedora core 5 and after that i configure the espresso but this error was written WARNING: the following problems were detected: unsupported C/Fortran compilers combination: CC=mpicc, F77=mpif77, F90=mpif90 you may not be able to compile this program. i don't know what this problem means.would you please help me. Thanks a lot On 11/22/06, Axel Kohlmeyer wrote: > > On 11/22/06, Amin Babazadeh wrote: > > dear users > > I use a computer with a dual core CPU.but my problem is that when i run > a > > program only one of the CPU work but i like to share the work (solving k > > points) between cpus. > > I think that i should install the parallel pwscf but when i run the > > ./configure only single pwscf installed. > > you have to install an MPI package (e.g. LAM/MPI, OpenMPI, MPICH) > first and then configure/compile quantum espresso with parallel support > as explained in the provided installation instructions. > > please make sure that your MPI package is compatible with or compiled > for the fortran compiler you want to use to compiled quantum espresso > with. > precompiled packages may be compiled for a different compiler and thus > need special flags or environment variables set to use the compiler of > your choice. > > axel. > > > would you please tell me or give me a document that solve my problem. > > > > > > -- > > AMIN > > > -- > ======================================================================= > 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 > -- AMIN -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061126/0c291425/attachment.htm From Gianni.Profeta at aquila.infn.it Sun Nov 26 14:26:14 2006 From: Gianni.Profeta at aquila.infn.it (Gianni Profeta) Date: Sun, 26 Nov 2006 14:26:14 +0100 (CET) Subject: [Pw_forum] AIX + XML files In-Reply-To: References: Message-ID: On Sat, 25 Nov 2006, Axel Kohlmeyer wrote: > CP> from write_rho_xml : error # 1 > CP> cannot opentmp/Al.save/charge-density.xml file for writing > CP> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > please check your input file. it looks as if you used > 'tmp' instead of '/tmp'. all QE codes will fail if you > try to write to a non-existing directory... > > axel. Dear Alex, I already did. I used : outdir='tmp/' and indeed all the previous versions (<= 3.0) work for the same input file. I think that if 'outdir' does not exist is checked before the writing, and the error message in that case is: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from outdir: : error # 1 tmp/ non existent or non writable %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% just after the execution starts. While in my case, the code arrives at the beginning of the first iteration and then stops. In fact, it can writes into the tmp/ directory : >ls -ltr tmp/ -rw-r--r-- 1 gianni staff 0 Nov 26 09:07 Al.wfc1 drwxr-xr-x 2 gianni staff 256 Nov 26 09:07 Al.save -rw-r--r-- 1 gianni staff 0 Nov 26 09:07 Al.igk but the files and directory (Al.save) are empty. Thank you. Gianni From akohlmey at cmm.chem.upenn.edu Sun Nov 26 20:08:20 2006 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sun, 26 Nov 2006 14:08:20 -0500 (EST) Subject: [Pw_forum] AIX + XML files In-Reply-To: Message-ID: On Sun, 26 Nov 2006, Gianni Profeta wrote: GP> On Sat, 25 Nov 2006, Axel Kohlmeyer wrote: GP> GP> > CP> from write_rho_xml : error # 1 GP> > CP> cannot opentmp/Al.save/charge-density.xml file for writing GP> > CP> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% GP> > GP> > please check your input file. it looks as if you used GP> > 'tmp' instead of '/tmp'. all QE codes will fail if you GP> > try to write to a non-existing directory... GP> > GP> > axel. GP> GP> Dear Alex, GP> I already did. I used : outdir='tmp/' GP> and indeed all the previous versions (<= 3.0) work for the same input GP> file. ok. that is indeed a hint, that you are seeing something related to iotk. please have a look at your make.sys file and compare the settings that configure has picked up with the following (most important are F90 and MPIF90, this is for parallel compilation...). there are some subtle semantics issue, that strike only the ibm compilers which do not seem to automatically switch to f95 mode on .f90 file name extensions (you _have_ to use xlf95 or mpxlf95... :-( ). # If the mass library is available, use -D__MASS DFLAGS = -D__AIX,-D__XLF,-D__ESSL,-D__MASS,-D__PARA,-D__MPI CPPFLAGS = -D__AIX -D__XLF -D__ESSL -D__MASS -D__PARA -D__MPI # Fortran and C compilers CC = mpcc MPICC = mpcc F90 = mpxlf95 MPIF90 = mpxlf95 F77 = mpxlf MPIF77 = mpxlf # Fortran flags FFLAGS = -qalias=noaryovrlp:nointptr -I../include \ -O3 -qstrict -qarch=auto -qtune=auto -qdpc -Q -qessl FFLAGS_NOOPT = -O0 F90FLAGS = $(FFLAGS) -qfree=f90 -qsuffix=cpp=f90 -WF,$(DFLAGS) \ -I../Modules -I../PW -I../PH -I../iotk/src CFLAGS = -I../include $(CPPFLAGS) -O2 # For essl and mass libraries LIBS = -lessl -lmassvp4 -lmass #-L/usr/local/lib -llapack # Linker LD = $(MPIF90) GP> GP> I think that if 'outdir' does not exist is checked before the writing, GP> and the error message in that case is: GP> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% GP> from outdir: : error # 1 GP> tmp/ non existent or non writable GP> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% GP> GP> just after the execution starts. GP> GP> While in my case, the code arrives at the beginning of the first iteration GP> and then stops. GP> In fact, it can writes into the tmp/ directory : GP> GP> >ls -ltr tmp/ GP> -rw-r--r-- 1 gianni staff 0 Nov 26 09:07 Al.wfc1 GP> drwxr-xr-x 2 gianni staff 256 Nov 26 09:07 Al.save GP> -rw-r--r-- 1 gianni staff 0 Nov 26 09:07 Al.igk GP> GP> but the files and directory (Al.save) are empty. ok. this was just a wild guess. if you intended to write to tmp/ you are correct. people usually do not use a path like that for scratch files, as it is generally NFS mounted, and thus lead to bad i/o performances... ciao, axel. GP> GP> Thank you. GP> Gianni GP> _______________________________________________ GP> Pw_forum mailing list GP> Pw_forum at pwscf.org GP> http://www.democritos.it/mailman/listinfo/pw_forum GP> -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From akohlmey at cmm.chem.upenn.edu Sun Nov 26 20:34:43 2006 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sun, 26 Nov 2006 14:34:43 -0500 (EST) Subject: [Pw_forum] use doual core In-Reply-To: Message-ID: On Sun, 26 Nov 2006, Amin Babazadeh wrote: dear amin, AB> Dear Dr.Kohlmeyer AB> Thanks a lot for your help.I did what you said.i installed AB> fftw-3.1.2.tar AB> lam-7.1.2.tar AB> mpich2-1.0.4p1.tar AB> openmpi-1.1.2.tar whoa there. you only need _one_ mpi library, not three of them. how did you plan to distinguish between them?? did you try to compile and run any MPI example programs to verify that your MPI installation is correct and working?? AB> on fedora core 5 and after that i configure the espresso but this error was AB> written AB> AB> WARNING: the following problems were detected: AB> AB> unsupported C/Fortran compilers combination: AB> CC=mpicc, F77=mpif77, F90=mpif90 AB> you may not be able to compile this program. AB> i don't know what this problem means.would you please help me. please read what it says. it is a _WARNING_. that means the configure script could not verify that you are using a know to work combination of MPI and compilers. i.e. you should go on and try to compile, but be very careful in checking that the resulting programs will work. good luck, axel. AB> Thanks a lot AB> AB> AB> -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From akohlmey at cmm.chem.upenn.edu Sun Nov 26 22:47:12 2006 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sun, 26 Nov 2006 16:47:12 -0500 (EST) Subject: [Pw_forum] portability of SCNDS subroutine in clib/cptimer.c Message-ID: hi everybody, just tried some testing on a fedora core 6 machine using the latest QE cvs version and timings do not work (again). after some searching, i found out that this is due to glibc following the POSIX specs, which have declared CLK_TCK to be obsolete and CLOCKS_PER_SEC to always return 1000000. hence the use of times(2) will not fly in the long run unless the subroutine is changed e.g. as follows. double F77_FUNC(scnds,SCNDS) ( ) /* Return the cpu time associated to the current process */ { static struct tms T; static int first = 1; static double init_cputime = 0.0; static double ticks_per_second; double cputime; if( first ) { ticks_per_second = sysconf(_SC_CLK_TCK); } times(&T); cputime = (double)(T.tms_utime) / ticks_per_second; if( first ) { first = 0; init_cputime = cputime; } return cputime - init_cputime; } furthermore, times are supposed to overflow rather frequently (about every 72min on a current 32bit machine...), so perhaps another source for time used, e.g. getrusage(2) might be an alternative. according to the linux manpage the fields ru_utime and ru_stime are specified by POSIX (apart from getrusage being conforming to SVr4 and 4.3BSD, i.e. rather widely in the *nix world). the corresponding code could be: #include #include double F77_FUNC(scnds,SCNDS) ( ) /* Return the cpu time associated to the current process */ { static struct rusage T; getrusage(RUSAGE_SELF, &T); return ((double)T.ru_utime.tv_sec + ((double)T.ru_utime.tv_usec)/1000000.0); } this would have the additional advantage, that out of #include #include #include only would be needed and potential conflicts (as it is know to happen on some OSes) from including overlapping headers. it would be nice, if people running on non-linux machines could verify that this change is portable across all of their platforms. any comments or suggestions are appreciated. ciao, axel. -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From hqzhou at nju.edu.cn Mon Nov 27 10:35:25 2006 From: hqzhou at nju.edu.cn (Huiqun Zhou) Date: Mon, 27 Nov 2006 17:35:25 +0800 Subject: [Pw_forum] use doual core References: <7b6913e90611220734s15038d2ei36e1e7a28166f146@mail.gmail.com> Message-ID: <002e01c71207$5e733cb0$1d00a8c0@solarflare> Have you set up you environment for using one the mpi libraries you installed in .bashrc? For example, for openmpi you need to set following in your .bashrc if you are using bash as your shell: ... export PATH=/opt/openmpi/bin:$PATH export LD_LIBRARY_PATH=/opt/openmpi/lib:$LD_LIBRARY_PATH ... Note: you need to replace /opt/openmpi with the path of your installation. Huiqun Zhou ----- Original Message ----- From: Amin Babazadeh To: pw_forum at pwscf.org Sent: Sunday, November 26, 2006 6:11 PM Subject: Re: [Pw_forum] use doual core Dear Dr.Kohlmeyer Thanks a lot for your help.I did what you said.i installed fftw-3.1.2.tar lam-7.1.2.tar mpich2-1.0.4p1.tar openmpi-1.1.2.tar on fedora core 5 and after that i configure the espresso but this error was written WARNING: the following problems were detected: unsupported C/Fortran compilers combination: CC=mpicc, F77=mpif77, F90=mpif90 you may not be able to compile this program. i don't know what this problem means.would you please help me. Thanks a lot On 11/22/06, Axel Kohlmeyer wrote: On 11/22/06, Amin Babazadeh < babazade at gmail.com> wrote: > dear users > I use a computer with a dual core CPU.but my problem is that when i run a > program only one of the CPU work but i like to share the work (solving k > points) between cpus. > I think that i should install the parallel pwscf but when i run the > ./configure only single pwscf installed. you have to install an MPI package (e.g. LAM/MPI, OpenMPI, MPICH) first and then configure/compile quantum espresso with parallel support as explained in the provided installation instructions. please make sure that your MPI package is compatible with or compiled for the fortran compiler you want to use to compiled quantum espresso with. precompiled packages may be compiled for a different compiler and thus need special flags or environment variables set to use the compiler of your choice. axel. > would you please tell me or give me a document that solve my problem. > > > -- > AMIN -- ======================================================================= 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 -- AMIN -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061127/c7d95497/attachment.htm From hqzhou at nju.edu.cn Mon Nov 27 11:05:09 2006 From: hqzhou at nju.edu.cn (Huiqun Zhou) Date: Mon, 27 Nov 2006 18:05:09 +0800 Subject: [Pw_forum] QHA code References: <20061125201717.67119.qmail@web60315.mail.yahoo.com> Message-ID: <005c01c7120b$857adb70$1d00a8c0@solarflare> I need it, Eyvaz. Huiqun ----- Original Message ----- From: "Eyvaz Isaev" To: Sent: Sunday, November 26, 2006 4:17 AM Subject: [Pw_forum] QHA code > Dear users, > > Below is a sketch from a README file for quasiharmonic calculations with > provided force constants matrix. Now all calculations can be done using > just a script which requires only a few input paramters. > > 1) phonon_DOS.x - phonon DOS and atom projected phonon DOS with g_x, g_y, > and g_z components. > 2) Atom_Integrated_DOS.x - integrates phonon DOS, and atom projected DOS > to find out > atom-specific contribution to the zero point vibration energy, phonon > DOS, vibration energy, specific heat, entropy, internal energy > NB!!! The limitation for the program is the presence of only one frequency > gap in the phonon spectrum, i.e. the program assumes that there are well > separated low-lying "acoustic" modes and optical branches. > > 3) Debye.x - calculates the Debye temperature > 4) F_QHA.x - Total vibrational energy, the specific heat, entropy and > internal energy in the > quasiharmonic approximation > > If you find these features interesting then I can make the code available > for the community , so that it can be distributed together with our > lovely code. > > Bests, > Eyvaz. > > > > > > > > > > ____________________________________________________________________________________ > Want to start your own business? > Learn how on Yahoo! Small Business. > http://smallbusiness.yahoo.com/r-index > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From gaoguoying at gmail.com Mon Nov 27 12:23:07 2006 From: gaoguoying at gmail.com (Guoying Gao) Date: Mon, 27 Nov 2006 19:23:07 +0800 Subject: [Pw_forum] question on structurerelax Message-ID: <6234c0c60611270323u1e92f47ds9619065907429542@mail.gmail.com> Dear everyone, When I did scf calculation with experimental parameters (experimental pressure 80GPa), the output file gave P=164GPa, the bond length is near equivalent to that of experiment, why so large diverge in pressure? Anyone who knows the reason is welcomed. Thanks in advance. gao From jin.zhang.pku at gmail.com Mon Nov 27 12:24:37 2006 From: jin.zhang.pku at gmail.com (Jin Zhang) Date: Mon, 27 Nov 2006 19:24:37 +0800 Subject: [Pw_forum] Confused about vibrational modes of isolated molecule from ph.x In-Reply-To: <5d233b6a0611200427w27ad0305pc63196f73eb8ea44@mail.gmail.com> References: <5d233b6a0611182329s46725cd4mf772a3a75f786c9a@mail.gmail.com> <456036D2.6030901@lg.ehu.es> <5d233b6a0611190542q19bcd0edq94b8c17917357c8@mail.gmail.com> <45618283.1030403@sissa.it> <5d233b6a0611200427w27ad0305pc63196f73eb8ea44@mail.gmail.com> Message-ID: <5d233b6a0611270324xdf38aefw159e7a526d588e31@mail.gmail.com> Hi all, I posted this issue one week ago. Now I have relaxed the structure and got total force less than 5.0D-5. I used a large cubic supercell with 30bohr alat and ecutwfc up to 50Ry. However, things only went worse. The highest frequency from ph.x dropped from 3187 cm^(-1) to 1849, while gaussian03 shows it should be around 3300cm^(-1). I also checked H2 molecule. In this case, I got three negative values and three positive ones( values are 898,898,3679cm^-1 ). But there should be only ONE vibrational frequency which corresponds to the stretching mode along bonding direction. And the highest freq. is also too low (should be around 43xx). I really couldn't figure out what's wrong. Can someone help me? Thanks in advance! The input files for H2 are: ---------------------------scf_input-----------------(run two times just like example07 )----- &control calculation='scf' restart_mode='from_scratch', prefix='h2', pseudo_dir = '$PSEUDO_DIR/', outdir='$TMP_DIR/' / &system ibrav= 8, celldm(1) =18, celldm(2)=1.0, celldm(3)=1.0, nat= 2, ntyp= 1, ecutwfc = 50.0, occupations='smearing', smearing='gausssian', degauss=0.01, la2F = .false., nspin=1, / &electrons conv_thr = 1.0d-8 mixing_beta = 0.7 / ATOMIC_SPECIES H 1.008 H.blyp-vbc.UPF ATOMIC_POSITIONS (bohr) H 0.000000000 0.000000000 -0.707244771 H 0.000000000 0.000000000 0.707244771 K_POINTS {automatic} 1 1 1 0 0 0 ------------------------------ph_input--------------------------- &inputph tr2_ph=1.0d-8, prefix='h2', fildvscf='h2dv', amass(1)=1.008, outdir='$TMP_DIR/', fildyn='h2.dyn', elph=.false., trans=.true., ldisp=.true., nq1=1, nq2=1, nq3=1 -------------------------------------------------------------------- Regards, Jin Zhang -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061127/b17f5b38/attachment.htm From ding at sissa.it Mon Nov 27 12:52:36 2006 From: ding at sissa.it (Xunlei Ding) Date: Mon, 27 Nov 2006 12:52:36 +0100 Subject: [Pw_forum] Confused about vibrational modes of isolated molecule from ph.x In-Reply-To: <5d233b6a0611270324xdf38aefw159e7a526d588e31@mail.gmail.com> References: <5d233b6a0611182329s46725cd4mf772a3a75f786c9a@mail.gmail.com> <456036D2.6030901@lg.ehu.es> <5d233b6a0611190542q19bcd0edq94b8c17917357c8@mail.gmail.com> <45618283.1030403@sissa.it> <5d233b6a0611200427w27ad0305pc63196f73eb8ea44@mail.gmail.com> <5d233b6a0611270324xdf38aefw159e7a526d588e31@mail.gmail.com> Message-ID: <456AD184.8000404@sissa.it> Dear Zhang, The tr2_ph is too large in you case. Just change the tr2_ph=1.0d-14, you will get: omega( 1) = 2.198936 [THz] = 73.349094 [cm-1] omega( 2) = 3.192379 [THz] = 106.487008 [cm-1] omega( 3) = 3.192379 [THz] = 106.487008 [cm-1] omega( 4) = 4.915826 [THz] = 163.975388 [cm-1] omega( 5) = 4.915826 [THz] = 163.975388 [cm-1] omega( 6) = 131.771213 [THz] = 4395.443762 [cm-1] Yours, Ding Jin Zhang wrote: > Hi all, > > I posted this issue one week ago. Now I have relaxed the structure > and got total force less than 5.0D-5. I used a large cubic supercell > with 30bohr alat and ecutwfc up to 50Ry. However, things only went > worse. The highest frequency from ph.x dropped from 3187 cm^(-1) to > 1849, while gaussian03 shows it should be around 3300cm^(-1). > > I also checked H2 molecule. In this case, I got three negative values > and three positive ones( values are 898,898,3679cm^-1 ). But there > should be only ONE vibrational frequency which corresponds to the > stretching mode along bonding direction. And the highest freq. is also > too low (should be around 43xx). > > I really couldn't figure out what's wrong. Can someone help me? Thanks > in advance! > > The input files for H2 are: > ---------------------------scf_input-----------------(run two times > just like example07 )----- > &control > calculation='scf' > restart_mode='from_scratch', > prefix='h2', > pseudo_dir = '$PSEUDO_DIR/', > outdir='$TMP_DIR/' > / > &system > ibrav= 8, celldm(1) =18, celldm(2)= 1.0, celldm(3)=1.0, > nat= 2, ntyp= 1, > ecutwfc = 50.0, > occupations='smearing', smearing='gausssian', degauss=0.01, > la2F = .false., nspin=1, > / > &electrons > conv_thr = 1.0d-8 > mixing_beta = 0.7 > / > ATOMIC_SPECIES > H 1.008 H.blyp-vbc.UPF > ATOMIC_POSITIONS (bohr) > H 0.000000000 0.000000000 -0.707244771 > H 0.000000000 0.000000000 0.707244771 > K_POINTS {automatic} > 1 1 1 0 0 0 > ------------------------------ph_input--------------------------- > &inputph > tr2_ph=1.0d-8, > prefix='h2', > fildvscf='h2dv', > amass(1)=1.008, > outdir='$TMP_DIR/', > fildyn=' h2.dyn', > elph=.false., > trans=.true., > ldisp=.true., > nq1=1, nq2=1, nq3=1 > -------------------------------------------------------------------- > > Regards, > Jin Zhang > From giannozz at nest.sns.it Mon Nov 27 13:05:45 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 27 Nov 2006 13:05:45 +0100 Subject: [Pw_forum] question on structurerelax In-Reply-To: <6234c0c60611270323u1e92f47ds9619065907429542@mail.gmail.com> References: <6234c0c60611270323u1e92f47ds9619065907429542@mail.gmail.com> Message-ID: On Nov 27, 2006, at 12:23 , Guoying Gao wrote: > When I did scf calculation with experimental parameters > (experimental pressure 80GPa), the output file gave P=164GPa, > the bond length is near equivalent to that of experiment, why so > large diverge in pressure? > Anyone who knows the reason is welcomed. one well-known potential reason is the slow convergence of pressure with respect to cutoff Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From gaoguoying at gmail.com Mon Nov 27 13:23:18 2006 From: gaoguoying at gmail.com (Guoying Gao) Date: Mon, 27 Nov 2006 20:23:18 +0800 Subject: [Pw_forum] question on structurerelax In-Reply-To: References: <6234c0c60611270323u1e92f47ds9619065907429542@mail.gmail.com> Message-ID: <6234c0c60611270423o5261b90fhcb705bd9fa09d87d@mail.gmail.com> Dear Polo: I'm sorry for my ambiguous expressions. I mean that I did a single scf calculation before relaxing. gao From giannozz at nest.sns.it Mon Nov 27 13:34:41 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 27 Nov 2006 13:34:41 +0100 Subject: [Pw_forum] question on structurerelax In-Reply-To: <6234c0c60611270423o5261b90fhcb705bd9fa09d87d@mail.gmail.com> References: <6234c0c60611270323u1e92f47ds9619065907429542@mail.gmail.com> <6234c0c60611270423o5261b90fhcb705bd9fa09d87d@mail.gmail.com> Message-ID: <2B82D1A3-7BBF-413F-956E-A3B50B36F202@nest.sns.it> On Nov 27, 2006, at 13:23 , Guoying Gao wrote: > I'm sorry for my ambiguous expressions. I mean that I did > a single scf calculation before relaxing. this is irrelevant: convergence of pressure wrt kinetic energy cutoff is notoriously slow Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From Gianni.Profeta at aquila.infn.it Mon Nov 27 13:37:10 2006 From: Gianni.Profeta at aquila.infn.it (Gianni Profeta) Date: Mon, 27 Nov 2006 13:37:10 +0100 (CET) Subject: [Pw_forum] AIX + XML files In-Reply-To: References: Message-ID: Alex, I changed the make.sys. The compilation goes straight, but at the linking phase I have the following : %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% ld: 0711-317 ERROR: Undefined symbol: .flush ld: 0711-345 Use the -bloadmap or -bnoquiet option to obtain more information. make: 1254-004 The error code from the last command is 8. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% I did not solve the problem by myself... G. From giannozz at nest.sns.it Mon Nov 27 13:49:15 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 27 Nov 2006 13:49:15 +0100 Subject: [Pw_forum] AIX + XML files In-Reply-To: References: Message-ID: On Nov 27, 2006, at 13:37 , Gianni Profeta wrote: > ld: 0711-317 ERROR: Undefined symbol: .flush for aix, you should use "flush_" instead of "flush" Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From jin.zhang.pku at gmail.com Mon Nov 27 14:14:22 2006 From: jin.zhang.pku at gmail.com (Jin Zhang) Date: Mon, 27 Nov 2006 21:14:22 +0800 Subject: [Pw_forum] Confused about vibrational modes of isolated molecule from ph.x In-Reply-To: <456AD184.8000404@sissa.it> References: <5d233b6a0611182329s46725cd4mf772a3a75f786c9a@mail.gmail.com> <456036D2.6030901@lg.ehu.es> <5d233b6a0611190542q19bcd0edq94b8c17917357c8@mail.gmail.com> <45618283.1030403@sissa.it> <5d233b6a0611200427w27ad0305pc63196f73eb8ea44@mail.gmail.com> <5d233b6a0611270324xdf38aefw159e7a526d588e31@mail.gmail.com> <456AD184.8000404@sissa.it> Message-ID: <5d233b6a0611270514p1cf17ef9uebfc72c1d84b24a@mail.gmail.com> Indeed! Thanks very much! BTW, is there a "tr1_ph"? This name convention is really interesting:) Best, Jin Zhang On 11/27/06, Xunlei Ding wrote: > > Dear Zhang, > > The tr2_ph is too large in you case. > Just change the tr2_ph=1.0d-14, you will get: > omega( 1) = 2.198936 [THz] = 73.349094 [cm-1] > omega( 2) = 3.192379 [THz] = 106.487008 [cm-1] > omega( 3) = 3.192379 [THz] = 106.487008 [cm-1] > omega( 4) = 4.915826 [THz] = 163.975388 [cm-1] > omega( 5) = 4.915826 [THz] = 163.975388 [cm-1] > omega( 6) = 131.771213 [THz] = 4395.443762 [cm-1] > > Yours, > Ding > > Jin Zhang wrote: > > > Hi all, > > > > I posted this issue one week ago. Now I have relaxed the structure > > and got total force less than 5.0D-5. I used a large cubic supercell > > with 30bohr alat and ecutwfc up to 50Ry. However, things only went > > worse. The highest frequency from ph.x dropped from 3187 cm^(-1) to > > 1849, while gaussian03 shows it should be around 3300cm^(-1). > > > > I also checked H2 molecule. In this case, I got three negative values > > and three positive ones( values are 898,898,3679cm^-1 ). But there > > should be only ONE vibrational frequency which corresponds to the > > stretching mode along bonding direction. And the highest freq. is also > > too low (should be around 43xx). > > > > I really couldn't figure out what's wrong. Can someone help me? Thanks > > in advance! > > > > The input files for H2 are: > > ---------------------------scf_input-----------------(run two times > > just like example07 )----- > > &control > > calculation='scf' > > restart_mode='from_scratch', > > prefix='h2', > > pseudo_dir = '$PSEUDO_DIR/', > > outdir='$TMP_DIR/' > > / > > &system > > ibrav= 8, celldm(1) =18, celldm(2)= 1.0, celldm(3)=1.0, > > nat= 2, ntyp= 1, > > ecutwfc = 50.0, > > occupations='smearing', smearing='gausssian', degauss=0.01, > > la2F = .false., nspin=1, > > / > > &electrons > > conv_thr = 1.0d-8 > > mixing_beta = 0.7 > > / > > ATOMIC_SPECIES > > H 1.008 H.blyp-vbc.UPF > > ATOMIC_POSITIONS (bohr) > > H 0.000000000 0.000000000 -0.707244771 > > H 0.000000000 0.000000000 0.707244771 > > K_POINTS {automatic} > > 1 1 1 0 0 0 > > ------------------------------ph_input--------------------------- > > &inputph > > tr2_ph=1.0d-8, > > prefix='h2', > > fildvscf='h2dv', > > amass(1)=1.008, > > outdir='$TMP_DIR/', > > fildyn=' h2.dyn', > > elph=.false., > > trans=.true., > > ldisp=.true., > > nq1=1, nq2=1, nq3=1 > > -------------------------------------------------------------------- > > > > Regards, > > Jin Zhang > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Dep. of Physics, Peking University Beijing, P.R.China Phone: 86-10-6275-3190 -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061127/a102bfc5/attachment.htm From jin.zhang.pku at gmail.com Mon Nov 27 14:30:44 2006 From: jin.zhang.pku at gmail.com (Jin Zhang) Date: Mon, 27 Nov 2006 21:30:44 +0800 Subject: [Pw_forum] question on structurerelax In-Reply-To: <2B82D1A3-7BBF-413F-956E-A3B50B36F202@nest.sns.it> References: <6234c0c60611270323u1e92f47ds9619065907429542@mail.gmail.com> <6234c0c60611270423o5261b90fhcb705bd9fa09d87d@mail.gmail.com> <2B82D1A3-7BBF-413F-956E-A3B50B36F202@nest.sns.it> Message-ID: <5d233b6a0611270530g17c2f61u5f6adc2013042e3a@mail.gmail.com> Hi, Paolo Gao mentioned he/she used experimental alat. I think this may be a better approach: 1. relax crytal, get alat A_0 and corresponding experimental alat B_0(under zero pressure) 2. get experimental alat B_1 under external pressure one's interested in, calculate the ratio c=B_1/B_0 3. set abinitio alat A_1 according to this ratio: A_1 = c * A_0, assuming they have same scaling behavior. I haven't done high-pressure calculations before. So this is just a guess. Correct me if I get on the wrong way. Thanks. Best, Jin Zhang On 11/27/06, Paolo Giannozzi wrote: > > > On Nov 27, 2006, at 13:23 , Guoying Gao wrote: > > > I'm sorry for my ambiguous expressions. I mean that I did > > a single scf calculation before relaxing. > > this is irrelevant: convergence of pressure wrt kinetic energy > cutoff is notoriously slow > > Paolo > --- > Paolo Giannozzi, Democritos and University of Udine, Italy > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Dep. of Physics, Peking University Beijing, P.R.China Phone: 86-10-62-753-190 -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061127/2b2bf9be/attachment.htm From giannozz at nest.sns.it Mon Nov 27 14:44:07 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 27 Nov 2006 14:44:07 +0100 Subject: [Pw_forum] Confused about vibrational modes of isolated molecule from ph.x In-Reply-To: <5d233b6a0611270514p1cf17ef9uebfc72c1d84b24a@mail.gmail.com> References: <5d233b6a0611182329s46725cd4mf772a3a75f786c9a@mail.gmail.com> <456036D2.6030901@lg.ehu.es> <5d233b6a0611190542q19bcd0edq94b8c17917357c8@mail.gmail.com> <45618283.1030403@sissa.it> <5d233b6a0611200427w27ad0305pc63196f73eb8ea44@mail.gmail.com> <5d233b6a0611270324xdf38aefw159e7a526d588e31@mail.gmail.com> <456AD184.8000404@sissa.it> <5d233b6a0611270514p1cf17ef9uebfc72c1d84b24a@mail.gmail.com> Message-ID: <48BD46EA-0B26-468D-88DF-58754694A496@nest.sns.it> On Nov 27, 2006, at 14:14 , Jin Zhang wrote: > BTW, is there a "tr1_ph"? This name convention is really interesting:) well, no, there isn't a "tr1_ph". The "2" in "tr2_ph" stands for "square". "tr2" used to be the name of the threshold on the square of the error of the potential during self-consistency in pwscf. When the phonon code was written, the threshold for its self-consistency loop was naturally (?) named "tr2_ph". Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From jibiaoli at gmail.com Mon Nov 27 14:51:24 2006 From: jibiaoli at gmail.com (Clark Lee) Date: Mon, 27 Nov 2006 21:51:24 +0800 Subject: [Pw_forum] How to calculate charge density at a specified energy level Message-ID: Dear QE users, Does anyone konows whether pp.x poccesses a function that is capable of calculating charge density at a specified energy level. -- Standing on the shoulders of Newton -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061127/9ca243ad/attachment.htm From gaoguoying at gmail.com Mon Nov 27 15:03:36 2006 From: gaoguoying at gmail.com (Guoying Gao) Date: Mon, 27 Nov 2006 22:03:36 +0800 Subject: [Pw_forum] question on structurerelax In-Reply-To: <2B82D1A3-7BBF-413F-956E-A3B50B36F202@nest.sns.it> References: <6234c0c60611270323u1e92f47ds9619065907429542@mail.gmail.com> <6234c0c60611270423o5261b90fhcb705bd9fa09d87d@mail.gmail.com> <2B82D1A3-7BBF-413F-956E-A3B50B36F202@nest.sns.it> Message-ID: <6234c0c60611270603x4a54e0d9vf9a6fbfd3ea8b632@mail.gmail.com> Dear Polo:, yes, I got it. Thank you a lot! gao From baroni at sissa.it Mon Nov 27 15:46:49 2006 From: baroni at sissa.it (Stefano Baroni) Date: Mon, 27 Nov 2006 15:46:49 +0100 Subject: [Pw_forum] question on structurerelax In-Reply-To: <6234c0c60611270323u1e92f47ds9619065907429542@mail.gmail.com> References: <6234c0c60611270323u1e92f47ds9619065907429542@mail.gmail.com> Message-ID: <217F7B14-2287-4BA5-A69F-1AED90AA7186@sissa.it> which system? we are decent physicists/chemists, but very poor as psychics ;-) S. On Nov 27, 2006, at 12:23 PM, Guoying Gao wrote: > Dear everyone, > When I did scf calculation with experimental parameters > (experimental pressure 80GPa), the output file gave P=164GPa, the bond > length is near equivalent to that of experiment, why so large diverge > in pressure? > Anyone who knows the reason is welcomed. Thanks in advance. > > gao > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061127/bb5202e2/attachment.htm From degironc at sissa.it Mon Nov 27 17:46:23 2006 From: degironc at sissa.it (degironc) Date: Mon, 27 Nov 2006 17:46:23 +0100 Subject: [Pw_forum] How to calculate charge density at a specified energy level In-Reply-To: References: Message-ID: <456B165F.2080907@sissa.it> what do you mean by a charge density at a specific energy level ? The square modulus of a specific wfc ? plotnum=3 does it at the fermi energy... should be simple to generalize The local density of states at a given energy ? plotnum=7 does it Please have a look to INPUT_PP stefano Clark Lee wrote: > Dear QE users, > > Does anyone konows whether pp.x poccesses a function that is capable > of calculating charge density at a specified energy level. > > -- > Standing on the shoulders of Newton From giannozz at nest.sns.it Mon Nov 27 17:58:19 2006 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 27 Nov 2006 17:58:19 +0100 Subject: [Pw_forum] If some geek wants some fun: pwscf on cell. In-Reply-To: <4568F89E.5080608@sissa.it> References: <4568F89E.5080608@sissa.it> Message-ID: <8C5EAEE7-F6A1-4EC9-BC47-4DF864E946DB@nest.sns.it> On Nov 26, 2006, at 3:14 , Antonio Suriano wrote: > This supercomputer costs only 599 euros, it is developed > by Sony and its name is Playstation 3. the major stumbling block is to convince the funding agency that you need a Playstation 3 for your research :-) Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From akohlmey at cmm.chem.upenn.edu Mon Nov 27 18:28:49 2006 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Mon, 27 Nov 2006 12:28:49 -0500 (EST) Subject: [Pw_forum] AIX + XML files In-Reply-To: Message-ID: On Mon, 27 Nov 2006, Gianni Profeta wrote: GP> Alex, GP> I changed the make.sys. GP> The compilation goes straight, but at the linking phase I have the GP> following : GP> GP> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% GP> ld: 0711-317 ERROR: Undefined symbol: .flush GP> ld: 0711-345 Use the -bloadmap or -bnoquiet option to obtain more GP> information. GP> make: 1254-004 The error code from the last command is 8. GP> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% GP> GP> I did not solve the problem by myself... hmmm, you may want to add the following to flib/flush_unit.f90 #if defined(__XLF) || defined(__ABSOFT) #define flush flush_ #endif that should be fixed in version 3.2, if i read the changelog correctly. axel. GP> GP> G. GP> GP> _______________________________________________ GP> Pw_forum mailing list GP> Pw_forum at pwscf.org GP> http://www.democritos.it/mailman/listinfo/pw_forum GP> -- ======================================================================= 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 suriano at sissa.it Mon Nov 27 19:33:46 2006 From: suriano at sissa.it (Antonio Suriano) Date: Mon, 27 Nov 2006 19:33:46 +0100 Subject: [Pw_forum] If some geek wants some fun: pwscf on cell. In-Reply-To: <8C5EAEE7-F6A1-4EC9-BC47-4DF864E946DB@nest.sns.it> References: <4568F89E.5080608@sissa.it> <8C5EAEE7-F6A1-4EC9-BC47-4DF864E946DB@nest.sns.it> Message-ID: <456B2F8A.2070205@sissa.it> Paolo Giannozzi ha scritto: > On Nov 26, 2006, at 3:14 , Antonio Suriano wrote: > >> This supercomputer costs only 599 euros, it is developed >> by Sony and its name is Playstation 3. > > the major stumbling block is to convince the funding agency > that you need a Playstation 3 for your research :-) > It is a good anti-stress. If well used can increase productivity of students :-D By the way a Xbox 360 is cheaper (399$) and has a three core evolution of G5 at 3 gigahertz as processors. It could be simpler to run espresso on this consolle. Chiacchiera con i tuoi amici in tempo reale! http://it.yahoo.com/mail_it/foot/*http://it.messenger.yahoo.com From wierzbom at ts.infn.it Tue Nov 28 11:11:16 2006 From: wierzbom at ts.infn.it (Malgorzata Wierzbowska) Date: Tue, 28 Nov 2006 11:11:16 +0100 (CET) Subject: [Pw_forum] question on structurerelax In-Reply-To: <217F7B14-2287-4BA5-A69F-1AED90AA7186@sissa.it> References: <6234c0c60611270323u1e92f47ds9619065907429542@mail.gmail.com> <217F7B14-2287-4BA5-A69F-1AED90AA7186@sissa.it> Message-ID: On Mon, 27 Nov 2006, Stefano Baroni wrote: > which system? > we are decent physicists/chemists, but very poor as psychics ;-) > S. I don't know what it means "psychics" but for sure we are all "psychopatic" because one of professors said: "Who works with legs has sic legs, who works with a head has a sic head". Gosia > > On Nov 27, 2006, at 12:23 PM, Guoying Gao wrote: > >> Dear everyone, >> When I did scf calculation with experimental parameters >> (experimental pressure 80GPa), the output file gave P=164GPa, the bond >> length is near equivalent to that of experiment, why so large diverge >> in pressure? >> Anyone who knows the reason is welcomed. Thanks in advance. >> >> gao >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum > > --- > Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste > [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) > > Please, if possible, don't send me MS Word or PowerPoint attachments > Why? See: http://www.gnu.org/philosophy/no-word-attachments.html > > > > From shuixblue at yahoo.com.cn Wed Nov 29 05:07:31 2006 From: shuixblue at yahoo.com.cn (shangyi) Date: Wed, 29 Nov 2006 12:07:31 +0800 (CST) Subject: [Pw_forum] Help on dipole and quadrupole moment! Message-ID: <20061129040731.93187.qmail@web15210.mail.cnb.yahoo.com> Dear all: I want to calculate the dipole and quadrupole moment of Ag atom when it adsorped on MgO slab using the one-side modle. How can I complete the calculation with pwscf? Though I examined the example 10 and searched the revelant information in maillist, I still have no ideas on it. So I try the calculation with H atom adsorped on 5Layer Al slab, the steps are: 1. relaxation the small system 2. contiune the calculation with the keyword "nscf" "lberry" "gdir" and "nppstr". I find it fails with information "Polarization is only for insulator and no empty band" if I using the "smearing" or the systme has odd electron. And it succeeds with even electron and "occupation=fixed". The results is: _______________________________________________________________________________________ P = -2.0127787 (mod 31.8198000) (e/Omega).bohr P = -0.0022491 (mod 0.0355556) e/bohr^2 P = -0.1285842 (mod 2.0327730) C/m^2 The polarization direction is: ( 0.00000 , 0.00000 , 1.00000 ) --------------------------------------------------------------------------------------- So I have three questions on the exercise: 1. Are the results diople moment with different units? 2. How can I get quadrupole moment? 3. How can I do if the system has odd electron? Any comments are appreciated! Thanks in advance! Best wishes! shuix _______________________________________ YM - ???? ??????????????????????????????????????????? http://messenger.yahoo.com.hk -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061129/18f74067/attachment.htm From rudrabnrj at gmail.com Thu Nov 30 06:26:06 2006 From: rudrabnrj at gmail.com (rudra Banerjee) Date: Thu, 30 Nov 2006 10:56:06 +0530 Subject: [Pw_forum] problem on grid Message-ID: <2e36f8e50611292126x4a98efa4k55a1316580d6d32b@mail.gmail.com> i am using static version of PWscf and i am trying to use it on grid(globusrun)...but it is showing from readpp : error # 29 file $HOME/rudra/pw/espresso-3.1/pseudo/Al.vbc.UPF not found can any1 plz help me on removing this possible error? thanking you -- rudra ------------------------- have a good time! -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061130/3be5b8f7/attachment.htm From baroni at sissa.it Thu Nov 30 08:01:19 2006 From: baroni at sissa.it (Stefano Baroni) Date: Thu, 30 Nov 2006 08:01:19 +0100 Subject: [Pw_forum] problem on grid In-Reply-To: <2e36f8e50611292126x4a98efa4k55a1316580d6d32b@mail.gmail.com> References: <2e36f8e50611292126x4a98efa4k55a1316580d6d32b@mail.gmail.com> Message-ID: Rudra: I know very little about globus, but this is one of the few instances where the PWscf error message is very clear. pw.x cannot simply find the pseudopotential file on the node where the application is actually being run. Most probably, you have installed QE one some machine and globus (or whatever) tries to execute the pw.x application on another ... S. On Nov 30, 2006, at 6:26 AM, rudra Banerjee wrote: > i am using static version of PWscf and i am trying to use it on grid > (globusrun)...but it is showing > > from readpp : error # 29 > file $HOME/rudra/pw/espresso-3.1/pseudo/Al.vbc.UPF not found > > can any1 plz help me on removing this possible error? > thanking you > -- > rudra > ------------------------- > have a good time! --- 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/20061130/76217d27/attachment.htm From babazade at gmail.com Thu Nov 30 08:17:47 2006 From: babazade at gmail.com (Amin Babazadeh) Date: Thu, 30 Nov 2006 10:47:47 +0330 Subject: [Pw_forum] use doual core In-Reply-To: <002e01c71207$5e733cb0$1d00a8c0@solarflare> References: <7b6913e90611220734s15038d2ei36e1e7a28166f146@mail.gmail.com> <002e01c71207$5e733cb0$1d00a8c0@solarflare> Message-ID: Thanks a lot for your answers.but until yet i couldn't solve my problem.Iconfused about mpi. there are three kind of mpi in the Internet (lam/mpi,openmpi,mpich) i don't know which one of them i should install. about making the environment i should say that i am beginner in using Linux so i use the ./configure commend and i didn't change any path because i don't know how i should do it. i will appreciate if answer my question? 1-which mpi do i need for my job? 2-how could i change the path? 3-is ifort compiler enough for installing mpi package? Thanks a lot On 11/27/06, Huiqun Zhou wrote: > > ? Have you set up you environment for using one the mpi libraries you > installed in .bashrc? > For example, for openmpi you need to set following in your .bashrc if you > are using bash > as your shell: > ... > export PATH=/opt/openmpi/bin:$PATH > export LD_LIBRARY_PATH=/opt/openmpi/lib:$LD_LIBRARY_PATH > ... > > Note: you need to replace /opt/openmpi with the path of your installation. > > Huiqun Zhou > > > ----- Original Message ----- > *From:* Amin Babazadeh > *To:* pw_forum at pwscf.org > *Sent:* Sunday, November 26, 2006 6:11 PM > *Subject:* Re: [Pw_forum] use doual core > > > Dear Dr.Kohlmeyer > Thanks a lot for your help.I did what you said.i installed > fftw-3.1.2.tar > lam-7.1.2.tar > mpich2-1.0.4p1.tar > openmpi-1.1.2.tar > on fedora core 5 and after that i configure the espresso but this error > was written > > WARNING: the following problems were detected: > > unsupported C/Fortran compilers combination: > CC=mpicc, F77=mpif77, F90=mpif90 > you may not be able to compile this program. > i don't know what this problem means.would you please help me. > Thanks a lot > > > > On 11/22/06, Axel Kohlmeyer wrote: > > > > On 11/22/06, Amin Babazadeh < babazade at gmail.com> wrote: > > > dear users > > > I use a computer with a dual core CPU.but my problem is that when i > > run a > > > program only one of the CPU work but i like to share the work (solving > > k > > > points) between cpus. > > > I think that i should install the parallel pwscf but when i run the > > > ./configure only single pwscf installed. > > > > you have to install an MPI package (e.g. LAM/MPI, OpenMPI, MPICH) > > first and then configure/compile quantum espresso with parallel support > > as explained in the provided installation instructions. > > > > please make sure that your MPI package is compatible with or compiled > > for the fortran compiler you want to use to compiled quantum espresso > > with. > > precompiled packages may be compiled for a different compiler and thus > > need special flags or environment variables set to use the compiler of > > your choice. > > > > axel. > > > > > would you please tell me or give me a document that solve my problem. > > > > > > > > > -- > > > AMIN > > > > > > -- > > ======================================================================= > > 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 > > > > > > -- > AMIN > > -- AMIN -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061130/c99d8943/attachment.htm From rudrabnrj at gmail.com Thu Nov 30 10:16:50 2006 From: rudrabnrj at gmail.com (rudra Banerjee) Date: Thu, 30 Nov 2006 14:46:50 +0530 Subject: [Pw_forum] problem on grid In-Reply-To: References: <2e36f8e50611292126x4a98efa4k55a1316580d6d32b@mail.gmail.com> Message-ID: <2e36f8e50611300116r270f2ccdx188de26184fe6ef9@mail.gmail.com> as i told, i have static version and compiled on P4. now i want it(my grid) is Xeon.should i compile it again? On 11/30/06, Stefano Baroni wrote: > > Rudra: I know very little about globus, but this is one of the few > instances where the PWscf error message is very clear. pw.x cannot simply > find the pseudopotential file on the node where the application is actually > being run. Most probably, you have installed QE one some machine and globus > (or whatever) tries to execute the pw.x application on another ... > S. > > On Nov 30, 2006, at 6:26 AM, rudra Banerjee wrote: > > i am using static version of PWscf and i am trying to use it on > grid(globusrun)...but it is showing > > from readpp : error # 29 > file $HOME/rudra/pw/espresso-3.1/pseudo/Al.vbc.UPF not found > > can any1 plz help me on removing this possible error? > thanking you > -- > rudra > ------------------------- > have a good time! > > > --- > 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 > > > > -- rudra ------------------------- have a good time! -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061130/bf9012e0/attachment.htm From degironc at sissa.it Thu Nov 30 11:07:59 2006 From: degironc at sissa.it (degironc) Date: Thu, 30 Nov 2006 11:07:59 +0100 Subject: [Pw_forum] problem on grid In-Reply-To: <2e36f8e50611300116r270f2ccdx188de26184fe6ef9@mail.gmail.com> References: <2e36f8e50611292126x4a98efa4k55a1316580d6d32b@mail.gmail.com> <2e36f8e50611300116r270f2ccdx188de26184fe6ef9@mail.gmail.com> Message-ID: <456EAD7F.1030703@sissa.it> Dear Rudra Banerjee, The error message says that the code does not find the file $HOME/rudra/pw/espresso-3.1/pseudo/Al.vbc.UPF It's not (yet) a problem of compilation, it is a problem of access to data. Either they aren't where you told the code they are, or they are read protected. Please notice that the code or the operative system didn't interpreted the $HOME variable in your file name. Try putting the explicit path. hope this help stefano rudra Banerjee wrote: > as i told, i have static version and compiled on P4. now i want it(my > grid) is Xeon.should i compile it again? > > On 11/30/06, *Stefano Baroni* < baroni at sissa.it > > wrote: > > Rudra: I know very little about globus, but this is one of the few > instances where the PWscf error message is very clear. pw.x cannot > simply find the pseudopotential file on the node where the > application is actually being run. Most probably, you have > installed QE one some machine and globus (or whatever) tries to > execute the pw.x application on another ... > > S. > > On Nov 30, 2006, at 6:26 AM, rudra Banerjee wrote: > >> i am using static version of PWscf and i am trying to use it on >> grid(globusrun)...but it is showing >> >> from readpp : error # 29 >> file $HOME/rudra/pw/espresso-3.1/pseudo/Al.vbc.UPF not found >> >> can any1 plz help me on removing this possible error? >> thanking you >> -- >> rudra >> ------------------------- >> have a good time! > > > --- > 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 > > > > > > > > -- > rudra > ------------------------- > have a good time! From ferretti.andrea at unimore.it Thu Nov 30 11:11:31 2006 From: ferretti.andrea at unimore.it (Andrea Ferretti) Date: Thu, 30 Nov 2006 11:11:31 +0100 (CET) Subject: [Pw_forum] problem on grid In-Reply-To: <2e36f8e50611300116r270f2ccdx188de26184fe6ef9@mail.gmail.com> References: <2e36f8e50611292126x4a98efa4k55a1316580d6d32b@mail.gmail.com> <2e36f8e50611300116r270f2ccdx188de26184fe6ef9@mail.gmail.com> Message-ID: maybe I am missing something, but it seems that the $HOME variable has not been substituted by the shell... try to put the actual path of the pseudofile in the input file... hope it helps andrea > as i told, i have static version and compiled on P4. now i want it(my grid) > is Xeon.should i compile it again? > > > from readpp : error # 29 > > file $HOME/rudra/pw/espresso-3.1/pseudo/Al.vbc.UPF not found > > > > can any1 plz help me on removing this possible error? > > thanking you From jin.zhang.pku at gmail.com Thu Nov 30 15:57:28 2006 From: jin.zhang.pku at gmail.com (Jin Zhang) Date: Thu, 30 Nov 2006 22:57:28 +0800 Subject: [Pw_forum] electron phonon interaction in ph.x Message-ID: <5d233b6a0611300657r143c8acek65a0c4fdc6e40777@mail.gmail.com> Dear all, I've read about sources in espresso/PH dir. Please correct me if my understanding about following quantities is incorrect. 1. elph_mat(nbnd, nbnd, ik, ipert): It is <\psi_ikq | dVscf(q) / dR_ipert | \psi_ik> while ipert is along one of three Cartesian direction for a particular atom. 2. dyn(iat, jat): The dynamical matrix which obeys -- dyn^{\diag} * amass * dyn = 1 Now, If I want get electron phonon matrix gkq( ibnd, jbnd, ik, imode) out, should I just do this? ---gkq(ibnd, jbnd, ik, imode) = \sum_{ipert} elph_mat(ibnd, jbnd, ik, ipert) * dyn(ipert, imode) The sqrt(mass) is hiden in dyn, so if I let gkq = gkq / sqrt( sqrt(w2(imode) ) ), will I get the final elph interaction matrix? Regards, Jin Zhang -- Dep. of Physics, Peking University Beijing, P.R.China Phone: 86-10-62-753-190 -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061130/a3cdb0cd/attachment.htm From jess.kondor at gmail.com Thu Nov 30 16:21:52 2006 From: jess.kondor at gmail.com (Jess Kondor) Date: Thu, 30 Nov 2006 09:21:52 -0600 Subject: [Pw_forum] smearing methods and relaxation Message-ID: <9d7293e10611300721p1d8d35b6tffa3dd8b28b58acb@mail.gmail.com> Hello, everyone, I have a question about smearing methods and relaxation, in particularly tetrahedron method and structural optimization in case of metals. VASP manual recommends do not use tetrahedron method in relaxation of metals because this method is not variational with respect to the partial occupancies, which means, that the calculated forces are wrong by 5 to 10 % in metals. Is it applicable in pwscf case? I am studying magnetic system (which is metal without U correction, and should be insulator with LSDA+U) and I am not sure it is good idea to use 'tetrahedra' in this case. What do you suggest? Regards, JK -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20061130/10261bf7/attachment.htm From marzari at MIT.EDU Thu Nov 30 16:32:58 2006 From: marzari at MIT.EDU (Nicola Marzari) Date: Thu, 30 Nov 2006 10:32:58 -0500 Subject: [Pw_forum] smearing methods and relaxation In-Reply-To: <9d7293e10611300721p1d8d35b6tffa3dd8b28b58acb@mail.gmail.com> References: <9d7293e10611300721p1d8d35b6tffa3dd8b28b58acb@mail.gmail.com> Message-ID: <456EF9AA.2010808@mit.edu> Hi Jess, you should use a smearing instead (usually methfessel-paxton or marzari-vanderbilt). You need to use a small value - try 0.4 eV (remember, pwscf uses Ry, so divide by 13.6), and make sure that whatever you want to calculate is converged with respect to k-points. 0.4eV might be too large for a lot of critical properties - so decrease it (and increase, by a lot, the kpoints) and see if things change. nicola Jess Kondor wrote: > > Hello, everyone, > > I have a question about smearing methods and relaxation, in > particularly tetrahedron method and structural optimization in case of > metals. VASP manual recommends do not use tetrahedron method in > relaxation of metals because this method is not variational with respect > to the partial occupancies, which means, that the calculated forces are > wrong by 5 to 10 % in metals. Is it applicable in pwscf case? I am > studying magnetic system (which is metal without U correction, and > should be insulator with LSDA+U) and I am not sure it is good idea to > use 'tetrahedra' in this case. What do you suggest? > > Regards, > > JK -- --------------------------------------------------------------------- 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 degironc at sissa.it Thu Nov 30 17:04:07 2006 From: degironc at sissa.it (degironc) Date: Thu, 30 Nov 2006 17:04:07 +0100 Subject: [Pw_forum] electron phonon interaction in ph.x In-Reply-To: <5d233b6a0611300657r143c8acek65a0c4fdc6e40777@mail.gmail.com> References: <5d233b6a0611300657r143c8acek65a0c4fdc6e40777@mail.gmail.com> Message-ID: <456F00F7.2050305@sissa.it> Jin Zhang wrote: > 2. dyn(iat, jat): The dynamical matrix which obeys -- dyn^{\diag} * > amass * dyn = 1 I don't think dymanical matric should obey this relation... EIGENVECTORS, U, of the generalized problem Dyn U = M U omega^2 will obey U^\dagger M U = 1 or something like that... while EIGENVECTORS, V=\sqrtM U of problem 1/sqrtM Dyn 1/sqrtM V = omega^2 V will obey V^\dagger V = 1 stefano From akohlmey at cmm.chem.upenn.edu Thu Nov 30 18:14:10 2006 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Thu, 30 Nov 2006 12:14:10 -0500 (EST) Subject: [Pw_forum] use doual core In-Reply-To: Message-ID: On Thu, 30 Nov 2006, Amin Babazadeh wrote: amin, AB> Thanks a lot for your answers.but until yet i couldn't solve my AB> problem.Iconfused about mpi. there are three kind of mpi in the AB> Internet there are _many_ more MPI implementations (some obsolete), not counting the commercial ones and platform/network specific implementations. MPI defines the programmers interface, i.e. what each function should do and what the arguments should look lile. (lam/mpi,openmpi,mpich) i don't know which one of them i should install. just one of them. since you have 'only' a single dual core node, it does not matter much. communication via shared memory is rather fast. i would recommend using LAM/MPI for convenience (make sure you defined -D__LAM in your make.sys file). AB> about making the environment i should say that i am beginner in AB> using Linux so i use the ./configure commend and i didn't change AB> any path because i don't know how i should do it. no need to hack the environment, this is only needed, if you have multiple MPI libraries installed (or a lazy sysadmin). AB> i will appreciate if answer my question? AB> 1-which mpi do i need for my job? (almost) any would do. see above. AB> 2-how could i change the path? you don't need to, just follow the installation instructions. AB> 3-is ifort compiler enough for installing mpi package? why not? you just have to make sure that you compile your mpi package, so that it uses ifort and not another fortran compiler. see the installation instructions of your MPI package of choice on how to specify your compiler preferences. please note, that all of this would be more appropriate to ask at an mpi support mailing list, as these are problems that are completely independent of quantum espresso. also, after installation of your mpi package of choice, please test it first with some _simple_ mpi programs. most packages supply a few testcases and examples. pay special attention to the fortran examples if you care about using mpi for quantum espresso. once you have verified that your mpi installation is working correctly (and only after that), you should move to trying to compiler quantum espresso in parallel... even if off-topic, i hope that there are some people around here, that will benefit from those clarifications. cheers, axel. -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot.