From van.pia at gmail.com Fri Feb 1 15:23:10 2008 From: van.pia at gmail.com (=?ISO-8859-1?Q?Fernando_C=E9sar_Lussani?=) Date: Fri, 1 Feb 2008 12:23:10 -0200 Subject: [Pw_forum] (no subject) Message-ID: <7dc15dee0802010623o609bce5ci2f79c5073c71ce9e@mail.gmail.com> Dear Pwscf users, I'm trying to generate a LDA pseudopotential for vanadium, but the program did not run correctly mainly due the lack of experience in generating PP. Too many nodes are appearing in the calculation of the pseudopotential when using ld1.x in Quantum Espresso package. The ld1.x doesn't generate the pseudopotential. It's seems to me that no one have build a pseudopotential of this kind using ld1.x, Opium or USPP codes. Have anyone build a LDA pseudo-potential for vanadium? What kind of trick could I use to get a pseudopotential like this? And if anyone have already created LDA PP's for V, could pass it to me ? my last input was: &input title='V', zed=23, rel=1, iswitch=3, beta=0.3, config='[Ar] 3d3 4s2', dft='PZ', prefix='psi', / &inputp lloc = 2, tm=.true., pseudotype = 1, file_pseudopw = 'v.lda.UPF', / 2 4S 4 0 2 0 2.0 2.0 3D 3 2 3 0 1.8 1.8 and i got in trouble a the test of my PP: ... ------------------------ End of All-electron run ------------------------ Solution not found in ascheqps for n,l= 4 0 Solution not found in ascheqps for n,l= 3 2 Solution not found in ascheqps for n,l= 4 0 Solution not found in ascheqps for n,l= 4 0 Solution not found in ascheqps for n,l= 4 0 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from ascheqps : error # 3 too many attempts %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... best regards, Fernando Lussani, Campinas, SP - Brazil -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080201/df47ca02/attachment.htm From Souraya.Goumri-Said at univ-lemans.fr Fri Feb 1 15:57:15 2008 From: Souraya.Goumri-Said at univ-lemans.fr (Souraya Goumri-Said) Date: Fri, 01 Feb 2008 15:57:15 +0100 Subject: [Pw_forum] how to run PWSCF using the graphical interface Message-ID: <47A3334B.2020907@univ-lemans.fr> Dear PWSCF user's, I am new user of PWSCF and I have 2 questions : 1- I am trying to run some examples under the graphical interface of PWSCF, but I have the following error message " can't read "::pwscf::settings(CHDENS)": no such element in array can't read "::pwscf::settings(CHDENS)": no such element in array while executing "::tclu::newset run(CHDENS,$moduleObj) $::pwscf::settings(CHDENS)" (procedure "_init" line 16) invoked from within "_init $guiObj" (procedure "::pwscf::run::run" line 7) invoked from within "::pwscf::run::run ::gUI0" invoked from within ".gui0.shellchildsite.menubar.menubar.run.menu invoke active" ("uplevel" body line 1) invoked from within "uplevel #0 [list $w invoke active]" (procedure "tk::MenuInvoke" line 47) invoked from within "tk::MenuInvoke .gui0.shellchildsite.menubar.menubar.run.menu 1" (command bound to event *2. *I a searching for the user guide related to the graphical interface of PWSCF : TK4GUIPWscf. Where I can find it, please? Sincerely yours, * *-- Dr. Souraya Goumri-Said Laboratoire de Physique de l'Etat Condens?, UMR CNRS 6087 Institut de Recherche en Ing?nierie Mol?culaire et Mat?riaux Fonctionnels, FR CNRS 2575 Universit? du Maine. Avenue Olivier Messiaen 72085 LE MANS Cedex 9 (France). Email : Souraya.Goumri-Said at univ-lemans.fr web : http://geocities.com/saidsouraya/ Tel : (+33) (0) 2 43 83 26 26 / Fax : (+33) (0) 2 02 43 83 35 18 From tone.kokalj at ijs.si Fri Feb 1 16:49:03 2008 From: tone.kokalj at ijs.si (Tone Kokalj) Date: Fri, 01 Feb 2008 16:49:03 +0100 Subject: [Pw_forum] how to run PWSCF using the graphical interface In-Reply-To: <47A3334B.2020907@univ-lemans.fr> References: <47A3334B.2020907@univ-lemans.fr> Message-ID: <1201880944.3299.43.camel@assam.ijs.si> On Fri, 2008-02-01 at 15:57 +0100, Souraya Goumri-Said wrote: > Dear PWSCF user's, > > I am new user of PWSCF and I have 2 questions : > > 1- I am trying to run some examples under the graphical interface of > PWSCF, but I have the following error message " > > can't read "::pwscf::settings(CHDENS)": no such element in array > can't read "::pwscf::settings(CHDENS)": no such element in array The PWgui is a bit outdated (new version is under preparation, though). To trick the gui, define the ::pwscf::settings(CHDENS) variable in your ~/.pwgui/pwgui.settings (if you don't have this directory/file, create one). Then add this line into the file: set ::pwscf::settings(CHDENS) echo Otherwise, you may save the prepared file by pwgui, and run manually. Regards, Tone From eyvaz_isaev at yahoo.com Sat Feb 2 01:23:56 2008 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Fri, 1 Feb 2008 16:23:56 -0800 (PST) Subject: [Pw_forum] how to run PWSCF using the graphical interface In-Reply-To: <47A3334B.2020907@univ-lemans.fr> Message-ID: <399742.97815.qm@web60322.mail.yahoo.com> Dear Souraya, Of course, PWGUI is an excellent tool, thanks Tone. From limkr at umich.edu Sat Feb 2 17:54:04 2008 From: limkr at umich.edu (Dong-Hee Lim) Date: Sat, 02 Feb 2008 11:54:04 -0500 Subject: [Pw_forum] NEB initial path change Message-ID: <20080202115404.p1lj7eqc8wsk4sww@web.mail.umich.edu> Dear PWscf members I was wondering if I can manually specify the initial path of NEB calculation. Manual says that the initial path is obtained interpolating between the specified configurations so that images are equispaced. Can I make the initial images unequispaced? Since I approximately know the final path of a reaction, I think the NEB calculation would be finised earlier if I make the initial path similar to the final path. Manual also says we can specify "intermediate_image". Can I do what I want using "intermediate_image" option, or does "intermediate_image" mean another geometry optimized configuration between the initial and final images? Thank you, Dong-Hee Lim -- Dong-Hee Lim, Ph.D. candidate Dept. of Civil and Environmental Engineering University of Michigan From hashem.yamani at gmail.com Sun Feb 3 18:54:19 2008 From: hashem.yamani at gmail.com (Hashem Al-Yamani) Date: Sun, 3 Feb 2008 19:54:19 +0200 Subject: [Pw_forum] plotrho error Message-ID: <82a593400802030954x205f8536i4299d291fa2905cd@mail.gmail.com> Dear All ; While am trying to plot charge density using plot rho i got the following error : forrtl: severe (24): end-of-file during read, unit 5, file stdin Image PC Routine Line Source plotrho.x 00000000008D0D5B Unknown Unknown Unknown plotrho.x 00000000008CF096 Unknown Unknown Unknown plotrho.x 00000000008A60B2 Unknown Unknown Unknown plotrho.x 000000000086BC9A Unknown Unknown Unknown plotrho.x 000000000086B8FB Unknown Unknown Unknown plotrho.x 00000000008889F2 Unknown Unknown Unknown plotrho.x 0000000000407C63 Unknown Unknown Unknown plotrho.x 0000000000406FAA Unknown Unknown Unknown libc.so.6 00002AAAAAFC85AA Unknown Unknown Unknown plotrho.x 0000000000406EEA Unknown Unknown Unknown Any suggestions would be very thankful ...... The out put file said : Input file > r0 : 0.0000 0.0000 0.0000 tau1 : 1.0000 1.0000 0.0000 tau2 : 0.0000 0.0000 1.0000 read 29 atomic positions output file > Read 60 * 60 grid Bounds: -0.101100 0.056120 min, max, # of levels > Out of Bounds! try again Bounds: -0.101100 0.056120 min, max, # of levels > Thanks alot and have a nice day . Sincerely ; Hashem Al-Yamani Computational Physics Department Jordan University -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080203/bca4873e/attachment.htm From akohlmey at cmm.chem.upenn.edu Sun Feb 3 19:12:54 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sun, 3 Feb 2008 13:12:54 -0500 (EST) Subject: [Pw_forum] plotrho error In-Reply-To: <82a593400802030954x205f8536i4299d291fa2905cd@mail.gmail.com> References: <82a593400802030954x205f8536i4299d291fa2905cd@mail.gmail.com> Message-ID: On Sun, 3 Feb 2008, Hashem Al-Yamani wrote: HA> Dear All ; HA> HA> While am trying to plot charge density using plot rho i got the following HA> error : HA> HA> forrtl: severe (24): end-of-file during read, unit 5, file stdin this means you have an error in the input. plotrho.x is an interactive program. don't run it with input redirection and check the output. axel. HA> Image PC Routine Line Source HA> plotrho.x 00000000008D0D5B Unknown Unknown Unknown HA> plotrho.x 00000000008CF096 Unknown Unknown Unknown HA> plotrho.x 00000000008A60B2 Unknown Unknown Unknown HA> plotrho.x 000000000086BC9A Unknown Unknown Unknown HA> plotrho.x 000000000086B8FB Unknown Unknown Unknown HA> plotrho.x 00000000008889F2 Unknown Unknown Unknown HA> plotrho.x 0000000000407C63 Unknown Unknown Unknown HA> plotrho.x 0000000000406FAA Unknown Unknown Unknown HA> libc.so.6 00002AAAAAFC85AA Unknown Unknown Unknown HA> plotrho.x 0000000000406EEA Unknown Unknown Unknown HA> HA> HA> HA> Any suggestions would be very thankful ...... HA> HA> The out put file said : HA> HA> HA> Input file > r0 : 0.0000 0.0000 0.0000 HA> tau1 : 1.0000 1.0000 0.0000 HA> tau2 : 0.0000 0.0000 1.0000 HA> read 29 atomic positions HA> output file > Read 60 * 60 grid HA> Bounds: -0.101100 0.056120 HA> min, max, # of levels > Out of Bounds! try again HA> Bounds: -0.101100 0.056120 HA> min, max, # of levels > HA> HA> HA> HA> Thanks alot and have a nice day . HA> HA> HA> Sincerely ; HA> Hashem Al-Yamani HA> Computational Physics Department HA> Jordan University HA> -- ======================================================================= 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 jkiitb at gmail.com Sun Feb 3 15:31:04 2008 From: jkiitb at gmail.com (jitendra mishra) Date: Sun, 03 Feb 2008 20:01:04 +0530 Subject: [Pw_forum] Send me an SMS Message-ID: f22d8f06a4403139e530843675eb7e12 An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080203/40f7f325/attachment.htm From fratesi at mater.unimib.it Mon Feb 4 09:17:04 2008 From: fratesi at mater.unimib.it (Guido Fratesi) Date: Mon, 4 Feb 2008 09:17:04 +0100 (CET) Subject: [Pw_forum] NEB initial path change In-Reply-To: <20080202115404.p1lj7eqc8wsk4sww@web.mail.umich.edu> References: <20080202115404.p1lj7eqc8wsk4sww@web.mail.umich.edu> Message-ID: Dear Dong-Hee, DHL> I was wondering if I can manually specify the initial path of NEB calculation. Ideally, you can prepare manually a prefix.path file, and use the "restart" option in the input of pw.x, but it's simpler to use intermediate images. DHL> Can I make the initial images unequispaced? Since I approximately know DHL> the final path of a reaction, I think the NEB calculation would be DHL> finised earlier if I make the initial path similar to the final path. DHL> Manual also says we can specify "intermediate_image". Can I do what I DHL> want using "intermediate_image" option, or does "intermediate_image" DHL> mean another geometry optimized configuration between the initial and DHL> final images? If you know a few configurations which are likely to occur during the reaction, in my opinion the best option is generally that you specify them as intermediate images. Those are just taken once, in order to construct the starting reaction path by some interpolation. Images in this path will still be equispaced, but will "pass close" to the images you have specified. This will generally be sufficient, unless you really know that the path will pass exactly by a given configuration. If you do exactly know this (how???), say for two configurations C1 C2, then you could split the path React-C1, C1-C2, C2-Products, and: * either run these separately * either merge these paths into a single one and run it: run each of them with pw.x with 0 neb steps (this will generate linear interpolation files), and then paste these files, manually or with path_merge.bash in pwtools/. Notice: if you know that the reaction proceeds through some (meta)stable intermediate, definitely the best option is to optimize this state, and split the NEB calculation before/after it. Hope this helps. Best wishes, Guido -- Guido Fratesi Dipartimento di Scienza dei Materiali Universita` degli Studi di Milano-Bicocca via Cozzi 53, 20125 Milano, Italy From chen_shao_hua197 at yahoo.com.tw Mon Feb 4 09:51:25 2008 From: chen_shao_hua197 at yahoo.com.tw (=?big5?q?=B3=AF=20=A4=D6=B5=D8?=) Date: Mon, 4 Feb 2008 16:51:25 +0800 (CST) Subject: [Pw_forum] problems in the el-ph/phonon calculations Message-ID: <743787.8403.qm@web73002.mail.tp2.yahoo.com> Dear forum's members, I got the error message from q2r.out : The transform length (ARG NO. 9) is not an allowed value. The next higher allowed value is (2). And I found the reply from Paolo in "search archive" : the code performs a FFT to bring force constants from reciprocal to real space (this is why it is called "q2r"). The FFTs in IBM ESSL libraries do not like to perform FFT with dimension 1 (or 3, or 5,...). The simplest soultion is to copy your dynamical matrices to another machine whose FFT's do not complain. Paolo So I copy dynamical matrices to another machine. But I got the error as below : forrtl: severe (59): list-directed I/O syntax error, unit 51, file /home/max/a2Fq2r.51 Image PC Routine Line Source q2r.x 0816D170 Unknown Unknown Unknown q2r.x 0816B671 Unknown Unknown Unknown q2r.x 08140BCE Unknown Unknown Unknown q2r.x 08109F4A Unknown Unknown Unknown q2r.x 0811FED5 Unknown Unknown Unknown q2r.x 0811F57C Unknown Unknown Unknown q2r.x 0804F8E3 Unknown Unknown Unknown q2r.x 0804B86F Unknown Unknown Unknown q2r.x 080498CA Unknown Unknown Unknown libc.so.6 40079EA0 Unknown Unknown Unknown q2r.x 080497F1 Unknown Unknown Unknown I still could get *.fc but could not get a2Fmatdyn.6?. Does anyone know why and how to solve this problem ? Thanks in advance max PhD candidate Physics department, Taiwan university, Taiwan ____________________________________________________________________________________ ????????????????http://tw.mini.yahoo.com/ From giannozz at nest.sns.it Mon Feb 4 15:16:31 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 04 Feb 2008 15:16:31 +0100 Subject: [Pw_forum] (no subject) In-Reply-To: <7dc15dee0802010623o609bce5ci2f79c5073c71ce9e@mail.gmail.com> References: <7dc15dee0802010623o609bce5ci2f79c5073c71ce9e@mail.gmail.com> Message-ID: <47A71E3F.1020009@nest.sns.it> Fernando C?sar Lussani wrote: > I'm trying to generate a LDA pseudopotential for vanadium [...] > What kind of trick could I use to get a pseudopotential like > this? for an element like V it is usually easier and safer to produce pseudopotentials with s and p semicore states in the valence. If not, the nonlinear core correction is a must. d as local part tends to produce "ghosts". And finally: > 4S 4 0 2 0 2.0 2.0 > 3D 3 2 3 0 1.8 1.8 should be 4S 1 0 2 0 2.0 2.0 3D 3 2 3 0 1.8 1.8 (pseudo S corresponds to N=1, pseudo D to N=3; P would be 2). All this and more is explained in the documentation. Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From giannozz at nest.sns.it Mon Feb 4 15:21:05 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 04 Feb 2008 15:21:05 +0100 Subject: [Pw_forum] How to run ground state molecular dynamics In-Reply-To: References: Message-ID: <47A71F51.3060002@nest.sns.it> Min Seung Kyu wrote: > Could you give me a sample input for ground state molecular dynamics > in NVT ensemble? I cannot, but I can tell you what to look for in the documentation: "ion_temperature", "tempw", "tolp", "delta_t", "nraise" . A number of different thermostats are implemented, but not Nose', in PWscf. Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From jakubas at ifpan.edu.pl Tue Feb 5 18:02:39 2008 From: jakubas at ifpan.edu.pl (Pawel Jakubas) Date: Tue, 5 Feb 2008 18:02:39 +0100 Subject: [Pw_forum] NEB vs 'direct' approach Message-ID: <000e01c86818$ea22c470$352c5194@jakubas> Dear Pwscf Users, I have started using CI NEB method implemented in espresso. I deal with native defects in bulk materials; let's say Cd interstitial in CdTe, which crystalizes in zinc blende structure. It is known that it diffuses along [111] direction between tetrahedral sites. Previously, to calculate barriers for diffusion I used 'direct' approach (I coined this name), which means I calculated total energies by fixing the position of an interstitial at a number of sites between metastable sites, and allowing all other atoms to relax (except one located at the outskirts of supercell, i. e., far away form the defect). The barriers heights calculated using NEB are usually higher than those obtained by 'direct' approach. I try to understand these differences. Because the diffusion occurs along [111] direction in straight line, so the path is to some extent in this case trivial (I am stressing it, because it is ok to choose nstep=1 in neb). Is there a clear, unanimous explanation why it is so? Are you familiarized with any paper in literature giving discussion of the dicrepancies between NEB results and those obtained using 'direct' approach? What is the physics behind it? Thank you very much in advance for any comments, hints and help. Pawel Jakubas PhD Student Piotr Boguslawski Group Institute of Physics Polish Academy of Sciences AL. Lotnikow 32/46 PL-02-668 Warsaw, POLAND group webpage: http://www.ifpan.edu.pl/SL-3/ my webpage: http://www.ifpan.edu.pl/SL-3/PJakubas/jakubas.html From giannozz at nest.sns.it Tue Feb 5 17:59:05 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 05 Feb 2008 17:59:05 +0100 Subject: [Pw_forum] Problem of bands data In-Reply-To: References: <479F7569.3000208@ncsu.edu> <47A053A5.3030304@nest.sns.it> Message-ID: <47A895D9.1070109@nest.sns.it> lan haiping wrote: > Dear All and Paolo, > I have changed kpoints, number of bands, smearing scheme, and cutoff > energy and density to try to obtain proper bandstructure of (7,0) nanotube. > But it seems my struggle does take effect on final result. A typical > bands plot is attached. such figure is plotted by plotbands.x > with bands data obtained from bands.x program. > Would you please give me some comments or hints ? don't use bands.x. Take the output from pw.x, remove all lines until the band structure starts, remove all "k =", add at the top a line like the following one, run plotbands.x . &plot nbnd= 80, nks= 21 / The bands.x code is there to help in plotting nice bands. If it doesn't help, as in this case, for rerasons I don't know, it is better not to use it. Reminder: messages larger than 40Kb are deleted (more exactly, they are held "for moderator approval" and automatically removed the day after) Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy -------------- next part -------------- A non-text attachment was scrubbed... Name: c70-nscf.ps.gz Type: application/x-gzip Size: 19273 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080205/bba4e6de/attachment-0002.bin -------------- next part -------------- A non-text attachment was scrubbed... Name: c70-nscf.dat.gz Type: application/x-gzip Size: 5839 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080205/bba4e6de/attachment-0003.bin From lanhaiping at gmail.com Tue Feb 5 19:20:21 2008 From: lanhaiping at gmail.com (lan haiping) Date: Tue, 5 Feb 2008 19:20:21 +0100 Subject: [Pw_forum] Problem of bands data In-Reply-To: <47A895D9.1070109@nest.sns.it> References: <479F7569.3000208@ncsu.edu> <47A053A5.3030304@nest.sns.it> <47A895D9.1070109@nest.sns.it> Message-ID: thank you, Paolo. i found this problem will occur for ifort 9.0 compilor, while no problem for pgf90 compilor. Bests, Hai-Ping On Feb 5, 2008 5:59 PM, Paolo Giannozzi wrote: > lan haiping wrote: > > > Dear All and Paolo, > > > I have changed kpoints, number of bands, smearing scheme, and cutoff > > energy and density to try to obtain proper bandstructure of (7,0) > nanotube. > > But it seems my struggle does take effect on final result. A typical > > bands plot is attached. such figure is plotted by plotbands.x > > with bands data obtained from bands.x program. > > > Would you please give me some comments or hints ? > > don't use bands.x. Take the output from pw.x, remove all lines > until the band structure starts, remove all "k =", add at the > top a line like the following one, run plotbands.x . > &plot nbnd= 80, nks= 21 / > The bands.x code is there to help in plotting nice bands. > If it doesn't help, as in this case, for rerasons I don't > know, it is better not to use it. > > Reminder: messages larger than 40Kb are deleted (more > exactly, they are held "for moderator approval" and > automatically removed the day after) > > Paolo > -- > Paolo Giannozzi, Democritos and University of Udine, Italy > -- Hai-Ping Lan Department of Electronics , Peking University , Bejing, 100871 lanhaiping at gmail.com, hplan at pku.edu.cn -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080205/39ba44f4/attachment.htm From baroni at sissa.it Tue Feb 5 20:24:14 2008 From: baroni at sissa.it (Stefano Baroni) Date: Tue, 05 Feb 2008 20:24:14 +0100 Subject: [Pw_forum] NEB vs 'direct' approach In-Reply-To: <000e01c86818$ea22c470$352c5194@jakubas> References: <000e01c86818$ea22c470$352c5194@jakubas> Message-ID: <47A8B7DE.6040208@sissa.it> Dear Pawel: your findings sound rather weird to me. I would have expected the contrary. At convergence, the NEB should produce the "Minimum Energy Path" (MEP), i.e. the path whose transition state has minimum energy (with respect to the transition states of all the other neighboring paths). I think this is discussed at length in the original literature, but it is enough to think to the defintinition of the MEP, a path for which the force is at any point tangent to the trajectory. Imagine of following such a path while climbing a mountain pass, and you will convince yourself easily that this path actually passes through the saddle. I do not know why you find a different result. An error in the QE implementation of the NEB would be an obvious explanation. Another possibility is that the discretization that you have (by necessity!) chosen to sample your "direct" path is so coarse that it actually misses the point of largest energy along the path, which is exactly what the prefix "CI" (for "climbing image") in the CI-NEB acronym is meant to avoid. Hope this helps Yours - Stefano B. Pawel Jakubas wrote: > Dear Pwscf Users, > > I have started using CI NEB method implemented in espresso. I deal with > native defects in bulk materials; let's say Cd interstitial in CdTe, which > crystalizes in zinc blende structure. > It is known that it diffuses along [111] direction between tetrahedral > sites. > > Previously, to calculate barriers for diffusion I used 'direct' approach (I > coined this name), which means I calculated total energies by fixing the > position of an interstitial at a number of sites between metastable sites, > and allowing all other atoms to relax (except one located at the outskirts > of supercell, i. e., far away form the defect). The barriers heights > calculated using NEB are usually higher than those obtained by 'direct' > approach. I try to understand these differences. Because the diffusion > occurs along [111] direction in straight line, so the path is to some extent > in this case trivial (I am stressing it, because it is ok to choose nstep=1 > in neb). > > Is there a clear, unanimous explanation why it is so? Are you familiarized > with any paper in literature giving discussion of the dicrepancies between > NEB results and those obtained using 'direct' approach? What is the physics > behind it? > > Thank you very much in advance for any comments, hints and help. > > Pawel Jakubas > PhD Student > Piotr Boguslawski Group > Institute of Physics > Polish Academy of Sciences > AL. Lotnikow 32/46 > PL-02-668 Warsaw, POLAND > group webpage: http://www.ifpan.edu.pl/SL-3/ > my webpage: http://www.ifpan.edu.pl/SL-3/PJakubas/jakubas.html > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From guido.roma at cea.fr Wed Feb 6 11:35:18 2008 From: guido.roma at cea.fr (Guido Roma) Date: Wed, 06 Feb 2008 11:35:18 +0100 Subject: [Pw_forum] NEB vs 'direct' approach In-Reply-To: <000e01c86818$ea22c470$352c5194@jakubas> References: <000e01c86818$ea22c470$352c5194@jakubas> Message-ID: <47A98D66.7030801@cea.fr> Dear Pawel, I am not familiar with CdTe, but I wouldn't be surprised if the energy profile that you are looking at shows a double peak structure. In that case, as suggested by Stefano, you can miss the real saddle point and only sample an intermediate local minimum, by such a coarse sampling (I assume that you take 1 point also for the calculation using constraints, that you call direct). In some comparison that I did for a different system (cubic SiC) I got the same saddle points with NEB or constrained relaxations, provided that I had enough points in both (I was using simple NEB, not CI). Guido Pawel Jakubas wrote: > Dear Pwscf Users, > > I have started using CI NEB method implemented in espresso. I deal with > native defects in bulk materials; let's say Cd interstitial in CdTe, which > crystalizes in zinc blende structure. > It is known that it diffuses along [111] direction between tetrahedral > sites. > > Previously, to calculate barriers for diffusion I used 'direct' approach (I > coined this name), which means I calculated total energies by fixing the > position of an interstitial at a number of sites between metastable sites, > and allowing all other atoms to relax (except one located at the outskirts > of supercell, i. e., far away form the defect). The barriers heights > calculated using NEB are usually higher than those obtained by 'direct' > approach. I try to understand these differences. Because the diffusion > occurs along [111] direction in straight line, so the path is to some extent > in this case trivial (I am stressing it, because it is ok to choose nstep=1 > in neb). > > Is there a clear, unanimous explanation why it is so? Are you familiarized > with any paper in literature giving discussion of the dicrepancies between > NEB results and those obtained using 'direct' approach? What is the physics > behind it? > > Thank you very much in advance for any comments, hints and help. > > Pawel Jakubas > PhD Student > Piotr Boguslawski Group > Institute of Physics > Polish Academy of Sciences > AL. Lotnikow 32/46 > PL-02-668 Warsaw, POLAND > group webpage: http://www.ifpan.edu.pl/SL-3/ > my webpage: http://www.ifpan.edu.pl/SL-3/PJakubas/jakubas.html > > _______________________________________________ > 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: guido_roma.vcf Type: text/x-vcard Size: 260 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080206/5364ec34/attachment.vcf From ierrea001 at ikasle.ehu.es Wed Feb 6 12:14:32 2008 From: ierrea001 at ikasle.ehu.es (Ion Errea) Date: Wed, 06 Feb 2008 12:14:32 +0100 Subject: [Pw_forum] format problem with "gam.lines" Message-ID: <1202296472.11653.36.camel@localhost> Dear PWscf users, I am using matdyn.x to calculate electron-phonon coefficients along some symmetry lines following the example07 of the quantum-espresso package. The system under study has 18 phonon modes, thus I should get in the output file gam.lines 18 different gamma values. That's exactly what I get but one of the gamma values appears as *****. The stars show up for the first gamma value after the line break. I attach the first lines of the file gam.lines: ------------------------------------ Gamma lines for all modes [THz] Broadening 0.0100 1 0.0000 0.0000 0.0000 0.0412 0.0249 0.0249 0.0382 0.0382 0.0002 ***** 0.2424 0.1969 0.1969 0.0003 0.0012 1.2793 0.6803 0.6803 2 0.0000 0.0005 0.0405 0.0007 0.0252 0.0262 0.0372 0.0378 0.0009 ***** 0.2402 0.1912 0.1904 0.0000 0.0013 1.2506 0.6757 0.6685 3 0.0379 0.0008 0.0019 0.0029 0.0259 0.0298 0.0356 0.0368 0.0017 ***** 0.2382 0.1702 0.1671 -0.0006 0.0016 0.7658 1.0627 0.6338 ------------------------------- (note that in the original output the line break is just before the stars) I guess this has to be a format problem with the writing of the output. I was wondering if any of the users has encounter this problem before. I would be very grateful if someone could give me some hints to solve the problem. Thanks in advance, Ion -- ======================================== Ion Errea Lope Materia Kondentsatuaren Fisika Saila Zientzia eta Teknologia Fakultatea Euskal Herriko Unibertsitatea 644 Posta Kutxatila, 48080 Bilbo, Spain Telefonoa: +34 946015322 Mail: ierrea001 at ikasle.ehu.es ======================================== From Vasile.Chis at physics.gu.se Wed Feb 6 13:26:19 2008 From: Vasile.Chis at physics.gu.se (Vasile Chis) Date: Wed, 06 Feb 2008 13:26:19 +0100 Subject: [Pw_forum] Phonons and charge density response Message-ID: <1202300779.12774.4.camel@zeeman.fy.chalmers.se> Dear users and developers, Running the ph.x one can extract the charge density response (fildrho) due to a phonon mode for an arbitrary phonon wave vector q. Thereafter one can presumably use pp.x to plot the induced charge density corresponding to mode in 1D, 2D or 3D (xcrysden). When this is done, the density is plotted for the super-cell no matter at which phonon wave vector q the linear response calculation was done. Then, I wonder if it is possible to obtain (visualizing, say with pp.x) the charge density of the corresponding wavelength? For example: One calculates a phonon mode with a wavelength which is twice the periodicity of the undistorted lattice in a particular direction. Let say that a top atom (in a surface slab) vibrates in the z-direction. Then when visualizing the charge density response for this mode in a 2D- plane along the wave vector, the first atom moves in the positive z- direction giving rise to a charge density pattern, then moving along the corresponding wave vector, the next surface atom should move in the negative z-direction giving rise to a charge density pattern with opposite sign. Although when extracting the above with pp.x and xcrysden, the charge density is just repeated (with the same sign) for the forthcoming atoms. It looks to me that the code consider the phonon induced density in the same way as a ground state density and thus benefiting from the symmetry of the unperturbed lattice. However, when a displacement field it set up by a phonon mode with a wave vector in the surface Brillouin zone this symmetry is broken and the periodicity deviates from the periodicity of the undistorted lattice. The information of the induced density variation caused by a phonon mode at an arbitrary q-vector should be possible to extract from a linear response calculation. The question is HOW? Looking forward to any replies. Best regards Vasse Chis -- Vasse Chis, Ph.D Student Department of Physics G?teborgs Universitet Fysikg?rden 6B S-412 96 G?teborg, SWEDEN tel :+46 (0) 31 7723338 email :Vasile.Chis at physics.gu.se From w2agz at pacbell.net Thu Feb 7 06:48:16 2008 From: w2agz at pacbell.net (Paul M. Grant) Date: Wed, 6 Feb 2008 21:48:16 -0800 Subject: [Pw_forum] Missing Libs? Message-ID: <000601c8694d$09969180$1cc3b480$@net> Have a look at below. I believe configure and make all ran smoothly. Am I missing a lib? pmpgrant at w2agz-hp6200 ~/O-sesame/examples/example05 $ ./run_example /home/pmpgrant/O-sesame/examples/example05 : starting This example shows how to use pw.x and postprocessing codes to make a contour plot in the [110] plane of the charge density for Si, and to plot the band structure of Si. executables directory: /home/pmpgrant/O-sesame/bin pseudo directory: /home/pmpgrant/O-sesame/pseudo temporary directory: /home/pmpgrant/tmp checking that needed directories and files exist... done running pw.x as: /home/pmpgrant/O-sesame/bin/pw.x running pp.x as: /home/pmpgrant/O-sesame/bin/pp.x running plotrho.x as: /home/pmpgrant/O-sesame/bin/plotrho.x running bands.x as: /home/pmpgrant/O-sesame/bin/bands.x running plotband.x as: /home/pmpgrant/O-sesame/bin/plotband.x cleaning /home/pmpgrant/tmp... done running the scf calculation... done running pp.x to do a 2-d plot of the charge density...######################## ############################################################################ #### ################ # FROM IOTK LIBRARY, VERSION 1.1.0development # UNRECOVERABLE ERROR (ierr=1) # ERROR IN: iotk_scan_end (iotk_scan.spp:211) # CVS Revision: 1.6 # foundl # ERROR IN: iotk_close_read (iotk_files.spp:589) # CVS Revision: 1.3 ############################################################################ #### ######################################## done running plotrho.x to generate rho.ps...At line 41 of file plotrho.f90 (unit = 1, file = '') Fortran runtime error: File '0 0.09 6' does not exist done running the band-structure calculation for Si... done running the post-processing for band structure...############################# ############################################################################ #### ########### # FROM IOTK LIBRARY, VERSION 1.1.0development # UNRECOVERABLE ERROR (ierr=1) # ERROR IN: iotk_scan_end (iotk_scan.spp:211) # CVS Revision: 1.6 # foundl # ERROR IN: iotk_close_read (iotk_files.spp:589) # CVS Revision: 1.3 ############################################################################ #### ######################################## done running plotband.x to generate sibands.ps...STOP Error reading file header done /home/pmpgrant/O-sesame/examples/example05: done pmpgrant at w2agz-hp6200 ~/O-sesame/examples/example05 $ Paul M. Grant, PhD Principal, W2AGZ Technologies Visiting Scholar, Applied Physics, Stanford University EPRI Science Fellow (Retired) IBM Research Staff Member Emeritus w2agz at pacbell.net http://www.w2agz.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080206/7b1097e1/attachment.htm From w2agz at pacbell.net Thu Feb 7 06:51:50 2008 From: w2agz at pacbell.net (Paul M. Grant) Date: Wed, 6 Feb 2008 21:51:50 -0800 Subject: [Pw_forum] espressocvs question(s) Message-ID: <000b01c8694d$89031ce0$9b0956a0$@net> I've noticed that the folders examples and pseudo (maybe some others?) are missing from the espressocvs distro.??? For some time I've been curious about the contents of the folder "tests".some of the "IN" files are interesting.is there a "guide" to tests and why it exists? Paul M. Grant, PhD Principal, W2AGZ Technologies Visiting Scholar, Applied Physics, Stanford University EPRI Science Fellow (Retired) IBM Research Staff Member Emeritus w2agz at pacbell.net http://www.w2agz.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080206/4c472c95/attachment-0001.htm From akohlmey at cmm.chem.upenn.edu Thu Feb 7 08:18:40 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Thu, 7 Feb 2008 02:18:40 -0500 (EST) Subject: [Pw_forum] Missing Libs? In-Reply-To: <000601c8694d$09969180$1cc3b480$@net> References: <000601c8694d$09969180$1cc3b480$@net> Message-ID: On Wed, 6 Feb 2008, Paul M. Grant wrote: hi paul, PG> Have a look at below. I believe configure and make all ran smoothly. Am I PG> missing a lib? which "version" is this? the current cvs? it is unlikely due to a library. it looks like your compile of iotk has problems reading back what it wrote or may even have been miscompiled. for the latter case it would be nice to know OS/compiler/settings etc. the same part of the calculation works for me. see below. cheers, axel. PG> PG> pmpgrant at w2agz-hp6200 ~/O-sesame/examples/example05 PG> $ ./run_example PG> PG> /home/pmpgrant/O-sesame/examples/example05 : starting PG> PG> This example shows how to use pw.x and postprocessing codes to make a PG> contour plot in the [110] plane of the charge density for Si, and to PG> plot the band structure of Si. PG> PG> executables directory: /home/pmpgrant/O-sesame/bin PG> pseudo directory: /home/pmpgrant/O-sesame/pseudo PG> temporary directory: /home/pmpgrant/tmp PG> checking that needed directories and files exist... done PG> PG> running pw.x as: /home/pmpgrant/O-sesame/bin/pw.x PG> running pp.x as: /home/pmpgrant/O-sesame/bin/pp.x PG> running plotrho.x as: /home/pmpgrant/O-sesame/bin/plotrho.x PG> running bands.x as: /home/pmpgrant/O-sesame/bin/bands.x PG> running plotband.x as: /home/pmpgrant/O-sesame/bin/plotband.x PG> PG> cleaning /home/pmpgrant/tmp... done PG> running the scf calculation... done PG> running pp.x to do a 2-d plot of the charge PG> density...######################## here is what i compiled off the cvs as of now with intel-10.1 on an x86 pentium M laptop running fedora 8: /home/akohlmey/compile/pwscf-cvs/examples/example05 : starting This example shows how to use pw.x and postprocessing codes to make a contour plot in the [110] plane of the charge density for Si, and to plot the band structure of Si. executables directory: /home/akohlmey/compile/pwscf-cvs/bin pseudo directory: /home/akohlmey/compile/pwscf-cvs/pseudo temporary directory: /home/akohlmey/tmp checking that needed directories and files exist... done running pw.x as: /home/akohlmey/compile/pwscf-cvs/bin/pw.x running pp.x as: /home/akohlmey/compile/pwscf-cvs/bin/pp.x running plotrho.x as: /home/akohlmey/compile/pwscf-cvs/bin/plotrho.x running bands.x as: /home/akohlmey/compile/pwscf-cvs/bin/bands.x running plotband.x as: /home/akohlmey/compile/pwscf-cvs/bin/plotband.x cleaning /home/akohlmey/tmp... done running the scf calculation... done running pp.x to do a 2-d plot of the charge density... done running plotrho.x to generate rho.ps... done running the band-structure calculation for Si... done .... PG> ############################################################################ PG> #### PG> ################ PG> # FROM IOTK LIBRARY, VERSION 1.1.0development PG> # UNRECOVERABLE ERROR (ierr=1) PG> # ERROR IN: iotk_scan_end (iotk_scan.spp:211) PG> # CVS Revision: 1.6 PG> # foundl PG> # ERROR IN: iotk_close_read (iotk_files.spp:589) PG> # CVS Revision: 1.3 PG> ############################################################################ PG> #### PG> ######################################## PG> done PG> running plotrho.x to generate rho.ps...At line 41 of file plotrho.f90 PG> (unit = PG> 1, file = '') PG> Fortran runtime error: File '0 0.09 6' does not exist PG> done PG> running the band-structure calculation for Si... done PG> running the post-processing for band PG> structure...############################# PG> ############################################################################ PG> #### PG> ########### PG> # FROM IOTK LIBRARY, VERSION 1.1.0development PG> # UNRECOVERABLE ERROR (ierr=1) PG> # ERROR IN: iotk_scan_end (iotk_scan.spp:211) PG> # CVS Revision: 1.6 PG> # foundl PG> # ERROR IN: iotk_close_read (iotk_files.spp:589) PG> # CVS Revision: 1.3 PG> ############################################################################ PG> #### PG> ######################################## PG> done PG> running plotband.x to generate sibands.ps...STOP Error reading file header PG> done PG> PG> /home/pmpgrant/O-sesame/examples/example05: done PG> PG> pmpgrant at w2agz-hp6200 ~/O-sesame/examples/example05 PG> $ PG> PG> Paul M. Grant, PhD PG> Principal, W2AGZ Technologies PG> Visiting Scholar, Applied Physics, Stanford University PG> EPRI Science Fellow (Retired) PG> IBM Research Staff Member Emeritus PG> w2agz at pacbell.net PG> http://www.w2agz.com PG> PG> PG> PG> PG> -- ======================================================================= 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 Thu Feb 7 08:24:48 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Thu, 7 Feb 2008 02:24:48 -0500 (EST) Subject: [Pw_forum] espressocvs question(s) In-Reply-To: <000b01c8694d$89031ce0$9b0956a0$@net> References: <000b01c8694d$89031ce0$9b0956a0$@net> Message-ID: On Wed, 6 Feb 2008, Paul M. Grant wrote: PG> I've noticed that the folders examples and pseudo (maybe some others?) are PG> missing from the espressocvs distro.??? i have them in my cvs checkout. you may have to explicitely check them out. if you check out the O-sesame module (the original code name of the QE package) it should be contained. PG> PG> For some time I've been curious about the contents of the folder PG> "tests".some of the "IN" files are interesting.is there a "guide" to tests PG> and why it exists? i guess paolo will be more qualified to answer this. the purpose is obvious from the name of the folder. in general you can get very useful information of what is going on in the cvs by running "perl cvs2cl.pl" in the main directory and reading through the resulting ChangeLog file. very interesting read at times and highly recommended for everybody using the cvs code. cheers, axel. PG> PG> Paul M. Grant, PhD PG> Principal, W2AGZ Technologies PG> Visiting Scholar, Applied Physics, Stanford University PG> EPRI Science Fellow (Retired) PG> IBM Research Staff Member Emeritus PG> w2agz at pacbell.net PG> http://www.w2agz.com PG> PG> PG> PG> PG> -- ======================================================================= 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 Thu Feb 7 10:26:13 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 7 Feb 2008 10:26:13 +0100 Subject: [Pw_forum] espressocvs question(s) In-Reply-To: <000b01c8694d$89031ce0$9b0956a0$@net> References: <000b01c8694d$89031ce0$9b0956a0$@net> Message-ID: <92EEEAA7-9635-4991-93B0-9A8DCF375FA7@nest.sns.it> On Feb 7, 2008, at 6:51 , Paul M. Grant wrote: > I?ve noticed that the folders examples and pseudo (maybe some > others?) are missing from the espressocvs distro???? > the cvs snapshot I uploaded some time ago on the download page is not complete with examples and pseudos, I don't remember why. As soon as the download page is updated to something more functional, I'll upload a new complete cvs snapshot, together with the stable version that has unfortunately disappeared into the "previous versions" section. Sorry for the mess. > For some time I?ve been curious about the contents of the folder > ?tests??some of the ?IN? files are interesting?is there a ?guide? > to tests and why it exists? > there is no "guide" and there will never be one, I guess. The tests directory is an attempt to write better automated tests than those presently available in the 'tests and examples' directory examples/. Tests should be - fast to execute (i.e. tens of minutes at most, not hours) - exhaustive (i.e. test all features) - modular (i.e. test a single code at the time) - boolean (i.e. the result should be ok or not ok) Examples should be - realistic (i.e. reflect real-life cases) - simple to read and to understand What is contained in examples/ iobeys to none of the above criteria. This is why I started to write tests/ . I am quite satisfied with the result, but unfortunately only pw.x is covered. Attached: the results of tonight's tests Paolo -------------- next part -------------- A non-text attachment was scrubbed... Name: RESULTS OF DAILY QE TEST Type: application/octet-stream Size: 4193 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080207/61d4ca6a/attachment.obj -------------- next part -------------- --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From giannozz at nest.sns.it Thu Feb 7 10:44:39 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 7 Feb 2008 10:44:39 +0100 Subject: [Pw_forum] Missing Libs? In-Reply-To: <000601c8694d$09969180$1cc3b480$@net> References: <000601c8694d$09969180$1cc3b480$@net> Message-ID: <9F74CE3C-5701-4074-99FF-4E1EAAF08E76@nest.sns.it> On Feb 7, 2008, at 6:48 , Paul M. Grant wrote: > Have a look at below. I believe configure and make all ran > smoothly. Am I missing a lib? > more likely, you are missing a compiler that works. Did you use gfortran by any chance? it has a number of problems in compiling the CVS version of q- e. For some of them, a workaround is implemented in the CVS version (see the __GFORTRAN preprocessing option), but others are still there. Among these: sometimes q-e cannot read the xml files it has just written. Since gfortran is present on all recent Linux distributions, and produces quite fast executables, it would be nice to have q-e 100% working out of the box. Unfortunately it looks like we'll have to wait until gfortran is more stable. Paolo --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From subhra at iitg.ernet.in Thu Feb 7 10:40:16 2008 From: subhra at iitg.ernet.in (Subhradip Ghosh) Date: Thu, 7 Feb 2008 15:10:16 +0530 (IST) Subject: [Pw_forum] partial phonon DOS Message-ID: <34273.172.16.48.18.1202377216.squirrel@webmail.iitg.ernet.in> Sometime ago, Eyvaz mentioned that he has submitted his code to calculate partial phonon DOS. Has it been incorporated in the present version of espresso? Regards, Subhradip ************************************************************************ Dr. Subhradip Ghosh Assistant Professor Department of Physics Indian Institute of Technology Guwahati,Assam-781039 India E-mail:subhra at iitg.ernet.in Phone: +91 361 2582717(O) +91 361 2584717(R) Fax: +91 361 2582749 (Physics) +91 361 2690762 (General) ************************************************************************ From igor at ffh.bg.ac.yu Thu Feb 7 12:42:53 2008 From: igor at ffh.bg.ac.yu (igor at ffh.bg.ac.yu) Date: Thu, 7 Feb 2008 12:42:53 +0100 Subject: [Pw_forum] problem with instalation Message-ID: <1202384573.47aaeebde2337@mail.ffh.bg.ac.yu> Hi, I have machine with two quadcore procesors and 16 GB RAM, and I am trying to make QE to work. When I run ./configure I get this -------------------------------------------------------------------- ESPRESSO can take advantage of several optimized numerical libraries (essl, fftw, mkl...). This configure script attempts to find them, but may fail if they have been installed in non-standard locations. The following libraries have been found: BLAS_LIBS= LAPACK_LIBS= FFT_LIBS= Please check if that is correct. If any libraries are missing, you may specify a list of directories to search and retry, as follows: ./configure LIBDIRS="list of directories, separated by spaces" Parallel environment detected successfully. Configured for compilation of parallel executables. For more info, read the ESPRESSO User's Guide (Doc/users-guide.tex). -------------------------------------------------------------------- configure: success but when I run make all error is returned: bp_c_phase.o: In function `c_phase_': bp_c_phase.F90:(.text+0x12c7): undefined reference to `ylm_q_' bp_radin.o: In function `radlg_': bp_radin.f:(.text+0x128): undefined reference to `s_wsle' bp_radin.f:(.text+0x146): undefined reference to `do_lio' bp_radin.f:(.text+0x150): undefined reference to `e_wsle' bp_radin.f:(.text+0x15f): undefined reference to `s_wsle' bp_radin.f:(.text+0x17d): undefined reference to `do_lio' bp_radin.f:(.text+0x199): undefined reference to `do_lio' bp_radin.f:(.text+0x1a3): undefined reference to `e_wsle' bp_radin.f:(.text+0x1b7): undefined reference to `s_stop' bp_radin.o: In function `radlg1_': bp_radin.f:(.text+0x278): undefined reference to `s_wsle' bp_radin.f:(.text+0x296): undefined reference to `do_lio' bp_radin.f:(.text+0x2a0): undefined reference to `e_wsle' bp_radin.f:(.text+0x2af): undefined reference to `s_wsle' bp_radin.f:(.text+0x2cd): undefined reference to `do_lio' bp_radin.f:(.text+0x2e9): undefined reference to `do_lio' bp_radin.f:(.text+0x2f3): undefined reference to `e_wsle' bp_radin.f:(.text+0x307): undefined reference to `s_stop' c_phase_field.o: In function `c_phase_field_': c_phase_field.F90:(.text+0x10af): undefined reference to `ylm_q_' h_epsi_her.o: In function `h_epsi_her_': h_epsi_her.F90:(.text+0x1071): undefined reference to `ylm_q_' h_epsi_her.F90:(.text+0x12f6): undefined reference to `ylm_q_' startup.o: In function `startup_': startup.F90:(.text+0x7a): undefined reference to `iargc_' startup.F90:(.text+0x23d): undefined reference to `iargc_' ../flib/flib.a(dlamch.o): In function `dlamc2_': dlamch.f:(.text+0x1283): undefined reference to `s_wsfe' dlamch.f:(.text+0x129c): undefined reference to `do_fio' dlamch.f:(.text+0x12a6): undefined reference to `e_wsfe' ../flib/flib.a(lapack.o): In function `xerbla_': lapack.f:(.text+0x4c9b): undefined reference to `s_wsfe' lapack.f:(.text+0x4cb2): undefined reference to `do_fio' lapack.f:(.text+0x4cc9): undefined reference to `do_fio' lapack.f:(.text+0x4cd3): undefined reference to `e_wsfe' lapack.f:(.text+0x4ce7): undefined reference to `s_stop' ../flib/flib.a(lapack.o): In function `zlanhp_': lapack.f:(.text+0xa7fe): undefined reference to `z_abs' lapack.f:(.text+0xa8c4): undefined reference to `z_abs' lapack.f:(.text+0xa9bb): undefined reference to `z_abs' lapack.f:(.text+0xaaec): undefined reference to `z_abs' ../flib/flib.a(lapack.o): In function `ilaenv_': lapack.f:(.text+0x10303): undefined reference to `s_copy' lapack.f:(.text+0x1035e): undefined reference to `s_copy' lapack.f:(.text+0x103c8): undefined reference to `s_copy' lapack.f:(.text+0x103e8): undefined reference to `s_copy' lapack.f:(.text+0x10405): undefined reference to `s_copy' lapack.f:(.text+0x10441): undefined reference to `s_cmp' lapack.f:(.text+0x10462): undefined reference to `s_cmp' lapack.f:(.text+0x1048e): undefined reference to `s_cmp' lapack.f:(.text+0x104ae): undefined reference to `s_cmp' lapack.f:(.text+0x104ce): undefined reference to `s_cmp' ../flib/flib.a(lapack.o):lapack.f:(.text+0x104ee): more undefined references to `s_cmp' follow ../flib/flib.a(lapack.o): In function `zlanhe_': lapack.f:(.text+0x1a03b): undefined reference to `z_abs' lapack.f:(.text+0x1a139): undefined reference to `z_abs' lapack.f:(.text+0x1a244): undefined reference to `z_abs' lapack.f:(.text+0x1a3b9): undefined reference to `z_abs' ../flib/flib.a(lapack.o): In function `zunmtr_': lapack.f:(.text+0x1e682): undefined reference to `s_cat' lapack.f:(.text+0x1e752): undefined reference to `s_cat' lapack.f:(.text+0x1e82e): undefined reference to `s_cat' lapack.f:(.text+0x1e8fe): undefined reference to `s_cat' ../flib/flib.a(lapack.o): In function `zunmql_': lapack.f:(.text+0x23241): undefined reference to `s_cat' ../flib/flib.a(lapack.o):lapack.f:(.text+0x23414): more undefined references to `s_cat' follow ../flib/flib.a(lapack.o): In function `zlange_': lapack.f:(.text+0x3534e): undefined reference to `z_abs' lapack.f:(.text+0x35400): undefined reference to `z_abs' lapack.f:(.text+0x354d1): undefined reference to `z_abs' ../flib/flib.a(lapack.o): In function `zggbal_': lapack.f:(.text+0x36c9b): undefined reference to `z_abs' lapack.f:(.text+0x36d27): undefined reference to `z_abs' ../flib/flib.a(lapack.o):lapack.f:(.text+0x36e6e): more undefined references to `z_abs' follow ../flib/flib.a(lapack.o): In function `zhgeqz_': lapack.f:(.text+0x3a97e): undefined reference to `z_sqrt' lapack.f:(.text+0x3bbb4): undefined reference to `z_abs' ../flib/flib.a(lapack.o): In function `zlanhs_': lapack.f:(.text+0x3fbff): undefined reference to `z_abs' lapack.f:(.text+0x3fcbe): undefined reference to `z_abs' lapack.f:(.text+0x3fda1): undefined reference to `z_abs' ../flib/flib.a(lapack.o): In function `ztrtri_': lapack.f:(.text+0x40ff5): undefined reference to `s_cat' ../flib/flib.a(lapack.o): In function `zgesvd_': lapack.f:(.text+0x42570): undefined reference to `s_cat' ../flib/flib.a(lapack.o): In function `zunmbr_': lapack.f:(.text+0x546ad): undefined reference to `s_cat' lapack.f:(.text+0x5477d): undefined reference to `s_cat' lapack.f:(.text+0x54858): undefined reference to `s_cat' ../flib/flib.a(lapack.o):lapack.f:(.text+0x54928): more undefined references to `s_cat' follow collect2: ld returned 1 exit status make[1]: *** [pw.x] Error 1 make[1]: Leaving directory `/quantumespresso/espresso-3.1.1/PW' make: *** [pw] Error 2 Can anybody help me? Thanks Igor From ceresoli at sissa.it Thu Feb 7 14:03:14 2008 From: ceresoli at sissa.it (Davide Ceresoli) Date: Thu, 07 Feb 2008 14:03:14 +0100 Subject: [Pw_forum] problem with instalation In-Reply-To: <1202384573.47aaeebde2337@mail.ffh.bg.ac.yu> References: <1202384573.47aaeebde2337@mail.ffh.bg.ac.yu> Message-ID: <47AB0192.5050903@sissa.it> igor at ffh.bg.ac.yu wrote: > bp_c_phase.o: In function `c_phase_': > bp_c_phase.F90:(.text+0x12c7): undefined reference to `ylm_q_' > bp_radin.o: In function `radlg_': > bp_radin.f:(.text+0x128): undefined reference to `s_wsle' This topic has been discussed several times in this mailing list. Search on google: undefined reference s_wsle site:democritos.it Davide From giannozz at nest.sns.it Thu Feb 7 14:10:47 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 7 Feb 2008 14:10:47 +0100 Subject: [Pw_forum] problem with instalation In-Reply-To: <47AB0192.5050903@sissa.it> References: <1202384573.47aaeebde2337@mail.ffh.bg.ac.yu> <47AB0192.5050903@sissa.it> Message-ID: On Feb 7, 2008, at 14:03 , Davide Ceresoli wrote: > igor at ffh.bg.ac.yu wrote: >> bp_c_phase.o: In function `c_phase_': >> bp_c_phase.F90:(.text+0x12c7): undefined reference to `ylm_q_' >> bp_radin.o: In function `radlg_': >> bp_radin.f:(.text+0x128): undefined reference to `s_wsle' > > This topic has been discussed several times in this mailing list. > > Search on google: undefined reference s_wsle site:democritos.it it might be more complicated than this. It looks like the f90 and the f77 compilers are not compatible. Older "configure" occasionally did this (that q-e version is quite old). Likely g77 was selected as f77 compiler. Paolo --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From eyvaz_isaev at yahoo.com Thu Feb 7 14:39:07 2008 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Thu, 7 Feb 2008 05:39:07 -0800 (PST) Subject: [Pw_forum] partial phonon DOS In-Reply-To: <34273.172.16.48.18.1202377216.squirrel@webmail.iitg.ernet.in> Message-ID: <292041.92615.qm@web60316.mail.yahoo.com> Hi, Unfortunately, not yet. But I can send it to you, of course. It is fully compatible with Espresso output fies. The problem is that I am "absolutely" busy with lots of projects, and had no time to check it for the last chance. Hopefully, the code will be incorporated into Espresso in the next future. Bests, Eyvaz. --- Subhradip Ghosh wrote: > Sometime ago, Eyvaz mentioned that he has submitted > his code to calculate > partial phonon DOS. Has it been incorporated in the > present version of > espresso? > > Regards, > > Subhradip > > > ************************************************************************ > Dr. Subhradip Ghosh > Assistant Professor > Department of Physics > Indian Institute of Technology > Guwahati,Assam-781039 > India > E-mail:subhra at iitg.ernet.in > Phone: +91 361 2582717(O) > +91 361 2584717(R) > Fax: +91 361 2582749 (Physics) > +91 361 2690762 (General) > ************************************************************************ > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, IFM, Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com ____________________________________________________________________________________ Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ From giannozz at nest.sns.it Thu Feb 7 14:55:27 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 7 Feb 2008 14:55:27 +0100 Subject: [Pw_forum] Missing Libs? In-Reply-To: <000601c8694d$09969180$1cc3b480$@net> References: <000601c8694d$09969180$1cc3b480$@net> Message-ID: <3A8059EA-0311-416D-AD6A-FD5C0C236CD5@nest.sns.it> On Feb 7, 2008, at 6:48 , Paul M. Grant wrote: > # FROM IOTK LIBRARY, VERSION 1.1.0development > > # UNRECOVERABLE ERROR (ierr=1) > quick fix: change Q_REAL in Q_REAL_SPACE in Modules/xml_io_base.f90 (thanks to Giovanni Bussi). This was a harmless q-e espresso bug in most cases, but it was (likely) acctivating a gfortran bug that affects iotk when a tag that is not present in the file is searched for... Paolo --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From subhra at iitg.ernet.in Thu Feb 7 15:06:17 2008 From: subhra at iitg.ernet.in (Subhradip Ghosh) Date: Thu, 7 Feb 2008 19:36:17 +0530 (IST) Subject: [Pw_forum] partial phonon DOS In-Reply-To: <292041.92615.qm@web60316.mail.yahoo.com> References: <292041.92615.qm@web60316.mail.yahoo.com> Message-ID: <1047.172.16.109.56.1202393177.squirrel@webmail.iitg.ernet.in> Dear Eyvaz, Please send it to me if possible. Subhradip > Hi, > > Unfortunately, not yet. But I can send it to you, of > course. It is fully compatible with Espresso output > fies. > The problem is that I am "absolutely" busy with lots > of projects, and had no time to check it for the last > chance. > > Hopefully, the code will be incorporated into Espresso > in the next future. > > Bests, > Eyvaz. > > --- Subhradip Ghosh wrote: > >> Sometime ago, Eyvaz mentioned that he has submitted >> his code to calculate >> partial phonon DOS. Has it been incorporated in the >> present version of >> espresso? >> >> Regards, >> >> Subhradip >> >> >> > ************************************************************************ >> Dr. Subhradip Ghosh >> Assistant Professor >> Department of Physics >> Indian Institute of Technology >> Guwahati,Assam-781039 >> India >> E-mail:subhra at iitg.ernet.in >> Phone: +91 361 2582717(O) >> +91 361 2584717(R) >> Fax: +91 361 2582749 (Physics) >> +91 361 2690762 (General) >> > ************************************************************************ >> >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> > > > ------------------------------------------------------------------- > Prof. Eyvaz Isaev, > Theoretical Physics Department, Moscow State Institute of Steel & Alloys, > Russia, > IFM, Linkoping University, Sweden > Condensed Matter Theory Group, Uppsala University, Sweden > Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com > > > ____________________________________________________________________________________ > Be a better friend, newshound, and > know-it-all with Yahoo! Mobile. Try it now. > http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > ************************************************************************ Dr. Subhradip Ghosh Assistant Professor Department of Physics Indian Institute of Technology Guwahati,Assam-781039 India E-mail:subhra at iitg.ernet.in Phone: +91 361 2582717(O) +91 361 2584717(R) Fax: +91 361 2582749 (Physics) +91 361 2690762 (General) ************************************************************************ From w2agz at pacbell.net Fri Feb 8 01:06:43 2008 From: w2agz at pacbell.net (Paul M. Grant) Date: Thu, 7 Feb 2008 16:06:43 -0800 Subject: [Pw_forum] Missing Libs? In-Reply-To: <3A8059EA-0311-416D-AD6A-FD5C0C236CD5@nest.sns.it> References: <000601c8694d$09969180$1cc3b480$@net> <3A8059EA-0311-416D-AD6A-FD5C0C236CD5@nest.sns.it> Message-ID: <003e01c869e6$7d1c2450$77546cf0$@net> Paolo, that was it! Thanks to you and your colleague Bussi. Details: -Box: HP 6200 I got at a fire sale (that means used and real cheap), AMD64x2 2.60 GHz, 2GB RAM -OS: Latest Cygwin/X running underneath 32-bit Vista Home Premium (ugh). -Fortran: gfortran -CVS O-Sesame I'm putting a Kubuntu 64 bit OS on the rest of the hard disk and it will interesting to see the performance differences (if any). BTW, I've been running O-sesame on my existing "Kubuntu production box" for about a month now, and, so far, it's been quite stable. But for one weird item...the check_failure.sh script in "examples" keeps bombing with an unexpected "(" error...which doesn't occur under Cygwin...different bash-es? Paul M. Grant, PhD Principal, W2AGZ Technologies Visiting Scholar, Applied Physics, Stanford University EPRI Science Fellow (Retired) IBM Research Staff Member Emeritus w2agz at pacbell.net http://www.w2agz.com ? ? -----Original Message----- From: pw_forum-bounces at pwscf.org [mailto:pw_forum-bounces at pwscf.org] On Behalf Of Paolo Giannozzi Sent: Thursday, February 07, 2008 5:55 AM To: PWSCF Forum Subject: Re: [Pw_forum] Missing Libs? On Feb 7, 2008, at 6:48 , Paul M. Grant wrote: > # FROM IOTK LIBRARY, VERSION 1.1.0development > > # UNRECOVERABLE ERROR (ierr=1) > quick fix: change Q_REAL in Q_REAL_SPACE in Modules/xml_io_base.f90 (thanks to Giovanni Bussi). This was a harmless q-e espresso bug in most cases, but it was (likely) acctivating a gfortran bug that affects iotk when a tag that is not present in the file is searched for... Paolo --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum From giannozz at nest.sns.it Fri Feb 8 09:05:41 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Fri, 08 Feb 2008 09:05:41 +0100 Subject: [Pw_forum] Missing Libs? In-Reply-To: <003e01c869e6$7d1c2450$77546cf0$@net> References: <000601c8694d$09969180$1cc3b480$@net> <3A8059EA-0311-416D-AD6A-FD5C0C236CD5@nest.sns.it> <003e01c869e6$7d1c2450$77546cf0$@net> Message-ID: <47AC0D55.9020403@nest.sns.it> Paul M. Grant wrote: > BTW, I've been running O-sesame on my existing "Kubuntu production box" for > about a month now, and, so far, it's been quite stable. But for one weird > item...the check_failure.sh script in "examples" keeps bombing with an > unexpected "(" error...which doesn't occur under Cygwin...different bash-es? different bash-es : remove the () in check_failure.sh . I have no idea how to make those scripts portable on all machines. They work for most but not all cases. Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From sagarambavale at yahoo.co.in Fri Feb 8 16:38:14 2008 From: sagarambavale at yahoo.co.in (ambavale sagar) Date: Fri, 8 Feb 2008 21:08:14 +0530 (IST) Subject: [Pw_forum] multi core Vs Single core Message-ID: <298798.6746.qm@web94608.mail.in2.yahoo.com> An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080208/c78ee662/attachment.htm From Giovanni.Cantele at na.infn.it Fri Feb 8 16:51:55 2008 From: Giovanni.Cantele at na.infn.it (Giovanni Cantele) Date: Fri, 08 Feb 2008 16:51:55 +0100 Subject: [Pw_forum] multi core Vs Single core In-Reply-To: <298798.6746.qm@web94608.mail.in2.yahoo.com> References: <298798.6746.qm@web94608.mail.in2.yahoo.com> Message-ID: <47AC7A9B.1040404@na.infn.it> ambavale sagar wrote: > > Is multi (quad) core processor more efficient than single core for our > abinitio calculations? How efficiently multi-threading is done while > working with serial version of code on multicore machine? > > Thanx. > > > > Sagar K. Ambavale > > PhD student, > > The M.S. University of Baroda, > > India > > > ------------------------------------------------------------------------ > 5, 50, 500, 5000 - Store N number of mails in your inbox. Click here. > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > dual core helps, quad core could give a negligible improvement with respect to dual core. Much can depend on your input, namely, the size of the system you are dealing with, I think. By the way, you can have a look to previous posts in the forum: http://www.democritos.it/pipermail/pw_forum/2006-August/004650.html http://www.democritos.it/pipermail/pw_forum/2006-December/005595.html http://www.democritos.it/pipermail/pw_forum/2007-November/007707.html Giovanni -- Dr. Giovanni Cantele Coherentia CNR-INFM and Dipartimento di Scienze Fisiche Universita' di Napoli "Federico II" Complesso Universitario di Monte S. Angelo - Ed. 6 Via Cintia, I-80126, Napoli, Italy Phone: +39 081 676910 Fax: +39 081 676346 E-mail: giovanni.cantele at cnr.it giovanni.cantele at na.infn.it Web: http://people.na.infn.it/~cantele Research Group: http://www.nanomat.unina.it From akohlmey at cmm.chem.upenn.edu Fri Feb 8 17:25:11 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Fri, 8 Feb 2008 11:25:11 -0500 (EST) Subject: [Pw_forum] multi core Vs Single core In-Reply-To: <298798.6746.qm@web94608.mail.in2.yahoo.com> References: <298798.6746.qm@web94608.mail.in2.yahoo.com> Message-ID: On Fri, 8 Feb 2008, ambavale sagar wrote: sagar, AS> Is multi (quad) core processor more efficient than single core for our please clarify what you mean by "efficient". more throughput for the same money, faster execution, better utilization of the cpu, less room needed for a given computational capability? AS> abinitio calculations? How efficiently multi-threading is done while AS> working with serial version of code on multicore machine? it is a common misconceptions that you need to use multi-threading to take advantage of multicore cpus. the vendors of those cpus are somewhat promoting it, because it is simpler to add multi-threading to a code than to parallelize it completely, and also because their new cpus don't have any advantage over the older ones without paralleization. fortunately, the methods used in quantum espresso are well parallelizable and the code offers multiple levels of MPI parallelism which i found to be almost always more efficient than using OpenMP (and thus multi-threading). even more so, QE does not support OpenMP (only indirectly via threaded BLAS/LAPACK libraries). depending on your hardware, problem set size and compiler/library combination, you can get a speedup of roughly 2.5x to 3.5x from a quad core cpu over a corresponding single core. when using multiple nodes via a network, however, the speed and the latency of the network (and the options of parallization, i.e. if you can parallelize over k-points or not) start to matter and can make it all very confusing. in the end, apart from a few general observations, there is no other way to get a definite answer than running tests with representative input files. cheers, axel. p.s.: this brings up the question again, that we should compile a list of such typical inputs and run them on mutually available machines so people can see what to expect.... AS> AS> Thanx. AS> AS> ? AS> AS> Sagar K. Ambavale AS> AS> PhD student, AS> AS> The M.S. University of Baroda, AS> AS> India AS> AS> AS> ________________________________________________________________________________ AS> 5, 50, 500, 5000 - Store N number of mails in your inbox. Click here. 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 marcel at physik.tu-berlin.de Fri Feb 8 18:10:58 2008 From: marcel at physik.tu-berlin.de (Marcel Mohr) Date: Fri, 8 Feb 2008 18:10:58 +0100 (CET) Subject: [Pw_forum] multi core Vs Single core In-Reply-To: References: <298798.6746.qm@web94608.mail.in2.yahoo.com> Message-ID: <20080208180343.T3645@rosa.physik-pool.tu-berlin.de> There are some posts about this topic some time ago. Planewave codes use a lot of memory bandwith, so an Intel quadcore is not much faster than a dual core on the same mainboard, because the memory bus becomes the bottleneck of the system, and all core share the same bus. Cheers Marcel On Fri, 8 Feb 2008, Axel Kohlmeyer wrote: > On Fri, 8 Feb 2008, ambavale sagar wrote: > > > sagar, > > AS> Is multi (quad) core processor more efficient than single core for our > > please clarify what you mean by "efficient". > more throughput for the same money, faster execution, > better utilization of the cpu, less room needed for > a given computational capability? > > AS> abinitio calculations? How efficiently multi-threading is done while > AS> working with serial version of code on multicore machine? > > it is a common misconceptions that you need to use multi-threading > to take advantage of multicore cpus. the vendors of those cpus > are somewhat promoting it, because it is simpler to add multi-threading > to a code than to parallelize it completely, and also because > their new cpus don't have any advantage over the older ones without > paralleization. fortunately, the methods used in quantum espresso > are well parallelizable and the code offers multiple levels of MPI > parallelism which i found to be almost always more efficient than > using OpenMP (and thus multi-threading). even more so, QE does not > support OpenMP (only indirectly via threaded BLAS/LAPACK libraries). > > depending on your hardware, problem set size and compiler/library > combination, you can get a speedup of roughly 2.5x to 3.5x from > a quad core cpu over a corresponding single core. when using multiple > nodes via a network, however, the speed and the latency of the > network (and the options of parallization, i.e. if you can parallelize > over k-points or not) start to matter and can make it all very > confusing. > > in the end, apart from a few general observations, there is no other > way to get a definite answer than running tests with representative > input files. > > cheers, > axel. > > p.s.: this brings up the question again, that we should compile a > list of such typical inputs and run them on mutually available > machines so people can see what to expect.... > > AS> > AS> Thanx. > AS> > AS> ? > AS> > AS> Sagar K. Ambavale > AS> > AS> PhD student, > AS> > AS> The M.S. University of Baroda, > AS> > AS> India > AS> > AS> > AS> ________________________________________________________________________________ > AS> 5, 50, 500, 5000 - Store N number of mails in your inbox. Click here. > 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 akohlmey at cmm.chem.upenn.edu Fri Feb 8 18:34:20 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Fri, 8 Feb 2008 12:34:20 -0500 Subject: [Pw_forum] multi core Vs Single core In-Reply-To: <20080208180343.T3645@rosa.physik-pool.tu-berlin.de> References: <298798.6746.qm@web94608.mail.in2.yahoo.com> <20080208180343.T3645@rosa.physik-pool.tu-berlin.de> Message-ID: <7b6913e90802080934l4a0371dpa609943279f831d0@mail.gmail.com> On Feb 8, 2008 12:10 PM, Marcel Mohr wrote: > There are some posts about this topic some time ago. > > Planewave codes use a lot of memory bandwith, so an Intel > quadcore is not much faster than a dual core on the same mainboard, > because the memory bus becomes the bottleneck of the system, and all core > share the same bus. for some additional numbers on multi-core. have a look at: http://groups.google.com/group/cp2k/msg/aa1ce54b56598aab?dmode=source FYI, CP2K/QuickStep is even more demanding in terms of memory bandwidth than QE. however, since some vendors are offering quad-core nodes for about the same price as dual-core nodes, it is actually a better deal to get quad-core and then use only half of the cores (and have twice the cache) if you get processor affinity set up right. a. > > Cheers > Marcel > > > On Fri, 8 Feb 2008, Axel Kohlmeyer wrote: > > > On Fri, 8 Feb 2008, ambavale sagar wrote: > > > > > > sagar, > > > > AS> Is multi (quad) core processor more efficient than single core for our > > > > please clarify what you mean by "efficient". > > more throughput for the same money, faster execution, > > better utilization of the cpu, less room needed for > > a given computational capability? > > > > AS> abinitio calculations? How efficiently multi-threading is done while > > AS> working with serial version of code on multicore machine? > > > > it is a common misconceptions that you need to use multi-threading > > to take advantage of multicore cpus. the vendors of those cpus > > are somewhat promoting it, because it is simpler to add multi-threading > > to a code than to parallelize it completely, and also because > > their new cpus don't have any advantage over the older ones without > > paralleization. fortunately, the methods used in quantum espresso > > are well parallelizable and the code offers multiple levels of MPI > > parallelism which i found to be almost always more efficient than > > using OpenMP (and thus multi-threading). even more so, QE does not > > support OpenMP (only indirectly via threaded BLAS/LAPACK libraries). > > > > depending on your hardware, problem set size and compiler/library > > combination, you can get a speedup of roughly 2.5x to 3.5x from > > a quad core cpu over a corresponding single core. when using multiple > > nodes via a network, however, the speed and the latency of the > > network (and the options of parallization, i.e. if you can parallelize > > over k-points or not) start to matter and can make it all very > > confusing. > > > > in the end, apart from a few general observations, there is no other > > way to get a definite answer than running tests with representative > > input files. > > > > cheers, > > axel. > > > > p.s.: this brings up the question again, that we should compile a > > list of such typical inputs and run them on mutually available > > machines so people can see what to expect.... > > > > AS> > > AS> Thanx. > > AS> > > AS> > > AS> > > AS> Sagar K. Ambavale > > AS> > > AS> PhD student, > > AS> > > AS> The M.S. University of Baroda, > > AS> > > AS> India > > AS> > > AS> > > AS> ________________________________________________________________________________ > > AS> 5, 50, 500, 5000 - Store N number of mails in your inbox. Click here. > > 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. > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From naromero at gmail.com Sat Feb 9 03:42:46 2008 From: naromero at gmail.com (Nichols A. Romero) Date: Fri, 8 Feb 2008 21:42:46 -0500 Subject: [Pw_forum] plotting charge density in espresso-4.0 Message-ID: <6ac064b60802081842s2cfce466ybc8cdaaad8b95a7e@mail.gmail.com> Hi, I am using the espresso-4.0cvs version. It looks like the charge density is being outputted into an XML format (though the file format is binary). Does one need to use pw_export.x to convert to a format that is readable by pp.x? My ultimate goal is to plot the 3D charge density from the XSF format or similar. -- Nichols A. Romero, Ph.D. DoD User Productivity Enhancement and Technology Transfer (PET) Group High Performance Technologies, Inc. Reston, VA 443-567-8328 (C) 410-278-2692 (O) -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080208/258df3dd/attachment-0001.htm From igor at ffh.bg.ac.yu Sat Feb 9 13:23:04 2008 From: igor at ffh.bg.ac.yu (igor at ffh.bg.ac.yu) Date: Sat, 9 Feb 2008 13:23:04 +0100 Subject: [Pw_forum] problems with installation Message-ID: <1202559784.47ad9b28dada4@mail.ffh.bg.ac.yu> Hi, all, I have some problems with installation of QE. First of all :/configure is not able to recognize parallel environment unless I define ./configure F77=gfortran F90=gfortran MPIF90=mpif90 but when I run make all it returns startup.f90:(.text+0x554): undefined reference to `iargc_' startup.f90:(.text+0x6b2): undefined reference to `iargc_' collect2: ld returned 1 exit status make[1]: *** [pw.x] Error 1 make[1]: Leaving directory `/quantumespresso/espresso-3.2.2/PW' make: *** [pw] Error 2 [root at xeon espresso-3.2.2]# This is problem with versions 3.1.1, 3.2.2 and 3.2.3. But when I define ./configure F77=gfortran F90=gfortran compilation is possible for 3.1.1 in serial version. (My operative system is Centos 5.1, I have 2 Intel Xeon processors). If anybody faced similar problems and managed to solve them I would really appreciate help. Thanks Igor From akohlmey at cmm.chem.upenn.edu Sat Feb 9 18:11:22 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sat, 9 Feb 2008 12:11:22 -0500 (EST) Subject: [Pw_forum] problems with installation In-Reply-To: <1202559784.47ad9b28dada4@mail.ffh.bg.ac.yu> References: <1202559784.47ad9b28dada4@mail.ffh.bg.ac.yu> Message-ID: On Sat, 9 Feb 2008, igor at ffh.bg.ac.yu wrote: I?> Hi, all, hi igor, I?> I have some problems with installation of QE. First of all I?> :/configure is not able to recognize parallel environment unless I I?> define I?> ./configure F77=gfortran F90=gfortran MPIF90=mpif90 that means, that your mpif90 is most likely using a compiler different from gfortran as underlying fortran 90 compiler. I?> but when I run I?> I?> make all I?> I?> it returns I?> I?> startup.f90:(.text+0x554): undefined reference to `iargc_' I?> startup.f90:(.text+0x6b2): undefined reference to `iargc_' I?> collect2: ld returned 1 exit status and this confirms this suspicion. different fortran compilers have different ways to detect command line arguments (and other stuff). all fortran parts of QE have to be compiler with compatible compilers. I?> make[1]: *** [pw.x] Error 1 I?> make[1]: Leaving directory `/quantumespresso/espresso-3.2.2/PW' I?> make: *** [pw] Error 2 I?> [root at xeon espresso-3.2.2]# I?> I?> This is problem with versions 3.1.1, 3.2.2 and 3.2.3. But when I define this is independent of the QE version. after QE has been around for so long and so many people have managed to compile and installed it regularly, it is safe to assume, that problems are usually within the machine that you want to compile it on. over the years there have been a lot of attempts to make the configure process smarter, but that works only to a certain degree. there are simply too many quirky installations around, particularly linux machines. I?> ./configure F77=gfortran F90=gfortran I?> compilation is possible for 3.1.1 in serial version. yes. because the compilers are compatible. as it is not version dependent, please always use the latest version. I?> (My operative system is Centos 5.1, I have 2 Intel Xeon processors). I?> If anybody faced similar problems and managed to solve them I would really I?> appreciate help. here's a number of things, that you can do yourself: you can have a closer look at output produced by configure. that should give you some hint on whether the parallel environment is detected or not. and you can look into the resulting config.log for a more detailed log (search for mpi_init). you can check your mpi installation (try 'mpif90 -v' or 'mpif90 -showme') to see what compiler you have "within" mpif90. you can try compiling and running a simple standalone mpi program. finally, please respect the policy of this forum and append your full name and affiliation. thanks. cheers, axel. I?> Thanks I?> I?> Igor I?> I?> I?> I?> _______________________________________________ I?> Pw_forum mailing list I?> Pw_forum at pwscf.org I?> http://www.democritos.it/mailman/listinfo/pw_forum I?> -- ======================================================================= 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 w2agz at pacbell.net Sat Feb 9 21:23:29 2008 From: w2agz at pacbell.net (Paul M. Grant) Date: Sat, 9 Feb 2008 12:23:29 -0800 Subject: [Pw_forum] multi core Vs Single core In-Reply-To: <7b6913e90802080934l4a0371dpa609943279f831d0@mail.gmail.com> References: <298798.6746.qm@web94608.mail.in2.yahoo.com> <20080208180343.T3645@rosa.physik-pool.tu-berlin.de> <7b6913e90802080934l4a0371dpa609943279f831d0@mail.gmail.com> Message-ID: <009801c86b59$a2c74990$e855dcb0$@net> For what it's worth, here're some anecdotal experiences I've had over the last several days on an AMD64x2 HP Pavilion box I got real cheap. I've haven't installed Kubuntu yet, but running some of my benchmarks under Cygwin/X (Win OS is Vista 32 bit), it seems "cpu intensive" tasks, like gfortran compile and pwscf nscf run only in one core...(~50% cpu as shown by Task Manager) with certain Vista "services" occasionally launching, presumably in the other core, ramping the total to 100%. BTW, the max commitment of RAM is only 50% of the that available, and there is very little disk activity. It will be interesting to see how Kubuntu handles this. Paul M. Grant, PhD Principal, W2AGZ Technologies Visiting Scholar, Applied Physics, Stanford University EPRI Science Fellow (Retired) IBM Research Staff Member Emeritus w2agz at pacbell.net http://www.w2agz.com ? ? -----Original Message----- From: pw_forum-bounces at pwscf.org [mailto:pw_forum-bounces at pwscf.org] On Behalf Of Axel Kohlmeyer Sent: Friday, February 08, 2008 9:34 AM To: PWSCF Forum Subject: Re: [Pw_forum] multi core Vs Single core On Feb 8, 2008 12:10 PM, Marcel Mohr wrote: > There are some posts about this topic some time ago. > > Planewave codes use a lot of memory bandwith, so an Intel > quadcore is not much faster than a dual core on the same mainboard, > because the memory bus becomes the bottleneck of the system, and all core > share the same bus. for some additional numbers on multi-core. have a look at: http://groups.google.com/group/cp2k/msg/aa1ce54b56598aab?dmode=source FYI, CP2K/QuickStep is even more demanding in terms of memory bandwidth than QE. however, since some vendors are offering quad-core nodes for about the same price as dual-core nodes, it is actually a better deal to get quad-core and then use only half of the cores (and have twice the cache) if you get processor affinity set up right. a. > > Cheers > Marcel > > > On Fri, 8 Feb 2008, Axel Kohlmeyer wrote: > > > On Fri, 8 Feb 2008, ambavale sagar wrote: > > > > > > sagar, > > > > AS> Is multi (quad) core processor more efficient than single core for our > > > > please clarify what you mean by "efficient". > > more throughput for the same money, faster execution, > > better utilization of the cpu, less room needed for > > a given computational capability? > > > > AS> abinitio calculations? How efficiently multi-threading is done while > > AS> working with serial version of code on multicore machine? > > > > it is a common misconceptions that you need to use multi-threading > > to take advantage of multicore cpus. the vendors of those cpus > > are somewhat promoting it, because it is simpler to add multi-threading > > to a code than to parallelize it completely, and also because > > their new cpus don't have any advantage over the older ones without > > paralleization. fortunately, the methods used in quantum espresso > > are well parallelizable and the code offers multiple levels of MPI > > parallelism which i found to be almost always more efficient than > > using OpenMP (and thus multi-threading). even more so, QE does not > > support OpenMP (only indirectly via threaded BLAS/LAPACK libraries). > > > > depending on your hardware, problem set size and compiler/library > > combination, you can get a speedup of roughly 2.5x to 3.5x from > > a quad core cpu over a corresponding single core. when using multiple > > nodes via a network, however, the speed and the latency of the > > network (and the options of parallization, i.e. if you can parallelize > > over k-points or not) start to matter and can make it all very > > confusing. > > > > in the end, apart from a few general observations, there is no other > > way to get a definite answer than running tests with representative > > input files. > > > > cheers, > > axel. > > > > p.s.: this brings up the question again, that we should compile a > > list of such typical inputs and run them on mutually available > > machines so people can see what to expect.... > > > > AS> > > AS> Thanx. > > AS> > > AS> > > AS> > > AS> Sagar K. Ambavale > > AS> > > AS> PhD student, > > AS> > > AS> The M.S. University of Baroda, > > AS> > > AS> India > > AS> > > AS> > > AS> ____________________________________________________________________________ ____ > > AS> 5, 50, 500, 5000 - Store N number of mails in your inbox. Click here. > > 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. > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum From giannozz at nest.sns.it Sat Feb 9 21:43:05 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Sat, 9 Feb 2008 21:43:05 +0100 Subject: [Pw_forum] problems with installation In-Reply-To: References: <1202559784.47ad9b28dada4@mail.ffh.bg.ac.yu> Message-ID: <73F55ECB-0CCC-415F-9B5F-B7EF093FCA23@nest.sns.it> On Feb 9, 2008, at 18:11 , Axel Kohlmeyer wrote: > I?> startup.f90:(.text+0x6b2): undefined reference to `iargc_' > I?> collect2: ld returned 1 exit status > > and this confirms this suspicion. different fortran compilers > have different ways to detect command line arguments (and other > stuff). all fortran parts of QE have to be compiler with compatible > compilers. correct, but the specific problem might be simpler than this: gfortran doesn't like "iargc" to be defined as "external". This was fixed in the CVS version some time ago: http://www.democritos.it:8888/O-sesame/chngview?cn=5150. Also, a real support for gfortran in "configure" is quite recent. Paolo --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From giannozz at nest.sns.it Sat Feb 9 21:44:44 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Sat, 9 Feb 2008 21:44:44 +0100 Subject: [Pw_forum] plotting charge density in espresso-4.0 In-Reply-To: <6ac064b60802081842s2cfce466ybc8cdaaad8b95a7e@mail.gmail.com> References: <6ac064b60802081842s2cfce466ybc8cdaaad8b95a7e@mail.gmail.com> Message-ID: On Feb 9, 2008, at 3:42 , Nichols A. Romero wrote: > I am using the espresso-4.0cvs version. It looks like the charge > density is > being outputted into an XML format (though the file format is > binary). Does one > need to use pw_export.x to convert to a format that is readable by > pp.x? no, pp.x will read the XML format: business as usual Paolo --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From w2agz at pacbell.net Sat Feb 9 22:23:37 2008 From: w2agz at pacbell.net (Paul M. Grant) Date: Sat, 9 Feb 2008 13:23:37 -0800 Subject: [Pw_forum] multi core Vs Single core In-Reply-To: <009801c86b59$a2c74990$e855dcb0$@net> References: <298798.6746.qm@web94608.mail.in2.yahoo.com> <20080208180343.T3645@rosa.physik-pool.tu-berlin.de> <7b6913e90802080934l4a0371dpa609943279f831d0@mail.gmail.com> <009801c86b59$a2c74990$e855dcb0$@net> Message-ID: <009901c86b62$0984e310$1c8ea930$@net> Added comment: Two cpu intensive Windows threads run parallel in the two cores...100% Paul M. Grant, PhD Principal, W2AGZ Technologies Visiting Scholar, Applied Physics, Stanford University EPRI Science Fellow (Retired) IBM Research Staff Member Emeritus w2agz at pacbell.net http://www.w2agz.com ? ? -----Original Message----- From: pw_forum-bounces at pwscf.org [mailto:pw_forum-bounces at pwscf.org] On Behalf Of Paul M. Grant Sent: Saturday, February 09, 2008 12:23 PM To: 'PWSCF Forum' Subject: Re: [Pw_forum] multi core Vs Single core For what it's worth, here're some anecdotal experiences I've had over the last several days on an AMD64x2 HP Pavilion box I got real cheap. I've haven't installed Kubuntu yet, but running some of my benchmarks under Cygwin/X (Win OS is Vista 32 bit), it seems "cpu intensive" tasks, like gfortran compile and pwscf nscf run only in one core...(~50% cpu as shown by Task Manager) with certain Vista "services" occasionally launching, presumably in the other core, ramping the total to 100%. BTW, the max commitment of RAM is only 50% of the that available, and there is very little disk activity. It will be interesting to see how Kubuntu handles this. Paul M. Grant, PhD Principal, W2AGZ Technologies Visiting Scholar, Applied Physics, Stanford University EPRI Science Fellow (Retired) IBM Research Staff Member Emeritus w2agz at pacbell.net http://www.w2agz.com ? ? -----Original Message----- From: pw_forum-bounces at pwscf.org [mailto:pw_forum-bounces at pwscf.org] On Behalf Of Axel Kohlmeyer Sent: Friday, February 08, 2008 9:34 AM To: PWSCF Forum Subject: Re: [Pw_forum] multi core Vs Single core On Feb 8, 2008 12:10 PM, Marcel Mohr wrote: > There are some posts about this topic some time ago. > > Planewave codes use a lot of memory bandwith, so an Intel > quadcore is not much faster than a dual core on the same mainboard, > because the memory bus becomes the bottleneck of the system, and all core > share the same bus. for some additional numbers on multi-core. have a look at: http://groups.google.com/group/cp2k/msg/aa1ce54b56598aab?dmode=source FYI, CP2K/QuickStep is even more demanding in terms of memory bandwidth than QE. however, since some vendors are offering quad-core nodes for about the same price as dual-core nodes, it is actually a better deal to get quad-core and then use only half of the cores (and have twice the cache) if you get processor affinity set up right. a. > > Cheers > Marcel > > > On Fri, 8 Feb 2008, Axel Kohlmeyer wrote: > > > On Fri, 8 Feb 2008, ambavale sagar wrote: > > > > > > sagar, > > > > AS> Is multi (quad) core processor more efficient than single core for our > > > > please clarify what you mean by "efficient". > > more throughput for the same money, faster execution, > > better utilization of the cpu, less room needed for > > a given computational capability? > > > > AS> abinitio calculations? How efficiently multi-threading is done while > > AS> working with serial version of code on multicore machine? > > > > it is a common misconceptions that you need to use multi-threading > > to take advantage of multicore cpus. the vendors of those cpus > > are somewhat promoting it, because it is simpler to add multi-threading > > to a code than to parallelize it completely, and also because > > their new cpus don't have any advantage over the older ones without > > paralleization. fortunately, the methods used in quantum espresso > > are well parallelizable and the code offers multiple levels of MPI > > parallelism which i found to be almost always more efficient than > > using OpenMP (and thus multi-threading). even more so, QE does not > > support OpenMP (only indirectly via threaded BLAS/LAPACK libraries). > > > > depending on your hardware, problem set size and compiler/library > > combination, you can get a speedup of roughly 2.5x to 3.5x from > > a quad core cpu over a corresponding single core. when using multiple > > nodes via a network, however, the speed and the latency of the > > network (and the options of parallization, i.e. if you can parallelize > > over k-points or not) start to matter and can make it all very > > confusing. > > > > in the end, apart from a few general observations, there is no other > > way to get a definite answer than running tests with representative > > input files. > > > > cheers, > > axel. > > > > p.s.: this brings up the question again, that we should compile a > > list of such typical inputs and run them on mutually available > > machines so people can see what to expect.... > > > > AS> > > AS> Thanx. > > AS> > > AS> > > AS> > > AS> Sagar K. Ambavale > > AS> > > AS> PhD student, > > AS> > > AS> The M.S. University of Baroda, > > AS> > > AS> India > > AS> > > AS> > > AS> ____________________________________________________________________________ ____ > > AS> 5, 50, 500, 5000 - Store N number of mails in your inbox. Click here. > > 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. > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum From w2agz at pacbell.net Sat Feb 9 23:18:04 2008 From: w2agz at pacbell.net (Paul M. Grant) Date: Sat, 9 Feb 2008 14:18:04 -0800 Subject: [Pw_forum] gfortran compiler errors on dforceb.f90 Message-ID: <009a01c86b69$a4ba73c0$ee2f5b40$@net> On a fresh make of espresso-3.2.3 with gfortran as the compiler, dforceb.f90 gives the following reproducible abend. Whatever is causing this seems to have been fixed in O-sesame and espressocvs. The dforceb source file sizes are different and of different dates, but a quick look didn't disclose (to me, anyway) what's causing the problem.looks like there should have been some integer declarations.somewhere. gfortran -O3 -x f95-cpp-input -D__FFTW -D__USE_INTERNAL_FFTW -I../include -I./ -I../Modules -I../iotk/src -I../PW -I../PH -I../CPV -c dforceb.f90 dforceb.f90:48.22: complex(8) c0(ngw, n), betae(ngw,nhsa), df(ngw),& 1 Error: Symbol 'n' at (1) has no IMPLICIT type dforceb.f90:8.37: subroutine dforceb(c0, i, betae, ipol, bec0, ctabin, gqq, gqqm, qmat, dq2, df) 1 Error: Symbol 'ipol' at (1) has no IMPLICIT type dforceb.f90:52.18: & dq2(nat,nhm,nhm,nspin), gmes 1 Error: Symbol 'nat' at (1) has no IMPLICIT type dforceb.f90:133.39: inl=ish(is)+(iv-1)*na(is)+ia 1 Error: Function 'na' at (1) has no IMPLICIT type dforceb.f90:134.39: jnl=ish(is)+(jv-1)*na(is)+ia 1 Error: Function 'na' at (1) has no IMPLICIT type dforceb.f90:132.23: do ia=1,na(is) 1 Error: Function 'na' at (1) has no IMPLICIT type make[1]: *** [dforceb.o] Error 1 make[1]: Leaving directory `/home/pmpgrant/espresso-3.2.3/CPV' make: *** [cp] Error 2 Paul M. Grant, PhD Principal, W2AGZ Technologies Visiting Scholar, Applied Physics, Stanford University EPRI Science Fellow (Retired) IBM Research Staff Member Emeritus w2agz at pacbell.net http://www.w2agz.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080209/d705f591/attachment-0001.htm From akohlmey at cmm.chem.upenn.edu Sat Feb 9 23:18:15 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sat, 9 Feb 2008 17:18:15 -0500 (EST) Subject: [Pw_forum] multi core Vs Single core In-Reply-To: <009801c86b59$a2c74990$e855dcb0$@net> References: <298798.6746.qm@web94608.mail.in2.yahoo.com> <20080208180343.T3645@rosa.physik-pool.tu-berlin.de> <7b6913e90802080934l4a0371dpa609943279f831d0@mail.gmail.com> <009801c86b59$a2c74990$e855dcb0$@net> Message-ID: On Sat, 9 Feb 2008, Paul M. Grant wrote: PG> haven't installed Kubuntu yet, but running some of my benchmarks under PG> Cygwin/X (Win OS is Vista 32 bit), it seems "cpu intensive" tasks, like PG> gfortran compile and pwscf nscf run only in one core...(~50% cpu as shown by paul, if you want to speed up compilations of QE (for all except iotk), you can try (provided you have a GNU make installed): make MFLAGS=-j2 dunno if windows is up to it, but it rocks under linux. ;-) 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 w2agz at pacbell.net Sun Feb 10 01:39:01 2008 From: w2agz at pacbell.net (Paul M. Grant) Date: Sat, 9 Feb 2008 16:39:01 -0800 Subject: [Pw_forum] CVS Changes in output file format Message-ID: <00aa01c86b7d$5512f950$ff38ebf0$@net> The scf and nscf output files produced by pw.x in release espresso-3.2.3 contain the following lines, nbndx = 28 nbnd = 7 natomwfc = 18 npwx = 5958 nelec = 6.00 nkb = 8 ngl = 2887 which are now (apparently) absent from pw.x results in the O-sesame and espressocvs distros. I imagine this has broken a few scripts (like mine!) that grep-ed for the default values of nbnd and nelec. No biggy, because these values can be input-ed directly, but I would think no changes are made in a core program like pw.x without a reason.what was it for this one? Paul M. Grant, PhD Principal, W2AGZ Technologies Visiting Scholar, Applied Physics, Stanford University EPRI Science Fellow (Retired) IBM Research Staff Member Emeritus w2agz at pacbell.net http://www.w2agz.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080209/c8c89003/attachment.htm From chenhanghuipwscf at gmail.com Sun Feb 10 16:51:18 2008 From: chenhanghuipwscf at gmail.com (alan chen) Date: Sun, 10 Feb 2008 10:51:18 -0500 Subject: [Pw_forum] Davidson diagonalization Message-ID: <22ae3ca40802100751j3c3a128cnbb7bcfbacb8d6b10@mail.gmail.com> Dear PWSCF users, I have encountered a problem recently. When I ran a self-consistent calculation of a large system, the Davidson diagonaliztion works well. However when I ran a non-self-consistent calculation of the same system with the same k-point density and 'temperature', the code collapses because 'from cdiaghg : error # 28286 info=/= 0'. It seems that the Davidson diagonalization does not work well here. What can I do to avoid that error? Is there anything wrong with the charge density? Thank you very much. Hanghui Department of Physics Yale University -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080210/87f6efab/attachment.htm From giannozz at nest.sns.it Sun Feb 10 17:38:40 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Sun, 10 Feb 2008 17:38:40 +0100 Subject: [Pw_forum] CVS Changes in output file format In-Reply-To: <00aa01c86b7d$5512f950$ff38ebf0$@net> References: <00aa01c86b7d$5512f950$ff38ebf0$@net> Message-ID: On Feb 10, 2008, at 1:39 , Paul M. Grant wrote: > The scf and nscf output files produced by pw.x in release > espresso-3.2.3 contain the following lines, > > nbndx = 28 nbnd = 7 natomwfc = 18 npwx = > 5958 > > nelec = 6.00 nkb = 8 ngl = 2887 > > which are now (apparently) absent from pw.x results in the O-sesame > and espressocvs distros. I imagine this has broken a few scripts > (like mine!) that grep-ed for the default values of nbnd and > nelec. No biggy, because these values can be input-ed directly, > but I would think no changes are made in a core program like pw.x > without a reason?what was it for this one? > the reason was to make the output more informative (or at least less obscure) and more consistent. I and you know what all those variables mean, but for most people those number mean close to nothing. "nbnd" and "nelec" are listed respectively as "number of electrons" and "number of Kohn-Sham states". About "dforceb" and gfortran, I think this is the relevant change: http://www.democritos.it:8888/O-sesame/chngview?cn=5551 Paolo --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From giannozz at nest.sns.it Sun Feb 10 17:44:25 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Sun, 10 Feb 2008 17:44:25 +0100 Subject: [Pw_forum] Davidson diagonalization In-Reply-To: <22ae3ca40802100751j3c3a128cnbb7bcfbacb8d6b10@mail.gmail.com> References: <22ae3ca40802100751j3c3a128cnbb7bcfbacb8d6b10@mail.gmail.com> Message-ID: On Feb 10, 2008, at 16:51 , alan chen wrote: > When I ran a self-consistent calculation of a large system, the > Davidson diagonalization works well. However when I ran a non-self- > consistent calculation of the same system with the same k-point > density and 'temperature', the code collapses because 'from > cdiaghg : error # 28286 info=/= 0'. see http://www.quantum-espresso.org/wiki/index.php/Troubleshooting_% 28PWscf%29#pw.x_stops_with_error_in_cdiaghg_or_rdiaghg and the many messages on this mailing list regarding similar errors --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From ghaderyan at gmail.com Mon Feb 11 10:19:42 2008 From: ghaderyan at gmail.com (saman ghaderyan) Date: Mon, 11 Feb 2008 12:49:42 +0330 Subject: [Pw_forum] pseudpotential Message-ID: hi I search NC-PP for Uranium for calculate phonon structure, where can i download it if available, thanks for your helping MSc ghaderyan ,iran From akohlmey at cmm.chem.upenn.edu Mon Feb 11 16:42:24 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Mon, 11 Feb 2008 10:42:24 -0500 (EST) Subject: [Pw_forum] pseudpotential In-Reply-To: References: Message-ID: On Mon, 11 Feb 2008, saman ghaderyan wrote: SG> hi SG> I search NC-PP for Uranium for calculate phonon structure, SG> where can i download it if available, there are some publications with troullier-martins pseudopotential generation parameters for uranium. you can use those to generate a potential yourself with the bundled atomic code. be advised, that due to the localized nature of the f-states you are running a high risk of getting completely nonsensical results with DFT, not to mention relativistic effects etc. axel. SG> thanks for your helping SG> MSc ghaderyan ,iran SG> _______________________________________________ SG> Pw_forum mailing list SG> Pw_forum at pwscf.org SG> http://www.democritos.it/mailman/listinfo/pw_forum SG> -- ======================================================================= 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 ghaderyan at gmail.com Mon Feb 11 19:41:08 2008 From: ghaderyan at gmail.com (saman ghaderyan) Date: Mon, 11 Feb 2008 22:11:08 +0330 Subject: [Pw_forum] pseudopotential Message-ID: thanks dear Axel for quickly answer (like always) where can I found that pseudopotentials? and I think for study phonon structure NC-pp are better because we use B-O approximation and studied core region with NC-pp are good results , and for exchange correlation energy for heavy atoms, LDA are better because the effective of core on valance region is weak and the orbitals are close recovery in valance region, is these true? and when l localized f in valance region then I have to use US-pp and I think the result for phonon structure are not true but the result for electronic structure (lattice parameter) are true ! -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080211/032ffd00/attachment.htm From akohlmey at cmm.chem.upenn.edu Mon Feb 11 20:23:51 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Mon, 11 Feb 2008 14:23:51 -0500 (EST) Subject: [Pw_forum] pseudopotential In-Reply-To: References: Message-ID: On Mon, 11 Feb 2008, saman ghaderyan wrote: SG> thanks dear Axel for quickly answer (like always) SG> where can I found that pseudopotentials? please check the literature. a quick search gives, e.g., Crocombette et al., Phys.Rev.B 64 (2002), 104107 SG> and I think for study phonon structure NC-pp are better because we use B-O SG> approximation and studied core region with NC-pp are good results, SG> and for exchange correlation energy for heavy atoms, LDA are better because SG> the effective of core on valance region is weak and the orbitals are close SG> recovery in valance region, SG> is these true? i cannot comment on this. SG> and when l localized f in valance region then I have to use US-pp and I SG> think the result for phonon structure are not true but the result for SG> electronic structure (lattice parameter) are true ! with d-, and even more so f-electrons DFT is always hit-or-miss. getting the right lattice parameter can just be an accident. i had this happen to me on other elements. 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 w2agz at pacbell.net Tue Feb 12 04:27:47 2008 From: w2agz at pacbell.net (Paul M. Grant) Date: Mon, 11 Feb 2008 19:27:47 -0800 Subject: [Pw_forum] CVS Changes in output file format In-Reply-To: References: <00aa01c86b7d$5512f950$ff38ebf0$@net> Message-ID: <010601c86d27$3da69aa0$b8f3cfe0$@net> Paolo, just don't change the description/location of the Fermi energy! I'm rather curious about your remark that dforceb.f may still contain errors...isn't this a key routine in Berry phase calculations? Paul M. Grant, PhD Principal, W2AGZ Technologies Visiting Scholar, Applied Physics, Stanford University EPRI Science Fellow (Retired) IBM Research Staff Member Emeritus w2agz at pacbell.net http://www.w2agz.com ? ? -----Original Message----- From: pw_forum-bounces at pwscf.org [mailto:pw_forum-bounces at pwscf.org] On Behalf Of Paolo Giannozzi Sent: Sunday, February 10, 2008 8:39 AM To: PWSCF Forum Subject: Re: [Pw_forum] CVS Changes in output file format On Feb 10, 2008, at 1:39 , Paul M. Grant wrote: > The scf and nscf output files produced by pw.x in release > espresso-3.2.3 contain the following lines, > > nbndx = 28 nbnd = 7 natomwfc = 18 npwx = > 5958 > > nelec = 6.00 nkb = 8 ngl = 2887 > > which are now (apparently) absent from pw.x results in the O-sesame > and espressocvs distros. I imagine this has broken a few scripts > (like mine!) that grep-ed for the default values of nbnd and > nelec. No biggy, because these values can be input-ed directly, > but I would think no changes are made in a core program like pw.x > without a reason what was it for this one? > the reason was to make the output more informative (or at least less obscure) and more consistent. I and you know what all those variables mean, but for most people those number mean close to nothing. "nbnd" and "nelec" are listed respectively as "number of electrons" and "number of Kohn-Sham states". About "dforceb" and gfortran, I think this is the relevant change: http://www.democritos.it:8888/O-sesame/chngview?cn=5551 Paolo --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum From shruba at gmail.com Tue Feb 12 05:53:50 2008 From: shruba at gmail.com (shruba at gmail.com) Date: Mon, 11 Feb 2008 23:53:50 -0500 Subject: [Pw_forum] Ultrasoft pseudopotential for lanthanide Message-ID: <822f4ec80802112053g3284a1c5m6a7ad47e842ea01f@mail.gmail.com> Dear pwscf forum members, I was looking for vanderbilt ultrasoft pseudopotential lanthanide elements, I have checked pwscf, dacapo and vanderbilt ultrasoft library ' http://www.physics.rutgers.edu/~dhv/uspp/" but I could not found any lanthanides except La, If you know any other source/ websites where I can find lanthanides ultrasoft pseudopotentials please let me know. Thanks in advance. shruba gangopadhyay -- shruba gangopadhyay graduate student department of chemistry university of central florida orlando, FL-32826 'friendship doubles joys and reduces sorrows by half' (Francis Bacon). -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080211/bac178de/attachment.htm From r95222066 at ntu.edu.tw Tue Feb 12 15:25:31 2008 From: r95222066 at ntu.edu.tw (r95222066 at ntu.edu.tw) Date: Tue, 12 Feb 2008 22:25:31 +0800 Subject: [Pw_forum] hello, I have a error message about Fermi surface by xcrysden ? Message-ID: <20080212222531.8ce43jcjs4w004cg@wmail1.cc.ntu.edu.tw> Dear all, I have a error message about Fermi surface by xcrysden Here is my error message ------------------------------------------------------------------------------- Number of atoms :0 Number of frames :0 Error: Error reading Band_Grid_3D_section,while reading Error: Error reading the BXSF file ------------------------------------------------------------------------------- Thanks a lot Bing-Hong Chen,National Taiwan University,Taiwan From akohlmey at cmm.chem.upenn.edu Tue Feb 12 15:41:20 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Tue, 12 Feb 2008 09:41:20 -0500 (EST) Subject: [Pw_forum] hello, I have a error message about Fermi surface by xcrysden ? In-Reply-To: <20080212222531.8ce43jcjs4w004cg@wmail1.cc.ntu.edu.tw> References: <20080212222531.8ce43jcjs4w004cg@wmail1.cc.ntu.edu.tw> Message-ID: On Tue, 12 Feb 2008, r95222066 at ntu.edu.tw wrote: BHC> Dear all, dear bing-hong chen, BHC> I have a error message about Fermi surface by xcrysden BHC> BHC> Here is my error message BHC> ------------------------------------------------------------------------------- BHC> Number of atoms :0 BHC> Number of frames :0 BHC> Error: Error reading Band_Grid_3D_section,while reading BHC> Error: Error reading the BXSF file BHC> ------------------------------------------------------------------------------- ... and i have an error message reading your mail! what kind of response did you expect? nobody here can read people's (or computer's) minds, so you have to tell use what you did and provide and example file to reproduce the error. other than that the only response has to be: if you get an error, you must have made something wrong! regards, axel. BHC> BHC> Thanks a lot BHC> BHC> Bing-Hong Chen,National Taiwan University,Taiwan BHC> BHC> _______________________________________________ BHC> Pw_forum mailing list BHC> Pw_forum at pwscf.org BHC> http://www.democritos.it/mailman/listinfo/pw_forum BHC> -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From eyvaz_isaev at yahoo.com Tue Feb 12 16:12:02 2008 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Tue, 12 Feb 2008 07:12:02 -0800 (PST) Subject: [Pw_forum] hello, I have a error message about Fermi surface by xcrysden ? In-Reply-To: <20080212222531.8ce43jcjs4w004cg@wmail1.cc.ntu.edu.tw> Message-ID: <987190.73501.qm@web60323.mail.yahoo.com> Hi, Most likely, your bxsf file is corrupted, or has inconvenient format. How did you get it? Bests, Eyvaz. --- r95222066 at ntu.edu.tw wrote: > Dear all, > I have a error message about Fermi surface by > xcrysden > > Here is my error message > ------------------------------------------------------------------------------- > Number of atoms :0 > Number of frames :0 > Error: Error reading Band_Grid_3D_section,while > reading > Error: Error reading the BXSF file > ------------------------------------------------------------------------------- > > Thanks a lot > > Bing-Hong Chen,National > Taiwan University,Taiwan > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, IFM, Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com ____________________________________________________________________________________ Looking for last minute shopping deals? Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping From giannozz at nest.sns.it Tue Feb 12 09:41:31 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 12 Feb 2008 09:41:31 +0100 Subject: [Pw_forum] CVS Changes in output file format In-Reply-To: <010601c86d27$3da69aa0$b8f3cfe0$@net> References: <00aa01c86b7d$5512f950$ff38ebf0$@net> <010601c86d27$3da69aa0$b8f3cfe0$@net> Message-ID: <47B15BBB.5070500@nest.sns.it> Paul M. Grant wrote: > Paolo, just don't change the description/location of the Fermi energy! the Fermi energy is printed in an appropriate way. The other variables were printed in a confused way. > I'm rather curious about your remark that dforceb.f may still contain > errors...isn't this a key routine in Berry phase calculations? apparently the incorrect calls are never called Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From giannozz at nest.sns.it Tue Feb 12 19:18:51 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 12 Feb 2008 19:18:51 +0100 Subject: [Pw_forum] Ultrasoft pseudopotential for lanthanide In-Reply-To: <822f4ec80802112053g3284a1c5m6a7ad47e842ea01f@mail.gmail.com> References: <822f4ec80802112053g3284a1c5m6a7ad47e842ea01f@mail.gmail.com> Message-ID: <47B1E30B.4060900@nest.sns.it> shruba at gmail.com wrote: > I was looking for vanderbilt ultrasoft pseudopotential lanthanide > elements, I have checked pwscf, dacapo and vanderbilt ultrasoft library > 'http://www.physics.rutgers.edu/~dhv/uspp/" but I could not found any > lanthanides except La, If you know any other source/ websites where I > can find lanthanides ultrasoft pseudopotentials please let me know. the fact that nobody provides USPP for lanthanides seems to be a little suspect, don't you think? in addition to technical difficulties in the generation, there are serious problems of DFT shortcomings. Did you read the message just before yours? http://www.democritos.it/pipermail/pw_forum/2008-February/008290.html I am not saying that it is impossible to make serious calculations in f-electron systems, but whoever wants to do them should either be an expert or be ready to become one. Such calculations are not yet of the "plug and play" kind. If any, they rather belong to the "plug and pray" category Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From aryjunior at gmail.com Tue Feb 12 21:21:05 2008 From: aryjunior at gmail.com (Ary Junior) Date: Tue, 12 Feb 2008 18:21:05 -0200 Subject: [Pw_forum] XCrySDen question Message-ID: <4c9c61c20802121221p44718655gaaa04a1158f53d68@mail.gmail.com> Hi, when I try: xcrysden --pwi input.pw.inp I get an error. I've checked my "xcrys_tmp/" directory and the " pwi2xsf.xsf_out" for the structure was generated, but the " pwi2xsf.xsf_out.raw" was not... Anybody can help me? The error is: Running on platform: unix Executing: /opt/XCrySDen-1.4.1bin-static/bin/ftnunit *** the hardware does not support the stereo *** TEXT-WIDGET: .a1.f1.t Executing: sh /opt/XCrySDen-1.4.1bin-static/scripts/pwi2xsf.sh /home/aryjr/DFT-GFQSI-ARY/Praticas/semicondutores/si-modelo.pw.inp Error in startup script: error opening files: ~/xcrys_tmp/xc_4618/pwi2xsf.xsf_out and ~/xcrys_tmp/xc_4618/pwi2xsf.xsf_out.raw while executing "exec $system(BINDIR)/xsf2xsf $filedir $filedir.raw 3" ("eval" body line 1) invoked from within "eval {exec $system(BINDIR)/xsf2xsf $filedir $filedir.raw} $xcMisc(reduce_to)" (procedure "xsfOpen" line 26) invoked from within "xsfOpen $program_output .mesa" (procedure "openExtStruct" line 183) invoked from within "openExtStruct 3 crystal external [list sh $system(TOPDIR)/scripts/pwi2xsf.sh] pwi2xsf.xsf_out {PWSCF Input File} ANGS -file $filedir -preset pwIn..." invoked from within "if [file exists $filedir] { # pop-up Viewer ViewMol . openExtStruct 3 crystal external [list sh $system(TOPDIR)/scripts/pwi2xsf.sh] pwi2xsf..." ("--pwi" arm line 2) invoked from within "switch -glob -- $tag { "-a" - "--attrib*" { set load_attributes 1 set file_attributes $filedir } "--xsf" - "--xcr..." (procedure "parseComLinArg" line 44) invoked from within "parseComLinArg [lrange $argv 2 end]" invoked from within "if { [llength $argv] > 2 } { parseComLinArg [lrange $argv 2 end] } else { ViewMol . }" (file "/opt/XCrySDen-1.4.1bin-static/Tcl/xcInit.tcl" line 532) Thanks very much!!! Ary Junior -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080212/71ca4c4d/attachment.htm From r95222066 at ntu.edu.tw Wed Feb 13 04:58:24 2008 From: r95222066 at ntu.edu.tw (r95222066 at ntu.edu.tw) Date: Wed, 13 Feb 2008 11:58:24 +0800 Subject: [Pw_forum] hello, I have a error message about Fermi surface by xcrysden ? Message-ID: <20080213115824.m9j6xf02og0sws0w@wmail1.cc.ntu.edu.tw> Dear all, I have a error message about Fermi surface by xcrysden Here is my input file detail (PS:the file is little larger) a.NaV2O4.scf &control calculation='scf' restart_mode='from_scratch', pseudo_dir = './', outdir='./', prefix='nav2o4' / &system ibrav = 8, celldm(1) =17.253962,celldm(2)=0.315912,celldm(3)=1.164067, nat=28 , ntyp= 3, ecutwfc = 40.0, starting_magnetization(1)=0.1, starting_magnetization(2)=1, starting_magnetization(3)=0.1, occupations='smearing', smearing='mp', degauss=0.02 / &electrons conv_thr = 1.0e-4, mixing_beta = 0.05, electron_maxstep=80, diagonalization= 'cg' / ATOMIC_SPECIES Na 22.9878 Na.pbe-sp-van_ak.UPF V 50.9415 V.pbe-n-van.UPF O 15.9994 O.pbe-rrkjus.UPF ATOMIC_POSITIONS (crystal) Na 0.24271000 0.25000000 0.34620000 Na 0.75729000 0.75000000 0.65380000 Na 0.74271000 0.25000000 0.15380000 Na 0.25729000 0.75000000 0.84620000 V 0.08309000 0.25000000 0.60374000 V 0.91691000 0.75000000 0.39626000 V 0.58309000 0.25000000 0.89626000 V 0.41691000 0.75000000 0.10374000 V 0.06330000 0.25000000 0.11198000 V 0.93670000 0.75000000 0.88802000 V 0.56330000 0.25000000 0.38802000 V 0.43670000 0.75000000 0.61198000 O 0.29100000 0.25000000 0.64828000 O 0.70900000 0.75000000 0.35172000 O 0.79100000 0.25000000 0.85172000 O 0.20900000 0.75000000 0.14828000 O 0.38720000 0.25000000 0.97947000 O 0.61280000 0.75000000 0.02053000 O 0.88720000 0.25000000 0.52053000 O 0.11280000 0.75000000 0.47947000 O 0.38720000 0.25000000 0.21757000 O 0.61280000 0.75000000 0.78243000 O 0.88720000 0.25000000 0.28243000 O 0.11280000 0.75000000 0.71757000 O 0.07850000 0.25000000 0.92880000 O 0.92150000 0.75000000 0.07120000 O 0.57850000 0.25000000 0.57120000 O 0.42150000 0.75000000 0.42880000 K_POINTS (automatic) 6 6 6 0 0 0 b.NaV2O4.fs.in &control calculation='nscf' pseudo_dir = './', outdir='./', prefix='nav2o4' / &system ibrav = 8, celldm(1) =17.253962,celldm(2)=0.315912,celldm(3)=1.164067, nat=28 , ntyp= 3,nbnd=103, ecutwfc = 40.0, starting_magnetization(1)=0.1, starting_magnetization(2)=1, starting_magnetization(3)=0.1, / &electrons conv_thr = 1.0e-4, mixing_beta = 0.05, / ATOMIC_SPECIES Na 22.9878 Na.pbe-sp-van_ak.UPF V 50.9415 V.pbe-n-van.UPF O 15.9994 O.pbe-rrkjus.UPF ATOMIC_POSITIONS (crystal) Na 0.24271000 0.25000000 0.34620000 Na 0.75729000 0.75000000 0.65380000 Na 0.74271000 0.25000000 0.15380000 Na 0.25729000 0.75000000 0.84620000 V 0.08309000 0.25000000 0.60374000 V 0.91691000 0.75000000 0.39626000 V 0.58309000 0.25000000 0.89626000 V 0.41691000 0.75000000 0.10374000 V 0.06330000 0.25000000 0.11198000 V 0.93670000 0.75000000 0.88802000 V 0.56330000 0.25000000 0.38802000 V 0.43670000 0.75000000 0.61198000 O 0.29100000 0.25000000 0.64828000 O 0.70900000 0.75000000 0.35172000 O 0.79100000 0.25000000 0.85172000 O 0.20900000 0.75000000 0.14828000 O 0.38720000 0.25000000 0.97947000 O 0.61280000 0.75000000 0.02053000 O 0.88720000 0.25000000 0.52053000 O 0.11280000 0.75000000 0.47947000 O 0.38720000 0.25000000 0.21757000 O 0.61280000 0.75000000 0.78243000 O 0.88720000 0.25000000 0.28243000 O 0.11280000 0.75000000 0.71757000 O 0.07850000 0.25000000 0.92880000 O 0.92150000 0.75000000 0.07120000 O 0.57850000 0.25000000 0.57120000 O 0.42150000 0.75000000 0.42880000 K_POINTS (produced from kvecs_FS.in by kvecs_FS.x ) c.kvecs_FS.in 1.000000 0.000000 0.000000 0.000000 3.165438 0.000000 0.000000 0.000000 0.859057 6 6 6 nav2o4 d.input_FS 85 88 9.2434 nav2o4 FS 6 6 6 1.000000 0.000000 0.000000 0.000000 3.165438 0.000000 0.000000 0.000000 0.859057 ---------------------------------------------------------------------- I had a nav2o4.fs.bxsf file,and I check PWSCF example08 ni.fs.bxsf file which has the same format with my nav2o4.fs.bxsf file. When I want to show the bxsf file by xcrysden,there is the error message Number of atoms :0 Number of frames :0 Error: Error reading Band_Grid_3D_section,while reading Error: Error reading the BXSF file Thanks a lot Bing-Hong Chen,National Taiwan University,Taiwan From eyvaz_isaev at yahoo.com Wed Feb 13 11:40:49 2008 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Wed, 13 Feb 2008 02:40:49 -0800 (PST) Subject: [Pw_forum] hello, I have a error message about Fermi surface by xcrysden ? In-Reply-To: <20080213115824.m9j6xf02og0sws0w@wmail1.cc.ntu.edu.tw> Message-ID: <927409.61706.qm@web60323.mail.yahoo.com> Hi, Can you please send me your output files to have a look at? Bests, Eyvaz. --- r95222066 at ntu.edu.tw wrote: > Dear all, > I have a error message about Fermi surface by > xcrysden > > Here is my input file detail (PS:the file is little > larger) > > a.NaV2O4.scf > > &control > calculation='scf' > restart_mode='from_scratch', > pseudo_dir = './', > outdir='./', > prefix='nav2o4' > / > &system > ibrav = 8, celldm(1) > =17.253962,celldm(2)=0.315912,celldm(3)=1.164067, > nat=28 , ntyp= 3, > ecutwfc = 40.0, > starting_magnetization(1)=0.1, > starting_magnetization(2)=1, > starting_magnetization(3)=0.1, > occupations='smearing', smearing='mp', > degauss=0.02 > / > &electrons > conv_thr = 1.0e-4, > mixing_beta = 0.05, > electron_maxstep=80, > diagonalization= 'cg' > / > ATOMIC_SPECIES > Na 22.9878 Na.pbe-sp-van_ak.UPF > V 50.9415 V.pbe-n-van.UPF > O 15.9994 O.pbe-rrkjus.UPF > ATOMIC_POSITIONS (crystal) > Na 0.24271000 0.25000000 0.34620000 > Na 0.75729000 0.75000000 0.65380000 > Na 0.74271000 0.25000000 0.15380000 > Na 0.25729000 0.75000000 0.84620000 > V 0.08309000 0.25000000 0.60374000 > V 0.91691000 0.75000000 0.39626000 > V 0.58309000 0.25000000 0.89626000 > V 0.41691000 0.75000000 0.10374000 > V 0.06330000 0.25000000 0.11198000 > V 0.93670000 0.75000000 0.88802000 > V 0.56330000 0.25000000 0.38802000 > V 0.43670000 0.75000000 0.61198000 > O 0.29100000 0.25000000 0.64828000 > O 0.70900000 0.75000000 0.35172000 > O 0.79100000 0.25000000 0.85172000 > O 0.20900000 0.75000000 0.14828000 > O 0.38720000 0.25000000 0.97947000 > O 0.61280000 0.75000000 0.02053000 > O 0.88720000 0.25000000 0.52053000 > O 0.11280000 0.75000000 0.47947000 > O 0.38720000 0.25000000 0.21757000 > O 0.61280000 0.75000000 0.78243000 > O 0.88720000 0.25000000 0.28243000 > O 0.11280000 0.75000000 0.71757000 > O 0.07850000 0.25000000 0.92880000 > O 0.92150000 0.75000000 0.07120000 > O 0.57850000 0.25000000 0.57120000 > O 0.42150000 0.75000000 0.42880000 > K_POINTS (automatic) > 6 6 6 0 0 0 > > > b.NaV2O4.fs.in > &control > calculation='nscf' > pseudo_dir = './', > outdir='./', > prefix='nav2o4' > / > &system > ibrav = 8, celldm(1) > =17.253962,celldm(2)=0.315912,celldm(3)=1.164067, > nat=28 , ntyp= 3,nbnd=103, > ecutwfc = 40.0, > starting_magnetization(1)=0.1, > starting_magnetization(2)=1, > starting_magnetization(3)=0.1, > / > &electrons > conv_thr = 1.0e-4, > mixing_beta = 0.05, > > / > ATOMIC_SPECIES > Na 22.9878 Na.pbe-sp-van_ak.UPF > V 50.9415 V.pbe-n-van.UPF > O 15.9994 O.pbe-rrkjus.UPF > ATOMIC_POSITIONS (crystal) > Na 0.24271000 0.25000000 0.34620000 > Na 0.75729000 0.75000000 0.65380000 > Na 0.74271000 0.25000000 0.15380000 > Na 0.25729000 0.75000000 0.84620000 > V 0.08309000 0.25000000 0.60374000 > V 0.91691000 0.75000000 0.39626000 > V 0.58309000 0.25000000 0.89626000 > V 0.41691000 0.75000000 0.10374000 > V 0.06330000 0.25000000 0.11198000 > V 0.93670000 0.75000000 0.88802000 > V 0.56330000 0.25000000 0.38802000 > V 0.43670000 0.75000000 0.61198000 > O 0.29100000 0.25000000 0.64828000 > O 0.70900000 0.75000000 0.35172000 > O 0.79100000 0.25000000 0.85172000 > O 0.20900000 0.75000000 0.14828000 > O 0.38720000 0.25000000 0.97947000 > O 0.61280000 0.75000000 0.02053000 > O 0.88720000 0.25000000 0.52053000 > O 0.11280000 0.75000000 0.47947000 > O 0.38720000 0.25000000 0.21757000 > O 0.61280000 0.75000000 0.78243000 > O 0.88720000 0.25000000 0.28243000 > O 0.11280000 0.75000000 0.71757000 > O 0.07850000 0.25000000 0.92880000 > O 0.92150000 0.75000000 0.07120000 > O 0.57850000 0.25000000 0.57120000 > O 0.42150000 0.75000000 0.42880000 > K_POINTS > (produced from kvecs_FS.in by kvecs_FS.x ) > > > c.kvecs_FS.in > 1.000000 0.000000 0.000000 > 0.000000 3.165438 0.000000 > 0.000000 0.000000 0.859057 > 6 6 6 > nav2o4 > > d.input_FS > 85 88 > 9.2434 > nav2o4 > FS > 6 6 6 > 1.000000 0.000000 0.000000 > 0.000000 3.165438 0.000000 > 0.000000 0.000000 0.859057 > > ---------------------------------------------------------------------- > I had a nav2o4.fs.bxsf file,and I check PWSCF > example08 ni.fs.bxsf > file which has the same format with my > nav2o4.fs.bxsf file. > > When I want to show the bxsf file by > xcrysden,there is the error message > > Number of atoms :0 > Number of frames :0 > Error: Error reading Band_Grid_3D_section,while > reading > Error: Error reading the BXSF file > > > Thanks a lot > > Bing-Hong Chen,National Taiwan > University,Taiwan > > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, IFM, Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com ____________________________________________________________________________________ Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs From miguel.martinez at ehu.es Wed Feb 13 11:52:47 2008 From: miguel.martinez at ehu.es (=?ISO-8859-1?Q?Miguel_Marti=ADnez?=) Date: Wed, 13 Feb 2008 11:52:47 +0100 Subject: [Pw_forum] hello, I have a error message about Fermi surface by xcrysden ? In-Reply-To: <927409.61706.qm@web60323.mail.yahoo.com> References: <927409.61706.qm@web60323.mail.yahoo.com> Message-ID: <47B2CBFF.3020806@ehu.es> Dear colleagues, I have found that, sometimes, when calculating Fermi surfaces using Eyvaz's code, the nscf run adds several kpoints to the output. This would not invalidate the results, but the bxsf would be unreadable in xcrysden. I've used two workarounds: 1) Perform the nscf run with nosym=.true. 2) Delete the excess k-points (the first ones seem to be correct) So far I've seen this happen in a P6/mmm Be-Li alloy and in Fddd Gallium. As Bing-Hong Chen already has the bxsf, I'd work out how many points should there be (kx*ky*kz) and then use any decent text editor to delete the hundreds of excess lines. -- ---------------------------------------- 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 5326 ---------------------------------------- "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 eyvaz_isaev at yahoo.com Wed Feb 13 12:43:14 2008 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Wed, 13 Feb 2008 03:43:14 -0800 (PST) Subject: [Pw_forum] hello, I have a error message about Fermi surface by xcrysden ? In-Reply-To: <47B2CBFF.3020806@ehu.es> Message-ID: <688963.3799.qm@web60322.mail.yahoo.com> Dear Miguel, Can you please send me an example of output file with additional "several k-points" from nscf. Then I can try fix the problem. Bests, Eyvaz. --- Miguel Marti???nez wrote: > Dear colleagues, > > I have found that, sometimes, when calculating Fermi > surfaces using Eyvaz's > code, the nscf run adds several kpoints to the > output. This would not > invalidate the results, but the bxsf would be > unreadable in xcrysden. I've > used two workarounds: > > 1) Perform the nscf run with nosym=.true. > 2) Delete the excess k-points (the first ones seem > to be correct) > > So far I've seen this happen in a P6/mmm Be-Li alloy > and in Fddd Gallium. > As Bing-Hong Chen already has the bxsf, I'd work out > how many points should > there be (kx*ky*kz) and then use any decent text > editor to delete the > hundreds of excess lines. > > > -- > ---------------------------------------- > 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 5326 > ---------------------------------------- > > "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 > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, IFM, Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com ____________________________________________________________________________________ Looking for last minute shopping deals? Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping From mousumi at jncasr.ac.in Wed Feb 13 19:19:31 2008 From: mousumi at jncasr.ac.in (Mousumi Upadhyay Kahaly) Date: Wed, 13 Feb 2008 23:49:31 +0530 (IST) Subject: [Pw_forum] Make file for JS20 power pc In-Reply-To: <20080213115824.m9j6xf02og0sws0w@wmail1.cc.ntu.edu.tw> References: <20080213115824.m9j6xf02og0sws0w@wmail1.cc.ntu.edu.tw> Message-ID: <41778.134.105.152.73.1202926771.squirrel@134.105.152.73> Dear All, I'm trying to compile espresso-3.2.3 (both serial and parallel versions) in a cluster with following configuration: 70 IBM JS20 PowerBlades with 2 PPC970 (1.6GHz) CPU's Operating system: SLES8 (Linux, Susse) Can any of you please provide me make.sys file for such system? Actually, I'm facing some problem during installation. While running ./configure , though it is completed successfully, I get a warning message, ------------------------------------------------------------------------- checking build system type... powerpc64-unknown-linux-gnu configure: WARNING: incorrect host name ? checking architecture... configure: WARNING: unsupported architecture? checking for xlf90... xlf90 checking for Fortran 77 compiler default output... a.out checking whether the Fortran 77 compiler works... yes checking whether we are cross compiling... yes checking for suffix of executables... checking for suffix of object files... o checking whether we are using the GNU Fortran 77 compiler... no ----------------------------------------------------------------------- However, after this, while running "make all" with the generated make.sys file, I get the following error message. -------------------------------------------------------------------------- cpp -P -traditional -D__FFTW -D__USE_INTERNAL_FFTW -D__MPI -D__PARA -I../include iotk_base.f90 -o iotk_base.F90 xlf90 -O -D__FFTW -D__USE_INTERNAL_FFTW -D__MPI -D__PARA -I../include -I./ -I../Modules -I../iotk/src -I../PW -I../PH -I../CPV -c iotk_base.F90 -o iotk_base.o xlf90: 1501-216 command option -__FFTW is not recognized - passed to ld xlf90: 1501-216 command option -__USE_INTERNAL_FFTW is not recognized - passed to ld xlf90: 1501-216 command option -__MPI is not recognized - passed to ld xlf90: 1501-216 command option -__PARA is not recognized - passed to ld xlf90: 1501-218 file iotk_base.F90 contains an incorrect file suffix make[2]: *** [iotk_base.o] Error 1 make[2]: Leaving directory `/home/users/mousumi/godot/espresso-3.2.3/iotk/src' make[1]: *** [libiotk.a] Error 2 make[1]: Leaving directory `/home/users/mousumi/godot/espresso-3.2.3/iotk' make: *** [libiotk] Error 2 ------------------------------------------------------------------------ Could you please help me to sort this out? Thanks and best regards, mousumi. Mousumi Upadhyay Kahaly Research Associate, Theoretical Sciences Unit, JNCASR, Bangalore, India. From ari.p.seitsonen at iki.fi Wed Feb 13 19:12:47 2008 From: ari.p.seitsonen at iki.fi (Ari P Seitsonen) Date: Wed, 13 Feb 2008 19:12:47 +0100 (CET) Subject: [Pw_forum] Make file for JS20 power pc In-Reply-To: <41778.134.105.152.73.1202926771.squirrel@134.105.152.73> References: <20080213115824.m9j6xf02og0sws0w@wmail1.cc.ntu.edu.tw> <41778.134.105.152.73.1202926771.squirrel@134.105.152.73> Message-ID: Dear Mousumi, The problem is the "strange" mixture of AIX vs linux options. Specifically, the xlf* compilers should NOT get the '-D...' options on the command line at all (because they are processed with the 'cpp' already). OR, you tell the 'configure' that it should process the *.f90 files directly (thus avoiding 'cpp'), but for this you need to check the compiler options for xlf* anyway. The fastest way is probably to take the '-D*' options away from the compiler flags (and thus leave them only for the pre-processor 'cpp'). The next problem will probably be, how to compile the parallel version: With 'mpxlf*' or with 'mpif90'; that depends on the installation of your system, but I guess it will be the second one. Good luck! Greetings from Paris, apsi On Wed, 13 Feb 2008, Mousumi Upadhyay Kahaly wrote: > Dear All, > > I'm trying to compile espresso-3.2.3 (both serial and parallel versions) > in a cluster with following configuration: > 70 IBM JS20 PowerBlades with 2 PPC970 (1.6GHz) CPU's > Operating system: SLES8 (Linux, Susse) > Can any of you please provide me make.sys file for such system? Actually, > I'm facing some problem during installation. > > While running ./configure , though it is completed successfully, I get a > warning message, > ------------------------------------------------------------------------- > checking build system type... powerpc64-unknown-linux-gnu > configure: WARNING: incorrect host name ? > checking architecture... > configure: WARNING: unsupported architecture? > checking for xlf90... xlf90 > checking for Fortran 77 compiler default output... a.out > checking whether the Fortran 77 compiler works... yes > checking whether we are cross compiling... yes > checking for suffix of executables... > checking for suffix of object files... o > checking whether we are using the GNU Fortran 77 compiler... no > ----------------------------------------------------------------------- > > However, after this, while running "make all" with the generated make.sys > file, I get the following error message. > -------------------------------------------------------------------------- > cpp -P -traditional -D__FFTW -D__USE_INTERNAL_FFTW -D__MPI -D__PARA > -I../include iotk_base.f90 -o iotk_base.F90 > xlf90 -O -D__FFTW -D__USE_INTERNAL_FFTW -D__MPI -D__PARA -I../include > -I./ -I../Modules -I../iotk/src -I../PW -I../PH -I../CPV -c > iotk_base.F90 -o iotk_base.o > xlf90: 1501-216 command option -__FFTW is not recognized - passed to ld > xlf90: 1501-216 command option -__USE_INTERNAL_FFTW is not recognized - > passed to ld > xlf90: 1501-216 command option -__MPI is not recognized - passed to ld > xlf90: 1501-216 command option -__PARA is not recognized - passed to ld > xlf90: 1501-218 file iotk_base.F90 contains an incorrect file suffix > make[2]: *** [iotk_base.o] Error 1 > make[2]: Leaving directory > `/home/users/mousumi/godot/espresso-3.2.3/iotk/src' > make[1]: *** [libiotk.a] Error 2 > make[1]: Leaving directory `/home/users/mousumi/godot/espresso-3.2.3/iotk' > make: *** [libiotk] Error 2 > ------------------------------------------------------------------------ > > Could you please help me to sort this out? > > Thanks and best regards, > mousumi. > > Mousumi Upadhyay Kahaly > Research Associate, > Theoretical Sciences Unit, > JNCASR, Bangalore, India. > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- Ari P Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/ IMPMC, CNRS & Universit? Pierre et Marie Curie Tel: +33-1-4427 7542, Fax: +33-1-4427 3785, GSM: +33-6-6736 3820 From giannozz at nest.sns.it Wed Feb 13 19:29:49 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Wed, 13 Feb 2008 19:29:49 +0100 Subject: [Pw_forum] Make file for JS20 power pc In-Reply-To: <41778.134.105.152.73.1202926771.squirrel@134.105.152.73> References: <20080213115824.m9j6xf02og0sws0w@wmail1.cc.ntu.edu.tw> <41778.134.105.152.73.1202926771.squirrel@134.105.152.73> Message-ID: On Feb 13, 2008, at 19:19 , Mousumi Upadhyay Kahaly wrote: > Could you please help me to sort this out? try the "configure" from the cvs version: it should support a machine in Barcelona that is almost as strange as yours. Paolo --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From nemeth at anl.gov Wed Feb 13 22:26:41 2008 From: nemeth at anl.gov (Karoly Nemeth) Date: Wed, 13 Feb 2008 15:26:41 -0600 (CST) Subject: [Pw_forum] fixed vs starting magnetization In-Reply-To: References: Message-ID: Guys, is there any way to use constrained_magnetization='atomic' such that for some atoms for which starting magnetization has been defined the magnetization will be allowed to vary during the SCF, for other ones the magnetization would be kept fixed? It appears from INPUT_PW, that once one uses constrained_magnetization='atomic', all starting magnetizations will be frozen. Thanks: Karoly From apczhang at yahoo.cn Thu Feb 14 03:46:14 2008 From: apczhang at yahoo.cn (Chao) Date: Thu, 14 Feb 2008 10:46:14 +0800 (CST) Subject: [Pw_forum] About parallel problem on SGI Altix 3700 Message-ID: <942459.15212.qm@web92201.mail.cnh.yahoo.com> Dear All I met a problem when using PWscf for parallel computation. I submitted my job in the following format: mpirun -np 8 pw.x -npool 4 < .in > .out But my job only ran on one CPU, and some messages were listed: MPI: could not run executable (case #3) MPI: No details available, no log files found When I complied PWscf, some errors appeared when make all: IPO Error: unresolved : mpi_conversion_fn_null_ Referenced in ../Modules/parallel_include.o Referenced in ../Modules/ptoolkit.o Referenced in libiotk.a(iotk_error.o) Can anyone help me? Best regards ZHANG, Chao City University of Hong Kong --------------------------------- ??????????????????? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080214/ee4b3232/attachment.htm From lawrence_lee_lee at yahoo.com.hk Thu Feb 14 04:02:28 2008 From: lawrence_lee_lee at yahoo.com.hk (Lawrence) Date: Thu, 14 Feb 2008 11:02:28 +0800 Subject: [Pw_forum] An error occur when I use another set of pseudopotential Message-ID: <1202958148.6464.6.camel@scg11> Dear all, I can run a cp.x run smoothly when using one set of pseudopotential. However, when I change to another set, an error as shown below occur. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from simpson : error # 8 few mesh points %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% And then the run stopped. What are the possible causes for this problem? Thanks very much for help! -- S.H. Lee (shlee at phy.cuhk.edu.hk) M. Phil Physics Division The Chinese University of Hong Kong Graduate School _______________________________________ ?x?????? - Yahoo! Messenger ???????]?????W?????????????????????????o???????????W?r???????????????????f???????????? http://messenger.yahoo.com.hk From akohlmey at cmm.chem.upenn.edu Thu Feb 14 06:53:48 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Thu, 14 Feb 2008 00:53:48 -0500 Subject: [Pw_forum] An error occur when I use another set of pseudopotential In-Reply-To: <1202958148.6464.6.camel@scg11> References: <1202958148.6464.6.camel@scg11> Message-ID: <7b6913e90802132153q7b5497aek13526905d68a0ab5@mail.gmail.com> On Wed, Feb 13, 2008 at 10:02 PM, Lawrence wrote: > Dear all, > > I can run a cp.x run smoothly when using one set of pseudopotential. > However, when I change to another set, an error as shown below occur. dear lawrence, please see the comments made to other error reports you have to tell people more details of how to reproduce the problem (input examples, pseudopotentials etc.) or else (almost) nobody will be able to tell you anything helpful. perhaps this is again a good time to post URLs to a couple of webpages that describe how to handle error reports for software in general: http://freshmeat.net/articles/view/149/ http://www.catb.org/~esr/faqs/smart-questions.html cheers, axel. > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > from simpson : error # 8 > few mesh points > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > And then the run stopped. What are the possible causes for this problem? > > Thanks very much for help! > > > -- > S.H. Lee (shlee at phy.cuhk.edu.hk) > M. Phil > Physics Division > The Chinese University of Hong Kong Graduate School > > > _______________________________________ > ?x??? ?? - Yahoo! Messenger > ???????]?????W????????????????????? ???o????(R)"?????W?r???????????????????f?'?????????? > http://messenger.yahoo.com.hk > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From sagarambavale at yahoo.co.in Thu Feb 14 09:01:57 2008 From: sagarambavale at yahoo.co.in (ambavale sagar) Date: Thu, 14 Feb 2008 13:31:57 +0530 (IST) Subject: [Pw_forum] error on zggev. Message-ID: <36432.33456.qm@web94604.mail.in2.yahoo.com> An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080214/b5498ce7/attachment.htm From jzwanzig at dal.ca Thu Feb 14 09:09:13 2008 From: jzwanzig at dal.ca (Josef Zwanziger) Date: Thu, 14 Feb 2008 09:09:13 +0100 Subject: [Pw_forum] GIPAW pseudopotentials Message-ID: <200802140909.13575.jzwanzig@dal.ca> Hello, I have been looking at the GIPAW modules in the 4.0cvs pre-release, and while I can compile and run the example without problem, it's not clear how to make pseudopotentials for GIPAW for other atoms. I notice that the given test pseudopotentials, for silicon for example, include blocks of GIPAW specific data, but I can't find any corresponding options in the ld1 input files or for that matter the ld1 source code that corresponds to these options. Also unclear is whether the GIPAW module as currently available works with non-norm-conserving data, or currently only with norm-conserving data. Any clarification would be helpful. thanks, Joe -- Josef W. Zwanziger Canada Research Chair in NMR Studies of Materials Director, Atlantic Region Magnetic Resonance Centre Dept. of Chemistry and Inst. for Research in Materials Dalhousie University Halifax B3H 4J3 Canada +1 (902) 494-1960 jzwanzig at dal.ca http://jwz.chem.dal.ca On sabbatical January-July 2008 Max Planck Institute for Polymer Research Mainz, Germany From dalcorso at sissa.it Thu Feb 14 09:40:21 2008 From: dalcorso at sissa.it (Dal Corso Andrea) Date: Thu, 14 Feb 2008 09:40:21 +0100 Subject: [Pw_forum] error on zggev. In-Reply-To: <36432.33456.qm@web94604.mail.in2.yahoo.com> References: <36432.33456.qm@web94604.mail.in2.yahoo.com> Message-ID: <1202978421.3397.8.camel@dhpc-5-48.sissa.it> Dear Ambavale Sagar, The two parameters ewind and epsproj control the accuracy of the pwcond calculation. Large ewind and small epsproj means better accuracy. In the example ewind is very small and epsproj very large. Typical values are ewind=3-4 Ry, and epsproj=1d-7, 1d-9. Your example already converges with ewind=2.0 and epsproj=1.d-4 but with these parameters the bands might be inaccurate. You need to check the convergence with respect to these two parameters. Hope this helps, Andrea On Thu, 2008-02-14 at 13:31 +0530, ambavale sagar wrote: > > > Sorry at first i sent this message to pw_users@ pwscf.org. > Where will it reach? > > I am trying to run pwcond.x to find complex band structure of > Si by simply following the example 12 - Al case. I am getting > the error describing it as error on zggev. > (P.S. I dont have fftw lib installed, but example for Al runs > well....) > > the detailed input for scf is > &control > calculation='scf' > restart_mode='from_scratch', > pseudo_dir = '/home/phy/Desktop/espresso-3.2.3/pseudo/', > outdir='/home/phy/tmp/' > wf_collect = .TRUE. > prefix='si' > / > &system > ibrav = 6, > celldm(1) =7.5, > celldm(3) =1.414, > nat= 2, > ntyp= 1, > ecutwfc = 15.07, > occupations='smearing', > smearing='methfessel-paxton', > degauss=0.01 > / > &electrons > conv_thr = 1.0e-8 > mixing_beta = 0.7 > / > ATOMIC_SPECIES > Si 28.086 Si.vbc.UPF > ATOMIC_POSITIONS > Si 0.0 0.0 0.0 > Si 0.5 0.5 0.707 > K_POINTS (automatic) > 4 4 4 1 1 1 > > and input for complex band calculation is > > &inputcond > outdir = '/home/phy/tmp/' > prefixl = 'si' > band_file = 'bands.si' > ikind = 0 > energy0 = 10.d0 > denergy = -0.4d0 > ewind = 1.d0 > epsproj = 1.d-3 > > > > > delgep = 1.d-12 > cutplot = 3.d0 > / > 1 > 0.0 0.0 1.0 > 60 > > and output file is > > Program POST-PROC v.3.2.3 starts ... > Today is 14Feb2008 at 12:37:30 > nbndx = 8 nbnd = 8 natomwfc = 8 npwx = > 595 > nelec = 8.00 nkb = 8 ngl = 375 > > ===== INPUT FILE containing the left lead ===== > > GEOMETRY: > > lattice parameter (a_0) = 7.5000 a.u. > the volume = 596.5313 (a.u.)^3 > the cross section = 56.2500 (a.u.)^2 > l of the unit cell = 1.4140 (a_0) > number of atoms/cell = 2 > number of atomic types = 1 > > crystal axes: (cart. coord. in units of a_0) > a(1) = ( 1.0000 0.0000 0.0000 ) > a(2) = ( 0.0000 1.0000 0.0000 ) > a(3) = ( 0.0000 0.0000 1.4140 ) > > > Cartesian axes > > site n. atom positions (a_0 units) > 1 Si tau( 1)=( 0.0000 0.0000 > 1.4140 ) > 2 Si tau( 2)=( 0.5000 0.5000 > 0.7070 ) > > nr1s = 20 > nr2s = 20 > nr3s = 27 > nrx1s = 20 > nrx2s = 20 > nrx3s = 27 > nr1 = 20 > nr2 = 20 > nr3 = 27 > nrx1 = 20 > nrx2 = 20 > nrx3 = 27 > > _______________________________ > Radii of nonlocal spheres: > > type ibeta ang. mom. radius (a_0 > units) > Si 1 0 0.3852 > Si 2 1 0.4050 > ----- General information ----- > > ----- Complex band structure calculation ----- > > nrx = 20 > nry = 20 > nz1 = 11 > > > energy0 = 1.0E+01 > denergy = -4.0E-01 > nenergy = 60 > ecut2d = 1.5E+01 > ewind = 1.0E+00 > epsproj = 1.0E-03 > > > number of k_|| points= 1 > cart. coord. in units 2pi/a_0 > k( 1) = ( 0.0000000 0.0000000), wk = > 1.0000000 > ----- Information about left lead ----- > > nocros = 4 > noins = 4 > norb = 12 > norbf = 12 > nrz = 27 > > iorb type ibeta ang. mom. m position > (a_0) > 1 1 1 0 1 taunew( 1)=( > 0.0000 0.0000 0.0000) > 2 1 2 1 1 taunew( 2)=( > 0.0000 0.0000 0.0000) > 3 1 2 1 2 taunew( 3)=( > 0.0000 0.0000 0.0000) > 4 1 2 1 3 taunew( 4)=( > 0.0000 0.0000 0.0000) > 5 1 1 0 1 taunew( 5)=( > 0.5000 0.5000 0.7070) > 6 1 2 1 1 taunew( 6)=( > 0.5000 0.5000 0.7070) > 7 1 2 1 2 taunew( 7)=( > 0.5000 0.5000 0.7070) > 8 1 2 1 3 taunew( 8)=( > 0.5000 0.5000 0.7070) > 9 1 1 0 1 taunew( 9)=( > 0.0000 0.0000 1.4140) > 10 1 2 1 1 taunew( 10)=( > 0.0000 0.0000 1.4140) > 11 1 2 1 2 taunew( 11)=( > 0.0000 0.0000 1.4140) > 12 1 2 1 3 taunew( 12)=( > 0.0000 0.0000 1.4140) > k slab z(k) z(k+1) crossing(iorb=1,norb) > 1 0.0000 0.0524 0.0524 111100000000 > 2 0.0524 0.1047 0.0524 111100000000 > 3 0.1047 0.1571 0.0524 111100000000 > 4 0.1571 0.2095 0.0524 111100000000 > 5 0.2095 0.2619 0.0524 111100000000 > 6 0.2619 0.3142 0.0524 111101110000 > 7 0.3142 0.3666 0.0524 111111110000 > 8 0.3666 0.4190 0.0524 111111110000 > 9 0.4190 0.4713 0.0524 000011110000 > 10 0.4713 0.5237 0.0524 000011110000 > 11 0.5237 0.5761 0.0524 000011110000 > 12 0.5761 0.6284 0.0524 000011110000 > 13 0.6284 0.6808 0.0524 000011110000 > 14 0.6808 0.7332 0.0524 000011110000 > 15 0.7332 0.7856 0.0524 000011110000 > 16 0.7856 0.8379 0.0524 000011110000 > 17 0.8379 0.8903 0.0524 000011110000 > 18 0.8903 0.9427 0.0524 000011110000 > 19 0.9427 0.9950 0.0524 000011110000 > 20 0.9950 1.0474 0.0524 000011111111 > 21 1.0474 1.0998 0.0524 000011111111 > 22 1.0998 1.1521 0.0524 000001111111 > 23 1.1521 1.2045 0.0524 000000001111 > 24 1.2045 1.2569 0.0524 000000001111 > 25 1.2569 1.3093 0.0524 000000001111 > 26 1.3093 1.3616 0.0524 000000001111 > 27 1.3616 1.4140 0.0524 000000001111 > > k( 1) = ( 0.0000000 0.0000000), wk = > 1.0000000 > > ngper, shell number = 69 13 > ngper, n2d = 69 39 > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > %%%%%%%%%%%%%%%%% > from gep_x : error # 48 > error on zggev > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > %%%%%%%%%%%%%%%%% > > stopping ... > > > > > Sagar K. Ambavale > Research student > The M.S. University of Baroda > India > > > > > ______________________________________________________________ > Chat on a cool, new interface. No download required. Click > here. > > > ______________________________________________________________________ > Bollywood, fun, friendship, sports and more. You name it, we have it. > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- Andrea Dal Corso Tel. 0039-040-3787428 SISSA, Via Beirut 2/4 Fax. 0039-040-3787528 34014 Trieste (Italy) e-mail: dalcorso at sissa.it From paulatto at sissa.it Thu Feb 14 09:49:30 2008 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Thu, 14 Feb 2008 09:49:30 +0100 (CET) Subject: [Pw_forum] About parallel problem on SGI Altix 3700 In-Reply-To: <942459.15212.qm@web92201.mail.cnh.yahoo.com> References: <942459.15212.qm@web92201.mail.cnh.yahoo.com> Message-ID: <16237.147.122.5.182.1202978970.squirrel@webmail.sissa.it> On Thu, February 14, 2008 03:46, Chao wrote: > But my job only ran on one CPU, and some messages were listed: > MPI: could not run executable (case #3) > MPI: No details available, no log files found It looks like you did not compile the parallel version of the code (which happens when ./configure cannot detect mpi libraries), or you did some mistake in the mpi configuration (is the binary present on all computers?). Does the configure script shows these lines at the end: Parallel environment detected successfully. Configured for compilation of parallel executables. > When I complied PWscf, some errors appeared when make all: > IPO Error: unresolved : mpi_conversion_fn_null_ > Referenced in ../Modules/parallel_include.o > Referenced in ../Modules/ptoolkit.o > Referenced in libiotk.a(iotk_error.o) You can disable Interprocedural Optimizations (IPO) with the option -no-ipo. IPO is usually a source of troubles; it didn't worked at all on some versions of ifort (maybe 9.0). Please give us more details, like you configure output, compiler and mpi flavour. regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) +39 040 3787 511 http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From giannozz at nest.sns.it Thu Feb 14 10:15:19 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 14 Feb 2008 10:15:19 +0100 Subject: [Pw_forum] An error occur when I use another set of pseudopotential In-Reply-To: <7b6913e90802132153q7b5497aek13526905d68a0ab5@mail.gmail.com> References: <1202958148.6464.6.camel@scg11> <7b6913e90802132153q7b5497aek13526905d68a0ab5@mail.gmail.com> Message-ID: On Feb 14, 2008, at 6:53 , Axel Kohlmeyer wrote: > On Wed, Feb 13, 2008 at 10:02 PM, Lawrence > wrote: > > perhaps this is again a good time to post URLs to > a couple of webpages that describe how to handle error > reports for software in general: > http://freshmeat.net/articles/view/149/ > http://www.catb.org/~esr/faqs/smart-questions.html ...and this one for Q-E (in particular, point 4): http://www.quantum-espresso.org/wiki/index.php/ Developer_Manual#How_to_Contribute_to_Quantum-ESPRESSO P. --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From smogunov at sissa.it Thu Feb 14 10:39:44 2008 From: smogunov at sissa.it (Alexander) Date: Thu, 14 Feb 2008 10:39:44 +0100 Subject: [Pw_forum] error on zggev. In-Reply-To: <1202978421.3397.8.camel@dhpc-5-48.sissa.it> References: <36432.33456.qm@web94604.mail.in2.yahoo.com> <1202978421.3397.8.camel@dhpc-5-48.sissa.it> Message-ID: <200802141039.44561.smogunov@sissa.it> Dear Ambavale Sagar I would also try to increase delgep, from 1.d-12 to say 1.d-10, now it seems to be very small. It should help to stabilize the solution of eigenvalue problem, the point where the run crashes. Best regards, Alexander Smogunov On Thursday 14 February 2008 09:40, Dal Corso Andrea wrote: > Dear Ambavale Sagar, > > The two parameters ewind and epsproj control the accuracy of the pwcond > calculation. Large ewind and small epsproj means better accuracy. > In the example ewind is very small and epsproj very large. Typical > values are ewind=3-4 Ry, and epsproj=1d-7, 1d-9. Your example already > converges with ewind=2.0 and epsproj=1.d-4 but with these parameters the > bands might be inaccurate. You need to check the convergence with > respect to these two parameters. > > Hope this helps, > > Andrea > > On Thu, 2008-02-14 at 13:31 +0530, ambavale sagar wrote: > > Sorry at first i sent this message to pw_users@ pwscf.org. > > Where will it reach? > > > > I am trying to run pwcond.x to find complex band structure of > > Si by simply following the example 12 - Al case. I am getting > > the error describing it as error on zggev. > > (P.S. I dont have fftw lib installed, but example for Al runs > > well....) > > > > the detailed input for scf is > > &control > > calculation='scf' > > restart_mode='from_scratch', > > pseudo_dir = '/home/phy/Desktop/espresso-3.2.3/pseudo/', > > outdir='/home/phy/tmp/' > > wf_collect = .TRUE. > > prefix='si' > > / > > &system > > ibrav = 6, > > celldm(1) =7.5, > > celldm(3) =1.414, > > nat= 2, > > ntyp= 1, > > ecutwfc = 15.07, > > occupations='smearing', > > smearing='methfessel-paxton', > > degauss=0.01 > > / > > &electrons > > conv_thr = 1.0e-8 > > mixing_beta = 0.7 > > / > > ATOMIC_SPECIES > > Si 28.086 Si.vbc.UPF > > ATOMIC_POSITIONS > > Si 0.0 0.0 0.0 > > Si 0.5 0.5 0.707 > > K_POINTS (automatic) > > 4 4 4 1 1 1 > > > > and input for complex band calculation is > > > > &inputcond > > outdir = '/home/phy/tmp/' > > prefixl = 'si' > > band_file = 'bands.si' > > ikind = 0 > > energy0 = 10.d0 > > denergy = -0.4d0 > > ewind = 1.d0 > > epsproj = 1.d-3 > > > > > > > > > > delgep = 1.d-12 > > cutplot = 3.d0 > > / > > 1 > > 0.0 0.0 1.0 > > 60 > > > > and output file is > > > > Program POST-PROC v.3.2.3 starts ... > > Today is 14Feb2008 at 12:37:30 > > nbndx = 8 nbnd = 8 natomwfc = 8 npwx = > > 595 > > nelec = 8.00 nkb = 8 ngl = 375 > > > > ===== INPUT FILE containing the left lead ===== > > > > GEOMETRY: > > > > lattice parameter (a_0) = 7.5000 a.u. > > the volume = 596.5313 (a.u.)^3 > > the cross section = 56.2500 (a.u.)^2 > > l of the unit cell = 1.4140 (a_0) > > number of atoms/cell = 2 > > number of atomic types = 1 > > > > crystal axes: (cart. coord. in units of a_0) > > a(1) = ( 1.0000 0.0000 0.0000 ) > > a(2) = ( 0.0000 1.0000 0.0000 ) > > a(3) = ( 0.0000 0.0000 1.4140 ) > > > > > > Cartesian axes > > > > site n. atom positions (a_0 units) > > 1 Si tau( 1)=( 0.0000 0.0000 > > 1.4140 ) > > 2 Si tau( 2)=( 0.5000 0.5000 > > 0.7070 ) > > > > nr1s = 20 > > nr2s = 20 > > nr3s = 27 > > nrx1s = 20 > > nrx2s = 20 > > nrx3s = 27 > > nr1 = 20 > > nr2 = 20 > > nr3 = 27 > > nrx1 = 20 > > nrx2 = 20 > > nrx3 = 27 > > > > _______________________________ > > Radii of nonlocal spheres: > > > > type ibeta ang. mom. radius (a_0 > > units) > > Si 1 0 0.3852 > > Si 2 1 0.4050 > > ----- General information ----- > > > > ----- Complex band structure calculation ----- > > > > nrx = 20 > > nry = 20 > > nz1 = 11 > > > > > > energy0 = 1.0E+01 > > denergy = -4.0E-01 > > nenergy = 60 > > ecut2d = 1.5E+01 > > ewind = 1.0E+00 > > epsproj = 1.0E-03 > > > > > > number of k_|| points= 1 > > cart. coord. in units 2pi/a_0 > > k( 1) = ( 0.0000000 0.0000000), wk = > > 1.0000000 > > ----- Information about left lead ----- > > > > nocros = 4 > > noins = 4 > > norb = 12 > > norbf = 12 > > nrz = 27 > > > > iorb type ibeta ang. mom. m position > > (a_0) > > 1 1 1 0 1 taunew( 1)=( > > 0.0000 0.0000 0.0000) > > 2 1 2 1 1 taunew( 2)=( > > 0.0000 0.0000 0.0000) > > 3 1 2 1 2 taunew( 3)=( > > 0.0000 0.0000 0.0000) > > 4 1 2 1 3 taunew( 4)=( > > 0.0000 0.0000 0.0000) > > 5 1 1 0 1 taunew( 5)=( > > 0.5000 0.5000 0.7070) > > 6 1 2 1 1 taunew( 6)=( > > 0.5000 0.5000 0.7070) > > 7 1 2 1 2 taunew( 7)=( > > 0.5000 0.5000 0.7070) > > 8 1 2 1 3 taunew( 8)=( > > 0.5000 0.5000 0.7070) > > 9 1 1 0 1 taunew( 9)=( > > 0.0000 0.0000 1.4140) > > 10 1 2 1 1 taunew( 10)=( > > 0.0000 0.0000 1.4140) > > 11 1 2 1 2 taunew( 11)=( > > 0.0000 0.0000 1.4140) > > 12 1 2 1 3 taunew( 12)=( > > 0.0000 0.0000 1.4140) > > k slab z(k) z(k+1) crossing(iorb=1,norb) > > 1 0.0000 0.0524 0.0524 111100000000 > > 2 0.0524 0.1047 0.0524 111100000000 > > 3 0.1047 0.1571 0.0524 111100000000 > > 4 0.1571 0.2095 0.0524 111100000000 > > 5 0.2095 0.2619 0.0524 111100000000 > > 6 0.2619 0.3142 0.0524 111101110000 > > 7 0.3142 0.3666 0.0524 111111110000 > > 8 0.3666 0.4190 0.0524 111111110000 > > 9 0.4190 0.4713 0.0524 000011110000 > > 10 0.4713 0.5237 0.0524 000011110000 > > 11 0.5237 0.5761 0.0524 000011110000 > > 12 0.5761 0.6284 0.0524 000011110000 > > 13 0.6284 0.6808 0.0524 000011110000 > > 14 0.6808 0.7332 0.0524 000011110000 > > 15 0.7332 0.7856 0.0524 000011110000 > > 16 0.7856 0.8379 0.0524 000011110000 > > 17 0.8379 0.8903 0.0524 000011110000 > > 18 0.8903 0.9427 0.0524 000011110000 > > 19 0.9427 0.9950 0.0524 000011110000 > > 20 0.9950 1.0474 0.0524 000011111111 > > 21 1.0474 1.0998 0.0524 000011111111 > > 22 1.0998 1.1521 0.0524 000001111111 > > 23 1.1521 1.2045 0.0524 000000001111 > > 24 1.2045 1.2569 0.0524 000000001111 > > 25 1.2569 1.3093 0.0524 000000001111 > > 26 1.3093 1.3616 0.0524 000000001111 > > 27 1.3616 1.4140 0.0524 000000001111 > > > > k( 1) = ( 0.0000000 0.0000000), wk = > > 1.0000000 > > > > ngper, shell number = 69 13 > > ngper, n2d = 69 39 > > > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > %%%%%%%%%%%%%%%%% > > from gep_x : error # 48 > > error on zggev > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > %%%%%%%%%%%%%%%%% > > > > stopping ... > > > > > > > > > > Sagar K. Ambavale > > Research student > > The M.S. University of Baroda > > India > > > > > > > > > > ______________________________________________________________ > > Chat on a cool, new interface. No download required. Click > > here. > > > > > > ______________________________________________________________________ > > Bollywood, fun, friendship, sports and more. You name it, we have it. > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum From apczhang at yahoo.cn Thu Feb 14 16:39:21 2008 From: apczhang at yahoo.cn (Chao) Date: Thu, 14 Feb 2008 23:39:21 +0800 (CST) Subject: [Pw_forum] About parallel problem on SGI Altix 3700 In-Reply-To: <16237.147.122.5.182.1202978970.squirrel@webmail.sissa.it> Message-ID: <613082.91157.qm@web92206.mail.cnh.yahoo.com> Dear Paulatto These are the last sentences as runing configure script. It seems that the configure is successful. -------------------------------------------------------------------- ESPRESSO can take advantage of several optimized numerical libraries (essl, fftw, mkl...). This configure script attempts to find them, but may fail if they have been installed in non-standard locations. If a required library is not found, the local copy will be compiled. The following libraries have been found: BLAS_LIBS=-L/opt/intel/mkl/8.0/lib/64 -lmkl_ipf -lguide -lpthread LAPACK_LIBS=-lmkl_lapack FFT_LIBS= Please check if this is what you expect. If any libraries are missing, you may specify a list of directories to search and retry, as follows: ./configure LIBDIRS="list of directories, separated by spaces" Parallel environment detected successfully. Configured for compilation of parallel executables. For more info, read the ESPRESSO User's Guide (Doc/users-guide.tex). -------------------------------------------------------------------- configure: success The attached is the configure output. Best regards ZHANG Chao City University of Hong Kong --------------------------------- ??????????????????? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080214/15d2fd26/attachment.htm -------------- next part -------------- A non-text attachment was scrubbed... Name: log1 Type: application/octet-stream Size: 5235 bytes Desc: 2460937548-log1 Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080214/15d2fd26/attachment.obj From giannozz at nest.sns.it Thu Feb 14 16:53:27 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 14 Feb 2008 16:53:27 +0100 Subject: [Pw_forum] About parallel problem on SGI Altix 3700 In-Reply-To: <942459.15212.qm@web92201.mail.cnh.yahoo.com> References: <942459.15212.qm@web92201.mail.cnh.yahoo.com> Message-ID: On Feb 14, 2008, at 3:46 , Chao wrote: > When I complied PWscf, some errors appeared when make all: > IPO Error: unresolved : mpi_conversion_fn_null_ > Referenced in ../Modules/parallel_include.o > Referenced in ../Modules/ptoolkit.o > Referenced in libiotk.a(iotk_error.o) > > Can anyone help me? short answer: no, it is a problem of your machine, not of quantum- espresso. It is either in IPO = interprocedural optimization or whatever it means (did you follow Lorenzo's advice? add -no-ipo to FFLAGS in make.sys, clean, recompile) or in your mpif90 script or in your mpi libraries. Paolo --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From ceresoli at sissa.it Fri Feb 15 16:19:39 2008 From: ceresoli at sissa.it (Davide Ceresoli) Date: Fri, 15 Feb 2008 16:19:39 +0100 (CET) Subject: [Pw_forum] GIPAW pseudopotentials In-Reply-To: <200802140909.13575.jzwanzig@dal.ca> References: <200802140909.13575.jzwanzig@dal.ca> Message-ID: On Thu, 14 Feb 2008, Josef Zwanziger wrote: > Hello, > I have been looking at the GIPAW modules in the 4.0cvs pre-release, and while > I can compile and run the example without problem, it's not clear how to make > pseudopotentials for GIPAW for other atoms. I notice that the given test > pseudopotentials, for silicon for example, include blocks of GIPAW specific > data, but I can't find any corresponding options in the ld1 input files or > for that matter the ld1 source code that corresponds to these options. Also > unclear is whether the GIPAW module as currently available works with > non-norm-conserving data, or currently only with norm-conserving data. Any > clarification would be helpful. > > thanks, > Joe > I've generated them with an ad-hoc version of ld1, in June 2007. Then, suddenly, the CVS version "diverged" and "conflicted" with my version. As soon as I'll some time, I'll try to merge my code, and put it in the current CVS version. 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 Skype: dceresoli Homepage: http://people.sissa.it/~ceresoli +----------------------------------------------------------+ From ari.p.seitsonen at iki.fi Sat Feb 16 11:04:52 2008 From: ari.p.seitsonen at iki.fi (Ari P Seitsonen) Date: Sat, 16 Feb 2008 11:04:52 +0100 (CET) Subject: [Pw_forum] GIPAW pseudopotentials In-Reply-To: <200802140909.13575.jzwanzig@dal.ca> References: <200802140909.13575.jzwanzig@dal.ca> Message-ID: Dear Joe, My humble answer to the question... The GIPAW module in the CVS might be somewhat outdated, as I haven't found the time to stay up-to-date recently due to the various changes in the CVS due to the incorporation of the PAW method (for self-consistent wave functions, ie. not to be confused with the GIPAW, that provides some properties due to magnetic response) into the main body of PWSCF/QE code. I'll try to do it still this month. A preliminary version of the whole GIPAW is found under http://www.impmc.jussieu.fr/~software/gipaw/qe-nmr-latest.tar.gz That site http://www.impmc.jussieu.fr/~software/gipaw/ provides a sub-minimal documentation on it use, a few GIPAW-compliant UPF files (the new version can read the data on reconstruction from the same UPF file with the pseudo potential itself). The tar file also contains the hacked version of the 'atomic' code to generate the pseudo potentials; I have been hesitating in checking that version into the CVS repository as the changes are not that clean (we introduced the generation of GIPAW projectors at arbitrary energies, ie. not only at eigenenergies - thus unbound states can be used as well). I'm sorry, all this needs to documented much better, this will happen soon (this month). Greetings, apsi On Thu, 14 Feb 2008, Josef Zwanziger wrote: > Hello, > I have been looking at the GIPAW modules in the 4.0cvs pre-release, and while > I can compile and run the example without problem, it's not clear how to make > pseudopotentials for GIPAW for other atoms. I notice that the given test > pseudopotentials, for silicon for example, include blocks of GIPAW specific > data, but I can't find any corresponding options in the ld1 input files or > for that matter the ld1 source code that corresponds to these options. Also > unclear is whether the GIPAW module as currently available works with > non-norm-conserving data, or currently only with norm-conserving data. Any > clarification would be helpful. > > thanks, > Joe > -- -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- Ari P Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/ IMPMC, CNRS & Universit? Pierre et Marie Curie Tel: +33-1-4427 7542, Fax: +33-1-4427 3785, GSM: +33-6-6736 3820 From jibiaoli at gmail.com Mon Feb 18 04:10:06 2008 From: jibiaoli at gmail.com (Clark Lee) Date: Mon, 18 Feb 2008 11:10:06 +0800 Subject: [Pw_forum] Adsorption at electrified metal surface Message-ID: Hi all, Is there any possibility to study adsorbates at electrified metal surfaces ( e.g. this is highly important in water-metal interface with applied potential) in QE package? -- Yours sincerely, Clark Lee State Key Lab of Corrosion and Protection (SKLCP) Institute of Metal Research (IMR) Chinese Academy of Sciences (CAS), China Phone: 024-23925323 Email: jibiaoli at imr.ac.cn or jibiaoli at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080218/9a5fc52c/attachment.htm From kddst2005 at yahoo.com.cn Mon Feb 18 06:10:43 2008 From: kddst2005 at yahoo.com.cn (Dongdong Kang) Date: Mon, 18 Feb 2008 13:10:43 +0800 (CST) Subject: [Pw_forum] structure optimization Message-ID: <3450.95308.qm@web92011.mail.cnb.yahoo.com> Dear all I'm a new user about QE. Recently I'm studying properties about some small clusters, I want to get the geometry structure of ground state using dynamic simulation annealing(DSA) strategy. Can QE implement the DSA strategy ? If yes, how can I implement this ? thanks Dongdong Kang Dongdong Kang National University of Defense Technology, P R China 410073, Hunan, Changsha --------------------------------- ??????????????????? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080218/1c548626/attachment.htm From r95222066 at ntu.edu.tw Mon Feb 18 10:31:47 2008 From: r95222066 at ntu.edu.tw (r95222066 at ntu.edu.tw) Date: Mon, 18 Feb 2008 17:31:47 +0800 Subject: [Pw_forum] hi, I wonder is the choice of pesudopotential cause the DOS(density of state) of 3d state region disappear? Message-ID: <20080218173147.9sbskwyi3o0000co@wmail1.cc.ntu.edu.tw> Dear all pwscf user, I have calculate the density of state on material which has V 3d states.But when I draw out to compare with the paper.I find the DOS of V 3d region disappear. Here is my pesudopotential below. ---------------------------------------------------------- V:Perdew-Burke-Ernzerhof (PBE) exch-corr nonlinear core-correction Vanderbilt ultrasoft O:Perdew-Burke-Ernzerhof (PBE) exch-corr Rabe Rappe Kaxiras Joannopoulos (ultrasoft) ----------------------------------------------------------- Or there may be another wrong place I misunderstood? PS:I had calculated nonmagnetic state and ferromagnetic state ,the DOS below Fermi level is well fittd. But the V 3d region up from the fermi level is disappeared. Thanks a lot Bong-Hong Chen,National Taiwan University,Taiwan From njuxuyuehua at gmail.com Mon Feb 18 14:08:42 2008 From: njuxuyuehua at gmail.com (xu yuehua) Date: Mon, 18 Feb 2008 21:08:42 +0800 Subject: [Pw_forum] about the feasibility of second order coefficient of Nonlinear Optical within PWSCF Message-ID: hi ,all here i am writing to enquire that is there a way to get the second order coefficient of Nonlinear Optical within PWSCF,directly or indirectly . if your could give me some help or suggestion ,i will be grateful. thank you in advance. -- Xu Yuehua physics Department of Nanjing university China -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080218/0ee97c8d/attachment.htm From giannozz at nest.sns.it Mon Feb 18 16:27:39 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 18 Feb 2008 16:27:39 +0100 Subject: [Pw_forum] Adsorption at electrified metal surface In-Reply-To: References: Message-ID: <47B9A3EB.3040101@nest.sns.it> Clark Lee wrote: > Is there any possibility to study adsorbates at electrified metal > surfaces (e.g. this is highly important in water-metal interface with > applied potential) in QE package? maybe if you explain what you mean by "electrified", you have better chances to get an answer P. -- Paolo Giannozzi, Democritos and University of Udine, Italy From giannozz at nest.sns.it Mon Feb 18 18:44:29 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 18 Feb 2008 18:44:29 +0100 Subject: [Pw_forum] hi, I wonder is the choice of pesudopotential cause the DOS(density of state) of 3d state region disappear? In-Reply-To: <20080218173147.9sbskwyi3o0000co@wmail1.cc.ntu.edu.tw> References: <20080218173147.9sbskwyi3o0000co@wmail1.cc.ntu.edu.tw> Message-ID: <47B9C3FD.8050007@nest.sns.it> r95222066 at ntu.edu.tw wrote: > I have calculate the density of state on material which has V 3d > states.But when I draw out to compare with the paper.I find the > DOS of V 3d region disappear. Here is my pesudopotential below. > ---------------------------------------------------------- > V:Perdew-Burke-Ernzerhof (PBE) exch-corr > nonlinear core-correction > Vanderbilt ultrasoft > O:Perdew-Burke-Ernzerhof (PBE) exch-corr > Rabe Rappe Kaxiras Joannopoulos (ultrasoft) > ----------------------------------------------------------- > Or there may be another wrong place I misunderstood? > > PS:I had calculated nonmagnetic state and ferromagnetic state ,the DOS > below Fermi level is well fittd. But the V 3d region up from the fermi > level is disappeared. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% error # 1 reading pw_forum message: not enough information provided to understand the problem stopping... %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% STOP 2 P. -- Paolo Giannozzi, Democritos and University of Udine, Italy From giannozz at nest.sns.it Mon Feb 18 23:02:08 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 18 Feb 2008 23:02:08 +0100 Subject: [Pw_forum] about the feasibility of second order coefficient of Nonlinear Optical within PWSCF In-Reply-To: References: Message-ID: <8495FFBE-218E-45B1-A21C-877977FD3816@nest.sns.it> On Feb 18, 2008, at 14:08 , xu yuehua wrote: > i am writing to enquire that is there a way to get the second order > coefficient of Nonlinear Optical within PWSCF, directly or indirectly. second-order nonlinear optical coefficients, i.e. \gamma in the following formula, where P is the induced polarization and E the electric field: P=\alpha E + \beta E^2 + \gamma E^3 ? this is the fourth-order derivative of the energy wrt electric field and can be calculated using the second-order response to an applied electric field. The latter is calculated by the phonon code for the Raman calculation. The code already calculates the nonlinear optical coefficient \beta (called "electro-optical coefficients": I do not think this is correct, by the way). Static calculations, though (\omega=0), and no ultrasoft PP. So all ingredients are there, but require some programming. It is something that I wanted to do (and still want to do: it is useful to estimate how nonlinear optical properties change in phtalocyanines when deposited on some surfaces and doped), but it requires an amount of patience that I am afraid I no longer have. Anybody interested? Paolo --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From jibiaoli at gmail.com Tue Feb 19 01:42:17 2008 From: jibiaoli at gmail.com (Clark Lee) Date: Tue, 19 Feb 2008 08:42:17 +0800 Subject: [Pw_forum] Adsorption at electrified metal surface In-Reply-To: <47B9A3EB.3040101@nest.sns.it> References: <47B9A3EB.3040101@nest.sns.it> Message-ID: On 2/18/08, Paolo Giannozzi wrote: > maybe if you explain what you mean by "electrified", you have > better chances to get an answer 'electrified' surface means a polarized surface with either positive or negative extra charges. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080219/9beca29c/attachment.htm From giannozz at nest.sns.it Tue Feb 19 11:48:22 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 19 Feb 2008 11:48:22 +0100 Subject: [Pw_forum] structure optimization In-Reply-To: <3450.95308.qm@web92011.mail.cnb.yahoo.com> References: <3450.95308.qm@web92011.mail.cnb.yahoo.com> Message-ID: <47BAB3F6.2090207@nest.sns.it> Dongdong Kang wrote: > I'm a new user about QE. Recently I'm studying properties about some > small clusters, I want to get the geometry structure of ground state > using dynamic simulation annealing(DSA) strategy. Can QE implement > the DSA strategy ? If yes, how can I implement this ? all needed ingredients are there (especially in the CP code that is especially designed for the simulated annealing approach): molecular dynamics and thermostats Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From njuxuyuehua at gmail.com Tue Feb 19 13:17:58 2008 From: njuxuyuehua at gmail.com (xu yuehua) Date: Tue, 19 Feb 2008 20:17:58 +0800 Subject: [Pw_forum] about the feasibility of second order coefficient of Nonlinear Optical within PWSCF In-Reply-To: <8495FFBE-218E-45B1-A21C-877977FD3816@nest.sns.it> References: <8495FFBE-218E-45B1-A21C-877977FD3816@nest.sns.it> Message-ID: thank you.it seems somewhat complex.hehe 2008/2/19, Paolo Giannozzi : > > > On Feb 18, 2008, at 14:08 , xu yuehua wrote: > > > i am writing to enquire that is there a way to get the second order > > coefficient of Nonlinear Optical within PWSCF, directly or indirectly. > > second-order nonlinear optical coefficients, i.e. \gamma in the > following formula, where P is the induced polarization and E the > electric field: P=\alpha E + \beta E^2 + \gamma E^3 ? this is > the fourth-order derivative of the energy wrt electric field and > can be calculated using the second-order response to an > applied electric field. The latter is calculated by the phonon > code for the Raman calculation. The code already calculates > the nonlinear optical coefficient \beta (called "electro-optical > coefficients": I do not think this is correct, by the way). Static > calculations, though (\omega=0), and no ultrasoft PP. So all > ingredients are there, but require some programming. It is > something that I wanted to do (and still want to do: it is useful > to estimate how nonlinear optical properties change in phtalocyanines > when deposited on some surfaces and doped), but it requires an > amount of patience that I am afraid I no longer have. Anybody > interested? > > Paolo > --- > Paolo Giannozzi, Dept of Physics, University of Udine > via delle Scienze 208, 33100 Udine, Italy > Phone +39-0432-558216, fax +39-0432-558222 > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Xu Yuehua physics Department of Nanjing university China -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080219/6cc575c5/attachment.htm From baroni at sissa.it Tue Feb 19 14:40:59 2008 From: baroni at sissa.it (Stefano Baroni) Date: Tue, 19 Feb 2008 14:40:59 +0100 Subject: [Pw_forum] about the feasibility of second order coefficient of Nonlinear Optical within PWSCF In-Reply-To: References: <8495FFBE-218E-45B1-A21C-877977FD3816@nest.sns.it> Message-ID: <47BADC6B.2010800@sissa.it> xu yuehua wrote: > thank you.it seems somewhat complex.hehe life is not an easy walk ... SB -- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html From lizhanfang at yahoo.com.cn Tue Feb 19 16:00:43 2008 From: lizhanfang at yahoo.com.cn (Xiaofang) Date: Tue, 19 Feb 2008 23:00:43 +0800 (CST) Subject: [Pw_forum] Hello, why the calculated polarization through Berry Phase is so large Message-ID: <421774.71540.qm@web15710.mail.cnb.yahoo.com> Dear all, I have calculated the polarization of KNbO3 using the Berry Phase. The atomic position and lattice constants of input are all same with the experimental values. The computed polarization is 1.1201929 C/m^2,but the experimental polarization is about 0.37 C/m^2. I even use the espressov.4.0cvs to calculate, but the result is similar. I have also tried to calculate the Born effective charge using ph.x, through the effective charge I get the polarization which is 0.44 C/m^2. I don?t know why there is so much difference between the two kinds of method and why the calculated polarization through Berry phase is so large. Any suggestion is appreciated. This is the Scf?s input &control calculation = 'scf' restart_mode = 'from_scratch' pseudo_dir = '/home/pseudo/' outdir = '/home/tmp/' / &system ibrav=6 celldm(1)=7.5532353215993782928038366689241 celldm(3)=1.016512384288216162121591193395 nat=5 ntyp=3 nbnd=20 ecutwfc=30.0 ecutrho =300.0 occupations = 'fixed' degauss=0.00 / &electrons diagonalization ='cg' conv_thr = 1e-12, mixing_beta=0.3, / ATOMIC_SPECIES K 39.0983 019-K-ca-sp-vgrp.uspp.UPF Nb 92.90638 041-Nb-ca-sp-vgrp.uspp.UPF O 15.9994 008-O-ca--vgrp.uspp.UPF ATOMIC_POSITIONS (crystal) K 0.5 0.5 0.518 Nb 0.000000000 0.000000000 0.0 O 0.500000000 0.00000000 0.04 O 0.00000000 0.500000000 0.04 O 0.00000000 0.00000000 0.544 K_POINTS {automatic} 6 6 3 1 1 1 This is the BP?s input &control calculation = 'nscf' pseudo_dir = '/home/pseudo/' outdir = '/home/tmp/' lberry = .true. gdir = 3 nppstr = 9 / &system ibrav=6 celldm(1)=7.5532353215993782928038366689241 celldm(3)=1.016512384288216162121591193395 nat=5 ntyp=3 nbnd=20 ecutwfc=30.0 ecutrho =300.0 occupations = 'fixed' degauss=0.00 / &electrons diagonalization ='cg' conv_thr = 1e-5 mixing_beta = 0.3 / ATOMIC_SPECIES K 39.0983 019-K-ca-sp-vgrp.uspp.UPF Nb 92.90638 041-Nb-ca-sp-vgrp.uspp.UPF O 15.9994 008-O-ca--vgrp. uspp.UPF ATOMIC_POSITIONS (crystal) K 0.5 0.5 0.518 Nb 0.000000000 0.000000000 0.0 O 0.500000000 0.00000000 0.04 O 0.00000000 0.500000000 0.04 O 0.00000000 0.00000000 0.544 K_POINTS {automatic} 6 6 9 1 1 1 Thanks in advance. Zhanfang Li Joint training student, Penn State University, USA --------------------------------- ??????????????????? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080219/12214f69/attachment.htm From chenhanghuipwscf at gmail.com Tue Feb 19 16:01:54 2008 From: chenhanghuipwscf at gmail.com (alan chen) Date: Tue, 19 Feb 2008 10:01:54 -0500 Subject: [Pw_forum] high energy bands Message-ID: <22ae3ca40802190701s7af6eafg5bbf5979155fa413@mail.gmail.com> Dear PWSCF users, Since I need to do an integral which involves atomic projected DOS from negative infinity to positive infinity, in principle I need to calculate all the bands. In practice I include many bands so that the integral hopefully converges well. My unit cell is composed of 82 atoms and I calculate 800 bands. I am just curious to ask whether those bands (their eigenvalues and eigenvectors) with very high energies (~15 eV above Fermi level) are reliable or not. Hanghui Departement of Physics, Yale -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080219/628ac279/attachment-0001.htm From stojic at sissa.it Tue Feb 19 16:31:07 2008 From: stojic at sissa.it (Natasa Stojic) Date: Tue, 19 Feb 2008 16:31:07 +0100 (CET) Subject: [Pw_forum] high energy bands In-Reply-To: <22ae3ca40802190701s7af6eafg5bbf5979155fa413@mail.gmail.com> References: <22ae3ca40802190701s7af6eafg5bbf5979155fa413@mail.gmail.com> Message-ID: Dear Hanghui, I would guess that at 15 eV above the Fermi level, you are really at the limit for good pseudopotentials, although chances are that it is not bad at all. To be sure, maybe you could compare the bands with some reliable all-electron calculation. If you find them unacceptable, you would have to create a new pseudopotential... Hope this helps, Natasa On Tue, 19 Feb 2008, alan chen wrote: > Dear PWSCF users, > Since I need to do an integral which involves atomic projected DOS from > negative infinity to positive infinity, in principle I need to calculate all > the bands. In practice I include many bands so that the integral hopefully > converges well. > My unit cell is composed of 82 atoms and I calculate 800 bands. I am > just curious to ask whether those bands (their eigenvalues and eigenvectors) > with very high energies (~15 eV above Fermi level) are reliable or not. > > Hanghui > Departement of Physics, Yale > From giannozz at nest.sns.it Tue Feb 19 16:43:54 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 19 Feb 2008 16:43:54 +0100 Subject: [Pw_forum] high energy bands In-Reply-To: <22ae3ca40802190701s7af6eafg5bbf5979155fa413@mail.gmail.com> References: <22ae3ca40802190701s7af6eafg5bbf5979155fa413@mail.gmail.com> Message-ID: <47BAF93A.2090902@nest.sns.it> alan chen wrote: > I am just curious to ask whether those bands (their eigenvalues and > eigenvectors) with very high energies (~15 eV above Fermi level) are > reliable or not. difficult to say. Pseudopotentials have a limited energy range of good transferability (but not so limited: a few Ry around the reference energy). They may have ghosts states at high energies (where nobody will notice except you). The higher the energy, the more your bands will be free-electron-like and the less the pseudopotential will be important, though. Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From chen_shao_hua197 at yahoo.com.tw Tue Feb 19 21:55:27 2008 From: chen_shao_hua197 at yahoo.com.tw (=?big5?q?=B3=AF=20=A4=D6=B5=D8?=) Date: Wed, 20 Feb 2008 04:55:27 +0800 (CST) Subject: [Pw_forum] negative el-ph coupling constant lambda Message-ID: <283162.35494.qm@web73015.mail.tp2.yahoo.com> Dear Eyvaz and forum's members, When I calculated the el-ph coupling constant \lambda of material.I find some negative value Electron-phonon coupling constant, lambda Broadening 0.0100 lambda 0.0008 dos_el 0.3719 Broadening 0.0200 lambda 0.0012 dos_el 0.3674 Broadening 0.0300 lambda 0.0021 dos_el 0.3565 Broadening 0.0400 lambda -0.0098 dos_el 0.3821 Broadening 0.0500 lambda -0.0303 dos_el 0.5231 Broadening 0.0600 lambda -0.0394 dos_el 0.7545 Broadening 0.0700 lambda -0.0408 dos_el 1.0409 Broadening 0.0800 lambda -0.0382 dos_el 1.3767 Broadening 0.0900 lambda -0.0333 dos_el 1.7603 Broadening 0.1000 lambda -0.0274 dos_el 2.1824 Also find some negative value in gam.line e.g. 24 0.0000 0.0000 0.0000 -0.0001 0.0000 -0.0001 ***** 0.0000 0.0000 But no negative frequency found in matdyn.freq I thought -\lambda comes from -\gamma (phonon linewidth), and -\gamma comes from -\omega (frequency) by its definitions. So I wonder why I got -\lambda and -\gamma (It should be >=0) even frequency \omega seems fine (no negative) ??? Should I just try check convergence of \gamma in dependent on cutoff energy and k-points sampling ? (from forum's "search archive", title " 3 questions about elph file of Calculation of electron-phonon interaction coefficients" suggestions ) Thanks in advance max PHD candidate Physics department, National Taiwan University,Taiwan ____________________________________________________________________________________________ 2G??????Yahoo??????2.0??????????? http://tw.mg0.mail.yahoo.com/dc/landing From eyvaz_isaev at yahoo.com Wed Feb 20 00:59:52 2008 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Tue, 19 Feb 2008 15:59:52 -0800 (PST) Subject: [Pw_forum] negative el-ph coupling constant lambda In-Reply-To: <283162.35494.qm@web73015.mail.tp2.yahoo.com> Message-ID: <302300.97458.qm@web60311.mail.yahoo.com> Dear Max, One can see that \lambda you got is quite small, just around zero. So, a reason might be numerical noise for such a kind small \lambda. \gamma depends also on production of two delta functions which couples two electronic states at the Fermi Surface and it converges slowly. I suggest you should perform tests with respect to number of k-points (more fine that you used), and higher cutoff energies. Bests, Eyvaz. P.S. Just today I have got \lambda for a metal, around 0.5, and everything went smoothly. --- ?? ???? wrote: > Dear Eyvaz and forum's members, > > When I calculated the el-ph coupling constant > \lambda > of material.I find some negative value > > Electron-phonon coupling constant, lambda > > Broadening 0.0100 lambda 0.0008 dos_el > 0.3719 > Broadening 0.0200 lambda 0.0012 dos_el > 0.3674 > Broadening 0.0300 lambda 0.0021 dos_el > 0.3565 > Broadening 0.0400 lambda -0.0098 dos_el > 0.3821 > Broadening 0.0500 lambda -0.0303 dos_el > 0.5231 > Broadening 0.0600 lambda -0.0394 dos_el > 0.7545 > Broadening 0.0700 lambda -0.0408 dos_el > 1.0409 > Broadening 0.0800 lambda -0.0382 dos_el > 1.3767 > Broadening 0.0900 lambda -0.0333 dos_el > 1.7603 > Broadening 0.1000 lambda -0.0274 dos_el > 2.1824 > > Also find some negative value in gam.line > e.g. > 24 0.0000 0.0000 0.0000 -0.0001 0.0000 > -0.0001 ***** 0.0000 0.0000 > > But no negative frequency found in matdyn.freq > > I thought -\lambda comes from -\gamma (phonon > linewidth), and -\gamma comes from -\omega > (frequency) by its definitions. > > So I wonder why I got -\lambda and -\gamma (It > should > be >=0) > even frequency \omega seems fine (no negative) ??? > > Should I just try check convergence of \gamma in > dependent on cutoff energy and k-points sampling ? > (from forum's "search archive", title " 3 questions > about elph file of Calculation of electron-phonon > interaction coefficients" suggestions ) > > Thanks in advance > > max > PHD candidate > Physics department, National Taiwan > University,Taiwan > > > > ____________________________________________________________________________________________ > 2G?i?H???h???HYahoo?_???q?l?H?c2.0???A???@???l?]???????I > http://tw.mg0.mail.yahoo.com/dc/landing > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, IFM, Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com ____________________________________________________________________________________ Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs From kddst2005 at yahoo.com.cn Wed Feb 20 06:33:53 2008 From: kddst2005 at yahoo.com.cn (Dongdong Kang) Date: Wed, 20 Feb 2008 13:33:53 +0800 (CST) Subject: [Pw_forum] structure optimization Message-ID: <586980.49175.qm@web92001.mail.cnb.yahoo.com> Dear Paolo Thank you for your answer, but I still have some questions. As you konw there are several options for ion_dynamics in input file of CP code , such as 'sd','cg','damp','verlet'. I want to know which option can implement dynamic simulation annealing. and is there option in input file to choose or change cooling rate with nose thermostat ? thanks Dongdong Kang National University of Defense Technology, P R China 410073, Hunan, Changsha --------------------------------- ??1G??????????? ????-????????? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080220/40557046/attachment.htm From akohlmey at cmm.chem.upenn.edu Wed Feb 20 13:49:25 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Wed, 20 Feb 2008 07:49:25 -0500 (EST) Subject: [Pw_forum] structure optimization In-Reply-To: <586980.49175.qm@web92001.mail.cnb.yahoo.com> References: <586980.49175.qm@web92001.mail.cnb.yahoo.com> Message-ID: On Wed, 20 Feb 2008, Dongdong Kang wrote: DK> Dear Paolo DK> DK> Thank you for your answer, but I still have some questions. As you DK> konw there are several options for ion_dynamics in input file of CP DK> code , such as 'sd','cg','damp','verlet'. I want to know which DK> option can implement dynamic simulation annealing. and is there DK> option in input file to choose or change cooling rate with nose DK> thermostat ? to get a specific annealing strategy, you will have to program it yourself. QE provides a framework, but it does not have every imaginable feature right from input parameters. if you are simply after geometry optimizations, both cg and damp will serve you well. one more general note, i've been noticing recently, that quite a few people tend to use the word "implement" when they actually mean "use". amongst people who write software, "implement" is usually interpreted as "program", i.e. writing source code and not using it. i hope that may clear some confusions. cheers, axel. DK> DK> thanks DK> DK> DK> Dongdong Kang DK> National University of Defense Technology, P R China DK> 410073, Hunan, Changsha DK> DK> DK> --------------------------------- DK> ????1G?????????????????????? DK> ????????-?????????????????? -- ======================================================================= 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 chen_shao_hua197 at yahoo.com.tw Wed Feb 20 17:11:05 2008 From: chen_shao_hua197 at yahoo.com.tw (=?big5?q?=B3=AF=20=A4=D6=B5=D8?=) Date: Thu, 21 Feb 2008 00:11:05 +0800 (CST) Subject: [Pw_forum] negative el-ph coupling constant lambda In-Reply-To: <302300.97458.qm@web60311.mail.yahoo.com> Message-ID: <304626.34726.qm@web73011.mail.tp2.yahoo.com> Dear Eyvaz, Thank you very much,I will try these two factors. max --- Eyvaz Isaev ?? > Dear Max, > > One can see that \lambda you got is quite small, > just > around zero. So, a reason might be numerical noise > for > such a kind small \lambda. \gamma depends also on > production of two delta functions which couples two > electronic states at the Fermi Surface and it > converges slowly. I suggest you should perform tests > with respect to number of k-points (more fine that > you > used), and higher cutoff energies. > > Bests, > Eyvaz. > P.S. Just today I have got \lambda for a metal, > around > 0.5, and everything went smoothly. > > --- ? ?? wrote: > > > Dear Eyvaz and forum's members, > > > > When I calculated the el-ph coupling constant > > \lambda > > of material.I find some negative value > > > > Electron-phonon coupling constant, lambda > > > > Broadening 0.0100 lambda 0.0008 dos_el > > 0.3719 > > Broadening 0.0200 lambda 0.0012 dos_el > > 0.3674 > > Broadening 0.0300 lambda 0.0021 dos_el > > 0.3565 > > Broadening 0.0400 lambda -0.0098 dos_el > > 0.3821 > > Broadening 0.0500 lambda -0.0303 dos_el > > 0.5231 > > Broadening 0.0600 lambda -0.0394 dos_el > > 0.7545 > > Broadening 0.0700 lambda -0.0408 dos_el > > 1.0409 > > Broadening 0.0800 lambda -0.0382 dos_el > > 1.3767 > > Broadening 0.0900 lambda -0.0333 dos_el > > 1.7603 > > Broadening 0.1000 lambda -0.0274 dos_el > > 2.1824 > > > > Also find some negative value in gam.line > > e.g. > > 24 0.0000 0.0000 0.0000 -0.0001 0.0000 > > -0.0001 ***** 0.0000 0.0000 > > > > But no negative frequency found in matdyn.freq > > > > I thought -\lambda comes from -\gamma (phonon > > linewidth), and -\gamma comes from -\omega > > (frequency) by its definitions. > > > > So I wonder why I got -\lambda and -\gamma (It > > should > > be >=0) > > even frequency \omega seems fine (no negative) ??? > > > > Should I just try check convergence of \gamma in > > dependent on cutoff energy and k-points sampling ? > > (from forum's "search archive", title " 3 > questions > > about elph file of Calculation of electron-phonon > > interaction coefficients" suggestions ) > > > > Thanks in advance > > > > max > > PHD candidate > > Physics department, National Taiwan > > University,Taiwan > > > > > > > > > ____________________________________________________________________________________________ > > > 2G??????Yahoo??????2.0??????????? > > http://tw.mg0.mail.yahoo.com/dc/landing > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > ------------------------------------------------------------------- > Prof. Eyvaz Isaev, > Theoretical Physics Department, Moscow State > Institute of Steel & Alloys, Russia, > IFM, Linkoping University, Sweden > Condensed Matter Theory Group, Uppsala University, > Sweden > Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com > > > > ____________________________________________________________________________________ > Never miss a thing. Make Yahoo your home page. > http://www.yahoo.com/r/hs > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ______________________________________________________________________________________________________ ?????????Yahoo!??????2.0????????????? http://tw.mg0.mail.yahoo.com/dc/landing From chenhanghuipwscf at gmail.com Thu Feb 21 00:12:41 2008 From: chenhanghuipwscf at gmail.com (alan chen) Date: Wed, 20 Feb 2008 18:12:41 -0500 Subject: [Pw_forum] two notations Message-ID: <22ae3ca40802201512t8270807if0bb04e91c026a2f@mail.gmail.com> Dear PWSCF users, I have a question about two notations used in the PWSCF output files. In the output file of atomic projected DOS, we have: 'state # 27: atom 5 (O ), wfc 2 (l=1 m= 3)'. Here 'l=1' means p-state. Physically we have 'm' quantum number which takes values of -1, 0 , 1. Is the correspondence of the notation of 'm' in PWSCF to real quantum number as follows: -1 => 1, 0 => 2, 1=> 3? Or you make some linear combinations of original quantum states and relabel them as m=1,2,3? Also in the output file of atomic projected DOS, we have: "psi = 0.974*[# 1]+0.014*[# 5]+0.007*[# 14]+0.002*[# 11]+0.001*[# 21]+0.001*[# 25]+''. Is 'psi' the wave function or the modulous square of the wave function? If 'psi' is the wave function, then why the coefficients in the decomposition are always positive? The coefficients are supposed to be the overlap of the Block state and one atomic orbital and are generally complex. Thank you very much. Hanghui Chen Department of Physics, Yale University -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080220/0352fdb2/attachment.htm From eyvaz_isaev at yahoo.com Thu Feb 21 03:42:14 2008 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Wed, 20 Feb 2008 18:42:14 -0800 (PST) Subject: [Pw_forum] two notations In-Reply-To: <22ae3ca40802201512t8270807if0bb04e91c026a2f@mail.gmail.com> Message-ID: <334176.80109.qm@web60315.mail.yahoo.com> Dear Alan, Here is a useful sketch from projwfc.f90 !In the non-collinear, spin-orbit case (i.e. if there is at least one ! fully relativistic pseudopotential) wavefunctions are projected ! onto eigenstates of the total angular-momentum. ! Projected DOS are written to file "filpdos".pdos_atm#N(X)_wfc#M(l_j), ! where N = atom number , X = atom symbol, M = wfc number, l=s,p,d,f ! and j is the value of the total angular momentum. ! In this case the format is ! E LDOS(E) PDOS_1(E) ... PDOS_2j+1(E) As concerns the second question the coefficients can be considered as the weight of appropriate atomic state to the energy band for a given k-point. Please read carefully your output file, projected atomic states are shown at the beginning of the file (**.pdos.out usually). In your case you can say that atomic state #1 (see atom type and symmetry) contributes mainly to a band you have chosen for a k-point (you have not shown here). Bests, Eyvaz. --- alan chen wrote: > Dear PWSCF users, > I have a question about two notations used in > the PWSCF output files. > In the output file of atomic projected DOS, we > have: 'state # 27: > atom 5 (O ), wfc 2 (l=1 m= 3)'. Here 'l=1' means > p-state. Physically we > have 'm' quantum number which takes values of -1, 0 > , 1. Is the > correspondence of the notation of 'm' in PWSCF to > real quantum number as > follows: -1 => 1, 0 => 2, 1=> 3? Or you make some > linear combinations of > original quantum states and relabel them as m=1,2,3? > Also in the output file of atomic projected > DOS, we have: "psi = > 0.974*[# 1]+0.014*[# 5]+0.007*[# 14]+0.002*[# > 11]+0.001*[# 21]+0.001*[# > 25]+''. Is 'psi' the wave function or the modulous > square of the wave > function? If 'psi' is the wave function, then why > the coefficients in the > decomposition are always positive? The coefficients > are supposed to be the > overlap of the Block state and one atomic orbital > and are generally complex. > Thank you very much. > > Hanghui Chen > Department of Physics, Yale University > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, IFM, Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com ____________________________________________________________________________________ Looking for last minute shopping deals? Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping From giannozz at nest.sns.it Thu Feb 21 10:24:02 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 21 Feb 2008 10:24:02 +0100 Subject: [Pw_forum] two notations In-Reply-To: <22ae3ca40802201512t8270807if0bb04e91c026a2f@mail.gmail.com> References: <22ae3ca40802201512t8270807if0bb04e91c026a2f@mail.gmail.com> Message-ID: <1CBE516B-B047-4C17-9C57-38987159D74D@nest.sns.it> On Feb 21, 2008, at 24:12 , alan chen wrote: > Is the correspondence of the notation of 'm' in PWSCF to real > quantum number as follows: -1 => 1, 0 => 2, 1=> 3? Or you make some > linear combinations of original quantum states and relabel them as > m=1,2,3? http://www.quantum-espresso.org/wiki/index.php/ Frequently_Asked_Questions#What_is_the_order_of_Ylm_components_in_projec ted_DOS_.2F_projection_of_atomic_wavefunctions.3F > > Also in the output file of atomic projected DOS, we have: > "psi = 0.974*[# 1]+0.014*[# 5]+0.007*[# 14]+0.002*[# 11]+0.001*[# > 21]+0.001*[# 25]+''. Is 'psi' the wave function or the modulous > square of the wave function? If 'psi' is the wave function, then > why the coefficients in the decomposition are always positive? The > coefficients are supposed to be the overlap of the Block state and > one atomic orbital and are generally complex. the printed numbers are the square moduli of the coefficients P. --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From giannozz at nest.sns.it Thu Feb 21 17:04:20 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 21 Feb 2008 17:04:20 +0100 Subject: [Pw_forum] structure optimization In-Reply-To: <586980.49175.qm@web92001.mail.cnb.yahoo.com> References: <586980.49175.qm@web92001.mail.cnb.yahoo.com> Message-ID: <31B84DD7-1691-47E5-A3E8-EEA7E3346312@nest.sns.it> On Feb 20, 2008, at 6:33 , Dongdong Kang wrote: > there are several options for ion_dynamics in input file of CP > code , such as 'sd','cg','damp','verlet'. I want to know which > option can implement dynamic > simulation annealing. none directly: as far as I know (i.e. very little), simulating annealing is achieved by performing an ordinary MD at high temperature, then cooling the system. The details "are left as an exercise to the reader". You choose what kind of MD you want to perform, for how long, at which temperature, with which thermostat, ... > and is there option in input file to choose or change cooling > rate with nose thermostat ? this is done by choosing the fictitious "thermostat mass" Paolo --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From marzari at MIT.EDU Thu Feb 21 20:05:58 2008 From: marzari at MIT.EDU (Nicola Marzari) Date: Thu, 21 Feb 2008 22:05:58 +0300 Subject: [Pw_forum] Fwd: Hello, why the calculated polarization through Berry Phase is so large In-Reply-To: <1758FF60-E8FB-4694-8937-F8AC762E777B@fisica.uniud.it> References: <421774.71540.qm@web15710.mail.cnb.yahoo.com> <1758FF60-E8FB-4694-8937-F8AC762E777B@fisica.uniud.it> Message-ID: <47BDCB96.1000800@mit.edu> First thing to check is the polarization quantum - P is not uniquely defined, but modulo 2e/omega R, where omega is the volume of the unit cell, and R any direct lattice vector. See the work of David Vanderbilt e.g. Ref 59 of http://www.physics.rutgers.edu/~dhv/pubs/pub_list.html), or the Rev. Mod. Phys. 1994 of Raffaele Resta. nicola PS: I would exercise some restraint on the significant digits in celldm. >> From: Xiaofang >> Date: February 19, 2008 4:00:43 GMT+01:00 >> To: pw_forum at pwscf.org >> Subject: [Pw_forum] Hello, why the calculated polarization through >> Berry Phase is so large >> Reply-To: PWSCF Forum >> >> Dear all, >> I have calculated the polarization of KNbO3 using the Berry Phase. The >> atomic position and lattice constants of input are all same with the >> experimental values. The computed polarization is 1.1201929 C/m^2,but >> the experimental polarization is about 0.37 C/m^2. I even use the >> espressov.4.0cvs to calculate, but the result is similar. I have also >> tried to calculate the Born effective charge using ph.x, through the >> effective charge I get the polarization which is 0.44 C/m^2. I don?t >> know why there is so much difference between the two kinds of method >> and why the calculated polarization through Berry phase is so large. >> Any suggestion is appreciated. >> This is the Scf?s input >> &control >> calculation = 'scf' >> restart_mode = 'from_scratch' >> pseudo_dir = '/home/pseudo/' >> outdir = '/home/tmp/' >> / >> &system >> ibrav=6 >> celldm(1)=7.5532353215993782928038366689241 >> celldm(3)=1.016512384288216162121591193395 >> nat=5 >> ntyp=3 >> nbnd=20 >> ecutwfc=30.0 >> ecutrho =300.0 >> occupations = 'fixed' >> degauss=0.00 >> / >> &electrons >> diagonalization ='cg' >> conv_thr = 1e-12, >> mixing_beta=0.3, >> / >> ATOMIC_SPECIES >> K 39.0983 019-K-ca-sp-vgrp.uspp.UPF >> Nb 92.90638 041-Nb-ca-sp-vgrp.uspp.UPF >> O 15.9994 008-O-ca--vgrp.uspp.UPF >> ATOMIC_POSITIONS (crystal) >> K 0.5 0.5 0.518 >> Nb 0.000000000 0.000000000 0.0 >> O 0.500000000 0.00000000 0.04 >> O 0.00000000 0.500000000 0.04 >> O 0.00000000 0.00000000 0.544 >> >> K_POINTS {automatic} >> 6 6 3 1 1 1 >> >> This is the BP?s input >> &control >> calculation = 'nscf' >> pseudo_dir = '/home/pseudo/' >> outdir = '/home/tmp/' >> lberry = .true. >> gdir = 3 >> nppstr = 9 >> / >> &system >> ibrav=6 >> celldm(1)=7.5532353215993782928038366689241 >> celldm(3)=1.016512384288216162121591193395 >> nat=5 >> ntyp=3 >> nbnd=20 >> ecutwfc=30.0 >> ecutrho =300.0 >> occupations = 'fixed' >> degauss=0.00 >> / >> &electrons >> diagonalization ='cg' >> conv_thr = 1e-5 >> mixing_beta = 0.3 >> / >> ATOMIC_SPECIES >> K 39.0983 019-K-ca-sp-vgrp.uspp.UPF >> Nb 92.90638 041-Nb-ca-sp-vgrp.uspp.UPF >> O 15.9994 008-O-ca--vgrp. uspp.UPF >> ATOMIC_POSITIONS (crystal) >> K 0.5 0.5 0.518 >> Nb 0.000000000 0.000000000 0.0 >> O 0.500000000 0.00000000 0.04 >> O 0.00000000 0.500000000 0.04 >> O 0.00000000 0.00000000 0.544 >> >> K_POINTS {automatic} >> 6 6 9 1 1 1 >> >> Thanks in advance. >> Zhanfang Li >> Joint training student, Penn State University, USA >> >> ??????????????????? >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum > > --- > Paolo Giannozzi, Dept of Physics, University of Udine > via delle Scienze 208, 33100 Udine, Italy > Phone +39-0432-558216, fax +39-0432-558222 > > > -- --------------------------------------------------------------------- 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 fornari at MIT.EDU Thu Feb 21 22:07:32 2008 From: fornari at MIT.EDU (Marco Fornari) Date: Thu, 21 Feb 2008 16:07:32 -0500 Subject: [Pw_forum] Fwd: Hello, why the calculated polarization through Berry Phase is so large In-Reply-To: <47BDCB96.1000800@mit.edu> References: <421774.71540.qm@web15710.mail.cnb.yahoo.com> <1758FF60-E8FB-4694-8937-F8AC762E777B@fisica.uniud.it> <47BDCB96.1000800@mit.edu> Message-ID: <47BDE814.6090607@mit.edu> Nic, sta diventando molto pressante la necessita' di studenti per il nuovo Ph.D. C'e' modo di pescare qualche etiope in tempi brevi? Se c'e' batti un colpo cosi' ti mando un po' di docs. Quando fate il workshop espresso? Mille grazie. Marco -- Dept. of Physics, Central Michigan University On leave at MIT-DMSE 13-4006 PH: 617-253-8701 -> http://www.phy.cmich.edu/people/fornari From chenhanghuipwscf at gmail.com Thu Feb 21 23:55:21 2008 From: chenhanghuipwscf at gmail.com (alan chen) Date: Thu, 21 Feb 2008 17:55:21 -0500 Subject: [Pw_forum] atomic projected DOS Message-ID: <22ae3ca40802211455y7593c958o474643b7afa78fef@mail.gmail.com> Dear PWSCF users, I have two questions about atomic projected DOS. In the subroutine projwfc.f90 there is a comment: "PDOS_m(E) = projected DOS on atomic wfc with component m". Is this "m" the quantum number of z-component of angular moment L? If it is, this 'm' is labeled from 1 to 2*L-1 instead of the standard '"-L, -L+1...L-1, L". Is there a one-to-one correspondence? Furthermore, what is the difference bwtween if "lsym" is set to .true. and .false.? I found that the coefficients in the output files are completely different? What does it really mean by 'symmetrize projection'? Thank you very much. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080221/8fdc76d8/attachment.htm From paulatto at sissa.it Fri Feb 22 09:29:54 2008 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Fri, 22 Feb 2008 09:29:54 +0100 (CET) Subject: [Pw_forum] atomic projected DOS In-Reply-To: <22ae3ca40802211455y7593c958o474643b7afa78fef@mail.gmail.com> References: <22ae3ca40802211455y7593c958o474643b7afa78fef@mail.gmail.com> Message-ID: <3791.147.122.5.182.1203668994.squirrel@webmail.sissa.it> On Gio, 21 Febbraio 2008 11:55 pm, alan chen wrote: > Dear PWSCF users, > I have two questions about atomic projected DOS. In the subroutine > projwfc.f90 there is a comment: "PDOS_m(E) = projected DOS on atomic wfc > with component m". Is this "m" the quantum number of z-component of > angular > moment L? If it is, this 'm' is labeled from 1 to 2*L-1 instead of the > standard '"-L, -L+1...L-1, L". Is there a one-to-one correspondence? Your question was answered just yesterday in this forum.. have a look here: > Furthermore, what is the difference bwtween if "lsym" is set to .true. > and .false.? I found that the coefficients in the output files are > completely different? What does it really mean by 'symmetrize projection'? > Thank you very much. I'm not 100% sure on this, I think that the projections are forced to have the same symmetry as the system, e.g. in a simple cubic system with a lonely atom px, py and pz components have to be equal. Bye, LP -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) +39 040 3787 511 http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From abotello at ipicyt.edu.mx Fri Feb 22 16:57:11 2008 From: abotello at ipicyt.edu.mx (Andres Botello) Date: Fri, 22 Feb 2008 09:57:11 -0600 Subject: [Pw_forum] Zn PP problem In-Reply-To: <251241210802210742n3c78a030w1d194ae45efd4f7e@mail.gmail.com> References: <251241210802210742n3c78a030w1d194ae45efd4f7e@mail.gmail.com> Message-ID: <251241210802220757k52b700b2j706c3d5460688e1@mail.gmail.com> Dear all, I'm trying to generate a NC pseudopotential for Zinc as described in *PRB 62 8802 (2000) *using ld1.x. However, I get a "too many nodes" message, when there are not nodes. I've read in previous posts that changin pseudotype to 1 worked for Ti pp generation, however, in my case, I get the very same error. Any help is appreciated. Regards, Andr?s. This is my input: &input title='Zn', zed=30., rel=1, iswitch=3, config = '[Ar] 3d10 4s1.75 4p0.25' lsd = 0, latt = 0, dft='revPBE', / &inputp pseudotype=2, lloc=1, file_pseudopw='Zn.rel.revPBE.UPF', / 3 3D 3 2 10.00 0.00 2.00 2.00 4S 1 0 1.75 0.00 1.10 1.10 4P 2 1 0.25 0.00 1.10 1.10 This is the message I get: --------------------- Generating NC pseudopotential --------------------- Generating local pot.: lloc=1, matching radius rcloc = 1.1000 Wfc 3D rcut= 2.000 Estimated cut-off energy= 58.77 Ry This function has 0 nodes for 0 < r < 2.011 from compute phi : info # -1 negative determinant 2 ld= 3.629293 f2ae 1.223718 faenor 0.031202 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from gener_pseudo : error # 1 too many nodes %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... -- Andr?s Rafael Botello M?ndez PhD Student Advanced Materials Department IPICyT, Mexico abotello at ipicyt.edu.mx -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080222/55d3653d/attachment.htm From nafiserb at gmail.com Sat Feb 23 05:42:33 2008 From: nafiserb at gmail.com (nafise rezaei) Date: Sat, 23 Feb 2008 08:12:33 +0330 Subject: [Pw_forum] wave-function In-Reply-To: <785b26a90802220827qf2ebc89v92b6e292382fd91f@mail.gmail.com> References: <785b26a90802220827qf2ebc89v92b6e292382fd91f@mail.gmail.com> Message-ID: <785b26a90802222042h6bf59bc4h5fd1035d434f816b@mail.gmail.com> : > > Dear all, > > I want to extract Kohan-Sham wave function. I know that the > wave-function is in general a complex (=real,imaginary) object. It is > clear that charge density can be visualized with pp.x, chden.x and > some softwares. > > However, I want to extract and to plot parts of real and imaginary > wave function. > Do you know any program to generate wave function? If there is no > program, I can modify what program to write it? > > > Thank you in advance, > Best regards > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080223/d5cddd5a/attachment.htm From paulatto at sissa.it Sat Feb 23 12:48:15 2008 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Sat, 23 Feb 2008 12:48:15 +0100 (CET) Subject: [Pw_forum] wave-function In-Reply-To: <785b26a90802222042h6bf59bc4h5fd1035d434f816b@mail.gmail.com> References: <785b26a90802220827qf2ebc89v92b6e292382fd91f@mail.gmail.com> <785b26a90802222042h6bf59bc4h5fd1035d434f816b@mail.gmail.com> Message-ID: <20816.82.50.175.59.1203767295.squirrel@webmail.sissa.it> On Sab, 23 Febbraio 2008 5:42 am, nafise rezaei wrote: >> However, I want to extract and to plot parts of real and imaginary >> wave function. >> Do you know any program to generate wave function? If there is no >> program, I can modify what program to write it? The postprocessing utility pp.x can extract the square modulus of a wavefunction (you choose it by specifying kpoint and band index), plotting real and imaginary parts is not implemented, but you may be able to do it youself. On the other hand, if nobody have felt the need to implement it in the last not so few years there could be a reason. If you tell us what you want to do, instead of how you want to do it, you will probably get some more useful advice. Regards, LP -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) +39 040 3787 511 http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From nafiserb at gmail.com Sat Feb 23 20:36:45 2008 From: nafiserb at gmail.com (nafise rezaei) Date: Sat, 23 Feb 2008 23:06:45 +0330 Subject: [Pw_forum] wave-function Message-ID: <785b26a90802231136x46dc4fbaod8ed77cd701992d1@mail.gmail.com> Hi I want , a program that generates wave function data on spatial grids for a given set of K-points and electronic bands not square modulus of a wavefunction . i.e I want to have periodic part of Bloch states represented on a grid. Which program generate and write it? Thank you in advance, Regards From cao at qtp.ufl.edu Sat Feb 23 20:50:08 2008 From: cao at qtp.ufl.edu (Chao Cao) Date: Sat, 23 Feb 2008 14:50:08 -0500 Subject: [Pw_forum] PAW in 4.0 CVS Message-ID: <47C078F0.1000905@qtp.ufl.edu> Hi all, It's really nice to see PAW implemented in 4.0 CVS. Thanks for your hard work! However, there are not many PAW pseudopotentials to test the feature in the package. Is there any upf tools to convert abinit PAW pseudopotentials, or any documents on the procedure to generate them using ld1.x? Any suggestions are welcome~! Best, Chao Cao -- Chao Cao Quantum Theory Project and Department of Physics University of Florida, Gainesville, FL 32611 U.S.A. From hushujun at mail.sdu.edu.cn Sun Feb 24 01:50:36 2008 From: hushujun at mail.sdu.edu.cn (=?gb2312?B?uvrK9778IFNodS1qdW4gSHU=?=) Date: Sun, 24 Feb 2008 08:50:36 +0800 Subject: [Pw_forum] the unit of charge or spin density (the isovalue) Message-ID: <403814236.27068@mail.sdu.edu.cn> Dear all, Anyone who can tell me the unit of charge or spin density (the isovalue) in the xsf file? Best wishes. Shujun Hu ------------------------------ Shujun Hu e-mail: hushujun at 163.com Shandong university Phone: +86/0531-88375097 Jinan, Shandong Province, China, 250100 From paulatto at sissa.it Sun Feb 24 11:14:32 2008 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Sun, 24 Feb 2008 11:14:32 +0100 (CET) Subject: [Pw_forum] the unit of charge or spin density (the isovalue) In-Reply-To: <403814236.27068@mail.sdu.edu.cn> References: <403814236.27068@mail.sdu.edu.cn> Message-ID: <28991.82.48.223.172.1203848072.squirrel@webmail.sissa.it> On Dom, 24 Febbraio 2008 1:50 am, ?????? Shu-jun Hu wrote: > > Dear all, > Anyone who can tell me the unit of charge or spin density (the isovalue) > in the > xsf file? The integral of the charge density over the unit cell should be the number of electrons, so its unit have to be 1 electron charge over the volume of the unit cell. Keep in mind that the unit of length is Angstrom. Spin polarization is just rho(up)-rho(down). Please note that xsf is not a format meant to be used for further postprocessing as the data is written with only 5 significant digits. Regards, LP -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) +39 040 3787 511 http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From paulatto at sissa.it Sun Feb 24 11:59:15 2008 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Sun, 24 Feb 2008 11:59:15 +0100 (CET) Subject: [Pw_forum] PAW in 4.0 CVS In-Reply-To: <47C078F0.1000905@qtp.ufl.edu> References: <47C078F0.1000905@qtp.ufl.edu> Message-ID: <29527.82.48.223.172.1203850755.squirrel@webmail.sissa.it> On Sab, 23 Febbraio 2008 8:50 pm, Chao Cao wrote: > Hi all, > It's really nice to see PAW implemented in 4.0 CVS. Thanks for your > hard work! However, there are not many PAW pseudopotentials to test the > feature in the package. Is there any upf tools to convert abinit PAW > pseudopotentials, or any documents on the procedure to generate them > using ld1.x? Any suggestions are welcome~! Dear Chao, its nice to see your interested in the PAW datasets; I'm working to generate some of them on time for the release of QE 4.0, but it takes time to test their quality. In the last few days I've been importing ld1.x documentation to the QE wiki (http://quantum-espresso.org/wiki/index.php/ld1.x), up to now it contains only a subset of what you can find in the atomic_doc/ directory, but I plan to add some sort of mini how-to during the next week. If you are rushing to generate some PAW dataset I can send you a few examples right now, just drop me an email. If after generation and testing you will be willing to contribute them back to the QE pseudopotentials library you will be warmly welcome! Goodbye, LP -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) +39 040 3787 511 http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From sagarambavale at yahoo.co.in Mon Feb 25 06:12:37 2008 From: sagarambavale at yahoo.co.in (ambavale sagar) Date: Mon, 25 Feb 2008 10:42:37 +0530 (IST) Subject: [Pw_forum] segmentation fault Message-ID: <869386.66027.qm@web94610.mail.in2.yahoo.com> An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080225/2d12bb79/attachment.htm From wangjunjie1981_0 at 163.com Mon Feb 25 07:37:48 2008 From: wangjunjie1981_0 at 163.com (wangjunjie1981_0) Date: Mon, 25 Feb 2008 14:37:48 +0800 (CST) Subject: [Pw_forum] WARNING: eigenvalues not converged in pw.x Message-ID: <21216119.280611203921468783.JavaMail.coremail@bj163app89.163.com> Dear pwscf users Recently i have such kind warnings. As people explained in the forum, this warning can be ignored, if it is not present in the last steps of self-consistency. But I got these warnings during NSCF calculations. Should I ignore them? Output file total cpu time spent up to now is 10348.76 secs Band Structure Calculation Davidson diagonalization (with overlap) WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 2 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 1 eigenvalues not converged ................. Thanks in advance, Junjie Wang -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080225/63cf4ea8/attachment-0001.htm From lanhaiping at gmail.com Mon Feb 25 08:18:59 2008 From: lanhaiping at gmail.com (lan haiping) Date: Mon, 25 Feb 2008 08:18:59 +0100 Subject: [Pw_forum] WARNING: eigenvalues not converged in pw.x In-Reply-To: <21216119.280611203921468783.JavaMail.coremail@bj163app89.163.com> References: <21216119.280611203921468783.JavaMail.coremail@bj163app89.163.com> Message-ID: why not just search the previous posts about this problem ? there are so many reports before. 2008/2/25 wangjunjie1981_0 : > > Dear pwscf users > Recently i have such kind warnings. As people explained in the forum, > this warning can be ignored, if it is not present in the last steps of > self-consistency. But I got these warnings during NSCF calculations. Should > I ignore them? > > Output file > > total cpu time spent up to now is 10348.76 secs > > Band Structure Calculation > > Davidson diagonalization (with overlap) > > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 2 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > ................. > > Thanks in advance, > Junjie Wang > > > > > > ------------------------------ > ? ? ? ? 100 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- Hai-Ping Lan Department of Electronics , Peking University , Bejing, 100871 lanhaiping at gmail.com, hplan at pku.edu.cn -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080225/1460fd3b/attachment.htm From yanyan1110_2 at yahoo.cn Mon Feb 25 09:14:50 2008 From: yanyan1110_2 at yahoo.cn (Yanyan) Date: Mon, 25 Feb 2008 16:14:50 +0800 (CST) Subject: [Pw_forum] WARNING: eigenvalues not converged in pw.x In-Reply-To: Message-ID: <577962.29463.qm@web92203.mail.cnh.yahoo.com> Thank you for your reply, i have searched this question in pw_forum, but i can not find the corresponding answers about it, so i submit the question to the forum. Junjie Wang why not just search the previous posts about this problem ? there are so many reports before. 2008/2/25 wangjunjie1981_0 : > > Dear pwscf users > Recently i have such kind warnings. As people explained in the forum, > this warning can be ignored, if it is not present in the last steps of > self-consistency. But I got these warnings during NSCF calculations. Should > I ignore them? > > Output file > > total cpu time spent up to now is 10348.76 secs > > Band Structure Calculation > > Davidson diagonalization (with overlap) > > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 2 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > WARNING: 1 eigenvalues not converged > ................. > > Thanks in advance, > Junjie Wang > > > > > > ------------------------------ > ? ? ? ? 100 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? > > _______________________________________________ > 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: http://www.democritos.it/pipermail/pw_forum/attachments/20080225/64eb6c39/attachment.htm From stenuit at sissa.it Mon Feb 25 09:47:03 2008 From: stenuit at sissa.it (Geoffrey Stenuit) Date: Mon, 25 Feb 2008 09:47:03 +0100 Subject: [Pw_forum] WARNING: eigenvalues not converged in pw.x In-Reply-To: <577962.29463.qm@web92203.mail.cnh.yahoo.com> References: <577962.29463.qm@web92203.mail.cnh.yahoo.com> Message-ID: <47C28087.8010403@sissa.it> Dear Junjie Wang, Please have a look in the forum list from November 2007, particulary the reply from Paolo Giannozzi: > From giannozz at nest.sns.it Fri Nov 30 13:21:05 2007 > From: giannozz at nest.sns.it (Paolo Giannozzi) > Date: Fri, 30 Nov 2007 13:21:05 +0100 > Subject: [Pw_forum] Warning in the 'nscf' calculation > In-Reply-To: <22ae3ca40711291520rfd5cb65kc0e5b07a061fe056 at mail.gmail.com> > References: <22ae3ca40711291520rfd5cb65kc0e5b07a061fe056 at mail.gmail.com> > Message-ID: > > > On Nov 30, 2007, at 24:20 , alan chen wrote: > > > WARNING: 3 eigenvalues not converged [...] > > The band structure is finally calculated at the end of output file > > but shall I believe in these E_{nk} values? > > the code counts unconverged eigenvalues and makes > another tour of iterative diagonalization, until the desired > number of eigenvalues (all of them for non-scf calculations) > have converged: see function test_exit_cond. So yes, you > should believe these values, or more exactly, you shouldn't > be any more suspicious that for all other numbers. > > > How can I get rid of these warning messages? > > try the "see no evil" approach: comment the line yielding the > warning message. The alternative is the "computer geek" > approach: find a better algorithm. > > Paolo > --- > Paolo Giannozzi, Dept of Physics, University of Udine > via delle Scienze 208, 33100 Udine, Italy > Phone +39-0432-558216, fax +39-0432-558222 Hope it helps, Joe Yanyan wrote: > > Thank you for your reply, i have searched this question in > pw_forum, but i can not find the corresponding answers about it, > so i submit the question to the forum. > Junjie Wang > > > why not just search the previous posts about this problem ? > there are so many reports before. > > > 2008/2/25 wangjunjie1981_0 : > > > > > Dear pwscf users > > Recently i have such kind warnings. As people explained in the > forum, > > this warning can be ignored, if it is not present in the last > steps of > > self-consistency. But I got these warnings during NSCF > calculations. Should > > I ignore them? > > > > Output file > > > > total cpu time spent up to now is 10348.76 secs > > > > Band Structure Calculation > > > > Davidson diagonalization (with overlap) > > > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 2 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > WARNING: 1 eigenvalues not converged > > ................. > > > > Thanks in advance, > > Junjie Wang > > > > > > > > > > > > ------------------------------ > > ? ? ? ? 100 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? > > > > _______________________________________________ > > 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 > -- -------------------------------------------------------------------- Geoffrey Stenuit Theory at Elettra Group CNR-INFM DEMOCRITOS National Simulation Center c/o Sincrotrone Trieste - SS14, Km 163,5 Basovizza, I-34012 TRIESTE email: stenuit at sissa.it tel: +39 040 375-8713 fax: - From helen at fh.huji.ac.il Mon Feb 25 09:56:49 2008 From: helen at fh.huji.ac.il (Helen) Date: Mon, 25 Feb 2008 10:56:49 +0200 Subject: [Pw_forum] Ti pseudopotential Message-ID: <000601c8778c$5b5ac020$cd604084@fh.huji.ac.il> Hello, I have a question about example 10, which does calculations on PbTiO_3. The number of valence electrons specified is 25, nbnd=25, corresponding to a Ti cation (4s2 3d1) and the Ti pseudopotential , Ti.vdb.UPF also states that it is generated for the cation, Wavefunctions nl l occ 3S 0 2.00 3P 1 6.00 4S 0 2.00 3D 2 1.00 Giving 11 valence electrons corresponding to 4s2 3d1. THough in the same file it states 12.00000000000 Z valence Is the calculation really being carried out for PbTiO_3 with a titanium cation and the rest neutral atoms? IF so why? Or is there something I am misunderstanding about how the pseudopotential and run file are made? Thank you, Dr. Helen Eisenberg, Fritz Haber Center, Hebrew University -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080225/0ceba2e5/attachment.htm From ylli at theory.issp.ac.cn Mon Feb 25 09:59:05 2008 From: ylli at theory.issp.ac.cn (Li yan-ling) Date: Mon, 25 Feb 2008 16:59:05 +0800 (CST) Subject: [Pw_forum] phonon calculation. Message-ID: <46284.137.189.4.2.1203929945.squirrel@210.73.16.4> Dear all, Now I try to calculate the superconducting transition temperature of materials by using pwscf code. By performing ph.x+q2r.x+matdyn.x, one can get gamma.lines, dyna2F, phonon.dos,and lamda. I have two questions. 1. I don't know how to plot the a2F(omega) vs. omega function curve? 2. I check the output file 'dyna2F'. It consists of all modes's a2F(omega), I think. But I can't understand its meaning. In general, a2F vs. omega is a 2D ( two dimentional) curve. I want to know how to transform it to the 2D form in order to ploting the a2F(omega) as the function of omega curve? Any answers or suggestions are welcome. Thanks a lot. Yanling Li From akohlmey at cmm.chem.upenn.edu Mon Feb 25 10:41:02 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Mon, 25 Feb 2008 04:41:02 -0500 Subject: [Pw_forum] Ti pseudopotential In-Reply-To: <000601c8778c$5b5ac020$cd604084@fh.huji.ac.il> References: <000601c8778c$5b5ac020$cd604084@fh.huji.ac.il> Message-ID: <7b6913e90802250141x3e09da4cnff803b526186a75e@mail.gmail.com> On Mon, Feb 25, 2008 at 3:56 AM, Helen wrote: > > > Hello, helen, > I have a question about example 10, which does calculations on PbTiO_3. The > number of valence electrons specified is 25, nbnd=25, corresponding to a Ti > cation (4s2 3d1) and the Ti pseudopotential , Ti.vdb.UPF also states that it > is generated for the cation, > Wavefunctions nl l occ > 3S 0 2.00 > 3P 1 6.00 > 4S 0 2.00 > 3D 2 1.00 > Giving 11 valence electrons corresponding to 4s2 3d1. > THough in the same file it states > 12.00000000000 Z valence > Is the calculation really being carried out for PbTiO_3 with a titanium it is quite common for transition metals to create pseudopotentials from a (slightly) oxidized reference configuration to have improved transferability. the use of resulting pseudopotential is independent of the reference configuration, i.e. zv is the core charge in the calculations. cheers, axel. > cation and the rest neutral atoms? IF so why? Or is there something I am > misunderstanding about how the pseudopotential and run file are made? > Thank you, > Dr. Helen Eisenberg, > Fritz Haber Center, > Hebrew University > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From eyvaz_isaev at yahoo.com Mon Feb 25 15:57:02 2008 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Mon, 25 Feb 2008 06:57:02 -0800 (PST) Subject: [Pw_forum] phonon calculation. In-Reply-To: <46284.137.189.4.2.1203929945.squirrel@210.73.16.4> Message-ID: <247116.77199.qm@web60323.mail.yahoo.com> Dear Yanling, Please respect the Netiquette accepted in this forum providing your affiliation. --- Li yan-ling wrote: > I have two questions. > 1. I don't know how to plot the a2F(omega) vs. omega > function curve? Use gnuplot or xmgrace to plot it. How to do it ask your experienced friend. It is not a subject for this forum. See www.gnuplot.info for more information. If you have done everything correctly you have a2F.dos.1 (2,3, ..., 10, by default) files. As you can read from header lines, the first column is frequency in Ry (find how to convert it to cm^{-1} or THz), then a2F_total. Next three columns are mode projected a2F (1 atom => 3 modes => 3 columns). Numbers from 1 to 10 specify smearing widths used in \gamma_{nk} calculations (from 0.01 to 0.10 by default). So, you can see how a2F is affected by this smearing parameter through a2F.dos.1 .... a2F.dos.10 > 2. I check the output file 'dyna2F'. It consists of > all modes's a2F(omega), I think. But I can't >understand its meaning. In general, a2F > vs. omega is a 2D ( two dimentional) curve. I want > to know how to transform it to the 2D form in order >to ploting the a2F(omega) as the function of omega >curve? > Any answers or suggestions are welcome. Actually, I do not think that you need to plot dyna2F file. Instead use gam.lines to plot \gamma_{nk} along high symmetry directions (numbers in the first column are indexes for q-vectors along these directions) you have chosen and in dependent on smearing width. You can plot them, again, using gnuplot or xmgrace (http://plasma-gate.weizmann.ac.il/Grace/). Bests, Eyvaz. --- Li yan-ling wrote: > I have two questions. > 1. I don't know how to plot the a2F(omega) vs. omega > function curve? Use gnuplot or xmgrace to plot it. How to do it ask your experienced friend. It is not a subject for this forum. See www.gnuplot.info for more information. If you have done everything correctly you have a2F.dos.1 (2,3, ..., 10, by default) files. As you can read from header lines, the first column is frequency in Ry (find how to convert it to cm^{-1} or THz), then a2F_total. Next three columns are mode projected a2F (1 atom => 3 modes => 3 columns). Numbers from 1 to 10 specify smearing widths used in \gamma_{nk} calculations (from 0.01 to 0.10 by default). So, you can see how a2F is affected by this smearing parameter through a2F.dos.1 .... a2F.dos.10 > 2. I check the output file 'dyna2F'. It consists of > all modes's a2F(omega), I think. But I can't >understand its meaning. In general, a2F > vs. omega is a 2D ( two dimentional) curve. I want > to know how to transform it to the 2D form in order >to ploting the a2F(omega) as the function of omega >curve? > Any answers or suggestions are welcome. Actually, I do not think that you need to plot dyna2F file. Instead use gam.lines to plot \gamma_{nk} along high symmetry directions you have chosen and in dependent on smearing width. You can plot them, again, using gnuplot or xmgrace (www. > Thanks a lot. > > Yanling Li > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, IFM, Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com ____________________________________________________________________________________ Looking for last minute shopping deals? Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping From sagars_daya at yahoo.co.in Mon Feb 25 17:39:27 2008 From: sagars_daya at yahoo.co.in (Daya sagar) Date: Mon, 25 Feb 2008 16:39:27 +0000 (GMT) Subject: [Pw_forum] Please unsubscribe me Message-ID: <318944.65401.qm@web8319.mail.in.yahoo.com> Hi Please unsubscribe me from this yahoo groups. Daya --------------------------------- Download prohibited? No problem. CHAT from any browser, without download. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080225/a82d2236/attachment.htm From akohlmey at cmm.chem.upenn.edu Mon Feb 25 17:47:50 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Mon, 25 Feb 2008 11:47:50 -0500 (EST) Subject: [Pw_forum] Please unsubscribe me In-Reply-To: <318944.65401.qm@web8319.mail.in.yahoo.com> References: <318944.65401.qm@web8319.mail.in.yahoo.com> Message-ID: On Mon, 25 Feb 2008, Daya sagar wrote: DS> Hi DS> DS> Please unsubscribe me from this yahoo groups. daya, this is not a yahoo group but a mailing list, and you can unsubscribe yourself at: http://www.democritos.it/mailman/listinfo/pw_forum posting to the list itself does not work. cheers, axel. DS> DS> Daya DS> DS> DS> --------------------------------- DS> Download prohibited? No problem. CHAT from any browser, without download. -- ======================================================================= 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 kddst2005 at yahoo.com.cn Tue Feb 26 06:05:41 2008 From: kddst2005 at yahoo.com.cn (Dongdong Kang) Date: Tue, 26 Feb 2008 13:05:41 +0800 (CST) Subject: [Pw_forum] radial distribution function Message-ID: <599330.40906.qm@web92014.mail.cnb.yahoo.com> Dear all I know that the 'md' code can calculate the radial distribution function, and I want to know how to calculate this function. But I can't find the source code which can calculate the 'rdf' function. Can anyone tell me which file include the source code ? thanks Dongdong Kang National University of Defense Technology, P R China 410073, Hunan, Changsha --------------------------------- ??????????????????? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080226/33ac3b74/attachment.htm From akohlmey at cmm.chem.upenn.edu Tue Feb 26 10:21:47 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Tue, 26 Feb 2008 04:21:47 -0500 Subject: [Pw_forum] radial distribution function In-Reply-To: <599330.40906.qm@web92014.mail.cnb.yahoo.com> References: <599330.40906.qm@web92014.mail.cnb.yahoo.com> Message-ID: <7b6913e90802260121k62ba41e1y5133e6873554c0dc@mail.gmail.com> 2008/2/26 Dongdong Kang : > Dear all > > I know that the 'md' code can calculate the radial distribution function, > and I want to know how to calculate this function. But I can't find the > source code which can calculate the 'rdf' function. Can anyone tell me > which file include the source code ? there is an example of a RDF code only in about every MD textbook on the market. ...and if that is not enough you can do a short web search. off the top of my head i'd know about 10 places (MD codes, MD tutorials) where one could find one. so please look again. cheers, axel. > > thanks > > Dongdong Kang > National University of Defense Technology, P R China > 410073, Hunan, Changsha > > ________________________________ > ??????????????????? > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From giannozz at nest.sns.it Tue Feb 26 11:45:38 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 26 Feb 2008 11:45:38 +0100 Subject: [Pw_forum] segmentation fault In-Reply-To: <869386.66027.qm@web94610.mail.in2.yahoo.com> References: <869386.66027.qm@web94610.mail.in2.yahoo.com> Message-ID: <47C3EDD2.9080004@nest.sns.it> ambavale sagar wrote: > whenever i run structure optimization code i.e &control have > card calculation = relax ; my program stops giving segmentation > fault. this occurs even with example 03. It could not give force > components. Is this problem related to RAM? 'segmentation fault' may happen because of insufficient ram, but it is VERY unlikely that a small calculation like the one in example 3 can exceed the available memory of today's smallest machines > if yes then how much RAM is necessary? I do have 512MB RAM in > IBM eserver xseries 206. it strongly depends on the size of the system under examination. See the "Performance Issue" section of the user guide. The only thing that can be said based on the scant available information is that it is a problem specific to your machine, since nobody else has reported any similar problem Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From miguel.martinez at ehu.es Tue Feb 26 19:55:18 2008 From: miguel.martinez at ehu.es (=?UTF-8?B?TWlndWVsIE1hcnRpwq1uZXo=?=) Date: Tue, 26 Feb 2008 19:55:18 +0100 Subject: [Pw_forum] Quick question on coordinates Message-ID: <47C46096.7050307@ehu.es> Dear colleages, I'm currently doing some stuff on face-centered monoclinic structures, and have found silly problems. For example, the documentation reads: espresso/Doc/INPUT_PW: monoclinic (p) ============================= a1 = (a,0,0), a2= (b*sin(gamma), b*cos(gamma), 0), a3 = (0, 0, c) where gamma is the angle between axis a and b base centered monoclinic ============================= a1 = ( a/2, 0, -c/2), a2 = (b*cos(gamma), b*sin(gamma), 0), a3 = ( a/2, 0, c/2), ?Is it right? ?Is it a typo? I fear it's a typo, because changing ibrav from 12 to 13 decreases cell volume by exactly a factor of 2. What's more, \vec{a}*\vec{b}=|a|*|b|*cos(ab), while according to the basis in the monoclinic case we'd have \vec{a}*\vec{b}=|a|*|b|*sin(ab) Thanks in advance, Miguel -- ---------------------------------------- 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 5326 ---------------------------------------- "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 giannozz at nest.sns.it Wed Feb 27 16:09:03 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Wed, 27 Feb 2008 16:09:03 +0100 Subject: [Pw_forum] Quick question on coordinates In-Reply-To: <47C46096.7050307@ehu.es> References: <47C46096.7050307@ehu.es> Message-ID: <47C57D0F.3050409@nest.sns.it> Miguel Marti?nez wrote: > monoclinic (p) > ============================= > a1 = (a,0,0), a2= (b*sin(gamma), b*cos(gamma), 0), a3 = (0, 0, c) > where gamma is the angle between axis a and b > > base centered monoclinic > ============================= > a1 = ( a/2, 0, -c/2), > a2 = (b*cos(gamma), b*sin(gamma), 0), > a3 = ( a/2, 0, c/2), > > ?Is it right? ?Is it a typo? what is right and what is a typo? As far as I know, the documentation exactly reflects what the code does (see flib/latgen.f90) : else if (ibrav == 12) then sen=sqrt(1.d0-celldm(4)**2) a1(1)=celldm(1) a2(1)=celldm(1)*celldm(2)*celldm(4) a2(2)=celldm(1)*celldm(2)*sen a3(3)=celldm(1)*celldm(3) else if (ibrav == 13) then sen = sqrt( 1.d0 - celldm(4) ** 2 ) a1(1) = 0.5d0 * celldm(1) a1(3) =-a1(1) * celldm(3) a2(1) = celldm(1) * celldm(2) * celldm(4) a2(2) = celldm(1) * celldm(2) * sen a3(1) = a1(1) a3(3) =-a1(3) The case ibrav=13 has been corrected more than once by the way. If you have evidence that there is something wrong above, please explain what exactly and why Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From miguel.martinez at ehu.es Wed Feb 27 17:21:30 2008 From: miguel.martinez at ehu.es (=?UTF-8?B?TWlndWVsIE1hcnRpwq1uZXo=?=) Date: Wed, 27 Feb 2008 17:21:30 +0100 Subject: [Pw_forum] Quick question on coordinates In-Reply-To: <47C57D0F.3050409@nest.sns.it> References: <47C46096.7050307@ehu.es> <47C57D0F.3050409@nest.sns.it> Message-ID: <47C58E0A.4060203@ehu.es> Dear Paolo, First of all thanks for your helpful answer. In any case, if flib/latgen.f90 says: else if (ibrav == 12) then sen=sqrt(1.d0-celldm(4)**2) a1(1)=celldm(1) a2(1)=celldm(1)*celldm(2)*celldm(4) a2(2)=celldm(1)*celldm(2)*sen a3(3)=celldm(1)*celldm(3) Then the documentation seems to have sin(gamma) and cos(gamma) mixed for ibrav=12. Nothing harmful, though. Paolo Giannozzi escribi?: > The case ibrav=13 has been corrected more than once by the way. > If you have evidence that there is something wrong above, please > explain what exactly and why Truth to be said, the only evidence I have is my lack of skill in putting the coordinates of a C2/C crystal in ibrav=13, when the full positions work OK in ibrav 12. I must be doing something very very silly. By the way, I'm sure there is a reason why the special axis in quantum espresso is c while the tables love to use axis b. Even if the answer is like, "if crystallographers don't put 2\pi everywere, we don't like their basis sets". -- ---------------------------------------- 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 5326 ---------------------------------------- "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 marzari at MIT.EDU Wed Feb 27 17:51:22 2008 From: marzari at MIT.EDU (Nicola Marzari) Date: Wed, 27 Feb 2008 11:51:22 -0500 Subject: [Pw_forum] Quick question on coordinates In-Reply-To: <47C58E0A.4060203@ehu.es> References: <47C46096.7050307@ehu.es> <47C57D0F.3050409@nest.sns.it> <47C58E0A.4060203@ehu.es> Message-ID: <47C5950A.4000901@mit.edu> > By the way, I'm sure there is a reason why the special axis in quantum > espresso is c while the tables love to use axis b. Even if the answer is > like, "if crystallographers don't put 2\pi everywere, we don't like their > basis sets". The crystallographic notation in Q-E is derived from the "Spiritual Exercises" of Saint Ignatius of Loyola - the rationale is that by forcing the novice to engage in the excruciating repetition of a mindless task, he/she will ultimately renounce its own ego and dissolve in the exalted state of acting within the grace. Now, of course, I'll be sent to hell by the P.owers to be. nicola -- --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu From giannozz at nest.sns.it Wed Feb 27 18:34:34 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Wed, 27 Feb 2008 18:34:34 +0100 Subject: [Pw_forum] Quick question on coordinates In-Reply-To: <47C5950A.4000901@mit.edu> References: <47C46096.7050307@ehu.es> <47C57D0F.3050409@nest.sns.it> <47C58E0A.4060203@ehu.es> <47C5950A.4000901@mit.edu> Message-ID: <47C59F2A.3040105@nest.sns.it> Nicola Marzari wrote: >> By the way, I'm sure there is a reason why the special axis in quantum >> espresso is c while the tables love to use axis b. Even if the answer is >> like, "if crystallographers don't put 2\pi everywere, we don't like their >> basis sets". the reason is that whoever decided, many and many a year ago, how to pass crystal data structure information to the original codes upon which quantum-espresso is built, chose the current logic (or lack of it). It was kept during the years because 1) the advantage of moving to a more "standard" description, if any, wasn't clear wrt the disadvantage of having people go ballistic because their input data found after so much hard labor was no longer valid; 2) nobody has ever volunteered to delve into the Wyckoff and to program whatever crystal axis are used there, or to implement space groups and the like. See also: http://www.democritos.it/pipermail/pw_forum/2007-August/007011.html > The crystallographic notation in Q-E is derived from the "Spiritual > Exercises" of Saint Ignatius of Loyola - the rationale is that by > forcing the novice to engage in the excruciating repetition > of a mindless task, he/she will ultimately renounce its own ego > and dissolve in the exalted state of acting within the grace. > > Now, of course, I'll be sent to hell by the P.owers to be. very funny. Point 2) above is your penance. P. -- Paolo Giannozzi, Democritos and University of Udine, Italy From scandolo at ictp.it Wed Feb 27 18:54:30 2008 From: scandolo at ictp.it (Scandolo Sandro) Date: Wed, 27 Feb 2008 18:54:30 +0100 (CET) Subject: [Pw_forum] Quick question on coordinates In-Reply-To: <47C59F2A.3040105@nest.sns.it> References: <47C46096.7050307@ehu.es> <47C57D0F.3050409@nest.sns.it> <47C58E0A.4060203@ehu.es> <47C5950A.4000901@mit.edu> <47C59F2A.3040105@nest.sns.it> Message-ID: <41419.10.50.27.135.1204134870.squirrel@webmail3.ictp.trieste.it> > the reason is that whoever decided, many and many a year ago, how > to pass crystal data structure information to the original codes > upon which quantum-espresso is built, chose the current logic (or > lack of it). It was kept during the years because > See also: > http://www.democritos.it/pipermail/pw_forum/2007-August/007011.html > and also http://www.democritos.it/pipermail/pw_forum/2007-September/007169.html Sandro From baroni at sissa.it Wed Feb 27 19:45:34 2008 From: baroni at sissa.it (Stefano Baroni) Date: Wed, 27 Feb 2008 19:45:34 +0100 Subject: [Pw_forum] Quick question on coordinates In-Reply-To: <47C59F2A.3040105@nest.sns.it> References: <47C46096.7050307@ehu.es> <47C57D0F.3050409@nest.sns.it> <47C58E0A.4060203@ehu.es> <47C5950A.4000901@mit.edu> <47C59F2A.3040105@nest.sns.it> Message-ID: On Feb 27, 2008, at 6:34 PM, Paolo Giannozzi wrote: > 2) nobody has ever volunteered to delve into the Wyckoff and to > program whatever crystal axis are used there, or to implement > space groups and the like. this is not exactly true. Moreno Baricevic has written a beatiful perl script that is expected to translate the info contained in Wyckoff's tables into something digestible by QE. Unfortunately, nobody has ever volunteered to test it thoroughly (not even occasionally), not to mention the little coding that is still necessary to turn the digestion process into actual metabolism ... If anybody out there is willing to help, I will gladly provide Moreno's script, along with info about what it is supposed to do ... Stefano --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080227/bd5fde0a/attachment.htm From sks.jnc at gmail.com Wed Feb 27 20:22:38 2008 From: sks.jnc at gmail.com (S. K. S.) Date: Thu, 28 Feb 2008 00:52:38 +0530 Subject: [Pw_forum] Gd LDA pseudo Message-ID: Dear Sir, Thanks a lot for your kind reply. First time, I'm trying to generate a pseudo myself. Being newbie, some thing I can not understand clearly. As for example, in pwscf suite it's given for Os that ***************************************************************** &input title='Os', zed=76., rel=1, iswitch=3, config='[Xe] 4f14.0 5d6.0 6s2.0 6p0.0', dft='LDA' / &inputp lloc=0, pseudotype=3, file_pseudopw='Os.pz-rrkjus.UPF', nlcc=.true., rcore=0.9, author='ADC', / 4 6P 2 1 0.00 0.00 2.60 2.60 1 5D 3 2 6.00 0.00 1.90 2.30 1 5D 3 2 0.00 -0.20 1.90 2.30 1 6S 1 0 2.00 0.00 2.50 2.50 1 ********************************************* Though I can understand the meaning of the most of the input parameters, however, in last part , why the numbers are like these 6P 0.00 2.60 2.60 1 5D 0.00 1.90 2.30 1 5D -0.20 1.90 2.30 1 6S 0.00 2.50 2.50 1 , I can not understand clearly. I know that for Gd it will be like that 6P 2 1 0.00 5D 3 2 1.00 5D 3 2 0.00 6S 1 0 2.00 but what will be the other numbers for Gd???? I'll be highly obliged if somebody kindly help me to understand the above points. First, I'll test the generated Gd pseudo and if it works well I'll defintely give it back to PWSCF website. Look forward to your kind help. With my best regards, SKS JNCASR BANGALORE ---------- Forwarded message ---------- From: Dal Corso Andrea Date: Feb 27, 2008 8:29 PM Subject: Re: Gd LDA pseudo To: "S. K. S." Dear S.K.S. Presently I do not have a Gd pseudopotential. Please post this question to the pw_forum mailing list, maybe someone will be able to provide it. Best wishes Andrea ---------- Forwarded message ---------- From: Dal Corso Andrea Date: Feb 27, 2008 8:29 PM Subject: Re: Gd LDA pseudo To: "S. K. S." Dear S.K.S. Presently I do not have a Gd pseudopotential. Please post this question to the pw_forum mailing list, maybe someone will be able to provide it. Best wishes Andrea On Wed, 2008-02-27 at 20:19 +0530, S. K. S. wrote: > > > Respected Sir, > > May I get a Gd pseudo (LDA) for PWSCF from > you or any of your students??? > > I know that the rare earth is poorly described > by the plane waves. However, there are lot of DFT calculations > on Lanthanides and Ceria in literature. > > Looking forward to your kind help. > > With my best regards, > SKS > JNCASR > Bangalore > > P.S. I'll be very happy to get someone as a coauthor if he/she > helps me by providing the above pseudo. -- Andrea Dal Corso Tel. 0039-040-3787428 SISSA, Via Beirut 2/4 Fax. 0039-040-3787528 34014 Trieste (Italy) e-mail: dalcorso at sissa.it On Wed, 2008-02-27 at 20:19 +0530, S. K. S. wrote: > > > Respected Sir, > > May I get a Gd pseudo (LDA) for PWSCF from > you or any of your students??? > > I know that the rare earth is poorly described > by the plane waves. However, there are lot of DFT calculations > on Lanthanides and Ceria in literature. > > Looking forward to your kind help. > > With my best regards, > SKS > JNCASR > Bangalore > > P.S. I'll be very happy to get someone as a coauthor if he/she > helps me by providing the above pseudo. -- Andrea Dal Corso Tel. 0039-040-3787428 SISSA, Via Beirut 2/4 Fax. 0039-040-3787528 34014 Trieste (Italy) e-mail: dalcorso at sissa.it -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080228/7cae5887/attachment-0001.htm From miguel.martinez at ehu.es Wed Feb 27 20:33:23 2008 From: miguel.martinez at ehu.es (=?ISO-8859-1?Q?Miguel_Mart=EDnez_Canales?=) Date: Wed, 27 Feb 2008 20:33:23 +0100 Subject: [Pw_forum] Quick question on coordinates In-Reply-To: References: <47C46096.7050307@ehu.es><47C57D0F.3050409@nest.sns.it> <47C58E0 A.4060203@ehu.es><47C5950A.4000901@mit.edu> <47C59F2A.3040105@nest.sns.it> Message-ID: <47C5BB03.9060002@ehu.es> Thanks everybody for your comments and suggestions. I'd be willing to give Moreno's script a go slightly later (I'm pretty busy at the moment), although my perl skills are, well, non-existent. Nicola might also be interested to know that my school belonged to the Marianists (society of Mary), so the pupils of St. Ignatius' school were actually our main rivals ;-). We're obviously the better ones. Stefano Baroni wrote: > > On Feb 27, 2008, at 6:34 PM, Paolo Giannozzi wrote: >> 2) nobody has ever volunteered to delve into the Wyckoff and to >> program whatever crystal axis are used there, or to implement >> space groups and the like. > > this is not exactly true. Moreno Baricevic has written a beatiful perl > script that is expected to translate the info contained in Wyckoff's > tables into something digestible by QE. Unfortunately, nobody has ever > volunteered to test it thoroughly (not even occasionally), not to > mention the little coding that is still necessary to turn the digestion > process into actual metabolism ... > > If anybody out there is willing to help, I will gladly provide Moreno's > script, along with info about what it is supposed to do ... > > Stefano -- ---------------------------------------- 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 chenhanghuipwscf at gmail.com Thu Feb 28 05:01:31 2008 From: chenhanghuipwscf at gmail.com (alan chen) Date: Wed, 27 Feb 2008 23:01:31 -0500 Subject: [Pw_forum] symmetrized projection in atomic projected DOS Message-ID: <22ae3ca40802272001i4ff4e3abu2ebb2c994e28fc75@mail.gmail.com> Dear PWSCF users, I have a question about the flag 'lsym' in the atomic projected DOS ( projwfc.f90). The flag 'lsym' is set .true. by default. I ran it both with 'lsym' = .true. or .false. and found that if the Block state is nondegenerate, the decomposition into atomic orbitals is the same no matter whether 'lsym'=.true. or .false. However for those degenerate states, the coefficients are different. For example, if there are three degenerate Block states and 'lsym'=.true., the coefficients in the decomposition (i.e. modulous square of the overlap) are the same for these three states. If 'lsym' = false, the coefficients are different in the three states. So why are the results different if 'lsym' = true and false? And which one is correct? What is symmetrized projection? Thank you in advance. Hanghui Chen Department of Physics, Yale University -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080227/89e9964c/attachment.htm From ylli at theory.issp.ac.cn Thu Feb 28 09:06:56 2008 From: ylli at theory.issp.ac.cn (Li yan-ling) Date: Thu, 28 Feb 2008 16:06:56 +0800 (CST) Subject: [Pw_forum] q2r.x error. Message-ID: <14261.137.189.4.2.1204186016.squirrel@210.73.16.4> Dear all users, When I performed the electron-phonon calculation, the errors happened as follows. ------------------------------------------------------------------------- forrtl: severe (59): list-directed I/O syntax error, unit 51, file /storage/SCSI/st12/home/user/isspzz/work/ylli/pwscf/ReN/new/test/a2Fq2r.51 Image PC Routine Line Source CPU time : 394830.47 sec. Max Memory : 1063 MB Max Swap : 2320 MB Max Processes : 22 Max Threads : 22 The output (if any) follows: forrtl: severe (59): list-directed I/O syntax error, unit 51, file /storage/SCSI/st12/home/user/isspzz/work/ylli/pwscf/ReN/new/test/a2Fq2r.51 Image PC Routine Line Source q2r.x 00000000007BC213 Unknown Unknown Unknown q2r.x 00000000007BA54E Unknown Unknown Unknown q2r.x 00000000007925DE Unknown Unknown Unknown q2r.x 0000000000755BD6 Unknown Unknown Unknown q2r.x 0000000000755837 Unknown Unknown Unknown q2r.x 000000000076C5D5 Unknown Unknown Unknown q2r.x 0000000000411888 Unknown Unknown Unknown q2r.x 000000000040FC70 Unknown Unknown Unknown q2r.x 000000000040B514 Unknown Unknown Unknown q2r.x 00000000004092EA Unknown Unknown Unknown libc.so.6 0000002A95BD967A Unknown Unknown Unknown q2r.x 000000000040922A Unknown Unknown Unknown forrtl: severe (24): end-of-file during read, unit 61, file /storage/SCSI/st12/home/user/isspzz/work/ylli/pwscf/ReN/new/test/a2Fmatdyn.61 Image PC Routine Line Source matdyn.x 00000000007D2F13 Unknown Unknown Unknown matdyn.x 00000000007D124E Unknown Unknown Unknown matdyn.x 00000000007A92CE Unknown Unknown Unknown matdyn.x 000000000076C8C6 Unknown Unknown Unknown matdyn.x 000000000076C527 Unknown Unknown Unknown matdyn.x 0000000000782916 Unknown Unknown Unknown matdyn.x 00000000004103C4 Unknown Unknown Unknown matdyn.x 000000000040B25C Unknown Unknown Unknown matdyn.x 00000000004092EA Unknown Unknown Unknown libc.so.6 0000002A95BD967A Unknown Unknown Unknown matdyn.x 000000000040922A Unknown Unknown Unknown forrtl: severe (24): end-of-file during read, unit 61, file /storage/SCSI/st12/home/user/isspzz/work/ylli/pwscf/ReN/new/test/a2Fmatdyn.61 Image PC Routine Line Source matdyn.x 00000000007D2F13 Unknown Unknown Unknown matdyn.x 00000000007D124E Unknown Unknown Unknown matdyn.x 00000000007A92CE Unknown Unknown Unknown matdyn.x 000000000076C8C6 Unknown Unknown Unknown matdyn.x 000000000076C527 Unknown Unknown Unknown matdyn.x 0000000000782916 Unknown Unknown Unknown matdyn.x 00000000004103C4 Unknown Unknown Unknown matdyn.x 000000000040B25C Unknown Unknown Unknown matdyn.x 00000000004092EA Unknown Unknown Unknown libc.so.6 0000002A95BD967A Unknown Unknown Unknown matdyn.x 000000000040922A Unknown Unknown Unknown --------------------------------------------------- By checking the output file 'q2r.out', I know the errors from the q2r.x program. For example, ......................... ......................... Dynamical Matrix in cartesian axes q = ( 0.000000000 -0.288675135 -0.236569803 ) 1 1 89.36769217 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 89.39381871 0.00000000 -0.01004736 0.00000000 0.00000000 0.00000000 -0.01004736 0.00000000 89.38396315 0.00000000 1 2 0.00000000************ 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000************ 0.00000000 -0.00300097 0.00000000 0.00000000 0.00000000 -0.00300097 0.00000000************ 1 3 44.65732502-50.41210189 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 44.65839582-50.41300456 50.40129204-44.66012171 0.00000000 0.00000000 50.39697063-44.65959080 44.66166361-50.41463893 1 4 50.41210182-44.65732510 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 50.41300449-44.65839590 44.66012163-50.40129211 0.00000000 0.00000000 44.65959072-50.39697070 50.41463886-44.66166369 2 1 0.00000000************ 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000************ 0.00000000 0.00300097 0.00000000 0.00000000 0.00000000 0.00300097 0.00000000************ .......................... ...................... That is to say, there are uncorrect data in a2Fq2r.51. I want to know the reason that results in these errors. I'm looking forward to solving it with your help. Thanks in advance. Yanling Li. From baris.malcioglu at gmail.com Thu Feb 28 10:23:16 2008 From: baris.malcioglu at gmail.com (=?ISO-8859-1?Q?Bar=3Fs=2C_Malc=3Fog=28lu?=) Date: Thu, 28 Feb 2008 11:23:16 +0200 Subject: [Pw_forum] how to cite pwscf In-Reply-To: <14261.137.189.4.2.1204186016.squirrel@210.73.16.4> References: <14261.137.189.4.2.1204186016.squirrel@210.73.16.4> Message-ID: <47C67D84.3020005@gmail.com> Dear all, Is there a preferred way for citing pwscf and individual pseudo potentials obtained from the web page? I was unable to find any information regarding this both in the wiki and the mailing list. I have encountered a number of different ways in the literature, and a little confused on which one to use. Best, Baris Malcioglu Ph.D. Candidate, METU Physics. From baris.malcioglu at gmail.com Thu Feb 28 10:27:18 2008 From: baris.malcioglu at gmail.com (=?ISO-8859-1?Q?Bar=3Fs=2C_Malc=3Fog=28lu?=) Date: Thu, 28 Feb 2008 11:27:18 +0200 Subject: [Pw_forum] q2r.x error. In-Reply-To: <14261.137.189.4.2.1204186016.squirrel@210.73.16.4> References: <14261.137.189.4.2.1204186016.squirrel@210.73.16.4> Message-ID: <47C67E76.80603@gmail.com> Dear Yanling Li, > > forrtl: severe (59): list-directed I/O syntax error, unit 51, file > > /storage/SCSI/st12/home/user/isspzz/work/ylli/pwscf/ReN/new/test/a2Fq2r.51 > > > This means you are unable to read or write the corresponding file, did you check your disk quota/free space/bad sectors/etc.? O. Baris Malcioglu Ph.D. Candidate METU Physics From eyvaz_isaev at yahoo.com Thu Feb 28 13:17:58 2008 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Thu, 28 Feb 2008 04:17:58 -0800 (PST) Subject: [Pw_forum] q2r.x error. In-Reply-To: <14261.137.189.4.2.1204186016.squirrel@210.73.16.4> Message-ID: <409122.92341.qm@web60316.mail.yahoo.com> Dear Yanling, Hopefully you paid attention that in a2Fq2r.51 there are lots of stars, meaning that something went in a wrong way. That is why q2r.x complained. So, look at closer to your input files and then try restart the job. It might be that some temporary files turned out to be corrupted. Bests, Eyvaz. --- Li yan-ling wrote: > --------------------------------------------------- > By checking the output file 'q2r.out', I know the > errors from the q2r.x > program. For example, > ......................... > ......................... > Dynamical Matrix in cartesian axes > > q = ( 0.000000000 -0.288675135 > -0.236569803 ) > > 1 1 > 89.36769217 0.00000000 0.00000000 0.00000000 > 0.00000000 0.00000000 > 0.00000000 0.00000000 89.39381871 0.00000000 > -0.01004736 0.00000000 > 0.00000000 0.00000000 -0.01004736 0.00000000 > 89.38396315 0.00000000 > 1 2 > 0.00000000************ 0.00000000 0.00000000 > 0.00000000 0.00000000 > 0.00000000 0.00000000 0.00000000************ > 0.00000000 -0.00300097 > 0.00000000 0.00000000 0.00000000 -0.00300097 > 0.00000000************ > 1 3 > 44.65732502-50.41210189 0.00000000 0.00000000 > 0.00000000 0.00000000 > 0.00000000 0.00000000 44.65839582-50.41300456 > 50.40129204-44.66012171 > 0.00000000 0.00000000 50.39697063-44.65959080 > 44.66166361-50.41463893 > 1 4 > 50.41210182-44.65732510 0.00000000 0.00000000 > 0.00000000 0.00000000 > 0.00000000 0.00000000 50.41300449-44.65839590 > 44.66012163-50.40129211 > 0.00000000 0.00000000 44.65959072-50.39697070 > 50.41463886-44.66166369 > 2 1 > 0.00000000************ 0.00000000 0.00000000 > 0.00000000 0.00000000 > 0.00000000 0.00000000 0.00000000************ > 0.00000000 0.00300097 > 0.00000000 0.00000000 0.00000000 0.00300097 > 0.00000000************ > .......................... > ...................... > > That is to say, there are uncorrect data in > a2Fq2r.51. I want to know > the reason that results in these errors. I'm looking > forward to solving > it with your help. > > Thanks in advance. > > > Yanling Li. > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, IFM, Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com ____________________________________________________________________________________ Looking for last minute shopping deals? Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping From jdogbe at gmail.com Thu Feb 28 14:22:27 2008 From: jdogbe at gmail.com (John K. Dogbe, Ph.D.) Date: Thu, 28 Feb 2008 08:22:27 -0500 Subject: [Pw_forum] Moreno's Script Message-ID: <47C6B593.2050804@gmail.com> Hi Stefano, Please, send me Moreno's script and I would test it and give you a feedback. I have a small group of students who could 'mis-handle' it as well and we can give you how it handles under those circumstances too. In any case, might not have all possible scenarios but I will give it my best shot and as I said give you/list a continuous feedback. John. -- ============================================================ John K. Dogbe, Visiting Assistant Professor Barton College, Dept. of Science & Mathematics 400 Atlantic Christian College Dr. NE, Wilson,NC 27893-2575 Tel.: (252) 399-6471, Fax.: (252) 399-6465 ============================================================ **The first is not necessarily the Leader** "If I have spoken evil, bear witness of the evil: but if well, why smitest thou me?" -- Jesus Christ (John, 18:23) From giannozz at nest.sns.it Thu Feb 28 19:39:08 2008 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 28 Feb 2008 19:39:08 +0100 Subject: [Pw_forum] symmetrized projection in atomic projected DOS In-Reply-To: <22ae3ca40802272001i4ff4e3abu2ebb2c994e28fc75@mail.gmail.com> References: <22ae3ca40802272001i4ff4e3abu2ebb2c994e28fc75@mail.gmail.com> Message-ID: <387C933E-7B39-4A82-B786-3CE6C96431CC@nest.sns.it> On Feb 28, 2008, at 5:01 , alan chen wrote: > Dear PWSCF users, > I have a question about the flag 'lsym' in the atomic > projected DOS (projwfc.f90). The flag 'lsym' is set .true. by > default. I ran it both with 'lsym' = .true. or .false. and found > that if the Block state is nondegenerate, the decomposition into > atomic orbitals is the same no matter whether 'lsym'=.true. > or .false. However for those degenerate states, the coefficients > are different. For example, if there are three degenerate Block > states and 'lsym'=.true., the coefficients in the decomposition > (i.e. modulous square of the overlap) are the same for these three > states. If 'lsym' = false, the coefficients are different in the > three states. > So why are the results different if 'lsym' = true and false? > And which one is correct? > What is symmetrized projection? hint: - your third question should be the first one (and the only one) - your second question doesn't make any sense: both are correct (hopefully) - your first question doesn't make any sense either: they differ because they are (slightly) different quantities - the answer is contained in your observations Disclaimer: I didn't write that part of the code Paolo --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From eyvaz_isaev at yahoo.com Thu Feb 28 20:43:32 2008 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Thu, 28 Feb 2008 11:43:32 -0800 (PST) Subject: [Pw_forum] how to cite pwscf In-Reply-To: <47C67D84.3020005@gmail.com> Message-ID: <434615.2588.qm@web60314.mail.yahoo.com> Dear Baris, --- Bar?s, Malc?og(lu wrote: > Is there a preferred way for citing pwscf and > individual pseudo potentials obtained from the web >page? > Please have a look at users_guide.tex in /Doc, section 1.4. Preliminary compilation of the tex-file by means of pdflatex (or any other TeX compiler) is highly recommended. Bests, Eyvaz. > I was unable to find any information regarding this > both in the wiki and > the mailing list. I have encountered a number of > different ways in the > literature, and a little confused on which one to > use. > > Best, > > Baris Malcioglu > Ph.D. Candidate, METU Physics. > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, IFM, Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com ____________________________________________________________________________________ Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ From naromero at gmail.com Thu Feb 28 21:02:13 2008 From: naromero at gmail.com (Nichols A. Romero) Date: Thu, 28 Feb 2008 15:02:13 -0500 Subject: [Pw_forum] radial distribution function In-Reply-To: <7b6913e90802260121k62ba41e1y5133e6873554c0dc@mail.gmail.com> References: <599330.40906.qm@web92014.mail.cnb.yahoo.com> <7b6913e90802260121k62ba41e1y5133e6873554c0dc@mail.gmail.com> Message-ID: <6ac064b60802281202k7348d2d7hefad7101e20305c0@mail.gmail.com> Try the textbook: Algorithm 7 in "Understanding Molecular Simulation" by Daan Frenkel & Berend Smit If you need a code that actually computes the rdf, I can probably find my old Fortran code. On Tue, Feb 26, 2008 at 4:21 AM, Axel Kohlmeyer wrote: > 2008/2/26 Dongdong Kang : > > Dear all > > > > I know that the 'md' code can calculate the radial distribution > function, > > and I want to know how to calculate this function. But I can't find the > > source code which can calculate the 'rdf' function. Can anyone tell me > > which file include the source code ? > > there is an example of a RDF code only in about every > MD textbook on the market. ...and if that is not enough you > can do a short web search. off the top of my head i'd know > about 10 places (MD codes, MD tutorials) where one could > find one. so please look again. > > cheers, > axel. > > > > > > thanks > > > > Dongdong Kang > > National University of Defense Technology, P R China > > 410073, Hunan, Changsha > > > > ________________________________ > > ??????????????????? > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > > -- > ======================================================================= > Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu > Center for Molecular Modeling -- University of Pennsylvania > Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 > tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 > ======================================================================= > If you make something idiot-proof, the universe creates a better idiot. > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Nichols A. Romero, Ph.D. DoD User Productivity Enhancement and Technology Transfer (PET) Group High Performance Technologies, Inc. Reston, VA 443-567-8328 (C) 410-278-2692 (O) -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080228/7e6e5a7b/attachment-0001.htm From ylli at theory.issp.ac.cn Fri Feb 29 06:27:33 2008 From: ylli at theory.issp.ac.cn (Li yan-ling) Date: Fri, 29 Feb 2008 13:27:33 +0800 (CST) Subject: [Pw_forum] q2r.x error Message-ID: <28803.137.189.4.2.1204262853.squirrel@210.73.16.4> Dear all users, Thanks the answer. Now I want to know which program results in the output data possessing the stars (**) in "a2Fq2r.5*" file. What is the nature of this case? I checked the input file, I think it's right. The input files is as follows. --------------------------------- ### scf.fit.in ### &control calculation='scf' restart_mode='from_scratch' pseudo_dir = '../', outdir='./tmp/' prefix='ren' / &system ibrav= 4, a = 2.7435, c = 5.7985,nat= 4, ntyp= 2,nbnd =32, ecutwfc =32.0, ecutrho = 400.0, occupations='smearing',smearing='methfessel-paxton', degauss=0.03,la2F=.true., / &electrons conv_thr = 1.0d-10 mixing_beta = 0.5 / ATOMIC_SPECIES Re 186.207 Re.pw91-n-van.UPF N 14.0067 N.pw91-van_ak.UPF ATOMIC_POSITIONS crystal Re 0.00 0.00 0.00 Re 0.00 0.00 0.50 N 0.333333333 0.666666666 0.25 N 0.666666666 0.333333333 0.75 K_POINTS {automatic} 36 36 16 0 0 0 ###scf.in#### &control calculation='scf' restart_mode='from_scratch' pseudo_dir = '../', outdir='./tmp/' prefix='ren' / &system ibrav= 4, a = 2.7435, c = 5.7985,nat= 4, ntyp= 2,nbnd=32, ecutwfc =32.0,ecutrho = 400.0, occupations='smearing',smearing='methfessel-paxton', degauss=0.03, / &electrons conv_thr = 1.0d-10 mixing_beta = 0.5 / ATOMIC_SPECIES Re 186.207 Re.pw91-n-van.UPF N 14.0067 N.pw91-van_ak.UPF ATOMIC_POSITIONS crystal Re 0.00 0.00 0.00 Re 0.00 0.00 0.50 N 0.333333333 0.666666666 0.25 N 0.666666666 0.333333333 0.75 K_POINTS {automatic} 16 16 8 0 0 0 ####ph.in#### Electron-phonon coefficients for ReN &inputph tr2_ph=1.0d-13, prefix='ren', fildvscf='rendv', amass(1)=186.207,amass(2)=14.0067, outdir='./tmp/', fildyn='ren.dyn', elph=.true., trans=.true., ldisp=.true. nq1=4,nq2=4,nq3=2 / #####q2r.in#### &input zasr='simple', fildyn='ren.dyn', flfrc='ren442.fc', la2F=.true. / --------------------------------------------- The system studied have the NiAs symmetry with the P63/mmc spacegroup. If you can tell me the answer, it's my pleasure. Thanks in advance. Yanling Li. From akohlmey at cmm.chem.upenn.edu Fri Feb 29 15:53:16 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Fri, 29 Feb 2008 09:53:16 -0500 (EST) Subject: [Pw_forum] q2r.x error In-Reply-To: <28803.137.189.4.2.1204262853.squirrel@210.73.16.4> References: <28803.137.189.4.2.1204262853.squirrel@210.73.16.4> Message-ID: On Fri, 29 Feb 2008, Li yan-ling wrote: YL> Dear all users, YL> Thanks the answer. Now I want to know which program results in the YL> output data possessing the stars (**) in "a2Fq2r.5*" file. What is the YL> nature of this case? the fact that you (literally) see stars means that the number computed cannot be written into the digits reserved for it. it can be either a very large number or an illegal "number". i.e. result of taking the square root of a negative number, division by zero or something similar. YL> I checked the input file, I think it's right. The input files is as YL> follows. well this could be a case of "weird atom, gotta check the pseudopotential". do you have any information of the origin of the Re potential? cheers, axel. YL> --------------------------------- YL> ### scf.fit.in ### YL> YL> &control YL> calculation='scf' YL> restart_mode='from_scratch' YL> pseudo_dir = '../', YL> outdir='./tmp/' YL> prefix='ren' YL> / YL> &system YL> ibrav= 4, a = 2.7435, c = 5.7985,nat= 4, ntyp= 2,nbnd =32, YL> ecutwfc =32.0, ecutrho = 400.0, YL> occupations='smearing',smearing='methfessel-paxton', YL> degauss=0.03,la2F=.true., YL> / YL> &electrons YL> conv_thr = 1.0d-10 YL> mixing_beta = 0.5 YL> / YL> ATOMIC_SPECIES YL> Re 186.207 Re.pw91-n-van.UPF YL> N 14.0067 N.pw91-van_ak.UPF YL> ATOMIC_POSITIONS crystal YL> Re 0.00 0.00 0.00 YL> Re 0.00 0.00 0.50 YL> N 0.333333333 0.666666666 0.25 YL> N 0.666666666 0.333333333 0.75 YL> K_POINTS {automatic} YL> 36 36 16 0 0 0 YL> YL> ###scf.in#### YL> &control YL> calculation='scf' YL> restart_mode='from_scratch' YL> pseudo_dir = '../', YL> outdir='./tmp/' YL> prefix='ren' YL> / YL> &system YL> ibrav= 4, a = 2.7435, c = 5.7985,nat= 4, ntyp= 2,nbnd=32, YL> ecutwfc =32.0,ecutrho = 400.0, YL> occupations='smearing',smearing='methfessel-paxton', YL> degauss=0.03, YL> / YL> &electrons YL> conv_thr = 1.0d-10 YL> mixing_beta = 0.5 YL> / YL> ATOMIC_SPECIES YL> Re 186.207 Re.pw91-n-van.UPF YL> N 14.0067 N.pw91-van_ak.UPF YL> ATOMIC_POSITIONS crystal YL> Re 0.00 0.00 0.00 YL> Re 0.00 0.00 0.50 YL> N 0.333333333 0.666666666 0.25 YL> N 0.666666666 0.333333333 0.75 YL> K_POINTS {automatic} YL> 16 16 8 0 0 0 YL> YL> ####ph.in#### YL> Electron-phonon coefficients for ReN YL> &inputph YL> tr2_ph=1.0d-13, YL> prefix='ren', YL> fildvscf='rendv', YL> amass(1)=186.207,amass(2)=14.0067, YL> outdir='./tmp/', YL> fildyn='ren.dyn', YL> elph=.true., YL> trans=.true., YL> ldisp=.true. YL> nq1=4,nq2=4,nq3=2 YL> / YL> YL> #####q2r.in#### YL> &input YL> zasr='simple', YL> fildyn='ren.dyn', YL> flfrc='ren442.fc', YL> la2F=.true. YL> / YL> --------------------------------------------- YL> The system studied have the NiAs symmetry with the P63/mmc spacegroup. YL> If you can tell me the answer, it's my pleasure. YL> YL> Thanks in advance. YL> YL> Yanling Li. YL> YL> YL> YL> YL> YL> _______________________________________________ YL> Pw_forum mailing list YL> Pw_forum at pwscf.org YL> http://www.democritos.it/mailman/listinfo/pw_forum YL> -- ======================================================================= 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 subhra at iitg.ernet.in Fri Feb 29 16:26:11 2008 From: subhra at iitg.ernet.in (Subhradip Ghosh) Date: Fri, 29 Feb 2008 20:56:11 +0530 (IST) Subject: [Pw_forum] symmetry operation # not compatible with FFT grid Message-ID: <41690.172.16.109.56.1204298771.squirrel@webmail.iitg.ernet.in> I am trying to run self-consistent calculations on a heusler structure with tetragonal distortion i.e. c/a=1.2. In the very beginning I am getting the following warnings: warning: symmetry operation # 2 not compatible with FFT grid. 0 1 -1 1 0 -1 0 0 -1 warning: symmetry operation # 3 not compatible with FFT grid. 0 -1 1 0 -1 0 1 -1 0 warning: symmetry operation # 4 not compatible with FFT grid. -1 0 0 -1 0 1 -1 1 0 warning: symmetry operation # 5 not compatible with FFT grid. What could be the cause for this and how to get rid of it? I ran calculations with c/a=0.9 and c/a=1-the warning didn't ocuur. But it is coming up with c/a=1 and c/a=1.2. Subhradip ************************************************************************ Dr. Subhradip Ghosh Assistant Professor Department of Physics Indian Institute of Technology Guwahati,Assam-781039 India E-mail:subhra at iitg.ernet.in Phone: +91 361 2582717(O) +91 361 2584717(R) Fax: +91 361 2582749 (Physics) +91 361 2690762 (General) ************************************************************************ From akohlmey at cmm.chem.upenn.edu Fri Feb 29 16:22:19 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Fri, 29 Feb 2008 10:22:19 -0500 (EST) Subject: [Pw_forum] symmetry operation # not compatible with FFT grid In-Reply-To: <41690.172.16.109.56.1204298771.squirrel@webmail.iitg.ernet.in> References: <41690.172.16.109.56.1204298771.squirrel@webmail.iitg.ernet.in> Message-ID: On Fri, 29 Feb 2008, Subhradip Ghosh wrote: SG> I am trying to run self-consistent calculations on a heusler structure SG> with tetragonal distortion i.e. c/a=1.2. In the very beginning I am SG> getting the following warnings: SG> SG> SG> warning: symmetry operation # 2 not compatible with FFT grid. please check out the troubleshooting guide: http://www.quantum-espresso.org/wiki/index.php/Troubleshooting_%28PWscf%29#warning:_.27symmetry_operation_.23_N_not_allowed.27 cheers, axel. SG> 0 1 -1 SG> 1 0 -1 SG> 0 0 -1 SG> warning: symmetry operation # 3 not compatible with FFT grid. SG> 0 -1 1 SG> 0 -1 0 SG> 1 -1 0 SG> warning: symmetry operation # 4 not compatible with FFT grid. SG> -1 0 0 SG> -1 0 1 SG> -1 1 0 SG> warning: symmetry operation # 5 not compatible with FFT grid. SG> SG> What could be the cause for this and how to get rid of it? I ran SG> calculations with c/a=0.9 and c/a=1-the warning didn't ocuur. But it is SG> coming up with c/a=1 and c/a=1.2. SG> SG> Subhradip SG> SG> ************************************************************************ SG> Dr. Subhradip Ghosh SG> Assistant Professor SG> Department of Physics SG> Indian Institute of Technology SG> Guwahati,Assam-781039 SG> India SG> E-mail:subhra at iitg.ernet.in SG> Phone: +91 361 2582717(O) SG> +91 361 2584717(R) SG> Fax: +91 361 2582749 (Physics) SG> +91 361 2690762 (General) SG> ************************************************************************ SG> SG> SG> _______________________________________________ SG> Pw_forum mailing list SG> Pw_forum at pwscf.org SG> http://www.democritos.it/mailman/listinfo/pw_forum SG> -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From eyvaz_isaev at yahoo.com Fri Feb 29 16:54:07 2008 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Fri, 29 Feb 2008 07:54:07 -0800 (PST) Subject: [Pw_forum] symmetry operation # not compatible with FFT grid In-Reply-To: <41690.172.16.109.56.1204298771.squirrel@webmail.iitg.ernet.in> Message-ID: <14895.97875.qm@web65714.mail.ac4.yahoo.com> Dear Subhradip, This is really FAQ in the forum. So, please, search the forum archive, and have a look at the users-guide in /DOC (Page 59 if PDF file, ?warning: symmetry operation # N not allowed?) To get rid of this message you can choose manually nr1,nr2, nr3 in &system or increase ecutrho. Bests, Eyvaz. --- Subhradip Ghosh wrote: > I am trying to run self-consistent calculations on a > heusler structure > with tetragonal distortion i.e. c/a=1.2. In the very > beginning I am > getting the following warnings: > > > warning: symmetry operation # 2 not compatible > with FFT grid. > 0 1 -1 > 1 0 -1 > 0 0 -1 > warning: symmetry operation # 3 not compatible > with FFT grid. > 0 -1 1 > 0 -1 0 > 1 -1 0 > warning: symmetry operation # 4 not compatible > with FFT grid. > -1 0 0 > -1 0 1 > -1 1 0 > warning: symmetry operation # 5 not compatible > with FFT grid. > > What could be the cause for this and how to get rid > of it? I ran > calculations with c/a=0.9 and c/a=1-the warning > didn't ocuur. But it is > coming up with c/a=1 and c/a=1.2. > > Subhradip > > ************************************************************************ > Dr. Subhradip Ghosh > Assistant Professor > Department of Physics > Indian Institute of Technology > Guwahati,Assam-781039 > India > E-mail:subhra at iitg.ernet.in > Phone: +91 361 2582717(O) > +91 361 2584717(R) > Fax: +91 361 2582749 (Physics) > +91 361 2690762 (General) > ************************************************************************ > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, IFM, Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com ____________________________________________________________________________________ Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs From subhra at iitg.ernet.in Fri Feb 29 17:51:54 2008 From: subhra at iitg.ernet.in (Subhradip Ghosh) Date: Fri, 29 Feb 2008 22:21:54 +0530 (IST) Subject: [Pw_forum] symmetry operation # not compatible with FFT grid In-Reply-To: <14895.97875.qm@web65714.mail.ac4.yahoo.com> References: <14895.97875.qm@web65714.mail.ac4.yahoo.com> Message-ID: <34755.172.16.109.56.1204303914.squirrel@webmail.iitg.ernet.in> Dear Eyvaz, Thanks. But, if I do not try to get rid of message, what could be the consequences? Subhradip > Dear Subhradip, > > This is really FAQ in the forum. So, please, search > the forum archive, and have a look at the users-guide > in /DOC (Page 59 if PDF file, ?warning: symmetry > operation # N not allowed?) > > To get rid of this message you can choose manually > nr1,nr2, nr3 in &system or increase ecutrho. > > Bests, > Eyvaz. > > --- Subhradip Ghosh wrote: > >> I am trying to run self-consistent calculations on a >> heusler structure >> with tetragonal distortion i.e. c/a=1.2. In the very >> beginning I am >> getting the following warnings: >> >> >> warning: symmetry operation # 2 not compatible >> with FFT grid. >> 0 1 -1 >> 1 0 -1 >> 0 0 -1 >> warning: symmetry operation # 3 not compatible >> with FFT grid. >> 0 -1 1 >> 0 -1 0 >> 1 -1 0 >> warning: symmetry operation # 4 not compatible >> with FFT grid. >> -1 0 0 >> -1 0 1 >> -1 1 0 >> warning: symmetry operation # 5 not compatible >> with FFT grid. >> >> What could be the cause for this and how to get rid >> of it? I ran >> calculations with c/a=0.9 and c/a=1-the warning >> didn't ocuur. But it is >> coming up with c/a=1 and c/a=1.2. >> >> Subhradip >> >> > ************************************************************************ >> Dr. Subhradip Ghosh >> Assistant Professor >> Department of Physics >> Indian Institute of Technology >> Guwahati,Assam-781039 >> India >> E-mail:subhra at iitg.ernet.in >> Phone: +91 361 2582717(O) >> +91 361 2584717(R) >> Fax: +91 361 2582749 (Physics) >> +91 361 2690762 (General) >> > ************************************************************************ >> >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> > > > ------------------------------------------------------------------- > Prof. Eyvaz Isaev, > Theoretical Physics Department, Moscow State Institute of Steel & Alloys, > Russia, > IFM, Linkoping University, Sweden > Condensed Matter Theory Group, Uppsala University, Sweden > Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com > > > ____________________________________________________________________________________ > Never miss a thing. Make Yahoo your home page. > http://www.yahoo.com/r/hs > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ************************************************************************ Dr. Subhradip Ghosh Assistant Professor Department of Physics Indian Institute of Technology Guwahati,Assam-781039 India E-mail:subhra at iitg.ernet.in Phone: +91 361 2582717(O) +91 361 2584717(R) Fax: +91 361 2582749 (Physics) +91 361 2690762 (General) ************************************************************************ From miguel.martinez at ehu.es Fri Feb 29 18:01:41 2008 From: miguel.martinez at ehu.es (=?ISO-8859-1?Q?Miguel_Mart=EDnez_Canales?=) Date: Fri, 29 Feb 2008 18:01:41 +0100 Subject: [Pw_forum] symmetry operation # not compatible with FFT grid In-Reply-To: <34755.172.16.109.56.1204303914.squirrel@webmail.iitg.ernet.in> References: <14895.97875.qm@web65714.mail.ac4.yahoo.com> <34755.172.16.109.56.1204303914.squirrel@webmail.iitg.ernet.in> Message-ID: <47C83A75.8060501@ehu.es> Subhradip Ghosh wrote: > Dear Eyvaz, > > Thanks. But, if I do not try to get rid of message, what could be the > consequences? The first that I could think of is that you are waisting CPU time for calculating unnecesary k-points. This also means that you'll have to work a bit with Vim if you want to get your fermi surface working on XCrySDen. There are probably more reasons, but there are also more knowledgeable people here. -- ---------------------------------------- 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 paulatto at sissa.it Fri Feb 29 18:05:21 2008 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Fri, 29 Feb 2008 18:05:21 +0100 (CET) Subject: [Pw_forum] symmetry operation # not compatible with FFT grid In-Reply-To: <34755.172.16.109.56.1204303914.squirrel@webmail.iitg.ernet.in> References: <14895.97875.qm@web65714.mail.ac4.yahoo.com> <34755.172.16.109.56.1204303914.squirrel@webmail.iitg.ernet.in> Message-ID: <10265.147.122.5.182.1204304721.squirrel@webmail.sissa.it> On Ven, 29 Febbraio 2008 5:51 pm, Subhradip Ghosh wrote: > Dear Eyvaz, > > Thanks. But, if I do not try to get rid of message, what could be the > consequences? None, but you may achieve a better precision while spending the same computational time if you do. -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) +39 040 3787 511 http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From eyvaz_isaev at yahoo.com Fri Feb 29 22:13:40 2008 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Fri, 29 Feb 2008 13:13:40 -0800 (PST) Subject: [Pw_forum] symmetry operation # not compatible with FFT grid In-Reply-To: <34755.172.16.109.56.1204303914.squirrel@webmail.iitg.ernet.in> Message-ID: <852864.9721.qm@web65706.mail.ac4.yahoo.com> Dear Subhradip, In general, you have been replyed by Lorenzo, and Miguel. Thanks them! I can just add that if you usually get rid of the message, you have correct number of symmetry elements which results in smaller number of k-points and, of course, faster convergence. Bests, Eyvaz. --- Subhradip Ghosh wrote: > Dear Eyvaz, > > Thanks. But, if I do not try to get rid of message, > what could be the > consequences? > > Subhradip > > > > > Dear Subhradip, > > > > This is really FAQ in the forum. So, please, > search > > the forum archive, and have a look at the > users-guide > > in /DOC (Page 59 if PDF file, ?warning: symmetry > > operation # N not allowed?) > > > > To get rid of this message you can choose manually > > nr1,nr2, nr3 in &system or increase ecutrho. > > > > Bests, > > Eyvaz. > > > > --- Subhradip Ghosh wrote: > > > >> I am trying to run self-consistent calculations > on a > >> heusler structure > >> with tetragonal distortion i.e. c/a=1.2. In the > very > >> beginning I am > >> getting the following warnings: > >> > >> > >> warning: symmetry operation # 2 not compatible > >> with FFT grid. > >> 0 1 -1 > >> 1 0 -1 > >> 0 0 -1 > >> warning: symmetry operation # 3 not > compatible > >> with FFT grid. > >> 0 -1 1 > >> 0 -1 0 > >> 1 -1 0 > >> warning: symmetry operation # 4 not > compatible > >> with FFT grid. > >> -1 0 0 > >> -1 0 1 > >> -1 1 0 > >> warning: symmetry operation # 5 not > compatible > >> with FFT grid. > >> > >> What could be the cause for this and how to get > rid > >> of it? I ran > >> calculations with c/a=0.9 and c/a=1-the warning > >> didn't ocuur. But it is > >> coming up with c/a=1 and c/a=1.2. > >> > >> Subhradip > >> > >> > > > ************************************************************************ > >> Dr. Subhradip Ghosh > >> Assistant Professor > >> Department of Physics > >> Indian Institute of Technology > >> Guwahati,Assam-781039 > >> India > >> E-mail:subhra at iitg.ernet.in > >> Phone: +91 361 2582717(O) > >> +91 361 2584717(R) > >> Fax: +91 361 2582749 (Physics) > >> +91 361 2690762 (General) > >> > > > ************************************************************************ > >> > >> > >> _______________________________________________ > >> Pw_forum mailing list > >> Pw_forum at pwscf.org > >> > http://www.democritos.it/mailman/listinfo/pw_forum > >> > > > > > > > ------------------------------------------------------------------- > > Prof. Eyvaz Isaev, > > Theoretical Physics Department, Moscow State > Institute of Steel & Alloys, > > Russia, > > IFM, Linkoping University, Sweden > > Condensed Matter Theory Group, Uppsala University, > Sweden > > Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com > > > > > > > ____________________________________________________________________________________ > > Never miss a thing. Make Yahoo your home page. > > http://www.yahoo.com/r/hs > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > ************************************************************************ > Dr. Subhradip Ghosh > Assistant Professor > Department of Physics > Indian Institute of Technology > Guwahati,Assam-781039 > India > E-mail:subhra at iitg.ernet.in > Phone: +91 361 2582717(O) > +91 361 2584717(R) > Fax: +91 361 2582749 (Physics) > +91 361 2690762 (General) > ************************************************************************ > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, IFM, Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com ____________________________________________________________________________________ Looking for last minute shopping deals? Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping From eyvaz_isaev at yahoo.com Fri Feb 29 23:25:50 2008 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Fri, 29 Feb 2008 14:25:50 -0800 (PST) Subject: [Pw_forum] Fermi Surface program In-Reply-To: <47C59F2A.3040105@nest.sns.it> Message-ID: <34903.52181.qm@web65709.mail.ac4.yahoo.com> Dear Paolo, Dear users, I have re-written (in F90 style) the Fermi Surface (FS) plot program so that now FS calculations for spin polarized cases are allowed. Actually I removed a part of the program for bands structure calculations "Calc_Type=Bands". I think this one deserves its own script. Input remains the same except "Calc_Type" line removed now. I apply also re-written script for Ni where one can find FS calculations for magnetic and nonmagnetic Ni. Please update these files in the QE distribution. Bests, Eyvaz. P.S. I have seen my responsibility for the QHA staff. Mostly it is ready to use. ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, IFM, Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com ____________________________________________________________________________________ Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs -------------- next part -------------- A non-text attachment was scrubbed... Name: bands_FS.f90 Type: text/x-fortran Size: 8639 bytes Desc: 3322185864-bands_FS.f90 Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080229/27f2cb44/attachment.bin -------------- next part -------------- A non-text attachment was scrubbed... Name: run_example Type: application/octet-stream Size: 10682 bytes Desc: 3506513086-run_example Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080229/27f2cb44/attachment.obj