From shrek_826 at yahoo.com.cn Mon Jun 2 04:06:22 2008 From: shrek_826 at yahoo.com.cn (Fan Yang) Date: Mon, 2 Jun 2008 10:06:22 +0800 (CST) Subject: [Pw_forum] About PWgui-4.0 and XCrySDen Message-ID: <110506.57140.qm@web15303.mail.cnb.yahoo.com> Thanks for Dr?Zhou's reply, I have?solved the installation of PWgui-4.0 and now it works fine. And also?many thanks to Madhura,?your suggestion is helpful. I have also tried the latest snapshot of?XCrySDen, it doesn't work. Regards ? ====================================== Fan Yang PH.D Candidate?in Electrochemistry????????????????????? College of Chemistry and Molecular Science??????????? Wuhan University,430072,Hubei Province,China? E-mail:shrek_826 at yahoo.com.cn ====================================== -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080602/9d73f1f6/attachment.htm From r95222066 at ntu.edu.tw Mon Jun 2 05:10:29 2008 From: r95222066 at ntu.edu.tw (r95222066 at ntu.edu.tw) Date: Mon, 02 Jun 2008 11:10:29 +0800 Subject: [Pw_forum] Did the wfc file must output during the calculation? Message-ID: <20080602111029.q7b4nyis088c08w0@wmail1.cc.ntu.edu.tw> Dear all, I had a little question about the wfc file. We all know during the igk.* file companyed with the calculation. I had a confused about did the wfc file must output during the calculation? Because the output wfc file was too big to disk space about 100G. That situation cost a lot of disk space. So I wondering to know that is there any parameter to control the output wfc file. Thanks a lot! Bing-Hong Chen,National Taiwan University,Taiwan From r95222066 at ntu.edu.tw Mon Jun 2 05:05:57 2008 From: r95222066 at ntu.edu.tw (r95222066 at ntu.edu.tw) Date: Mon, 02 Jun 2008 11:05:57 +0800 Subject: [Pw_forum] How to calculated about supercell of AFM structure? Message-ID: <20080602110557.yparsg56sgs8ockk@wmail1.cc.ntu.edu.tw> Dear all, I encountered a problem about AFM supercell setting.In one unit cell,the magnetism calculation is fine. But I try many possibility to calculated the supercell, it didn't work very well. Here is 1*1*1 uncell input file --------------------------------------------------------------------------- &control calculation='scf', restart_mode='from_scratch', pseudo_dir = './', outdir='./', prefix='SmOFeAs' / &system ibrav = 6,celldm(1)=7.626031,celldm(3)=2.1656 nat= 8, ntyp= 5,nbnd=140,nspin=2, ecutwfc = 60.0,ecutrho=300, starting_magnetization(1)=0.0, starting_magnetization(2)=0.0, starting_magnetization(3)=-0.3, starting_magnetization(3)=0.3, starting_magnetization(4)=0.0, occupations='smearing', smearing='mp', degauss=0.005, / &electrons conv_thr = 1.0e-5, mixing_beta = 0.05, / ATOMIC_SPECIES Sm 138.9055 Sm.pbe-nsp-van.UPF O 15.9994 O.pbe-van_ak.UPF Fe1 55.845 Fe.pbe-sp-van_ak.UPF Fe2 55.845 Fe.pbe-sp-van_ak.UPF As 74.92160 As.pbe-n-van.UPF ATOMIC_POSITIONS (crystal) Sm 0.25000000 0.25000000 0.14045310 Sm 0.75000000 0.75000000 0.85954689 O 0.75000000 0.25000000 0.00000000 O 0.25000000 0.75000000 0.00000000 Fe1 0.75000000 0.25000000 0.50000000 Fe2 0.25000000 0.75000000 0.50000000 As 0.25000000 0.25000000 0.63676752 As 0.75000000 0.75000000 0.36323247 K_POINTS (automatic) 12 12 6 0 0 0 ----------------------------------------------------------- Here is my 2*2*1 input file &control calculation='scf', restart_mode='from_scratch', pseudo_dir = './', outdir='./', prefix='SmOFeAs' / &system ibrav = 0, nat= 32, ntyp= 5,nbnd=320,nspin=2, ecutwfc = 60.0,ecutrho=300, starting_magnetization(1)=0.0, starting_magnetization(2)=0.0, starting_magnetization(3)=-0.3, starting_magnetization(3)=0.3, starting_magnetization(4)=0.0, occupations='smearing', smearing='mp', degauss=0.02, / &electrons conv_thr = 1.0e-4, mixing_beta = 0.05, / CELL_PARAMETERS (simple tetragonal) 15.252062 0.000000 0.000000 0.000000 15.252062 0.000000 0.000000 0.000000 16.514932 ATOMIC_SPECIES Sm 138.9055 Sm.pbe-nsp-van.UPF O 15.9994 O.pbe-van_ak.UPF Fe1 55.845 Fe.pbe-sp-van_ak.UPF Fe2 55.845 Fe.pbe-sp-van_ak.UPF As 74.92160 As.pbe-n-van.UPF ATOMIC_POSITIONS (angstrom) La 1.00889 1.00889 1.22746 La 1.00889 5.0444 1.22746 La 5.04440 1.00889 1.22746 La 5.04440 5.04440 1.22746 La 3.02664 3.02664 7.51186 La 3.02664 7.06216 7.51886 La 7.06216 3.02664 7.51886 La 7.06216 7.06216 7.51886 O 1.00889 3.02664 0.00000 O 1.00889 7.06216 0.00000 O 3.02664 1.00889 0.00000 O 3.02664 5.04440 0.00000 O 5.04440 3.02664 0.00000 O 5.04440 7.06216 0.00000 O 7.06216 1.00889 0.00000 O 7.06216 5.04440 0.00000 Fe1 1.00889 3.02664 4.36966 Fe2 1.00889 7.06216 4.36966 Fe1 3.02664 1.00889 4.36966 Fe2 3.02664 5.04440 4.36966 Fe2 5.04440 3.02664 4.36966 Fe1 5.04440 7.06216 4.36966 Fe2 7.06216 1.00889 4.36966 Fe1 7.06216 5.04440 4.36966 As 3.02664 3.02664 3.17440 As 3.02664 7.06216 3.17440 As 7.06216 3.02664 3.17440 As 7.06216 7.06216 3.17440 As 1.00889 1.00889 5.56491 As 1.00889 5.04440 5.56491 As 5.04440 1.00889 5.56491 As 5.04440 5.04440 5.56491 K_POINTS (automatic) 10 10 6 0 0 0 ---------------------------------------------------------------- Thanks a lot! Bing-Hong Chen,National Taiwan University,Taiwan From r95222066 at ntu.edu.tw Mon Jun 2 05:30:27 2008 From: r95222066 at ntu.edu.tw (r95222066 at ntu.edu.tw) Date: Mon, 02 Jun 2008 11:30:27 +0800 Subject: [Pw_forum] Convergence problem about AFM total magnetization? Message-ID: <20080602113027.lof8nj4xgkc4k0c4@wmail1.cc.ntu.edu.tw> Dear all, I had a question about the convergence problem of AFM total magnetization. The situation is the AFM total magnetization supposed to converge to "0" value. But the output file show "0.01" or "-0.01" value. I wondered to ask how to improve the situation like that ? Thanks a lot! Bing-Hong Chen, National Taiwan University, Taiwan From hqzhou at nju.edu.cn Mon Jun 2 09:41:30 2008 From: hqzhou at nju.edu.cn (Huiqun Zhou) Date: Mon, 2 Jun 2008 15:41:30 +0800 Subject: [Pw_forum] How to calculated about supercell of AFM structure? References: <20080602110557.yparsg56sgs8ockk@wmail1.cc.ntu.edu.tw> Message-ID: <002201c8c484$12f96300$a600a8c0@solarflare> At first glance of your input file, I think that the two lines below starting_magnetization(3)=0.3 starting_magnetization(4)=0.0 should be starting_magnetization(4)=0.3, starting_magnetization(5)=0.0 Huiqun Zhou @Nanjing University, China ----- Original Message ----- From: To: Sent: Monday, June 02, 2008 11:05 AM Subject: [Pw_forum] How to calculated about supercell of AFM structure? > Dear all, > I encountered a problem about AFM supercell setting.In one unit > cell,the magnetism calculation is fine. But I try many possibility to > calculated the supercell, it didn't work very well. > Here is 1*1*1 uncell input file > --------------------------------------------------------------------------- > > &control > calculation='scf', > restart_mode='from_scratch', > pseudo_dir = './', > outdir='./', > prefix='SmOFeAs' > / > &system > ibrav = 6,celldm(1)=7.626031,celldm(3)=2.1656 > nat= 8, ntyp= 5,nbnd=140,nspin=2, > ecutwfc = 60.0,ecutrho=300, > starting_magnetization(1)=0.0, > starting_magnetization(2)=0.0, > starting_magnetization(3)=-0.3, > starting_magnetization(3)=0.3, > starting_magnetization(4)=0.0, > occupations='smearing', smearing='mp', degauss=0.005, > / > &electrons > conv_thr = 1.0e-5, > mixing_beta = 0.05, > / > > ATOMIC_SPECIES > Sm 138.9055 Sm.pbe-nsp-van.UPF > O 15.9994 O.pbe-van_ak.UPF > Fe1 55.845 Fe.pbe-sp-van_ak.UPF > Fe2 55.845 Fe.pbe-sp-van_ak.UPF > As 74.92160 As.pbe-n-van.UPF > ATOMIC_POSITIONS (crystal) > Sm 0.25000000 0.25000000 0.14045310 > Sm 0.75000000 0.75000000 0.85954689 > O 0.75000000 0.25000000 0.00000000 > O 0.25000000 0.75000000 0.00000000 > Fe1 0.75000000 0.25000000 0.50000000 > Fe2 0.25000000 0.75000000 0.50000000 > As 0.25000000 0.25000000 0.63676752 > As 0.75000000 0.75000000 0.36323247 > K_POINTS (automatic) > 12 12 6 0 0 0 > ----------------------------------------------------------- > Here is my 2*2*1 input file > &control > calculation='scf', > restart_mode='from_scratch', > pseudo_dir = './', > outdir='./', > prefix='SmOFeAs' > / > &system > ibrav = 0, > nat= 32, ntyp= 5,nbnd=320,nspin=2, > ecutwfc = 60.0,ecutrho=300, > starting_magnetization(1)=0.0, > starting_magnetization(2)=0.0, > starting_magnetization(3)=-0.3, > starting_magnetization(3)=0.3, > starting_magnetization(4)=0.0, > occupations='smearing', smearing='mp', degauss=0.02, > / > &electrons > conv_thr = 1.0e-4, > mixing_beta = 0.05, > / > CELL_PARAMETERS (simple tetragonal) > 15.252062 0.000000 0.000000 > 0.000000 15.252062 0.000000 > 0.000000 0.000000 16.514932 > ATOMIC_SPECIES > Sm 138.9055 Sm.pbe-nsp-van.UPF > O 15.9994 O.pbe-van_ak.UPF > Fe1 55.845 Fe.pbe-sp-van_ak.UPF > Fe2 55.845 Fe.pbe-sp-van_ak.UPF > As 74.92160 As.pbe-n-van.UPF > ATOMIC_POSITIONS (angstrom) > La 1.00889 1.00889 1.22746 > La 1.00889 5.0444 1.22746 > La 5.04440 1.00889 1.22746 > La 5.04440 5.04440 1.22746 > La 3.02664 3.02664 7.51186 > La 3.02664 7.06216 7.51886 > La 7.06216 3.02664 7.51886 > La 7.06216 7.06216 7.51886 > O 1.00889 3.02664 0.00000 > O 1.00889 7.06216 0.00000 > O 3.02664 1.00889 0.00000 > O 3.02664 5.04440 0.00000 > O 5.04440 3.02664 0.00000 > O 5.04440 7.06216 0.00000 > O 7.06216 1.00889 0.00000 > O 7.06216 5.04440 0.00000 > Fe1 1.00889 3.02664 4.36966 > Fe2 1.00889 7.06216 4.36966 > Fe1 3.02664 1.00889 4.36966 > Fe2 3.02664 5.04440 4.36966 > Fe2 5.04440 3.02664 4.36966 > Fe1 5.04440 7.06216 4.36966 > Fe2 7.06216 1.00889 4.36966 > Fe1 7.06216 5.04440 4.36966 > As 3.02664 3.02664 3.17440 > As 3.02664 7.06216 3.17440 > As 7.06216 3.02664 3.17440 > As 7.06216 7.06216 3.17440 > As 1.00889 1.00889 5.56491 > As 1.00889 5.04440 5.56491 > As 5.04440 1.00889 5.56491 > As 5.04440 5.04440 5.56491 > K_POINTS (automatic) > 10 10 6 0 0 0 > ---------------------------------------------------------------- > > 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 > From sauluck at iitk.ac.in Mon Jun 2 12:41:05 2008 From: sauluck at iitk.ac.in (Prof. Sushil Auluck) Date: Mon, 2 Jun 2008 16:11:05 +0530 (IST) Subject: [Pw_forum] Convergence problem about AFM total magnetization? In-Reply-To: <20080602113027.lof8nj4xgkc4k0c4@wmail1.cc.ntu.edu.tw> References: <20080602113027.lof8nj4xgkc4k0c4@wmail1.cc.ntu.edu.tw> Message-ID: <1336.122.161.84.179.1212403265.squirrel@newwebmail.iitk.ac.in> hi, if you make the convergence criteria more stringent you will get moment very close to zero. i.e. if your convergence ws 0.001 then make it 0.0001 or even smaller. s.auluck > Dear all, > I had a question about the convergence problem of AFM total magnetization. > The situation is the AFM total magnetization supposed to converge to > "0" value. > But the output file show "0.01" or "-0.01" value. > I wondered to ask how to improve the situation like that ? > > 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. Sushil Auluck Phone: +91-512-2597092(Work) Department of Physics +91-512-2596148(Work) Indian Institute of Technology- Mobile:+91-9305548667 Kanpur Fax :+91-512-2590914 Kanpur 208016 (UP) e-mail: sauluck at iitk.ac.in India e-mail: sauluck at gmail.com http://www.iitk.ac.in/phy/New01/profile_SA.html http://www.iitk.ac.in/phy/People/phy_facvis.html ************************************************************************** From tone.kokalj at ijs.si Mon Jun 2 14:46:37 2008 From: tone.kokalj at ijs.si (Tone Kokalj) Date: Mon, 02 Jun 2008 14:46:37 +0200 Subject: [Pw_forum] About PWgui-4.0 and XCrySDen In-Reply-To: <674993.61975.qm@web15303.mail.cnb.yahoo.com> References: <674993.61975.qm@web15303.mail.cnb.yahoo.com> Message-ID: <1212410797.5020.18.camel@walk.ijs.si> On Sat, 2008-05-31 at 10:30 +0800, Fan Yang wrote: > > Dear everyone, > > I have two questions about the additional tools for > espresso-4.0. First is the compilation of PWgui-4.0, namely, when > executing ./pwui > I got error message: Can't find package Itcl while executing. PWgui (source package, i.e. the one available on-line) needs some additional software, in particular: Tcl/Tk and Itcl/Itk/Iwidgets (i.e. full names are [Incr Tcl]/[Incr Tk]/[Incr widgets]). For example, in Debian-based linux distributions this can be installed by (as root or sudo): apt-get install tcl tk itcl3 itk3 iwidgets4 Alternatively, you may use the standalone executable: 64 bit, linux: http://www-k3.ijs.si/kokalj/pwgui/download/pwgui-linux-x86_64.zip 32 bit, linux: http://www-k3.ijs.si/kokalj/pwgui/download/pwgui-linux-x86.zip > > The second is about XCrySDen, when runing espresso-4.0, it > seems that only the initial structure could be displayed, while the > subsequent configurations could not be imported. If I use the previous > version, espresso-3.2.3, then everything goes well. It is due to change in output format of QE-4.0. The version 1.4 cannot handle it, but the "latest snapshot" can. If you use linux, you may try this: http://www.xcrysden.org/download/xc-latest-linuxPC-semishared.tar.gz Regards, Tone -- Tone Kokalj J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia (tel: +386-1-477-3523 // fax:+386-1-477-3822) From wangqj1 at 126.com Tue Jun 3 10:08:29 2008 From: wangqj1 at 126.com (wangqj1) Date: Tue, 3 Jun 2008 16:08:29 +0800 (CST) Subject: [Pw_forum] charge density difference 1 Message-ID: <1366051.244211212480509946.JavaMail.coremail@bj126app29.126.com> Dear pwscf users I want to plot the charge density difference of ZnO on (1 1 0) surface ,but when I open the inputpp ,I meet some troubles. first, I don't understand the 'filplot',how to set it , I know it from the self-consistent,but in self-consistent which file is it ? second , I know that 'nfile' is the number of files will be dealed with ,but I don't know how to set the filepp(1)=? filepp(2)=? ...(nfile>1) Could somebody tell me how to set 'filplot'and filpp( ) ? It will be good if give me the example of charge density difference of ZnO on (1 1 0) surface . Everybody's advice will be appreciate . thanks -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080603/30dfe697/attachment.htm From jibiaoli at gmail.com Tue Jun 3 11:07:38 2008 From: jibiaoli at gmail.com (Clark Lee) Date: Tue, 3 Jun 2008 17:07:38 +0800 Subject: [Pw_forum] Vibration Spectrum from QE Message-ID: Hi all, I am eager to know detailed steps to calculate vibration spectrum,particularlly, those of supported molecules, from QE. Has methods of calculating vibration spectrum implemented in QE? How? -- Yours sincerely, Jibiao Li State Key Lab of Corrosion and Protection (SKLCP) Institute of Metal Research (IMR) Chinese Academy of Sciences (CAS), China Email: jibiaoli at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080603/67c7c11e/attachment.htm From yccheng.nju at gmail.com Tue Jun 3 11:17:44 2008 From: yccheng.nju at gmail.com (=?UTF-8?B?56iL6L+O5pil?=) Date: Tue, 3 Jun 2008 17:17:44 +0800 Subject: [Pw_forum] Vibration Spectrum from QE In-Reply-To: References: Message-ID: Dear Jibiao, Your questions are very general and hard to answer. I think you should read the manual and the input files in Doc directory, and then ask some special questions. 2008/6/3, Clark Lee : > > Hi all, > I am eager to know detailed steps to calculate vibration > spectrum,particularlly, those of supported molecules, from QE. > Has methods of calculating vibration spectrum implemented in QE? How? > > -- > Yours sincerely, > > Jibiao Li > > State Key Lab of Corrosion and Protection (SKLCP) > Institute of Metal Research (IMR) > Chinese Academy of Sciences (CAS), China > Email: jibiaoli at gmail.com > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- Y. C. Cheng Department of Phyics Nanjing University Nanjing 210093 P. R. China Tel: 86-25-83592907 Email: yccheng.nju at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080603/1014b124/attachment.htm From hande at newton.physics.metu.edu.tr Tue Jun 3 11:50:55 2008 From: hande at newton.physics.metu.edu.tr (Hande Ustunel) Date: Tue, 3 Jun 2008 12:50:55 +0300 (EEST) Subject: [Pw_forum] charge density difference 1 In-Reply-To: <1366051.244211212480509946.JavaMail.coremail@bj126app29.126.com> Message-ID: Hi, I'm assuming you've read the file INPUT_PP under the Doc directory of your espresso distribution. The meaning of the parameters are explained in there. But to clarify, the &inputpp structure produces a file designated by you contained in the filplot variable, so you can call this any name you wish. The program will read the relevant quantity (say density) from your outdir with your prefix, convert it into a format that is suitable for further processing and write it into the file you have named in the variable filplot. You should give separate names to all the densities you want to take into account in the difference calculation. Once you have all the density files, you should use filepp(n) to read in those files that you have named previously with the correct weights. For instance if your system is a surface+adsorbant, you may consider three different calculations with filplot = 'surface+ad' filplot = 'surface' filplot = 'ad' Then when it's time to take the difference you could do filepp(1)='surface+ad' filepp(2)='surface' filepp(3)='ad' with the obvious weights weight(1)=1 weight(2)=-1 weight(3)=-1 Make sure to be consistent in your naming scheme. Hope this helps, Hande On Tue, 3 Jun 2008, wangqj1 wrote: > > Dear pwscf users > I want to plot the charge density difference of ZnO on (1 1 0) surface ,but when I open the inputpp ,I meet some troubles. > first, I don't understand the 'filplot',how to set it , I know it from the self-consistent,but in self-consistent which file is it ? > second , I know that 'nfile' is the number of files will be dealed with ,but I don't know how to set the filepp(1)=? filepp(2)=? ...(nfile>1) > Could somebody tell me how to set 'filplot'and filpp( ) ? It will be good if give me the example of charge density difference of ZnO on (1 1 0) surface . > Everybody's advice will be appreciate . > thanks > -- Hande Ustunel Department of Physics Office 439 Middle East Technical University Ankara 06531, Turkey Tel : +90 312 210 3264 http://www.physics.metu.edu.tr/~hande From eyvaz_isaev at yahoo.com Tue Jun 3 12:55:00 2008 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Tue, 3 Jun 2008 03:55:00 -0700 (PDT) Subject: [Pw_forum] Vibration Spectrum from QE In-Reply-To: Message-ID: <99208.38670.qm@web65713.mail.ac4.yahoo.com> Please have a look at example06 in /examples. Of course, you should use a supercell, calculate phonons and then projected phonon dos. If you mean to calculate vibrations of selected atoms I think you can try "nat_todo" keyword for phonons, but not sure it will work in your particular case, as this keyword has been used for a special structure (fulleren, if I remember correctly from Paolo's reply). By the way, in this way you will never be sure whether a structure you consider is dynamically stable as you do not have a complete picture for pohonon dispersions. Bests, Eyvaz. --- Clark Lee wrote: > Hi all, > I am eager to know detailed steps to calculate > vibration > spectrum,particularlly, those of supported > molecules, from QE. > Has methods of calculating vibration spectrum > implemented in QE? How? > > -- > Yours sincerely, > > Jibiao Li > > State Key Lab of Corrosion and Protection (SKLCP) > Institute of Metal Research (IMR) > Chinese Academy of Sciences (CAS), China > Email: jibiaoli at gmail.com > > _______________________________________________ > 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, Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com From sei.ou0722 at gmail.com Wed Jun 4 06:39:25 2008 From: sei.ou0722 at gmail.com (sei ou) Date: Wed, 4 Jun 2008 13:39:25 +0900 Subject: [Pw_forum] Convergence Problem on dipole correction and XCrysden problem on X86_64 Message-ID: <161e5bd50806032139p4910552dj69f58643d8fa9e35@mail.gmail.com> Dear all, Would you please answer my following questions? (1) ZnO polar surface simulation using dipole correction: Recently I performed test calculation on ZnO polar surface, when using dipole correction, the convergence is very difficult to achieve, even in a single scf cycle. Could anyone tell me where is wrong in my attached input file? (2) Xcrysden installation problem on 64 bit machine. Xcrysden-pre1.5 source code can be compiled on Ubuntu 8.04 x86_64, but when I run xcrysden, it always crashs. by using xcrysden -d, I got the follwing error messages: This GDB was configured as "x86_64-linux-gnu"... [Thread debugging using libthread_db enabled] [New Thread 0x7f1ab074b6f0 (LWP 9336)] [New Thread 0x40ac4950 (LWP 9339)] Running on platform: unix Taking settings for 1280x1024 screen-resolution !!! ------------------- Small Font Size set to: 10 Normal Font Size set to: 12 Executing: ~/linux/XCrySDen-pre1.5bs-all/bin/ftnunit Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 0x7f1ab074b6f0 (LWP 9336)] 0x0000000000746e69 in ?? () Where is the problem, also when compiling the code, there are warning messages on 'imcompatible pointer type'. Is there a 64 bit binary files on version 1.5? Thank you in advance Sei Ou Simulatio Co. Yokohama Japan -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080604/2c60ff2f/attachment.htm -------------- next part -------------- A non-text attachment was scrubbed... Name: 000-1.ml1-8dl.inp Type: chemical/x-gamess-input Size: 1965 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080604/2c60ff2f/attachment.inp From huangchunlin1984 at 163.com Wed Jun 4 06:35:32 2008 From: huangchunlin1984 at 163.com (huangchunlin1984) Date: Wed, 4 Jun 2008 12:35:32 +0800 (CST) Subject: [Pw_forum] error : unable to creat directory *.save Message-ID: <16100290.785751212554132369.JavaMail.coremail@bj163app106.163.com> Dear Doctor I have installed the Quantum Espresso 4.0. I encountered a error as follows when running the example 03 and other input files: error: from create_directory : error #4294967295 unable to create directory ./FeNi1.save thanks for any help! student h -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080604/4e2abaad/attachment.htm From zhaohscas at yahoo.com.cn Wed Jun 4 06:49:29 2008 From: zhaohscas at yahoo.com.cn (zhaohscas) Date: Wed, 4 Jun 2008 12:49:29 +0800 Subject: [Pw_forum] Convergence Problem on dipole correction and XCrysdenproblem on X86_64 References: <161e5bd50806032139p4910552dj69f58643d8fa9e35@mail.gmail.com> Message-ID: <200806041249264475662@yahoo.com.cn> >(2) Xcrysden installation problem on 64 bit machine. >Xcrysden-pre1.5 source code can be compiled on Ubuntu 8.04 x86_64, but when >I run xcrysden, it always crashs. >by using xcrysden -d, I got the follwing error messages: > [bla bla bla] I recommend you post the Xcrysden's install-and-compile relative questions on the Xcrysden's list. Perhaps there're so many experts in solving the things you meet :-) Sincerely yours, Hongsheng. __________________________________________________ ??????????????? http://cn.mail.yahoo.com From zhaohscas at yahoo.com.cn Wed Jun 4 06:49:29 2008 From: zhaohscas at yahoo.com.cn (zhaohscas) Date: Wed, 4 Jun 2008 12:49:29 +0800 Subject: [Pw_forum] Convergence Problem on dipole correction and XCrysdenproblem on X86_64 References: <161e5bd50806032139p4910552dj69f58643d8fa9e35@mail.gmail.com> Message-ID: <200806041249264475662@yahoo.com.cn> >(2) Xcrysden installation problem on 64 bit machine. >Xcrysden-pre1.5 source code can be compiled on Ubuntu 8.04 x86_64, but when >I run xcrysden, it always crashs. >by using xcrysden -d, I got the follwing error messages: > [bla bla bla] I recommend you post the Xcrysden's install-and-compile relative questions on the Xcrysden's list. Perhaps there're so many experts in solving the things you meet :-) Sincerely yours, Hongsheng. __________________________________________________ ??????????????? http://cn.mail.yahoo.com From huangchunlin1984 at 163.com Wed Jun 4 06:47:49 2008 From: huangchunlin1984 at 163.com (huangchunlin1984) Date: Wed, 4 Jun 2008 12:47:49 +0800 (CST) Subject: [Pw_forum] unable to create directory prefix.save Message-ID: <26372581.790931212554869925.JavaMail.coremail@bj163app106.163.com> Dear Doctor I have installed the 4.0.when running ,I have encounted a error as follows : from create_directory : error #4294967295 unable to create directory ./FeNi1.save thanks for any help! student h -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080604/d006e2d6/attachment.htm From wangqj1 at 126.com Wed Jun 4 08:17:36 2008 From: wangqj1 at 126.com (wangqj1) Date: Wed, 4 Jun 2008 14:17:36 +0800 (CST) Subject: [Pw_forum] charge density difference 1 Message-ID: <15990537.601131212560256765.JavaMail.coremail@bj126app27.126.com> Thanks for your advice about charge density difference . but I still have a problem : whether I get the filepp(1)='ZnOcharge',filepp(2)='ZnOcharge',filepp(3)='Ocharge'in one 'scf' or I must do them seperately (i.e,three times) ? thank you ! -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080604/051cbe95/attachment.htm From hande at newton.physics.metu.edu.tr Wed Jun 4 09:10:27 2008 From: hande at newton.physics.metu.edu.tr (Hande Ustunel) Date: Wed, 4 Jun 2008 10:10:27 +0300 (EEST) Subject: [Pw_forum] charge density difference 1 In-Reply-To: <15990537.601131212560256765.JavaMail.coremail@bj126app27.126.com> Message-ID: Because they are different systems, you should first do three scf calculations : One with the combined system and two with the components. Make sure the components are located in the same location as they are in the combined system. Best, Hande On Wed, 4 Jun 2008, wangqj1 wrote: > Thanks for your advice about charge density difference . > but I still have a problem : > whether I get the filepp(1)='ZnOcharge',filepp(2)='ZnOcharge',filepp(3)='Ocharge'in one 'scf' or I must do them seperately (i.e,three times) ? > thank you ! > > > > -- Hande Ustunel Department of Physics Office 439 Middle East Technical University Ankara 06531, Turkey Tel : +90 312 210 3264 http://www.physics.metu.edu.tr/~hande From baroni at sissa.it Wed Jun 4 09:21:16 2008 From: baroni at sissa.it (Stefano Baroni) Date: Wed, 4 Jun 2008 09:21:16 +0200 Subject: [Pw_forum] Convergence Problem on dipole correction and XCrysdenproblem on X86_64 In-Reply-To: <200806041249264475662@yahoo.com.cn> References: <161e5bd50806032139p4910552dj69f58643d8fa9e35@mail.gmail.com> <200806041249264475662@yahoo.com.cn> Message-ID: nothing wrong in posting xcrysden-related questions on this list SB On Jun 4, 2008, at 6:49 AM, zhaohscas wrote: >> (2) Xcrysden installation problem on 64 bit machine. >> Xcrysden-pre1.5 source code can be compiled on Ubuntu 8.04 x86_64, >> but when >> I run xcrysden, it always crashs. >> by using xcrysden -d, I got the follwing error messages: >> [bla bla bla] > > I recommend you post the Xcrysden's install-and-compile relative > questions on the Xcrysden's list. > > Perhaps there're so many experts in solving the things you meet :-) > > Sincerely yours, > Hongsheng. > > __________________________________________________ > ??????????????? > http://cn.mail.yahoo.com > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget 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/20080604/2b71f8da/attachment.htm From paulatto at sissa.it Wed Jun 4 10:03:10 2008 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Wed, 4 Jun 2008 10:03:10 +0200 (CEST) Subject: [Pw_forum] unable to create directory prefix.save In-Reply-To: <26372581.790931212554869925.JavaMail.coremail@bj163app106.163.com> References: <26372581.790931212554869925.JavaMail.coremail@bj163app106.163.com> Message-ID: <22779.82.48.141.31.1212566590.squirrel@webmail.sissa.it> On Wed, June 4, 2008 06:47, huangchunlin1984 wrote: > Dear Doctor > I have installed the 4.0.when running ,I have encounted a error as > follows : > from create_directory : error #4294967295 > unable to create directory ./FeNi1.save Dear Huang Chun Lin, you have to provide more detail to receive any help! How can we know why directory creation fails? Have you tried to create the directory by hand? Is it possible? Does a file with that name (FeNi1.save) exists? Are you running on a single computer or parallel? Have you run the same job with QE 3.2.*, and haven't cleaned the tmp files? -- 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 tone.kokalj at ijs.si Wed Jun 4 10:12:24 2008 From: tone.kokalj at ijs.si (Tone Kokalj) Date: Wed, 04 Jun 2008 10:12:24 +0200 Subject: [Pw_forum] Convergence Problem on dipole correction and XCrysden problem on X86_64 In-Reply-To: <161e5bd50806032139p4910552dj69f58643d8fa9e35@mail.gmail.com> References: <161e5bd50806032139p4910552dj69f58643d8fa9e35@mail.gmail.com> Message-ID: <1212567144.19067.5.camel@walk.ijs.si> On Wed, 2008-06-04 at 13:39 +0900, sei ou wrote: > (2) Xcrysden installation problem on 64 bit machine. > Xcrysden-pre1.5 source code can be compiled on Ubuntu 8.04 x86_64, but > when I run xcrysden, it always crashs. > by using xcrysden -d, I got the follwing error messages: There are some very tricky issues with the 64bit compilation of xcrysden. However, the 32bit compiled code works well on 64bit machines. I suggest you try with the following: http://www.xcrysden.org/download/xc-latest-linuxPC-semishared.tar.gz (and if for some reason this will not work, you may try with: http://www.xcrysden.org/download/xc-latest-linuxPC-static.tar.gz http://www.xcrysden.org/download/xc-latest-linuxPC-shared.tar.gz ) Regards, Tone -- Tone Kokalj J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia (tel: +386-1-477-3523 // fax:+386-1-477-3822) From zhaohscas at yahoo.com.cn Wed Jun 4 10:17:04 2008 From: zhaohscas at yahoo.com.cn (zhaohscas) Date: Wed, 4 Jun 2008 16:17:04 +0800 Subject: [Pw_forum] About the visualization of charge density plot on arbitrary Miller plane of general triclinic crystal system. References: <200805300018388129944@yahoo.com.cn> Message-ID: <200806041616509645551@yahoo.com.cn> Hi forks, I have noticed that the Xcrysden only can visualize the charge density plot on the (100), (010), or (001) Miller plane, not the arbitrary ones. If I want to visualize the charge density plot on arbitrary Miller plane of general triclinic crystal system. Who can give some hints on this thing? Any help would be appreciated, Sincerely yours, Hongsheng. __________________________________________________ ??????????????? http://cn.mail.yahoo.com From sei.ou0722 at gmail.com Wed Jun 4 10:47:37 2008 From: sei.ou0722 at gmail.com (sei ou) Date: Wed, 4 Jun 2008 17:47:37 +0900 Subject: [Pw_forum] Convergence Problem on dipole correction and XCrysden problem on X86_64 In-Reply-To: <1212567144.19067.5.camel@walk.ijs.si> References: <161e5bd50806032139p4910552dj69f58643d8fa9e35@mail.gmail.com> <1212567144.19067.5.camel@walk.ijs.si> Message-ID: <161e5bd50806040147mf3f7e8elf637d8981308cfae@mail.gmail.com> Dear Tone, Thank you very much! Xcrysden can work now after downloading the binary files. Regards, Sei Ou 2008/6/4 Tone Kokalj : > On Wed, 2008-06-04 at 13:39 +0900, sei ou wrote: > > > (2) Xcrysden installation problem on 64 bit machine. > > Xcrysden-pre1.5 source code can be compiled on Ubuntu 8.04 x86_64, but > > when I run xcrysden, it always crashs. > > by using xcrysden -d, I got the follwing error messages: > > There are some very tricky issues with the 64bit compilation of > xcrysden. However, the 32bit compiled code works well on 64bit machines. > I suggest you try with the following: > > http://www.xcrysden.org/download/xc-latest-linuxPC-semishared.tar.gz > > (and if for some reason this will not work, you may try with: > http://www.xcrysden.org/download/xc-latest-linuxPC-static.tar.gz > http://www.xcrysden.org/download/xc-latest-linuxPC-shared.tar.gz > ) > > Regards, Tone > -- > Tone Kokalj > J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia (tel: > +386-1-477-3523 // fax:+386-1-477-3822) > > _______________________________________________ > 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/20080604/2ed7c128/attachment.htm From tone.kokalj at ijs.si Wed Jun 4 11:03:28 2008 From: tone.kokalj at ijs.si (Tone Kokalj) Date: Wed, 04 Jun 2008 11:03:28 +0200 Subject: [Pw_forum] About the visualization of charge density plot on arbitrary Miller plane of general triclinic crystal system. In-Reply-To: <200806041616509645551@yahoo.com.cn> References: <200805300018388129944@yahoo.com.cn> <200806041616509645551@yahoo.com.cn> Message-ID: <1212570208.19067.18.camel@walk.ijs.si> On Wed, 2008-06-04 at 16:17 +0800, zhaohscas wrote: > Hi forks, > > I have noticed that the Xcrysden only can visualize the charge density plot on the (100), (010), or (001) Miller plane, not the arbitrary ones. > > If I want to visualize the charge density plot on arbitrary Miller > plane of general triclinic crystal system. Who can give some hints on this thing? You need to calculate 2D mesh of points by pp.x that lies on the plane of your interest. Then xcrysden will visualize contours on this pre-calculated plane. Regars, Tone > > Any help would be appreciated, > Sincerely yours, > Hongsheng. > > __________________________________________________ > ??????????????? > http://cn.mail.yahoo.com > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- Tone Kokalj J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia (tel: +386-1-477-3523 // fax:+386-1-477-3822) From zhaohscas at yahoo.com.cn Wed Jun 4 11:55:11 2008 From: zhaohscas at yahoo.com.cn (zhaohscas) Date: Wed, 4 Jun 2008 17:55:11 +0800 Subject: [Pw_forum] About the visualization of charge density ploton arbitrary Miller plane of general triclinic crystal References: <200805300018388129944@yahoo.com.cn>, <200806041616509645551@yahoo.com.cn> Message-ID: <200806041755076060408@yahoo.com.cn> >You need to calculate 2D mesh of points by pp.x that lies on the plane >of your interest. Then xcrysden will visualize contours on this >pre-calculated plane. Because of my machine have some issue in installing the linux OS, currently, I only have the cygwin and the xcrysden on windows, would you like give me a small examle xsf data of the you obtained of charge density ploton arbitrary Miller plane? Regars, Hongsheng. __________________________________________________ ??????????????? http://cn.mail.yahoo.com From chuchun.fu at cea.fr Wed Jun 4 14:02:49 2008 From: chuchun.fu at cea.fr (chu-chun fu) Date: Wed, 04 Jun 2008 14:02:49 +0200 Subject: [Pw_forum] post-proccesing using pwscf4.0 and PAW Message-ID: <48468469.3010700@cea.fr> Dear PWscf dvelopers, I am using the new released PWscf-4.0 within quantum espresso. In particular the new implemented PAW option to study FeCr system. I realized that to compute density of electronic states DOS and projected ones PDOS among the post-procceding utilities, there is format incompability when performing PAW calculations. I found that the problem comes from calling the file_data.xml file and reading the 'pseudo' UPF file. I will appreciate if you could fix this problem if possible ? Kind Regards Chu Chun FU -- Chu Chun FU DEN/DMN/SRMP CEA-Saclay 91191 Gif-sur-yvette Cedex France Tel: +33 1 69 08 29 32 Fax: +33 1 69 08 68 67 E-mail: chuchun.fu at cea.fr From giannozz at democritos.it Wed Jun 4 14:56:25 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Wed, 04 Jun 2008 14:56:25 +0200 Subject: [Pw_forum] Did the wfc file must output during the calculation? In-Reply-To: <20080602111029.q7b4nyis088c08w0@wmail1.cc.ntu.edu.tw> References: <20080602111029.q7b4nyis088c08w0@wmail1.cc.ntu.edu.tw> Message-ID: <484690F9.1040207@democritos.it> r95222066 at ntu.edu.tw wrote: > I had a little question about the wfc file. We all know during the > igk.* file companyed with the calculation. do we? > Because the output > wfc file was too big to disk space about 100G. That situation cost a > lot of disk space. So I wondering to know that is there any parameter > to control the output wfc file. there are several parameters controlling disk space requirements: http://www.quantum-espresso.org/wiki/index.php/Performance_issues This will also be useful: http://www.quantum-espresso.org/wiki/index.php/Running_on_parallel_machines#Understanding_parallel_I.2FO Note that if your wavefunction file is 100Gb, you will need at least 100Gb RAM (total, i.e. summed over the processors) Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From jlz101 at unlserve.unl.edu Wed Jun 4 13:56:35 2008 From: jlz101 at unlserve.unl.edu (J. D. Burton) Date: Wed, 4 Jun 2008 06:56:35 -0500 Subject: [Pw_forum] generating a La US-PP with 4f valence Message-ID: <000001c8c63a$0a11ca40$0200a8c0@BurLap> Dear everyone, I am trying to apply LDA+U to the unoccupied 4f states of La, but the US-PP for La on the pwscf website (La.pbe-nsp-van.UPF) is generated with maximum angular momentum component = 2. I need to do this to push the unoccupied La 4f states further up in energy away from the Ti d states in LaTiO3. The hybridization between La 4f and Ti d states in this material is unphysical, and causes problems. Anyway... I've been trying to use the ld1.x program to generate such a creature but I've had no success. The main problem seems to have nothing to do with the 4f states, but rather the "semicore" 5s and 5p states which must be included. Here is my input file: &input title='La', zed=57., rel=1, config='[Xe] 5d1.0 6s1.5 6p0.25 4f0.25', iswitch=3, dft='PZ', verbosity='high', / &inputp zval=11, lloc=-1, rcloc=3.0, pseudotype=3, file_pseudopw='La.pbe-rrkjus.UPF', author='JDB', / 6 5S 1 0 2.00 0.00 2.20 2.20 5P 2 1 6.00 0.00 2.00 2.00 6S 3 0 1.50 0.00 2.20 2.20 5D 4 2 1.00 0.00 2.20 2.20 4F 5 3 0.25 0.00 2.20 2.20 6P 6 1 0.25 0.00 2.00 2.00 I took the cutoff radii directly from the original La.pbe-nsp-van.UPF file, thinking they should work just fine, but... no luck. The all electron calculation seems to run just fine, but the pseudo-generator gives the following error messages: Wfc 5P rcut= 1.978 Estimated cut-off energy= 45.51 Ry This function has 0 nodes for 0 < r < 1.978 Message from routine compute phi: negative determinant ld= 2.977509 f2ae 0.324572 faenor 0.048542 0 1254 1253 100.178012872869 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% from compute_chi : error # 1 n is too large %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% Could someone tell me what I am doing wrong here? Thanks, J. D. **************************** J. D. Burton, Ph.D. jlz101 at unlserve.unl.edu University of Nebraska Lincoln Physics and Astronomy Office Ph. (402) 472 0546 214CC Ferguson Hall **************************** "The job of a scientist is to generate wrong ideas as fast as possible." -- Murray Gell-Mann From bnrj.rudra at yahoo.com Wed Jun 4 16:02:13 2008 From: bnrj.rudra at yahoo.com (Rudra Banerjee) Date: Wed, 4 Jun 2008 19:32:13 +0530 (IST) Subject: [Pw_forum] Did the wfc file must output during the calculation? Message-ID: <34619.24809.qm@web94112.mail.in2.yahoo.com> there are several parameters controlling disk space requirements: http://www.quantum-espresso.org/wiki/index.php/Performance_issues This link is still blank -- Rudra JRF; SNBNCBS http://www.bose.res.in/~rudra A bus station is where a bus stops. A train station is where a train stops. On my desk I have a work station. Please, if possible, don't send me MS Word or PowerPoint attachments Why?See: http://www.gnu.org/philosophy/no-word-attachments.html Explore your hobbies and interests. Go to http://in.promos.yahoo.com/groups/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080604/2783e056/attachment.htm From andrea.ferretti at unimore.it Wed Jun 4 16:08:38 2008 From: andrea.ferretti at unimore.it (Andrea Ferretti) Date: Wed, 4 Jun 2008 16:08:38 +0200 (CEST) Subject: [Pw_forum] unable to create directory prefix.save Message-ID: Dear Huang Chun Lin, > > from create_directory : error #4294967295 > unable to create directory ./FeNi1.save looking at the error number, which looks like an overflow, it may be that your problem is related to a mismatch between the default integer of fortran and the type "int" of c... let me exaplin better: such error message is given after a call to a c-interface (c_mkdir); the output of the routine in c is defined as int retval; and it is passed to a default integer fortran variable (INTEGER :: ierr) in the routine create_directory. Thus, your problem can be related to the usage of 8byte default integer when compiling the fortran code (I met the a similar problem in the past). To patch it, try to avoid the use of any -i8 flags or, if you have downloaded a compiler with i8 default integers, try to force -i4 or similar... hope it helps andrea -- Andrea Ferretti National Research Center S3, CNR-INFM ( http://s3.infm.it ) Dip. Fisica, Univ. di Modena e Reggio E. (Italy) Tel: +39 059 2055301; Fax: +39 059 374794; Skype: andrea_ferretti URL: http://www.nanoscience.unimo.it Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html From degironc at sissa.it Wed Jun 4 16:08:16 2008 From: degironc at sissa.it (Stefano de Gironcoli) Date: Wed, 04 Jun 2008 16:08:16 +0200 Subject: [Pw_forum] Did the wfc file must output during the calculation? In-Reply-To: <34619.24809.qm@web94112.mail.in2.yahoo.com> References: <34619.24809.qm@web94112.mail.in2.yahoo.com> Message-ID: <4846A1D0.5060903@sissa.it> try http://www.quantum-espresso.org/wiki/index.php/Performance_issues_(PWscf) Rudra Banerjee wrote: > there are several parameters controlling disk space requirements: > http://www.quantum-espresso.org/wiki/index.php/Performance_issues > This link is still blank > -- > Rudra > JRF; SNBNCBS > http://www.bose.res.in/~rudra > > > A bus station is where a bus stops. A train station is where a train > stops. On my desk I have a work station. > > Please, if possible, don't send me MS Word or PowerPoint attachments > Why?See: http://www.gnu.org/philosophy/no-word-attachments.html > > > > > ------------------------------------------------------------------------ > 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 giannozz at democritos.it Wed Jun 4 16:09:52 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Wed, 04 Jun 2008 16:09:52 +0200 Subject: [Pw_forum] Did the wfc file must output during the calculation? In-Reply-To: <34619.24809.qm@web94112.mail.in2.yahoo.com> References: <34619.24809.qm@web94112.mail.in2.yahoo.com> Message-ID: <4846A230.5000207@democritos.it> Rudra Banerjee wrote: > This link is still blank this is the correct link: http://www.quantum-espresso.org/wiki/index.php/Performance_issues_%28PWscf%29 -- Paolo Giannozzi, Democritos and University of Udine, Italy From conor.hogan at roma2.infn.it Wed Jun 4 22:19:18 2008 From: conor.hogan at roma2.infn.it (Conor Hogan) Date: Wed, 4 Jun 2008 22:19:18 +0200 (CEST) Subject: [Pw_forum] Convergence Problem on dipole correction and XCrysden problem on X86_64 In-Reply-To: <161e5bd50806032139p4910552dj69f58643d8fa9e35@mail.gmail.com> References: <161e5bd50806032139p4910552dj69f58643d8fa9e35@mail.gmail.com> Message-ID: <39902.83.184.20.251.1212610758.squirrel@83.184.20.251> On Wed, June 4, 2008 6:39 am, sei ou said: > Would you please answer my following questions? Yes SIR! Three bags full, sir. > (1) ZnO polar surface simulation using dipole correction: > Recently I performed test calculation on ZnO polar surface, when using > dipole correction, the convergence is very difficult to achieve, even in a > single scf cycle. Could anyone tell me where is wrong in my attached input > file? Try putting the slab symmetrically around the origin (the bulk of the density should be within the voronoi cell), this helped me in the past. Good luck with that Conor ---- Dr. Conor Hogan --- ----- --- --- Dipartimento di Fisica e CNR-INFM | : | : Universita' di Roma "Tor Vergata" |-- : --- |--- Tel: +39 06 7259 4121 *NEW* | : | : Fax: +39 06 2023507 --- : --- : http://www.fisica.uniroma2.it/~cmtheo-group/ European Theoretical Spectroscopy Facility http://www.etsf.eu Q. What do you call a fish with no eyes? A. A fsh. (Tommy Cooper) The early bird gets the worm, but the second mouse gets the cheese. - S. Wright If you go through a lot of hammers each month, I don't think it necessarily means you're a hard worker. It may just mean that you have a lot to learn about proper hammer maintenance - J. Handey From Mohamed.Majdoub at mail.uh.edu Wed Jun 4 23:23:59 2008 From: Mohamed.Majdoub at mail.uh.edu (Mohamed.Majdoub at mail.uh.edu) Date: Wed, 04 Jun 2008 16:23:59 -0500 Subject: [Pw_forum] sawlike potential in supercell calculation Message-ID: Dear users, I really find difficulties in understanding and ambiguities on how the saw like potential is implemented in pwscf. I need to apply an external electric filed to a slab structure in supercell (including vacuum region) in the z direction (edir =3) of amplitude eamp.According to the manual, besides setting tefiled and dipfield (dipole correction for slab) to be true, we need to specify the following variables in &system: ------------------------------------------------------------------------ edir?????????? INTEGER ?????????????? The direction of the electric field or dipole correction is ?????????????? parallel to the bg(:,edir) reciprocal lattice vector, so the ?????????????? potential is constant in planes defined by FFT grid points; ?????????????? edir = 1, 2 or 3. Used only if tefield is .TRUE. emaxpos??????? REAL ( default = 0.5D0 ) ?????????????? Position of the maximum of the sawlike potential along crystal ?????????????? axis "edir", within the? unit cell (see below), 0 < emaxpos < 1 ?????????????? Used only if tefield is .TRUE. eopreg???????? REAL( default = 0.1D0 ) ?????????????? Zone in the unit cell where the sawlike potential decreases. ?????????????? ( see below, 0 < eopreg < 1 ). Used only if tefield is .TRUE. eamp?????????? REAL ( default = 0.001 a.u. ) 0.000001 a.u ?????????????? Amplitude of the electric field (in a.u. = 51.44 10^10 V/m ) ?????????????? The sawlike potential increases with slope "eamp" in the ?????????????? region from (emaxpos+eopreg-1) to (emaxpos), then decreases ?????????????? to 0 until (emaxpos+eopreg), in units of the crystal ?????????????? vector "edir". Used only if tefield is .TRUE. --------------------------------------------------------------------- Since it is external electric field it shoud be only applied to the vacuum region in the supercell as it was pointed out before (http://www.democritos.it/pipermail/pw_forum/2004-July/001169.html). However, it is not clear if the 0< x<1 interval is the vacuum interval or the full length of the supercell (structure + vacuum). How the decrease in the sawlike potential is defined? Is it in the slab or the vaccum? Can any one familiar with this kind of calculation explain how it is implemented? Thank you in advance! Mohamed Sabri Majdoub PhD mechanical engineering University of Houston Houston, Tx, USA -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080604/811a522d/attachment-0001.htm From lanhaiping at gmail.com Thu Jun 5 05:51:58 2008 From: lanhaiping at gmail.com (lan haiping) Date: Thu, 5 Jun 2008 11:51:58 +0800 Subject: [Pw_forum] sawlike potential in supercell calculation In-Reply-To: References: Message-ID: you can draw a saw potential by hand, which will help you understand. for you question , ?0?1? of coz define the whole supercell under your examination. On Thu, Jun 5, 2008 at 5:23 AM, wrote: > Dear users, > > I really find difficulties in understanding and ambiguities on how the saw > like potential is implemented in pwscf. I need to apply an external electric > filed to a slab structure in supercell (including vacuum region) in the z > direction (edir =3) of amplitude eamp.According to the manual, besides > setting tefiled and dipfield (dipole correction for slab) to be true, we > need to specify the following variables in &system: > ------------------------------------------------------------------------ > edir INTEGER > > The direction of the electric field or dipole correction is > parallel to the bg(:,edir) reciprocal lattice vector, so the > potential is constant in planes defined by FFT grid points; > edir = 1, 2 or 3. Used only if tefield is .TRUE. > > emaxpos REAL ( default = 0.5D0 ) > > Position of the maximum of the sawlike potential along > crystal > axis "edir", within the unit cell (see below), 0 < emaxpos > < 1 > Used only if tefield is .TRUE. > > eopreg REAL( default = 0.1D0 ) > > Zone in the unit cell where the sawlike potential decreases. > ( see below, 0 < eopreg < 1 ). Used only if tefield is > .TRUE. > > eamp REAL ( default = 0.001 a.u. ) 0.000001 a.u > > Amplitude of the electric field (in a.u. = 51.44 10^10 V/m ) > The sawlike potential increases with slope "eamp" in the > region from (emaxpos+eopreg-1) to (emaxpos), then decreases > to 0 until (emaxpos+eopreg), in units of the crystal > vector "edir". Used only if tefield is .TRUE. > --------------------------------------------------------------------- > Since it is external electric field it shoud be only applied to the vacuum > region in the supercell as it was pointed out before ( > http://www.democritos.it/pipermail/pw_forum/2004-July/001169.html). > However, it is not clear if the 0< x<1 interval is the vacuum interval or > the full length of the supercell (structure + vacuum). How the decrease in > the sawlike potential is defined? Is it in the slab or the vaccum? > Can any one familiar with this kind of calculation explain how it is > implemented? > > Thank you in advance! > > Mohamed Sabri Majdoub > PhD mechanical engineering > University of Houston > Houston, Tx, USA > > _______________________________________________ > 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/20080605/a5f622af/attachment.htm From huangchunlin1984 at 163.com Thu Jun 5 05:39:05 2008 From: huangchunlin1984 at 163.com (huangchunlin1984) Date: Thu, 5 Jun 2008 11:39:05 +0800 (CST) Subject: [Pw_forum] unable to create directory prefix.save In-Reply-To: <22779.82.48.141.31.1212566590.squirrel@webmail.sissa.it> References: <22779.82.48.141.31.1212566590.squirrel@webmail.sissa.it> <26372581.790931212554869925.JavaMail.coremail@bj163app106.163.com> Message-ID: <10659781.211491212637145868.JavaMail.coremail@bj163app80.163.com> Dear Doctor: I have installed 3.2.3 and 4.0 with g95 compiler on a single AMD 64 bits computer. when running ,I encountered an error as follows : unable to creat *.save directory today, I change g95 with pgf90 ,it works. thanks for you help. students h -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080605/4b7d2846/attachment.htm From akohlmey at cmm.chem.upenn.edu Thu Jun 5 08:02:31 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Thu, 5 Jun 2008 02:02:31 -0400 (EDT) Subject: [Pw_forum] unable to create directory prefix.save In-Reply-To: <10659781.211491212637145868.JavaMail.coremail@bj163app80.163.com> References: <22779.82.48.141.31.1212566590.squirrel@webmail.sissa.it> <26372581.790931212554869925.JavaMail.coremail@bj163app106.163.com> <10659781.211491212637145868.JavaMail.coremail@bj163app80.163.com> Message-ID: On Thu, 5 Jun 2008, huangchunlin1984 wrote: > Dear Doctor: > I have installed 3.2.3 and 4.0 with g95 compiler on a single AMD 64 bits computer. when running ,I encountered an error as follows : > unable to creat *.save directory > today, I change g95 with pgf90 ,it works. so you probably downloaded the wrong version of g95. you need the one with 32-bit default integer. the 64-bit default integer version is non-standard... cheers, axel. > thanks for you help. > students h > -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From degironc at sissa.it Thu Jun 5 09:47:52 2008 From: degironc at sissa.it (Stefano de Gironcoli) Date: Thu, 05 Jun 2008 09:47:52 +0200 Subject: [Pw_forum] sawlike potential in supercell calculation In-Reply-To: References: Message-ID: <48479A28.1040908@sissa.it> Dear Mohamed, the saw-like potential mimic an uniform external electric field ... that is a field that is generated by a planar capacitor... actually an infinite array of planar capacitors ... to this end you need to specify where you want to put the capacitor plates (typically in the vacuum region) and what is the potential drop between them. emaxpos says WHERE to put the negative capacitor plate in unit of the relevant unit cell. eopreg says WHERE to to the positive capacitor plate (relative to the negative one). eamp is related to the electric field amplitude i.e. the potential drop across the capacitor. stefano Mohamed.Majdoub at mail.uh.edu wrote: > Dear users, > > I really find difficulties in understanding and ambiguities on how the > saw like potential is implemented in pwscf. I need to apply an > external electric filed to a slab structure in supercell (including > vacuum region) in the z direction (edir =3) of amplitude > eamp.According to the manual, besides setting tefiled and dipfield > (dipole correction for slab) to be true, we need to specify the > following variables in &system: > ------------------------------------------------------------------------ > edir INTEGER > > The direction of the electric field or dipole > correction is > parallel to the bg(:,edir) reciprocal lattice vector, > so the > potential is constant in planes defined by FFT grid points; > edir = 1, 2 or 3. Used only if tefield is .TRUE. > > emaxpos REAL ( default = 0.5D0 ) > > Position of the maximum of the sawlike potential along > crystal > axis "edir", within the unit cell (see below), 0 < > emaxpos < 1 > Used only if tefield is .TRUE. > > eopreg REAL( default = 0.1D0 ) > > Zone in the unit cell where the sawlike potential > decreases. > ( see below, 0 < eopreg < 1 ). Used only if tefield is > .TRUE. > > eamp REAL ( default = 0.001 a.u. ) 0.000001 a.u > > Amplitude of the electric field (in a.u. = 51.44 10^10 > V/m ) > The sawlike potential increases with slope "eamp" in the > region from (emaxpos+eopreg-1) to (emaxpos), then decreases > to 0 until (emaxpos+eopreg), in units of the crystal > vector "edir". Used only if tefield is .TRUE. > --------------------------------------------------------------------- > Since it is external electric field it shoud be only applied to the > vacuum region in the supercell as it was pointed out before > (http://www.democritos.it/pipermail/pw_forum/2004-July/001169.html). > However, it is not clear if the 0< x<1 interval is the vacuum interval > or the full length of the supercell (structure + vacuum). How the > decrease in the sawlike potential is defined? Is it in the slab or the > vaccum? > Can any one familiar with this kind of calculation explain how it is > implemented? > > Thank you in advance! > > Mohamed Sabri Majdoub > PhD mechanical engineering > University of Houston > Houston, Tx, USA > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From paulatto at sissa.it Thu Jun 5 15:08:38 2008 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Thu, 5 Jun 2008 15:08:38 +0200 (CEST) Subject: [Pw_forum] post-proccesing using pwscf4.0 and PAW In-Reply-To: <48468469.3010700@cea.fr> References: <48468469.3010700@cea.fr> Message-ID: <17596.147.122.5.216.1212671318.squirrel@webmail.sissa.it> On Wed, June 4, 2008 14:02, chu-chun fu wrote: > I will appreciate if you could fix this problem if possible ? We will surely fix it, I'm working on it right now. It will probably take a few days, than I'll take care to send you the patch to test. 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 w2agz at pacbell.net Fri Jun 6 06:16:45 2008 From: w2agz at pacbell.net (Paul M. Grant) Date: Thu, 5 Jun 2008 21:16:45 -0700 Subject: [Pw_forum] Innards of ph.f90 and Matdyn.f90 Message-ID: <00f401c8c78c$260a0980$721e1c80$@net> It is with great trepidation that I post these kinds of questions on PWforum. I always fear being scolded by Paolo and/or Axel for not looking at past fora postings or within the intricacies of "the code." (No, I missed seeing that covered in the posting of 28 June 1914.I was otherwise engaged). To be serious, my present posting is motivated by the late May discussions on "Phonon DOS in noncubic materials." I have started some e-p ph.x calculations on several "interesting" transition metal oxides using SCF eigenstates computed with LDA+U implemented, of cubic and/or tetragonal rocksalt symmetry. The end story intention is to calculate the Eliashberg lambda. Why would one ever want to do this? Think superconductivity. Issues: 1) I assume ph.x can handle spin polarized states, because that's the basis of Example02 and it seems to work OK. 2) It (ph.x) also "seems to work" with la2F and lda_plus_u set .true., that is, it generates gammas and lambdas without barfing, for cubic and tetragonal AFII primitive cells. So far, so good. 3) Question.why is it necessary to do two scf runs a la Example07, one to set a DOS flag for separate use later, and another to set the k-point grid for a subsequent el-ph run with ph.x? I know there might some resource issues for "serious" calculations, but for Example07, one scf with la2F "on" and an MP grid of order 12, "seems to" work just fine. At least for a "non-serious" calculation. 4) Next we bring C(q) from the output of ph.x to "real space" with q2r.x. 5) Now it gets dicey. The following I haven't tried yet. 6) The matdyn.x calculation to compute the frequency dependent component of the Eliashberg kernel uses a "real space" sampling grid.in the case of Example07 of size 19. How is this grid determined? It looks to me like it's spanning some high symmetry directions within the Al unit cell. Is there an algorithm to calculate such for general symmetry other than Al (or fcc) (e.g., an analog to MP sampling, or FS BZ sampling), or is such just a user-selected symmeterized "wet finger in the wind?" 7) Lastly, here's where I really get confused and appreciate Dr. Prasad's posting. Which "dos" calculation is being referred to.the scf run with la2F set true, or a restriction on the input to the "final" matdyn.x run with dos set true? What's needed is a user's manual for matdyn (sorry, Paolo), or, better yet, three or four more examples like 07 on non-cubic but otherwise simple symmetries and simple basis orbital sets with more than one atom per unit cell.like MgB2! Ciao, -Paul 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/20080605/3bbb174b/attachment-0001.htm From yrehem at mac.com Fri Jun 6 19:06:00 2008 From: yrehem at mac.com (Yaser Rehem) Date: Fri, 06 Jun 2008 10:06:00 -0700 Subject: [Pw_forum] Seeking an LDA Fluorine USPP Message-ID: <2B5BD15D-011A-1000-877C-7B4423332C98-Webmail-10012@mac.com> We don't see to have any LDA F pp's on the site. Could someone provide me with an LDA Fluorine USPP---no warranties expected. Your kindness would be much appreciated! Cheers, -Yaser Rehem Rehem Research & Consulting From wangqj1 at 126.com Sat Jun 7 13:17:10 2008 From: wangqj1 at 126.com (wangqj1) Date: Sat, 7 Jun 2008 19:17:10 +0800 (CST) Subject: [Pw_forum] The criteria of relax ? Message-ID: <12920016.164321212837430258.JavaMail.coremail@bj126app27.126.com> dear pwscf users I want to know the criteria of relax ,after I completing one relax ,I don't know whether the result conform to request or not .How to set the key parameter of relax , e.g ,the trust_radius_max,trust_radius_min,trust_radius_ini,and so on .what's the basis of set them ? I long for someone to help me . Thank you . -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080607/c19da785/attachment.htm From 041019019 at fudan.edu.cn Sat Jun 7 13:17:11 2008 From: 041019019 at fudan.edu.cn (Wang Qinjing) Date: Sat, 07 Jun 2008 19:17:11 +0800 Subject: [Pw_forum] How to solve the problem: "problems computing cholesky decomposition" Message-ID: <003101c8c890$07d31920$f9045c0a@WWWA0A18A7CBA2> Dear PWSCF users: I am a new user of the PWSCF. I met a problem when I did a simple SCF calculation. The CRASH file is as follows: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% task # 6 from pzpotrf : error # 54 problems computing cholesky decomposition %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% If I change the number of atoms "nat" from 9 to 3 in my input file, there is not any problem. What could I do to solve this problem? Thank you in advance. Yours Wang -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080607/f552c3b1/attachment.htm From eyvaz_isaev at yahoo.com Sat Jun 7 15:01:58 2008 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Sat, 7 Jun 2008 06:01:58 -0700 (PDT) Subject: [Pw_forum] How to solve the problem: "problems computing cholesky decomposition" In-Reply-To: <003101c8c890$07d31920$f9045c0a@WWWA0A18A7CBA2> Message-ID: <833262.50381.qm@web65708.mail.ac4.yahoo.com> Dear Wang, --- Wang Qinjing <041019019 at fudan.edu.cn> wrote: > Dear PWSCF users: > > I am a new user of the PWSCF. You are welcome! >I met a problem when I did a simple SCF calculation. The CRASH file is as > follows: > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > task # 6 > from pzpotrf : error # 54 > problems computing cholesky decomposition > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > If I change the number of atoms "nat" from 9 to 3 in > my input file, there is not any problem. > > What could I do to solve this problem? Thank you in > advance. I suppose nobody can help you until you post your input file. Besides, please provide your affiliation. This is accepted Netiquette by default in this forum. Bests, Eyvaz. > Yours Wang > > _______________________________________________ > 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, Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com From paulatto at sissa.it Sat Jun 7 16:54:56 2008 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Sat, 7 Jun 2008 16:54:56 +0200 (CEST) Subject: [Pw_forum] Seeking an LDA Fluorine USPP In-Reply-To: <2B5BD15D-011A-1000-877C-7B4423332C98-Webmail-10012@mac.com> References: <2B5BD15D-011A-1000-877C-7B4423332C98-Webmail-10012@mac.com> Message-ID: <28363.82.48.141.31.1212850496.squirrel@webmail.sissa.it> On Fri, June 6, 2008 19:06, Yaser Rehem wrote: > We don't see to have any LDA F pp's on the site. Could someone provide me > with an LDA Fluorine USPP---no warranties expected. Your kindness would > be much appreciated! There is an input file for ld1 in the pseudo library. If you are using QE 4.0 you will find it in espresso/atomic_doc/pseudo_library/LDA/SR/ otherwise you can download it from here: 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 wangqj1 at 126.com Sat Jun 7 16:34:55 2008 From: wangqj1 at 126.com (wangqj1) Date: Sat, 7 Jun 2008 22:34:55 +0800 (CST) Subject: [Pw_forum] convergence NOT achieved, stopping Message-ID: <16081824.202961212849295271.JavaMail.coremail@bj126app104.126.com> Dear PWSCF users When I made the 'scf'of 'O',the convergence is very slow,and at last turn up 'convergence NOT achieved, stopping',my inputfile as following: &CONTROL title = 'O' , calculation = 'scf' , restart_mode = 'from_scratch' , outdir = '/root/work/ZnO_0_0_1/O' , wfcdir = '/root/work/ZnO_0_0_1/O' , pseudo_dir = '/root/tool/espresso-3.2.3/pseudo/' , prefix = 'O' , etot_conv_thr = 0.0001 , forc_conv_thr = 0.0001 , / &SYSTEM ibrav = 14, A = 6.49854 , B = 6.49854 , C = 26.427677 , cosAB = 0 , cosAC = 0 , cosBC = 0.5 , nat = 16, ntyp = 1, ecutwfc = 30 , ecutrho = 120 , nosym = .true. , occupations = 'fixed' , degauss = 0.02 , / &ELECTRONS mixing_mode = 'plain' , diagonalization = 'david' , diago_full_acc = .false. , / ATOMIC_SPECIES O 16.00000 O.pbe-van_bm.UPF ATOMIC_POSITIONS angstrom O 1.624773263 0.573114301 0.485838289 O 3.249224094 3.822615460 0.485946263 O 4.873764156 0.573115030 0.485836746 O 3.249294779 1.804680145 2.723816512 O 1.624804430 0.573863860 13.703775132 O 3.249230400 3.823233855 13.703791642 O 4.873742964 0.573863275 13.703791170 O 3.249304260 1.805307421 15.942064019 O 6.498493290 3.822459082 0.485574395 O 4.873652639 5.054116368 2.724105761 O 6.498566063 1.804935512 2.723993591 O 8.123380444 5.054113329 2.724106775 O 6.498495372 3.823229705 13.703092130 O 4.873728849 5.054758560 15.941611312 O 6.498573263 1.805673494 15.941687153 O 8.123315037 5.054760018 15.941598059 K_POINTS automatic 1 1 1 0 0 0 What should I do ? Thanks for your advice! -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080607/723f1800/attachment.htm From j.antonio.montoya at gmail.com Sat Jun 7 17:30:25 2008 From: j.antonio.montoya at gmail.com (j.antonio.montoya at gmail.com) Date: Sat, 7 Jun 2008 15:30:25 +0000 Subject: [Pw_forum] The criteria of relax ? In-Reply-To: <12920016.164321212837430258.JavaMail.coremail@bj126app27.126.com> References: <12920016.164321212837430258.JavaMail.coremail@bj126app27.126.com> Message-ID: <134053490-1212852631-cardhu_decombobulator_blackberry.rim.net-802791043-@bxe114.bisx.prod.on.blackberry> Hi, If you relax ions probably you want to check the forces on the atoms and make sure that they are as low as possible. In the case of cell relaxation you will want to check the stress tensor and make sure that it only contains elements in the diagonal and that they are as close to each other as possible and coinciding with the target pressure. Please include your affiliation in your posts. Javier Montoya Geophysical Lab. Carnegie Institution of Washington Washington DC Sent via BlackBerry by AT&T -----Original Message----- From: wangqj1 Date: Sat, 7 Jun 2008 19:17:10 To:pw_forum Subject: [Pw_forum] The criteria of relax ? ? ?dear pwscf users ?I want to know the criteria of relax ,after I completing one relax ,I don't know whether the result conform to request or not .How to set the key parameter of relax , e.g ,the trust_radius_max,trust_radius_min,trust_radius_ini,and so on .what's the basis of set them ? I long for someone to help me . Thank you . ? ? ---------------- ??????3D?????2??6?6????? _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum From j.antonio.montoya at gmail.com Sat Jun 7 17:39:13 2008 From: j.antonio.montoya at gmail.com (j.antonio.montoya at gmail.com) Date: Sat, 7 Jun 2008 15:39:13 +0000 Subject: [Pw_forum] convergence NOT achieved, stopping In-Reply-To: <16081824.202961212849295271.JavaMail.coremail@bj126app104.126.com> References: <16081824.202961212849295271.JavaMail.coremail@bj126app104.126.com> Message-ID: <710458371-1212853159-cardhu_decombobulator_blackberry.rim.net-777694581-@bxe114.bisx.prod.on.blackberry> There is no general solution to guarantee fast convergence for the electrons, but there are many options that can be used in "ELECTRONS" and that you should explore if you are willing to do some research. Please read the documentation regarding the input for pw. Javier Montoya Geophysical Lab. Carnegie Institution of Washington Washington DC Sent via BlackBerry by AT&T -----Original Message----- From: wangqj1 Date: Sat, 7 Jun 2008 22:34:55 To:pw_forum Subject: [Pw_forum] convergence NOT achieved, stopping ? ?Dear PWSCF users ?When I made the 'scf'of 'O',the convergence is very slow,and at last turn up 'convergence NOT achieved, stopping',my inputfile as following: ? ?&CONTROL ?????????????????????? title = 'O' , ???????????????? calculation = 'scf' , ??????????????? restart_mode = 'from_scratch' , ????????????????????? outdir = '/root/work/ZnO_0_0_1/O' , ????????????????????? wfcdir = '/root/work/ZnO_0_0_1/O' , ????????????????? pseudo_dir = '/root/tool/espresso-3.2.3/pseudo/' , ????????????????????? prefix = 'O' , ?????????????? etot_conv_thr = 0.0001 , ?????????????? forc_conv_thr = 0.0001 , ?/ ?&SYSTEM ?????????????????????? ibrav = 14, ?????????????????????????? A = 6.49854 , ?????????????????????????? B = 6.49854 , ?????????????????????????? C = 26.427677 , ?????????????????????? cosAB = 0 , ?????????????????????? cosAC = 0 , ?????????????????????? cosBC = 0.5 , ???????????????????????? nat = 16, ??????????????????????? ntyp = 1, ???????????????????? ecutwfc = 30 , ???????????????????? ecutrho = 120 , ?????????????????????? nosym = .true. , ???????????????? occupations = 'fixed' , ???????????????????? degauss = 0.02 , ?/ ?&ELECTRONS ???????????????? mixing_mode = 'plain' , ???????????? diagonalization = 'david' , ????????????? diago_full_acc = .false. , ?/ ATOMIC_SPECIES ??? O?? 16.00000? O.pbe-van_bm.UPF ATOMIC_POSITIONS angstrom ??? O????? 1.624773263??? 0.573114301??? 0.485838289??? ??? O????? 3.249224094??? 3.822615460??? 0.485946263??? ??? O????? 4.873764156??? 0.573115030??? 0.485836746??? ??? O????? 3.249294779??? 1.804680145??? 2.723816512??? ??? O????? 1.624804430??? 0.573863860?? 13.703775132??? ??? O????? 3.249230400??? 3.823233855?? 13.703791642??? ??? O????? 4.873742964??? 0.573863275?? 13.703791170??? ??? O????? 3.249304260??? 1.805307421?? 15.942064019??? ??? O????? 6.498493290??? 3.822459082??? 0.485574395??? ??? O????? 4.873652639??? 5.054116368??? 2.724105761??? ??? O????? 6.498566063??? 1.804935512??? 2.723993591??? ??? O????? 8.123380444??? 5.054113329??? 2.724106775??? ??? O????? 6.498495372??? 3.823229705?? 13.703092130??? ??? O????? 4.873728849??? 5.054758560?? 15.941611312??? ??? O????? 6.498573263??? 1.805673494?? 15.941687153??? ??? O????? 8.123315037??? 5.054760018?? 15.941598059??? K_POINTS automatic ? 1 1 1?? 0 0 0 What should I do ? Thanks for your advice! ? ---------------- ??????3D?????2??6?6????? _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum From zhaohscas at yahoo.com.cn Sat Jun 7 18:57:50 2008 From: zhaohscas at yahoo.com.cn (=?utf-8?B?emhhb2hzY2Fz?=) Date: Sun, 8 Jun 2008 00:57:50 +0800 Subject: [Pw_forum] =?utf-8?q?About_the_generate_uspp_of_hafnium_for_using?= =?utf-8?q?_within_pwscf=2E?= References: <16081824.202961212849295271.JavaMail.coremail@bj126app104.126.com> Message-ID: <200806080057479569123@yahoo.com.cn> Hi forks, I'm now using the pwscf for my work, in the study of mine, the hafnium is one of the elements in the system. I've foundd that the CASTEP, which is my another tool for my work, has the uspp's for hafnium, the Hf 5d^2 6s^2 was treated as the valence states. Who can give me some hints to generate the corresponding uspp of hafnium for using within pwscf? Sincerely yours, Hongsheng. ------------------ Hongsheng Zhao Xinjiang Technical Institute of Physics and Chemistry Chinese Academy of Sciences GnuPG DSA: 0xD108493 2008-06-08 __________________________________________________ ????????????????????????????? http://cn.mail.yahoo.com From yrehem at mac.com Sat Jun 7 23:09:29 2008 From: yrehem at mac.com (Yaser Rehem) Date: Sat, 07 Jun 2008 14:09:29 -0700 Subject: [Pw_forum] Seeking an LDA Fluorine USPP In-Reply-To: <28363.82.48.141.31.1212850496.squirrel@webmail.sissa.it> References: <2B5BD15D-011A-1000-877C-7B4423332C98-Webmail-10012@mac.com> <28363.82.48.141.31.1212850496.squirrel@webmail.sissa.it> Message-ID: <5CB6E11A-6B6C-431D-83D2-6B86AE94E901@mac.com> Thank you Lorenzo. Cheers, -Yaser On Jun 7, 2008, at 7:54 AM, Lorenzo Paulatto wrote: > > On Fri, June 6, 2008 19:06, Yaser Rehem wrote: >> We don't see to have any LDA F pp's on the site. Could someone >> provide me >> with an LDA Fluorine USPP---no warranties expected. Your kindness >> would >> be much appreciated! > > There is an input file for ld1 in the pseudo library. If you are > using QE > 4.0 you will find it in espresso/atomic_doc/pseudo_library/LDA/SR/ > otherwise you can download it from here: > > > > 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/ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From zhaohscas at yahoo.com.cn Sun Jun 8 04:05:31 2008 From: zhaohscas at yahoo.com.cn (zhaohscas) Date: Sun, 8 Jun 2008 10:05:31 +0800 Subject: [Pw_forum] pwscf's compiling error under cygwin. Message-ID: <200806081005266098270@yahoo.com.cn> Hi all, I using the latest cygwin, i.e., cygwin-1.5.25-14, and the g95-x86-cygwin-41 is also installled on my cygwin box. When I compile the espresso-4.0, the configure can be performed successfully. BTW, I just use the default configure options for my case, that is, ./configure. When I perform the make all command, I meet the following errors: -------------------------------- [...] make[1]: Entering directory `/cygdrive/f/zhaohs/cygwinhomewerner/espresso-4.0/Mo dules' g95 -O3 -cpp -D__FFTW -D__USE_INTERNAL_FFTW -I../include -I./ -I../Modules -I .../iotk/src -I../PW -I../PH -c constraints_module.f90 constraints_module.f90: In function 'set_structure_factor__': constraints_module.f90:661: internal compiler error: in find_lattice_value, at t ree-complex.c:133 Please submit a full bug report, with preprocessed source if appropriate. See http://www.g95.org or mail andyv at firstinter.net for instructions. make[1]: *** [constraints_module.o] Error 1 make[1]: Leaving directory `/cygdrive/f/zhaohs/cygwinhomewerner/espresso-4.0/Mod ules' make: *** [mods] Error 2 -------------------------------- Who can give me some hints on this thing. Any help would be appreciated. Sincerely yours, -------------- Hongsheng Zhao Xinjiang Technical Institute of Physics and Chemistry Chinese Academy of Sciences GnuPG DSA: 0xD108493 2008-06-08 __________________________________________________ ??????????????? http://cn.mail.yahoo.com From zhaohscas at yahoo.com.cn Sun Jun 8 04:05:31 2008 From: zhaohscas at yahoo.com.cn (zhaohscas) Date: Sun, 8 Jun 2008 10:05:31 +0800 Subject: [Pw_forum] pwscf's compiling error under cygwin. Message-ID: <200806081005266098270@yahoo.com.cn> Hi all, I using the latest cygwin, i.e., cygwin-1.5.25-14, and the g95-x86-cygwin-41 is also installled on my cygwin box. When I compile the espresso-4.0, the configure can be performed successfully. BTW, I just use the default configure options for my case, that is, ./configure. When I perform the make all command, I meet the following errors: -------------------------------- [...] make[1]: Entering directory `/cygdrive/f/zhaohs/cygwinhomewerner/espresso-4.0/Mo dules' g95 -O3 -cpp -D__FFTW -D__USE_INTERNAL_FFTW -I../include -I./ -I../Modules -I .../iotk/src -I../PW -I../PH -c constraints_module.f90 constraints_module.f90: In function 'set_structure_factor__': constraints_module.f90:661: internal compiler error: in find_lattice_value, at t ree-complex.c:133 Please submit a full bug report, with preprocessed source if appropriate. See http://www.g95.org or mail andyv at firstinter.net for instructions. make[1]: *** [constraints_module.o] Error 1 make[1]: Leaving directory `/cygdrive/f/zhaohs/cygwinhomewerner/espresso-4.0/Mod ules' make: *** [mods] Error 2 -------------------------------- Who can give me some hints on this thing. Any help would be appreciated. Sincerely yours, -------------- Hongsheng Zhao Xinjiang Technical Institute of Physics and Chemistry Chinese Academy of Sciences GnuPG DSA: 0xD108493 2008-06-08 __________________________________________________ ??????????????? http://cn.mail.yahoo.com From akohlmey at cmm.chem.upenn.edu Sun Jun 8 04:05:30 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sat, 7 Jun 2008 22:05:30 -0400 (EDT) Subject: [Pw_forum] pwscf's compiling error under cygwin. In-Reply-To: <200806081005266098270@yahoo.com.cn> References: <200806081005266098270@yahoo.com.cn> Message-ID: On Sun, 8 Jun 2008, zhaohscas wrote: HZ> Hi all, hi, HZ> I using the latest cygwin, i.e., cygwin-1.5.25-14, and the HZ> g95-x86-cygwin-41 is also installled on my cygwin box. ouch. there is one born every minute ... HZ> When I compile the espresso-4.0, the configure can be performed HZ> successfully. BTW, I just use the default configure options for my HZ> case, that is, ./configure. HZ> When I perform the make all command, I meet the following errors: HZ> HZ> -------------------------------- HZ> [...] HZ> make[1]: Entering directory `/cygdrive/f/zhaohs/cygwinhomewerner/espresso-4.0/Mo HZ> dules' HZ> g95 -O3 -cpp -D__FFTW -D__USE_INTERNAL_FFTW -I../include -I./ -I../Modules -I HZ> .../iotk/src -I../PW -I../PH -c constraints_module.f90 ==================================================== HZ> constraints_module.f90: In function 'set_structure_factor__': HZ> constraints_module.f90:661: internal compiler error: in find_lattice_value, at t HZ> ree-complex.c:133 HZ> Please submit a full bug report, HZ> with preprocessed source if appropriate. HZ> See http://www.g95.org or mail andyv at firstinter.net for instructions. ===================================================== 5BHZ> make[1]: *** [constraints_module.o] Error 1 HZ> make[1]: Leaving directory `/cygdrive/f/zhaohs/cygwinhomewerner/espresso-4.0/Mod HZ> ules' HZ> make: *** [mods] Error 2 HZ> -------------------------------- HZ> HZ> Who can give me some hints on this thing. HZ> Any help would be appreciated. how about _reading_ what the error message says??? check out the block of text marked with ============ you have encountered an "internal compiler error", which is a _definitely_ a bug in the compiler. so contact the g95 folk(s) and help them to reproduce the bug so they can fix it. this is _not_ a problem of quantum espresso! cheers, axel. HZ> Sincerely yours, HZ> -------------- HZ> Hongsheng Zhao HZ> Xinjiang Technical Institute of Physics and Chemistry HZ> Chinese Academy of Sciences HZ> GnuPG DSA: 0xD108493 HZ> 2008-06-08 HZ> HZ> __________________________________________________ HZ> ??????????????? HZ> http://cn.mail.yahoo.com HZ> HZ> _______________________________________________ HZ> Pw_forum mailing list HZ> Pw_forum at pwscf.org HZ> http://www.democritos.it/mailman/listinfo/pw_forum HZ> -- ======================================================================= 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 041019019 at fudan.edu.cn Sun Jun 8 07:04:45 2008 From: 041019019 at fudan.edu.cn (Wang Qinjing) Date: Sun, 08 Jun 2008 13:04:45 +0800 Subject: [Pw_forum] How to solve the problem: "problems computing cholesky decomposition" References: <833262.50381.qm@web65708.mail.ac4.yahoo.com> Message-ID: <00d201c8c925$2b567b40$f9045c0a@WWWA0A18A7CBA2> Dear Eyvaz: I've solved the problem by changing 'diagonalization' from default value to 'CG' mode. That means the default diagonalization mode is unstable. Thank you for your advice. Yours Wang ------------------------------------------------------- Wang Qinjing Physics Department, Fudan University, Shanghai, China ??? ??????? 014019019 at fudan.edu.cn ----- Original Message ----- From: "Eyvaz Isaev" To: "PWSCF Forum" Sent: Saturday, June 07, 2008 9:01 PM Subject: Re: [Pw_forum] How to solve the problem: "problems computing cholesky decomposition" > Dear Wang, > > --- Wang Qinjing <041019019 at fudan.edu.cn> wrote: > >> Dear PWSCF users: >> >> I am a new user of the PWSCF. > > You are welcome! > >>I met a problem when I did a simple SCF calculation. > The CRASH file is as >> follows: >> >> > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% >> task # 6 >> from pzpotrf : error # 54 >> problems computing cholesky decomposition >> >> > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% >> >> If I change the number of atoms "nat" from 9 to 3 in >> my input file, there is not any problem. >> >> What could I do to solve this problem? Thank you in >> advance. > > I suppose nobody can help you until you post your > input file. > Besides, please provide your affiliation. This is > accepted Netiquette by default in this forum. > > Bests, > Eyvaz. > >> Yours Wang >> > _______________________________________________ >> 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, > Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden > Condensed Matter Theory Group, Uppsala University, Sweden > Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From zhaohscas at yahoo.com.cn Sun Jun 8 13:46:31 2008 From: zhaohscas at yahoo.com.cn (zhaohscas) Date: Sun, 8 Jun 2008 19:46:31 +0800 Subject: [Pw_forum] A bug in the pp.x? Message-ID: <200806081946261265750@yahoo.com.cn> Hi forks, I find the following things: According to the description in the pwgui, the variables: e1(i), e2(i), e3(i), i=1,3 which are 3D vectors determining the plotting parallelepiped (if present, must be orthogonal). Then, I use the sample05 for test. I find that that, if I use the following input files for pp.x, the pp.x will also work: ------------------ &INPUTPP prefix = 'si' , outdir = '/home/werner/tmp/' , filplot = 'sicharge' , plot_num = 0, / &PLOT nfile = 1 , filepp(1) = 'sicharge', weight(1) = 1.0, fileout = 'si.rho.xsf' , iflag = 3 , output_format = 3 , e1(1) = 1.0, e1(2) = 1.0, e1(3) = 0.0, e2(1) = 0.0, e2(2) = 0.0, e2(3) = 1.0, e3(1) = -2, e3(2) = 1, e3(3) = 0, x0(1) = 0, x0(2) = 0, x0(3) = 0, nx = 56 , ny = 40 , nz = 40 , / ------------------ In the abobe case, the e1 \bot e2, and e2 \bot e3, but e1 isn' perpendicular to e3. So, this will violate orthogonal condition. Who can give some hints on this thing. Sincerely yours, Hongsheng. PS. the attachments are the inpput file and the partial output xsf file of pp.x. -------------- Hongsheng Zhao Xinjiang Technical Institute of Physics and Chemistry Chinese Academy of Sciences GnuPG DSA: 0xD108493 2008-06-08 -------------- next part -------------- A non-text attachment was scrubbed... Name: sample.pp.inp Type: application/octet-stream Size: 982 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080608/1e517ede/attachment.obj -------------- next part -------------- A non-text attachment was scrubbed... Name: si.rho.xsf Type: application/octet-stream Size: 631 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080608/1e517ede/attachment-0001.obj From zhaohscas at yahoo.com.cn Sun Jun 8 13:46:31 2008 From: zhaohscas at yahoo.com.cn (zhaohscas) Date: Sun, 8 Jun 2008 19:46:31 +0800 Subject: [Pw_forum] A bug in the pp.x? Message-ID: <200806081946261265750@yahoo.com.cn> Hi forks, I find the following things: According to the description in the pwgui, the variables: e1(i), e2(i), e3(i), i=1,3 which are 3D vectors determining the plotting parallelepiped (if present, must be orthogonal). Then, I use the sample05 for test. I find that that, if I use the following input files for pp.x, the pp.x will also work: ------------------ &INPUTPP prefix = 'si' , outdir = '/home/werner/tmp/' , filplot = 'sicharge' , plot_num = 0, / &PLOT nfile = 1 , filepp(1) = 'sicharge', weight(1) = 1.0, fileout = 'si.rho.xsf' , iflag = 3 , output_format = 3 , e1(1) = 1.0, e1(2) = 1.0, e1(3) = 0.0, e2(1) = 0.0, e2(2) = 0.0, e2(3) = 1.0, e3(1) = -2, e3(2) = 1, e3(3) = 0, x0(1) = 0, x0(2) = 0, x0(3) = 0, nx = 56 , ny = 40 , nz = 40 , / ------------------ In the abobe case, the e1 \bot e2, and e2 \bot e3, but e1 isn' perpendicular to e3. So, this will violate orthogonal condition. Who can give some hints on this thing. Sincerely yours, Hongsheng. PS. the attachments are the inpput file and the partial output xsf file of pp.x. -------------- Hongsheng Zhao Xinjiang Technical Institute of Physics and Chemistry Chinese Academy of Sciences GnuPG DSA: 0xD108493 2008-06-08 -------------- next part -------------- A non-text attachment was scrubbed... Name: sample.pp.inp Type: application/octet-stream Size: 982 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080608/1e517ede/attachment-0002.obj -------------- next part -------------- A non-text attachment was scrubbed... Name: si.rho.xsf Type: application/octet-stream Size: 631 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080608/1e517ede/attachment-0003.obj From paulatto at sissa.it Sun Jun 8 16:55:16 2008 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Sun, 8 Jun 2008 16:55:16 +0200 (CEST) Subject: [Pw_forum] A bug in the pp.x? In-Reply-To: <200806081946261265750@yahoo.com.cn> References: <200806081946261265750@yahoo.com.cn> Message-ID: <16644.79.16.7.234.1212936916.squirrel@webmail.sissa.it> On Sun, June 8, 2008 13:46, zhaohscas wrote: > Hi forks, Hi chopstick ;-), > In the abobe case, the e1 \bot e2, and e2 \bot e3, but e1 isn' > perpendicular to e3. So, this will violate orthogonal condition. You are right: the actual check performed by the code is: e_i \cdot \e_j > 1.d-6 which wrongly allows non-orthogonal vectors if their pairwise scalar products are negative. Thank you for reporting, it will be fixed in the next release. bye -- 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 zhaohscas at yahoo.com.cn Sun Jun 8 17:29:42 2008 From: zhaohscas at yahoo.com.cn (zhaohscas) Date: Sun, 8 Jun 2008 23:29:42 +0800 Subject: [Pw_forum] A bug in the pp.x? References: <200806081946261265750@yahoo.com.cn> Message-ID: <200806082329365953078@yahoo.com.cn> >Hi chopstick ;-), Aha, I'm using both forks and chopsticks once in a while. >> In the abobe case, the e1 \bot e2, and e2 \bot e3, but e1 isn' >> perpendicular to e3. So, this will violate orthogonal condition. > >You are right: the actual check performed by the code is: > e_i \cdot \e_j > 1.d-6 >which wrongly allows non-orthogonal vectors if their pairwise scalar >products are negative. Thank you for reporting, it will be fixed in the >next release. But, in my mind, as for the study of the distribution of the charge density, we generallly want to have the distribution w.r.t a specific miller plane. In the periodic crystal system, this can be represented on the plane determined by a parallelogram whose two sides are the corresponding primitive plane vectors. If we consider this requirement, I think we should use miller plane vectors to contruct the grid, which are not always perpendicular indeed. Under the above consideration, the orthogonal condition will be not so good. Yours sincerely, Hongsheng. __________________________________________________ ??????????????? http://cn.mail.yahoo.com From akohlmey at cmm.chem.upenn.edu Sun Jun 8 17:52:11 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sun, 8 Jun 2008 11:52:11 -0400 (EDT) Subject: [Pw_forum] A bug in the pp.x? In-Reply-To: <200806082329365953078@yahoo.com.cn> References: <200806081946261265750@yahoo.com.cn> <200806082329365953078@yahoo.com.cn> Message-ID: On Sun, 8 Jun 2008, zhaohscas wrote: HZ> >You are right: the actual check performed by the code is: HZ> > e_i \cdot \e_j > 1.d-6 HZ> >which wrongly allows non-orthogonal vectors if their pairwise scalar HZ> >products are negative. Thank you for reporting, it will be fixed in the HZ> >next release. HZ> HZ> But, in my mind, as for the study of the distribution of the charge HZ> density, we generallly want to have the distribution w.r.t a HZ> specific miller plane. In the periodic crystal system, this can be HZ> represented on the plane determined by a parallelogram whose two HZ> sides are the corresponding primitive plane vectors. If we consider HZ> this requirement, I think we should use miller plane vectors to HZ> contruct the grid, which are not always perpendicular indeed. HZ> Under the above consideration, the orthogonal condition will be not HZ> so good. that is all find and good, but the code actually has to be able to do this. those checks have the purpose to stop people from getting bogus output (and publishing them a _new_ science). so before lifting one of those restrictions, one has to validate that the code in question gives the correct results, and - most likely - the missing parts will have to be implemented. there are essentially two major forces that drive open source scientific software: 1) people that want to do something specific and sit down and program it, 2) people that want a degree or a salary or a professor position or ..., so they sit down and write what their bosses/advisor/peers ask them to do. unfortunately, there is a large difference between the number of good and useful ideas and the people that fall under those two categories, and on top of that for some strange reasion they prefer to implement their own ideas (or those of their bosses/advisors/...) ;-) cheers, axel. HZ> HZ> Yours sincerely, HZ> Hongsheng. HZ> HZ> __________________________________________________ HZ> ??????????????? HZ> http://cn.mail.yahoo.com HZ> HZ> _______________________________________________ HZ> Pw_forum mailing list HZ> Pw_forum at pwscf.org HZ> http://www.democritos.it/mailman/listinfo/pw_forum HZ> -- ======================================================================= 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 zhaohscas at yahoo.com.cn Mon Jun 9 13:08:36 2008 From: zhaohscas at yahoo.com.cn (zhaohscas) Date: Mon, 9 Jun 2008 19:08:36 +0800 Subject: [Pw_forum] A query about the PRIMCOORD parameter of the Xcrysden. References: <200806081946261265750@yahoo.com.cn> Message-ID: <200806091908337851952@yahoo.com.cn> Hi all, I've seen the Xcrysden documentions at the url: http://www.xcrysden.org/doc/XSF.html On the above url, there're the following descriptions: --------------------------------------------- Here is an example of ZnS crystal structure: CRYSTAL see (1) PRIMVEC 0.0000000 2.7100000 2.7100000 see (2) 2.7100000 0.0000000 2.7100000 2.7100000 2.7100000 0.0000000 CONVVEC 5.4200000 0.0000000 0.0000000 see (3) 0.0000000 5.4200000 0.0000000 0.0000000 0.0000000 5.4200000 PRIMCOORD 2 1 see (4) 16 0.0000000 0.0000000 0.0000000 see (5) 30 1.3550000 -1.3550000 -1.3550000 --------------------------------------------- In the above descriptions, if we take the sulfur atom position as the origin, than the Zinc atom should have the coordinates (-1.3550000 -1.3550000 -1.3550000). I don't konw why it should be (1.3550000 -1.3550000 -1.3550000) in the Xcrysden's documents. Who can give me some hints on this? Sincerely yours, Hongsheng. ------------------ Hongsheng Zhao Xinjiang Technical Institute of Physics and Chemistry Chinese Academy of Sciences GnuPG DSA: 0xD108493 __________________________________________________ ??????????????? http://cn.mail.yahoo.com From zhaohscas at yahoo.com.cn Mon Jun 9 13:08:36 2008 From: zhaohscas at yahoo.com.cn (zhaohscas) Date: Mon, 9 Jun 2008 19:08:36 +0800 Subject: [Pw_forum] A query about the PRIMCOORD parameter of the Xcrysden. References: <200806081946261265750@yahoo.com.cn> Message-ID: <200806091908337851952@yahoo.com.cn> Hi all, I've seen the Xcrysden documentions at the url: http://www.xcrysden.org/doc/XSF.html On the above url, there're the following descriptions: --------------------------------------------- Here is an example of ZnS crystal structure: CRYSTAL see (1) PRIMVEC 0.0000000 2.7100000 2.7100000 see (2) 2.7100000 0.0000000 2.7100000 2.7100000 2.7100000 0.0000000 CONVVEC 5.4200000 0.0000000 0.0000000 see (3) 0.0000000 5.4200000 0.0000000 0.0000000 0.0000000 5.4200000 PRIMCOORD 2 1 see (4) 16 0.0000000 0.0000000 0.0000000 see (5) 30 1.3550000 -1.3550000 -1.3550000 --------------------------------------------- In the above descriptions, if we take the sulfur atom position as the origin, than the Zinc atom should have the coordinates (-1.3550000 -1.3550000 -1.3550000). I don't konw why it should be (1.3550000 -1.3550000 -1.3550000) in the Xcrysden's documents. Who can give me some hints on this? Sincerely yours, Hongsheng. ------------------ Hongsheng Zhao Xinjiang Technical Institute of Physics and Chemistry Chinese Academy of Sciences GnuPG DSA: 0xD108493 __________________________________________________ ??????????????? http://cn.mail.yahoo.com From degironc at sissa.it Mon Jun 9 13:15:56 2008 From: degironc at sissa.it (Stefano de Gironcoli) Date: Mon, 09 Jun 2008 13:15:56 +0200 Subject: [Pw_forum] A query about the PRIMCOORD parameter of the Xcrysden. In-Reply-To: <200806091908337851952@yahoo.com.cn> References: <200806081946261265750@yahoo.com.cn> <200806091908337851952@yahoo.com.cn> Message-ID: <484D10EC.8010200@sissa.it> the two choices are equivalent (not identical). You may want to try to visualize the two and see the difference. Stefano de Gironcoli - SISSA and DEMOCRITOS > Here is an example of ZnS crystal structure: > > CRYSTAL see (1) > PRIMVEC > 0.0000000 2.7100000 2.7100000 see (2) > 2.7100000 0.0000000 2.7100000 > 2.7100000 2.7100000 0.0000000 > CONVVEC > 5.4200000 0.0000000 0.0000000 see (3) > 0.0000000 5.4200000 0.0000000 > 0.0000000 0.0000000 5.4200000 > PRIMCOORD > 2 1 see (4) > 16 0.0000000 0.0000000 0.0000000 see (5) > 30 1.3550000 -1.3550000 -1.3550000 > --------------------------------------------- > > In the above descriptions, if we take the sulfur atom position as the origin, than the Zinc atom should have the coordinates (-1.3550000 -1.3550000 -1.3550000). I don't konw why it should be (1.3550000 -1.3550000 -1.3550000) in the Xcrysden's documents. > > Who can give me some hints on this? > From zhaohscas at yahoo.com.cn Mon Jun 9 13:36:29 2008 From: zhaohscas at yahoo.com.cn (zhaohscas) Date: Mon, 9 Jun 2008 19:36:29 +0800 Subject: [Pw_forum] Xcysden cann't give all atoms in the conventional cell. References: <200806081946261265750@yahoo.com.cn> Message-ID: <200806091936233323357@yahoo.com.cn> Hi all, I noticed the following things in Xcysden: it cann't give all atoms in the conventional cell based on the PRIMVEC, CONVVEC, and PRIMCOORD parameters only. In detail, the space group number of a specific crystal is of importance in determining the atoms in a cell. From the Xcrysen's descriptions, it just use the above three parameters to generate the atoms in the conventional cell. So, I think it is inadequate to decide all the atoms in the conventional cell corresponding to a primitive cell. Sincerely yours, Hongsheng. __________________________________________________ ??????????????? http://cn.mail.yahoo.com From akohlmey at cmm.chem.upenn.edu Mon Jun 9 13:14:07 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Mon, 9 Jun 2008 07:14:07 -0400 (EDT) Subject: [Pw_forum] Xcysden cann't give all atoms in the conventional cell. In-Reply-To: <200806091936233323357@yahoo.com.cn> References: <200806081946261265750@yahoo.com.cn> <200806091936233323357@yahoo.com.cn> Message-ID: On Mon, 9 Jun 2008, zhaohscas wrote: HY> Hi all, HY> HY> I noticed the following things in Xcysden: it cann't give all atoms HY> in the conventional cell based on the PRIMVEC, CONVVEC, and HY> PRIMCOORD parameters only. In detail, the space group number of a HY> specific crystal is of importance in determining the atoms in a HY> cell. From the Xcrysen's descriptions, it just use the above three HY> parameters to generate the atoms in the conventional cell. So, I HY> think it is inadequate to decide all the atoms in the conventional HY> cell corresponding to a primitive cell. if you have a fundamental doubt like this on a software that has been around for quite a while and that a lot of people have used without noticing it, wouldn't it be more probable, that you don't get how it is meant? the primcoord section contains the atoms in the primary cell. all of them, not only the symmetry invariant ones. so you can switch between them, the system is just not as sophisticated as you expected it to be... ;-) cheers, axel. HY> HY> Sincerely yours, HY> Hongsheng. HY> HY> __________________________________________________ HY> ??????????????? HY> http://cn.mail.yahoo.com HY> HY> _______________________________________________ HY> Pw_forum mailing list HY> Pw_forum at pwscf.org HY> http://www.democritos.it/mailman/listinfo/pw_forum HY> -- ======================================================================= 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 anjmanqi at yahoo.com.cn Mon Jun 9 14:44:16 2008 From: anjmanqi at yahoo.com.cn (jiming an) Date: Mon, 9 Jun 2008 20:44:16 +0800 (CST) Subject: [Pw_forum] =?gb2312?b?u9i4tKO6IFB3X2ZvcnVtIERpZ2VzdCwgVm9sIDEy?= =?gb2312?b?LCBJc3N1ZSAxMQ==?= Message-ID: <749950.91682.qm@web15001.mail.cnb.yahoo.com> 1. The criteria of relax ? (wangqj1) dear pwscf users I want to know the criteria of relax ,after I completing one relax ,I don't know whether the result conform to request or not .How to set the key parameter of relax , e.g ,the trust_radius_max,trust_radius_min,trust_radius_ini,and so on .what's the basis of set them ? I long for someone to help me . Thank you . ----------------------------------------------------------------------- the criteria of relax? is Total foce equal 0. Jiming An ___________________________________________________________ ???????????? http://cn.mail.yahoo.com/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080609/627fa21f/attachment.htm From tone.kokalj at ijs.si Mon Jun 9 14:52:31 2008 From: tone.kokalj at ijs.si (Tone Kokalj) Date: Mon, 09 Jun 2008 14:52:31 +0200 Subject: [Pw_forum] Xcysden cann't give all atoms in the conventional cell. In-Reply-To: <200806091936233323357@yahoo.com.cn> References: <200806081946261265750@yahoo.com.cn> <200806091936233323357@yahoo.com.cn> Message-ID: <1213015951.18909.14.camel@walk.ijs.si> On Mon, 2008-06-09 at 19:36 +0800, zhaohscas wrote: > Hi all, > > I noticed the following things in Xcysden: it cann't give all atoms in the conventional cell based on the > PRIMVEC, CONVVEC, and PRIMCOORD parameters only. In detail, the space group number of a specific crystal > is of importance in determining the atoms in a cell. From the Xcrysen's descriptions, it just use the > above three parameters to generate the atoms in the conventional cell. So, I think it is inadequate to > decide all the atoms in the conventional cell corresponding to a primitive cell. First, note that the PRIMVEC and CONVVEC share the same origin (0,0,0). If one have ALL (and not only irreducible ones) atoms within the primitive unit cell and the associated latice vectors, then an infinite perfect crystal can be generated. Once the crystal is build there is an infinite number of choices for the cell. One can pick the one he/she likes. Some will be bigger, some will be smaller. From this perspective convetional cell is nothing but a small supercell. So the primvec/convec/primcoord should be sufficient. If you have find a structure that xcrysden fails to generate proper convetional-cell build structure, I woule appriciate if you send me the associated file, so that I can investigate. Best regards, Tone PS: if you would like to know how the alrgorithm works, here is the simplified recipe: xcrysden generate "big-enough" portion of the crystal based on primvec/primcoord, and then determines the atoms within the convvec specified cell. The program of course performs some checks that convvec are OK. -- Tone Kokalj J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia (tel: +386-1-477-3523 // fax:+386-1-477-3822) From yccheng.nju at gmail.com Mon Jun 9 15:46:42 2008 From: yccheng.nju at gmail.com (=?UTF-8?B?56iL6L+O5pil?=) Date: Mon, 9 Jun 2008 21:46:42 +0800 Subject: [Pw_forum] band structure Message-ID: Dear pwscf users, I have a problem about the band structure. When I calculate the band structure of an kind of semiconductor, I found an inconsistent problem between using conventional cell and primitive cell. The conventional cell shows the direct band gap ,while the primitive one shows indirect band gap. My Question: what kind of band gap of this semiconductor is. Any suggestion is appreciated. Thank you in advance! -- Y. C. Cheng Department of Phyics Nanjing University Nanjing 210093 P. R. China Tel: 86-25-83592907 Email: yccheng.nju at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080609/81016c9f/attachment.htm From tone.kokalj at ijs.si Mon Jun 9 18:02:14 2008 From: tone.kokalj at ijs.si (Tone Kokalj) Date: Mon, 09 Jun 2008 18:02:14 +0200 Subject: [Pw_forum] A bug in the pp.x? In-Reply-To: <200806082329365953078@yahoo.com.cn> References: <200806081946261265750@yahoo.com.cn> <200806082329365953078@yahoo.com.cn> Message-ID: <1213027334.18909.38.camel@walk.ijs.si> On Sun, 2008-06-08 at 23:29 +0800, zhaohscas wrote: > >Hi chopstick ;-), > > Aha, I'm using both forks and chopsticks once in a while. > > >> In the abobe case, the e1 \bot e2, and e2 \bot e3, but e1 isn' > >> perpendicular to e3. So, this will violate orthogonal condition. > > > >You are right: the actual check performed by the code is: > > e_i \cdot \e_j > 1.d-6 > >which wrongly allows non-orthogonal vectors if their pairwise scalar > >products are negative. Thank you for reporting, it will be fixed in the > >next release. > > But, in my mind, as for the study of the distribution of the charge density, we generallly want to > have the distribution w.r.t a specific miller plane. In the periodic crystal system, this can be > represented on the plane determined by a parallelogram whose two sides are the corresponding > primitive plane vectors. If we consider this requirement, I think we should use miller plane > vectors to contruct the grid, which are not always perpendicular indeed. Under the above consideration, > the orthogonal condition will be not so good. I believe the primary reason for orthogonality constraint was/is that not all drawing software can handle non-orthogonal mesh. Otherwise your considerations are justified: in many instances non-orthogonal 2D region would make sense. I belive orthogonality constraint could be lifted at least for XSF formatted output file. Regards, Tone -- Tone Kokalj J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia (tel: +386-1-477-3523 // fax:+386-1-477-3822) From asafis at yahoo.com.br Mon Jun 9 18:04:15 2008 From: asafis at yahoo.com.br (=?iso-8859-1?Q?=C1lvaro_Alves?=) Date: Mon, 9 Jun 2008 09:04:15 -0700 (PDT) Subject: [Pw_forum] Gamma point versus spin-orbit calculation Message-ID: <209357.88893.qm@web52304.mail.re2.yahoo.com> Hi all. Is there any incompatibility between the gamma point and spin-orbit calculation ?                        Regards,  ?lvaro Abra sua conta no Yahoo! Mail, o ?nico sem limite de espa?o para armazenamento! http://br.mail.yahoo.com/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080609/8f65c8ad/attachment.htm From zhaohscas at yahoo.com.cn Mon Jun 9 18:27:51 2008 From: zhaohscas at yahoo.com.cn (zhaohscas) Date: Tue, 10 Jun 2008 00:27:51 +0800 Subject: [Pw_forum] Xcysden cann't give all atoms in theconventional cell. References: <200806081946261265750@yahoo.com.cn>, <200806091936233323357@yahoo.com.cn>, <1213015951.18909.14.camel@walk.ijs.si> Message-ID: <200806100027489102810@yahoo.com.cn> >If you have find a structure that xcrysden fails to generate proper >convetional-cell build structure, I woule appriciate if you send me the >associated file, so that I can investigate. I give you the folllowing xsf file of Si, which is generated by me: ------------------------ CRYSTAL PRIMVEC 3.819990 0.000000 0.000000 1.909995 3.308208 0.000000 1.909995 1.102736 3.119009 CONVVEC 0 4.41094429 3.11900862 3.81998981 -2.20547214 3.11900862 3.81998981 2.20547214 -3.11900862 PRIMCOORD 9 1 14 1.909995 1.102736 0.779752 0.000000 0.000000 0.000000 14 0 0 0 14 1.9100 1.1027 3.1190 14 1.9100 3.3082 0 14 3.8200 4.4109 3.1190 14 3.8200 0 0 14 5.7300 1.1027 3.1190 14 5.7300 3.3082 0 14 7.6400 4.4109 3.1190 -------------------------- Using the above xsf file, the xcrysden fails to generate one of the face-centered atom. Please check it. Best regards, Honsheng. P.S. I also give the above xsf in the attachment, FYI. -------------- next part -------------- A non-text attachment was scrubbed... Name: si-conventional-cell.xsf Type: application/octet-stream Size: 618 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080610/4c3a4127/attachment.obj From zhaohscas at yahoo.com.cn Mon Jun 9 18:27:51 2008 From: zhaohscas at yahoo.com.cn (zhaohscas) Date: Tue, 10 Jun 2008 00:27:51 +0800 Subject: [Pw_forum] Xcysden cann't give all atoms in theconventional cell. References: <200806081946261265750@yahoo.com.cn>, <200806091936233323357@yahoo.com.cn>, <1213015951.18909.14.camel@walk.ijs.si> Message-ID: <200806100027489102810@yahoo.com.cn> >If you have find a structure that xcrysden fails to generate proper >convetional-cell build structure, I woule appriciate if you send me the >associated file, so that I can investigate. I give you the folllowing xsf file of Si, which is generated by me: ------------------------ CRYSTAL PRIMVEC 3.819990 0.000000 0.000000 1.909995 3.308208 0.000000 1.909995 1.102736 3.119009 CONVVEC 0 4.41094429 3.11900862 3.81998981 -2.20547214 3.11900862 3.81998981 2.20547214 -3.11900862 PRIMCOORD 9 1 14 1.909995 1.102736 0.779752 0.000000 0.000000 0.000000 14 0 0 0 14 1.9100 1.1027 3.1190 14 1.9100 3.3082 0 14 3.8200 4.4109 3.1190 14 3.8200 0 0 14 5.7300 1.1027 3.1190 14 5.7300 3.3082 0 14 7.6400 4.4109 3.1190 -------------------------- Using the above xsf file, the xcrysden fails to generate one of the face-centered atom. Please check it. Best regards, Honsheng. P.S. I also give the above xsf in the attachment, FYI. -------------- next part -------------- A non-text attachment was scrubbed... Name: si-conventional-cell.xsf Type: application/octet-stream Size: 618 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080610/4c3a4127/attachment-0001.obj From zhaohscas at yahoo.com.cn Mon Jun 9 18:37:19 2008 From: zhaohscas at yahoo.com.cn (zhaohscas) Date: Tue, 10 Jun 2008 00:37:19 +0800 Subject: [Pw_forum] Xcysden cann't give all atoms in theconventional cell. References: <200806081946261265750@yahoo.com.cn>, <200806091936233323357@yahoo.com.cn>, <1213015951.18909.14.camel@walk.ijs.si> Message-ID: <200806100037163632629@yahoo.com.cn> >If you have find a structure that xcrysden fails to generate proper >convetional-cell build structure, I woule appriciate if you send me the >associated file, so that I can investigate. I give you the folllowing xsf file of Si, which is generated by me: ------------------------ CRYSTAL PRIMVEC 3.819990 0.000000 0.000000 1.909995 3.308208 0.000000 1.909995 1.102736 3.119009 CONVVEC 0 4.41094429 3.11900862 3.81998981 -2.20547214 3.11900862 3.81998981 2.20547214 -3.11900862 PRIMCOORD 9 1 14 1.909995 1.102736 0.779752 0.000000 0.000000 0.000000 14 0 0 0 14 1.9100 1.1027 3.1190 14 1.9100 3.3082 0 14 3.8200 4.4109 3.1190 14 3.8200 0 0 14 5.7300 1.1027 3.1190 14 5.7300 3.3082 0 14 7.6400 4.4109 3.1190 -------------------------- Using the above xsf file, the xcrysden fails to generate one of the face-centered atom. Please check it. Best regards, Honsheng. P.S. I also give the above xsf and the snapshot of the conventional cell in the attachment, FYI. -------------- next part -------------- A non-text attachment was scrubbed... Name: si-conventional-cell.xsf Type: application/octet-stream Size: 618 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080610/ea886731/attachment-0001.obj -------------- next part -------------- A non-text attachment was scrubbed... Name: si-conventional-cell.jpg Type: image/jpeg Size: 12492 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080610/ea886731/attachment-0001.jpg From zhaohscas at yahoo.com.cn Mon Jun 9 18:37:19 2008 From: zhaohscas at yahoo.com.cn (zhaohscas) Date: Tue, 10 Jun 2008 00:37:19 +0800 Subject: [Pw_forum] Xcysden cann't give all atoms in theconventional cell. References: <200806081946261265750@yahoo.com.cn>, <200806091936233323357@yahoo.com.cn>, <1213015951.18909.14.camel@walk.ijs.si> Message-ID: <200806100037163632629@yahoo.com.cn> >If you have find a structure that xcrysden fails to generate proper >convetional-cell build structure, I woule appriciate if you send me the >associated file, so that I can investigate. I give you the folllowing xsf file of Si, which is generated by me: ------------------------ CRYSTAL PRIMVEC 3.819990 0.000000 0.000000 1.909995 3.308208 0.000000 1.909995 1.102736 3.119009 CONVVEC 0 4.41094429 3.11900862 3.81998981 -2.20547214 3.11900862 3.81998981 2.20547214 -3.11900862 PRIMCOORD 9 1 14 1.909995 1.102736 0.779752 0.000000 0.000000 0.000000 14 0 0 0 14 1.9100 1.1027 3.1190 14 1.9100 3.3082 0 14 3.8200 4.4109 3.1190 14 3.8200 0 0 14 5.7300 1.1027 3.1190 14 5.7300 3.3082 0 14 7.6400 4.4109 3.1190 -------------------------- Using the above xsf file, the xcrysden fails to generate one of the face-centered atom. Please check it. Best regards, Honsheng. P.S. I also give the above xsf and the snapshot of the conventional cell in the attachment, FYI. -------------- next part -------------- A non-text attachment was scrubbed... Name: si-conventional-cell.xsf Type: application/octet-stream Size: 618 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080610/ea886731/attachment-0002.obj -------------- next part -------------- A non-text attachment was scrubbed... Name: si-conventional-cell.jpg Type: image/jpeg Size: 12492 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080610/ea886731/attachment-0002.jpg From akohlmey at cmm.chem.upenn.edu Mon Jun 9 18:06:16 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Mon, 9 Jun 2008 12:06:16 -0400 (EDT) Subject: [Pw_forum] Xcysden cann't give all atoms in theconventional cell. In-Reply-To: <200806100037163632629@yahoo.com.cn> References: <200806081946261265750@yahoo.com.cn>, <200806091936233323357@yahoo.com.cn>, <1213015951.18909.14.camel@walk.ijs.si> <200806100037163632629@yahoo.com.cn> Message-ID: On Tue, 10 Jun 2008, zhaohscas wrote: hongsheng, have you considered, that this "missing" atom could just be "too far away" (due to numerical inaccuracy) so that the code building the conventional cell, doesn't consider it on the boundary anymore but in the next cell. please see tone's explanation of how the algorithm works. cheers, axel. ps.: can you please fix your mailer to not send n+1 copies for the same mail. thanks. a. HZ> >If you have find a structure that xcrysden fails to generate proper HZ> >convetional-cell build structure, I woule appriciate if you send me the HZ> >associated file, so that I can investigate. HZ> HZ> I give you the folllowing xsf file of Si, which is generated by me: HZ> HZ> ------------------------ HZ> CRYSTAL HZ> PRIMVEC HZ> 3.819990 0.000000 0.000000 HZ> 1.909995 3.308208 0.000000 HZ> 1.909995 1.102736 3.119009 HZ> CONVVEC HZ> 0 4.41094429 3.11900862 HZ> 3.81998981 -2.20547214 3.11900862 HZ> 3.81998981 2.20547214 -3.11900862 HZ> PRIMCOORD HZ> 9 1 HZ> 14 1.909995 1.102736 0.779752 0.000000 0.000000 0.000000 HZ> 14 0 0 0 HZ> 14 1.9100 1.1027 3.1190 HZ> 14 1.9100 3.3082 0 HZ> 14 3.8200 4.4109 3.1190 HZ> 14 3.8200 0 0 HZ> 14 5.7300 1.1027 3.1190 HZ> 14 5.7300 3.3082 0 HZ> 14 7.6400 4.4109 3.1190 HZ> -------------------------- HZ> HZ> Using the above xsf file, the xcrysden fails to generate one of the face-centered atom. HZ> Please check it. HZ> HZ> Best regards, HZ> Honsheng. HZ> HZ> P.S. I also give the above xsf and the snapshot of the conventional cell in the attachment, FYI. HZ> -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From baroni at sissa.it Mon Jun 9 20:12:02 2008 From: baroni at sissa.it (Stefano Baroni) Date: Mon, 9 Jun 2008 20:12:02 +0200 Subject: [Pw_forum] band structure In-Reply-To: References: Message-ID: Dear YC Cheng, rather than answering your question, I will give you a hint, and let you answer yourself. I do not know what "conventional cell" means, but I wil assume that for a semiconductor with an FCC Bravais lattice (say, with the diamond or zincblende structure) you mean the smallest simple cubic cell that can tile your crystal structure. Take Silicon, for instance. The point X={2\pi\over a}(100) is the X point of the FCC crystal structure, but it is a reciprocal-lattice vector of the SC one. The bottom of the conduction band is on the \Delta (\Gamma-X) line, very near to the X point, whereas the top of the valence band is at \Gamma. the gap is therefore indirect. What happens if you consider the crystal as SC, rather than FCC? Please, post your answer to this forum. Cheers - Stefano B On Jun 9, 2008, at 3:46 PM, ??? wrote: > Dear pwscf users, > I have a problem about the band structure. When I calculate > the band structure of an kind of semiconductor, I found an > inconsistent problem between using conventional cell and primitive > cell. The conventional cell shows the direct band gap ,while the > primitive one shows > indirect band gap. My Question: what kind of band gap of this > semiconductor is. Any suggestion is appreciated. > Thank you in advance! > > -- > Y. C. Cheng > Department of Phyics > Nanjing University > Nanjing 210093 > P. R. China > Tel: 86-25-83592907 > Email: yccheng.nju at gmail.com > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste http://www.sissa.it/~baroni / [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget 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/20080609/b34eff9a/attachment-0001.htm From zhaohscas at yahoo.com.cn Tue Jun 10 05:55:48 2008 From: zhaohscas at yahoo.com.cn (zhaohscas) Date: Tue, 10 Jun 2008 11:55:48 +0800 Subject: [Pw_forum] The erro of pwgui on windows by using the activetcl: can not find channel named... Message-ID: <200806101155457033091@yahoo.com.cn> Hi all, I installed the PWgui-4.0.1 on my widows xp with ActiveTcl8.4.19.0.285137. When I start the PWgui-4.0.1 to open a file, it will always give the following errors: ---------------------------------- can not find channel named "file1194b30" can not find channel named "file1194b30" while executing "close $fileChannel" (object "::gUI0" method "::guib::GUI::_openInput" body line 37) invoked from within "$cmd $moduleIdent $moduleName $moduleFile" (object "::gUI0" method "::guib::GUI::_toolbarCmd" body line 3) invoked from within "::gUI0 _toolbarCmd .gUI0_toolbaropen _openInput pp PP.X C:/PWgui-4.0.1/modules/pp/pp.tcl" (in namespace inscope "::guib::GUI" script line 1) invoked from within "namespace inscope ::guib::GUI {::gUI0 _toolbarCmd .gUI0_toolbaropen _openInput pp PP.X C:/PWgui-4.0.1/modules/pp/pp.tcl}" invoked from within ".gUI0_toolbaropen.2 invoke" ("uplevel" body line 1) invoked from within "uplevel #0 [list $w invoke]" (procedure "tk::ButtonUp" line 24) invoked from within "tk::ButtonUp .gUI0_toolbaropen.2" (command bound to event) ---------------------------------- Who can give me some hints on solving this issue? Thanks in advance. Hongsheng. __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From yccheng.nju at gmail.com Tue Jun 10 07:21:40 2008 From: yccheng.nju at gmail.com (=?UTF-8?B?56iL6L+O5pil?=) Date: Tue, 10 Jun 2008 13:21:40 +0800 Subject: [Pw_forum] band structure In-Reply-To: References: Message-ID: What happens if you consider the crystal as SC, rather than FCC? > The gap is direct when the crystal is considered as SC. The bottom of conduction band and the top of valence band are both at Gamma. The first Brillouin zone of the FCC is larger than that of the SC. The reduction of the brillouin zone of the SC crystal may induce the K point near X in FCC superpose with Gamma in SC, so the gap of SC is direct . Am I right? -- Y. C. Cheng Department of Phyics Nanjing University Nanjing 210093 P. R. China Tel: 86-25-83592907 Email: yccheng.nju at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080610/d9ede866/attachment.htm From baroni at sissa.it Tue Jun 10 08:07:08 2008 From: baroni at sissa.it (Stefano Baroni) Date: Tue, 10 Jun 2008 08:07:08 +0200 Subject: [Pw_forum] band structure In-Reply-To: References: Message-ID: <21259D9B-4B8C-44EB-99A9-B184253737B5@sissa.it> Tha facts are right, the explanation is missing (or not clearly stated) The fact is, the SC Gamma pooint (or any other k vector, for that matter) contains four times as many electron states as the FCC Gamma point (have you noticed this?). This is because the BZ is four times as small, and the total number of k point in the BZ cannot depend on one's ability to identify the proper crystal symmetries. Can you see which FCC k points are *folded* into the SC gamma point? SB On Jun 10, 2008, at 7:21 AM, ??? wrote: > > > What happens if you consider the crystal as SC, rather than FCC? > > The gap is direct when the crystal is considered as SC. The bottom > of conduction band and the top of valence band are both at Gamma. > The first Brillouin zone of the FCC is larger than that of the SC. > The reduction of the brillouin zone of the SC crystal may induce the > K point near X in FCC superpose with Gamma in SC, so the gap of SC > is direct . > Am I right? > > > -- > Y. C. Cheng > Department of Phyics > Nanjing University > Nanjing 210093 > P. R. China > Tel: 86-25-83592907 > Email: yccheng.nju at gmail.com > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste http://www.sissa.it/~baroni / [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget 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/20080610/59ffafce/attachment.htm From yccheng.nju at gmail.com Tue Jun 10 10:00:00 2008 From: yccheng.nju at gmail.com (=?UTF-8?B?56iL6L+O5pil?=) Date: Tue, 10 Jun 2008 16:00:00 +0800 Subject: [Pw_forum] band structure In-Reply-To: <21259D9B-4B8C-44EB-99A9-B184253737B5@sissa.it> References: <21259D9B-4B8C-44EB-99A9-B184253737B5@sissa.it> Message-ID: > Can you see which FCC k points are *folded* into the SC gamma point? > Sorry, I can't see which FCC k points are folded into the SC gamma point. I failed to find the relationship of the BZ between FCC and SC, because I can't put the two BZ with the same Gamma. Is there any way to find the folded k points? -- Y. C. Cheng Department of Phyics Nanjing University Nanjing 210093 P. R. China Tel: 86-25-83592907 Email: yccheng.nju at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080610/b29dbec8/attachment.htm From tone.kokalj at ijs.si Tue Jun 10 10:05:48 2008 From: tone.kokalj at ijs.si (Tone Kokalj) Date: Tue, 10 Jun 2008 10:05:48 +0200 Subject: [Pw_forum] Xcysden cann't give all atoms in theconventional cell. In-Reply-To: <200806100027489102810@yahoo.com.cn> References: <200806081946261265750@yahoo.com.cn> , <200806091936233323357@yahoo.com.cn> , <1213015951.18909.14.camel@walk.ijs.si> <200806100027489102810@yahoo.com.cn> Message-ID: <1213085148.9841.16.camel@walk.ijs.si> On Tue, 2008-06-10 at 00:27 +0800, zhaohscas wrote: > >If you have find a structure that xcrysden fails to generate proper > >convetional-cell build structure, I woule appriciate if you send me the > >associated file, so that I can investigate. Thanks for your file and picture. It is not that the structure is wrongly generated, but due to some round-off error, there is indeed one missing atom in the "nicely cut unit cell" display mode. However if you display 1x2x1 cells you will see that atom (see attached picture). Actually, the "missing" atom is the periodic replica in the 2nd cell (but is on the border plane of the 1st). I admit this is annoying. One more thing concerning the visualization of so-called "nicely cut unit cell". Consider for example a plain fcc structure. There are 4-atoms in the conventional unit cell, however the "nicely cut unit cell" mode displays 9 atoms, but 5 of them are periodic replicas ... Regards, Tone -- Tone Kokalj J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia (tel: +386-1-477-3523 // fax:+386-1-477-3822) -------------- next part -------------- A non-text attachment was scrubbed... Name: 2cell.jpg Type: image/jpeg Size: 12507 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080610/d36bf685/attachment.jpg From tone.kokalj at ijs.si Tue Jun 10 10:07:20 2008 From: tone.kokalj at ijs.si (Tone Kokalj) Date: Tue, 10 Jun 2008 10:07:20 +0200 Subject: [Pw_forum] The erro of pwgui on windows by using the activetcl: can not find channel named... In-Reply-To: <200806101155457033091@yahoo.com.cn> References: <200806101155457033091@yahoo.com.cn> Message-ID: <1213085240.9841.20.camel@walk.ijs.si> On Tue, 2008-06-10 at 11:55 +0800, zhaohscas wrote: > Hi all, > > I installed the PWgui-4.0.1 on my widows xp with ActiveTcl8.4.19.0.285137. > When I start the PWgui-4.0.1 to open a file, it will always give the following errors: > > ---------------------------------- > can not find channel named "file1194b30" > can not find channel named "file1194b30" > while executing > "close $fileChannel" > > Who can give me some hints on solving this issue? Actually, I cannot, because, honesty, I have no idea what is going on. What does it say when you want to save a file? Does it produce similar error? Regards, Tone -- Tone Kokalj J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia (tel: +386-1-477-3523 // fax:+386-1-477-3822) From tone.kokalj at ijs.si Tue Jun 10 10:09:36 2008 From: tone.kokalj at ijs.si (Tone Kokalj) Date: Tue, 10 Jun 2008 10:09:36 +0200 Subject: [Pw_forum] Xcysden cann't give all atoms in theconventional cell. In-Reply-To: <1213085148.9841.16.camel@walk.ijs.si> References: <200806081946261265750@yahoo.com.cn> , <200806091936233323357@yahoo.com.cn> , <1213015951.18909.14.camel@walk.ijs.si> <200806100027489102810@yahoo.com.cn> <1213085148.9841.16.camel@walk.ijs.si> Message-ID: <1213085376.9841.22.camel@walk.ijs.si> On Tue, 2008-06-10 at 10:05 +0200, Tone Kokalj wrote: > One more thing concerning the visualization of so-called "nicely cut > unit cell". Consider for example a plain fcc structure. There are > 4-atoms in the conventional unit cell, however the "nicely cut unit > cell" mode displays 9 atoms, but 5 of them are periodic replicas ... Actually 14 (and not 9), and of course 10 of them are periodic replicas ... -- Tone Kokalj J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia (tel: +386-1-477-3523 // fax:+386-1-477-3822) From degironc at sissa.it Tue Jun 10 10:16:49 2008 From: degironc at sissa.it (Stefano de Gironcoli) Date: Tue, 10 Jun 2008 10:16:49 +0200 Subject: [Pw_forum] band structure In-Reply-To: References: <21259D9B-4B8C-44EB-99A9-B184253737B5@sissa.it> Message-ID: <484E3871.5020203@sissa.it> dear YC Cheng ??? wrote: > > Can you see which FCC k points are *folded* into the SC gamma point? > > > Sorry, I can't see which FCC k points are folded into the SC gamma > point. I failed to find the relationship of the BZ between FCC and SC, > because I can't put the two BZ with the same Gamma. Is there any way > to find the folded k points? write down the general plane wave |PW> as a |k_FCC + G_FCC> and also as |k_SC + G_SC>... and see which k_FCC's correspond to a given k_SC... that amounts to understand which G_SC are also G_FCC and which not. hope this helps. stefano From holger.hesske at chem.ethz.ch Tue Jun 10 11:53:59 2008 From: holger.hesske at chem.ethz.ch (Holger Hesske) Date: Tue, 10 Jun 2008 11:53:59 +0200 Subject: [Pw_forum] Problems concerning fhi2upf and cpmd2upf Message-ID: <200806101154.00087.holger.hesske@chem.ethz.ch> ? Dear pwscf-users, (1) I tried to convert a NC-PP for Nitrogen from the fhi-website. It fails with segfault (forrtl: severe (174): SIGSEGV, segmentation fault occurred). To test wheter it is a micompiled fhi2upf.x I convert several other PPs and it worked fine for Oxygen, Carbon, Barium, etc. but it fails also, if I choose an atom with odd electrons (fluorine/sodium; no semicore for all of them). The question arose: Is it my non-ability (probably), does it not work for that special case (I couldn't find another NC_PBE_PP) or is it a general bug (probably not)? Or even more interesting: How could I fix it and obtain a working Nitrogen_NC_PBE_PP in UPF format? (2) After failing with the fhi conversion, I tried to get one from the cpmd package (there is one in the pp_ext_package) and convert it to UPF (works well so far). But when starting test calculations the code stops and complains after/with: ------------------------------------ Current dimensions of program pwscf are: ? ? ?Max number of different atomic species (ntypx) = 10 ? ? ?Max number of k-points (npk) = ?40000 ? ? ?Max angular momentum in pseudopotentials (lmaxx) = ?3 %%%%%%%%%%%%%%%%%%%%%%%%% ? ? ?task # ? ? ? ? 4 ? ? ?from set_dft_value : error # ? ? ? ? 1 ? ? ?two conflicting matching values ?%%%%%%%%%%%%%%%%%%%%%%%% -------------------------------------- I tried to attached the "original" PPs and the cpmd2UPF converted one for reproduction tests but the mail is to big to be send directly to the list. In any case every kind of help is appreciated! Thanks in advance, -- Dr. Holger Hesske Inst.f.Chemie-/Bioingenieurwissenschaft. Wolfgang-Pauli-Str. 10 ETH H?nggerberg, HCI E 133 CH-8093 Z?rich Tel: +41 44 633 66 19 Fax: +41 44 632 11 63 E-Mail: holger.hesske at chem.ethz.ch From akohlmey at cmm.chem.upenn.edu Tue Jun 10 11:29:47 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Tue, 10 Jun 2008 05:29:47 -0400 (EDT) Subject: [Pw_forum] Problems concerning fhi2upf and cpmd2upf In-Reply-To: <200806101154.00087.holger.hesske@chem.ethz.ch> References: <200806101154.00087.holger.hesske@chem.ethz.ch> Message-ID: On Tue, 10 Jun 2008, Holger Hesske wrote: HH> Dear pwscf-users, HH> HH> (1) I tried to convert a NC-PP for Nitrogen from the fhi-website. It fails HH> with segfault (forrtl: severe (174): SIGSEGV, segmentation fault occurred). HH> To test wheter it is a micompiled fhi2upf.x I convert several other PPs and it HH> worked fine for Oxygen, Carbon, Barium, etc. but it fails also, if I choose HH> an atom with odd electrons (fluorine/sodium; no semicore for all of them). HH> The question arose: Is it my non-ability (probably), does it not work for that HH> special case (I couldn't find another NC_PBE_PP) or is it a general bug HH> (probably not)? Or even more interesting: How could I fix it and obtain a HH> working Nitrogen_NC_PBE_PP in UPF format? the fhi2upf.x converter always has been the most problematic one... HH> (2) After failing with the fhi conversion, I tried to get one from the cpmd HH> package (there is one in the pp_ext_package) and convert it to UPF (works HH> well so far). But when starting test calculations the code stops and HH> complains after/with: HH> ------------------------------------ HH> Current dimensions of program pwscf are: HH> ? ? ?Max number of different atomic species (ntypx) = 10 HH> ? ? ?Max number of k-points (npk) = ?40000 HH> ? ? ?Max angular momentum in pseudopotentials (lmaxx) = ?3 HH> HH> %%%%%%%%%%%%%%%%%%%%%%%%% HH> ? ? ?task # ? ? ? ? 4 HH> ? ? ?from set_dft_value : error # ? ? ? ? 1 HH> ? ? ?two conflicting matching values HH> ?%%%%%%%%%%%%%%%%%%%%%%%% cpmd uses a special encoding for PBE (it has exchange and correlation in one special subroutine). you can fix the resulting pseudopotentia with a text editor by changing SLA PW PBE PBE in the UPF file into SLA PW PBX PBC cheers, axel. HH> -------------------------------------- HH> I tried to attached the "original" PPs and the cpmd2UPF converted one for HH> reproduction tests but the mail is to big to be send directly to the list. HH> HH> HH> In any case every kind of help is appreciated! HH> Thanks in advance, HH> -- ======================================================================= 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 zhaohscas at yahoo.com.cn Tue Jun 10 13:54:35 2008 From: zhaohscas at yahoo.com.cn (zhaohscas) Date: Tue, 10 Jun 2008 19:54:35 +0800 Subject: [Pw_forum] Can the Xcrysden give the values labels for the 2D charge density coutour plots? References: <200806101154.00087.holger.hesske@chem.ethz.ch>, Message-ID: <200806101954314682407@yahoo.com.cn> Hi all, I've use the Xcrysden to generate the coutour plots for the 2D charge density coutour plots of my system. Nowm, I want to give the charge density values as the labels for some of the contour lines in the plot, who can give mes some hints? Sincerely yours, Hongsheng. ps. The attachments are the 2D charge density coutour plot and the the corresponding setting for it. ------------------ zhaohscas 2008-06-10 -------------- next part -------------- A non-text attachment was scrubbed... Name: contour1.png Type: application/octet-stream Size: 11211 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080610/35f2dba0/attachment-0002.obj -------------- next part -------------- A non-text attachment was scrubbed... Name: contour2.png Type: application/octet-stream Size: 6362 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080610/35f2dba0/attachment-0003.obj From tone.kokalj at ijs.si Tue Jun 10 14:43:52 2008 From: tone.kokalj at ijs.si (Tone Kokalj) Date: Tue, 10 Jun 2008 14:43:52 +0200 Subject: [Pw_forum] Can the Xcrysden give the values labels for the 2D charge density coutour plots? In-Reply-To: <200806101954314682407@yahoo.com.cn> References: <200806101154.00087.holger.hesske@chem.ethz.ch> , <200806101954314682407@yahoo.com.cn> Message-ID: <1213101832.9841.29.camel@walk.ijs.si> On Tue, 2008-06-10 at 19:54 +0800, zhaohscas wrote: > Hi all, > > I've use the Xcrysden to generate the coutour plots for the 2D charge density coutour plots of my system. > Nowm, I want to give the charge density values as the labels for some of the contour lines in the plot, > who can give mes some hints? xcrysden cannot do it. It can print a separate color "thermometer" instead (see: http://www.xcrysden.org/doc/plane1control.html#__toc__6 ) As an alternative, you can use some image manipulation program, where you put the labels manually. Regards, Tone -- Tone Kokalj J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia (tel: +386-1-477-3523 // fax:+386-1-477-3822) From ylli at theory.issp.ac.cn Wed Jun 11 02:15:10 2008 From: ylli at theory.issp.ac.cn (Li yan-ling) Date: Wed, 11 Jun 2008 08:15:10 +0800 (CST) Subject: [Pw_forum] Electron-phonon interaction energy (EPIE). Message-ID: <1266.210.73.16.91.1213143310.squirrel@210.73.16.4> Dear all users, It is well known that the electron-phonon interaction plays an important role in superconductors. Now, I want to know how to calculate (or obtain) the enthalpic contributions from electron-phonon coupling. Thanks in advance. Yanling Li. Institute of Solid state Physics, CAS, China. From ihsanas at yahoo.com Wed Jun 11 18:13:50 2008 From: ihsanas at yahoo.com (=?iso-8859-1?Q?=C7=CD=D3=C7=E4_=DA=D1=ED=DE=C7=CA?=) Date: Wed, 11 Jun 2008 09:13:50 -0700 (PDT) Subject: [Pw_forum] neb Message-ID: <53315.52445.qm@web52308.mail.re2.yahoo.com> Hi In the output of the neb what is the physical meaning of that the activation energy ->0.0000 and activation energy <-7.8eV is it a range that activation energy lie in between ... In example 17 at the end of the output neb file ->..=<-... my question activation energy that we have from the neb must have single value or it will be withen a range of values as ..< activation energy >.. Best regards Ihsan A.Erikat Jordan Physics department From degironc at sissa.it Wed Jun 11 18:31:19 2008 From: degironc at sissa.it (Stefano de Gironcoli) Date: Wed, 11 Jun 2008 18:31:19 +0200 Subject: [Pw_forum] neb In-Reply-To: <53315.52445.qm@web52308.mail.re2.yahoo.com> References: <53315.52445.qm@web52308.mail.re2.yahoo.com> Message-ID: <484FFDD7.50707@sissa.it> -> means from left to right (the froward reaction, from the first to the last image) <- means from right to left (the reverse reaction). If the initial and final configurations are equivalent (as in example 17) then -> = <- , otherwise it is unlikely to happen. hope this helps, stefano ????? ?????? wrote: > Hi > In the output of the neb what is the physical meaning of that the activation energy ->0.0000 and activation energy <-7.8eV is it a range that activation energy lie in between ... > In example 17 at the end of the output neb file ->..=<-... > my question activation energy that we have from the neb must have single value or it will be withen a range of values as ..< activation energy >.. > Best regards > Ihsan A.Erikat > Jordan Physics department > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From asafis at yahoo.com.br Wed Jun 11 22:50:34 2008 From: asafis at yahoo.com.br (=?iso-8859-1?Q?=C1lvaro_Alves?=) Date: Wed, 11 Jun 2008 13:50:34 -0700 (PDT) Subject: [Pw_forum] Xcrysden Message-ID: <335209.63017.qm@web52310.mail.re2.yahoo.com> Hello everyone. After calculating the polarization and create a file *. xsf, trying to open it in Xcrysden. However, the molecule appears damage. Does anyone know what is happening?                  Sincerely, Alvaro Abra sua conta no Yahoo! Mail, o ?nico sem limite de espa?o para armazenamento! http://br.mail.yahoo.com/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080611/c80fa475/attachment.htm From akohlmey at cmm.chem.upenn.edu Thu Jun 12 01:31:09 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Wed, 11 Jun 2008 19:31:09 -0400 (EDT) Subject: [Pw_forum] Xcrysden In-Reply-To: <335209.63017.qm@web52310.mail.re2.yahoo.com> References: <335209.63017.qm@web52310.mail.re2.yahoo.com> Message-ID: On Wed, 11 Jun 2008, ?lvaro Alves wrote: AA> Hello everyone. After calculating the polarization and create a file *. hello alvaro, AA> xsf, trying to open it in Xcrysden. However, the molecule appears AA> damage. Does anyone know what is happening? how should we? if you don't provide a way to reproduce or demonstrate what you are seeing, and explain what you understand under "damaged", it will be difficult to help you. cheers, axel. BTW: you mail looks "damaged", too. please configure your mail client to send plain ascii text and add a signature with your name and affiliation. thanks. AA> AA>                  Sincerely, Alvaro -- ======================================================================= 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 lfhuang at theory.issp.ac.cn Thu Jun 12 08:00:53 2008 From: lfhuang at theory.issp.ac.cn (lfhuang at theory.issp.ac.cn) Date: Thu, 12 Jun 2008 14:00:53 +0800 (CST) Subject: [Pw_forum] eigenvalues not converged in ph.x Message-ID: <33378.210.73.17.131.1213250453.squirrel@210.73.16.4> Dear all pwscf friends: Although there are so much messages on "WARNING: # eigenvalues not converged" in pw.x, this problem in ph.x is not frequently talked. This time, I calculate the phonon dispersions in graphene. The convergence can be easily obtained in pw.x without any warnings, but when ph.x is executed, there always appear hundreds warning lines: G cutoff = 215.4158 ( 47155 G-vectors) FFT grid: ( 30, 30,125) G cutoff = 86.1663 ( 11933 G-vectors) smooth grid: ( 20, 20, 80) nbndx = 32 nbnd = 8 natomwfc = 8 npwx = 152 nelec = 8.00 nkb = 16 ngl = 1449 The potential is recalculated from file : graphene.save/charge-density.xml negative rho (up, down): 0.326E-05 0.000E+00 Starting wfc are atomic total cpu time spent up to now is 387.33 secs Band Structure Calculation Davidson diagonalization with overlap WARNING: 2 eigenvalues not converged WARNING: 1 eigenvalues not converged ................... (nearly 900 lines!!!) ................... WARNING: 2 eigenvalues not converged WARNING: 1 eigenvalues not converged ethr = 1.25E-10, avg # of iterations = 23.1 total cpu time spent up to now is 732.55 secs End of band structure calculation I have taken so much pain testing according to the userguide and some message on the mailing list(maybe those discriptions are not exactly suitable for ph.x). Or maybe there is something important/subtle I have missed. Thanks a lot for reading my letter! Appreciate greatly the one who can help me!! From min0220 at postech.ac.kr Thu Jun 12 08:31:04 2008 From: min0220 at postech.ac.kr (Min Seung Kyu) Date: Thu, 12 Jun 2008 15:31:04 +0900 (KST) Subject: [Pw_forum] converting CPMD pp's to UPF Message-ID: Dear users and developers, I converted CPMD pseudopotentials to UPF, and I optimized geometries of water and benzene for pseudopotential test. However, during the optimization, the bond length were getting longer both LDA and GGA cases. Does anyone succeed in converting CPMD pseudopotential to UPFs? Thanks. Sincerely, Min. -------------------------------Sig.--------------------------- Center for Superfunctional Materials, Department of Chemistry, Pohang University of Science and Technology (POSTECH), S. Korea Mr. Seung Kyu, Min Ph.D. Candidate E-Mail: min0220 at postech.ac.kr Tel. : 82-54-279-5858 ---------------------------------------------- From w2agz at pacbell.net Thu Jun 12 09:45:33 2008 From: w2agz at pacbell.net (Paul M. Grant) Date: Thu, 12 Jun 2008 00:45:33 -0700 Subject: [Pw_forum] phq_readin and LDA+U Message-ID: <000001c8cc60$4cc673a0$e6535ae0$@net> To All: I just tried running ph.x on the scf output of some runs on antiferromagnetic transition metal oxides and got the following message: From eyvaz_isaev at yahoo.com Thu Jun 12 10:03:45 2008 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Thu, 12 Jun 2008 01:03:45 -0700 (PDT) Subject: [Pw_forum] phq_readin and LDA+U In-Reply-To: <000001c8cc60$4cc673a0$e6535ae0$@net> Message-ID: <624111.25993.qm@web65708.mail.ac4.yahoo.com> Hi Paul, As far as I know LDA+U and phonon calculations are not allowed yet, and as Matteo noticed before there are few bugs they are fighting. Hopefully Matteo can tell us more precisely about the status of LDA+U and phonons combination. Bests, Eyvaz. --- "Paul M. Grant" wrote: > To All: > > I just tried running ph.x on the scf output of some > runs on > antiferromagnetic transition metal oxides and got > the following message: > > From phq_readin error # 1 > The phonon code with LDA+U is not yet available > > I infer from the use of the adverb "yet" that such > code is in the works? I > see anticipatory flags exist phq_readin.f90. > > I could really use such a feature and would be quite > willing to help create > and test it. > > -Paul > > 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 > > > > > > _______________________________________________ > 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, Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com From w2agz at pacbell.net Thu Jun 12 12:38:36 2008 From: w2agz at pacbell.net (Paul Grant) Date: Thu, 12 Jun 2008 03:38:36 -0700 (PDT) Subject: [Pw_forum] phq_readin and LDA+U Message-ID: <946419.12409.qm@web82507.mail.mud.yahoo.com> Hi Eyvaz, Thanks for getting back so fast. When I saw Matteo at the APS Meeting in New Orleans in March, he warned me there could be "issues" and in a couple of subseqent e-mails also...but I decided to try anyway! In the meantime, I'll do a "rough fake" by calculating the total energy for a few values of lattice constant and U. The three of us should get together because I think there's an intriguing problem underneath all this. I'm at an MRS meeting in Chongqing right now and off to give a talk at the CAS/IP in Beijing next Monday. The bandwidth here is pretty narrow...I wonder why ;) Have a look at next week's Nature...I'll have an N&V on the ferrous oxypnictides. -Paul ----- Original Message ---- From: Eyvaz Isaev To: PWSCF Forum Sent: Thursday, June 12, 2008 1:03:45 AM Subject: Re: [Pw_forum] phq_readin and LDA+U Hi Paul, As far as I know LDA+U and phonon calculations are not allowed yet, and as Matteo noticed before there are few bugs they are fighting. Hopefully Matteo can tell us more precisely about the status of LDA+U and phonons combination. Bests, Eyvaz. --- "Paul M. Grant" wrote: > To All: > > I just tried running ph.x on the scf output of some > runs on > antiferromagnetic transition metal oxides and got > the following message: > > From phq_readin error # 1 > The phonon code with LDA+U is not yet available > > I infer from the use of the adverb "yet" that such > code is in the works? I > see anticipatory flags exist phq_readin.f90. > > I could really use such a feature and would be quite > willing to help create > and test it. > > -Paul > > 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 > > > > > > _______________________________________________ > 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, Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080612/7bd54ba9/attachment.htm From Jeremie.Teyssier at physics.unige.ch Thu Jun 12 13:20:09 2008 From: Jeremie.Teyssier at physics.unige.ch (=?iso-8859-1?Q?J=E9r=E9mie_Teyssier?=) Date: Thu, 12 Jun 2008 13:20:09 +0200 Subject: [Pw_forum] PB mixing PP produced with ld1 with virtual.x Message-ID: <000001c8cc7e$466a3410$d33e9c30$%Teyssier@physics.unige.ch> Dear all I have produced a PP for Bi that works pretty well. I would like to mix it with another PP but virtual returns this error message : ./virtual.x Generate the UPF pseudopotential for a virtual atom combining two pseudopootentials in UPF format Input PP file # 1 in UPF format > Bi.rel-pbe-rrkj.UPF IOS= 0 1 4 Reading pseudopotential file in UPF format... %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% from scan_begin : error # 1 No HEADER block %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% I is true that there is a different header here but the tool readUPF works Could someone help me? What is the normal format as an input for virtual? Thanks. J?r?mie Generated using "atomic" code by A. Dal Corso (espresso distribution) Author: JT Generation date: 11Jun2008 Pseudopotential type: NC Element: Bi Functional: PBE Suggested minimum cutoff for wavefunctions: 108. Ry Suggested minimum cutoff for charge density: 434. Ry The Pseudo was generated with a Fully-Relativistic Calculation L component and cutoff radius for Local Potential: 0 2.5000 Pseudopotential contains additional information for spin-orbit calculations. Valence configuration: nl pn l occ Rcut Rcut US E pseu 6P 2 1 2.00 3.200 3.200 -0.408528 6P 2 1 1.00 3.200 3.200 -0.272552 5D 3 2 4.00 1.600 1.600 -2.056590 5D 3 2 6.00 1.600 1.600 -1.836857 6S 1 0 2.00 2.256 2.256 -1.018341 J?r?mie Teyssier DPMC 24 quai E. Ensermet 1211 Geneve Switzerland phone : +41(0)22 379 35 68 fax : +41(0)22 379 35 02 email : jeremie.teyssier at physics.unige.ch -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080612/7ac34a06/attachment-0001.htm From marzari at MIT.EDU Thu Jun 12 14:37:51 2008 From: marzari at MIT.EDU (Nicola Marzari) Date: Thu, 12 Jun 2008 15:37:51 +0300 Subject: [Pw_forum] PB mixing PP produced with ld1 with virtual.x In-Reply-To: <000001c8cc7e$466a3410$d33e9c30$%Teyssier@physics.unige.ch> References: <000001c8cc7e$466a3410$d33e9c30$%Teyssier@physics.unige.ch> Message-ID: <4851189F.4080600@mit.edu> J?r?mie Teyssier wrote: > Dear all > > > > I have produced a PP for Bi that works pretty well. I would like to mix > it with another PP but virtual returns this error message : > > > > ./virtual.x > > > > Generate the UPF pseudopotential for a virtual atom > > combining two pseudopootentials in UPF format Dear Jeremie, hopefully someone will reply about your specific error message (looks to me just a matter of compatibility between reading upf formats). Do note though that virtual.x has some (simple) bugs - we are working on it these days. Be in touch with Jivtesh Garg . 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 ihsanas at yahoo.com Thu Jun 12 16:29:53 2008 From: ihsanas at yahoo.com (=?iso-8859-1?Q?=C7=CD=D3=C7=E4_=DA=D1=ED=DE=C7=CA?=) Date: Thu, 12 Jun 2008 07:29:53 -0700 (PDT) Subject: [Pw_forum] neb Message-ID: <166984.92508.qm@web52306.mail.re2.yahoo.com> Hi thank you for replying , I have another question from the output of the neb there is a file ---.int it consists of two columns .what is the meaning of each column . best regards Ihsan A.Erikat Jordan -physics department From prasenjit.jnc at gmail.com Thu Jun 12 16:37:53 2008 From: prasenjit.jnc at gmail.com (Prasenjit Ghosh) Date: Thu, 12 Jun 2008 16:37:53 +0200 Subject: [Pw_forum] neb In-Reply-To: <166984.92508.qm@web52306.mail.re2.yahoo.com> References: <166984.92508.qm@web52306.mail.re2.yahoo.com> Message-ID: <627e0ffa0806120737gc4d6d8ladefc50f15ce0a46@mail.gmail.com> The first column gives the position along the reaction coordinate while the second column gives the energy difference between the first image and the images corresponding to the position. Note that this is obtained by doing an interpolation. Prasenjit 2008/6/12 ????? ?????? : > Hi > thank you for replying , I have another question from the output of the neb > there is a file ---.int it consists of two columns .what is the meaning of > each column . > best regards > Ihsan A.Erikat > Jordan -physics department > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- PRASENJIT GHOSH, POST-DOC, ROOM NO: 265, MAIN BUILDING, CM SECTION, ICTP, STRADA COSTERIA 11, TRIESTE, 34104, ITALY PHONE: +39 040 2240 369 (O) +39 3807528672 (M) -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080612/1926692c/attachment.htm From zhaohscas at yahoo.com.cn Thu Jun 12 16:55:41 2008 From: zhaohscas at yahoo.com.cn (zhaohscas) Date: Thu, 12 Jun 2008 22:55:41 +0800 Subject: [Pw_forum] neb References: <166984.92508.qm@web52306.mail.re2.yahoo.com> Message-ID: <200806122255389247313@yahoo.com.cn> >The first column gives the position along the reaction coordinate while the >second column gives the energy difference between the first image and the >images corresponding to the position. Note that this is obtained by doing an >interpolation. Which type of interpolation method should be taken in order to obtain *good* values of these? ---------------- Sincerely yours, Hongsheng Zhao Xinjiang Technical Institute of Physics and Chemistry Chinese Academy of Sciences __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From prasenjit.jnc at gmail.com Thu Jun 12 17:19:10 2008 From: prasenjit.jnc at gmail.com (Prasenjit Ghosh) Date: Thu, 12 Jun 2008 17:19:10 +0200 Subject: [Pw_forum] neb In-Reply-To: <200806122255389247313@yahoo.com.cn> References: <166984.92508.qm@web52306.mail.re2.yahoo.com> <200806122255389247313@yahoo.com.cn> Message-ID: <627e0ffa0806120819j126c21c1q951952e5f4adc60a@mail.gmail.com> As far as I know, you don't have an option to choose the type of interpolation to obtain the data in the *.int file. The code uses a linear interpolation technique to obtain the above mentioned data. Prasenjit. 2008/6/12 zhaohscas : > >The first column gives the position along the reaction coordinate while > the > >second column gives the energy difference between the first image and the > >images corresponding to the position. Note that this is obtained by doing > an > >interpolation. > > Which type of interpolation method should be taken in order to obtain > *good* values of these? > > ---------------- > Sincerely yours, > Hongsheng Zhao > Xinjiang Technical Institute of Physics and Chemistry > Chinese Academy of Sciences > > __________________________________________________ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- PRASENJIT GHOSH, POST-DOC, ROOM NO: 265, MAIN BUILDING, CM SECTION, ICTP, STRADA COSTERIA 11, TRIESTE, 34104, ITALY PHONE: +39 040 2240 369 (O) +39 3807528672 (M) -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080612/c6daa224/attachment.htm From degironc at sissa.it Thu Jun 12 17:22:40 2008 From: degironc at sissa.it (Stefano de Gironcoli) Date: Thu, 12 Jun 2008 17:22:40 +0200 Subject: [Pw_forum] neb In-Reply-To: <627e0ffa0806120819j126c21c1q951952e5f4adc60a@mail.gmail.com> References: <166984.92508.qm@web52306.mail.re2.yahoo.com> <200806122255389247313@yahoo.com.cn> <627e0ffa0806120819j126c21c1q951952e5f4adc60a@mail.gmail.com> Message-ID: <48513F40.4060406@sissa.it> let me add that the .int file is for plotting purpose only. The actual data are the ones in the corresponding .dat file. stefano Prasenjit Ghosh wrote: > As far as I know, you don't have an option to choose the type of > interpolation to obtain the data in the *.int file. The code uses a > linear interpolation technique to obtain the above mentioned data. > > Prasenjit. > > 2008/6/12 zhaohscas >: > > >The first column gives the position along the reaction coordinate > while the > >second column gives the energy difference between the first image > and the > >images corresponding to the position. Note that this is obtained > by doing an > >interpolation. > > Which type of interpolation method should be taken in order to > obtain *good* values of these? > > ---------------- > Sincerely yours, > Hongsheng Zhao > > Xinjiang Technical Institute of Physics and Chemistry > Chinese Academy of Sciences > > __________________________________________________ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > -- > PRASENJIT GHOSH, > POST-DOC, > ROOM NO: 265, MAIN BUILDING, > CM SECTION, ICTP, > STRADA COSTERIA 11, > TRIESTE, 34104, > ITALY > PHONE: +39 040 2240 369 (O) > +39 3807528672 (M) > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From tone.kokalj at ijs.si Thu Jun 12 17:29:01 2008 From: tone.kokalj at ijs.si (Tone Kokalj) Date: Thu, 12 Jun 2008 17:29:01 +0200 Subject: [Pw_forum] neb In-Reply-To: <627e0ffa0806120819j126c21c1q951952e5f4adc60a@mail.gmail.com> References: <166984.92508.qm@web52306.mail.re2.yahoo.com> <200806122255389247313@yahoo.com.cn> <627e0ffa0806120819j126c21c1q951952e5f4adc60a@mail.gmail.com> Message-ID: <1213284541.6084.43.camel@walk.ijs.si> On Thu, 2008-06-12 at 17:19 +0200, Prasenjit Ghosh wrote: > As far as I know, you don't have an option to choose the type of > interpolation to obtain the data in the *.int file. The code uses a > linear interpolation technique to obtain the above mentioned data. This is not true. As far as I know, the code uses the interpolation as suggested in the original NEB paper. This interpolation is kind of cubic interpolation, which uses the information also from the forces (=slope) provided by the code. According to my experience it's rather good (see the appendix in G. Henkelman, J. Chem. Phys. 113, 9978 (2000)). Regards, Tone -- Tone Kokalj J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia (tel: +386-1-477-3523 // fax:+386-1-477-3822) From afloris at physik.fu-berlin.de Thu Jun 12 17:22:26 2008 From: afloris at physik.fu-berlin.de (Andrea Floris) Date: Thu, 12 Jun 2008 17:22:26 +0200 (CEST) Subject: [Pw_forum] phq_readin and LDA+U Message-ID: Dear Paul, regarding the combination LDA+U and phonons, I am implementing, together with Prof. Matteo Cococcioni and Prof. Stefano de Gironcoli, the extension of the linear response theory to include the extra terms coming from the Hubbard energy. The implementation is not complete, but a lot of work has been done, and it will be complete soon. cheers, Andrea --------------- Dr. Andrea Floris Institut f"ur Theoretische Physik Freie Universit"at Berlin Arnimallee 14 D-14195 Berlin Germany Tel +49-30-838-53029 Fax +49-30-838-55258 e-mail: afloris at physik.fu-berlin.de From paulatto at sissa.it Thu Jun 12 18:11:53 2008 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Thu, 12 Jun 2008 18:11:53 +0200 (CEST) Subject: [Pw_forum] PB mixing PP produced with ld1 with virtual.x In-Reply-To: <000001c8cc7e$466a3410$d33e9c30$%Teyssier@physics.unige.ch> References: <000001c8cc7e$466a3410$d33e9c30$%Teyssier@physics.unige.ch> Message-ID: <4580.147.122.5.216.1213287113.squirrel@webmail.sissa.it> On Thu, June 12, 2008 13:20, J?r?mie Teyssier wrote: > from scan_begin : error # 1 Are you using version QE 4.0 or 3.2.3? In the former case be sure to generate the pseudopotentials with the option upf_v1_format=.true. in inputp namelist, as virtual.x has not been updated (and does not update itself) to support newer pseudopotential format version introduced in QE 4.0. bye -- 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 prasenjit.jnc at gmail.com Thu Jun 12 23:51:56 2008 From: prasenjit.jnc at gmail.com (Prasenjit Ghosh) Date: Thu, 12 Jun 2008 23:51:56 +0200 Subject: [Pw_forum] neb In-Reply-To: <1213284541.6084.43.camel@walk.ijs.si> References: <166984.92508.qm@web52306.mail.re2.yahoo.com> <200806122255389247313@yahoo.com.cn> <627e0ffa0806120819j126c21c1q951952e5f4adc60a@mail.gmail.com> <1213284541.6084.43.camel@walk.ijs.si> Message-ID: <627e0ffa0806121451p64cf24cdia463ad8a65917922@mail.gmail.com> Hi Tone, Thanks for correcting me............ Is the linear interpolation in the Modules/path_base.f90 file used by the code only for finding the initial guess of the path (if not specified in the input). Can you please refer me to the specific subroutine where the interpolation is done in order to generate the data for the *.int file? Prasenjit. 2008/6/12 Tone Kokalj : > On Thu, 2008-06-12 at 17:19 +0200, Prasenjit Ghosh wrote: > > As far as I know, you don't have an option to choose the type of > > interpolation to obtain the data in the *.int file. The code uses a > > linear interpolation technique to obtain the above mentioned data. > > This is not true. As far as I know, the code uses the interpolation as > suggested in the original NEB paper. This interpolation is kind of cubic > interpolation, which uses the information also from the forces (=slope) > provided by the code. According to my experience it's rather good (see > the appendix in G. Henkelman, J. Chem. Phys. 113, 9978 (2000)). > > Regards, Tone > -- > Tone Kokalj > J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia (tel: > +386-1-477-3523 // fax:+386-1-477-3822) > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- PRASENJIT GHOSH, POST-DOC, ROOM NO: 265, MAIN BUILDING, CM SECTION, ICTP, STRADA COSTERIA 11, TRIESTE, 34104, ITALY PHONE: +39 040 2240 369 (O) +39 3807528672 (M) -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080612/bb431032/attachment-0001.htm From w2agz at pacbell.net Fri Jun 13 02:07:27 2008 From: w2agz at pacbell.net (Paul Grant) Date: Thu, 12 Jun 2008 17:07:27 -0700 (PDT) Subject: [Pw_forum] phq_readin and LDA+U Message-ID: <273139.48660.qm@web82504.mail.mud.yahoo.com> Dear Andrea, Great news. Please feel free to call on me as a "beta tester" when appropriate. -Paul ----- Original Message ---- From: Andrea Floris To: PWSCF_forum Sent: Thursday, June 12, 2008 8:22:26 AM Subject: [Pw_forum] phq_readin and LDA+U Dear Paul, regarding the combination LDA+U and phonons, I am implementing, together with Prof. Matteo Cococcioni and Prof. Stefano de Gironcoli, the extension of the linear response theory to include the extra terms coming from the Hubbard energy. The implementation is not complete, but a lot of work has been done, and it will be complete soon. cheers, Andrea --------------- Dr. Andrea Floris Institut f"ur Theoretische Physik Freie Universit"at Berlin Arnimallee 14 D-14195 Berlin Germany Tel +49-30-838-53029 Fax +49-30-838-55258 e-mail: afloris at physik.fu-berlin.de _______________________________________________ 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/20080612/a18435e1/attachment.htm From giannozz at democritos.it Fri Jun 13 08:52:18 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Fri, 13 Jun 2008 08:52:18 +0200 Subject: [Pw_forum] Gamma point versus spin-orbit calculation In-Reply-To: <209357.88893.qm@web52304.mail.re2.yahoo.com> References: <209357.88893.qm@web52304.mail.re2.yahoo.com> Message-ID: <1180F9BE-4D87-47F9-B56D-B82BFF3E71A8@democritos.it> On Jun 9, 2008, at 18:04 , ?lvaro Alves wrote: > Hi all. Is there any incompatibility between the gamma point and > spin-orbit calculation ? are you asking whether "there is presently an incompatibility" (the answer is yes) or whether "there is a fundamental reason why there is a incompatibility"? 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 democritos.it Fri Jun 13 08:52:27 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Fri, 13 Jun 2008 08:52:27 +0200 Subject: [Pw_forum] =?utf-8?b?5Zue5aSN77yaIFB3X2ZvcnVtIERpZ2VzdCwgVm9sIDEy?= =?utf-8?q?=2C_Issue_11?= In-Reply-To: <749950.91682.qm@web15001.mail.cnb.yahoo.com> References: <749950.91682.qm@web15001.mail.cnb.yahoo.com> Message-ID: On Jun 9, 2008, at 14:44 , jiming an wrote: > the criteria of relax is Total foce equal 0. that would be a bad criterion, since forces are never exactly zero except in some high-symmetry crystals! 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 democritos.it Fri Jun 13 08:52:51 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Fri, 13 Jun 2008 08:52:51 +0200 Subject: [Pw_forum] How to solve the problem: "problems computing cholesky decomposition" In-Reply-To: <00d201c8c925$2b567b40$f9045c0a@WWWA0A18A7CBA2> References: <833262.50381.qm@web65708.mail.ac4.yahoo.com> <00d201c8c925$2b567b40$f9045c0a@WWWA0A18A7CBA2> Message-ID: On Jun 8, 2008, at 7:04 , Wang Qinjing wrote: > I've solved the problem by changing 'diagonalization' from default > value > to 'CG' mode. That means the default diagonalization mode is unstable. no sir, that means something different. It means that instead of helping the community by providing the needed information to reproduce the problem, if any (type of machine, number of processors used, input data and related output, pseudopotentials if not available in the database) and to fix it, you choose a workaround that will considerably slow down your calculations. Too bad for you. --- 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 democritos.it Fri Jun 13 08:53:11 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Fri, 13 Jun 2008 08:53:11 +0200 Subject: [Pw_forum] Innards of ph.f90 and Matdyn.f90 In-Reply-To: <00f401c8c78c$260a0980$721e1c80$@net> References: <00f401c8c78c$260a0980$721e1c80$@net> Message-ID: <6047E3F4-1BD7-471F-9EAC-4CF757BFCD8A@democritos.it> On Jun 6, 2008, at 6:16 , Paul M. Grant wrote: > I always fear being scolded by Paolo and/or Axel for not looking at > past fora postings or within the intricacies of ?the code.? (No, I > missed seeing that covered in the posting of 28 June 1914?I was > otherwise engaged). > nobody who posts not-too-stupid questions is at risk of being scolded here. > 3) Question?why is it necessary to do two scf runs a la > Example07, one to set a DOS flag for separate use later, and > another to set the k-point grid for a subsequent el-ph run with ph.x? > it is sort of explained in the README that comes with example 07 (and with all examples). The first scf calculates \epsilon_k on a dense grid, needed only in the estimate of the electron-phonon coefficients, saves the \epsilon_k to a specific file. The second scf calculates the \psi_k on a "normal" grid, used in all further calculations, saves data to the data directory, overwriting all files written by the previous run, except the special one containing the \epsilon_k on a dense grid. > I know there might some resource issues for ?serious? > calculations, but for Example07, one scf with la2F ?on? and an MP > grid of order 12, ?seems to? work just fine. At least for a ?non- > serious? calculation. > in that case, the dense and normal grid just coincide. The double- delta integral needed for electron-phonon coefficients is calculated without interpolation over the dense grid (or, more exactly, with interpolation, but the original and interpolated grids are the same). > 6) The matdyn.x calculation to compute the frequency dependent > component of the Eliashberg kernel uses a ?real space? sampling > grid?in the case of Example07 of size 19. > those are q-vectors in the BZ, not real-space points. You supply the points you are interested in. Typicaly people draws phonons on high-symmetry lines in the BZ and corresponding el-ph coefiicients > 7) Lastly, here?s where I really get confused and appreciate > Dr. Prasad?s posting. Which ?dos? calculation is being referred to? > phonon DOS. You need phonons calculated in a uniform q-point grid. matdyn.x does this, starting from interatomic forces constants in real space > What?s needed is a user?s manual for matdyn (sorry, Paolo) > if you write a manual for matdyn, I promise that I won't feel sorry at all. Essential documentation is available in the header of file PH/dynmat.f90 > , or, better yet, three or four more examples like 07 on non-cubic > but otherwise simple symmetries and simple basis orbital sets with > more than one atom per unit cell?like MgB2! > the problem with el-ph is that the current algorithm is not especially satisfactory. There are much better implementations around, but until now thay have failed to materialize into the "official" release. Spending time and effort for the old algorithm when a new one might be released soon(er or later) is not an appealing perspective. 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 democritos.it Fri Jun 13 08:53:20 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Fri, 13 Jun 2008 08:53:20 +0200 Subject: [Pw_forum] Problems concerning fhi2upf and cpmd2upf In-Reply-To: <200806101154.00087.holger.hesske@chem.ethz.ch> References: <200806101154.00087.holger.hesske@chem.ethz.ch> Message-ID: <2EDEA88B-CE8D-45CC-93A3-FC874F7FA528@democritos.it> On Jun 10, 2008, at 11:53 , Holger Hesske wrote: > (1) I tried to convert a NC-PP for Nitrogen from the fhi-website. > It fails > with segfault [...] The question arose: Is it my non-ability > (probably) since you managed to convert other PP's, your ability has a nonzero value > does it not work for that special case this is what the experimental observation would suggest > Or even more interesting: How could I fix it and obtain a > working Nitrogen_NC_PBE_PP in UPF format? you can easily locate the point in which the code segfaults: the code itself is quite easy to follow. Maybe it is something trivial, maybe it is not: the FHI format doesn't seem to contain much information on its contents, so one has to guess what is what 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 tone.kokalj at ijs.si Fri Jun 13 10:29:58 2008 From: tone.kokalj at ijs.si (Tone Kokalj) Date: Fri, 13 Jun 2008 10:29:58 +0200 Subject: [Pw_forum] neb In-Reply-To: <627e0ffa0806121451p64cf24cdia463ad8a65917922@mail.gmail.com> References: <166984.92508.qm@web52306.mail.re2.yahoo.com> <200806122255389247313@yahoo.com.cn> <627e0ffa0806120819j126c21c1q951952e5f4adc60a@mail.gmail.com> <1213284541.6084.43.camel@walk.ijs.si> <627e0ffa0806121451p64cf24cdia463ad8a65917922@mail.gmail.com> Message-ID: <1213345798.6540.24.camel@walk.ijs.si> On Thu, 2008-06-12 at 23:51 +0200, Prasenjit Ghosh wrote: > Hi Tone, > > Thanks for correcting me............ > Is the linear interpolation in the Modules/path_base.f90 file used by > the code only for finding the initial guess of the path (if not > specified in the input). Yes, the linear interpolation is used to generate images for initial path. > Can you please refer me to the specific subroutine where the > interpolation is done in order to generate the data for the *.int > file? See the file: path_io_routines.f90, subroutine write_dat_files(). There also exists the same type of interpolation for the images: for example using a given neb path restart file you can generate a new path with different number of images, see pwtools/path_int.sh Regards, Tone -- Tone Kokalj J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia (tel: +386-1-477-3523 // fax:+386-1-477-3822) From prasenjit.jnc at gmail.com Fri Jun 13 10:41:32 2008 From: prasenjit.jnc at gmail.com (Prasenjit Ghosh) Date: Fri, 13 Jun 2008 10:41:32 +0200 Subject: [Pw_forum] neb In-Reply-To: <1213345798.6540.24.camel@walk.ijs.si> References: <166984.92508.qm@web52306.mail.re2.yahoo.com> <200806122255389247313@yahoo.com.cn> <627e0ffa0806120819j126c21c1q951952e5f4adc60a@mail.gmail.com> <1213284541.6084.43.camel@walk.ijs.si> <627e0ffa0806121451p64cf24cdia463ad8a65917922@mail.gmail.com> <1213345798.6540.24.camel@walk.ijs.si> Message-ID: <627e0ffa0806130141l21a5c4eex37c64d9b7aed4154@mail.gmail.com> thanks a lot for the details. :) Prasenjit 2008/6/13 Tone Kokalj : > On Thu, 2008-06-12 at 23:51 +0200, Prasenjit Ghosh wrote: > > Hi Tone, > > > > Thanks for correcting me............ > > Is the linear interpolation in the Modules/path_base.f90 file used by > > the code only for finding the initial guess of the path (if not > > specified in the input). > > Yes, the linear interpolation is used to generate images for initial > path. > > > Can you please refer me to the specific subroutine where the > > interpolation is done in order to generate the data for the *.int > > file? > > See the file: path_io_routines.f90, subroutine write_dat_files(). > > There also exists the same type of interpolation for the images: for > example using a given neb path restart file you can generate a new path > with different number of images, see pwtools/path_int.sh > > Regards, Tone > -- > Tone Kokalj > J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia (tel: > +386-1-477-3523 // fax:+386-1-477-3822) > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- PRASENJIT GHOSH, POST-DOC, ROOM NO: 265, MAIN BUILDING, CM SECTION, ICTP, STRADA COSTERIA 11, TRIESTE, 34104, ITALY PHONE: +39 040 2240 369 (O) +39 3807528672 (M) -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080613/c54e6b8f/attachment.htm From ezadshojaee at hotmail.com Fri Jun 13 18:53:26 2008 From: ezadshojaee at hotmail.com (Ezad Shojaee) Date: Fri, 13 Jun 2008 16:53:26 +0000 Subject: [Pw_forum] EOS for Tetragonal Systems Message-ID: Hi I have noticed the following answer in the archive about EOS in Tetragonal lattices(a=b=/c): You have to optimize first c/a keeping the volume fixed. The discontinuities of the total energy as a function of c/a (due to the fact that the number of PWs change as you change c/a) are less critical than the discontinuities in E(V). Once you have E(V), you can fit it to an equation of state, much in the same way you would for a cubic system. a silly question: should'nt be the same way we do for a hexagonal system?! and what if the Tetragonal Lattice be a Body-centered one? Thanx for any comment _________________________________________________________________ Explore the seven wonders of the world http://search.msn.com/results.aspx?q=7+wonders+world&mkt=en-US&form=QBRE -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080613/182a8d50/attachment.htm From xwzhang at uestc.edu.cn Fri Jun 13 20:57:48 2008 From: xwzhang at uestc.edu.cn (xwzhang at uestc.edu.cn) Date: Sat, 14 Jun 2008 02:57:48 +0800 Subject: [Pw_forum] about Phonon results Message-ID: Dear my pwscf friends, Recently, I am trying to do phonon calculations in Heusler. As a first step, I tried to reproduce the results of a paper by Zayak(PRB, 72, 054113, 2005). I choose Ni2MnGe as an example. To my surprise, the acoustic branches at Gamma point gives a frequency of about 1 THz, which should be zero. Other branches agree with the published data. I wonder what might be the reason? I attached my scf, and phonon input files, and the phonon out put files. I have another question about the output: What is the difference between the total magnetic moment and the absolute magnetic moment. Thank you and wish you a happy weekend! Wenxu Zhang -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080614/c7e9ea22/attachment-0001.htm -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: ni2mnge.phG.in Url: http://www.democritos.it/pipermail/pw_forum/attachments/20080614/c7e9ea22/attachment-0003.txt -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: ni2mnge.scf.in Url: http://www.democritos.it/pipermail/pw_forum/attachments/20080614/c7e9ea22/attachment-0004.txt -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: Phon.Gamma Url: http://www.democritos.it/pipermail/pw_forum/attachments/20080614/c7e9ea22/attachment-0005.txt From lanhaiping at gmail.com Fri Jun 13 21:18:27 2008 From: lanhaiping at gmail.com (lan haiping) Date: Sat, 14 Jun 2008 03:18:27 +0800 Subject: [Pw_forum] about Phonon results In-Reply-To: References: Message-ID: Have you ever employed acoustic sum rule ? you can do a search on this topic over the forum archives. PS, please supply your affiliation . On Sat, Jun 14, 2008 at 2:57 AM, wrote: > Dear my pwscf friends, > > Recently, I am trying to do phonon calculations in Heusler. As a first > step, I tried to reproduce the results of a paper by Zayak(PRB, 72, 054113, > 2005). I choose Ni2MnGe as an example. To my surprise, the acoustic branches > at Gamma point gives a frequency of about 1 THz, which should be zero. Other > branches agree with the published data. > > I wonder what might be the reason? I attached my scf, and phonon input > files, and the phonon out put files. > > I have another question about the output: What is the difference between > the total magnetic moment and the absolute magnetic moment. > > Thank you and wish you a happy weekend! > > Wenxu Zhang > > _______________________________________________ > 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/20080614/a8674160/attachment.htm From beaudet at uiuc.edu Fri Jun 13 21:35:53 2008 From: beaudet at uiuc.edu (beaudet at uiuc.edu) Date: Fri, 13 Jun 2008 14:35:53 -0500 (CDT) Subject: [Pw_forum] new user problem running Si Message-ID: <20080613143553.BGR01895@expms2.cites.uiuc.edu> experienced PWscf users, I have compiled the code. When I run the Si example the code crashes after the second iteration... iteration # 2 ecut= 18.00 Ry beta=0.70 CG style diagonalization ethr = 7.69E-04, avg # of iterations = 3.0 p0_3521: p4_error: interrupt SIGSEGV: 11 when I backtrace using a debugger I get the following: (gdb) backtrace #0 0x000000000050ed55 in __scf__mix_type_axpy () #1 0x00000000004e2770 in mix_rho_ () #2 0x00000000004cae58 in electrons_ () #3 0x000000000040f39c in MAIN__ () I suspect a problem with the environment when I compiled the code. I'm using Intel Core2 Quad CPU Q6600 and Ubuntu 8.04 - the Hardy Heron. Below is the input file I was using. Does anyone have a idea on how I might proceed? I can provide more information if necessary. &control calculation = 'scf' restart_mode='from_scratch', prefix='silicon', tstress = .true. tprnfor = .true. pseudo_dir = '/home/beaudet/espresso/espresso-4.0/pseudo/', outdir='/tmp/' / &system ibrav= 2, celldm(1) =10.20, nat= 2, ntyp= 1, ecutwfc =18.0, / &electrons diagonalization='cg' mixing_mode = 'plain' mixing_beta = 0.7 conv_thr = 1.0d-8 / ATOMIC_SPECIES Si 28.086 Si.vbc.UPF ATOMIC_POSITIONS Si 0.00 0.00 0.00 Si 0.25 0.25 0.25 K_POINTS 10 0.1250000 0.1250000 0.1250000 1.00 0.1250000 0.1250000 0.3750000 3.00 0.1250000 0.1250000 0.6250000 3.00 0.1250000 0.1250000 0.8750000 3.00 0.1250000 0.3750000 0.3750000 3.00 0.1250000 0.3750000 0.6250000 6.00 0.1250000 0.3750000 0.8750000 6.00 0.1250000 0.6250000 0.6250000 3.00 0.3750000 0.3750000 0.3750000 1.00 0.3750000 0.3750000 0.6250000 3.00 Thanks, Todd Beaudet From naromero at gmail.com Fri Jun 13 21:53:37 2008 From: naromero at gmail.com (Nichols A. Romero) Date: Fri, 13 Jun 2008 15:53:37 -0400 Subject: [Pw_forum] new user problem running Si In-Reply-To: <20080613143553.BGR01895@expms2.cites.uiuc.edu> References: <20080613143553.BGR01895@expms2.cites.uiuc.edu> Message-ID: <6ac064b60806131253i1a11458cjc853f00938dd3684@mail.gmail.com> Todd, You should tells us the following information. 1. compiler 2. math library 3. serial or parallel run 4. PWscf version On Fri, Jun 13, 2008 at 3:35 PM, wrote: > experienced PWscf users, > > I have compiled the code. When I run the Si example the code crashes after > the second iteration... > > iteration # 2 ecut= 18.00 Ry beta=0.70 > CG style diagonalization > ethr = 7.69E-04, avg # of iterations = 3.0 > p0_3521: p4_error: interrupt SIGSEGV: 11 > > when I backtrace using a debugger I get the following: > > (gdb) backtrace > #0 0x000000000050ed55 in __scf__mix_type_axpy () > #1 0x00000000004e2770 in mix_rho_ () > #2 0x00000000004cae58 in electrons_ () > #3 0x000000000040f39c in MAIN__ () > > I suspect a problem with the environment when I compiled the code. I'm > using Intel Core2 Quad CPU Q6600 and Ubuntu 8.04 - the Hardy Heron. Below is > the input file I was using. Does anyone have a idea on how I might proceed? > I can provide more information if necessary. > > &control > calculation = 'scf' > restart_mode='from_scratch', > prefix='silicon', > tstress = .true. > tprnfor = .true. > pseudo_dir = '/home/beaudet/espresso/espresso-4.0/pseudo/', > outdir='/tmp/' > / > &system > ibrav= 2, celldm(1) =10.20, nat= 2, ntyp= 1, > ecutwfc =18.0, > / > &electrons > diagonalization='cg' > mixing_mode = 'plain' > mixing_beta = 0.7 > conv_thr = 1.0d-8 > / > ATOMIC_SPECIES > Si 28.086 Si.vbc.UPF > ATOMIC_POSITIONS > Si 0.00 0.00 0.00 > Si 0.25 0.25 0.25 > K_POINTS > 10 > 0.1250000 0.1250000 0.1250000 1.00 > 0.1250000 0.1250000 0.3750000 3.00 > 0.1250000 0.1250000 0.6250000 3.00 > 0.1250000 0.1250000 0.8750000 3.00 > 0.1250000 0.3750000 0.3750000 3.00 > 0.1250000 0.3750000 0.6250000 6.00 > 0.1250000 0.3750000 0.8750000 6.00 > 0.1250000 0.6250000 0.6250000 3.00 > 0.3750000 0.3750000 0.3750000 1.00 > 0.3750000 0.3750000 0.6250000 3.00 > > Thanks, > > Todd Beaudet > _______________________________________________ > 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/20080613/7d3a310a/attachment.htm From akohlmey at cmm.chem.upenn.edu Fri Jun 13 22:04:14 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Fri, 13 Jun 2008 16:04:14 -0400 (EDT) Subject: [Pw_forum] new user problem running Si In-Reply-To: <20080613143553.BGR01895@expms2.cites.uiuc.edu> References: <20080613143553.BGR01895@expms2.cites.uiuc.edu> Message-ID: On Fri, 13 Jun 2008, beaudet at uiuc.edu wrote: TB> experienced PWscf users, hi todd, congratulations for providing an almost perfect problem report! from the backtrace it looks very much like you are running into a compiler bug in gfortran(?). please make sure you have the latest version of QE and all updates of your OS installed. if that does not make the problem go away, you could try using a different compiler (e.g. g95). the crash is in an overloaded function and that is a "new-ish" feature of the fortran compiler and g95 typically more reliable with these. cheers, axel. TB> I have compiled the code. When I run the Si example the code crashes TB> after the second iteration... TB> TB> iteration # 2 ecut= 18.00 Ry beta=0.70 TB> CG style diagonalization TB> ethr = 7.69E-04, avg # of iterations = 3.0 TB> p0_3521: p4_error: interrupt SIGSEGV: 11 TB> TB> when I backtrace using a debugger I get the following: TB> TB> (gdb) backtrace TB> #0 0x000000000050ed55 in __scf__mix_type_axpy () TB> #1 0x00000000004e2770 in mix_rho_ () TB> #2 0x00000000004cae58 in electrons_ () TB> #3 0x000000000040f39c in MAIN__ () TB> TB> I suspect a problem with the environment when I compiled the code. I'm using Intel Core2 Quad CPU Q6600 and Ubuntu 8.04 - the Hardy Heron. Below is the input file I was using. Does anyone have a idea on how I might proceed? I can provide more information if necessary. TB> TB> &control TB> calculation = 'scf' TB> restart_mode='from_scratch', TB> prefix='silicon', TB> tstress = .true. TB> tprnfor = .true. TB> pseudo_dir = '/home/beaudet/espresso/espresso-4.0/pseudo/', TB> outdir='/tmp/' TB> / TB> &system TB> ibrav= 2, celldm(1) =10.20, nat= 2, ntyp= 1, TB> ecutwfc =18.0, TB> / TB> &electrons TB> diagonalization='cg' TB> mixing_mode = 'plain' TB> mixing_beta = 0.7 TB> conv_thr = 1.0d-8 TB> / TB> ATOMIC_SPECIES TB> Si 28.086 Si.vbc.UPF TB> ATOMIC_POSITIONS TB> Si 0.00 0.00 0.00 TB> Si 0.25 0.25 0.25 TB> K_POINTS TB> 10 TB> 0.1250000 0.1250000 0.1250000 1.00 TB> 0.1250000 0.1250000 0.3750000 3.00 TB> 0.1250000 0.1250000 0.6250000 3.00 TB> 0.1250000 0.1250000 0.8750000 3.00 TB> 0.1250000 0.3750000 0.3750000 3.00 TB> 0.1250000 0.3750000 0.6250000 6.00 TB> 0.1250000 0.3750000 0.8750000 6.00 TB> 0.1250000 0.6250000 0.6250000 3.00 TB> 0.3750000 0.3750000 0.3750000 1.00 TB> 0.3750000 0.3750000 0.6250000 3.00 TB> TB> Thanks, TB> TB> Todd Beaudet TB> _______________________________________________ TB> Pw_forum mailing list TB> Pw_forum at pwscf.org TB> http://www.democritos.it/mailman/listinfo/pw_forum TB> -- ======================================================================= 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 Jun 13 23:53:23 2008 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Fri, 13 Jun 2008 14:53:23 -0700 (PDT) Subject: [Pw_forum] EOS for Tetragonal Systems In-Reply-To: Message-ID: <318252.32190.qm@web65704.mail.ac4.yahoo.com> Hi Ezad, --- Ezad Shojaee wrote: > a silly question: should'nt be the same way we do > for a hexagonal system?! > and what if the Tetragonal Lattice be a > Body-centered one? Yes, the same way for both hcp and bct lattices. Bests, Eyvaz. P.S. Please do not forget provide your affiliation. > Thanx for any comment > _________________________________________________________________ > Explore the seven wonders of the world > http://search.msn.com/results.aspx?q=7+wonders+world&mkt=en-US&form=QBRE> _______________________________________________ > 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, Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com From ezadshojaee at hotmail.com Sat Jun 14 08:19:47 2008 From: ezadshojaee at hotmail.com (Ezad Shojaee) Date: Sat, 14 Jun 2008 06:19:47 +0000 Subject: [Pw_forum] EOS for Tetragonal Systems Message-ID: Thank You Eyvaz for EOS in Tetragonal Systems, it was mentioned in the archive: You have to optimize first c/a keeping the volume fixed ... Once you have E(V), you can fit it to an equation of state, much in the same way you would for a cubic system. imagine we have optimized c/a for a range of volumes in a Tetragonal lattice. The program ev.x whice fits the data to an equation of state, asks to specify the lattice. Does the sentence "much in the same way you would for a cubic system" means that we should provide (a E_tot) as an input (ofcourse 'a' is according to the optimized c/a for each volume) and treat the Tetragonal like a Sc? or we should provide (V E_tot) as we do for Hexagonal systems and treat it like a Hexa? If it is the first, so we should treat BCT's like a BCC? maybe wrong, but i have much more reasonable results in the second way(treating like a Hexa)! Thanx again _________________________________________________________________ Connect to the next generation of MSN Messenger? http://imagine-msn.com/messenger/launch80/default.aspx?locale=en-us&source=wlmailtagline -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080614/667200ab/attachment.htm From baroni at sissa.it Sat Jun 14 09:02:47 2008 From: baroni at sissa.it (Stefano Baroni) Date: Sat, 14 Jun 2008 09:02:47 +0200 Subject: [Pw_forum] EOS for Tetragonal Systems In-Reply-To: References: Message-ID: <72C67828-F530-45AE-B27F-7CD773C34DEC@sissa.it> yes, the same would apply to any non-cubic structure. SB On Jun 13, 2008, at 6:53 PM, Ezad Shojaee wrote: > Hi > I have noticed the following answer in the archive about EOS in > Tetragonal lattices(a=b=/c): > > You have to optimize first c/a keeping the volume fixed. The > discontinuities of the total energy as a > function of c/a (due to the fact that the number of PWs change as > you change c/a) are less critical than the discontinuities in E(V). > Once you have E(V), you can fit it to an equation of state, much in > the same way you would for a cubic system. > > a silly question: should'nt be the same way we do for a hexagonal > system?! > and what if the Tetragonal Lattice be a Body-centered one? > Thanx for any comment > > > Explore the seven wonders of the world Learn more! > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste http://www.sissa.it/~baroni / [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget 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/20080614/d671fa35/attachment.htm From glevita at units.it Sat Jun 14 15:09:25 2008 From: glevita at units.it (glevita at units.it) Date: Sat, 14 Jun 2008 15:09:25 +0200 Subject: [Pw_forum] ELP map with VMD In-Reply-To: References: <765273.92348.qm@web33508.mail.mud.yahoo.com> Message-ID: <20080614150925.jpt6vex2nj4kg00s@webmail.units.it> Dear All, I'd like to map the Electrostatic potential of a molecule onto its density surface. I can't make it with XCrysDen, but I found out that VMD can handle its .xsf files. I managed to actually import in VMD the .xsf file of the molecular potential I had generated with pp.x... but I got stuck with the generation of the density surface and the mapping of the potential. It seems I have to download the ASPD plugin, but after having done that, I am still unable to make it work. Did anyone encountered similar problem? Or can suggest me some tutorial to read? (unfortunately I wasn't able to get much clarification from those I found on the web) Thank you very much, Giacomo Levita (INFM- Univ. of Trieste) ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program. From akohlmey at cmm.chem.upenn.edu Sat Jun 14 16:59:53 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sat, 14 Jun 2008 10:59:53 -0400 (EDT) Subject: [Pw_forum] ELP map with VMD In-Reply-To: <20080614150925.jpt6vex2nj4kg00s@webmail.units.it> References: <765273.92348.qm@web33508.mail.mud.yahoo.com> <20080614150925.jpt6vex2nj4kg00s@webmail.units.it> Message-ID: On Sat, 14 Jun 2008, glevita at units.it wrote: GL> Dear All, dear giacomo, GL> I'd like to map the Electrostatic potential of a molecule onto its GL> density surface. GL> I can't make it with XCrysDen, but I found out that VMD can handle its GL> .xsf files. yep. ...and if you set the output to .plt or .cube format instead of .xsf there are many other programs out there that can do it. GL> I managed to actually import in VMD the .xsf file of the molecular GL> potential I had generated with pp.x... but I got stuck with the GL> generation of the density surface and the mapping of the potential. It GL> seems I have to download the ASPD plugin, but after having done that, how did you determine that you need this plugin? please have a look at: http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/cpmd-vmd/part5.html#chap7_sect1 for an example (using a different code, but the principle is the same). please note, that VMD requires a correctly working OpenGL for this mapping to work. GL> I am still unable to make it work. GL> Did anyone encountered similar problem? Or can suggest me some GL> tutorial to read? (unfortunately I wasn't able to get much GL> clarification from those I found on the web) the process is extremely straightforward, but it is one of these things that can be very confusing unless you realize how trivial it is. so please explain what you did, what didn't work and where you run into problems. cheers, axel. GL> Thank you very much, GL> Giacomo Levita (INFM- Univ. of Trieste) GL> GL> GL> ---------------------------------------------------------------- GL> This message was sent using IMP, the Internet Messaging Program. GL> GL> GL> _______________________________________________ GL> Pw_forum mailing list GL> Pw_forum at pwscf.org GL> http://www.democritos.it/mailman/listinfo/pw_forum GL> -- ======================================================================= 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 djaithania at hotmail.com Sat Jun 14 17:02:37 2008 From: djaithania at hotmail.com (hania djani-ait aissa) Date: Sat, 14 Jun 2008 15:02:37 +0000 Subject: [Pw_forum] cell optimization Message-ID: dear users, I am trying to optimize lattices parameters in tetragonal centered structure. putting c/a=cte and a=7.28, 7.29, 7.30 ......7.36, I've got the results on total energy: a=7.28 E=-1282.635 a=7.29 E=-1282.639 a=7.30 E=-1282.658 a=7.31 E=-1282.684 a=7.32 E=-1282.654 a=7.33 E=-1282.624 a=7.34 E=-1282.679 a=7.35 E=-1282.669 a=7.36 E=-1282.652 I've done the calculations twice and in each case convergence is well achieved giving same results. so, why above a=7.31, total energy rises and drops. is it because ecutoff is too small? (ecutoff=25Ry) thanks a lot for any answer. Hania Djani-AitAissa Center of development of advanced technologies, Algiers, Algeria _________________________________________________________________ Lancez des recherches en toute s?curit? depuis n'importe quelle page Web. T?l?chargez GRATUITEMENT Windows Live Toolbar aujourd'hui ! http://toolbar.live.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080614/3ac29941/attachment.htm From akohlmey at cmm.chem.upenn.edu Sat Jun 14 17:17:59 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sat, 14 Jun 2008 11:17:59 -0400 (EDT) Subject: [Pw_forum] cell optimization In-Reply-To: References: Message-ID: On Sat, 14 Jun 2008, hania djani-ait aissa wrote: HDA> dear users, dear hania, HDA> I am trying to optimize lattices parameters in tetragonal centered HDA> structure. putting c/a=cte and a=7.28, 7.29, 7.30 ......7.36, I've HDA> got the results on total energy: HDA> a=7.28 E=-1282.635 HDA> a=7.29 E=-1282.639 HDA> a=7.30 E=-1282.658 HDA> a=7.31 E=-1282.684 HDA> a=7.32 E=-1282.654 HDA> a=7.33 E=-1282.624 HDA> a=7.34 E=-1282.679 HDA> a=7.35 E=-1282.669 HDA> a=7.36 E=-1282.652 HDA> I've done the calculations twice and in each case convergence is HDA> well achieved giving same results. so, why above a=7.31, total HDA> energy rises and drops. is it because ecutoff is too small? please have a look at the part of your output and check out the number of grid points. HDA> (ecutoff=25Ry) thanks a lot for any answer. whether your cutoff is too low depends on the pseudopotential files that you are using. 25ry suggests you are using ultra soft pseudopotentials, for those you have to use a much higher density cutoff than the default (which is 4x ecutwfc). please note, it is easier to give qualified advice, if you make your input file available... cheers, axel. HDA> HDA> Hania Djani-AitAissa HDA> Center of development of advanced technologies, Algiers, Algeria HDA> _________________________________________________________________ HDA> Lancez des recherches en toute s?curit? depuis n'importe quelle page Web. T?l?chargez GRATUITEMENT Windows Live Toolbar aujourd'hui ! HDA> http://toolbar.live.com -- ======================================================================= 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 democritos.it Sat Jun 14 17:59:59 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Sat, 14 Jun 2008 17:59:59 +0200 Subject: [Pw_forum] converting CPMD pp's to UPF In-Reply-To: References: Message-ID: <4B68C186-4F58-44D0-8B1F-B786AFD2C3A5@democritos.it> On Jun 12, 2008, at 8:31 , Min Seung Kyu wrote: > However, during the optimization, the bond length were getting > longer both LDA and GGA cases. longer by how much, and with respect to what? 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 democritos.it Sat Jun 14 18:02:51 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Sat, 14 Jun 2008 18:02:51 +0200 Subject: [Pw_forum] eigenvalues not converged in ph.x In-Reply-To: <33378.210.73.17.131.1213250453.squirrel@210.73.16.4> References: <33378.210.73.17.131.1213250453.squirrel@210.73.16.4> Message-ID: <6A29B73D-F618-4BC1-81A2-61F316A8ADF8@democritos.it> On Jun 12, 2008, at 8:00 , lfhuang at theory.issp.ac.cn wrote: > Although there are so much messages on "WARNING: # eigenvalues not > converged" in pw.x, this problem in ph.x is not frequently talked. there is no reasn why it should "be talked", since it is the same problem (or not-a-problem) in ph.x as in pw.x . In the last release those messages have been suppressed. 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 democritos.it Sat Jun 14 18:14:55 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Sat, 14 Jun 2008 18:14:55 +0200 Subject: [Pw_forum] About the generate uspp of hafnium for using within pwscf. In-Reply-To: <200806080057479569123@yahoo.com.cn> References: <16081824.202961212849295271.JavaMail.coremail@bj126app104.126.com> <200806080057479569123@yahoo.com.cn> Message-ID: On Jun 7, 2008, at 18:57 , zhaohscas wrote: > I've found that the CASTEP, which is my another tool for my work, > has the uspp's for hafnium, the Hf 5d^2 6s^2 was treated as the > valence states. > > Who can give me some hints to generate the corresponding uspp > of hafnium for using within pwscf? in order to generate an USPP you need to know - the reference electronic configuration - the number of projectors per l - the corresponding energies and matching radii I seriously doubt, however, that it is worth to try to generate an USPP for 4 valence electrons in Hf. It is worth if you include semicore states as well (5s^2 5p^6 for Hf) 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 zhaohscas at yahoo.com.cn Sun Jun 15 04:42:14 2008 From: zhaohscas at yahoo.com.cn (zhaohscas) Date: Sun, 15 Jun 2008 10:42:14 +0800 Subject: [Pw_forum] Compulate the spanning vectors for 2D and 3D scalar-fields data grid in Xcrysden. Message-ID: <200806151042114842758@yahoo.com.cn> Hi all, Although the Xcrysden give some specifications of its 2D and 3D scalar-fields data grid, I'm in a puzzle as to how to determine the corresponding two or three spanning vectors w.r.t the 2D and 3D scalar-fields data grid. Who can give me hints on this? Sincerely yours, -------------- Hongsheng Zhao Xinjiang Technical Institute of Physics and Chemistry Chinese Academy of Sciences GnuPG DSA: 0xD108493 2008-06-15 __________________________________________________ ??????????????? http://cn.mail.yahoo.com From djaithania at hotmail.com Sun Jun 15 10:30:46 2008 From: djaithania at hotmail.com (hania djani-ait aissa) Date: Sun, 15 Jun 2008 08:30:46 +0000 Subject: [Pw_forum] cell optimization Message-ID: Hi, thanks for your fast response.actually, I am using Bi-mt-PPs, so 25Ry is too small.here is my input: &control calculation = 'scf' restart_mode = 'from_scratch' pseudo_dir = '/root/pwscf/pseudo/' outdir = '/root/tmp/' / &system ibrav=7 celldm(1)=7.33, celldm(3)=8.54 nat=19 ntyp=3 ecutwfc=25.0 ecutrho=100.0 / &electrons conv_thr = 1e-6, mixing_beta=0.2, /ATOMIC_SPECIES Bi 208.98 Bi.pbe-d-mt.UPF Ti 47.867 Ti.pbe-sp-van_ak.UPF O 15.9994 O.pbe-van_ak.UPFATOMIC_POSITIONS Bi 0.00000000 0.00000000 0.5713404 Bi 0.00000000 0.00000000 7.9816596 Bi 0.00000000 0.00000000 1.8072489 Bi 0.00000000 0.00000000 6.7457511 Ti 0.00000000 0.00000000 3.1757289 Ti 0.00000000 0.00000000 5.3772711 Ti 0.00000000 0.00000000 4.2765000 O 0.50000000 0.00000000 0.0000 O 0.00000000 0.50000000 0.0000 O 0.50000000 0.00000000 2.13825 O 0.50000000 0.00000000 6.41475 O 0.00000000 0.00000000 3.771873 O 0.00000000 0.00000000 4.781127 O 0.00000000 0.00000000 2.7241305 O 0.00000000 0.00000000 5.8288695 O 0.50000000 0.00000000 1.0049775 O 0.50000000 0.00000000 7.5480225 O 0.00000000 0.50000000 1.0049775 O 0.00000000 0.50000000 7.5480225K_POINTS {automatic} 4 4 4 1 1 1 Hania Djani-AitAissa CDTA, Algiers, Algeria _________________________________________________________________ D?couvrez Windows Live Spaces et cr?ez votre site Web perso en quelques clics ! http://spaces.live.com/signup.aspx -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080615/db6278ce/attachment.htm From rgebauer at ictp.it Sun Jun 15 10:41:50 2008 From: rgebauer at ictp.it (Gebauer Ralph) Date: Sun, 15 Jun 2008 10:41:50 +0200 (CEST) Subject: [Pw_forum] cell optimization In-Reply-To: References: Message-ID: <51394.151.65.25.161.1213519310.squirrel@webmail.ictp.it> Hania, Apart from the (too small) cutoff you are using, I think that you experience the well known problem of a discontinuous energy-volume graph. See for example here: http://www.quantum-espresso.org/wiki/index.php/Methodological_Background#Stress and in the reference given in that wiki. Greetings! Ralph ----- Ralph Gebauer The Abdus Salam International Centre for Theoretical Physics (ICTP) Condensed Matter Section Tel: (+39).040.2240.344 Strada Costiera 11 Fax: (+39).040.224163 I-34014 Trieste (Italy) e-mail: rgebauer at ictp.trieste.it -- > > Hi, thanks for your fast response.actually, I am using Bi-mt-PPs, so 25Ry > is too small.here is my input: > &control calculation = 'scf' restart_mode = 'from_scratch' > pseudo_dir = '/root/pwscf/pseudo/' outdir = '/root/tmp/' / > &system ibrav=7 celldm(1)=7.33, celldm(3)=8.54 nat=19 ntyp=3 > ecutwfc=25.0 ecutrho=100.0 / &electrons conv_thr = 1e-6, > mixing_beta=0.2, /ATOMIC_SPECIES Bi 208.98 Bi.pbe-d-mt.UPF Ti > 47.867 Ti.pbe-sp-van_ak.UPF O 15.9994 > O.pbe-van_ak.UPFATOMIC_POSITIONS Bi 0.00000000 0.00000000 0.5713404 Bi > 0.00000000 0.00000000 7.9816596 Bi 0.00000000 0.00000000 1.8072489 Bi > 0.00000000 0.00000000 6.7457511 Ti 0.00000000 0.00000000 3.1757289 Ti > 0.00000000 0.00000000 5.3772711 Ti 0.00000000 0.00000000 4.2765000 O > 0.50000000 0.00000000 0.0000 O 0.00000000 0.50000000 0.0000 O > 0.50000000 0.00000000 2.13825 O 0.50000000 0.00000000 6.41475 O > 0.00000000 0.00000000 3.771873 O 0.00000000 0.00000000 4.781127 O > 0.00000000 0.00000000 2.7241305 O 0.00000000 0.00000000 5.8288695 O > 0.50000000 0.00000000 1.0049775 O 0.50000000 0.00000000 7.5480225 O > 0.00000000 0.50000000 1.0049775 O 0.00000000 0.50000000 > 7.5480225K_POINTS {automatic} 4 4 4 1 1 1 > Hania Djani-AitAissa > CDTA, Algiers, Algeria From djaithania at hotmail.com Sun Jun 15 16:05:59 2008 From: djaithania at hotmail.com (hania djani-ait aissa) Date: Sun, 15 Jun 2008 14:05:59 +0000 Subject: [Pw_forum] cell optimization Message-ID: dear Ralph, nice to read from you. I am trying new calculation with ecutoff=40Ry, and will see if the discontinuous energy-volume persist. however for better understanding, i would ask if someone could send me: M.Bernasconi et al. J.Phys.Chem.Solids 56,501(1995). thanks Hania Djani-aitaissa CDTA, Algiers, Algeria _________________________________________________________________ T?l?chargez le nouveau Windows Live Messenger ! http://get.live.com/messenger/overview -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080615/d83fa154/attachment.htm From baroni at sissa.it Sun Jun 15 18:32:49 2008 From: baroni at sissa.it (Stefano Baroni) Date: Sun, 15 Jun 2008 18:32:49 +0200 Subject: [Pw_forum] cell optimization In-Reply-To: References: Message-ID: <33994644-9595-4CDB-B221-1169FC1E9C2A@sissa.it> done - SB On Jun 15, 2008, at 4:05 PM, hania djani-ait aissa wrote: > dear Ralph, > nice to read from you. > I am trying new calculation with ecutoff=40Ry, and will see if the > discontinuous energy-volume persist. however for better > understanding, i would ask if someone could send me: M.Bernasconi > et al. J.Phys.Chem.Solids 56,501(1995). > thanks > > Hania Djani-aitaissa > CDTA, Algiers, Algeria > > T?l?chargez le nouveau Windows Live Messenger ! T?l?chargez > Messenger, c'est gratuit ! > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste http://www.sissa.it/~baroni / [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget 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/20080615/3c25e0b0/attachment.htm From akohlmey at cmm.chem.upenn.edu Sun Jun 15 19:57:34 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sun, 15 Jun 2008 13:57:34 -0400 (EDT) Subject: [Pw_forum] cell optimization In-Reply-To: References: Message-ID: On Sun, 15 Jun 2008, hania djani-ait aissa wrote: HDA> HDA> dear Ralph, HDA> nice to read from you. HDA> I am trying new calculation with ecutoff=40Ry, and will see if the HDA> discontinuous energy-volume persist. however for better yes, but i'd expect the discontinuities to shrink. if i may make a suggestion to approach the problem more systematically and with less computional effort: how about doing a test run for bulk Bi, Ti first and see what the required cutoffs for wavefunction and desity are to get converged results. the cutoff requirements are in general a property of the the individual pseudopotential and the proper choice is set of cutoff that satisfies the minimal requirements for all of them. for the norm conserving potentials there is no need to raise the density cutoff beyond 4x the wavefunction cutoff, but for the ultra-soft potentials you may need up to 10x the cutoff. doing those calculations with a small unit cell and plenty k-points is usually _very_ fast. once you know how each potential's requirements are, you have a much better understanding for what to choose for your compound. HDA> understanding, i would ask if someone could send me: M.Bernasconi HDA> et al. J.Phys.Chem.Solids 56,501(1995). please note, that this technique is mainly intended for variable cell dynamics, where you are (for technical reasons) using a constant number of plane waves (and not a constant cutoff). with individual calculations, like you were doing, you are running calculations with "constant cutoff" (within the limits of the available grids and matching number of plane waves, hence the "jumps"). without the bernasconi trick you need to choose an insanely high wavefunction cutoff to make sure the stress tensor is tightly converged wrt. cutoff, so that (small) changes in the effective cutoff do not matter. hope that helps, axel. HDA> thanks HDA> HDA> Hania Djani-aitaissa HDA> CDTA, Algiers, Algeria HDA> _________________________________________________________________ HDA> T?l?chargez le nouveau Windows Live Messenger ! HDA> http://get.live.com/messenger/overview -- ======================================================================= 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 wangqj1 at 126.com Mon Jun 16 07:28:43 2008 From: wangqj1 at 126.com (wangqj1) Date: Mon, 16 Jun 2008 13:28:43 +0800 (CST) Subject: [Pw_forum] cell optimization and relax Message-ID: <21807808.386751213594123432.JavaMail.coremail@bj126app75.126.com> dear PWSCF users what's the relationship between cell optimization and relax ? After relaxing ,should I have to make cell optimization ? thanks for your advice ! -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080616/f41e0c56/attachment.htm From tone.kokalj at ijs.si Mon Jun 16 09:24:55 2008 From: tone.kokalj at ijs.si (Tone Kokalj) Date: Mon, 16 Jun 2008 09:24:55 +0200 Subject: [Pw_forum] Compulate the spanning vectors for 2D and 3D scalar-fields data grid in Xcrysden. In-Reply-To: <200806151042114842758@yahoo.com.cn> References: <200806151042114842758@yahoo.com.cn> Message-ID: <1213601095.12061.2.camel@walk.ijs.si> On Sun, 2008-06-15 at 10:42 +0800, zhaohscas wrote: > Hi all, > > Although the Xcrysden give some specifications of its 2D and 3D scalar-fields data grid, > I'm in a puzzle as to how to determine the corresponding two or three spanning vectors w.r.t > the 2D and 3D scalar-fields data grid. pp.x will produce the scalar-field, all you need to do is to enter origin and the spanning vectors into pp.x input file? What is puzzling with that? Regards, Tone -- Tone Kokalj J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia (tel: +386-1-477-3523 // fax:+386-1-477-3822) From marzari at MIT.EDU Mon Jun 16 09:24:31 2008 From: marzari at MIT.EDU (Nicola Marzari) Date: Mon, 16 Jun 2008 10:24:31 +0300 Subject: [Pw_forum] cell optimization In-Reply-To: References: Message-ID: <4856152F.4080508@mit.edu> > On Sun, 15 Jun 2008, hania djani-ait aissa wrote: > > > HDA> I am trying new calculation with ecutoff=40Ry, and will see if the > HDA> discontinuous energy-volume persist. however for better A quick comment - the discontinuous energy-volume issue is one where the energy *drops down* discontinuously when the lattice parameter is increased, due to a discontinuous increase in the size of the basis set when new plane waves enter your calculation. Here: a=7.28 E=-1282.635 a=7.29 E=-1282.639 a=7.30 E=-1282.658 a=7.31 E=-1282.684 a=7.32 E=-1282.654 a=7.33 E=-1282.624 a=7.34 E=-1282.679 a=7.35 E=-1282.669 a=7.36 E=-1282.652 in going from 7.31 to 7.32 the energy *increases*; same from 7.32 to 7.33 . While increasing the cutoff might make your calculations physically accurate, it will not explain why the energy jumps up in going from 7.31 to 7.32. My suggestion (did you try it ?) was to make sure that the calculations were converged to the correct ground state - in exotic cases (seen only twice by me - in graphene and in the cubic barium titanate perovskite), and depending on the algorithm, the top band at some k-point might get stuck in a higher eigenstate. My limited experience is that startingwfc="atomic+random" cured this problem in those two cases. 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 djaithania at hotmail.com Mon Jun 16 10:38:19 2008 From: djaithania at hotmail.com (hania djani-ait aissa) Date: Mon, 16 Jun 2008 08:38:19 +0000 Subject: [Pw_forum] cell optimization Message-ID: Dear Axel, Nicola, let me time to digest, to understand and to test your suggestions and i will give you news as soon as i can. thanks for Bernasconi paper. a+ Hania _________________________________________________________________ Lancez des recherches en toute s?curit? depuis n'importe quelle page Web. T?l?chargez GRATUITEMENT Windows Live Toolbar aujourd'hui ! http://toolbar.live.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080616/a5e0ea8e/attachment.htm From zhaohscas at yahoo.com.cn Mon Jun 16 10:51:59 2008 From: zhaohscas at yahoo.com.cn (zhaohscas) Date: Mon, 16 Jun 2008 16:51:59 +0800 Subject: [Pw_forum] Compulate the spanning vectors for 2D and3D scalar-fields data grid in Xcrysden. References: <200806151042114842758@yahoo.com.cn>, <1213601095.12061.2.camel@walk.ijs.si> Message-ID: <200806161651595931796@yahoo.com.cn> > On Sun, 2008-06-15 at 10:42 +0800, zhaohscas wrote: >> Hi all, >> >> Although the Xcrysden give some specifications of its 2D and 3D scalar-fields data grid, >> I'm in a puzzle as to how to determine the corresponding two or three spanning vectors w.r.t >> the 2D and 3D scalar-fields data grid. > pp.x will produce the scalar-field, all you need to do is to enter > origin and the spanning vectors into pp.x input file? What is puzzling > with that? I sometimes use ab initio codes other than pwscf, e.g., CASTEP, which will give the charge density on a 3d scalar-field in just one-column format in the result file. The pp.x will produce the scalar-field in a matrix format into the result xsf file, not in the one-column format. If I can reshape all these data into one-column format in the xsf file generated by pp.x, should I read them in the row-majored or column-majored order? So that, I can convert the CASTEP's scalar-field result to the correct xsf format and visualize the CASTEP's tresult also by using Xcrysden. Regards, Hongsheng. -- Hongsheng Zhao Xinjiang Technical Institute of Physics and Chemistry Chinese Academy of Sciences GnuPG DSA: 0xD108493 2008-6-16 From paulatto at sissa.it Mon Jun 16 11:32:21 2008 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Mon, 16 Jun 2008 11:32:21 +0200 (CEST) Subject: [Pw_forum] Compulate the spanning vectors for 2D and3D scalar-fields data grid in Xcrysden. In-Reply-To: <200806161651595931796@yahoo.com.cn> References: <200806151042114842758@yahoo.com.cn>, <1213601095.12061.2.camel@walk.ijs.si> <200806161651595931796@yahoo.com.cn> Message-ID: <28229.147.122.5.216.1213608741.squirrel@webmail.sissa.it> On Lun, Giugno 16, 2008 10:51, zhaohscas wrote: > should I read them in the > row-majored or column-majored order? It is explained in detail on xcrysden homepage; have a look here: http://www.xcrysden.org/doc/XSF.html#__toc__13 What you have to keep in mind is that xsf is not a quantum-espresso-specific file format, but the xcrysden file format. bye -- 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 democritos.it Mon Jun 16 15:59:42 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Mon, 16 Jun 2008 15:59:42 +0200 Subject: [Pw_forum] Vibration Spectrum from QE In-Reply-To: <99208.38670.qm@web65713.mail.ac4.yahoo.com> References: <99208.38670.qm@web65713.mail.ac4.yahoo.com> Message-ID: <485671CE.8050000@democritos.it> Eyvaz Isaev wrote: > Please have a look at example06 in /examples. > Of course, you should use a supercell, calculate > phonons and then projected phonon dos. > If you mean to calculate vibrations of selected atoms > I think you can try "nat_todo" keyword for phonons, > but not sure it will work in your particular case, > as this keyword has been used for a special structure > (fullerene, if I remember correctly from Paolo's > reply). no, 'nat_todo' is used to select a subset of atoms for which to perform the linera response calculation. For instance, if you have a molecule on a surface, you may calculate the dynamical matrix for the atoms in the molecule only, if you want. The specialized code "phcg.x" (example in examples/example09) was originally written for fullerenes and molecules: it uses k=0 sampling only, calculates phonons at q=0 only Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From shruba at gmail.com Mon Jun 16 17:26:18 2008 From: shruba at gmail.com (shruba at gmail.com) Date: Mon, 16 Jun 2008 11:26:18 -0400 Subject: [Pw_forum] error message - problems computing cholesky decomposition Message-ID: <822f4ec80806160826h127d87dawcd8b512d7f21b63c@mail.gmail.com> Dear all pwscf users, I am running a calculation for an isolated Mn(IV)-Mn(IV) bimettalic system, using the pseudopotential for Mn (from pwscf.org website) Mn.pbe-sp-van.UPF Perdew-Burke-Ernzerhof (PBE) exch-corr semicore state s in valence semicore state p in valence Vanderbilt ultrasoft all other pseudopotential is also taken from pwscf library, this system has +4 charge the detailed input is given below,This calculation is terminating by showing *the eroor message* Self-consistent Calculation iteration # 1 ecut= 40.00 Ry beta=0.70 Davidson diagonalization with overlap %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from pzpotrf : error # 358 problems computing cholesky decomposition %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% when I was checking the forum for similar error I saw Dr Giannozzis mail, according to that I am providing the details, I am sorry if I have missed any *Details of machine* where I ran the calculation ( i also have checked the disk spaces) CPU -Intel Xeon OS- Platform 4 / Redhat MPI- mpich2-1.0.5p4 I am using 16 processors npool=4 *INPUT* &CONTROL calculation = 'scf', restart_mode = 'from_scratch', tprnfor = .TRUE., prefix = ' pseudo_dir = ' outdir=' / &SYSTEM ibrav = 1, celldm(1) = 32.00 nat =92, ntyp =5, nelec=302 ecutwfc = 40, ecutrho = 360, nspin=2, multiplicity=7, occupations = 'smearing', smearing ='gaussian', degauss=0.03, / &ELECTRONS electron_maxstep =200, startingwfc ='atomic', / ATOMIC_SPECIES Mn 54.93 Mn.pbe-sp-van.UPF C 12.00 C.pbe-van_ak.UPF N 15.00 N.pbe-van_bm.UPF O 16.00 O.pbe-van-bm.UPF H 1.00 H.pbe-van_bm.UPF ATOMIC_POSITIONS (angstrom) ............................ N 2.569522 -0.039142 3.993887 C 2.720000 -0.192415 5.309385 C 1.616406 -0.338367 6.145259 C 0.352491 -0.290342 5.628298 Mn 3.984087 0.453313 2.673149 O 3.623204 2.216950 2.565823 Mn 4.611705 2.802338 3.947754 N 5.447377 3.027934 5.827957 ........................ K_POINTS {automatic} 4 4 4 0 0 0 Sorry for this too long mail, but I prefer to give as much detail as I can. Thanks again for your time and help Regards Shruba Gangopadhyay -- Shruba Gangopadhyay Graduate Student Department of Chemistry, Nanoscience Technology Center 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/20080616/fb0fcb9e/attachment.htm From giannozz at democritos.it Mon Jun 16 18:54:26 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Mon, 16 Jun 2008 18:54:26 +0200 Subject: [Pw_forum] error message - problems computing cholesky decomposition In-Reply-To: <822f4ec80806160826h127d87dawcd8b512d7f21b63c@mail.gmail.com> References: <822f4ec80806160826h127d87dawcd8b512d7f21b63c@mail.gmail.com> Message-ID: <48569AC2.5090605@democritos.it> shruba at gmail.com wrote: > from pzpotrf : error # 358 > problems computing cholesky decomposition please try "pw.x -ndiag 1": it will use serial instead of parallel subspace diagonalization > when I was checking the forum for similar error I saw Dr Giannozzis > mail, according to that I am providing the details, I am sorry if I have > missed any you missed to provide atomic positions. If this is a real problem, it has to be reproduced. Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From eariel99 at gmail.com Mon Jun 16 19:57:05 2008 From: eariel99 at gmail.com (Eduardo Ariel Menendez Proupin) Date: Mon, 16 Jun 2008 13:57:05 -0400 Subject: [Pw_forum] about Phonon results Message-ID: This happens because a) 0 frecuencies at Gamma occurs because the energy is is invariant with respect to arbitrary translations of the whole crystal (the same translation for all atoms). From this assumption the accoustic sum rule (ASR) is obtained theoretically. b) The plane wave DFT calculations do not have this invariance exactly, because of the plane wave cutoff. The energy is invariant not for arbitrary translations, but for translations in the FFT grid. c) I guess that the absolute translation invariance is lost in the back and forth FFTs of the exchange-correlation potential, which is non linear, and probably in some part related with the ultrasoft pseudopotentials. Someone correct me if I am wrong, please. You may increase the density cutoff (ecutrho) and see if the first three frequencies get closer to zero. First do it with one of the ESPRESSO examples, that run in minutes. I guess that I made that test some years ago. Using larger cutoffs increase a lot the calculation time. That is why you have the tricky option to impose the ASR in postprocessing (matdyn.x). Practical way. If your "wrong" frecuencies are smaller than 50 cm^-1, this is normal. Apply the ASR in the input for matdyn. Check that the other frequencies do not change more than 1 or 2 cm^-1. Then, if this makes you happy, go to write your report >I have another question about the output: What is the >difference between the total magnetic moment and the >absolute magnetic moment. This has been asked in the forum before. Please, google for this. If I remember well, the absolute magnetic moment is the integral of the absolute value of the magnetization. Regards Eduardo Menendez On Sat, Jun 14, 2008 at 2:57 AM, wrote: > Dear my pwscf friends, > > Recently, I am trying to do phonon calculations in Heusler. As a first > step, I tried to reproduce the results of a paper by Zayak(PRB, 72, 054113, > 2005). I choose Ni2MnGe as an example. To my surprise, the acoustic branches > at Gamma point gives a frequency of about 1 THz, which should be zero. Other > branches agree with the published data. > > I wonder what might be the reason? I attached my scf, and phonon input > files, and the phonon out put files. > > I have another question about the output: What is the difference between > the total magnetic moment and the absolute magnetic moment. > > Thank you and wish you a happy weekend! > > Wenxu Zhang > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080616/ed790697/attachment-0001.htm From beaudet at uiuc.edu Mon Jun 16 21:58:36 2008 From: beaudet at uiuc.edu (beaudet at uiuc.edu) Date: Mon, 16 Jun 2008 14:58:36 -0500 (CDT) Subject: [Pw_forum] new user problem running Si Message-ID: <20080616145836.BGS83973@expms2.cites.uiuc.edu> Nick, Sorry it has taken a while to get back to you. After I posted I had a meeting, and several other things. I tried recompiling several different ways and while I get through the compile SEEMINGLY OK the code crashes at various points - usually at the very beginning. I've got: Quantum Espresso 4.0 Intel compilers version 10.1.011 and mkl version 10.0.1.014 BLAS and LAPACK are up to date on my Umbuntu distribution I suspect it has to do with the parallel environment more than the particular compiler - but I could be wrong. The simplest thing I've tried is using intel compilers. I used the following configure command: ./configure CXX=icpc CC=icc F90=ifort F77=ifort MPIF90=ifort see attachment configintel When I run an example I get: forrtl: severe (174): SIGSEGV, segmentation fault occurred Image PC Routine Line Source libc.so.6 00007F49D8CD6980 Unknown Unknown Unknown libc.so.6 00007F49D8CD7481 Unknown Unknown Unknown libmpich.so.1.0 00007F49D94C75BF Unknown Unknown Unknown libmpich.so.1.0 00007F49D94D40E2 Unknown Unknown Unknown libmpich.so.1.0 00007F49D94D4259 Unknown Unknown Unknown libmpich.so.1.0 00007F49D94A3DC8 Unknown Unknown Unknown libmpich.so.1.0 00007F49D9494A43 Unknown Unknown Unknown libmpich.so.1.0 00007F49D94B86C9 Unknown Unknown Unknown libmpich.so.1.0 00007F49D94B7FE4 Unknown Unknown Unknown pw.x 000000000048F539 Unknown Unknown Unknown pw.x 00000000005D02FD Unknown Unknown Unknown pw.x 000000000040BE64 Unknown Unknown Unknown pw.x 000000000040BDE2 Unknown Unknown Unknown libc.so.6 00007F49D8C7A1C4 Unknown Unknown Unknown pw.x 000000000040BD29 Unknown Unknown Unknown It doesn't matter if I just execute ./pw.x or "mpirun -n 4 ./pw.x". It crashes immediately with or without redirected input. In the post to the forum pw.x ran for a little bit and then crashed (this is the best so far). In that case the following configure was used: ./configure CXX=mpic++ CC=mpicc F90=mpif90 F77=mpif77 MPIF90=mpif90 see attachment configmpi It crashes serial or parallel (i.e. with or without "mpirun -n 4"). Also, I tried gfortran but it basically went like intel where the crash was immediate. Thanks for any help, Todd Beaudet PS. Are you married now? My wife and I are expecting our second son around Aug. 15! I hope all is well. ---- Original message ---- >Date: Fri, 13 Jun 2008 15:53:37 -0400 >From: "Nichols A. Romero" >Subject: Re: [Pw_forum] new user problem running Si >To: "PWSCF Forum" > > Todd, > > You should tells us the following information. > 1. compiler > 2. math library > 3. serial or parallel run > 4. PWscf version > > On Fri, Jun 13, 2008 at 3:35 PM, > wrote: > > experienced PWscf users, > > I have compiled the code. When I run the Si > example the code crashes after the second > iteration... > > iteration # 2 ecut= 18.00 Ry beta=0.70 > CG style diagonalization > ethr = 7.69E-04, avg # of iterations = 3.0 > p0_3521: p4_error: interrupt SIGSEGV: 11 > > when I backtrace using a debugger I get the > following: > > (gdb) backtrace > #0 0x000000000050ed55 in __scf__mix_type_axpy () > #1 0x00000000004e2770 in mix_rho_ () > #2 0x00000000004cae58 in electrons_ () > #3 0x000000000040f39c in MAIN__ () > > I suspect a problem with the environment when I > compiled the code. I'm using Intel Core2 Quad CPU > Q6600 and Ubuntu 8.04 - the Hardy Heron. Below is > the input file I was using. Does anyone have a > idea on how I might proceed? I can provide more > information if necessary. > > &control > calculation = 'scf' > restart_mode='from_scratch', > prefix='silicon', > tstress = .true. > tprnfor = .true. > pseudo_dir = > '/home/beaudet/espresso/espresso-4.0/pseudo/', > outdir='/tmp/' > / > &system > ibrav= 2, celldm(1) =10.20, nat= 2, ntyp= 1, > ecutwfc =18.0, > / > &electrons > diagonalization='cg' > mixing_mode = 'plain' > mixing_beta = 0.7 > conv_thr = 1.0d-8 > / > ATOMIC_SPECIES > Si 28.086 Si.vbc.UPF > ATOMIC_POSITIONS > Si 0.00 0.00 0.00 > Si 0.25 0.25 0.25 > K_POINTS > 10 > 0.1250000 0.1250000 0.1250000 1.00 > 0.1250000 0.1250000 0.3750000 3.00 > 0.1250000 0.1250000 0.6250000 3.00 > 0.1250000 0.1250000 0.8750000 3.00 > 0.1250000 0.3750000 0.3750000 3.00 > 0.1250000 0.3750000 0.6250000 6.00 > 0.1250000 0.3750000 0.8750000 6.00 > 0.1250000 0.6250000 0.6250000 3.00 > 0.3750000 0.3750000 0.3750000 1.00 > 0.3750000 0.3750000 0.6250000 3.00 > > Thanks, > > Todd Beaudet > _______________________________________________ > 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) >________________ >_______________________________________________ >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: configintel Type: application/octet-stream Size: 4761 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080616/5cbb3457/attachment.obj -------------- next part -------------- A non-text attachment was scrubbed... Name: configmpi Type: application/octet-stream Size: 5027 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080616/5cbb3457/attachment-0001.obj From degironc at sissa.it Tue Jun 17 09:48:48 2008 From: degironc at sissa.it (Stefano de Gironcoli) Date: Tue, 17 Jun 2008 09:48:48 +0200 Subject: [Pw_forum] new user problem running Si In-Reply-To: <20080616145836.BGS83973@expms2.cites.uiuc.edu> References: <20080616145836.BGS83973@expms2.cites.uiuc.edu> Message-ID: <48576C60.1090701@sissa.it> have you tried to make a scalar installation first, in order to verify whether the problem is the compiler or the parallel configuration ? stefano beaudet at uiuc.edu wrote: > Nick, > > Sorry it has taken a while to get back to you. After I posted I had a meeting, and several other things. I tried recompiling several different ways and while I get through the compile SEEMINGLY OK the code crashes at various points - usually at the very beginning. > > I've got: > Quantum Espresso 4.0 > Intel compilers version 10.1.011 and mkl version 10.0.1.014 > BLAS and LAPACK are up to date on my Umbuntu distribution > > I suspect it has to do with the parallel environment more than the particular compiler - but I could be wrong. > > The simplest thing I've tried is using intel compilers. I used the following configure command: > > ./configure CXX=icpc CC=icc F90=ifort F77=ifort MPIF90=ifort > > see attachment configintel > > When I run an example I get: > > forrtl: severe (174): SIGSEGV, segmentation fault occurred > Image PC Routine Line Source > libc.so.6 00007F49D8CD6980 Unknown Unknown Unknown > libc.so.6 00007F49D8CD7481 Unknown Unknown Unknown > libmpich.so.1.0 00007F49D94C75BF Unknown Unknown Unknown > libmpich.so.1.0 00007F49D94D40E2 Unknown Unknown Unknown > libmpich.so.1.0 00007F49D94D4259 Unknown Unknown Unknown > libmpich.so.1.0 00007F49D94A3DC8 Unknown Unknown Unknown > libmpich.so.1.0 00007F49D9494A43 Unknown Unknown Unknown > libmpich.so.1.0 00007F49D94B86C9 Unknown Unknown Unknown > libmpich.so.1.0 00007F49D94B7FE4 Unknown Unknown Unknown > pw.x 000000000048F539 Unknown Unknown Unknown > pw.x 00000000005D02FD Unknown Unknown Unknown > pw.x 000000000040BE64 Unknown Unknown Unknown > pw.x 000000000040BDE2 Unknown Unknown Unknown > libc.so.6 00007F49D8C7A1C4 Unknown Unknown Unknown > pw.x 000000000040BD29 Unknown Unknown Unknown > > It doesn't matter if I just execute ./pw.x or "mpirun -n 4 ./pw.x". It crashes immediately with or without redirected input. > > In the post to the forum pw.x ran for a little bit and then crashed (this is the best so far). In that case the following configure was used: > > ./configure CXX=mpic++ CC=mpicc F90=mpif90 F77=mpif77 MPIF90=mpif90 > > see attachment configmpi > > It crashes serial or parallel (i.e. with or without "mpirun -n 4"). > > Also, I tried gfortran but it basically went like intel where the crash was immediate. > > Thanks for any help, > > Todd Beaudet > > PS. Are you married now? My wife and I are expecting our second son around Aug. 15! I hope all is well. > > > ---- Original message ---- > >> Date: Fri, 13 Jun 2008 15:53:37 -0400 >> From: "Nichols A. Romero" >> Subject: Re: [Pw_forum] new user problem running Si >> To: "PWSCF Forum" >> >> Todd, >> >> You should tells us the following information. >> 1. compiler >> 2. math library >> 3. serial or parallel run >> 4. PWscf version >> >> On Fri, Jun 13, 2008 at 3:35 PM, >> wrote: >> >> experienced PWscf users, >> >> I have compiled the code. When I run the Si >> example the code crashes after the second >> iteration... >> >> iteration # 2 ecut= 18.00 Ry beta=0.70 >> CG style diagonalization >> ethr = 7.69E-04, avg # of iterations = 3.0 >> p0_3521: p4_error: interrupt SIGSEGV: 11 >> >> when I backtrace using a debugger I get the >> following: >> >> (gdb) backtrace >> #0 0x000000000050ed55 in __scf__mix_type_axpy () >> #1 0x00000000004e2770 in mix_rho_ () >> #2 0x00000000004cae58 in electrons_ () >> #3 0x000000000040f39c in MAIN__ () >> >> I suspect a problem with the environment when I >> compiled the code. I'm using Intel Core2 Quad CPU >> Q6600 and Ubuntu 8.04 - the Hardy Heron. Below is >> the input file I was using. Does anyone have a >> idea on how I might proceed? I can provide more >> information if necessary. >> >> &control >> calculation = 'scf' >> restart_mode='from_scratch', >> prefix='silicon', >> tstress = .true. >> tprnfor = .true. >> pseudo_dir = >> '/home/beaudet/espresso/espresso-4.0/pseudo/', >> outdir='/tmp/' >> / >> &system >> ibrav= 2, celldm(1) =10.20, nat= 2, ntyp= 1, >> ecutwfc =18.0, >> / >> &electrons >> diagonalization='cg' >> mixing_mode = 'plain' >> mixing_beta = 0.7 >> conv_thr = 1.0d-8 >> / >> ATOMIC_SPECIES >> Si 28.086 Si.vbc.UPF >> ATOMIC_POSITIONS >> Si 0.00 0.00 0.00 >> Si 0.25 0.25 0.25 >> K_POINTS >> 10 >> 0.1250000 0.1250000 0.1250000 1.00 >> 0.1250000 0.1250000 0.3750000 3.00 >> 0.1250000 0.1250000 0.6250000 3.00 >> 0.1250000 0.1250000 0.8750000 3.00 >> 0.1250000 0.3750000 0.3750000 3.00 >> 0.1250000 0.3750000 0.6250000 6.00 >> 0.1250000 0.3750000 0.8750000 6.00 >> 0.1250000 0.6250000 0.6250000 3.00 >> 0.3750000 0.3750000 0.3750000 1.00 >> 0.3750000 0.3750000 0.6250000 3.00 >> >> Thanks, >> >> Todd Beaudet >> _______________________________________________ >> 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) >> ________________ >> _______________________________________________ >> 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 degironc at sissa.it Tue Jun 17 10:03:51 2008 From: degironc at sissa.it (Stefano de Gironcoli) Date: Tue, 17 Jun 2008 10:03:51 +0200 Subject: [Pw_forum] about Phonon results In-Reply-To: References: Message-ID: <48576FE7.9040707@sissa.it> Eduardo Ariel Menendez Proupin wrote: > This happens because > a) 0 frecuencies at Gamma occurs because the energy is is invariant > with respect to arbitrary translations of the whole crystal (the same > translation for all atoms). From this assumption the accoustic sum > rule (ASR) is obtained theoretically. > b) The plane wave DFT calculations do not have this invariance > exactly, because of the plane wave cutoff. The energy is invariant not > for arbitrary translations, but for translations in the FFT grid. > c) I guess that the absolute translation invariance is lost in the > back and forth FFTs of the exchange-correlation potential, which is > non linear, and probably in some part related with the ultrasoft > pseudopotentials. Someone correct me if I am wrong, please. everything correct, except for the clarification of a detail. The problem is actually not in the finite kinetic energy cutoff but in the fact that the exchange and correlation energy is computed in real space on a discrete grid and hence the total energy is invariant (as you said) only for translation in the FFT grid. Increasing the charge density cutoff increases the grid density thus making the integral more exact thus reducing the problem.... unfortunately rather slowly... This problem is usually more severe for GGA than with LDA because the GGA functionals have functional forms that vary more strongly with the position; particularly so for isolated molecules or system with significant portions of "vaccum" because in the exponential tail of the charge density a) the finite cutoff (hence there is an effect due to cutoff) induces oscillations in rho and b) the reduced gradient is diverging stefano > You may increase the density cutoff (ecutrho) and see if the first > three frequencies get closer to zero. First do it with one of the > ESPRESSO examples, that run in minutes. I guess that I made that test > some years ago. > Using larger cutoffs increase a lot the calculation time. That is why > you have the tricky option to impose the ASR in postprocessing (matdyn.x). > > Practical way. If your "wrong" frecuencies are smaller than 50 cm^-1, > this is normal. Apply the ASR in the input for matdyn. Check that the > other frequencies do not change more than 1 or 2 cm^-1. Then, if this > makes you happy, go to write your report > > >I have another question about the output: What is the >difference > between the total magnetic moment and the >absolute magnetic moment. > This has been asked in the forum before. Please, google for this. If I > remember well, the absolute magnetic moment is the integral of the > absolute value of the magnetization. > > Regards > Eduardo Menendez > > On Sat, Jun 14, 2008 at 2:57 AM, > wrote: > > Dear my pwscf friends, > > Recently, I am trying to do phonon calculations in Heusler. As a > first step, I tried to reproduce the results of a paper by > Zayak(PRB, 72, 054113, 2005). I choose Ni2MnGe as an example. To > my surprise, the acoustic branches at Gamma point gives a > frequency of about 1 THz, which should be zero. Other branches > agree with the published data. > > I wonder what might be the reason? I attached my scf, and phonon > input files, and the phonon out put files. > > I have another question about the output: What is the difference > between the total magnetic moment and the absolute magnetic moment. > > Thank you and wish you a happy weekend! > > Wenxu Zhang > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From T.Maitra at tnw.utwente.nl Tue Jun 17 15:17:05 2008 From: T.Maitra at tnw.utwente.nl (T.Maitra at tnw.utwente.nl) Date: Tue, 17 Jun 2008 15:17:05 +0200 Subject: [Pw_forum] regarding Wannier transport code Message-ID: <9E7F7554631C0047AB598527A4E9569B01EF66A4@tnx4.dynamic.tnw.utwente.nl> Hello, I am a new user of wannier transport code (wanT). I want to ask two (probably naive) questions to the experts of the code. 1. How does the programme 'kgrid.f90' which is supposed to calculate the k-points in the full Brillouin Zone take into account the lattice structure? I mean, for example the hexagonal lattice and simple cubic should have different set of k-points. 2. When I try to do a transport calculation with single k-point along the transport direction the code doesn't calculate R = (0,0,1) block of the Hamiltonian matrix. But with more than k-points along the transport direction it does it. Why is that so? Thanks in advance for your reply. Best regards Tulika Maitra ----------------------------------------- Tulika Maitra Computational Materials Science Faculty of Science and Technology University of Twente P.O. Box 217 7500 AE Enschede The Netherlands Tel. +31-53-4893168 Fax: +31-53-4892910 E-mail: t.maitra at tnw.utwente.nl ------------------------------------------- From giannozz at democritos.it Tue Jun 17 15:44:32 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Tue, 17 Jun 2008 15:44:32 +0200 Subject: [Pw_forum] new user problem running Si In-Reply-To: <20080616145836.BGS83973@expms2.cites.uiuc.edu> References: <20080616145836.BGS83973@expms2.cites.uiuc.edu> Message-ID: <4857BFC0.2080808@democritos.it> beaudet at uiuc.edu wrote: > ./configure CXX=icpc CC=icc F90=ifort F77=ifort MPIF90=ifort try "./configure --disable-parallel CC=icc F90=ifort" first. If it works, download and compile (with ifort as frotran compiler) openmpi. Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From degironc at sissa.it Tue Jun 17 16:00:04 2008 From: degironc at sissa.it (Stefano de Gironcoli) Date: Tue, 17 Jun 2008 16:00:04 +0200 Subject: [Pw_forum] regarding Wannier transport code In-Reply-To: <9E7F7554631C0047AB598527A4E9569B01EF66A4@tnx4.dynamic.tnw.utwente.nl> References: <9E7F7554631C0047AB598527A4E9569B01EF66A4@tnx4.dynamic.tnw.utwente.nl> Message-ID: <4857C364.2010601@sissa.it> Dear T. Maitra, are you addressing your question to the right forum ? I found no file named kgrid.f90 in the QE distribution. Stefano de Gironcoli - SISSA and DEMOCRITOS T.Maitra at tnw.utwente.nl wrote: > Hello, > > I am a new user of wannier transport code (wanT). I want to ask two (probably naive) questions > to the experts of the code. > > 1. How does the programme 'kgrid.f90' which is supposed to calculate the k-points > in the full Brillouin Zone take into account the lattice structure? I mean, for example > the hexagonal lattice and simple cubic should have different set of k-points. > > 2. When I try to do a transport calculation with single k-point along the transport direction > the code doesn't calculate R = (0,0,1) block of the Hamiltonian matrix. But with more than > k-points along the transport direction it does it. Why is that so? > > Thanks in advance for your reply. > > Best regards > Tulika Maitra > > > ----------------------------------------- > Tulika Maitra > Computational Materials Science > Faculty of Science and Technology > University of Twente > P.O. Box 217 > 7500 AE Enschede > The Netherlands > > Tel. +31-53-4893168 > Fax: +31-53-4892910 > E-mail: t.maitra at tnw.utwente.nl > ------------------------------------------- > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From T.Maitra at tnw.utwente.nl Tue Jun 17 16:04:56 2008 From: T.Maitra at tnw.utwente.nl (T.Maitra at tnw.utwente.nl) Date: Tue, 17 Jun 2008 16:04:56 +0200 Subject: [Pw_forum] regarding Wannier transport code References: <9E7F7554631C0047AB598527A4E9569B01EF66A4@tnx4.dynamic.tnw.utwente.nl> <4857C364.2010601@sissa.it> Message-ID: <9E7F7554631C0047AB598527A4E9569B01EF66A5@tnx4.dynamic.tnw.utwente.nl> Dear Dr. Stefano de Gironcoli, Sorry, I thought Pwscf forum also addresses questions on Wannier transport (wanT) code. Please let me know if I am wrong. -Thanks & regards Tulika -----Original Message----- From: pw_forum-bounces at pwscf.org on behalf of Stefano de Gironcoli Sent: Tue 6/17/2008 4:00 PM To: PWSCF Forum Subject: Re: [Pw_forum] regarding Wannier transport code Dear T. Maitra, are you addressing your question to the right forum ? I found no file named kgrid.f90 in the QE distribution. Stefano de Gironcoli - SISSA and DEMOCRITOS T.Maitra at tnw.utwente.nl wrote: > Hello, > > I am a new user of wannier transport code (wanT). I want to ask two (probably naive) questions > to the experts of the code. > > 1. How does the programme 'kgrid.f90' which is supposed to calculate the k-points > in the full Brillouin Zone take into account the lattice structure? I mean, for example > the hexagonal lattice and simple cubic should have different set of k-points. > > 2. When I try to do a transport calculation with single k-point along the transport direction > the code doesn't calculate R = (0,0,1) block of the Hamiltonian matrix. But with more than > k-points along the transport direction it does it. Why is that so? > > Thanks in advance for your reply. > > Best regards > Tulika Maitra > > > ----------------------------------------- > Tulika Maitra > Computational Materials Science > Faculty of Science and Technology > University of Twente > P.O. Box 217 > 7500 AE Enschede > The Netherlands > > Tel. +31-53-4893168 > Fax: +31-53-4892910 > E-mail: t.maitra at tnw.utwente.nl > ------------------------------------------- > > _______________________________________________ > 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 degironc at sissa.it Tue Jun 17 16:42:45 2008 From: degironc at sissa.it (Stefano de Gironcoli) Date: Tue, 17 Jun 2008 16:42:45 +0200 Subject: [Pw_forum] regarding Wannier transport code In-Reply-To: <9E7F7554631C0047AB598527A4E9569B01EF66A5@tnx4.dynamic.tnw.utwente.nl> References: <9E7F7554631C0047AB598527A4E9569B01EF66A4@tnx4.dynamic.tnw.utwente.nl> <4857C364.2010601@sissa.it> <9E7F7554631C0047AB598527A4E9569B01EF66A5@tnx4.dynamic.tnw.utwente.nl> Message-ID: <4857CD65.8060000@sissa.it> Dear T. Maitra, I think there are WanT users (and developers) among the subscribers to PWscf forum and therefore your query may get some reply. I just wanted to remind you that you are asking information about a code we have no control of, so maybe there is some other more effective mean to get help. stefano T.Maitra at tnw.utwente.nl wrote: > Dear Dr. Stefano de Gironcoli, > > Sorry, I thought Pwscf forum also addresses questions on Wannier transport (wanT) code. > Please let me know if I am wrong. > > -Thanks & regards > Tulika > > > -----Original Message----- > From: pw_forum-bounces at pwscf.org on behalf of Stefano de Gironcoli > Sent: Tue 6/17/2008 4:00 PM > To: PWSCF Forum > Subject: Re: [Pw_forum] regarding Wannier transport code > > Dear T. Maitra, > are you addressing your question to the right forum ? > I found no file named kgrid.f90 in the QE distribution. > Stefano de Gironcoli - SISSA and DEMOCRITOS > > T.Maitra at tnw.utwente.nl wrote: > >> Hello, >> >> I am a new user of wannier transport code (wanT). I want to ask two (probably naive) questions >> to the experts of the code. >> >> 1. How does the programme 'kgrid.f90' which is supposed to calculate the k-points >> in the full Brillouin Zone take into account the lattice structure? I mean, for example >> the hexagonal lattice and simple cubic should have different set of k-points. >> >> 2. When I try to do a transport calculation with single k-point along the transport direction >> the code doesn't calculate R = (0,0,1) block of the Hamiltonian matrix. But with more than >> k-points along the transport direction it does it. Why is that so? >> >> Thanks in advance for your reply. >> >> Best regards >> Tulika Maitra >> >> >> ----------------------------------------- >> Tulika Maitra >> Computational Materials Science >> Faculty of Science and Technology >> University of Twente >> P.O. Box 217 >> 7500 AE Enschede >> The Netherlands >> >> Tel. +31-53-4893168 >> Fax: +31-53-4892910 >> E-mail: t.maitra at tnw.utwente.nl >> ------------------------------------------- >> >> _______________________________________________ >> 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 > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From andrea.ferretti at unimore.it Tue Jun 17 17:22:41 2008 From: andrea.ferretti at unimore.it (Andrea Ferretti) Date: Tue, 17 Jun 2008 17:22:41 +0200 (CEST) Subject: [Pw_forum] regarding Wannier transport code In-Reply-To: <4857CD65.8060000@sissa.it> References: <9E7F7554631C0047AB598527A4E9569B01EF66A4@tnx4.dynamic.tnw.utwente.nl> <4857C364.2010601@sissa.it> <9E7F7554631C0047AB598527A4E9569B01EF66A5@tnx4.dynamic.tnw.utwente.nl> <4857CD65.8060000@sissa.it> Message-ID: Dear Stefano and T. Maitra, > I think there are WanT users (and developers) among the subscribers > to PWscf forum and therefore your query may get some reply. > I just wanted to remind you that you are asking information > about a code we have no control of, so maybe there is some other more > effective mean to get help. > stefano > WanT is a different project wrt Q-Espresso and at the moment is not distributted with it, while it is tightly interfaced to, and it is supposed to work with, Q-E. Said that, I think that while questions about the Q-E/WanT interface may be proper to this forum, other questions purely related to WanT can be probably better suited for the WanT developers (myself, or anybody else from the page http://www.wannier-transport.org/contacts.html )... >> T.Maitra at tnw.utwente.nl wrote: >> >>> 1. How does the programme 'kgrid.f90' which is supposed to calculate the k-points >>> in the full Brillouin Zone take into account the lattice structure? I mean, for example >>> the hexagonal lattice and simple cubic should have different set of k-points. basically kgrid.f90 generates a kpt mesh expressed in terms of the reciprocal lattice vectors ("crystal" units), avoiding any symmetrization of the kpts... such procedure is needed because the usage of symmetries is not currently implemented in want-2.1 (released version of the code), while it is going on in the cvs version... >>> >>> 2. When I try to do a transport calculation with single k-point along the transport direction >>> the code doesn't calculate R = (0,0,1) block of the Hamiltonian matrix. But with more than >>> k-points along the transport direction it does it. Why is that so? >>> basically, the inequivalent R vectors (direct lattice vect) you can obtain are related to the monkhorst-pack grid you are using.. 1kpt -> 1 R vect = (0 0 0) this is because the Hamiltonian matrix elements on the localized basis are obtained as a discrete fourier transform: H(R) = 1/Nk \sum_k H(k) e^-ikR this is why we need to setup more that one kpt in the transport direction... From baroni at sissa.it Tue Jun 17 17:28:40 2008 From: baroni at sissa.it (Stefano Baroni) Date: Tue, 17 Jun 2008 17:28:40 +0200 Subject: [Pw_forum] regarding Wannier transport code In-Reply-To: References: <9E7F7554631C0047AB598527A4E9569B01EF66A4@tnx4.dynamic.tnw.utwente.nl> <4857C364.2010601@sissa.it> <9E7F7554631C0047AB598527A4E9569B01EF66A5@tnx4.dynamic.tnw.utwente.nl> <4857CD65.8060000@sissa.it> Message-ID: still, I would find it perfectly legitamate that people needing a helping hand on WanT seek it here cheers - SB On Jun 17, 2008, at 5:22 PM, Andrea Ferretti wrote: > > Dear Stefano and T. Maitra, > >> I think there are WanT users (and developers) among the >> subscribers >> to PWscf forum and therefore your query may get some reply. >> I just wanted to remind you that you are asking information >> about a code we have no control of, so maybe there is some other more >> effective mean to get help. >> stefano >> > > WanT is a different project wrt Q-Espresso and at the moment is not > distributted with it, while it is tightly > interfaced to, and it is supposed to work with, Q-E. > > Said that, I think that while questions about the Q-E/WanT interface > may > be proper to this forum, other questions purely related to > WanT can be probably better suited for the WanT developers (myself, or > anybody else from the page http://www.wannier-transport.org/contacts.html > )... > > >>> T.Maitra at tnw.utwente.nl wrote: >>> >>>> 1. How does the programme 'kgrid.f90' which is supposed to >>>> calculate the k-points >>>> in the full Brillouin Zone take into account the lattice >>>> structure? I mean, for example >>>> the hexagonal lattice and simple cubic should have different set >>>> of k-points. > > basically kgrid.f90 generates a kpt mesh expressed in terms of the > reciprocal lattice vectors ("crystal" units), avoiding any > symmetrization > of the kpts... > > such procedure is needed because the usage of symmetries is not > currently implemented in want-2.1 (released version of the code), > while it is going on in the cvs version... > >>>> >>>> 2. When I try to do a transport calculation with single k-point >>>> along the transport direction >>>> the code doesn't calculate R = (0,0,1) block of the Hamiltonian >>>> matrix. But with more than >>>> k-points along the transport direction it does it. Why is that so? >>>> > > basically, the inequivalent R vectors (direct > lattice vect) you can obtain are related to the monkhorst-pack grid > you are using.. > 1kpt -> 1 R vect = (0 0 0) > > this is because the Hamiltonian matrix elements on the localized > basis are > obtained as a discrete fourier transform: > > H(R) = 1/Nk \sum_k H(k) e^-ikR > > this is why we need to setup more that one kpt in the transport > direction... > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget 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/20080617/0c29a635/attachment.htm From T.Maitra at tnw.utwente.nl Tue Jun 17 19:19:12 2008 From: T.Maitra at tnw.utwente.nl (T.Maitra at tnw.utwente.nl) Date: Tue, 17 Jun 2008 19:19:12 +0200 Subject: [Pw_forum] regarding Wannier transport code References: <9E7F7554631C0047AB598527A4E9569B01EF66A4@tnx4.dynamic.tnw.utwente.nl><4857C364.2010601@sissa.it><9E7F7554631C0047AB598527A4E9569B01EF66A5@tnx4.dynamic.tnw.utwente.nl><4857CD65.8060000@sissa.it> Message-ID: <9E7F7554631C0047AB598527A4E9569B01EF66A6@tnx4.dynamic.tnw.utwente.nl> Dear Dr. Ferretti, Thank you very much for your e-mail with detail answers to my questions. Now it is more clear to me. Best regards Tulika ----------------------------------------- Tulika Maitra Computational Materials Science Faculty of Science and Technology University of Twente P.O. Box 217 7500 AE Enschede The Netherlands Tel. +31-53-4893168 Fax: +31-53-4892910 E-mail: t.maitra at tnw.utwente.nl ------------------------------------------- -----Original Message----- From: pw_forum-bounces at pwscf.org on behalf of Andrea Ferretti Sent: Tue 6/17/2008 5:22 PM To: PWSCF Forum Subject: Re: [Pw_forum] regarding Wannier transport code Dear Stefano and T. Maitra, > I think there are WanT users (and developers) among the subscribers > to PWscf forum and therefore your query may get some reply. > I just wanted to remind you that you are asking information > about a code we have no control of, so maybe there is some other more > effective mean to get help. > stefano > WanT is a different project wrt Q-Espresso and at the moment is not distributted with it, while it is tightly interfaced to, and it is supposed to work with, Q-E. Said that, I think that while questions about the Q-E/WanT interface may be proper to this forum, other questions purely related to WanT can be probably better suited for the WanT developers (myself, or anybody else from the page http://www.wannier-transport.org/contacts.html )... >> T.Maitra at tnw.utwente.nl wrote: >> >>> 1. How does the programme 'kgrid.f90' which is supposed to calculate the k-points >>> in the full Brillouin Zone take into account the lattice structure? I mean, for example >>> the hexagonal lattice and simple cubic should have different set of k-points. basically kgrid.f90 generates a kpt mesh expressed in terms of the reciprocal lattice vectors ("crystal" units), avoiding any symmetrization of the kpts... such procedure is needed because the usage of symmetries is not currently implemented in want-2.1 (released version of the code), while it is going on in the cvs version... >>> >>> 2. When I try to do a transport calculation with single k-point along the transport direction >>> the code doesn't calculate R = (0,0,1) block of the Hamiltonian matrix. But with more than >>> k-points along the transport direction it does it. Why is that so? >>> basically, the inequivalent R vectors (direct lattice vect) you can obtain are related to the monkhorst-pack grid you are using.. 1kpt -> 1 R vect = (0 0 0) this is because the Hamiltonian matrix elements on the localized basis are obtained as a discrete fourier transform: H(R) = 1/Nk \sum_k H(k) e^-ikR this is why we need to setup more that one kpt in the transport direction... _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum From kondor.jess at gmail.com Tue Jun 17 19:59:04 2008 From: kondor.jess at gmail.com (Jess Kondor) Date: Tue, 17 Jun 2008 12:59:04 -0500 Subject: [Pw_forum] Berry Phase calculation for gdir=1,2 in parallel Message-ID: <1d9d5d9d0806171059i35c5e09bj551656de8e5f26df@mail.gmail.com> Dear All, I have noticed that calculation of polarization using Berry Phase method and pwscf-4.0 doesn't work in parallel for gdir=1 and gdir=2. For example, I took tetragonal phase of BaTiO3 with atomic displacements along 'x' direction and (c,a,a) lattice vectors. For nprocs =1, it works fine: VALUES OF POLARIZATION ~~~~~~~~~~~~~~~~~~~~~~ The calculation of phases done along the direction of vector 1 of the reciprocal lattice gives the following contribution to the polarization vector (in different units, and being Omega the volume of the unit cell): P = 2.2164982 (mod 15.2155768) (e/Omega).bohr P = 0.0051349 (mod 0.0352497) e/bohr^2 P = 0.2935730 (mod 2.0152882) C/m^2 The polarization direction is: ( 1.00000 , 0.00000 , 0.00000 ) but for nprocs > 1 (for example, 16) it doesn't: ================================================== POLARIZATION CALCULATION !!! NOT THOROUGHLY TESTED !!! -------------------------------------------------- error: translated G= -0.2483941319370271 0.000000000000000 0.000000000000000 with crystal coordinates -1 0 0 corresponds to ng= 0 but G(ng)= 2.8071327799762095E-319 0.000000000000000 1.4927122354773060E-316 probably because G_par is NOT a reciprocal lattice vector Possible choices as smallest G_par: i= 1 G= 0.000000000000000 0.000000000000000 0.000000000000000 i= 2 G= 0.000000000000000 0.000000000000000 -0.2508780732563974 i= 3 G= 0.000000000000000 0.000000000000000 0.2508780732563974 i= 4 G= 0.000000000000000 0.000000000000000 -0.5017561465127947 i= 5 G= 0.000000000000000 0.000000000000000 0.5017561465127947 i= 6 G= 0.4967882638740543 -0.2508780732563974 0.000000000000000 i= 7 G= 0.4967882638740543 -0.2508780732563974 -0.2508780732563974 ....... is there a way to solve this problem? Best regards, -- ===================================== Jess Kondor, PhD student PICYT, Av. Venustiano Carranza 2425-A, San Luis Potos? 78210, M?xico ===================================== -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080617/c07a15af/attachment-0001.htm From umari at democritos.it Wed Jun 18 15:32:06 2008 From: umari at democritos.it (umari at democritos.it) Date: Wed, 18 Jun 2008 15:32:06 +0200 Subject: [Pw_forum] Berry Phase calculation for gdir=1, 2 in parallel (Paolo Umari) Message-ID: <20080618153206.tjzu67gu0wogc4o4@mail.democritos.it> Dear Jess, the Berry's phase routines are not parallelized yet but for gdir=3. However, the Berry's phase polarization is also calculated when finite electric fields are applied. This part of the code has been recently parallelized for any direction. You should have a look at the example31 of the CVS version Greetings, Paolo Umari, Democritos ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program. From xwzhang at uestc.edu.cn Wed Jun 18 16:53:34 2008 From: xwzhang at uestc.edu.cn (xwzhang at uestc.edu.cn) Date: Wed, 18 Jun 2008 22:53:34 +0800 Subject: [Pw_forum] my understanding about ASR Message-ID: Dear Eduardo and Stefano, Thank you for clear and careful explaination of my phonon problem. I am now doing some tests and getting more experiences, because I am a freshman in phonon calculation. It seems to me that ASR can "force" the acoustic phonon at Gamma point to zero. A naive thought is that can we just ignore it? In this case, are there any problem that the slope of the phonon around Gamma is incorrect, because ASR cure the problem of k=0, but what about other points near it? Regards Wenxu Zhang State Key Lab. of thin films and integrated devices , Univ. of Elec. Sci. & Tech. of China. Chengdu, 610054, China >Eduardo Ariel Menendez Proupin wrote: >> This happens because >> a) 0 frecuencies at Gamma occurs because the energy is is invariant >> with respect to arbitrary translations of the whole crystal (the same >> translation for all atoms). From this assumption the accoustic sum >> rule (ASR) is obtained theoretically. >> b) The plane wave DFT calculations do not have this invariance >> exactly, because of the plane wave cutoff. The energy is invariant not >> for arbitrary translations, but for translations in the FFT grid. >> c) I guess that the absolute translation invariance is lost in the >> back and forth FFTs of the exchange-correlation potential, which is >> non linear, and probably in some part related with the ultrasoft >> pseudopotentials. Someone correct me if I am wrong, please. >everything correct, except for the clarification of a detail. >The problem is actually not in the finite kinetic energy cutoff but in >the fact that the exchange and correlation energy is computed in real >space on a discrete grid and hence the total energy is invariant (as you >said) only for translation in the FFT grid. >Increasing the charge density cutoff increases the grid density thus >making the integral more exact thus reducing the problem.... >unfortunately rather slowly... >This problem is usually more severe for GGA than with LDA because the >GGA functionals have functional forms that vary more strongly with the >position; particularly so for isolated molecules or system with >significant portions of "vaccum" because in the exponential tail of the >charge density a) the finite cutoff (hence there is an effect due to >cutoff) induces oscillations in rho and b) the reduced gradient is diverging > >stefano > >> You may increase the density cutoff (ecutrho) and see if the first >> three frequencies get closer to zero. First do it with one of the >> ESPRESSO examples, that run in minutes. I guess that I made that test >> some years ago. >> Using larger cutoffs increase a lot the calculation time. That is why >> you have the tricky option to impose the ASR in postprocessing (matdyn.x). >> >> Practical way. If your "wrong" frecuencies are smaller than 50 cm^-1, >> this is normal. Apply the ASR in the input for matdyn. Check that the >> other frequencies do not change more than 1 or 2 cm^-1. Then, if this >> makes you happy, go to write your report >> >> >I have another question about the output: What is the >difference >> between the total magnetic moment and the >absolute magnetic moment. >> This has been asked in the forum before. Please, google for this. If I >> remember well, the absolute magnetic moment is the integral of the >> absolute value of the magnetization. >> >> Regards >> Eduardo Menendez >> >> On Sat, Jun 14, 2008 at 2:57 AM, >> > wrote: >> >> Dear my pwscf friends, >> >> Recently, I am trying to do phonon calculations in Heusler. As a >> first step, I tried to reproduce the results of a paper by >> Zayak(PRB, 72, 054113, 2005). I choose Ni2MnGe as an example. To >> my surprise, the acoustic branches at Gamma point gives a >> frequency of about 1 THz, which should be zero. Other branches >> agree with the published data. >> >> I wonder what might be the reason? I attached my scf, and phonon >> input files, and the phonon out put files. >> >> I have another question about the output: What is the difference >> between the total magnetic moment and the absolute magnetic moment. >> >> Thank you and wish you a happy weekend! >> >> Wenxu Zhang >> >> >> ------------------------------------------------------------------------ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080618/88a739f1/attachment.htm From zhaohscas at yahoo.com.cn Wed Jun 18 17:48:03 2008 From: zhaohscas at yahoo.com.cn (zhaohscas) Date: Wed, 18 Jun 2008 23:48:03 +0800 Subject: [Pw_forum] About the spline interpolation for charge density. Message-ID: <200806182348021565918@yahoo.com.cn> Hi all, I have the following 18*18*18 charge dentity datagrid as follows: ------------------------ a b c Charge 1 1 1 0.430034694385803 2 1 1 0.433590624875083 3 1 1 0.444263507693938 4 1 1 0.461808085757678 5 1 1 0.485486354576373 6 1 1 0.514303497336158 ... ... ... 18 18 18 ------------------------ Where, the a, b, and c is FFT grid point used for sample the charge, and the last column is the corresponding charge density at the FFT grid point defined by crystal vectors: (a,b,c). I want to make a spline interpolation to this datagid to get a more dense and smooth charge density datagrid, I think this should be done by spline interpolation. I've read the the source plotrho.f90 of pp.x, and cann't figure it out how can I make my issue solved. Who can give me some hints? Sincerely yours, -------------- Hongsheng Zhao Xinjiang Technical Institute of Physics and Chemistry Chinese Academy of Sciences GnuPG DSA: 0xD108493 2008-06-18 From giannozz at democritos.it Wed Jun 18 21:21:20 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Wed, 18 Jun 2008 21:21:20 +0200 Subject: [Pw_forum] About the spline interpolation for charge density. In-Reply-To: <200806182348021565918@yahoo.com.cn> References: <200806182348021565918@yahoo.com.cn> Message-ID: On Jun 18, 2008, at 17:48 , zhaohscas wrote: > I have the following 18*18*18 charge dentity datagrid as follows: > > ------------------------ > a b c Charge > 1 1 1 0.430034694385803 > 2 1 1 0.433590624875083 > 3 1 1 0.444263507693938 > 4 1 1 0.461808085757678 > 5 1 1 0.485486354576373 > 6 1 1 0.514303497336158 > ... ... ... > 18 18 18 > ------------------------ > > Where, the a, b, and c is FFT grid point used for sample the > charge, and the last column is the corresponding charge density at > the FFT grid point defined by crystal vectors: (a,b,c). > > I want to make a spline interpolation to this datagrid to get a > more dense and smooth charge density datagrid, I think this should > be done by spline interpolation. > > I've read the the source plotrho.f90 of pp.x, and can't figure it > out how can I make my issue solved. plotrho.f90 performs simple linear interpolation. You may have a look at code voronoy.f90: it performs a Fourier interpolation, i.e. transforms to real to reciprocal space, pads additional Fourier components with zero, transforms back to real space onto a denser grid. 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 eariel99 at gmail.com Thu Jun 19 14:26:42 2008 From: eariel99 at gmail.com (Eduardo Ariel Menendez Proupin) Date: Thu, 19 Jun 2008 08:26:42 -0400 Subject: [Pw_forum] my understanding about ASR Message-ID: Dear Zhang, The ASR "cures" the dynamical matrix or the force constant matrix., or maybe both. The ASR can be applied by q2r.x (calculation of force constants) or matdyn.x (calculation and diagonalization of the dynamical matrices). I do not know , and I wuld like to know, if one should apply it twice or only in one of the steps. In any case, the 0 frequency at Gamma and the slopes are consistent. However, the you must check that frequencies that are far from zero change little due to the ASR. Moreover, if you study the convergence vs cutoffs (this may be long and painful), check that the convergence trend of the frequencies is towards the values given with the ASR. Note that negative frequencies may indicate instabilities in the structure, e.g., atomic positions not relaxed. This does not seems to be the case in the ouputs that you sent. Negative frequencies at points other than Gamma uncover the so called dynamical instability. The dynamical instability indicates that using a larger supercell you can find relaxations of the atomic positions (and maybe a phase transition) that are avoided by the symmetry imposed by the smaller supercell. I do not know if the ASR can mask the dynamical instabilities. Hope this helps. eduardo Dear Eduardo and Stefano, Thank you for clear and careful explaination of my phonon problem. I am now doing some tests and getting more experiences, because I am a freshman in phonon calculation. It seems to me that ASR can "force" the acoustic phonon at Gamma point to zero. A naive thought is that can we just ignore it? In this case, are there any problem that the slope of the phonon around Gamma is incorrect, because ASR cure the problem of k=0, but what about other points near it? -- Eduardo Menendez -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080619/c307a91d/attachment.htm From giannozz at democritos.it Thu Jun 19 15:36:41 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Thu, 19 Jun 2008 15:36:41 +0200 Subject: [Pw_forum] my understanding about ASR In-Reply-To: References: Message-ID: <485A60E9.7080209@democritos.it> xwzhang at uestc.edu.cn wrote: > are there any problem that the slope of the phonon around Gamma is > incorrect, because ASR cure the problem of k=0, but what about other > points near it? for phonons at k/=0, the error due to ASR violation seems to be usually negligible. I never tried to make calculations with very small q, though. Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From stewart at cnf.cornell.edu Thu Jun 19 16:17:41 2008 From: stewart at cnf.cornell.edu (stewart at cnf.cornell.edu) Date: Thu, 19 Jun 2008 10:17:41 -0400 Subject: [Pw_forum] my understanding about ASR In-Reply-To: <485A60E9.7080209@democritos.it> References: <485A60E9.7080209@democritos.it> Message-ID: <20080619141741.59250.qmail@mail.spidergraphics.com> I recently studied Kohn anomalies in palladium using Quantum Espresso. For this, I calculated the phonon dispersion with a 8x8x8 q-grid and I also did a series of direct calculations along the direction with the Kohn anomaly. Overall the agreement was very good. I found a small deviation between the two approaches for small q as you would expect. The maximum error at the Gamma point was 0.07 THz. For a comparison along the Gamma to X line, see Fig 2 (D. A. Stewart, New. J. Phys. 10 043025 (2008)). Of course, the difference you get between ASR results and direct calculations will depend strongly on how well relaxed your structure is. http://dx.doi.org/10.1088/1367-2630/10/4/043025 Best regards, Derek Paolo Giannozzi writes: > xwzhang at uestc.edu.cn wrote: > >> are there any problem that the slope of the phonon around Gamma is >> incorrect, because ASR cure the problem of k=0, but what about other >> points near it? > > for phonons at k/=0, the error due to ASR violation seems to be > usually negligible. I never tried to make calculations with > very small q, though. > > Paolo > -- > Paolo Giannozzi, Democritos and University of Udine, Italy > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ################################ Derek Stewart, Ph. D. Scientific Computation Associate http://www.people.cornell.edu/pages/das248/ 250 Duffield Hall Cornell Nanoscale Facility (CNF) Ithaca, NY 14853 stewart (at) cnf.cornell.edu (607) 255-2856 From baroni at sissa.it Thu Jun 19 18:16:02 2008 From: baroni at sissa.it (Stefano Baroni) Date: Thu, 19 Jun 2008 18:16:02 +0200 Subject: [Pw_forum] my understanding about ASR In-Reply-To: <20080619141741.59250.qmail@mail.spidergraphics.com> References: <485A60E9.7080209@democritos.it> <20080619141741.59250.qmail@mail.spidergraphics.com> Message-ID: On Jun 19, 2008, at 4:17 PM, stewart at cnf.cornell.edu wrote: > Fig 2 (D. A. Stewart, New. J. Phys. 10 043025 (2008)). Of course, the > difference you get between ASR results and direct calculations will > depend > strongly on how well relaxed your structure is. hmmm: this may well be the case, but if this is so, it must be for a nontrivial reason (which, if true, would qualify the "of course" as slightly misplaced). the total force acting on any isolated system is of course zero, even though the system is not at equilibrium ("not well relaxed"). in particular, this total force is invariant ("invariantly zero!") with respect to a homogeneous translation of the system. therefore, the system must have a zero mode (actually three of them: one for each cartesian coordinate), irrespective on how well (or how bad) the internal degrees of freedom have been optimized. is this wrong? SB > > > http://dx.doi.org/10.1088/1367-2630/10/4/043025 > > Best regards, > > Derek > > > Paolo Giannozzi writes: > >> xwzhang at uestc.edu.cn wrote: >> >>> are there any problem that the slope of the phonon around Gamma is >>> incorrect, because ASR cure the problem of k=0, but what about other >>> points near it? >> >> for phonons at k/=0, the error due to ASR violation seems to be >> usually negligible. I never tried to make calculations with >> very small q, though. >> >> Paolo >> -- >> Paolo Giannozzi, Democritos and University of Udine, Italy >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> > > > > ################################ > Derek Stewart, Ph. D. > Scientific Computation Associate > http://www.people.cornell.edu/pages/das248/ > 250 Duffield Hall > Cornell Nanoscale Facility (CNF) > Ithaca, NY 14853 > stewart (at) cnf.cornell.edu > (607) 255-2856 > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget 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/20080619/98008550/attachment.htm From stewart at cnf.cornell.edu Thu Jun 19 19:10:45 2008 From: stewart at cnf.cornell.edu (stewart at cnf.cornell.edu) Date: Thu, 19 Jun 2008 13:10:45 -0400 Subject: [Pw_forum] my understanding about ASR In-Reply-To: References: <485A60E9.7080209@democritos.it> <20080619141741.59250.qmail@mail.spidergraphics.com> Message-ID: <20080619171045.76038.qmail@mail.spidergraphics.com> Hi Stefano, You raise a good point and perhaps I should have been more careful in my previous statement. I agree that based on symmetry, the total forces on the system should be zero and there should be a set of zero mode. Applying the acoustic sum rule reinstates this fact. From my practical experience, I recall having greater problems with negative frequencies at the Gamma point for systems that had the lattice constant greater than the equilibrium value (expanded system). This occurred for direct calculations at small q and also based on interpolating the phonon dispersion from the q grid. Here you get phonon softening which shifts the phonon dispersion curves down and perhaps this leads to problems with calculating the phonon dispersion from the given q grid near Gamma. I took this as an indication of the instability of the expanded structure, but perhaps this is not the best way to view it. Best regards, Derek Stefano Baroni writes: > > On Jun 19, 2008, at 4:17 PM, stewart at cnf.cornell.edu wrote: > >> Fig 2 (D. A. Stewart, New. J. Phys. 10 043025 (2008)). Of course, the >> difference you get between ASR results and direct calculations will >> depend >> strongly on how well relaxed your structure is. > > hmmm: this may well be the case, but if this is so, it must be for a > nontrivial reason (which, if true, would qualify the "of course" as > slightly misplaced). > > the total force acting on any isolated system is of course zero, even > though the system is not at equilibrium ("not well relaxed"). in > particular, this total force is invariant ("invariantly zero!") with > respect to a homogeneous translation of the system. therefore, the system > must have a zero mode (actually three of them: one for each cartesian > coordinate), irrespective on how well (or how bad) the internal degrees > of freedom have been optimized. > > is this wrong? > > SB > >> >> >> http://dx.doi.org/10.1088/1367-2630/10/4/043025 >> >> Best regards, >> >> Derek >> >> >> Paolo Giannozzi writes: >> >>> xwzhang at uestc.edu.cn wrote: >>> >>>> are there any problem that the slope of the phonon around Gamma is >>>> incorrect, because ASR cure the problem of k=0, but what about other >>>> points near it? >>> >>> for phonons at k/=0, the error due to ASR violation seems to be >>> usually negligible. I never tried to make calculations with >>> very small q, though. >>> >>> Paolo >>> -- >>> Paolo Giannozzi, Democritos and University of Udine, Italy >>> _______________________________________________ >>> Pw_forum mailing list >>> Pw_forum at pwscf.org >>> http://www.democritos.it/mailman/listinfo/pw_forum >>> >> >> >> >> ################################ >> Derek Stewart, Ph. D. >> Scientific Computation Associate >> http://www.people.cornell.edu/pages/das248/ >> 250 Duffield Hall >> Cornell Nanoscale Facility (CNF) >> Ithaca, NY 14853 >> stewart (at) cnf.cornell.edu >> (607) 255-2856 >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum > > --- > Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - > Trieste > [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) > > La morale est une logique de l'action comme la logique est une morale de > la pens?e - Jean Piaget > > Please, if possible, don't send me MS Word or PowerPoint attachments > Why? See: http://www.gnu.org/philosophy/no-word-attachments.html > > > > ################################ Derek Stewart, Ph. D. Scientific Computation Associate 250 Duffield Hall Cornell Nanoscale Facility (CNF) Ithaca, NY 14853 stewart (at) cnf.cornell.edu (607) 255-2856 From baroni at sissa.it Thu Jun 19 20:11:51 2008 From: baroni at sissa.it (Stefano Baroni) Date: Thu, 19 Jun 2008 20:11:51 +0200 Subject: [Pw_forum] my understanding about ASR In-Reply-To: <20080619171045.76038.qmail@mail.spidergraphics.com> References: <485A60E9.7080209@democritos.it> <20080619141741.59250.qmail@mail.spidergraphics.com> <20080619171045.76038.qmail@mail.spidergraphics.com> Message-ID: <8358AB1F-15FE-41DA-9BFE-DF047B4B74C1@sissa.it> Soft modes at small (but finite) q can indeed be an indication of an elastic or a mechanical instability. At q=0, there should be exactly thre zero modes, though, irrespective of the occurrence of such an instability (I believe). Imaginary frequencies can also occur at q=0, but _in addition_ to the three zero modes mentioned above. cheers - SB On Jun 19, 2008, at 7:10 PM, stewart at cnf.cornell.edu wrote: > Hi Stefano, > > You raise a good point and perhaps I should have been more careful > in my > previous statement. I agree that based on symmetry, the total > forces on the > system should be zero and there should be a set of zero mode. > Applying the > acoustic sum rule reinstates this fact. From my practical > experience, I > recall having greater problems with negative frequencies at the > Gamma point > for systems that had the lattice constant greater than the > equilibrium value > (expanded system). This occurred for direct calculations at small q > and > also based on interpolating the phonon dispersion from the q grid. > Here you > get phonon softening which shifts the phonon dispersion curves down > and > perhaps this leads to problems with calculating the phonon > dispersion from > the given q grid near Gamma. I took this as an indication of the > instability of the expanded structure, but perhaps this is not the > best way > to view it. > > Best regards, > > Derek > > > Stefano Baroni writes: > >> >> On Jun 19, 2008, at 4:17 PM, stewart at cnf.cornell.edu wrote: >> >>> Fig 2 (D. A. Stewart, New. J. Phys. 10 043025 (2008)). Of course, >>> the >>> difference you get between ASR results and direct calculations will >>> depend >>> strongly on how well relaxed your structure is. >> >> hmmm: this may well be the case, but if this is so, it must be for a >> nontrivial reason (which, if true, would qualify the "of course" as >> slightly misplaced). >> >> the total force acting on any isolated system is of course zero, even >> though the system is not at equilibrium ("not well relaxed"). in >> particular, this total force is invariant ("invariantly zero!") with >> respect to a homogeneous translation of the system. therefore, the >> system >> must have a zero mode (actually three of them: one for each >> cartesian >> coordinate), irrespective on how well (or how bad) the internal >> degrees >> of freedom have been optimized. >> >> is this wrong? >> >> SB >> >>> >>> >>> http://dx.doi.org/10.1088/1367-2630/10/4/043025 >>> >>> Best regards, >>> >>> Derek >>> >>> >>> Paolo Giannozzi writes: >>> >>>> xwzhang at uestc.edu.cn wrote: >>>> >>>>> are there any problem that the slope of the phonon around Gamma is >>>>> incorrect, because ASR cure the problem of k=0, but what about >>>>> other >>>>> points near it? >>>> >>>> for phonons at k/=0, the error due to ASR violation seems to be >>>> usually negligible. I never tried to make calculations with >>>> very small q, though. >>>> >>>> Paolo >>>> -- >>>> Paolo Giannozzi, Democritos and University of Udine, Italy >>>> _______________________________________________ >>>> Pw_forum mailing list >>>> Pw_forum at pwscf.org >>>> http://www.democritos.it/mailman/listinfo/pw_forum >>>> >>> >>> >>> >>> ################################ >>> Derek Stewart, Ph. D. >>> Scientific Computation Associate >>> http://www.people.cornell.edu/pages/das248/ >>> 250 Duffield Hall >>> Cornell Nanoscale Facility (CNF) >>> Ithaca, NY 14853 >>> stewart (at) cnf.cornell.edu >>> (607) 255-2856 >>> _______________________________________________ >>> Pw_forum mailing list >>> Pw_forum at pwscf.org >>> http://www.democritos.it/mailman/listinfo/pw_forum >> >> --- >> Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - >> Trieste >> [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) >> >> La morale est une logique de l'action comme la logique est une >> morale de >> la pens?e - Jean Piaget >> >> Please, if possible, don't send me MS Word or PowerPoint attachments >> Why? See: http://www.gnu.org/philosophy/no-word-attachments.html >> >> >> >> > > > > ################################ > Derek Stewart, Ph. D. > Scientific Computation Associate > 250 Duffield Hall > Cornell Nanoscale Facility (CNF) > Ithaca, NY 14853 > stewart (at) cnf.cornell.edu > (607) 255-2856 > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste http://www.sissa.it/~baroni / [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget 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/20080619/aa95c74e/attachment-0001.htm From eariel99 at gmail.com Thu Jun 19 21:39:36 2008 From: eariel99 at gmail.com (Eduardo Ariel Menendez Proupin) Date: Thu, 19 Jun 2008 15:39:36 -0400 Subject: [Pw_forum] my understanding about ASR Message-ID: The forces are zero also if the system is rotated. So, why dont' we get 5 zero frecuencies in the crystal, as in a dimer or 6 as in a complex molecule? Is this is forced by the periodic boundary conditions? -- Eduardo Menendez -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080619/aa962123/attachment.htm From javier.fuhr at gmail.com Thu Jun 19 22:04:37 2008 From: javier.fuhr at gmail.com (Javier Fuhr) Date: Thu, 19 Jun 2008 17:04:37 -0300 Subject: [Pw_forum] wfcdir and oldwfc Message-ID: Dear PWscf users, I am relaxing a rather large system with pw.x in parallel mode in a cluster. I set wfcdir pointing to directories in the local hard drives of the machines. However, the big *.oldwfc* and *.old2wfc* files, necessary for the wfc extrapolation (I set wfc_extrapolation = 'second_order'), are written in outdir which is pointing to a directory in the NFS server. Therefore, the extrapolation takes a lot of time. Is there a way to write these files in the local directories?, or is it advised not to use wfc_extrapolation? Thanks, Javier -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080619/11ee788a/attachment.htm From w2agz at pacbell.net Fri Jun 20 06:55:05 2008 From: w2agz at pacbell.net (Paul M. Grant) Date: Thu, 19 Jun 2008 21:55:05 -0700 Subject: [Pw_forum] Print Modules in pw.x Message-ID: <01e901c8d291$cf6f7c20$6e4e7460$@net> What .f90 sources compile to whatever .o modules that control/direct printing to the output text files for pw.x? I want to recompile it to set additional grep tags similar to the "!" flag for total energy. Paul M. Grant, PhD Principal, W2AGZ Technologies Visiting Scholar, Applied Physics, Stanford (2005-2008) 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/20080619/9db18d14/attachment.htm From baroni at sissa.it Fri Jun 20 10:00:02 2008 From: baroni at sissa.it (Stefano Baroni) Date: Fri, 20 Jun 2008 10:00:02 +0200 Subject: [Pw_forum] my understanding about ASR In-Reply-To: References: Message-ID: <8A4EF750-DDD6-4FFB-8930-8404F11F7389@sissa.it> yes SB On Jun 19, 2008, at 9:39 PM, Eduardo Ariel Menendez Proupin wrote: > The forces are zero also if the system is rotated. So, why dont' we > get 5 zero frecuencies in the crystal, as in a dimer or 6 as in a > complex molecule? Is this is forced by the periodic boundary > conditions? > > -- > Eduardo Menendez _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget 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/20080620/574507d8/attachment.htm From giannozz at democritos.it Fri Jun 20 10:24:41 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Fri, 20 Jun 2008 10:24:41 +0200 Subject: [Pw_forum] wfcdir and oldwfc In-Reply-To: References: Message-ID: <485B6949.9030004@democritos.it> Javier Fuhr wrote: > I set wfcdir pointing to directories in the local hard drives > of the machines. However, the big *.oldwfc* and *.old2wfc* files, > necessary for the wfc extrapolation (I set wfc_extrapolation = > 'second_order'), are written in outdir [...] Is there a way to > write these files in the local directories? yes, but you have to modify the code. See how variables 'wfc_dir' and 'tmp_dir' are used in PW/openfilf90, do the same where files 'oldwfc' and 'old2wfc' are opened in PW/update_pot.f90 . It is clumsy and obscure, it will be fixed sooner or later > or is it advised not to use wfc_extrapolation? it doesn't make a big difference. Note that wavefunction extrapolation doesn't work for USPP in versions < 4. Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From giannozz at democritos.it Fri Jun 20 10:36:18 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Fri, 20 Jun 2008 10:36:18 +0200 Subject: [Pw_forum] Print Modules in pw.x In-Reply-To: <01e901c8d291$cf6f7c20$6e4e7460$@net> References: <01e901c8d291$cf6f7c20$6e4e7460$@net> Message-ID: <485B6C02.6000105@democritos.it> Paul M. Grant wrote: > What .f90 sources compile to whatever .o modules that control/direct > printing to the output text files for pw.x? printing is (unfortunately) done in many different places. Most of the interesting stuff is printed in routines "electrons.f90" and "print_ks_energies.f90" Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From cyrille.barreteau at cea.fr Fri Jun 20 09:59:39 2008 From: cyrille.barreteau at cea.fr (Cyrille Barreteau) Date: Fri, 20 Jun 2008 09:59:39 +0200 Subject: [Pw_forum] wfcdir and oldwfc In-Reply-To: References: Message-ID: <485B636B.6050905@cea.fr> Dear Javier I had the same problem on our cluster since by default the home directory was the NSF server (master node). You should be able to write on each node. In our case the /local/ directory is mounted on each node therefore choosing the outdir as /local/javier.... should solve your problem. best regards cyrille -- ================================================================== Cyrille Barreteau | phone : +33 (0)1 69 08 29 51 CEA Saclay | fax : +33 (0)1 69 08 84 46 DSM/IRAMIS/SPCSI | email cyrille.barreteau at cea.fr Batiment 462 | 91191 Gif sur Yvette Cedex FRANCE ~~~~~~~~~~~~~~~~~~~~~~~~ http://iramis.cea.fr/Images/Pisp/cbarreteau/cbarreteau_fr.html ================================================================== Javier Fuhr wrote: > Dear PWscf users, > > I am relaxing a rather large system with pw.x in parallel mode in a > cluster. I set wfcdir pointing to directories in the local hard drives > of the machines. However, the big *.oldwfc* and *.old2wfc* files, > necessary for the wfc extrapolation (I set wfc_extrapolation = > 'second_order'), are written in outdir which is pointing to a > directory in the NFS server. Therefore, the extrapolation takes a lot > of time. Is there a way to write these files in the local > directories?, or is it advised not to use wfc_extrapolation? > > Thanks, > Javier > >------------------------------------------------------------------------ > >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum > > From ggokoglu at hacettepe.edu.tr Fri Jun 20 09:58:27 2008 From: ggokoglu at hacettepe.edu.tr (Gokhan GOKOGLU) Date: Fri, 20 Jun 2008 07:58:27 +0000 (GMT) Subject: [Pw_forum] Phonon anomaly in GaSb Message-ID: Dear PWSCF users, I tried to calculate phonon dipersion of zincblende GaSb, but I found large negative modes at Gamma-point. The same system was studied by Giannozzi et al. in a previous study (PRB 43, 7231 (1991)) using LDA functionals and the obtained dispersion was quite nice without any anomalous vibrational property. I used GGA (PW91) vanderbilt US PP. I also checked the kinetic energy cut-off (70 Ry) and k-mesh (16x16x16) for convergence. I didn't find any phonon anomaly in AlSb with similar geometry. I give the dynamical matrix below. I would be very grateful, if someone could give an idea. Thank you very much.. Gokhan Gokoglu Dept. Physics Bilkent University, Ankara, Turkey Diagonalizing the dynamical matrix q = ( 0.000000000 0.000000000 0.000000000 ) ************************************************************************** omega( 1) = -9.779567 [THz] = -326.213411 [cm-1] ( -0.646207 0.000000 0.460941 0.000000 -0.354330 0.000000 ) ( 0.367516 0.000000 -0.262150 0.000000 0.201517 0.000000 ) omega( 2) = -9.779567 [THz] = -326.213411 [cm-1] ( 0.504782 0.000000 0.707669 0.000000 0.000000 0.000000 ) ( -0.287083 0.000000 -0.402471 0.000000 0.000000 0.000000 ) omega( 3) = -9.779567 [THz] = -326.213411 [cm-1] ( -0.288464 0.000000 0.205762 0.000000 0.793757 0.000000 ) ( 0.164057 0.000000 -0.117023 0.000000 -0.451432 0.000000 ) omega( 4) = -0.606056 [THz] = -20.215980 [cm-1] ( 0.172018 0.000000 0.676699 0.000000 0.095247 0.000000 ) ( 0.173197 0.000000 0.681338 0.000000 0.095900 0.000000 ) omega( 5) = -0.606056 [THz] = -20.215980 [cm-1] ( -0.434105 0.000000 0.184061 0.000000 -0.523697 0.000000 ) ( -0.437081 0.000000 0.185323 0.000000 -0.527287 0.000000 ) omega( 6) = -0.606056 [THz] = -20.215980 [cm-1] ( 0.527775 0.000000 -0.069163 0.000000 -0.461794 0.000000 ) ( 0.531394 0.000000 -0.069637 0.000000 -0.464960 0.000000 ) ************************************************************************** From w2agz at pacbell.net Fri Jun 20 15:13:44 2008 From: w2agz at pacbell.net (Paul Grant) Date: Fri, 20 Jun 2008 06:13:44 -0700 (PDT) Subject: [Pw_forum] Print Modules in pw.x Message-ID: <756122.85184.qm@web82506.mail.mud.yahoo.com> Bingo! Think I found what I want in electrons.f90. Molte Grazie di Nuovo, Paolo. ----- Original Message ---- From: Paolo Giannozzi To: PWSCF Forum Sent: Friday, June 20, 2008 1:36:18 AM Subject: Re: [Pw_forum] Print Modules in pw.x Paul M. Grant wrote: > What .f90 sources compile to whatever .o modules that control/direct > printing to the output text files for pw.x? printing is (unfortunately) done in many different places. Most of the interesting stuff is printed in routines "electrons.f90" and "print_ks_energies.f90" Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080620/3da77665/attachment.htm From amosleffler at sbcglobal.net Fri Jun 20 20:59:35 2008 From: amosleffler at sbcglobal.net (Amos Leffler) Date: Fri, 20 Jun 2008 11:59:35 -0700 Subject: [Pw_forum] Running examples in espresso-4.0 Message-ID: Dear Forum, I just compiled espresso-4.0 using gfortran with OpenSuSE 10.2 in 64 bit mode. The entire process consisted of the two commands: ./configure make all The compilation went without error. However, when I tried to run some examples (1,2,7) they all failed with the same type message as given below. example/example01 :starting This example shows how to use pw.x to calculate the total energy and the band structure of four simple systems: Si, Al, Cu, Ni. executables directory ../espresso-4.0/bin pseudo directory ../espresso-4.0/pseudo temporary directory: /root/tmp checking that needed directories and files exist... done running pw.x as: ../espresso-4.0/bin/pw.x running bands.x as ../espresso-4.0/bin/bands.x cleaning /root/tmp... done running the scf calcuation for Si... ..espresso-4.0/bin/pw.x : error while loading shared libraries :libguide.so cannot open shared object file: No such file or directory Error conditin encountered during test: exit status = 127 Aborting Since espresso-4.0 is so new I couldn't find anything about this problem. Hopefully there is a simple fix. Amos Leffler -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080620/50556e11/attachment-0001.htm -------------- next part -------------- A non-text attachment was scrubbed... Name: example01.out Type: application/octet-stream Size: 1476 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080620/50556e11/attachment-0001.obj From akohlmey at cmm.chem.upenn.edu Fri Jun 20 21:15:19 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Fri, 20 Jun 2008 15:15:19 -0400 (EDT) Subject: [Pw_forum] Running examples in espresso-4.0 In-Reply-To: References: Message-ID: On Fri, 20 Jun 2008, Amos Leffler wrote: AL> running pw.x as: ../espresso-4.0/bin/pw.x AL> running bands.x as ../espresso-4.0/bin/bands.x AL> AL> cleaning /root/tmp... done AL> running the scf calcuation for Si... ..espresso-4.0/bin/pw.x : error while loading shared AL> libraries :libguide.so cannot open shared object file: No such file or directory amos, you are using MKL right? please read the MKL installation instructions or link MKL statically. AL> Error conditin encountered during test: exit status = 127 AL> Aborting AL> AL> Since espresso-4.0 is so new I couldn't find anything about this AL> problem. Hopefully there is a simple fix. this has nothing to do with QE. axel. AL> AL> Amos Leffler AL> -- ======================================================================= 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 sauluck at iitk.ac.in Sat Jun 21 06:07:21 2008 From: sauluck at iitk.ac.in (Sushil Auluck) Date: Sat, 21 Jun 2008 09:37:21 +0530 Subject: [Pw_forum] Running examples in espresso-4.0 In-Reply-To: References: Message-ID: <485C7E79.6060706@iitk.ac.in> amos, it seems that libguide.so is not in your path. so either give tha library path in your .cshrc or .bashrc files.....OR just issue the appropriate command below....... it seems you have compiled with ifort and not gfortran as you mentioned in your email > > source /opt/intel/fc/*/bin/ifortvars.sh where * is the number > > source /opt/intel/cc/*/bin/iccvars.sh where * is the > compiler s.auluck -- ....................................................................... Prof. Sushil Auluck Phone:+91-512-6797092/6148 Department of Physics +91-512-6798177(Home) Indian Institute of Technology Cell :+91-9305548667 Kanpur 208016 (UP) Fax :+91-512-6790914 India E-mail:sauluck at iitk.ac.in ...............................................:sauluck at gmail.com http://www.iitk.ac.in/phy/People/phy_facvis.html http://www.iitk.ac.in/phy/New01/profile_SA.html ....................................................................... From raamesh123 at rediffmail.com Sat Jun 21 20:34:29 2008 From: raamesh123 at rediffmail.com (ramesh kumar) Date: 21 Jun 2008 18:34:29 -0000 Subject: [Pw_forum] spin polarized DOS Message-ID: <20080621183429.17759.qmail@f5mail-236-227.rediffmail.com> Dear Users i am new to PWSCF learning how to do band structure for magnetic materials (metals). can anybody kindly give link or tutorial file to run spin polarized DOS and band structure.. with regards ramesh -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080621/ef65bb14/attachment.htm From w2agz at pacbell.net Sun Jun 22 07:39:47 2008 From: w2agz at pacbell.net (Paul M. Grant) Date: Sat, 21 Jun 2008 22:39:47 -0700 Subject: [Pw_forum] Running examples in espresso-4.0 In-Reply-To: <485C7E79.6060706@iitk.ac.in> References: <485C7E79.6060706@iitk.ac.in> Message-ID: <003d01c8d42a$6287efd0$2797cf70$@net> BTW, it seems the "bug" in check_failure.sh still remains in espresso-4.0 ...at least re the version of bash included in the ubuntu distros. Paul M. Grant, PhD Principal, W2AGZ Technologies Visiting Scholar, Applied Physics, Stanford (2005-2008) 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 Sushil Auluck Sent: Friday, June 20, 2008 9:07 PM To: Amos Leffler; PWSCF Forum Subject: Re: [Pw_forum] Running examples in espresso-4.0 amos, it seems that libguide.so is not in your path. so either give tha library path in your .cshrc or .bashrc files.....OR just issue the appropriate command below....... it seems you have compiled with ifort and not gfortran as you mentioned in your email > > source /opt/intel/fc/*/bin/ifortvars.sh where * is the number > > source /opt/intel/cc/*/bin/iccvars.sh where * is the > compiler s.auluck -- ....................................................................... Prof. Sushil Auluck Phone:+91-512-6797092/6148 Department of Physics +91-512-6798177(Home) Indian Institute of Technology Cell :+91-9305548667 Kanpur 208016 (UP) Fax :+91-512-6790914 India E-mail:sauluck at iitk.ac.in ...............................................:sauluck at gmail.com http://www.iitk.ac.in/phy/People/phy_facvis.html http://www.iitk.ac.in/phy/New01/profile_SA.html ....................................................................... _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum From djaithania at hotmail.com Sun Jun 22 11:06:38 2008 From: djaithania at hotmail.com (hania djani-ait aissa) Date: Sun, 22 Jun 2008 09:06:38 +0000 Subject: [Pw_forum] cell optimization Message-ID: Dear Nicola, I have tried startingwfc='random' and it gave me the same energies values as for startingwfc='atomic' with still the same problem of drop and rise of total energy. I also increase Ecutoff to 40Ry, the energies values are actually more accurate but the problem persist. Hania &control calculation = 'scf' restart_mode = 'from_scratch' pseudo_dir = '/root/pwscf/pseudo/' outdir = '/root/tmp/' / &system ibrav=7 celldm(1)=7.33, celldm(3)=8.54 nat=19 ntyp=3 ecutwfc=25.0 ecutrho=100.0 / &electrons conv_thr = 1e-6, mixing_beta=0.2, startingwfc=?random? /ATOMIC_SPECIES Bi 208.98 Bi.pbe-d-mt.UPF Ti 47.867 Ti.pbe-sp-van_ak.UPF O 15.9994 O.pbe-van_ak.UPFATOMIC_POSITIONS Bi 0.00000000 0.00000000 0.5713404 Bi 0.00000000 0.00000000 7.9816596 Bi 0.00000000 0.00000000 1.8072489 Bi 0.00000000 0.00000000 6.7457511 Ti 0.00000000 0.00000000 3.1757289 Ti 0.00000000 0.00000000 5.3772711 Ti 0.00000000 0.00000000 4.2765000 O 0.50000000 0.00000000 0.0000 O 0.00000000 0.50000000 0.0000 O 0.50000000 0.00000000 2.13825 O 0.50000000 0.00000000 6.41475 O 0.00000000 0.00000000 3.771873 O 0.00000000 0.00000000 4.781127 O 0.00000000 0.00000000 2.7241305 O 0.00000000 0.00000000 5.8288695 O 0.50000000 0.00000000 1.0049775 O 0.50000000 0.00000000 7.5480225 O 0.00000000 0.50000000 1.0049775 O 0.00000000 0.50000000 7.5480225K_POINTS {automatic} 4 4 4 1 1 1 _________________________________________________________________ Sur Windows Live Ideas, d?couvrez en exclusivit? de nouveaux services en ligne... si nouveaux qu'ils ne sont pas encore sortis officiellement sur le march? ! http://ideas.live.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080622/5fac8723/attachment.htm From baroni at sissa.it Sun Jun 22 14:32:02 2008 From: baroni at sissa.it (Stefano Baroni) Date: Sun, 22 Jun 2008 14:32:02 +0200 Subject: [Pw_forum] cell optimization In-Reply-To: References: Message-ID: I may be wrong, but your problem could be due to subtelties of the US PP formalism. Using US PPs, I am afraid that PW calculations are variational with respect to PW expansion coefficients only when the augmentation charge is at convergence. Convergence with respect to the augmentation charge usually is a rather delicate issue. For sure, the cutoff for the charge-density distribution (ecutrho) should be taken larger (sometimes much larger) than the default value for norm- conserving PPs (4 x ecutwfc). In yor calculation, you seem to use just this value, which may turn out to be inadequate. Have you tried to test the convergence of your calcs wrt echutrho (keeping ecutwfc fixed)? SB On Jun 22, 2008, at 11:06 AM, hania djani-ait aissa wrote: > Dear Nicola, > I have tried startingwfc='random' and it gave me the same energies > values as for startingwfc='atomic' with still the same problem of > drop and rise of total energy. I also increase Ecutoff to 40Ry, the > energies values are actually more accurate but the problem persist. > > Hania > > > &control > calculation = 'scf' > restart_mode = 'from_scratch' > pseudo_dir = '/root/pwscf/pseudo/' > outdir = '/root/tmp/' > / > &system > ibrav=7 > celldm(1)=7.33, celldm(3)=8.54 > nat=19 > ntyp=3 > ecutwfc=25.0 > ecutrho=100.0 > / > &electrons > conv_thr = 1e-6, > mixing_beta=0.2, > startingwfc=?random? > / > ATOMIC_SPECIES > Bi 208.98 Bi.pbe-d-mt.UPF > Ti 47.867 Ti.pbe-sp-van_ak.UPF > O 15.9994 O.pbe-van_ak.UPF > ATOMIC_POSITIONS > Bi 0.00000000 0.00000000 0.5713404 > Bi 0.00000000 0.00000000 7.9816596 > Bi 0.00000000 0.00000000 1.8072489 > Bi 0.00000000 0.00000000 6.7457511 > Ti 0.00000000 0.00000000 3.1757289 > Ti 0.00000000 0.00000000 5.3772711 > Ti 0.00000000 0.00000000 4.2765000 > O 0.50000000 0.00000000 0.0000 > O 0.00000000 0.50000000 0.0000 > O 0.50000000 0.00000000 2.13825 > O 0.50000000 0.00000000 6.41475 > O 0.00000000 0.00000000 3.771873 > O 0.00000000 0.00000000 4.781127 > O 0.00000000 0.00000000 2.7241305 > O 0.00000000 0.00000000 5.8288695 > O 0.50000000 0.00000000 1.0049775 > O 0.50000000 0.00000000 7.5480225 > O 0.00000000 0.50000000 1.0049775 > O 0.00000000 0.50000000 7.5480225 > K_POINTS {automatic} > 4 4 4 1 1 1 > > > > T?l?chargez le nouveau Windows Live Messenger ! T?l?chargez > Messenger, c'est gratuit ! > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste http://www.sissa.it/~baroni / [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget 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/20080622/4d2ede7c/attachment.htm From djaithania at hotmail.com Sun Jun 22 17:35:18 2008 From: djaithania at hotmail.com (hania djani-ait aissa) Date: Sun, 22 Jun 2008 15:35:18 +0000 Subject: [Pw_forum] cell optimization In-Reply-To: References: Message-ID: Dear Baroni, thank you, but it seems i am not using US pps. However, I will increase ecutrho and see. thanks again hania From: baroni at sissa.itTo: pw_forum at pwscf.orgDate: Sun, 22 Jun 2008 14:32:02 +0200Subject: Re: [Pw_forum] cell optimizationI may be wrong, but your problem could be due to subtelties of the US PP formalism. Using US PPs, I am afraid that PW calculations are variational with respect to PW expansion coefficients only when the augmentation charge is at convergence. Convergence with respect to the augmentation charge usually is a rather delicate issue. For sure, the cutoff for the charge-density distribution (ecutrho) should be taken larger (sometimes much larger) than the default value for norm-conserving PPs (4 x ecutwfc). In yor calculation, you seem to use just this value, which may turn out to be inadequate. Have you tried to test the convergence of your calcs wrt echutrho (keeping ecutwfc fixed)? SB On Jun 22, 2008, at 11:06 AM, hania djani-ait aissa wrote: Dear Nicola,I have tried startingwfc='random' and it gave me the same energies values as for startingwfc='atomic' with still the same problem of drop and rise of total energy. I also increase Ecutoff to 40Ry, the energies values are actually more accurate but the problem persist. Hania &control calculation = 'scf' restart_mode = 'from_scratch' pseudo_dir = '/root/pwscf/pseudo/' outdir = '/root/tmp/' / &system ibrav=7 celldm(1)=7.33, celldm(3)=8.54 nat=19 ntyp=3 ecutwfc=25.0 ecutrho=100.0 / &electrons conv_thr = 1e-6, mixing_beta=0.2, startingwfc=?random? /ATOMIC_SPECIES Bi 208.98 Bi.pbe-d-mt.UPF Ti 47.867 Ti.pbe-sp-van_ak.UPF O 15.9994 O.pbe-van_ak.UPFATOMIC_POSITIONS Bi 0.00000000 0.00000000 0.5713404 Bi 0.00000000 0.00000000 7.9816596 Bi 0.00000000 0.00000000 1.8072489 Bi 0.00000000 0.00000000 6.7457511 Ti 0.00000000 0.00000000 3.1757289 Ti 0.00000000 0.00000000 5.3772711 Ti 0.00000000 0.00000000 4.2765000 O 0.50000000 0.00000000 0.0000 O 0.00000000 0.50000000 0.0000 O 0.50000000 0.00000000 2.13825 O 0.50000000 0.00000000 6.41475 O 0.00000000 0.00000000 3.771873 O 0.00000000 0.00000000 4.781127 O 0.00000000 0.00000000 2.7241305 O 0.00000000 0.00000000 5.8288695 O 0.50000000 0.00000000 1.0049775 O 0.50000000 0.00000000 7.5480225 O 0.00000000 0.50000000 1.0049775 O 0.00000000 0.50000000 7.5480225K_POINTS {automatic} 4 4 4 1 1 1 T?l?chargez le nouveau Windows Live Messenger ! T?l?chargez Messenger, c'est gratuit !_______________________________________________Pw_forum mailing listPw_forum at pwscf.orghttp://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste http://www.sissa.it/~baroni / [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html _________________________________________________________________ T?l?chargez le nouveau Windows Live Messenger ! http://get.live.com/messenger/overview -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080622/7552460d/attachment-0001.htm From baroni at sissa.it Sun Jun 22 17:41:40 2008 From: baroni at sissa.it (Stefano Baroni) Date: Sun, 22 Jun 2008 17:41:40 +0200 Subject: [Pw_forum] cell optimization In-Reply-To: References: Message-ID: <3CF8A96D-6C8E-412C-9A1D-652786F8D8C1@sissa.it> Dear Hania: it seems you do not know what you are doing. The acronym "van" in the PP file name in your input data means "Vanderbilt", i.e. "ultrasoft". Please, check the QE PP table for more details on the PP's you are using. SB On Jun 22, 2008, at 5:35 PM, hania djani-ait aissa wrote: > Dear Baroni, > thank you, but it seems i am not using US pps. However, I will > increase ecutrho and see. > thanks again > hania > > > From: baroni at sissa.it > To: pw_forum at pwscf.org > Date: Sun, 22 Jun 2008 14:32:02 +0200 > Subject: Re: [Pw_forum] cell optimization > > I may be wrong, but your problem could be due to subtelties of the > US PP formalism. Using US PPs, I am afraid that PW calculations are > variational with respect to PW expansion coefficients only when the > augmentation charge is at convergence. Convergence with respect to > the augmentation charge usually is a rather delicate issue. For > sure, the cutoff for the charge-density distribution (ecutrho) > should be taken larger (sometimes much larger) than the default > value for norm-conserving PPs (4 x ecutwfc). In yor calculation, you > seem to use just this value, which may turn out to be inadequate. > Have you tried to test the convergence of your calcs wrt echutrho > (keeping ecutwfc fixed)? SB > > On Jun 22, 2008, at 11:06 AM, hania djani-ait aissa wrote: > > Dear Nicola, > I have tried startingwfc='random' and it gave me the same energies > values as for startingwfc='atomic' with still the same problem of > drop and rise of total energy. I also increase Ecutoff to 40Ry, the > energies values are actually more accurate but the problem persist. > > Hania > > > &control > calculation = 'scf' > restart_mode = 'from_scratch' > pseudo_dir = '/root/pwscf/pseudo/' > outdir = '/root/tmp/' > / > &system > ibrav=7 > celldm(1)=7.33, celldm(3)=8.54 > nat=19 > ntyp=3 > ecutwfc=25.0 > ecutrho=100.0 > / > &electrons > conv_thr = 1e-6, > mixing_beta=0.2, > startingwfc=?random? > / > ATOMIC_SPECIES > Bi 208.98 Bi.pbe-d-mt.UPF > Ti 47.867 Ti.pbe-sp-van_ak.UPF > O 15.9994 O.pbe-van_ak.UPF > ATOMIC_POSITIONS > Bi 0.00000000 0.00000000 0.5713404 > Bi 0.00000000 0.00000000 7.9816596 > Bi 0.00000000 0.00000000 1.8072489 > Bi 0.00000000 0.00000000 6.7457511 > Ti 0.00000000 0.00000000 3.1757289 > Ti 0.00000000 0.00000000 5.3772711 > Ti 0.00000000 0.00000000 4.2765000 > O 0.50000000 0.00000000 0.0000 > O 0.00000000 0.50000000 0.0000 > O 0.50000000 0.00000000 2.13825 > O 0.50000000 0.00000000 6.41475 > O 0.00000000 0.00000000 3.771873 > O 0.00000000 0.00000000 4.781127 > O 0.00000000 0.00000000 2.7241305 > O 0.00000000 0.00000000 5.8288695 > O 0.50000000 0.00000000 1.0049775 > O 0.50000000 0.00000000 7.5480225 > O 0.00000000 0.50000000 1.0049775 > O 0.00000000 0.50000000 7.5480225 > K_POINTS {automatic} > 4 4 4 1 1 1 > > > > T?l?chargez le nouveau Windows Live Messenger ! T?l?chargez > Messenger, c'est gratuit ! > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > > --- > Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - > Trieste > http://www.sissa.it/~baroni / [+39] 040 3787 406 (tel) -528 (fax) / > stefanobaroni (skype) > > La morale est une logique de l'action comme la logique est une > morale de la pens?e - Jean Piaget > > Please, if possible, don't send me MS Word or PowerPoint attachments > Why? See: http://www.gnu.org/philosophy/no-word-attachments.html > > > > > > > > T?l?chargez le nouveau Windows Live Messenger ! T?l?chargez > Messenger, c'est gratuit ! > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste http://www.sissa.it/~baroni / [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget 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/20080622/af83ba1f/attachment.htm From eyvaz_isaev at yahoo.com Sun Jun 22 20:46:22 2008 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Sun, 22 Jun 2008 11:46:22 -0700 (PDT) Subject: [Pw_forum] cell optimization In-Reply-To: Message-ID: <611344.49606.qm@web65713.mail.ac4.yahoo.com> Dear Hania, First of all I would recommend you read at least Richard Martin's textbook. There are also nice review papers of W.Pickett (old, but good to understand the basic ideas), S.Baroni et al. (2002, Rev.Mod.Phys), Arias et al. in RMP (?) etc....... I suggest your problem raised by "ATOMIC_POSITIONS " keyword. By default it is "alat", i.e. atomic positions are in units of the lattice parameter "a". But looking at atomic positions provided it seems you use Angstrom. I.e. you used cartesian coordiantes in Angstrom. So, use ATOMIC_POSITIONS {angstrom} and see what happens. Bests, Eyvaz. > > > &control calculation = 'scf' restart_mode = > 'from_scratch' pseudo_dir = > '/root/pwscf/pseudo/' outdir = '/root/tmp/' > / &system ibrav=7 celldm(1)=7.33, > celldm(3)=8.54 nat=19 ntyp=3 ecutwfc=25.0 > ecutrho=100.0 / &electrons conv_thr = 1e-6, > mixing_beta=0.2, > startingwfc=?random? /ATOMIC_SPECIES Bi > 208.98 Bi.pbe-d-mt.UPF Ti 47.867 > Ti.pbe-sp-van_ak.UPF O 15.9994 > O.pbe-van_ak.UPFATOMIC_POSITIONS Bi 0.00000000 > 0.00000000 0.5713404 Bi 0.00000000 0.00000000 > 7.9816596 Bi 0.00000000 0.00000000 1.8072489 Bi > 0.00000000 0.00000000 6.7457511 Ti 0.00000000 > 0.00000000 3.1757289 Ti 0.00000000 0.00000000 > 5.3772711 Ti 0.00000000 0.00000000 4.2765000 O > 0.50000000 0.00000000 0.0000 O 0.00000000 > 0.50000000 0.0000 O 0.50000000 0.00000000 2.13825 O > 0.50000000 0.00000000 6.41475 O 0.00000000 > 0.00000000 3.771873 O 0.00000000 0.00000000 > 4.781127 O 0.00000000 0.00000000 2.7241305 O > 0.00000000 0.00000000 5.8288695 O 0.50000000 > 0.00000000 1.0049775 O 0.50000000 0.00000000 > 7.5480225 O 0.00000000 0.50000000 1.0049775 O > 0.00000000 0.50000000 7.5480225K_POINTS {automatic} > 4 4 4 1 1 1 > ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com From sks.jnc at gmail.com Sun Jun 22 20:53:38 2008 From: sks.jnc at gmail.com (S. K. S.) Date: Mon, 23 Jun 2008 00:23:38 +0530 Subject: [Pw_forum] "recover" tag Message-ID: Dear All, I'm running a phonon calculation on parallel cluster. The job got killed on the midway after reaching run time limit. Now while I'm trying to restart it using "recover" tag, I'm getting the following error: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from broyden : error # 1 factorization %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... I'm sure, the above error is because of "recover" tag. The same job gets over without any error while I submit it on the nodes having unlimited time-slot. It would be really nice if somebody tell me how to get rid of the above error. regards, SKS JNCASR Bangalore. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080623/e7aa07a9/attachment.htm From eyvaz_isaev at yahoo.com Sun Jun 22 21:05:13 2008 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Sun, 22 Jun 2008 12:05:13 -0700 (PDT) Subject: [Pw_forum] cell optimization In-Reply-To: Message-ID: <149855.59394.qm@web65716.mail.ac4.yahoo.com> Hi again, Please discard my previous mail regarding "alat". It seems to be correct, as I did not pay attention that you used celldim(1) and celldim(3). Please remember that celldim(3) is c/a. Nevertheless "conv_thr = 1e-6" looks quite high, use 1.e-8. For complex systems "mixing_mode=local-TF" keyword is very useful. Bests, Eyvaz. --- Stefano Baroni wrote: > I may be wrong, but your problem could be due to > subtelties of the US > PP formalism. Using US PPs, I am afraid that PW > calculations are > variational with respect to PW expansion > coefficients only when the > augmentation charge is at convergence. Convergence > with respect to the > augmentation charge usually is a rather delicate > issue. For sure, the > cutoff for the charge-density distribution (ecutrho) > should be taken > larger (sometimes much larger) than the default > value for norm- > conserving PPs (4 x ecutwfc). In yor calculation, > you seem to use just > this value, which may turn out to be inadequate. > Have you tried to > test the convergence of your calcs wrt echutrho > (keeping ecutwfc > fixed)? SB > > On Jun 22, 2008, at 11:06 AM, hania djani-ait aissa > wrote: > > > Dear Nicola, > > I have tried startingwfc='random' and it gave me > the same energies > > values as for startingwfc='atomic' with still the > same problem of > > drop and rise of total energy. I also increase > Ecutoff to 40Ry, the > > energies values are actually more accurate but the > problem persist. > > > > Hania > > > > > > &control > > calculation = 'scf' > > restart_mode = 'from_scratch' > > pseudo_dir = '/root/pwscf/pseudo/' > > outdir = '/root/tmp/' > > / > > &system > > ibrav=7 > > celldm(1)=7.33, celldm(3)=8.54 > > nat=19 > > ntyp=3 > > ecutwfc=25.0 > > ecutrho=100.0 > > / > > &electrons > > conv_thr = 1e-6, > > mixing_beta=0.2, > > startingwfc=?random? > > / > > ATOMIC_SPECIES > > Bi 208.98 Bi.pbe-d-mt.UPF > > Ti 47.867 Ti.pbe-sp-van_ak.UPF > > O 15.9994 O.pbe-van_ak.UPF > > ATOMIC_POSITIONS > > Bi 0.00000000 0.00000000 0.5713404 > > Bi 0.00000000 0.00000000 7.9816596 > > Bi 0.00000000 0.00000000 1.8072489 > > Bi 0.00000000 0.00000000 6.7457511 > > Ti 0.00000000 0.00000000 3.1757289 > > Ti 0.00000000 0.00000000 5.3772711 > > Ti 0.00000000 0.00000000 4.2765000 > > O 0.50000000 0.00000000 0.0000 > > O 0.00000000 0.50000000 0.0000 > > O 0.50000000 0.00000000 2.13825 > > O 0.50000000 0.00000000 6.41475 > > O 0.00000000 0.00000000 3.771873 > > O 0.00000000 0.00000000 4.781127 > > O 0.00000000 0.00000000 2.7241305 > > O 0.00000000 0.00000000 5.8288695 > > O 0.50000000 0.00000000 1.0049775 > > O 0.50000000 0.00000000 7.5480225 > > O 0.00000000 0.50000000 1.0049775 > > O 0.00000000 0.50000000 7.5480225 > > K_POINTS {automatic} > > 4 4 4 1 1 1 > > > > > > > > T?l?chargez le nouveau Windows Live Messenger ! > T?l?chargez > > Messenger, c'est gratuit ! > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > --- > Stefano Baroni - SISSA & DEMOCRITOS National > Simulation Center - > Trieste > http://www.sissa.it/~baroni / [+39] 040 3787 406 > (tel) -528 (fax) / > stefanobaroni (skype) > > La morale est une logique de l'action comme la > logique est une morale > de la pens?e - Jean Piaget > > Please, if possible, don't send me MS Word or > PowerPoint attachments > Why? See: > http://www.gnu.org/philosophy/no-word-attachments.html > > > > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com From sks.jnc at gmail.com Sun Jun 22 21:34:19 2008 From: sks.jnc at gmail.com (S. K. S.) Date: Mon, 23 Jun 2008 01:04:19 +0530 Subject: [Pw_forum] Phonon paralization Message-ID: Dear All, Is it possible in PWscf to do (phonon) representation number paralization as like as k-point or neb image paralization??? regards, SKS JNCASR Bangalore -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080623/85db7cd7/attachment.htm From giannozz at democritos.it Sun Jun 22 22:55:24 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Sun, 22 Jun 2008 22:55:24 +0200 Subject: [Pw_forum] Phonon paralization In-Reply-To: References: Message-ID: <618F78E5-2C1D-4A9D-A86A-DB8BF0D8F007@democritos.it> On Jun 22, 2008, at 21:34 , S. K. S. wrote: > Is it possible in PWscf to do (phonon) representation number > paralization as like as k-point or neb image paralization??? it is possible, it should be done, but it hasn't been done (yet) > 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 democritos.it Sun Jun 22 22:58:17 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Sun, 22 Jun 2008 22:58:17 +0200 Subject: [Pw_forum] "recover" tag In-Reply-To: References: Message-ID: On Jun 22, 2008, at 20:53 , S. K. S. wrote: > I'm running a phonon calculation on parallel cluster. [...] > Now while I'm trying to restart it using "recover" tag, > I'm getting the following error: > from broyden : error # 1 > factorization well known problem, solution not easy. The restarting mechanism in the phonon code is not very robust and occasionally gives raise to unexpected results. Several cases of bad restart have been fixed in v.4, but maybe not all of them. In any event: if your code crashes while writing a file, restarting will go wrong for sure 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 democritos.it Sun Jun 22 22:59:24 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Sun, 22 Jun 2008 22:59:24 +0200 Subject: [Pw_forum] Running examples in espresso-4.0 In-Reply-To: <003d01c8d42a$6287efd0$2797cf70$@net> References: <485C7E79.6060706@iitk.ac.in> <003d01c8d42a$6287efd0$2797cf70$@net> Message-ID: On Jun 22, 2008, at 7:39 , Paul M. Grant wrote: > BTW, it seems the "bug" in check_failure.sh still remains in > espresso-4.0 > ...at least re the version of bash included in the ubuntu distros. if I remember correctly the problem, there is no way to fix it without breaking it on some other machine 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 akohlmey at cmm.chem.upenn.edu Sun Jun 22 23:05:26 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sun, 22 Jun 2008 17:05:26 -0400 (EDT) Subject: [Pw_forum] Running examples in espresso-4.0 In-Reply-To: <003d01c8d42a$6287efd0$2797cf70$@net> References: <485C7E79.6060706@iitk.ac.in> <003d01c8d42a$6287efd0$2797cf70$@net> Message-ID: On Sat, 21 Jun 2008, Paul M. Grant wrote: PG> BTW, it seems the "bug" in check_failure.sh still remains in espresso-4.0 PG> ...at least re the version of bash included in the ubuntu distros. paul, to the best of my knowledge, the syntax in check_failure.sh is correct and the problem is most likely a "feature" in ubuntu. are you certain that /bin/sh is actually pointing to bash? unfortunately, there is no good solution. you could make /bin/sh a symlink to /bin/bash, but that will probably break other scripts, e.g., those that control your startup. that reminds be a lot of the situation way back when i had an IBM RS/6000 320h as desktop. the proper bourne shell was in /bin/bsh, but if you put that in your shell scripts, they become non-portable, and you couldn't change use /bin/bsh as /bin/sh as it would break the many and elaborate shell scripts that controlled the machine. . cheers, axel. PG> PG> Paul M. Grant, PhD PG> Principal, W2AGZ Technologies PG> Visiting Scholar, Applied Physics, Stanford (2005-2008) PG> EPRI Science Fellow (Retired) PG> IBM Research Staff Member Emeritus PG> w2agz at pacbell.net PG> http://www.w2agz.com 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 ceresoli at MIT.EDU Sun Jun 22 23:18:20 2008 From: ceresoli at MIT.EDU (Davide Ceresoli) Date: Sun, 22 Jun 2008 17:18:20 -0400 (EDT) Subject: [Pw_forum] Running examples in espresso-4.0 In-Reply-To: References: <485C7E79.6060706@iitk.ac.in> <003d01c8d42a$6287efd0$2797cf70$@net> Message-ID: On Sun, 22 Jun 2008, Axel Kohlmeyer wrote: > to the best of my knowledge, the syntax in check_failure.sh is > correct and the problem is most likely a "feature" in ubuntu. > > are you certain that /bin/sh is actually pointing to bash? > > unfortunately, there is no good solution. you could make /bin/sh a symlink to > /bin/bash, but that will probably break other scripts, e.g., those that > control your startup. Actually, I'm running Ubuntu 8.04 Hardy and linking /bin/sh to /bin/bash solves the problem and does not affect the startup scripts. HTH Davide From w2agz at pacbell.net Mon Jun 23 03:31:25 2008 From: w2agz at pacbell.net (Paul M. Grant) Date: Sun, 22 Jun 2008 18:31:25 -0700 Subject: [Pw_forum] Running examples in espresso-4.0 In-Reply-To: References: <485C7E79.6060706@iitk.ac.in> <003d01c8d42a$6287efd0$2797cf70$@net> Message-ID: <009501c8d4d0$da4992c0$8edcb840$@net> I'm in the middle of a two-day run, but will try Davide's solution when it finishes. It turns out Ubuntu is a clone of Debian, which is supposed to be "the" Linux distro, and they use (for some reason), a different command interpreter, dash (I wonder what the "d" is for ;)). Where's a good place to park the symlink so that it invokes on boot-up? On the subject of command interpreters, I spend a lot of time writing scripts to generate and analyze text I/O from the PWscf executables, and bash is tedious, to say the least. What's been the experience with "semi-smart" interpreters, like Perl? How I long for good old APL2, which even has an unconditional goto! (I actually have IBM's latest on my machine...maybe I should try it out) For anyone interested in superconductivity, I had an N&V in last week's Nature on the recently discovered "iron age" superconductors...barbs are welcome. Paul M. Grant, PhD Principal, W2AGZ Technologies Visiting Scholar, Applied Physics, Stanford (2005-2008) 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 Davide Ceresoli Sent: Sunday, June 22, 2008 2:18 PM To: Axel Kohlmeyer; PWSCF Forum Cc: Paul M. Grant Subject: Re: [Pw_forum] Running examples in espresso-4.0 On Sun, 22 Jun 2008, Axel Kohlmeyer wrote: > to the best of my knowledge, the syntax in check_failure.sh is > correct and the problem is most likely a "feature" in ubuntu. > > are you certain that /bin/sh is actually pointing to bash? > > unfortunately, there is no good solution. you could make /bin/sh a symlink to > /bin/bash, but that will probably break other scripts, e.g., those that > control your startup. Actually, I'm running Ubuntu 8.04 Hardy and linking /bin/sh to /bin/bash solves the problem and does not affect the startup scripts. HTH Davide _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum From lukuilin at gmail.com Mon Jun 23 04:41:18 2008 From: lukuilin at gmail.com (kuilin lu) Date: Mon, 23 Jun 2008 10:41:18 +0800 Subject: [Pw_forum] Pw_forum Digest, Vol 12, Issue 40 In-Reply-To: References: Message-ID: <4303e8ec0806221941h3d49674fhe4c6702ce318ed65@mail.gmail.com> ============= Dear Grant and all who has the same problem: Do you mean the "bug" is the following? please note the round bracket after "function check_failure" //This is the check_failure.sh file in veriosn 4.0 ------------------ function check_failure() { # usage: check_failure $? if test $1 != 0; then $ECHO "Error condition encountered during test: exit status = $1" $ECHO "Aborting" exit 1 fi } ------------------ // and I modified this to the following to make it running under AIX 5.2, note I deleted the round bracket after "function check_failure". ------------------ # function to test the exit status of a job function check_failure { # usage: check_failure $? if test $1 != 0; then $ECHO "Error condition encountered during test: exit status = $1" $ECHO "Aborting" exit 1 fi } ------------------ It seems to me that some *uix distribution didn't realize some function in sh script, so if you meet the problem, please try this "dirty" solution if you like. Hope it save someone's time ;-) Best Wishes Kuilin Lu > ------------------------------ > > Message: 2 > Date: Sat, 21 Jun 2008 22:39:47 -0700 > From: "Paul M. Grant" > Subject: Re: [Pw_forum] Running examples in espresso-4.0 > To: "'PWSCF Forum'" > Message-ID: <003d01c8d42a$6287efd0$2797cf70$@net> > Content-Type: text/plain; charset="iso-8859-1" > > BTW, it seems the "bug" in check_failure.sh still remains in espresso-4.0 > ...at least re the version of bash included in the ubuntu distros. > > Paul M. Grant, PhD > Principal, W2AGZ Technologies > Visiting Scholar, Applied Physics, Stanford (2005-2008) > EPRI Science Fellow (Retired) > IBM Research Staff Member Emeritus > w2agz at pacbell.net > http://www.w2agz.com > ? > ? > > ---------------------------------------------------------------------- Kuilin Lu, Ph.D student College of Materials Science and Engineering Hunan University Changsha 410082 China E-Mail: lukuilin at gmail.com ---------------------------------------------------------------------- From lukuilin at gmail.com Mon Jun 23 04:54:14 2008 From: lukuilin at gmail.com (kuilin lu) Date: Mon, 23 Jun 2008 10:54:14 +0800 Subject: [Pw_forum] Running examples in espresso-4.0 Message-ID: <4303e8ec0806221954o65410532ib1eeaccebae6042d@mail.gmail.com> Dear Grant and all who has the same problem: Do you mean the "bug" is the following? please note the round bracket after "function check_failure" //This is the check_failure.sh file in veriosn 4.0 ------------------ function check_failure() { # usage: check_failure $? if test $1 != 0; then $ECHO "Error condition encountered during test: exit status = $1" $ECHO "Aborting" exit 1 fi } ------------------ // and I modified this to the following to make it running under AIX 5.2, note I deleted the round bracket after "function check_failure". ------------------ # function to test the exit status of a job function check_failure { # usage: check_failure $? if test $1 != 0; then $ECHO "Error condition encountered during test: exit status = $1" $ECHO "Aborting" exit 1 fi } ------------------ It seems to me that some *uix distribution didn't realize some function in sh script, so if you meet the problem, please try this "dirty" solution if you like. Hope it save someone's time ;-) Best Wishes Kuilin Lu > ------------------------------ > > Message: 2 > Date: Sat, 21 Jun 2008 22:39:47 -0700 > From: "Paul M. Grant" > Subject: Re: [Pw_forum] Running examples in espresso-4.0 > To: "'PWSCF Forum'" > Message-ID: <003d01c8d42a$6287efd0$2797cf70$@net> > Content-Type: text/plain; charset="iso-8859-1" > > BTW, it seems the "bug" in check_failure.sh still remains in espresso-4.0 > ...at least re the version of bash included in the ubuntu distros. > > Paul M. Grant, PhD > Principal, W2AGZ Technologies > Visiting Scholar, Applied Physics, Stanford (2005-2008) > EPRI Science Fellow (Retired) > IBM Research Staff Member Emeritus > w2agz at pacbell.net > http://www.w2agz.com > ? > ? > > ---------------------------------------------------------------------- Kuilin Lu, Ph.D student College of Materials Science and Engineering Hunan University Changsha 410082 China E-Mail: lukuilin at gmail.com ---------------------------------------------------------------------- From giannozz at democritos.it Mon Jun 23 16:49:21 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Mon, 23 Jun 2008 16:49:21 +0200 Subject: [Pw_forum] Running examples in espresso-4.0 In-Reply-To: <009501c8d4d0$da4992c0$8edcb840$@net> References: <485C7E79.6060706@iitk.ac.in> <003d01c8d42a$6287efd0$2797cf70$@net> <009501c8d4d0$da4992c0$8edcb840$@net> Message-ID: On Jun 23, 2008, at 3:31 , Paul M. Grant wrote: > On the subject of command interpreters, I spend a lot of time writing > scripts to generate and analyze text I/O from the PWscf executables, > and bash is tedious, to say the least. it would be interesting to know what kind of information you want to extract from text, and if minor changes in output could make your (and other people's) task easier. 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 Mon Jun 23 21:14:09 2008 From: w2agz at pacbell.net (Paul M. Grant) Date: Mon, 23 Jun 2008 12:14:09 -0700 Subject: [Pw_forum] Running examples in espresso-4.0 In-Reply-To: References: <485C7E79.6060706@iitk.ac.in> <003d01c8d42a$6287efd0$2797cf70$@net> <009501c8d4d0$da4992c0$8edcb840$@net> Message-ID: <010201c8d565$51058ab0$f310a010$@net> Paolo, one "standardization" which would be nice would be to assign a "grep tag" to the final values of an scf run that are otherwise iterated under the same name during the course of its execution. For example, you do this the final value of "total energy" using an "!". (This nice little feature is undocumented it seems). With your help, I was able to go into electrons.f90 and do similar for "estimated scf accuracy" as it uses a separate FORMAT statement to print the final value, but not the "magnetizations" or "spin eigenvalues." I suppose a real "bash-er" could find something close by like "Fermi energy" and slog around to get the others. Paul M. Grant, PhD Principal, W2AGZ Technologies Visiting Scholar, Applied Physics, Stanford (2005-2008) 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: Monday, June 23, 2008 7:49 AM To: PWSCF Forum Subject: Re: [Pw_forum] Running examples in espresso-4.0 On Jun 23, 2008, at 3:31 , Paul M. Grant wrote: > On the subject of command interpreters, I spend a lot of time writing > scripts to generate and analyze text I/O from the PWscf executables, > and bash is tedious, to say the least. it would be interesting to know what kind of information you want to extract from text, and if minor changes in output could make your (and other people's) task easier. 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 akohlmey at cmm.chem.upenn.edu Mon Jun 23 21:25:47 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Mon, 23 Jun 2008 15:25:47 -0400 (EDT) Subject: [Pw_forum] Running examples in espresso-4.0 In-Reply-To: <009501c8d4d0$da4992c0$8edcb840$@net> References: <485C7E79.6060706@iitk.ac.in> <003d01c8d42a$6287efd0$2797cf70$@net> <009501c8d4d0$da4992c0$8edcb840$@net> Message-ID: On Sun, 22 Jun 2008, Paul M. Grant wrote: PG> I'm in the middle of a two-day run, but will try Davide's solution when it PG> finishes. PG> PG> It turns out Ubuntu is a clone of Debian, which is supposed to be "the" PG> Linux distro, and they use (for some reason), a different command PG> interpreter, dash (I wonder what the "d" is for ;)). PG> Where's a good place to park the symlink so that it invokes on boot-up? paul, if my understanding is correct, you only need to do it once, or unless you update "dash" (well maybe not even then). i just checked it out (on wikipedia) and it turns out to be a renamed version of "ash". redhat used "ash" for a while as /bin/sh. the reason being that it is a much smaller executable and runs much faster than /bin/bash (even in reduced posix mode). there also is a (lengthy) explanation on http://wiki.ubuntu.com/DashAsBinSh since i was using ash previously a lot to detect "bash-isms" in shell scripts and to keep them portable, i trust that it actually _is_ posix compliant. i checked out the posix specs and indeed "function" is a "bash-ism" (i've obviously been using bash-only systems for too long to forget about this). i compared with some of my older scripts that passed the "/bin/ash" acid test and i would like to suggest the attached change. this should work on any posix compliant shell. PG> On the subject of command interpreters, I spend a lot of time writing PG> scripts to generate and analyze text I/O from the PWscf executables, and PG> bash is tedious, to say the least. What's been the experience with PG> "semi-smart" interpreters, like Perl? How I long for good old APL2, which perl is great for sifting through tons of text, particularly if you like using regular expressions. i'm basically writing anything that goes beyond simple shell scripts and that does not warrant writing a compiled executable in perl or - gasp - tcl. if you can give me a rough description of what you are looking for and how you'd like the output to be formatted, i may be able to whip up some templates that you can adapt to your needs. if you can find a copy of the old "camel book" (o'reilly book on perl describing perl 4) you have a very good, compact and quite entertaining introduction to the most important stuff. the newer edition covering perl 5 is IMO too bloated and the syntax is basically the same. a lot of people like to write scripts in python these days, but i find that i can do the equivalent stuff in perl with half the lines of code (and every line of code you don't have to type is automatically correct... ;-) ). PG> even has an unconditional goto! (I actually have IBM's latest on my PG> machine...maybe I should try it out) cheers, axel. PG> For anyone interested in superconductivity, I had an N&V in last week's PG> Nature on the recently discovered "iron age" superconductors...barbs are PG> welcome. PG> PG> Paul M. Grant, PhD PG> Principal, W2AGZ Technologies PG> Visiting Scholar, Applied Physics, Stanford (2005-2008) 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> -----Original Message----- PG> From: pw_forum-bounces at pwscf.org [mailto:pw_forum-bounces at pwscf.org] On PG> Behalf Of Davide Ceresoli PG> Sent: Sunday, June 22, 2008 2:18 PM PG> To: Axel Kohlmeyer; PWSCF Forum PG> Cc: Paul M. Grant PG> Subject: Re: [Pw_forum] Running examples in espresso-4.0 PG> PG> On Sun, 22 Jun 2008, Axel Kohlmeyer wrote: PG> PG> > to the best of my knowledge, the syntax in check_failure.sh is PG> > correct and the problem is most likely a "feature" in ubuntu. PG> > PG> > are you certain that /bin/sh is actually pointing to bash? PG> > PG> > unfortunately, there is no good solution. you could make /bin/sh a symlink PG> to PG> > /bin/bash, but that will probably break other scripts, e.g., those that PG> > control your startup. PG> Actually, I'm running Ubuntu 8.04 Hardy and linking /bin/sh to PG> /bin/bash solves the problem and does not affect the startup PG> scripts. PG> PG> HTH PG> PG> Davide PG> PG> _______________________________________________ PG> Pw_forum mailing list PG> Pw_forum at pwscf.org PG> http://www.democritos.it/mailman/listinfo/pw_forum PG> PG> _______________________________________________ PG> Pw_forum mailing list PG> Pw_forum at pwscf.org PG> http://www.democritos.it/mailman/listinfo/pw_forum 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. -------------- next part -------------- A non-text attachment was scrubbed... Name: qe-posix-sh.diff.gz Type: application/x-gzip Size: 1058 bytes Desc: Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080623/9511f7ac/attachment.bin From padmaja_patnaik at yahoo.co.uk Mon Jun 23 21:29:28 2008 From: padmaja_patnaik at yahoo.co.uk (Padmaja Patnaik) Date: Mon, 23 Jun 2008 19:29:28 +0000 (GMT) Subject: [Pw_forum] Problem in running jobs Message-ID: <359444.86490.qm@web25404.mail.ukl.yahoo.com> Dear PWscf-users and -authors I am facing a problem while trying to run the examples of pwscf. the error is as follows, Error: Need to obtain the job magic number in MXMPI_MAGIC ! [0] Error: write to socket failed ! Error: Need to obtain the job magic number in MXMPI_MAGIC ! [0] Error: write to socket failed ! Error: Need to obtain the job magic number in MXMPI_MAGIC ! [0] Error: write to socket failed ! Error: Need to obtain the job magic number in MXMPI_MAGIC ! [0] Error: write to socket failed ! Error: Need to obtain the job magic number in MXMPI_MAGIC ! [0] Error: write to socket failed ! Error: Need to obtain the job magic number in MXMPI_MAGIC ! [0] Error: write to socket failed ! Error: Need to obtain the job magic number in MXMPI_MAGIC ! [0] Error: write to socket failed ! Error: Need to obtain the job magic number in MXMPI_MAGIC ! [0] Error: write to socket failed ! Error: Need to obtain the job magic number in MXMPI_MAGIC ! [0] Error: write to socket failed ! Error: Need to obtain the job magic number in MXMPI_MAGIC ! [0] Error: write to socket failed ! tried to see the users guide but could not solve the problem. Need help Regards Padmaja __________________________________________________________ Sent from Yahoo! Mail. A Smarter Email http://uk.docs.yahoo.com/nowyoucan.html -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080623/075cf0b4/attachment.htm From akohlmey at cmm.chem.upenn.edu Mon Jun 23 21:29:59 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Mon, 23 Jun 2008 15:29:59 -0400 (EDT) Subject: [Pw_forum] Problem in running jobs In-Reply-To: <359444.86490.qm@web25404.mail.ukl.yahoo.com> References: <359444.86490.qm@web25404.mail.ukl.yahoo.com> Message-ID: On Mon, 23 Jun 2008, Padmaja Patnaik wrote: PP> Dear PWscf-users and -authors PP> dear padmaja, PP> I am facing a problem while trying to run the examples of pwscf. the PP> error is as follows, [...] PP> PP> PP> Error: Need to obtain the job magic number in MXMPI_MAGIC ! PP> PP> PP> [0] Error: write to socket failed ! you cannot run a binary that has been compiled with myrinet support on a node that does not have a working and configured myrinet card. PP> tried to see the users guide but could not solve the problem. this is not an error from QE but from your (myrinet) MPI library. before reporting MPI related problems, please first make sure you can actually run MPI based executables _at all_. cheers, axel. PP> PP> PP> Need help PP> PP> PP> PP> PP> PP> Regards PP> PP> PP> Padmaja PP> PP> __________________________________________________________ Sent from Yahoo! Mail. A Smarter Email http://uk.docs.yahoo.com/nowyoucan.html -- ======================================================================= 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 manoj at phys.ufl.edu Mon Jun 23 22:19:14 2008 From: manoj at phys.ufl.edu (Manoj Srivastava) Date: Mon, 23 Jun 2008 16:19:14 -0400 (EDT) Subject: [Pw_forum] complex band of Cu(111) Message-ID: Dear PWSCF users, I am a new user of PWSCF and I have encountered a problem in calculating complex band of Cu(111). I see a band gap and there are no bands crossing Fermi energy. which is obviously wrong :( I have used hexagonal unit cells containing 3 atoms. Cu.pz-d-rrkjus.UPF pseudo potential was used in the calculation. It would be great if somebody can suggest me something. Following is the input file I used - &system ibrav =4, celldm(1) =4.78076379, celldm(3)=2.44948974, nat= 3, ntyp= 1, nelec=33.0, nbnd=24, ecutwfc =25.0, occupations= 'smearing', smearing='gaussian', degauss=0.02 / &electrons diagonalization='$diago' conv_thr = 1.0e-8 mixing_beta=0.7 / ATOMIC_SPECIES Cu 63.55 Cu.pz-d-rrkjus.UPF ATOMIC_POSITIONS {bohr} Cu 0.0 0.0 0.0 Cu 2.390381895 1.380087631 3.903477289 Cu 4.780763790 2.760175261 7.806954578 K_POINTS (automatic) 8 8 8 0 0 0 EOF # complex bands of cu(111) along the 001 direction K_perp=0 cat > cu.cond.in << EOF &inputcond outdir='$TMP_DIR/' prefixl='cu' band_file ='bands.cu' ikind=0 energy0=10.d0 denergy=-0.2d0 ewind=3.d0 epsproj=1.d-7 / 1 -1.0 1.732 -0.8165 100 EOF Regards, Manoj From vegalew at hotmail.com Tue Jun 24 11:12:12 2008 From: vegalew at hotmail.com (lewvega) Date: Tue, 24 Jun 2008 17:12:12 +0800 Subject: [Pw_forum] something wrong with vc-relax & relax on my clusters Message-ID: Dear Users I'm a new user of pwscf, and I have some experience of castep, which seems to be similar with pwscf. First I tested paralell vision of pwscf with anatase-TiO2 lattice cell optimization calculation. the input file as follows: &CONTROL title = 'Anatase lattice BFGS' , calculation = 'vc-relax' , restart_mode = 'from_scratch' , outdir = '/home/vega/tmp/' , pseudo_dir = '/home/vega/espresso-4.0/pseudo/' , prefix = 'Anatase lattice default' , etot_conv_thr = 0.000000735 , forc_conv_thr = 0.0011668141375 , nstep = 1000 , / &SYSTEM ibrav = 6, celldm(1) = 7.135605333, celldm(3) = 2.5121822033898305084745762711864, nat = 12, ntyp = 2, ecutwfc = 25 , ecutrho = 200 , / &ELECTRONS conv_thr = 7.3D-8 , / &IONS / &CELL / ATOMIC_SPECIES Ti 47.86700 Ti.pw91-sp-van_ak.UPF O 15.99940 O.pw91-van_ak.UPF ATOMIC_POSITIONS angstrom Ti 0.000000000 0.000000000 0.000000000 Ti 1.888000000 1.888000000 4.743000000 Ti 0.000000000 1.888000000 2.372000000 Ti 1.888000000 0.000000000 7.115000000 O 0.000000000 0.000000000 1.973000000 O 1.888000000 1.888000000 6.716000000 O 0.000000000 1.888000000 4.345000000 O 1.888000000 0.000000000 9.088000000 O 1.888000000 0.000000000 5.141000000 O 0.000000000 1.888000000 0.398000000 O 1.888000000 1.888000000 2.770000000 O 0.000000000 0.000000000 7.513000000 K_POINTS automatic 7 7 3 1 1 1 then I checked the initial structure in Xcrysden, and type paralell command like this: mpiexec -n 20 pw.x < inputfile > outputfile the calculation ran immediately. After calculation finished, the last CELL_PARAMETERS section in outputfile is CELL_PARAMETERS (alat) 1.000000000 0.000000000 0.000000000 0.000000000 1.000000000 0.000000000 0.000000000 0.000000000 2.512182203 this indicate the 'vc-relax' didn't take effect, no cell parameters was not optimized during the vc-relax strategy. And the atom positons also changed a lot, the ATOMIC_POSITIONS sections like this: ATOMIC_POSITIONS (angstrom) Ti 0.000000000 0.000000000 0.000108235 Ti 1.888000000 1.888000000 4.743111514 Ti 0.000000000 1.888000000 2.371065034 Ti 1.888000000 0.000000000 7.115503582 O 0.000000000 0.000000000 1.981862393 O 1.888000000 1.888000000 6.726003352 O 0.000000000 1.888000000 4.353977562 O 1.888000000 0.000000000 9.096778202 O 1.888000000 0.000000000 5.132185229 O 0.000000000 1.888000000 0.389375062 O 1.888000000 1.888000000 2.760157286 O 0.000000000 0.000000000 7.504294082 Both x,y remain as the initial value. All of this indicate my above calculation is a 'relax' calculation not a 'vc-relax' strategy. Also if I set the ionic and cell optimization both using BFGS schedule as the castep did, the cell parameters changed so much that an error occured said 'non orthogonal operation'. Does anyone encounter the same problem? Second, I uesed the optimized anatase cell via pwscf's default vc-relax damped MD schedule to cleave the 101 surface and make the further calculation. But another serious problem happened. the input file as follows: &CONTROL title = 'Anatase101_2x' , calculation = 'relax' , restart_mode = 'from_scratch' , outdir = '/home/vega/tmp/' , pseudo_dir = '/home/vega/espresso-4.0/pseudo/' , prefix = 'Anatase101_2x' , etot_conv_thr = 0.000000735 , forc_conv_thr = 0.0011668141375 , nstep = 1000 , / &SYSTEM ibrav = 8, celldm(1) = 19.293941717, celldm(2) = 1.109509628630600744472407458828, celldm(3) = 1.7484889932972017886534804648227, nat = 72, ntyp = 2, ecutwfc = 30 , ecutrho = 240 , / &ELECTRONS conv_thr = 7.3D-8 , / &IONS ion_dynamics = 'bfgs' , / ATOMIC_SPECIES Ti 47.86700 Ti.pw91-sp-van_ak.UPF O 15.99940 O.pw91-van_ak.UPF ATOMIC_POSITIONS angstrom Ti 2.649000000 -1.888000000 0.733000000 Ti -1.060000000 0.000000000 4.242000000 Ti 0.446000000 0.000000000 1.610000000 Ti -3.264000000 -1.888000000 5.119000000 O 0.807000000 -1.888000000 1.466000000 O -2.902000000 0.000000000 4.975000000 O -1.397000000 0.000000000 2.344000000 O -5.105000000 -1.888000000 5.852000000 O -1.421000000 -1.888000000 4.386000000 O 2.287000000 0.000000000 0.877000000 O 0.783000000 0.000000000 3.508000000 O -4.323000000 -1.888000000 3.509000000 Ti 2.649000000 5.664000000 0.733000000 Ti -1.060000000 7.552000000 4.242000000 Ti 0.446000000 7.552000000 1.610000000 Ti -3.264000000 5.664000000 5.119000000 O 0.807000000 5.664000000 1.466000000 O -2.902000000 7.552000000 4.975000000 O -1.397000000 7.552000000 2.344000000 O -5.105000000 5.664000000 5.852000000 O -1.421000000 5.664000000 4.386000000 O 2.287000000 7.552000000 0.877000000 O 0.783000000 7.552000000 3.508000000 O -4.323000000 5.664000000 3.509000000 Ti 2.649000000 1.888000000 0.733000000 Ti -1.060000000 3.776000000 4.242000000 Ti 0.446000000 3.776000000 1.610000000 Ti -3.264000000 1.888000000 5.119000000 O 0.807000000 1.888000000 1.466000000 O -2.902000000 3.776000000 4.975000000 O -1.397000000 3.776000000 2.344000000 O -5.105000000 1.888000000 5.852000000 O -1.421000000 1.888000000 4.386000000 O 2.287000000 3.776000000 0.877000000 O 0.783000000 3.776000000 3.508000000 O -4.323000000 1.888000000 3.509000000 Ti 4.045000000 -1.888000000 4.242000000 Ti 1.842000000 0.000000000 5.119000000 O 2.204000000 -1.888000000 4.975000000 O 0.000000000 0.000000000 5.852000000 O 3.684000000 0.000000000 4.386000000 Ti 4.045000000 5.664000000 4.242000000 Ti 1.842000000 7.552000000 5.119000000 O 2.204000000 5.664000000 4.975000000 O 0.000000000 7.552000000 5.852000000 O 3.684000000 7.552000000 4.386000000 Ti 4.045000000 1.888000000 4.242000000 Ti 1.842000000 3.776000000 5.119000000 O 2.204000000 1.888000000 4.975000000 O 0.000000000 3.776000000 5.852000000 O 3.684000000 3.776000000 4.386000000 Ti 7.754000000 3.776000000 0.733000000 Ti 5.549000000 1.888000000 1.611000000 O 5.911000000 3.776000000 1.467000000 O 3.708000000 1.888000000 2.344000000 O 7.392000000 1.888000000 0.877000000 O 9.596000000 3.776000000 -0.000000000 O 4.490000000 1.888000000 0.000000000 Ti 7.754000000 7.552000000 0.733000000 Ti 5.549000000 5.664000000 1.611000000 O 5.911000000 7.552000000 1.467000000 O 3.708000000 5.664000000 2.344000000 O 7.392000000 5.664000000 0.877000000 O 9.596000000 7.552000000 -0.000000000 O 4.490000000 5.664000000 0.000000000 Ti 7.754000000 0.000000000 0.733000000 Ti 5.549000000 -1.888000000 1.611000000 O 5.911000000 0.000000000 1.467000000 O 3.708000000 -1.888000000 2.344000000 O 7.392000000 -1.888000000 0.877000000 O 9.596000000 0.000000000 -0.000000000 O 4.490000000 -1.888000000 0.000000000 K_POINTS gamma after checking the initial structure via Xcrysden, I typed 'mpiexec -n 20 pw.x < inputfile > outputfile'. But unfortunately nothing happened, even the first few lines of pwscf was not printed, and all the processes of pw.x triggered by mpich2 are sleeping. No calculation is running at all. Then I run this task in serial version of pwscf by typing 'pw.x < inputfile > outputfile', the calculation ran immediately. then I tried again and again, the paralell calculation of the input file always failed with no message out. And to make sure whether the environment is suitable for paralell calculation, I ran the first input file again in 20 paralell process. Everything seems to be ok with the paralell environment. I had tested the same task(same number of atoms, same atom, same convergence threshold et al.) in CASTEP successfully, so the error is not because of hardware. Is there any one can try my input file, and give me some suggestions. thank you so much for reading. by the way, I also will tell you something about my hardware and softeware: hardware: five computers with intel coreII Q6600 CPU (quadcore) and 40G menmory (8G memory per computer) on intel S3000AH system board. Computers linked with 1Gbit/s network by only one Swich( TP-Link ) software: Readhat linux enterprise 4 Advanced Server Update 4, mpich2, intel fortran compiler 10.1.008 (64 bit-edition), intel c++ 10.1.008 (64 bit-edition), intel MKL 10.0.011 (64 bit-edition), fftw2.1. And I the fftw in MKL can't be recognized by 'configure' of espresso, even the lib dictionary is included in LD_LIBRARY_PATH. thank you all again for reading and I'm looking forward to your suggestions. Vega Lew PH.D Candidate in Chemical Engineering College of Chemistry and Chemical Engineering Nanjing University of Technology, 210009, Nanjing, Jiangsu, China _________________________________________________________________ ?????????live mail???????? http://get.live.cn/product/mail.html -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080624/51ea2d1e/attachment-0001.htm From paulatto at sissa.it Tue Jun 24 12:15:11 2008 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Tue, 24 Jun 2008 12:15:11 +0200 (CEST) Subject: [Pw_forum] something wrong with vc-relax & relax on my clusters In-Reply-To: References: Message-ID: <28807.147.122.5.216.1214302511.squirrel@webmail.sissa.it> On Mar, Giugno 24, 2008 11:12, lewvega wrote: > &CELL > / > [...] > this indicate the 'vc-relax' didn't take effect, no cell parameters was > not optimized during the vc-relax strategy. Dear Vega Lew, default cell dynamic is "none", which means that cell won't change unless you specify it. bye -- 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 baroni at sissa.it Tue Jun 24 12:27:53 2008 From: baroni at sissa.it (Stefano Baroni) Date: Tue, 24 Jun 2008 12:27:53 +0200 Subject: [Pw_forum] something wrong with vc-relax & relax on my clusters In-Reply-To: References: Message-ID: <7E4A897E-D918-4D12-AF42-8B0F29899F08@sissa.it> On Jun 24, 2008, at 11:12 AM, lewvega wrote: > Dear Users > > I'm a new user of pwscf, and I have some experience of castep, which > seems to be similar with pwscf. please, watch your tongue: we are all very kind to each other in this forum ;-) S. --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget 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/20080624/2acc4afe/attachment.htm From akohlmey at cmm.chem.upenn.edu Tue Jun 24 14:54:50 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Tue, 24 Jun 2008 08:54:50 -0400 (EDT) Subject: [Pw_forum] something wrong with vc-relax & relax on my clusters In-Reply-To: References: Message-ID: On Tue, 24 Jun 2008, lewvega wrote: VL> Dear Users dear vega lew, thanks for your detailed e-mail, but before making any claims that something is broken, you should fix the obviously broken stuff. in this case it is your compilers. the intel 10.1.008 compilers are _horribly_ broken and miscompile almost any scientific code that i came across. you have to update to at least 10.1.015 and try again. btw: don't worry about MKL's FFT not being detected. there is no directly supported anyways and both the fftw3 and the fftw2 wrappers don't work reliably. in all my tests it turned out to be slower than the bundled FFTW, except for power of two grids (which confirms results with BenchFFT). more importantly, you have to make certain that the OMP_NUM_THREADS environment variable is set to 1 or else it will multi-thread across all local cpu cores, regardless of whether you parallelize with MPI (see the MKL documentation for details). [...] VL> celldm(1) = 19.293941717, VL> celldm(2) = 1.109509628630600744472407458828, VL> celldm(3) = 1.7484889932972017886534804648227, another remark: please read up double precision floating point arithmetic (e.g. on wikipedia) and realize that specifying more than 14 significant digits is a completely pointless execise... (not to mention that the errors from DFT, using grids and an incomplete basis set by far exceed the possible numerical accuracy. cheers, axel. -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From giannozz at democritos.it Tue Jun 24 15:26:01 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Tue, 24 Jun 2008 15:26:01 +0200 Subject: [Pw_forum] something wrong with vc-relax & relax on my clusters In-Reply-To: <28807.147.122.5.216.1214302511.squirrel@webmail.sissa.it> References: <28807.147.122.5.216.1214302511.squirrel@webmail.sissa.it> Message-ID: <4860F5E9.4010709@democritos.it> Lorenzo Paulatto wrote: > On Mar, Giugno 24, 2008 11:12, lewvega wrote: >> &CELL >> / >> [...] >> this indicate the 'vc-relax' didn't take effect, no cell parameters was >> not optimized during the vc-relax strategy. > > Dear Vega Lew, > default cell dynamic is "none", which means that cell won't change unless > you specify it. in addition, one has to choose a suitable time step: see http://www.quantum-espresso.org/wiki/index.php/Frequently_Asked_Questions item 1.2.5 > I typed 'mpiexec -n 20 pw.x < inputfile > outputfile'. > But unfortunately nothing happened the usual problem with misconfigured mpi libraries and redirection? http://www.quantum-espresso.org/wiki/index.php/Running_on_parallel_machines#Trouble_with_input_files Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From bsnebiha at yahoo.fr Tue Jun 24 17:33:33 2008 From: bsnebiha at yahoo.fr (nebiha ben sedrine) Date: Tue, 24 Jun 2008 08:33:33 -0700 (PDT) Subject: [Pw_forum] Run ni.scf.in Example Message-ID: <959916.40548.qm@web52901.mail.re2.yahoo.com> Dear Pwscf users, I'm just beginning with PWscf, and after all the installation process with libraairies and related softs, I am tryig to run as example "ni.scf.in". I have copied pw.x into the example directory, but no succes. Here you can find the error messages after running in both PWgui (1) and from a command line (2). 1) couldn't execute "pw.x": no such file or directory couldn't execute "pw.x": no such file or directory while executing "open [concat | $args] r" ("exec" arm line 18) invoked from within "switch -exact -- $mode { open { set count [incr nonblocking(counter)] set nonblocking(status,$count) 1 return $count } exec { ..." (procedure "::tclu::nonblocking" line 4) invoked from within "::tclu::nonblocking exec 12 pw.x < /home/tgharbi/espresso-4.0/PWgui-4.0/examples/pw/ni.scf.in 2> /home/tgharbi/espresso-4.0/PWgui-4.0/examples/pw/ni.s..." ("eval" body line 1) invoked from within "eval ::tclu::nonblocking exec $execID $runCmd < $run(inpFile,$moduleObj) 2> $run(errFile,$moduleObj)" (procedure "::pwscf::run::run" line 89) 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 50) invoked from within "tk::MenuInvoke .gui0.shellchildsite.menubar.menubar.run.menu 1" (command bound to event) 2) bash: pw.x: command not found thank you in advance for your reply Best Regards N.B.Sedrine _____________________________________________________________________________ Envoyez avec Yahoo! Mail. Une boite mail plus intelligente http://mail.yahoo.fr -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080624/2f0afa5a/attachment.htm From bsnebiha at yahoo.fr Tue Jun 24 17:28:15 2008 From: bsnebiha at yahoo.fr (nebiha ben sedrine) Date: Tue, 24 Jun 2008 08:28:15 -0700 (PDT) Subject: [Pw_forum] Run ni.scf.in Example Message-ID: <11934.26321.qm@web52905.mail.re2.yahoo.com> Dear Pwscf users, I'm just beginning with PWscf, and after all the installation process with libraairies and related softs, I am tryig to run as example "ni.scf.in". I have copied pw.x into the example directory, but no succes. Here you can find the error messages after running in both PWgui (1) and from a command line (2). 1) couldn't execute "pw.x": no such file or directory couldn't execute "pw.x": no such file or directory while executing "open [concat | $args] r" ("exec" arm line 18) invoked from within "switch -exact -- $mode { open { set count [incr nonblocking(counter)] set nonblocking(status,$count) 1 return $count } exec { ..." (procedure "::tclu::nonblocking" line 4) invoked from within "::tclu::nonblocking exec 12 pw.x < /home/tgharbi/espresso-4.0/PWgui-4.0/examples/pw/ni.scf.in 2> /home/tgharbi/espresso-4.0/PWgui-4.0/examples/pw/ni.s..." ("eval" body line 1) invoked from within "eval ::tclu::nonblocking exec $execID $runCmd < $run(inpFile,$moduleObj) 2> $run(errFile,$moduleObj)" (procedure "::pwscf::run::run" line 89) 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 50) invoked from within "tk::MenuInvoke .gui0.shellchildsite.menubar.menubar.run.menu 1" (command bound to event) 2) bash: pw.x: command not found thank you in advance for your reply Best Regards N.B.Sedrine _____________________________________________________________________________ Envoyez avec Yahoo! Mail. Une boite mail plus intelligente http://mail.yahoo.fr -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080624/51d8372c/attachment-0001.htm From vegalew at hotmail.com Tue Jun 24 18:28:26 2008 From: vegalew at hotmail.com (vega lew) Date: Wed, 25 Jun 2008 00:28:26 +0800 Subject: [Pw_forum] something wrong with vc-relax & relax on my clusters Message-ID: Dear Users Thank your for Lorenzo Paulatto, Paolo Giannozz, Stefano Baroni and Axel Kohlmeyer's kindly responding. According Lorenzo Paulatto's suggestion, I typed mpiexec -n 20 pw.x -in inputfile > outputfile, the job runing successfully. I'm so happy ^_^ Thanks to Axel Kohlmeyer's responding, I learned a lot from it. I't very kind of you to reply me in detail. I'll update my intel compilers to 10.1.015 or later, and recompile my espresso. Thanks to Lorenzo Paulatto's responding, I note that the default cell dynamic is "none". Thank your all again for responding so soon. cheers, Vega Lew PH.D Candidate in Chemical Engineering College of Chemistry and Chemical Engineering Nanjing University of Technology, 210009, Nanjing, Jiangsu, China _________________________________________________________________ News, entertainment and everything you care about at Live.com. Get it now! http://www.live.com/getstarted.aspx -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080625/0e57217b/attachment.htm From letiziachiodo at gmail.com Tue Jun 24 18:44:06 2008 From: letiziachiodo at gmail.com (Letizia Chiodo) Date: Tue, 24 Jun 2008 18:44:06 +0200 Subject: [Pw_forum] complex band of Cu(111) In-Reply-To: References: Message-ID: <5edd2b2b0806240944q265b0c78r605c0e46474a6744@mail.gmail.com> Dear Manoj at a first sight, I can see some errors in your input: - if you want to model a surface, in the supercell approach, you need to create a slab + vacuum geometry, I'm not sure you are doing this. - in general, 3 layers is a very bad choice for describing (semiconductor or metal) surfaces, you will be very far from convergence for structural, electronic, etc properties. - ecutwfc sounds to me too low for Cu, and for ultrasoft pseudopotentials, you are missing the cutoff on rho. - and probably I'm forgetting other things, e.g. mixing_beta=0.7 maybe is too large. Hope it helps Letizia NNL of CNR-INFM, Lecce, Italy, UPV-EHU, San Sebastian, Spain On Mon, Jun 23, 2008 at 10:19 PM, Manoj Srivastava wrote: > Dear PWSCF users, > I am a new user of PWSCF and I have encountered a problem in calculating > complex band of Cu(111). I see a band gap and there are no bands crossing > Fermi energy. which is obviously wrong :( > I have used hexagonal unit cells containing 3 atoms. Cu.pz-d-rrkjus.UPF > pseudo potential was used in the calculation. It would be great if > somebody can suggest me something. > Following is the input file I used - > > &system > ibrav =4, > celldm(1) =4.78076379, celldm(3)=2.44948974, > nat= 3, > ntyp= 1, > nelec=33.0, > nbnd=24, > ecutwfc =25.0, > occupations= 'smearing', > smearing='gaussian', > degauss=0.02 > > / > &electrons > diagonalization='$diago' > conv_thr = 1.0e-8 > mixing_beta=0.7 > / > ATOMIC_SPECIES > Cu 63.55 Cu.pz-d-rrkjus.UPF > ATOMIC_POSITIONS {bohr} > Cu 0.0 0.0 0.0 > Cu 2.390381895 1.380087631 3.903477289 > Cu 4.780763790 2.760175261 7.806954578 > K_POINTS (automatic) > 8 8 8 0 0 0 > EOF > # complex bands of cu(111) along the 001 direction K_perp=0 > cat > cu.cond.in << EOF > &inputcond > outdir='$TMP_DIR/' > prefixl='cu' > band_file ='bands.cu' > ikind=0 > energy0=10.d0 > denergy=-0.2d0 > ewind=3.d0 > epsproj=1.d-7 > / > 1 > -1.0 1.732 -0.8165 > 100 > EOF > > > Regards, > Manoj > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From baroni at sissa.it Tue Jun 24 19:28:05 2008 From: baroni at sissa.it (Stefano Baroni) Date: Tue, 24 Jun 2008 19:28:05 +0200 Subject: [Pw_forum] Run ni.scf.in Example In-Reply-To: <11934.26321.qm@web52905.mail.re2.yahoo.com> References: <11934.26321.qm@web52905.mail.re2.yahoo.com> Message-ID: it seems that your current directory is not in your path this has nothing to do with Quantum ESPRESSO may I suggest that you seek some advice from a colleague proficient in unix? good luck! SB On Jun 24, 2008, at 5:28 PM, nebiha ben sedrine wrote: > Dear Pwscf users, > I'm just beginning with PWscf, and after all the installation > process with libraairies and related softs, > I am tryig to run as example "ni.scf.in". I have copied pw.x into > the example directory, but no succes. > > Here you can find the error messages after running in both PWgui (1) > and from a command line (2). > 1) couldn't execute "pw.x": no such file or directory > couldn't execute "pw.x": no such file or directory > while executing > "open [concat | $args] r" > ("exec" arm line 18) > invoked from within > "switch -exact -- $mode { > open { > set count [incr nonblocking(counter)] > set nonblocking(status,$count) 1 > return $count > } > > exec { > ..." > (procedure "::tclu::nonblocking" line 4) > invoked from within > "::tclu::nonblocking exec 12 pw.x < /home/tgharbi/espresso-4.0/ > PWgui-4.0/examples/pw/ni.scf.in 2> /home/tgharbi/espresso-4.0/ > PWgui-4.0/examples/pw/ni.s..." > ("eval" body line 1) > invoked from within > "eval ::tclu::nonblocking exec $execID $runCmd < $run(inpFile, > $moduleObj) 2> $run(errFile,$moduleObj)" > (procedure "::pwscf::run::run" line 89) > 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 50) > invoked from within > "tk::MenuInvoke .gui0.shellchildsite.menubar.menubar.run.menu 1" > (command bound to event) > > 2) bash: pw.x: command not found > > thank you in advance for your reply > Best Regards > N.B.Sedrine > > > Envoy? avec Yahoo! Mail. > Une boite mail plus intelligente. > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget 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/20080624/6bf413c9/attachment.htm From jibiaoli at gmail.com Wed Jun 25 02:31:48 2008 From: jibiaoli at gmail.com (Clark Lee) Date: Wed, 25 Jun 2008 08:31:48 +0800 Subject: [Pw_forum] Methods of analyzing chemical bonds Message-ID: Hi all, Recently I learned methods which are now widely used for analyzing the chemical bonds (Chem.Rev, 2000, 100, 717). Partitioning schemes based on the electronic charge distribution include the natural bondorbital (NBO), thecharge distribution analysis (CDA), the atoms-in-molecules (AIM) method. There are also methods which define the partitioning of the energy of a chemical bond in different contributions: the energy decomposition analysis (EDA), the extended transition state (ETS) and the constrained space orbital variation (CSOV) method. The basic ideas of the EDA, ETS and CSOV methods are the following. The total bond energy ?E of a bond A-B is partitioned into four components which are calculated separately in four consecutive steps: ?E=?Eprep + ?Eels + ?EPauli + ?Eorb. Like the EDA and ETS methods, the CSOV partitioning scheme uses the frozen orbitals of the fragments as the starting point for the energy decomposition by superimposing them fixed at their separated unit character. Starting from the superposition of the fragments, the wave function of the complex is optimized in a series of steps. The energy changes which are associated with changing the size of the variational space indicate the importance of the different orbital interactions for the bond energy. These methods provide insights into the chemical bondings; in contrast, method like charge density difference is not so effective as these methods. Here goes my question: has any of these methods been implemented in the QE package? How can I carry out CSOV method in PWscf to study a bonding of an adsobate-substrate interaction. Another question: I noticed that electron localizaiton function analysis was not implemented in previous versions of QE package. I don't know whether it is included or not. -- Yours sincerely, Jibiao Li State Key Lab of Corrosion and Protection (SKLCP) Institute of Metal Research (IMR) Chinese Academy of Sciences (CAS), China Phone: 024-23904856 Email: jibiaoli at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080625/35f40ac4/attachment-0001.htm From meheut at impmc.jussieu.fr Wed Jun 25 03:16:47 2008 From: meheut at impmc.jussieu.fr (merlin meheut) Date: Wed, 25 Jun 2008 03:16:47 +0200 Subject: [Pw_forum] Generating VBC pseudo of Mg with ld1? Message-ID: <20080625031647.qig6b93ew4occo84@www-ext.impmc.jussieu.fr> Dear PWSCF users, I would like to use the best-possible Mg pseudopotential, and had been a bit disappointed by those I have generated so far. Looking at the litterature, I realized that a pseudopotential generated using Von-Barth Car scheme gave pretty good results in LDA (actually, this is the Mg.pz-n-vbc.UPF pseudo on PWscf webpage). I would like to try a similar approach with PBE functional. I don't know if it is possible to use this scheme with ld1.x. If it is I did not find how. Apparently, on O-sesame webpage it is mentioned that this possibility is "temporarily removed". Is there a way to get old versions of ld1 code that do it? Or is there a more recent scheme doing things similarly? Thank you in advance for any hints, Regards, Merlin Merlin Meheut 24 rue de la division Leclerc 92140 CLAMART 06 82 17 42 55 01 44 27 75 03 ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program. From smogunov at sissa.it Wed Jun 25 08:38:22 2008 From: smogunov at sissa.it (Alexander) Date: Wed, 25 Jun 2008 08:38:22 +0200 Subject: [Pw_forum] complex band of Cu(111) In-Reply-To: References: Message-ID: <200806250838.22413.smogunov@sissa.it> Dear Manoj On my PC with espresso-4.0 I do get one band around the Fermi energy as it should be. My output for energies around Ef looks like this: ------------------------------ k( 1) = ( -1.0000000 1.7320000), wk = 1.0000000 ngper, shell number = 39 24 ngper, n2d = 39 39 Nchannels of the left tip = 1 k1(2pi/a) k2(2pi/a) E-Ef (eV) 0.0891385 0.0000000 0.4000000 Nchannels of the left tip = 1 k1(2pi/a) k2(2pi/a) E-Ef (eV) 0.0568822 0.0000000 0.2000000 Nchannels of the left tip = 1 k1(2pi/a) k2(2pi/a) E-Ef (eV) 0.0228076 0.0000000 0.0000000 Nchannels of the left tip = 1 k1(2pi/a) k2(2pi/a) E-Ef (eV) -0.0133793 0.0000000 -0.2000000 Nchannels of the left tip = 1 k1(2pi/a) k2(2pi/a) E-Ef (eV) -0.0520218 0.0000000 -0.4000000 ------------------------------------------------------ The problem is however that though the code calculates correctly all the complex bands, the subroutine (summary_band.f90) writing the bands in the file 'bands.cu' and in the common output file confuses two branches at k_z and -k_z in the present case where no mirror symmetry in the XY plane is present so that k_z is not equivalent to -k_z. I will fix this problem in the CVS version as soon as possible. Thank you for reporting the problem. Best regards, Alexander Smogunov On Monday 23 June 2008 22:19, Manoj Srivastava wrote: > Dear PWSCF users, > I am a new user of PWSCF and I have encountered a problem in calculating > complex band of Cu(111). I see a band gap and there are no bands crossing > Fermi energy. which is obviously wrong :( > I have used hexagonal unit cells containing 3 atoms. Cu.pz-d-rrkjus.UPF > pseudo potential was used in the calculation. It would be great if > somebody can suggest me something. > Following is the input file I used - > > &system > ibrav =4, > celldm(1) =4.78076379, celldm(3)=2.44948974, > nat= 3, > ntyp= 1, > nelec=33.0, > nbnd=24, > ecutwfc =25.0, > occupations= 'smearing', > smearing='gaussian', > degauss=0.02 > > / > &electrons > diagonalization='$diago' > conv_thr = 1.0e-8 > mixing_beta=0.7 > / > ATOMIC_SPECIES > Cu 63.55 Cu.pz-d-rrkjus.UPF > ATOMIC_POSITIONS {bohr} > Cu 0.0 0.0 0.0 > Cu 2.390381895 1.380087631 3.903477289 > Cu 4.780763790 2.760175261 7.806954578 > K_POINTS (automatic) > 8 8 8 0 0 0 > EOF > # complex bands of cu(111) along the 001 direction K_perp=0 > cat > cu.cond.in << EOF > &inputcond > outdir='$TMP_DIR/' > prefixl='cu' > band_file ='bands.cu' > ikind=0 > energy0=10.d0 > denergy=-0.2d0 > ewind=3.d0 > epsproj=1.d-7 > / > 1 > -1.0 1.732 -0.8165 > 100 > EOF > > > Regards, > Manoj > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From giannozz at democritos.it Wed Jun 25 10:20:28 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Wed, 25 Jun 2008 10:20:28 +0200 Subject: [Pw_forum] Generating VBC pseudo of Mg with ld1? In-Reply-To: <20080625031647.qig6b93ew4occo84@www-ext.impmc.jussieu.fr> References: <20080625031647.qig6b93ew4occo84@www-ext.impmc.jussieu.fr> Message-ID: <4861FFCC.6020404@democritos.it> merlin meheut wrote: > [...] I realized that a pseudopotential generated using > Von-Barth Car scheme gave pretty good results in LDA [...] > I would like to try a similar approach with PBE functional. > [...] Apparently, on O-sesame webpage there is no such thing as an "O-sesame webpage" > it is mentioned that this possibility is "temporarily removed". nothing is more permanent than "temporary" changes. It was "temporarily removed" no less than 5 years ago. It isn't easy to generate pseudotentials, the core correction is implemented in a lousy way, and a simple analytical form for PP isn't any longer as important as it used to be > Is there a way to get old versions of ld1 code that do it? try this (no warranty): http://www.fisica.uniud.it/~giannozz/public/atom.tar.gz Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From giannozz at democritos.it Wed Jun 25 10:25:44 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Wed, 25 Jun 2008 10:25:44 +0200 Subject: [Pw_forum] Methods of analyzing chemical bonds In-Reply-To: References: Message-ID: <48620108.7020609@democritos.it> Clark Lee wrote: > Another question: I noticed that electron localization function > analysis was not implemented in previous versions of QE package. it is there since at least 2003. It doesn't work for ultrasoft PP, though (more exactly: the contribution from the augmentation part is missing). See INPUT_PP* Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From akohlmey at cmm.chem.upenn.edu Wed Jun 25 14:35:02 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Wed, 25 Jun 2008 08:35:02 -0400 (EDT) Subject: [Pw_forum] Methods of analyzing chemical bonds In-Reply-To: References: Message-ID: On Wed, 25 Jun 2008, Clark Lee wrote: clark, a number of comments on your mail. my observeration on open source software projects are: if something is not documented, it usually is not working well or well maintained. if that doesn't scare you, you'd have to search the source code. if you want something bad, you'll have to implement it yourself or make a good case that that people will do it for you. 'i want it', or 'i think it is useful' is not very convincing to most... if they wouldn't want it or thought it was useful they would not be working on what they are working on right now. so you'd have to come up with something that will make somebody drop what he or she is doing and do something else instead. just think of what would make you do that? please keep in mind that the ratio of what you get from this package to what you paid for it is infinitely large. in some cases people has offered a "bounty" to programmers to implement certain features into their code of choice... a couple more details, AIM can be done with outside packages, all you have to do is to create an _all-electron_ density. this is of course a problems for pseudopotential calculations and you'd need to see if the PAW reconstruction can write this out for you or you'd have to fudge it somehow. you'll also have to keep in mind that all methods derived for local basis sets first require a projection from the plane wave basis on local orbitals, which can be a big problem, as the projection is not complete due to the diffuseness of the plane wave basis and the lack thereof of the corresponding local basis sets. this in turn can reduce the accuracy of the methods you describe significantly and make them less useful than density differences. they don't have to, but they can. ... and again you are missing the core orbitals which may (or may not) be important. i hope this puts a few things into perspective... cheers, axel. -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From preface767 at sina.com Wed Jun 25 15:07:13 2008 From: preface767 at sina.com (preface767 at sina.com) Date: Wed, 25 Jun 2008 21:07:13 +0800 Subject: [Pw_forum] How to calculate partial density of phonon states? Message-ID: <20080625130713.263BC10C7404@mail3-118.sinamail.sina.com.cn> Dear all   I'm a freshman in QE. I have calculated density of phonon states  of  TiMoH using QE 3.2.3. But I don?t know how to calculate the partial density of  phonon states of H in TiMoH. What should I do to get it started? Is there any code in QE 3.2.3 can be used for this task? Or, should any set-up be done in the input files? Is there anybody can give me some suggestions? Regards ------------------------------------------------ Xu xu Shenyang National Laboratory for Materials Sciences (SYNL) Institute of Metal Research (IMR)  Chinese Academy of Sciences (CAS) E-mail: preface767 at sina.com ------------------------------------------------------------------- 08???????????????????(http://d1.sina.com.cn/sina/limeng3/mail_zhuiyu/2008/mail_zhuiyu_20080623.html) ------------------------------------------------------------------- ????2G????(http://mail.sina.com.cn/) -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080625/c5867226/attachment.htm From manoj at phys.ufl.edu Wed Jun 25 16:10:09 2008 From: manoj at phys.ufl.edu (Manoj Srivastava) Date: Wed, 25 Jun 2008 10:10:09 -0400 (EDT) Subject: [Pw_forum] complex band of Cu(111) In-Reply-To: <200806250838.22413.smogunov@sissa.it> Message-ID: Dear Alaxendar and Letizia Chiodo, Thank you very much for your reply. Did you use my input file for this calculation, or you have your own input file, in that case would you mind sending me your input file? The thing is I am using 3 atoms hexagonal unit cell for the Cu bulk. I guess the problem in my input file is about the direction in which i am trying to calculate the bands. In the last few lines of 'inputcond' 1 0 0 1 100 I assume that the second line is the direction, and last line dictates how fine my k points are in that direction. Does the second line always have to be always in Kx or Ky or Kz direction. Can't I use any combination of Kx and Ky or Kz? In my calculation to make sure that I was trying to sample the right K points in BZ, I used example 1, given in the PWSCF manual, as reference where they calculate real bands structure of Cu with 1 atom fcc unit cell. Then, I transformed the most symmetric directions i.e. delta, sigma and lambda lines, used in BZ of example 1 to the BZ of hexagonal lattice. But unfortunately I did not get the real part of complex bands to match up with the one in example1! Also, in this direction, I did not get any imaginary part of K ! I appreciate your help in this matter. Also, would you mind sending references on comples bands, if you know any,specially on which the code summary_band.f90 is based. Regards, Manoj On Wed, 25 Jun 2008, Alexander wrote: > Dear Manoj > On my PC with espresso-4.0 I do get one band around the Fermi energy > as it should be. My output for energies around Ef looks like this: > ------------------------------ > k( 1) = ( -1.0000000 1.7320000), wk = 1.0000000 > > ngper, shell number = 39 24 > ngper, n2d = 39 39 > Nchannels of the left tip = 1 > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > 0.0891385 0.0000000 0.4000000 > Nchannels of the left tip = 1 > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > 0.0568822 0.0000000 0.2000000 > Nchannels of the left tip = 1 > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > 0.0228076 0.0000000 0.0000000 > Nchannels of the left tip = 1 > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > -0.0133793 0.0000000 -0.2000000 > Nchannels of the left tip = 1 > k1(2pi/a) k2(2pi/a) E-Ef (eV) > > -0.0520218 0.0000000 -0.4000000 > ------------------------------------------------------ > The problem is however that though the code calculates > correctly all the complex bands, the subroutine (summary_band.f90) > writing the bands in the file 'bands.cu' and in the common > output file confuses two branches at k_z and -k_z in the present > case where no mirror symmetry in the XY plane is present so that k_z > is not equivalent to -k_z. I will fix this problem in the CVS version > as soon as possible. Thank you for reporting the problem. > Best regards, > Alexander Smogunov > > On Monday 23 June 2008 22:19, Manoj Srivastava wrote: > > Dear PWSCF users, > > I am a new user of PWSCF and I have encountered a problem in calculating > > complex band of Cu(111). I see a band gap and there are no bands crossing > > Fermi energy. which is obviously wrong :( > > I have used hexagonal unit cells containing 3 atoms. Cu.pz-d-rrkjus.UPF > > pseudo potential was used in the calculation. It would be great if > > somebody can suggest me something. > > Following is the input file I used - > > > > &system > > ibrav =4, > > celldm(1) =4.78076379, celldm(3)=2.44948974, > > nat= 3, > > ntyp= 1, > > nelec=33.0, > > nbnd=24, > > ecutwfc =25.0, > > occupations= 'smearing', > > smearing='gaussian', > > degauss=0.02 > > > > / > > &electrons > > diagonalization='$diago' > > conv_thr = 1.0e-8 > > mixing_beta=0.7 > > / > > ATOMIC_SPECIES > > Cu 63.55 Cu.pz-d-rrkjus.UPF > > ATOMIC_POSITIONS {bohr} > > Cu 0.0 0.0 0.0 > > Cu 2.390381895 1.380087631 3.903477289 > > Cu 4.780763790 2.760175261 7.806954578 > > K_POINTS (automatic) > > 8 8 8 0 0 0 > > EOF > > # complex bands of cu(111) along the 001 direction K_perp=0 > > cat > cu.cond.in << EOF > > &inputcond > > outdir='$TMP_DIR/' > > prefixl='cu' > > band_file ='bands.cu' > > ikind=0 > > energy0=10.d0 > > denergy=-0.2d0 > > ewind=3.d0 > > epsproj=1.d-7 > > / > > 1 > > -1.0 1.732 -0.8165 > > 100 > > EOF > > > > > > Regards, > > Manoj > > > > _______________________________________________ > > 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 ihsanas at yahoo.com Wed Jun 25 17:04:52 2008 From: ihsanas at yahoo.com (=?iso-8859-1?Q?=C7=CD=D3=C7=E4_=DA=D1=ED=DE=C7=CA?=) Date: Wed, 25 Jun 2008 08:04:52 -0700 (PDT) Subject: [Pw_forum] neb Message-ID: <622416.90977.qm@web52309.mail.re2.yahoo.com> Hi In order to test how much time can parallel cluster take time when working on neb job and comparing it with the time that can one CPU take ,we put neb job with 15 atom on 1-a cluster (MPICH2) of 4 nodes (2 CPU in one station ). 2- a cluster (MPICH2) OF 2 nodes (2 CPU in one station ) 3- On one CPU we used for 1 &2 mpiexec -n 4 pw.x <-->-- mpiexec -n 2 pw.x <-->-- All CPU have same properties (8G for Ram ),opteron AMD. We found that the job on one CPU (3rd one )took less time than others.!!! Why this happen could anybody tell us please? best regards Ihsan Erikat Jordan -Physics department From eyvaz_isaev at yahoo.com Wed Jun 25 17:10:14 2008 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Wed, 25 Jun 2008 08:10:14 -0700 (PDT) Subject: [Pw_forum] How to calculate partial density of phonon states? In-Reply-To: <20080625130713.263BC10C7404@mail3-118.sinamail.sina.com.cn> Message-ID: <400956.44627.qm@web65714.mail.ac4.yahoo.com> Hi, I will send you the code which is exactly what is you need. Bests, Eyvaz. ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com --- On Wed, 6/25/08, preface767 at sina.com wrote: > From: preface767 at sina.com > Subject: [Pw_forum] How to calculate partial density of phonon states? > To: pw_forum at pwscf.org > Date: Wednesday, June 25, 2008, 5:07 PM > Dear all >   I'm a freshman in > QE. I have calculated density of phonon > states  of  TiMoH using QE > 3.2.3. But I don?t know how to calculate the partial > density of  phonon states of H in > TiMoH. What should I do to get it started? Is > there any code in QE 3.2.3 can be used > for this task? Or, should any set-up be done > in the input files? Is there anybody can > give me some suggestions? > Regards > ------------------------------------------------ > Xu xu > Shenyang National Laboratory for Materials Sciences (SYNL) > Institute of Metal Research (IMR)  > Chinese Academy of Sciences (CAS) > E-mail: preface767 at sina.com > > > ------------------------------------------------------------------- > 08???????????????????(http://d1.sina.com.cn/sina/limeng3/mail_zhuiyu/2008/mail_zhuiyu_20080623.html) > > ------------------------------------------------------------------- > ????2G????(http://mail.sina.com.cn/)_______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From akohlmey at cmm.chem.upenn.edu Wed Jun 25 17:16:53 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Wed, 25 Jun 2008 11:16:53 -0400 (EDT) Subject: [Pw_forum] neb In-Reply-To: <622416.90977.qm@web52309.mail.re2.yahoo.com> References: <622416.90977.qm@web52309.mail.re2.yahoo.com> Message-ID: On Wed, 25 Jun 2008, ????? ?????? wrote: IE> Hi dear ihsan, IE> In order to test how much time can parallel cluster take time when working on neb job and comparing it with the time that can one CPU take ,we put neb job with 15 atom on IE> 1-a cluster (MPICH2) of 4 nodes (2 CPU in one station ). IE> 2- a cluster (MPICH2) OF 2 nodes (2 CPU in one station ) IE> 3- On one CPU IE> we used for 1 &2 IE> mpiexec -n 4 pw.x <-->-- IE> IE> mpiexec -n 2 pw.x <-->-- IE> IE> IE> All CPU have same properties (8G for Ram ),opteron AMD. IE> We found that the job on one CPU (3rd one )took less time than others.!!! IE> Why this happen could anybody tell us please? what kind of interconnect do you have between the nodes? please have a look at the part of the QE documentation that explains how to run efficiently. by default you parallelize over g-space only. that is what always works, but what also demands a _very_ fast network (myrinet, infiniband etc.). if you run with k-points you can parallelize across those, but you have to use a command line flag to _tell_ QE how many pools of k-points you want to use per MPI task, and similarly for NEB, you have tell QE how many groups of images you want to have. in principle one could implement some heuristics to try and figure out the best way of running a job, but that can go as wrong as not doing it and trusting the user that he or she knows what he or she is doing. this is all documented and discussed on this list many times... cheers, axel. IE> best regards IE> Ihsan Erikat IE> Jordan -Physics department IE> IE> IE> IE> IE> _______________________________________________ IE> Pw_forum mailing list IE> Pw_forum at pwscf.org IE> http://www.democritos.it/mailman/listinfo/pw_forum IE> -- ======================================================================= 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 democritos.it Wed Jun 25 17:35:24 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Wed, 25 Jun 2008 17:35:24 +0200 Subject: [Pw_forum] something wrong with vc-relax & relax on my clusters In-Reply-To: References: Message-ID: <486265BC.1080007@democritos.it> Axel Kohlmeyer wrote: > btw: don't worry about MKL's FFT not being detected. > there is no direct support anyways and both the fftw3 > and the fftw2 wrappers don't work reliably. > in all my tests it turned out to be slower than the > bundled FFTW, except for power of two grids (which confirms > results with BenchFFT). more importantly, you have to make > certain that the OMP_NUM_THREADS environment variable is set > to 1 or else it will multi-thread across all local cpu cores, > regardless of whether you parallelize with MPI (see the MKL > documentation for details). in serial execution (which uses 3d FFT) with no threading, I notice a sizable speedup of MKL (FFTW3 interface) wrt bundled FFTW. I instead never noticed any improvement in FFTW v.3 wrt either FFTW v.2 or the bundled FFTW. Disclaimer: there are few things I dislike more, and trust less, than benchmarking. Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From vegalew at hotmail.com Thu Jun 26 05:20:43 2008 From: vegalew at hotmail.com (vega lew) Date: Thu, 26 Jun 2008 11:20:43 +0800 Subject: [Pw_forum] questions about intel CPU and vc-relax using bfgs cell optimization Message-ID: Dear all, I built a cluster of 5 computers with intel Core TM 2 Q6600 CPU (quadcore), and 40G memory total (8G each) on S3000AH system board. The network is 1Gbit Ethernet. I also checked the em64t option in BIOS is on, so I think Q6600 is a cpu using em64t technology. For more information about my CPU, see http://processorfinder.intel.com/details.aspx?sSpec=SL9UM also I typed "more /proc/cpuinfo", the information for my cpu and OS as follows: LSB Version: :core-3.0-amd64:core-3.0-ia32:core-3.0-noarch:graphics-3.0-amd64:graphics-3.0-ia32:graphics-3.0-noarch Distributor ID: RedHatEnterpriseAS Description: Red Hat Enterprise Linux AS release 4 (Nahant Update 4) Release: 4 Codename: NahantUpdate4 processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 15 model name : Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz stepping : 11 cpu MHz : 2400.150 cache size : 4096 KB physical id : 0 siblings : 4 core id : 0 cpu cores : 4 fpu : yes fpu_exception : yes cpuid level : 10 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm pni monitor ds_cpl est tm2 cx16 xtpr bogomips : 4806.03 clflush size : 64 cache_alignment : 64 address sizes : 36 bits physical, 48 bits virtual power management: processor : 1 vendor_id : GenuineIntel cpu family : 6 model : 15 model name : Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz stepping : 11 cpu MHz : 2400.150 cache size : 4096 KB physical id : 0 siblings : 4 core id : 2 cpu cores : 4 fpu : yes fpu_exception : yes cpuid level : 10 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm pni monitor ds_cpl est tm2 cx16 xtpr bogomips : 4798.75 clflush size : 64 cache_alignment : 64 address sizes : 36 bits physical, 48 bits virtual power management: processor : 2 vendor_id : GenuineIntel cpu family : 6 model : 15 model name : Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz stepping : 11 cpu MHz : 2400.150 cache size : 4096 KB physical id : 0 siblings : 4 core id : 1 cpu cores : 4 fpu : yes fpu_exception : yes cpuid level : 10 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm pni monitor ds_cpl est tm2 cx16 xtpr bogomips : 4799.49 clflush size : 64 cache_alignment : 64 address sizes : 36 bits physical, 48 bits virtual power management: processor : 3 vendor_id : GenuineIntel cpu family : 6 model : 15 model name : Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz stepping : 11 cpu MHz : 2400.150 cache size : 4096 KB physical id : 0 siblings : 4 core id : 3 cpu cores : 4 fpu : yes fpu_exception : yes cpuid level : 10 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm pni monitor ds_cpl est tm2 cx16 xtpr bogomips : 4799.52 clflush size : 64 cache_alignment : 64 address sizes : 36 bits physical, 48 bits virtual power management: Therefore, I updated my intel C++ and Fortran Compilers from 10.1.008 to latest vision 10.1.017 for Intel(R) 64 and MKL from 10.0.011 to latest 10.0.3.020, file names displayed on website were l_cc_p_10.1.017_intel64.tar.gz, l_cc_p_10.1.017_intel64.tar.gz and l_mkl_p_10.0.3.020.tgz. After installation of the three, I compiled for em64t vision of blas95 lapack95 in /opt/intel/mkl/10.0.3.020/interfaces/ using ifort under /opt/intel/fce/10.1.017/bin/. Then compiled mpich2 using ifort and icc. But when I compile fftw 2.1.5 an error occurred, so I compile the fftw 2.1.5 using 10.1.008 ifort and icc on other node with same hardware, the scp it to master node. After all above done, I turned to compile QE. But to my surprise, QE detected my architecture as amd64, not ia32 or ia64. My first question is does QE support the intel EM64T technology and take advantages from it ? At last, I compile the QE using amd64 architecture schedule by intel C++ and Fortran 10.1.017 vision and MKL 10.0.3.020 library, but I find it less efficienct the the QE compiled by intel C++ and Fortran 10.1.008 vision and 10.0.011 library. The efficiency of QE compiled by 10.1.008 compiler and 10.0.011 is about 60% but the QE compiled by 10.1.017compiler is 10% tested by input file like this: &CONTROL title = 'Anatase lattice BFGS' , calculation = 'vc-relax' , restart_mode = 'from_scratch' , outdir = '/home/vega/tmp/' , pseudo_dir = '/home/vega/espresso-4.0/pseudo/' , prefix = 'Anatase lattice default' , etot_conv_thr = 0.000000735 , forc_conv_thr = 0.0011668141375 , nstep = 1000 , / &SYSTEM ibrav = 6, celldm(1) = 7.135605333, celldm(3) = 2.5121822033898305084745762711864, nat = 12, ntyp = 2, ecutwfc = 25 , ecutrho = 200 , / &ELECTRONS conv_thr = 7.3D-8 , / &IONS ion_dynamics = 'bfgs' , / &CELL cell_dynamics = 'bfgs' , cell_dofree = 'xyz' , / ATOMIC_SPECIES Ti 47.86700 Ti.pw91-sp-van_ak.UPF O 15.99940 O.pw91-van_ak.UPF ATOMIC_POSITIONS angstrom Ti 0.000000000 0.000000000 0.000000000 Ti 1.888000000 1.888000000 4.743000000 Ti 0.000000000 1.888000000 2.372000000 Ti 1.888000000 0.000000000 7.115000000 O 0.000000000 0.000000000 1.973000000 O 1.888000000 1.888000000 6.716000000 O 0.000000000 1.888000000 4.345000000 O 1.888000000 0.000000000 9.088000000 O 1.888000000 0.000000000 5.141000000 O 0.000000000 1.888000000 0.398000000 O 1.888000000 1.888000000 2.770000000 O 0.000000000 0.000000000 7.513000000 K_POINTS automatic 7 7 3 1 1 1 My second question is about the efficiency: Which compiler and MKL vision is the best one for my cluster? Why I updated my MKL and compilers brings me less efficiency? What is the best efficiency of my cluster can reach ? 60% is low or high for QE? Third question is about bfgs cell optimization. When I run the above input file with the 'cell_dofree = 'xyz'' in &CELL section. I think it mean only a,b,c of the lattice are changeable, and three angles, alpha, beta, gamma is fixed to 90 degrees according to the PWgui. So that, the lattice will remain orthogonal. But the results showed the angles were still changing. the results file as follows: ...... entering subroutine stress ... total stress (Ry/bohr**3) (kbar) P= -1.85 0.00004989 0.00000000 0.00000000 7.34 0.00 0.00 0.00000000 0.00005560 0.00000000 0.00 8.18 0.00 0.00000000 0.00000000 -0.00014314 0.00 0.00 -21.06 number of scf cycles = 5 number of bfgs steps = 2 enthalpy old = -725.4093146855 Ry enthalpy new = -725.4093492895 Ry CASE: enthalpy_new < enthalpy_old new trust radius = 0.0190701467 bohr new conv_thr = 0.0000000074 Ry CELL_PARAMETERS (alat) 0.992368528 0.000000000 -0.000000009 0.000000000 0.992410788 0.000000037 -0.000000021 0.000000091 2.503790203 ATOMIC_POSITIONS (angstrom) Ti 0.000000000 0.000000000 0.000040364 Ti 1.873591740 1.873671740 4.727209100 Ti -0.000000020 1.873671654 2.362997426 Ti 1.873591720 0.000000258 7.091885462 O -0.000000017 0.000000072 1.972736887 O 1.873591723 1.873671812 6.700607058 O -0.000000037 1.873671726 4.336885456 O 1.873591703 0.000000330 9.064014758 O 1.873591737 0.000000186 5.117531758 O -0.000000004 1.873671582 0.390380409 O 1.873591756 1.873671668 2.753778039 O -0.000000064 0.000000273 7.481645190 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from checkallsym : error # 2 not orthogonal operation %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0[cli_0]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0 rank 0 in job 3 node5_32785 caused collective abort of all ranks exit status of rank 0: killed by signal 9 Do you think, I shloud never using BFGS to optimize Anatase lattice? But CASTEP can do so,why? thanking for reading. I'm looking forward to responding. Vega Lew PH.D Candidate in Chemical Engineering College of Chemistry and Chemical Engineering Nanjing University of Technology, 210009, Nanjing, Jiangsu, China _________________________________________________________________ Discover the new Windows Vista http://search.msn.com/results.aspx?q=windows+vista&mkt=en-US&form=QBRE -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080626/e73977db/attachment-0001.htm From akohlmey at cmm.chem.upenn.edu Thu Jun 26 05:50:38 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Wed, 25 Jun 2008 23:50:38 -0400 (EDT) Subject: [Pw_forum] questions about intel CPU and vc-relax using bfgs cell optimization In-Reply-To: References: Message-ID: On Thu, 26 Jun 2008, vega lew wrote: VL> VL> Dear all, VL> VL> I built a cluster of 5 computers with intel Core TM 2 Q6600 CPU VL> (quadcore), and 40G memory total (8G each) on S3000AH system board. VL> The network is 1Gbit Ethernet. I also checked the em64t option in VL> BIOS is on, so I think Q6600 is a cpu using em64t technology. For VL> more information about my CPU, see more important is to determine whether you installed a 32-bit or a 64-bit version of the OS. you can find that out with 'uname -a'. for 32-bit you get (amongst others) i386 and i686 whereas for 64-bit you get x86_64. regardless of bios options or what cpuinfo shows, the cpu can handle both. [...] VL> Therefore, I updated my intel C++ and Fortran Compilers from VL> 10.1.008 to latest vision 10.1.017 for Intel(R) 64 and MKL from VL> 10.0.011 to latest 10.0.3.020, file names displayed on website were VL> l_cc_p_10.1.017_intel64.tar.gz, l_cc_p_10.1.017_intel64.tar.gz and VL> l_mkl_p_10.0.3.020.tgz. After installation of the three, I compiled VL> for em64t vision of blas95 lapack95 in those are not needed. VL> /opt/intel/mkl/10.0.3.020/interfaces/ using ifort under VL> /opt/intel/fce/10.1.017/bin/. Then compiled mpich2 using ifort and VL> icc. But when I compile fftw 2.1.5 an error occurred, so I compile it is much better to compiler fftw with the native gcc compiler. but since QE actually contains FFTW there is no need to install or compile it. VL> the fftw 2.1.5 using 10.1.008 ifort and icc on other node with same VL> hardware, the scp it to master node. After all above done, I turned VL> to compile QE. VL> But to my surprise, QE detected my architecture as amd64, not ia32 VL> or ia64. My first question is does QE support the intel EM64T VL> technology and take advantages from it ? it is neither ia32 nor ia64. amd actually invented this 64-bit mode and then intel named it EM64t (to avoid having to call it amd64). the official linux architecture is x86_64. VL> At last, I compile the QE using amd64 architecture schedule by intel VL> C++ and Fortran 10.1.017 vision and MKL 10.0.3.020 library, but I VL> find it less efficienct the the QE compiled by intel C++ and Fortran VL> 10.1.008 vision and 10.0.011 library. The efficiency of QE compiled VL> by 10.1.008 compiler and 10.0.011 is about 60% but the QE compiled VL> by 10.1.017compiler is 10% tested by input file like this: how do you determine this "efficiency"? how do you run your job? since MKL will automatically multi-thread across all cores, you have to set the environment variable OMP_NUM_THREADS to 1 or else you'll be oversubscribing each cpu 4x. secondly, for efficient operation across gigabit ethernet (which is quite slow and has very high latencies), you have to parallelize across k-point pools, at least between nodes or else your performance will be horrible. if you have not taken care of mkl multithreading all performance data will be bogus. VL> My second question is about the efficiency: VL> Which compiler and MKL vision is the best one for my cluster? the one that runs corectly. the performance difference between a different optimized implementations of BLAS are on average of the order of 10% of the total time. compiler impact (e.g. between g95/gfortran and intel 10) is of the same order. VL> Why I updated my MKL and compilers brings me less efficiency? that has most likely other reasons (runaway processes?, other users?) VL> What is the best efficiency of my cluster can reach ? 60% is low or high for QE? you should be able to do better, if you run your job the right way. please check the documentation on how to run QE properly in parallel. to determine the performance baseline, you should first run a test with only one MPI task and set OMP_NUM_THREADS to 1 (read the intel docs about this). cheers, axel. -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From vegalew at hotmail.com Thu Jun 26 07:25:35 2008 From: vegalew at hotmail.com (vega lew) Date: Thu, 26 Jun 2008 13:25:35 +0800 Subject: [Pw_forum] questions about intel CPU and vc-relax using bfgs cell optimization Message-ID: Dear Axel, First, I wanna express my acknowledgement for your kindly responding. From it I learned a lot. I typed 'uname -a', the information 'Linux node5 2.6.9-42.ELsmp #1 SMP Wed Jul 12 23:32:02 EDT 2006 x86_64 x86_64 x86_64 GNU/Linux' displayed. So, it's a x86_64 OS. > VL> Why I updated my MKL and compilers brings me less efficiency? > > that has most likely other reasons (runaway processes?, other users?) I'm sure no other users, I'm the admin of that cluster. Have your tested the intel C++ and Fortran 10.1.017 Compilers for intel 64 vision and MKL 10.0.3.020? Do you think it's really suitable for my clusters and QE? Do you think it's better than 10.1.008 compilers and 10.0.011 MKL? > VL> At last, I compile the QE using amd64 architecture schedule by intel > VL> C++ and Fortran 10.1.017 vision and MKL 10.0.3.020 library, but I > VL> find it less efficienct the the QE compiled by intel C++ and Fortran > VL> 10.1.008 vision and 10.0.011 library. The efficiency of QE compiled > VL> by 10.1.008 compiler and 10.0.011 is about 60% but the QE compiled > VL> by 10.1.017compiler is 10% tested by input file like this: > > how do you determine this 'efficiency'? how do you run your job? I run the inputfile by mpi command 'mpiexec -n 20 pw.x < inputfile > outputfile', the job started immediately. The input file is like this: &CONTROL title = 'Anatase lattice BFGS' , calculation = 'vc-relax' , restart_mode = 'from_scratch' , outdir = '/home/vega/tmp/' , pseudo_dir = '/home/vega/espresso-4.0/pseudo/' , prefix = 'Anatase lattice default' , etot_conv_thr = 0.000000735 , forc_conv_thr = 0.0011668141375 , nstep = 1000 , / &SYSTEM ibrav = 6, celldm(1) = 7.135605333, celldm(3) = 2.5121822033898305084745762711864, nat = 12, ntyp = 2, ecutwfc = 25 , ecutrho = 200 , / &ELECTRONS conv_thr = 7.3D-8 , / &IONS ion_dynamics = 'bfgs' , / &CELL cell_dynamics = 'bfgs' , cell_dofree = 'xyz' , / ATOMIC_SPECIES Ti 47.86700 Ti.pw91-sp-van_ak.UPF O 15.99940 O.pw91-van_ak.UPF ATOMIC_POSITIONS angstrom Ti 0.000000000 0.000000000 0.000000000 Ti 1.888000000 1.888000000 4.743000000 Ti 0.000000000 1.888000000 2.372000000 Ti 1.888000000 0.000000000 7.115000000 O 0.000000000 0.000000000 1.973000000 O 1.888000000 1.888000000 6.716000000 O 0.000000000 1.888000000 4.345000000 O 1.888000000 0.000000000 9.088000000 O 1.888000000 0.000000000 5.141000000 O 0.000000000 1.888000000 0.398000000 O 1.888000000 1.888000000 2.770000000 O 0.000000000 0.000000000 7.513000000 K_POINTS automatic 7 7 3 1 1 1 And I see 4 process on each node by 'top' command. So I think each core has a process. Is there anything I misunderstanding? I think I should also use the mpi command like this 'mpiexec -n 20 pw.x -npool 5 < inputfile > outputfile' I has tried it. But no obvious improvement, the usage of the CPU count by 'sar' command still about 60%, 7% for the system, and the rest is idle. > it is much better to compiler fftw with the native gcc compiler. > but since QE actually contains FFTW there is no need to install > or compile it. QE contains FFTW? Where is it? Does it should be detected by the QE configure? if I compile the FFTW by gcc or don't compile FFTW, QE configure can't detect it and show the information about the FFTW in QE configure process even I have put the dirs to the environment variable. So, I think only the fftw compiled by ifort can be found by my QE. Do your think intel MKL is needed? When I configured the QE without intel MKL, the QE configure also could find the BLAS and LAPACK under itself folder. You mentioned OMP_NUM_THREADS again. I'm sorry I know little about it. Should I use the export command like 'export OMP_NUM_THREADS=1'? If the command is enough, could you please tell me, when I should type it? before configuring the QE? or before using the mpiexec command? And could you give me some hints about the optimization of Anatase lattice using BFGS schedule? Why the 'cell_dofree = 'xyz'' in &CELL section take no effect to aviod lattice angles changing, and result in a 'not orthogonal operation' error? Do you think, I shloud never using BFGS to optimize Anatase lattice? thank your again for your so detail responding. Vega Lew PH.D Candidate in Chemical Engineering College of Chemistry and Chemical Engineering Nanjing University of Technology, 210009, Nanjing, Jiangsu, China > Date: Wed, 25 Jun 2008 23:50:38 -0400 > From: akohlmey at cmm.chem.upenn.edu > To: vegalew at hotmail.com > CC: pw_forum at pwscf.org > Subject: Re: [Pw_forum] questions about intel CPU and vc-relax using bfgs cell optimization > > On Thu, 26 Jun 2008, vega lew wrote: > > VL> > VL> Dear all, > VL> > > VL> I built a cluster of 5 computers with intel Core TM 2 Q6600 CPU > VL> (quadcore), and 40G memory total (8G each) on S3000AH system board. > VL> The network is 1Gbit Ethernet. I also checked the em64t option in > VL> BIOS is on, so I think Q6600 is a cpu using em64t technology. For > VL> more information about my CPU, see > > more important is to determine whether you installed a 32-bit or > a 64-bit version of the OS. you can find that out with 'uname -a'. > for 32-bit you get (amongst others) i386 and i686 whereas for > 64-bit you get x86_64. regardless of bios options or what cpuinfo > shows, the cpu can handle both. > > [...] > > VL> Therefore, I updated my intel C++ and Fortran Compilers from > VL> 10.1.008 to latest vision 10.1.017 for Intel(R) 64 and MKL from > VL> 10.0.011 to latest 10.0.3.020, file names displayed on website were > VL> l_cc_p_10.1.017_intel64.tar.gz, l_cc_p_10.1.017_intel64.tar.gz and > VL> l_mkl_p_10.0.3.020.tgz. After installation of the three, I compiled > VL> for em64t vision of blas95 lapack95 in > > those are not needed. > > VL> /opt/intel/mkl/10.0.3.020/interfaces/ using ifort under > VL> /opt/intel/fce/10.1.017/bin/. Then compiled mpich2 using ifort and > VL> icc. But when I compile fftw 2.1.5 an error occurred, so I compile > > it is much better to compiler fftw with the native gcc compiler. > but since QE actually contains FFTW there is no need to install > or compile it. > > VL> the fftw 2.1.5 using 10.1.008 ifort and icc on other node with same > VL> hardware, the scp it to master node. After all above done, I turned > VL> to compile QE. > > VL> But to my surprise, QE detected my architecture as amd64, not ia32 > VL> or ia64. My first question is does QE support the intel EM64T > VL> technology and take advantages from it ? > > it is neither ia32 nor ia64. amd actually invented this 64-bit mode > and then intel named it EM64t (to avoid having to call it amd64). > the official linux architecture is x86_64. > > VL> At last, I compile the QE using amd64 architecture schedule by intel > VL> C++ and Fortran 10.1.017 vision and MKL 10.0.3.020 library, but I > VL> find it less efficienct the the QE compiled by intel C++ and Fortran > VL> 10.1.008 vision and 10.0.011 library. The efficiency of QE compiled > VL> by 10.1.008 compiler and 10.0.011 is about 60% but the QE compiled > VL> by 10.1.017compiler is 10% tested by input file like this: > > how do you determine this 'efficiency'? how do you run your job? > > since MKL will automatically multi-thread across all cores, > you have to set the environment variable OMP_NUM_THREADS to 1 > or else you'll be oversubscribing each cpu 4x. secondly, for > efficient operation across gigabit ethernet (which is quite slow > and has very high latencies), you have to parallelize across > k-point pools, at least between nodes or else your performance > will be horrible. if you have not taken care of mkl multithreading > all performance data will be bogus. > > > VL> My second question is about the efficiency: > VL> Which compiler and MKL vision is the best one for my cluster? > > the one that runs corectly. the performance difference between > a different optimized implementations of BLAS are on average > of the order of 10% of the total time. compiler impact (e.g. > between g95/gfortran and intel 10) is of the same order. > > VL> Why I updated my MKL and compilers brings me less efficiency? > > that has most likely other reasons (runaway processes?, other users?) > > VL> What is the best efficiency of my cluster can reach ? 60% is low or high for QE? > > you should be able to do better, if you run your job the right way. > please check the documentation on how to run QE properly in parallel. > > to determine the performance baseline, you should first run a test > with only one MPI task and set OMP_NUM_THREADS to 1 (read the intel > docs about this). > > cheers, > axel. > > > -- > ======================================================================= > Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu > Center for Molecular Modeling -- University of Pennsylvania > Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 > tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 > ======================================================================= > If you make something idiot-proof, the universe creates a better idiot. _________________________________________________________________ Connect to the next generation of MSN Messenger? http://imagine-msn.com/messenger/launch80/default.aspx?locale=en-us&source=wlmailtagline -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080626/3ddefe21/attachment-0001.htm From marzari at MIT.EDU Thu Jun 26 07:46:40 2008 From: marzari at MIT.EDU (Nicola Marzari) Date: Thu, 26 Jun 2008 07:46:40 +0200 Subject: [Pw_forum] questions about intel CPU and vc-relax using bfgs cell optimization In-Reply-To: References: Message-ID: <48632D40.8040405@mit.edu> vega lew wrote: > Dear all, > > I built a cluster of 5 computers with intel Core TM 2 Q6600 CPU > (quadcore), and 40G memory total (8G each) on S3000AH system board. Dear Vega Law, you can find here tests and CPU timings for Q-E: http://quasiamore.mit.edu/pmwiki/index.php?n=Main.CP90Timings This should give you a baseline on what performance to expect. 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 eariel99 at gmail.com Thu Jun 26 14:15:44 2008 From: eariel99 at gmail.com (Eduardo Ariel Menendez Proupin) Date: Thu, 26 Jun 2008 08:15:44 -0400 Subject: [Pw_forum] linking with Intel MKL 10 in Mac OS X Message-ID: Hi, Has anybody compiled Espresso in a Mac Pro linking with the Intel MKL 10.0.3 or similar library? -- Eduardo Menendez -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080626/ded26934/attachment.htm From Daniele.Passerone at empa.ch Thu Jun 26 14:54:44 2008 From: Daniele.Passerone at empa.ch (Daniele Passerone) Date: Thu, 26 Jun 2008 14:54:44 +0200 Subject: [Pw_forum] cp.x and "second dimension too small" Message-ID: <4863ADB4.11F0.001D.0@empa.ch> Dear developers, I am trying to run cp.x [version 3.2.3] within ensemble dft for a molecule on a metallic surface. Unfortunately I observe an error that has to do with the dimensions of the fft grid. During the CG minimization of electronic states, I get from rhoofr : error # 9 c second dimension too small This has to do with the calculation of the self-consistent charge, but also inspecting the code I could not understand the origin of this behavior. Thank you for any help, Daniele This is my input file: &control calculation='cp' restart_mode='from_scratch' prefix='cp_mol' pseudo_dir = './' outdir='./tmp' nstep=5, iprint=20, isave=20, dt=10.0, wf_collect=.true. ndr=90, ndw=91, verbosity='high' / &system ibrav=0, celldm(1) =39.9317, nat=9,ntyp=2, ecutwfc = 20.0, ecutrho=160., occupations='ensemble', smearing='fd', degauss=0.004, nspin=1, nelec=18 nr1b=20,nr2b=20,nr3b=20 / &electrons conv_thr = 1.0e-8 electron_dynamics='damp', electron_damping=0.2, startingwfc='random', ampre=0.01, emass=700., emass_cutoff=3., orthogonalization='Gram-Schmidt' passop=0.3, ortho_max=15, n_inner =8 startingwfc='random', ampre=0.02, tcg=.true., / &ions ion_dynamics = 'damp', ion_damping = 0., ion_positions = 'from_input', greasp=1.0, ion_radius(1) = 1.8d0, ion_radius(2) = 1.8d0, ion_temperature = 'not_controlled', / ATOMIC_SPECIES C 12.0107 C.pbe-rrkjus.UPF H 1.008 H.pbe-rrkjus.UPF ATOMIC_POSITIONS alat H 0.36317733 0.11811776 -0.05126820 1 1 1 H 0.33347823 0.03794802 -0.07312380 1 1 1 C 0.34054779 0.05622752 0.02693170 1 1 1 C 0.34090358 0.07386291 -0.03630660 1 1 1 C 0.33055349 -0.01080546 0.04894040 1 1 1 H 0.30317609 -0.03849456 0.01423330 1 1 1 H 0.30576881 -0.01323450 0.09475600 1 1 1 H 0.37723626 -0.03389693 0.05512170 1 1 1 H 0.36108514 0.08926878 0.06142280 1 1 1 CELL_PARAMETERS 1 0 0 0.099994 0.582845 0 0 0 0.9797 From giannozz at democritos.it Thu Jun 26 22:24:19 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Thu, 26 Jun 2008 22:24:19 +0200 Subject: [Pw_forum] questions about intel CPU and vc-relax using bfgs cell optimization In-Reply-To: References: Message-ID: On Jun 26, 2008, at 5:20 , vega lew wrote: > from checkallsym : error # 2 > not orthogonal operation try this: http://www.democritos.it:8888/O-sesame/chngview?cn=5972 About the rest of your message: if a process is not running close to 100%, it is either swapping, or waiting for I/O, or for memory access, or for the CPU. It is exceedingly unlikely that this depends on usage of compiler version 10.1.000.0.1234567890.sin(pi/2) instead of > version 10.1.000.0.0987654321.cos(pi/2) 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 democritos.it Thu Jun 26 22:25:54 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Thu, 26 Jun 2008 22:25:54 +0200 Subject: [Pw_forum] cp.x and "second dimension too small" In-Reply-To: <4863ADB4.11F0.001D.0@empa.ch> References: <4863ADB4.11F0.001D.0@empa.ch> Message-ID: Hi Daniele > I am trying to run cp.x [version 3.2.3] you shouldn't: the new version is out 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 eyvaz_isaev at yahoo.com Thu Jun 26 22:59:02 2008 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Thu, 26 Jun 2008 13:59:02 -0700 (PDT) Subject: [Pw_forum] Phonon anomaly in GaSb In-Reply-To: Message-ID: <976675.57633.qm@web65708.mail.ac4.yahoo.com> Dear Gokhan, A sketch of the dynamical matrix you provided is not enough to draw a conclusion. It would be better if you post your input file. If you are sure your input file is correct, check another PsPs. Bests, Eyvaz. ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com --- On Fri, 6/20/08, Gokhan GOKOGLU wrote: > From: Gokhan GOKOGLU > Subject: [Pw_forum] Phonon anomaly in GaSb > To: pw_forum at pwscf.org > Date: Friday, June 20, 2008, 11:58 AM > Dear PWSCF users, > I tried to calculate phonon dipersion of zincblende GaSb, > but I found large > negative modes at Gamma-point. The same system was studied > by Giannozzi et al. > in a previous study (PRB 43, 7231 (1991)) using LDA > functionals and the obtained > dispersion was quite nice without any anomalous > vibrational property. I used > GGA (PW91) vanderbilt US PP. I also checked the kinetic > energy cut-off (70 Ry) > and k-mesh (16x16x16) for convergence. I didn't find > any phonon anomaly in AlSb > with similar geometry. I give the dynamical matrix below. I > would be very > grateful, if someone could give an idea. Thank you very > much.. > > Gokhan Gokoglu > Dept. Physics > Bilkent University, Ankara, Turkey > > > Diagonalizing the dynamical matrix > > q = ( 0.000000000 0.000000000 0.000000000 ) > > > ************************************************************************** > omega( 1) = -9.779567 [THz] = -326.213411 > [cm-1] > ( -0.646207 0.000000 0.460941 0.000000 -0.354330 > 0.000000 ) > ( 0.367516 0.000000 -0.262150 0.000000 0.201517 > 0.000000 ) > omega( 2) = -9.779567 [THz] = -326.213411 > [cm-1] > ( 0.504782 0.000000 0.707669 0.000000 0.000000 > 0.000000 ) > ( -0.287083 0.000000 -0.402471 0.000000 0.000000 > 0.000000 ) > omega( 3) = -9.779567 [THz] = -326.213411 > [cm-1] > ( -0.288464 0.000000 0.205762 0.000000 0.793757 > 0.000000 ) > ( 0.164057 0.000000 -0.117023 0.000000 -0.451432 > 0.000000 ) > omega( 4) = -0.606056 [THz] = -20.215980 > [cm-1] > ( 0.172018 0.000000 0.676699 0.000000 0.095247 > 0.000000 ) > ( 0.173197 0.000000 0.681338 0.000000 0.095900 > 0.000000 ) > omega( 5) = -0.606056 [THz] = -20.215980 > [cm-1] > ( -0.434105 0.000000 0.184061 0.000000 -0.523697 > 0.000000 ) > ( -0.437081 0.000000 0.185323 0.000000 -0.527287 > 0.000000 ) > omega( 6) = -0.606056 [THz] = -20.215980 > [cm-1] > ( 0.527775 0.000000 -0.069163 0.000000 -0.461794 > 0.000000 ) > ( 0.531394 0.000000 -0.069637 0.000000 -0.464960 > 0.000000 ) > > ************************************************************************** > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From akohlmey at cmm.chem.upenn.edu Fri Jun 27 01:18:33 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Thu, 26 Jun 2008 19:18:33 -0400 (EDT) Subject: [Pw_forum] questions about intel CPU and vc-relax using bfgs cell optimization In-Reply-To: References: Message-ID: On Thu, 26 Jun 2008, vega lew wrote: VL> Dear Axel, VL> VL> First, I wanna express my acknowledgement for your kindly VL> responding. From it I learned a lot. if you want to do me a personal favor. please don't use "wanna". i'm quite old fashioned and this kind of shortcut writing always makes me cringe (same goes for "plz" and "U" and so on). if i spend time explaining something, i appreciate somebody taking the time of writing normal language. i understand that many people don't think anything about it, and given my own refusal to write capital letters and the many, many typos when i'm tired it sounds a bit odd, but i cannot help it. thanks. [...] VL> And I see 4 process on each node by 'top' command. So I think each VL> core has a process. Is there anything I misunderstanding? i cannot tell. there are too many ways to mess up a cluster installation. neither 'top' nor 'sar' are good measures for performance. i would first run some MPI benchmarks/tests to make sure everything is set up correctly. VL> I think I should also use the mpi command like this 'mpiexec -n 20 VL> pw.x -npool 5 < inputfile > outputfile' VL> I has tried it. But no obvious improvement, the usage of the CPU VL> count by 'sar' command still about 60%, 7% for the system, and the VL> rest is idle. there is something wrong. hard to tell if it is in the way how you have set up the machine, or compiled your executables or running the jobs. i just did a series of runs using your input on one of our clusters and you can see from that roughly what kind of performance and scaling you can expect. you'll find a summary at the end of this mail. perhaps it is of use to other people looking for optimal ways of running QE jobs. the main lesson: it helps to read the documentation and experiment a little bit. [...] VL> QE contains FFTW? Where is it? Does it should be detected by the QE configure? it is in clib and it is used when FFTW is _not_ detected. VL> if I compile the FFTW by gcc or don't compile FFTW, QE configure VL> can't detect it and show the information about the FFTW in QE VL> configure process even I have put the dirs to the environment VL> variable. So, I think only the fftw compiled by ifort can be found VL> by my QE. no. fftw is c code and QE uses its own c wrappers so using ifort has no impact at all. VL> Do your think intel MKL is needed? When I configured the QE without VL> intel MKL, the QE configure also could find the BLAS and LAPACK VL> under itself folder. this is only a matter of last resort. using MKL will speed up calculations quite a bit. VL> You mentioned OMP_NUM_THREADS again. I'm sorry I know little about it. you have to read the MKL documentation. VL> Should I use the export command like 'export OMP_NUM_THREADS=1'? VL> If the command is enough, could you please tell me, when I should that will only set it locally. this has nothing to do with compiling QE, but running it. so it has to be set _always_ when running in parallel, _unless_ for some very special cases (gamma point only, many states, serial executable) where that may actually be faster than running MPI-parallel. VL> type it? before configuring the QE? or before using the mpiexec VL> command? whether that is enough or not, i cannot tell. that depends on your MPI library. you have to read the MPI library documentation how you export local environment variables. i'm using OpenMPI as mpi library and i start my parallel jobs with mpirun -x OMP_NUM_THREADS=1 ... to set the environment for all MPI tasks. all MPI packages are different, so you have to read the documentation of your MPI package (or use OpenMPI ;-) ). VL> And could you give me some hints about the optimization of Anatase VL> lattice using BFGS schedule? Why the 'cell_dofree = 'xyz'' in &CELL VL> section take no effect to aviod lattice angles changing, and result VL> in a 'not orthogonal operation' error? VL> Do you think, I shloud never using BFGS to optimize Anatase lattice? well, you got already paolo's recommendation, but i would also like to point out that this feature is flagged as "experimental"... in any case, here is the protocol of my test runs of your system. enjoy, axel. the following tests were run on dual processor AMD opteron 248 nodes with a 2.2GHz clock and 2GB RAM. The machine has a myrinet 2000 rev04 interconnect and a GigaBit Ethernet with a (pretty crappy) SMC gigabit switch and Broadcom BCM5704 ethernet controllers (tg3 driver). the machine runs scientific linux 4.x (equivalent to RHEL 4.x). i've changed the input to do only one SCF calculation. that run converges with the given settings in 17 SCF cycles. timings are taken from the PWSCF line at the end of the job. performance percentages are taken relative to wall time (this is all we care about, after, right?). first the serial performance of the parallel binary (a non-parallel compiled binary should be ~10% faster, but i don't have one at hand right now). 13m31.08s CPU time, 13m37.49s wall time now running in parallel across the two cpus within one node 7m57.19s CPU time, 8m 2.49s wall time ( ~59%) => not perfect scaling, but quite nice now running across 10 nodes (i.e. 20 cpus) equivalent to 5 nodes with quad-core using the gigabit network: 12m23.00s CPU time, 35m59.80s wall time (~265%) => OUCH! gigabit and g-space parallization don't mix well. not only does it take much more cpu time. there is a lot of waiting involved (2/3 of the time). now using the myrinet instead of gigabit with the same settings: 5m37.57s CPU time, 5m59.54s wall time ( ~44%) => _much_ better. so we didn't waste our money on the fast network but the speedup is only large compared to gigabit. perhaps we are pushing to hard. lets try less cpus, i.e. only two nodes (i.e. 4 cpus total). 5m37.41s CPU time, 5m48.58s wall time ( ~43%) => indeed. same speedup with only two nodes. lets see if it can go faster with twice as many nodes (8 cpus). 4m19.55s CPU time, 4m27.07s wall time ( ~32%) => a bit faster but it looks like we're getting close to the end of the line... lets see if it can go faster with a few more nodes (12 cpus). 5m13.68s CPU time, 5m29.59s wall time ( ~40%) => no. we're out of luck. so g-space parallelization scales out even on a fast network between 8 and 12 cpus for this system. now lets move on to using -npools. again we start with 10 nodes of 2 cpu. we have 32 k-points in total, so we can use 20 pools. using gigabit first. 2m32.99s CPU time, 2m48.98s wall time ( ~20%) => NICE. almost 5 times faster and the gigabit does not hurt us anymore. cool. can we go any faster? trying 16-nodes/32 cpus with 32 pools 2m 8.31s CPU time, 2m25.25s wall time ( ~18%) => a-ha... and matching the number of pools with the k-points makes this optimal. how about the fast network? 2m18.56s CPU time, 2m21.71s wall time ( ~17%) => a tiny bit faster, but not much. can we go even faster? now trying 32 nodes with 32 pools, i.e. two cpus g-space parallel, gigabit: 0m50.14s CPU time, 1m 0.91s wall time (~7.5%) => perfect. and the same with myrinet: 53.08s CPU time, 55.82s wall time (~6.8%) => even better. too bad our cluster does not have more empty nodes, or else the job might run even faster... VL> VL> thank your again for your so detail responding. VL> VL> Vega Lew VL> PH.D Candidate in Chemical Engineering VL> College of Chemistry and Chemical Engineering VL> Nanjing University of Technology, 210009, Nanjing, Jiangsu, China VL> -- ======================================================================= 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 vegalew at hotmail.com Fri Jun 27 08:12:29 2008 From: vegalew at hotmail.com (vega lew) Date: Fri, 27 Jun 2008 14:12:29 +0800 Subject: [Pw_forum] How to handle with the I/O wait Message-ID: Dear all, First thank you all again to spend a lot time reading my mail, dealing with my questions, and giving me so much help. It's meanful for my resarch work. Paolo said "if a process is not running close to 100%, it is either swapping, or waiting for I/O, or for memory access, or for the CPU. This remind me the case of CASTEP. CASTEP is so good a software, but it more suitable for commercial use. So I gived it up. According to many high quality scientific paper such as nature materials Journal of the American Society, Q-E is the best choice for academic using. According to my CASTEP experiences, I think swapping is due to either lacking of memory or settings of the job. CASTEP can run in speed option or momory option. In speed mode, calculation will cost relative high memory but faster, while memory mode, the calculation will cost additional I/O to the hard-disk with much slower run. Does Q-E has the same function? I wonder whether the disk_io and verboity command in &CONTROL section has the similar function as CASTEP does? Where could I find the detail explaination of all the command in Q-E? I can't find the document with all the command illustrated. Where is it? Also ccording to my CASTEP experiences, if the temporary file is stored in a NFS space, it will suffer a lot from I/O data and result in a long time of CPU waiting. The best CASTEP I/O wait percentage showed by 'sar' command is about 0.01. But the Q-E is about 0.20 and 0.30 occasionally, I think this probably the reason for less efficiency. Another thing I want to ask is how to place the temporary file correctly. There are two types of temporary file,outdir and wfcdir. Which one should be placed at NFS space for all the node can read and write? Which one should be placed at local disk to avoid additional unnecessary communication I/O data via network. Some temporary data is very important for further analysis such as DOS, pDOS, Charge et al. Whether I set the wf_collect=yes in &CONTROL the information for DOS,pDOS, Charge won't be lost? All the information for DoS analysis et al. is in the folder of ~/outdir/prefix.save/? best wishes, Vega Lew PH.D Candidate in Chemical Engineering State Key Laboratory of Materials-oriented Chemical Engineering College of Chemistry and Chemical Engineering Nanjing University of Technology, 210009, Nanjing, Jiangsu, China _________________________________________________________________ Discover the new Windows Vista http://search.msn.com/results.aspx?q=windows+vista&mkt=en-US&form=QBRE -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080627/69cf49d6/attachment-0001.htm From wangxinquan at tju.edu.cn Fri Jun 27 11:08:51 2008 From: wangxinquan at tju.edu.cn (wangxinquan) Date: Fri, 27 Jun 2008 17:08:51 +0800 Subject: [Pw_forum] EPR g tensor calculation Message-ID: <414557744.02007@tju.edu.cn> Dear all: I am about to calculate g tensor of EPR using espresso package,but I do not know the keywords of the input file how to obtain (Only NMR examples for gipaw.x).Is there some examples? Thank you for your attention! Yuan Wang School of Chemical Eng. & Techn., Tianjin Univ. E-mail: only1xiaoyuan at tju.edu.cn -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080627/0d628cb2/attachment.htm From sathya_sheela1985001 at rediffmail.com Fri Jun 27 11:51:12 2008 From: sathya_sheela1985001 at rediffmail.com (sathya subramanyan) Date: 27 Jun 2008 09:51:12 -0000 Subject: [Pw_forum] negative value of polarization and Z* for Pb in Tetragonal PbTiO3 and discrepancy in calculation of Message-ID: <20080627095112.30200.qmail@f4mail-235-129.rediffmail.com> ? Dear Developers, I am trying to calculate the Berry Phaze Polarization and effective charge on Pb(with the displacement of .01*ao) for PbTiO3(tetragonal) but am getting a negative value, using quantum espresso ver 4.0. When I run the example 10 for cubic phase the polarization and the Z* comes out to be alright. But, when I try for tetragonal phase, I am getting negative ionic phase leading to negative polarization for Pb. If I compare the out put file for both cubic and tertragonal phase, to my surprise in the cubic phase the total ionic phase comes out to be positive (actually, if you sum it is negative), where as in the case of tetragonal phase the ionic phase comes out to be negative (but, it is positive if you sum the same way as for the cubic). I do not understand this differrent summing rule for different phases. I am attaching both the input and out files for cubic and tetragonal phases. I would appreciate any useful suggestions to clear this point. I am very sorry for the long mail and data files. INPUT FILE FOR CUBIC PbTiO34 &control calculation = 'nscf' pseudo_dir = '/home/nbaski/espresso-4.0/pseudo/' outdir = '/home/nbaski/tmp/' lberry = .true. gdir = 3 nppstr = 7 / &system ibrav = 1 celldm(1) = 7.3699 nat = 5 ntyp = 3 nbnd = 22 ecutwfc = 30.0 occupations = 'fixed' degauss = 0.00 / &electrons conv_thr = 1e-5 mixing_beta = 0.3 / ATOMIC_SPECIES Pb 207.2 Pb.vdb.UPF Ti 47.867 Ti.vdb.UPF O 15.9994 O.vdb.UPF ATOMIC_POSITIONS Pb 0.000 0.000 0.010 Ti 0.500 0.500 0.500 O 0.000 0.500 0.500 O 0.500 0.500 0.000 O 0.500 0.000 0.500 K_POINTS {automatic} 4 4 7 1 1 1 THE O/P FILE FOR CUBIC PbTiO3 (as it is a very big file am attaching only the ionic phase calculation section) ================================================== POLARIZATION CALCULATION !!! NOT THOROUGHLY TESTED !!! -------------------------------------------------- K-POINTS STRINGS USED IN CALCULATIONS ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ G-vector along string (2 pi/a): 0.00000 0.00000 1.00000 Modulus of the vector (1/bohr): 0.85255 Number of k-points per string: 7 Number of different strings : 3 IONIC POLARIZATION ~~~~~~~~~~~~~~~~~~ Note: (mod 1) means that the phases (angles ranging from -pi to pi) have been mapped to the interval [-1/2,+1/2) by dividing by 2*pi; (mod 2) refers to the interval [-1,+1) ============================================================================ Ion Species Charge Position Phase ---------------------------------------------------------------------------- 1 Pb 14.000 0.0000 0.0000 0.0100 0.14000 (mod 2) 2 Ti 12.000 0.5000 0.5000 0.5000 0.00000 (mod 2) 3 O 6.000 0.0000 0.5000 0.5000 -1.00000 (mod 2) 4 O 6.000 0.5000 0.5000 0.0000 0.00000 (mod 2) 5 O 6.000 0.5000 0.0000 0.5000 -1.00000 (mod 2) ---------------------------------------------------------------------------- IONIC PHASE: 0.14000 (mod 2) ============================================================================ ELECTRONIC POLARIZATION ~~~~~~~~~~~~~~~~~~~~~~~ Note: (mod 1) means that the phases (angles ranging from -pi to pi) have been mapped to the interval [-1/2,+1/2) by dividing by 2*pi; (mod 2) refers to the interval [-1,+1) ============================================================================ Spin String Weight First k-point in string Phase ---------------------------------------------------------------------------- up 1 0.250000 0.1250 0.1250 -0.5000 -0.05427 (mod 1) up 2 0.500000 0.1250 0.3750 -0.5000 -0.04876 (mod 1) up 3 0.250000 0.3750 0.3750 -0.5000 -0.05048 (mod 1) ---------------------------------------------------------------------------- down 1 0.250000 0.1250 0.1250 -0.5000 -0.05427 (mod 1) down 2 0.500000 0.1250 0.3750 -0.5000 -0.04876 (mod 1) down 3 0.250000 0.3750 0.3750 -0.5000 -0.05048 (mod 1) ---------------------------------------------------------------------------- Average phase (up): -0.05057 (mod 1) Average phase (down): -0.05057 (mod 1) ELECTRONIC PHASE: -0.10114 (mod 2) ============================================================================ SUMMARY OF PHASES ~~~~~~~~~~~~~~~~~ Ionic Phase: 0.14000 (mod 2) Electronic Phase: -0.10114 (mod 2) TOTAL PHASE: 0.03886 (mod 2) VALUES OF POLARIZATION ~~~~~~~~~~~~~~~~~~~~~~ The calculation of phases done along the direction of vector 3 of the reciprocal lattice gives the following contribution to the polarization vector (in different units, and being Omega the volume of the unit cell): P = 0.2864184 (mod 14.7398000) (e/Omega).bohr P = 0.0007155 (mod 0.0368220) e/bohr^2 P = 0.0409070 (mod 2.1051744) C/m^2 The polarization direction is: ( 0.00000 , 0.00000 , 1.00000 ) ================================================== THE I/P FILE FOR TETRAGONAL PbTiO3 &CONTROL calculation = 'nscf' pseudo_dir = '/home/nbaski/espresso-4.0/pseudo/' outdir = '/home/nbaski/tmp/' lberry = .true. gdir = 3 nppstr = 7 / &SYSTEM ibrav = 6 celldm(1) = 7.3415 celldm(3) = 1.0653 nat = 5 ntyp = 3 nbnd = 22 ecutwfc = 30.0 occupations = 'fixed' degauss = 0.00 / &ELECTRONS conv_thr = 1e-5 mixing_beta = 0.3 / ATOMIC_SPECIES Pb 207.2 Pb.vdb.UPF Ti 47.867 Ti.vdb.UPF O 15.9994 O.vdb.UPF ATOMIC_POSITIONS Pb 0.00000 0.00000 0.01000 Ti 0.50000 0.50000 0.53770 O 0.50000 0.50000 0.11180 O 0.00000 0.50000 0.61740 O 0.50000 0.00000 0.61740 K_POINTS {automatic} 4 4 7 1 1 1 O/P FILE FOR PbTiO3 ================================================== POLARIZATION CALCULATION !!! NOT THOROUGHLY TESTED !!! -------------------------------------------------- K-POINTS STRINGS USED IN CALCULATIONS ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ G-vector along string (2 pi/a): 0.00000 0.00000 0.93870 Modulus of the vector (1/bohr): 0.80338 Number of k-points per string: 7 Number of different strings : 3 IONIC POLARIZATION ~~~~~~~~~~~~~~~~~~ Note: (mod 1) means that the phases (angles ranging from -pi to pi) have been mapped to the interval [-1/2,+1/2) by dividing by 2*pi; (mod 2) refers to the interval [-1,+1) ============================================================================ Ion Species Charge Position Phase ---------------------------------------------------------------------------- 1 Pb 14.000 0.0000 0.0000 0.0100 0.13142 (mod 2) 2 Ti 12.000 0.5000 0.5000 0.5377 0.05689 (mod 2) 3 O 6.000 0.5000 0.5000 0.1118 0.62968 (mod 2) 4 O 6.000 0.0000 0.5000 0.6174 -0.52267 (mod 2) 5 O 6.000 0.5000 0.0000 0.6174 -0.52267 (mod 2) ---------------------------------------------------------------------------- IONIC PHASE: -0.22735 (mod 2) ============================================================================ ELECTRONIC POLARIZATION ~~~~~~~~~~~~~~~~~~~~~~~ Note: (mod 1) means that the phases (angles ranging from -pi to pi) have been mapped to the interval [-1/2,+1/2) by dividing by 2*pi; (mod 2) refers to the interval [-1,+1) ============================================================================ Spin String Weight First k-point in string Phase ---------------------------------------------------------------------------- up 1 0.250000 0.1250 0.1250 -0.4694 -0.30118 (mod 1) up 2 0.500000 0.1250 0.3750 -0.4694 -0.28922 (mod 1) up 3 0.250000 0.3750 0.3750 -0.4694 -0.25437 (mod 1) ---------------------------------------------------------------------------- down 1 0.250000 0.1250 0.1250 -0.4694 -0.30118 (mod 1) down 2 0.500000 0.1250 0.3750 -0.4694 -0.28922 (mod 1) down 3 0.250000 0.3750 0.3750 -0.4694 -0.25437 (mod 1) ---------------------------------------------------------------------------- Average phase (up): -0.28353 (mod 1) Average phase (down): -0.28353 (mod 1) ELECTRONIC PHASE: -0.56706 (mod 2) ============================================================================ SUMMARY OF PHASES ~~~~~~~~~~~~~~~~~ Ionic Phase: -0.22735 (mod 2) Electronic Phase: -0.56706 (mod 2) TOTAL PHASE: -0.79441 (mod 2) VALUES OF POLARIZATION ~~~~~~~~~~~~~~~~~~~~~~ The calculation of phases done along the direction of vector 3 of the reciprocal lattice gives the following contribution to the polarization vector (in different units, and being Omega the volume of the unit cell): P = -6.2130096 (mod 15.6417999) (e/Omega).bohr P = -0.0147393 (mod 0.0371074) e/bohr^2 P = -0.8426689 (mod 2.1214933) C/m^2 The polarization direction is: ( 0.00000 , 0.00000 , 1.00000 ) ================================================== Thank you in advance for any help SATHYA SHEELA.S Grad. Student Department of Physics National Institute of Technology Tiruchirapalli - 620015 India -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080627/757fde5c/attachment-0001.htm From baroni at sissa.it Fri Jun 27 12:36:05 2008 From: baroni at sissa.it (Stefano Baroni) Date: Fri, 27 Jun 2008 12:36:05 +0200 Subject: [Pw_forum] negative value of polarization and Z* for Pb in Tetragonal PbTiO3 and discrepancy in calculation of In-Reply-To: <20080627095112.30200.qmail@f4mail-235-129.rediffmail.com> References: <20080627095112.30200.qmail@f4mail-235-129.rediffmail.com> Message-ID: the sum of the all the effective charges should be zero for any structure, due dynamical charge neutrality (or acoustic sum rule, or whatever you want to call the fact that a homogeneouse electric field will not exert a net (and infinite!) force on a crystal. Vioation of this charge neutrality/acoustic sum rule are often due to numerical inaccuracy, often inaccurate k-point sampling. you may want to check the convergence of your calculations with respect to the number of k points and/or try to calculate the effective charges using linear- response theory. SB On Jun 27, 2008, at 11:51 AM, sathya subramanyan wrote: > > Dear Developers, > I am trying to calculate the Berry Phaze > Polarization and effective charge on Pb(with the displacement of . > 01*ao) for PbTiO3(tetragonal) but am getting a negative value, using > quantum espresso ver 4.0. > > When I run the example 10 for cubic phase the > polarization and the Z* comes out to be alright. But, when I try for > tetragonal phase, I am getting negative ionic phase leading to > negative polarization for Pb. > If I compare the out put file for both cubic and > tertragonal phase, to my surprise in the cubic phase the total ionic > phase comes out to be positive (actually, if you sum it is > negative), where as in the case of tetragonal phase the ionic phase > comes out to be negative (but, it is positive if you sum the same > way as for the cubic). > I do not understand this differrent summing rule for > different phases. I am attaching both the input and out files for > cubic and tetragonal phases. I would appreciate any useful > suggestions to clear this point. > I am very sorry for the long mail and data files. > ... > > Thank you in advance for any help > SATHYA SHEELA.S > Grad. Student > Department of Physics > National Institute of Technology > Tiruchirapalli - 620015 > India > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste http://www.sissa.it/~baroni / [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget 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/20080627/f9f1ec7d/attachment.htm From umari at democritos.it Fri Jun 27 12:53:08 2008 From: umari at democritos.it (umari at democritos.it) Date: Fri, 27 Jun 2008 12:53:08 +0200 Subject: [Pw_forum] cp.x and "second dimension too small" (Daniele Passerone) Message-ID: <20080627125308.ub4q7opmsgww4swo@mail.democritos.it> Dear Daniele, The error was due to a bug for cg calculations with cp , for the case of no spin multiplicity (i.e. nspin=1) with odd number of bands. Note that for performing dft calculation of metallic systems you MUST specify the total number of bands (it is the parameter "nbnd" which should be larger than nelec/2 for nspin==1). In the case of your example, you can now set nbnd to an even number or if you want to set it to an odd one you should use the latest CVS version of the code where the bug has been corrected. Best wishes, Paolo Umari, Democritos. ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program. From ceresoli at MIT.EDU Fri Jun 27 14:56:13 2008 From: ceresoli at MIT.EDU (Davide Ceresoli) Date: Fri, 27 Jun 2008 08:56:13 -0400 Subject: [Pw_forum] EPR g tensor calculation In-Reply-To: <414557744.02007@tju.edu.cn> References: <414557744.02007@tju.edu.cn> Message-ID: <4864E36D.4000605@mit.edu> wangxinquan wrote: > Dear all: > I am about to calculate g tensor of EPR using espresso package,but I > do not > know the keywords of the input file how to obtain (Only NMR examples for > gipaw.x).Is there some examples? > Thank you for your attention! > It is in the documentation (doc/INPUT_GIPAW.def): &inputgipaw job='g_tensor' .... / Provided you use gipaw-pseudopotentials, this is all! Davide -- +----------------------------------------------------------+ Davide Ceresoli DMSE 13-4084 Massachusetts Institute of Technology 77 Massachusetts Avenue Cambridge, MA 02139-4307 Phone: (617) 253-6026 Mobile: +39-347-1001570 Skype: dceresoli +----------------------------------------------------------+ From sathya_sheela1985001 at rediffmail.com Fri Jun 27 15:06:13 2008 From: sathya_sheela1985001 at rediffmail.com (sathya subramanyan) Date: 27 Jun 2008 13:06:13 -0000 Subject: [Pw_forum] negative value of polarization and Z* for Pb in Tetragonal PbTiO3 Message-ID: <20080627130613.12697.qmail@f4mail-235-220.rediffmail.com> ? Dear Sir, Thank you for your reply. I do understand that the sum of all the effective charges for a system should be zero but I would like to know why there is a discrepancy in calculating the ionic phase of the cubic( EXAMPLE10 OF ESPRESSO-4.0) and tetragonal PbTiO3 i.e. in cubic phase the total ionic phase should be -1.86 whereas it is 0.14(if it is sum of all the individual ionic phases of the constituent elements). In the tetra phase the total ionic phase is the sum of the individidual elements which is not so in cubic.I thus would like to know if this is the reason for me getting a negative value of effective charge for Pb for the displacement.I am attaching the ionic phase section for comparison.Please do enlighten me and thank you in advance FOR CUBIC PHASE IONIC POLARIZATION ~~~~~~~~~~~~~~~~~~ Note: (mod 1) means that the phases (angles ranging from -pi to pi) have been mapped to the interval [-1/2,+1/2) by dividing by 2*pi; (mod 2) refers to the interval [-1,+1) ============================================================================ Ion Species Charge Position Phase ---------------------------------------------------------------------------- 1 Pb 14.000 0.0000 0.0000 0.0100 0.14000 (mod 2) 2 Ti 12.000 0.5000 0.5000 0.5000 0.00000 (mod 2) 3 O 6.000 0.0000 0.5000 0.5000 -1.00000 (mod 2) 4 O 6.000 0.5000 0.5000 0.0000 0.00000 (mod 2) 5 O 6.000 0.5000 0.0000 0.5000 -1.00000 (mod 2) ---------------------------------------------------------------------------- IONIC PHASE: 0.14000 (mod 2) ============================================================================ FOR TETRAGONAL PbTiO3 IONIC POLARIZATION ~~~~~~~~~~~~~~~~~~ Note: (mod 1) means that the phases (angles ranging from -pi to pi) have been mapped to the interval [-1/2,+1/2) by dividing by 2*pi; (mod 2) refers to the interval [-1,+1) ============================================================================ Ion Species Charge Position Phase ---------------------------------------------------------------------------- 1 Pb 14.000 0.0000 0.0000 0.0100 0.13142 (mod 2) 2 Ti 12.000 0.5000 0.5000 0.5377 0.05689 (mod 2) 3 O 6.000 0.5000 0.5000 0.1118 0.62968 (mod 2) 4 O 6.000 0.0000 0.5000 0.6174 -0.52267 (mod 2) 5 O 6.000 0.5000 0.0000 0.6174 -0.52267 (mod 2) ---------------------------------------------------------------------------- IONIC PHASE: -0.22735 (mod 2) ============================================================================ Thanking You SATHYA SHEELA.S Grad. Student Department of Physics National Institute of Technology Tiruchirapalli - 620015 India -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080627/42b6e337/attachment-0001.htm From akohlmey at cmm.chem.upenn.edu Fri Jun 27 16:21:30 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Fri, 27 Jun 2008 10:21:30 -0400 (EDT) Subject: [Pw_forum] How to handle with the I/O wait In-Reply-To: References: Message-ID: On Fri, 27 Jun 2008, vega lew wrote: dear vega, VL> Paolo said "if a process is not running close to 100%, it is either VL> swapping, or waiting for I/O, or for memory access, or for the CPU. VL> This remind me the case of CASTEP. CASTEP is so good a software, but VL> it more suitable for commercial use. So I gived it up. According to VL> many high quality scientific paper such as nature materials Journal VL> of the American Society, Q-E is the best choice for academic using. please explain the difference between "commercial" calculations and "academic" calculations. i only know "accurate" and "inaccurate". in my experience, it is not the software that makes a good publication, but the skill of the person using it and how he or she interprets the results. VL> According to my CASTEP experiences, I think swapping is due to VL> either lacking of memory or settings of the job. CASTEP can run in VL> speed option or momory option. In speed mode, calculation will cost VL> relative high memory but faster, while memory mode, the calculation VL> will cost additional I/O to the hard-disk with much slower run. Does VL> Q-E has the same function? I wonder whether the disk_io and verboity VL> command in &CONTROL section has the similar function as CASTEP does? VL> Where could I find the detail explaination of all the command in VL> Q-E? I can't find the document with all the command illustrated. VL> Where is it? the detailed documentation is called "the source code", for less detail you can check out the quantum espresso wiki and the INPUT_XXX.def files and employ your common sense. if you feel those are not detailed enough, you are welcome to the club and encouraged to help improving it. VL> Also ccording to my CASTEP experiences, if the temporary file is VL> stored in a NFS space, it will suffer a lot from I/O data and result VL> in a long time of CPU waiting. The best CASTEP I/O wait percentage VL> showed by 'sar' command is about 0.01. But the Q-E is about 0.20 and VL> 0.30 occasionally, I think this probably the reason for less VL> efficiency. VL> VL> Another thing I want to ask is how to place the temporary file VL> correctly. There are two types of temporary file,outdir and wfcdir. VL> Which one should be placed at NFS space for all the node can read VL> and write? Which one should be placed at local disk to avoid VL> additional unnecessary communication I/O data via network. Some VL> temporary data is very important for further analysis such as DOS, VL> pDOS, Charge et al. Whether I set the wf_collect=yes in &CONTROL the VL> information for DOS,pDOS, Charge won't be lost? All the information VL> for DoS analysis et al. is in the folder of ~/outdir/prefix.save/? most of this is documented and for the rest, you can try it out. cheers, axel. VL> VL> best wishes, VL> VL> Vega Lew VL> PH.D Candidate in Chemical Engineering VL> State Key Laboratory of Materials-oriented Chemical Engineering VL> College of Chemistry and Chemical Engineering VL> Nanjing University of Technology, 210009, Nanjing, Jiangsu, China VL> _________________________________________________________________ VL> Discover the new Windows Vista VL> http://search.msn.com/results.aspx?q=windows+vista&mkt=en-US&form=QBRE -- ======================================================================= 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 democritos.it Fri Jun 27 16:35:16 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Fri, 27 Jun 2008 16:35:16 +0200 Subject: [Pw_forum] How to handle with the I/O wait In-Reply-To: References: Message-ID: <15CC96ED-829D-4A74-BF6F-9764DEF27E75@democritos.it> On Jun 27, 2008, at 8:12 , vega lew wrote: > I can't find the document with all the command illustrated. Where > is it? in Doc/ and in the wiki: http://www.quantum-espresso.org/wiki/index.php/Main_Page > Discover the new Windows Vista Learn more! no thanks --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From mazar83 at gmail.com Fri Jun 27 21:57:11 2008 From: mazar83 at gmail.com (Mark Mazar) Date: Fri, 27 Jun 2008 14:57:11 -0500 Subject: [Pw_forum] Generating Psuedopotentials Message-ID: <802dfcc30806271257o443d046l4157e6e7128de2dc@mail.gmail.com> Hello all, I am in need of an ultrasoft pbe pseudo potential for Mo. After reviewing the vanderbuilt website (http://www.physics.rutgers.edu/~dhv/uspp/), I realized that the generation process is quite complicated and that none have been constructed there. Further, the pwscf website has a Mo, but it is for pw91. Does anyone have a working ultrasoft pbe Mo psuedopotential? Or does someone know an easy way to generate one that will work with the Quantum Espresso distribution? Thanks, -- Mark Mazar Graduate Student in Materials Science Chemical Engineering and Materials Science University of Minnesota Office: (612)624-3311 From hsuhan at cems.umn.edu Fri Jun 27 23:36:14 2008 From: hsuhan at cems.umn.edu (Han Hsu) Date: Fri, 27 Jun 2008 16:36:14 -0500 (CDT) Subject: [Pw_forum] mixing_fixed_ns in LDA+U calculation Message-ID: <2273.134.84.165.15.1214602574.squirrel@webmail.cems.umn.edu> Dear Pwscf users, I am currently running some LDA+U calculations, and have a question about specifying mixing_fixed_ns in the input file. In my scf calculation, I specicied starting_ns_eigenvalue to some value I wanted, and also specifed mixing_fixed_ns trying to fix the occupation ns values for several iteratioins. However, when checking the *.scf.out file, I only see "Modify starting ns matrices according to input values" once, right after the first iteratioin is completed. Then the occupation numbers in the following iteartions are not the same as what I have already specified. I am thus wondering how I can see and be sure that the ns values were fixed for mixing_fixed_ns iterations. No matter how many times I requested, I only saw the message "Modify starting ns matrices according to input values" for once. I appreciate any reponses and hints. Thanks! Han ******************************************************* Han Hsu, Ph.D. Postdoctoral Associate Department of Chemical Engineering & Materials Science University of Minnesota http://www.cems.umn.edu/~hsuhan ******************************************************* From meheut at impmc.jussieu.fr Sat Jun 28 00:52:55 2008 From: meheut at impmc.jussieu.fr (merlin meheut) Date: Sat, 28 Jun 2008 00:52:55 +0200 Subject: [Pw_forum] Generating pseudo of Mg with ld1: how to deal with unbound states? In-Reply-To: <4861FFCC.6020404@democritos.it> References: <20080625031647.qig6b93ew4occo84@www-ext.impmc.jussieu.fr> <4861FFCC.6020404@democritos.it> Message-ID: <20080628005255.apmlln9qaokw4w8o@www-ext.impmc.jussieu.fr> Hi Paolo, > there is no such thing as an "O-sesame webpage" Sorry, I have confounded with the Atom "README" file that I found on your weppage actually: http://www.fisica.uniud.it/~giannozz/Atom/README > nothing is more permanent than "temporary" changes. It was > "temporarily removed" no less than 5 years ago. It isn't easy > to generate pseudotentials, the core correction is implemented > in a lousy way, and a simple analytical form for PP isn't any > longer as important as it used to be I followed your advice, and tried to make the pseudos otherwise. Actually, my main problem is how to deal with the 3d unbound states in Mg. I want to try the different recipies you propose to accomodate unbound states on your "Notes on Pseudopotential generation", beginning with the third one: "generate PP's on unbound states". There are several things concerning that procedure that I would like to understand. My first question is: is that procedure corresponding to the Hamann's scheme? If it is, how do you choose the outermost matching radius for the 3d orbital? I do not see a flag corresponding to it. A second, not very important question is : is that possible to plot the corresponding unbound 3d orbitals? Such as in fig.1 of Fuchs and Scheffler 1999 (Comput. Phys. Comm. 119 p67)? My third and principal question is: how do you choose the "reference energy". If it is corresponding to the Hamann's scheme, such as implemented in fhi, Fuchs and Scheffer 1999 state: "The value of eps_l should be chosen in the energy range where the valence states are expected to form bands or molecular orbitals, as a default, the program psgen employs the highest eigenvalue of the occupied states, e.g., for Al to eps_l = eps_3p" This seems reasonable, so if I understood well (which I am really not sure), for Al, in PBE, eps_l = eps_3p ~ -0.20Ry . Would it be correct to take a -0.20Ry value for eps_l? It is strange to take a negative energy for an unbound state, so maybe I have to take +0.20Ry? This would correspond to the reference energy taken for the Al pseudo in /atomic_doc/pseudo-gen/al.in, which is +0.15Ry. Is that the case? Are these two energies (eps_l and the "reference energy") the same thing? Or the opposite? In the case of Mg, 3s2 is -0.44Ry, whereas 3p0 is -0.24Ry. Since 3p is empty, the reference energy to take should be around -0.4Ry? +0.4Ry? Does that make sense? Thanks again for any hints, Best regards, Merlin Merlin Meheut Earth and Space Sciences Dept. University of California at Los Angeles USA ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program. From degironc at sissa.it Sat Jun 28 09:26:17 2008 From: degironc at sissa.it (Stefano de Gironcoli) Date: Sat, 28 Jun 2008 09:26:17 +0200 Subject: [Pw_forum] mixing_fixed_ns in LDA+U calculation In-Reply-To: <2273.134.84.165.15.1214602574.squirrel@webmail.cems.umn.edu> References: <2273.134.84.165.15.1214602574.squirrel@webmail.cems.umn.edu> Message-ID: <4865E799.9080607@sissa.it> Dear Han Hsu, I think the code behave as you want even if it does not print any reassuring message... in PW/electrons.f90 you can find the following bit of code: === IF ( lda_plus_u ) THEN ! hwf_energy = hwf_energy + eth ! IF ( iverbosity > 0 .OR. first ) CALL write_ns() ! IF ( first .AND. istep == 0 .AND. & startingpot == 'atomic' ) CALL ns_adj() IF ( iter <= niter_with_fixed_ns ) rho%ns = rhoin%ns ! END IF ==== ... so you can see that the code print the current "output" value of the ns variable (the values computed from the current estimate of the wavefunctions) but then reset it to the initial value for the number of iterations required. I'll add a line of output stating this fact in the cvs version but it should be OK. best regards, Stefano de Gironcoli -- SISSA and DEMOCRITOS Han Hsu wrote: > Dear Pwscf users, > > I am currently running some LDA+U calculations, and have a question about > specifying mixing_fixed_ns in the input file. > > In my scf calculation, I specicied starting_ns_eigenvalue to some value I > wanted, and also specifed mixing_fixed_ns trying to fix the occupation ns > values for several iteratioins. However, when checking the *.scf.out file, > I only see "Modify starting ns matrices according to input values" once, > right after the first iteratioin is completed. Then the occupation numbers > in the following iteartions are not the same as what I have already > specified. > > I am thus wondering how I can see and be sure that the ns values were > fixed for mixing_fixed_ns iterations. No matter how many times I > requested, I only saw the message "Modify starting ns matrices according > to input values" for once. > > I appreciate any reponses and hints. Thanks! > > Han > > ******************************************************* > Han Hsu, Ph.D. > Postdoctoral Associate > Department of Chemical Engineering & Materials Science > University of Minnesota > http://www.cems.umn.edu/~hsuhan > ******************************************************* > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From vegalew at hotmail.com Sat Jun 28 16:39:45 2008 From: vegalew at hotmail.com (vega lew) Date: Sat, 28 Jun 2008 22:39:45 +0800 Subject: [Pw_forum] How to using gamma point calculation with high efficiency Message-ID: Dear all, Thanks to the help of Axel Kohlmeyer and Paolo Giannozzi. I could calculate with high efficiency when using 30-40 k-points. I compile Q-E on my cluster with 10.1.015 vision of intel compilers successfully and correctly. Now my cluster can calculate very fast when calculating the structure relaxtion with 30-40 k-points. But on my cluster which has 5 quad-core CPU, I must using 20 pools to get the highest CPU usage (most of time 90%+, but it's unstable. 70%+ in average was shown by 'sar' command). Thanks to Axel's advices, I set the environmental variable OMP_NUM_THREADS=1. The CPU usage in every 5 computers was always the same case. The calculation can be achieved fast. If I using 10 or 5 pools the CPU usage can't reach that high. Is this up to snuff? After testing the lattice optimizations, another questions rises. I need to calculate the surface structure with gamma point only, because of the system composed of ~80 atoms ( scientists always calculate gamma point optimization in my area of researching). But when I calculate the surface structure with gamma point only, I couldn't use many pools. Therefore the cpu usage for gamma point calculation is coming down, about ~20% again. How could I calculation with a high cpu useage? Thank you for reading. regards, Vega Lew PH.D Candidate in Chemical Engineeringhh State Key Laboratory of Materials-oriented Chemical Engineering College of Chemistry and Chemical Engineering Nanjing University of Technology, 210009, Nanjing, Jiangsu, China _________________________________________________________________ Invite your mail contacts to join your friends list with Windows Live Spaces. It's easy! http://spaces.live.com/spacesapi.aspx?wx_action=create&wx_url=/friends.aspx&mkt=en-us -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080628/995004bd/attachment.htm From akohlmey at cmm.chem.upenn.edu Sat Jun 28 18:14:41 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sat, 28 Jun 2008 12:14:41 -0400 (EDT) Subject: [Pw_forum] How to using gamma point calculation with high efficiency In-Reply-To: References: Message-ID: On Sat, 28 Jun 2008, vega lew wrote: dear vega, [...] VL> I compile Q-E on my cluster with 10.1.015 vision of intel compilers VL> successfully and correctly. Now my cluster can calculate very fast VL> when calculating the structure relaxtion with 30-40 k-points. But on VL> my cluster which has 5 quad-core CPU, I must using 20 pools to get VL> the highest CPU usage (most of time 90%+, but it's unstable. 70%+ VL> in average was shown by 'sar' command). Thanks to Axel's advices, I VL> set the environmental variable OMP_NUM_THREADS=1. The CPU usage in VL> every 5 computers was always the same case. The calculation can be VL> achieved fast. If I using 10 or 5 pools the CPU usage can't reach VL> that high. Is this up to snuff? that just means you are not using your MPI library correctly. QE assumes that consecutive MPI task ids are on the same node. most MPI libraries decide this based on the host/machine file, you have to read your MPI documentation how to arrange it the way it should would. you've seen from my timings, that it should work quite well. i am using OpenMPI and that package has on top of using hosts in the hostfile order the option to explicitely follow a round-robin or a by-node scheme when assigning MPI tasks. VL> After testing the lattice optimizations, another questions rises. I VL> need to calculate the surface structure with gamma point only, VL> because of the system composed of ~80 atoms ( scientists always VL> calculate gamma point optimization in my area of researching). But whether or not a gamma point calculation is sufficient is not determined by what other people do, but by the physics of the problem. that reminds me, you still owe us an explanation about the difference between "commercial" and "academic" calculations. i am _very_ curious about this. VL> when I calculate the surface structure with gamma point only, I VL> couldn't use many pools. Therefore the cpu usage for gamma point VL> calculation is coming down, about ~20% again. How could I VL> calculation with a high cpu useage? you cannot. as you should have already seen from the numbers i sent you. g-space parallelization doesn't scale across gigabit ethernet using TCP/IP networking. just run the job within one node and then you can run 5 jobs at the same time. axel. VL> VL> Thank you for reading. VL> VL> regards, VL> VL> Vega Lew VL> PH.D Candidate in Chemical Engineeringhh VL> State Key Laboratory of Materials-oriented Chemical Engineering VL> College of Chemistry and Chemical Engineering VL> Nanjing University of Technology, 210009, Nanjing, Jiangsu, China VL> _________________________________________________________________ VL> Invite your mail contacts to join your friends list with Windows Live Spaces. It's easy! VL> http://spaces.live.com/spacesapi.aspx?wx_action=create&wx_url=/friends.aspx&mkt=en-us -- ======================================================================= 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 flitimein at yahoo.fr Sat Jun 28 22:40:26 2008 From: flitimein at yahoo.fr (Fatma Litimein) Date: Sat, 28 Jun 2008 22:40:26 +0200 (CEST) Subject: [Pw_forum] R-3c symmetry Message-ID: <413571.32596.qm@web23102.mail.ird.yahoo.com> Dear all, I'm a new user and I'm trying to calculate the electronic properties of rhombohedral al2o3 under R-3C symmetry group. When I optimize these atoms, the code can't find the symmetry group (No symmetry !). I suspect that I must be doing something wrong here. Below is my input file. Thanks in advance, ------------------------------------------------------------------------------------------- &control calculation = "vc-relax" , restart_mode = 'from_scratch' , outdir='$TMP_DIR/' , pseudo_dir = '$PSEUDO_DIR' , disk_io = 'default' , verbosity = 'default' , tstress = .true. , tprnfor = .true. , nstep = 55 , etot_conv_thr = 1.0E-5 , forc_conv_thr = 1.0D-4 , iprint = 1 , max_seconds = 6000 , dt = 150 , / &system ibrav =4, celldm(1) =8.98943, celldm(3) = 2.73029, nat=10, ntyp= 2, ecutwfc =25.0 / &electrons conv_thr = 1.0d-8 mixing_beta = 0.7 / &IONS / &CELL cell_dynamics = 'damp-w' , press = 0.00 , wmass = 0.00700000 , / ATOMIC_SPECIES Al 26.98 Al.vbc.UPF O 15.999 O.pw91-van_ak.UPF ATOMIC_POSITIONS {crystal} Al 0.3554 0.3554 0.3554 Al 0.1480 0.1480 0.1480 Al 0.6480 0.6480 0.6480 Al 0.8519 0.8519 0.8519 O 0.2500 0.9437 0.5562 O 0.9437 0.5562 0.2500 O 0.5562 0.2500 0.9437 O 0.7500 0.0562 0.4437 O 0.4437 0.7500 0.0562 O 0.0562 0.4437 0.7500 K_POINTS {automatic} 4 4 4 1 1 1 EOF ------------------------------------------------------------------------------- ======================================================= Fatma Litimein Computational Materials Science Laboratory, Physics Department, University of Sidi Bel Abbes Sidi Bel Abbes, 22000 - Algeria Tel:+213 48-54-95-46 Fax:+213 48-54-19-28 flitimein at yahoo.fr ======================================================= --------------------------------- Envoy? avec Yahoo! Mail. Une boite mail plus intelligente. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080628/c7d483dc/attachment.htm From vegalew at hotmail.com Sun Jun 29 04:44:17 2008 From: vegalew at hotmail.com (vega lew) Date: Sun, 29 Jun 2008 10:44:17 +0800 Subject: [Pw_forum] How to using gamma point calculation with high efficiency Message-ID: Dear Axel, > that just means you are not using your MPI library correctly. > QE assumes that consecutive MPI task ids are on the same node. > most MPI libraries decide this based on the host/machine file, > you have to read your MPI documentation how to arrange it the > way it should would. you've seen from my timings, that it should > work quite well. i am using OpenMPI and that package has on top > of using hosts in the hostfile order the option to explicitely > follow a round-robin or a by-node scheme when assigning MPI tasks. I'm using MPICH2. So before using it, I should type command ' mpdboot -n 5 -f mpd.mashines'. Maybe MPICH2 is not arrange the MPI task IDs in the way that Q-E assumes. So I'll try OPENMPI. By the way, what vision does you recommend me to install? > whether or not a gamma point calculation is sufficient is not > determined by what other people do, but by the physics of the > problem. I have realized that. Late time maybe I doesn't say it in a right way. I'm using gamma point calculation is not all because other people do so, but also many scientific papers of high impact and high quality, such as paper of American Physical Society and American Chemical Society said using 2X2X1 mesh of k points result in no large difference in energy of geometry from the gamma point only calculation. It's accepted by most of scientists do the similar job for many years. I think it's not necessary to validate it by my self because of the weakness of our calculation condition. >that reminds me, you still owe us an explanation about > the difference between 'commercial' and 'academic' calculations. > i am _very_ curious about this. I don't have enough authority to illustrate the difference beteen the two. I only saying something from my own brain. I think Q-E is more suitable for scientific research, because of its free and open-source. And another very important thing is Q-E also accepted by many resarchers who used Q-E and doing excellent calculation. Also many pieces of work published on high quality and high impact Journals, like Nature Materials, Journal of the American Chemical Society et al... But for comercial software, it cost a lot of money and the source code is not open for their copyright. So it's much more like a black box. We can't know any detail that how the results comes out. Maybe it's more accurate, but we can't believe it until we see the source code. But if we know the source code, their secret is out. As far as I know, commercial software is more suitable for industry researcher such as China National Petroleum Corporation or China Petrochemical Corporation et al. The industry resarchers is often unwilling to say whether the resarch work using commercial simulation software is accurate or not. We only see the product of the industry, but can't see what's the role of the software in the product exploitation. Over all, people especially the scientists maybe couldn't trust the commercial software until the see the source. Most scientist can't bear the big uncertain factor that the software brings to their resarching. > you cannot. as you should have already seen from the numbers > i sent you. g-space parallelization doesn't scale across gigabit > ethernet using TCP/IP networking. just run the job within one > node and then you can run 5 jobs at the same time. Do you think the Q-E can achieve ~80 atoms gamma point calculation in one computer with single quad-core CPU and 8Gmemory in less than 1 month running? Maybe it's too unbelievable task. I'll try OPENMPI later, thank you for helping. Vega Lew PH.D Candidate in Chemical Engineering State Key Laboratory of Materials-oriented Chemical Engineering College of Chemistry and Chemical Engineering Nanjing University of Technology, 210009, Nanjing, Jiangsu, China _________________________________________________________________ News, entertainment and everything you care about at Live.com. Get it now! http://www.live.com/getstarted.aspx -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080629/d1e242e3/attachment-0001.htm From baroni at sissa.it Sun Jun 29 08:00:38 2008 From: baroni at sissa.it (Stefano Baroni) Date: Sun, 29 Jun 2008 08:00:38 +0200 Subject: [Pw_forum] R-3c symmetry In-Reply-To: <413571.32596.qm@web23102.mail.ird.yahoo.com> References: <413571.32596.qm@web23102.mail.ird.yahoo.com> Message-ID: On Jun 28, 2008, at 10:40 PM, Fatma Litimein wrote: > Dear all, > I'm a new user and I'm trying to calculate the electronic > properties > of rhombohedral al2o3 under R-3C symmetry group. When I optimize > these atoms, the code can't find the symmetry group (No symmetry !). > I suspect that I must be doing something wrong here. Dear Fatma Litimein: You are right! What is wring here is that you failed to browse the archive of this mailing list, were similar issues have been discussed quite a number of times. Have a good time browsing and, please, revert to us when your problem is solved, or if you fail to solve it. Cheers - SB > Below is my input file. > Thanks in advance, > ------------------------------------------------------------------------------------------- > &control > calculation = "vc-relax" > , > restart_mode = 'from_scratch' , > outdir='$TMP_DIR/' , > pseudo_dir = '$PSEUDO_DIR' , > disk_io = 'default' , > verbosity = 'default' , > tstress = .true. , > tprnfor = .true. , > nstep = 55 , > etot_conv_thr = 1.0E-5 , > forc_conv_thr = 1.0D-4 , > iprint = 1 , > max_seconds = 6000 , > dt = 150 , > / > &system > ibrav =4, > celldm(1) =8.98943, > celldm(3) = 2.73029, > nat=10, ntyp= 2, > ecutwfc =25.0 > / > &electrons > conv_thr = 1.0d-8 > mixing_beta = 0.7 > / > &IONS > / > &CELL > cell_dynamics = 'damp-w' , > press = 0.00 , > wmass = 0.00700000 , > / > ATOMIC_SPECIES > Al 26.98 Al.vbc.UPF > O 15.999 O.pw91-van_ak.UPF > ATOMIC_POSITIONS > {crystal} > Al 0.3554 0.3554 0.3554 > Al 0.1480 0.1480 0.1480 > Al 0.6480 0.6480 0.6480 > Al 0.8519 0.8519 0.8519 > O 0.2500 0.9437 0.5562 > O 0.9437 0.5562 0.2500 > O 0.5562 0.2500 0.9437 > O 0.7500 0.0562 0.4437 > O 0.4437 0.7500 0.0562 > O 0.0562 0.4437 0.7500 > K_POINTS {automatic} > 4 4 4 1 1 1 > EOF > > ------------------------------------------------------------------------------- > > > > ======================================================= > Fatma Litimein > Computational Materials Science Laboratory, > Physics Department, University of Sidi Bel Abbes > Sidi Bel Abbes, 22000 - Algeria > Tel:+213 48-54-95-46 Fax:+213 48-54-19-28 flitimein at yahoo.fr > ======================================================= > > Envoy? avec Yahoo! Mail. > Une boite mail plus intelligente. > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste http://www.sissa.it/~baroni / [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget 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/20080629/cd453d3c/attachment.htm From baroni at sissa.it Sun Jun 29 08:18:20 2008 From: baroni at sissa.it (Stefano Baroni) Date: Sun, 29 Jun 2008 08:18:20 +0200 Subject: [Pw_forum] How to using gamma point calculation with high efficiency In-Reply-To: References: Message-ID: <82A90436-9492-42FF-8D0B-273AA5BD1902@sissa.it> On Jun 29, 2008, at 4:44 AM, vega lew wrote: > > whether or not a gamma p oint calculation is sufficient is not > > determined by what other people do, but by the physics of the > > problem. > > I have realized that. Late time maybe I doesn't say it in a right > way. I'm using gamma point calculation > is not all because other people do so, but also many scientific > papers of high impact and high quality, > such as paper of American Physical Society > and American Chemical Society even worse. i would trust more a colleague next dore (whom, at least I could tease and ask questions) than a journal, ANY journal, no matter its impact factor, publishing company, title ... > said using 2X2X1 mesh of k points > result in no large difference in energy of geometry from the gamma > point only calculation. I am afraid you do not exactly what you are talking about. who says what? what is good fo a system may be totally wrong for another, unless you know exactly WHY it is good in one case and you are sure that the same argument holds for the second. this only YOU can say, after sweating quite a lot to understand by yourself, which, I understand, you are not quite prepared to do > It's accepted > by most of scientists do the similar job for many years. I think > it's not necessary to validate it by my self > because of the weakness of our calculation condition. if you are not "strong" enough (as opposed to "weak", to use your words) to accomplish a task fairly, you simply have to choose a different task that suits your strength more ... > >that reminds me, you still owe us an explanation about > > the difference between 'commercial' and 'academic' calculations. > > i am _very_ curious about this. > > I don't have enough authority to illustrate the difference beteen th > e two. I only saying something from my > own brain. I think Q-E is more suitable for scientific research, > because of its free and open-source. And > another very important thing is Q-E also accepted by many resarchers > who used Q-E and doing excellent > calculation. Also many pieces of work published on high quality and > high impact Journals, like Nature Materials, > Journal of the American Chemical Society et al... it sounds like somebody led you to believe that the purpose of science is to publish in high-impact journals. are you sure this is really the case? > But for comercial software, it cost a lot of money and the > source code is not open for their copyright. So it's much more like > a black box. We can't know any detail that > how the results comes out. Maybe it's more accurate, demonstrabily, it is not > but we can't believe it until we see the source code. finally, we agree on at least one point If I were to describe in one sentence what makes the difference between modern science and ancient knowledge systems, it is the role played in the former by the "doubt": never trust a truth unless you can give by yourself a correct answer to a legitimate question. and reproducibly so. I am sure you are studying to become a modern scientist, don't you? regards - SB --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste http://www.sissa.it/~baroni / [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget 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/20080629/eafbe0a1/attachment-0001.htm From vegalew at hotmail.com Sun Jun 29 09:58:37 2008 From: vegalew at hotmail.com (vega lew) Date: Sun, 29 Jun 2008 15:58:37 +0800 Subject: [Pw_forum] How to using gamma point calculation with high efficiency In-Reply-To: <82A90436-9492-42FF-8D0B-273AA5BD1902@sissa.it> References: <82A90436-9492-42FF-8D0B-273AA5BD1902@sissa.it> Message-ID: Dear STEFANO BARONI, Thank your for your enlightening me about how to do a good research work. Your words are so similar to my tutorial. I'll a try my best to be a good modern scientist. Vega Lew PH.D Candidate in Chemical Engineering State Key Laboratory of Materials-oriented Chemical Engineering College of Chemistry and Chemical Engineering Nanjing University of Technology, 210009, Nanjing, Jiangsu, China _________________________________________________________________ Connect to the next generation of MSN Messenger? http://imagine-msn.com/messenger/launch80/default.aspx?locale=en-us&source=wlmailtagline From alessandro.bencini at unifi.it Sun Jun 29 11:30:04 2008 From: alessandro.bencini at unifi.it (Alessandro Bencini) Date: Sun, 29 Jun 2008 11:30:04 +0200 Subject: [Pw_forum] Pseudopotentials Message-ID: <4867561C.80607@unifi.it> Hi All, I'm new to this mailing list. I' trying to run QE calculation for magnetic structure of transition metal clusters.Has anyone a working example of input for ld1.x to generate a pseudopotential for copper? Have you also some comments on the importance of including l=3 wavefunctions for transition metals? Thanks, Alessandro -- ___________________________________________________ Alessandro Bencini Dipartimento di Chimica Polo Scientifico dell'Universit? di Firenze via della Lastruccia, 3 50019 Sesto Fiorentino (FI) tel: 00390554573271 fax: 00390554573372 From janos.kiss at theochem.ruhr-uni-bochum.de Sun Jun 29 13:40:51 2008 From: janos.kiss at theochem.ruhr-uni-bochum.de (Janos Kiss) Date: Sun, 29 Jun 2008 13:40:51 +0200 Subject: [Pw_forum] How to using gamma point calculation with high efficiency In-Reply-To: References: Message-ID: <200806291340.51526.janos.kiss@theochem.ruhr-uni-bochum.de> Dear Vega Lew, >I compile Q-E on my cluster with 10.1.015 vision of intel compilers >successfully and correctly. Now my cluster can calculate >very fast when calculating the structure relaxtion with 30-40 k-points. But >on my cluster which has 5 quad-core CPU, I must >using 20 pools to get the highest CPU usage (most of time 90%+, but it's >unstable. 70%+ in average was shown by 'sar' >command). Does this mean, that you have single socket quad core CPU/machines, and you have 5 nodes in your cluster? What is the interconnect between the machines? I'm a novice user of PWSCF by myself, but as far as i understood (actually it is made relative clear in the manual) if you do a calculation with K-points, than you have two ways of parallelization: One over the G-space, and one additional over k-points. I show you my own example how this cold be exploited: I use dual socket Xeon machines, each CPU have 4 CPU cores (8 CPU cores/machine). The nodes (machines) are communicating via Gigabit LAN. The mesh in the z direction of my supercell is having 180 points. If you look into your output file, at the beginning you will see something like: Planes per process (thick) : nr3 =180 npp = 30 ncplane = 6400 Proc/ planes cols G planes cols G columns G Pool (dense grid) (smooth grid) (wavefct grid) 1 30 711 83109 30 711 83109 189 11381 2 30 711 83111 30 711 83111 189 11379 3 30 711 83109 30 711 83109 190 11386 4 30 711 83111 30 711 83111 189 11377 5 30 712 83114 30 712 83114 189 11381 6 30 711 83107 30 711 83107 189 11377 0 180 4267 498661 180 4267 498661 1135 68281 This means that i use the parallelization over the G-space with 6 CPU cores (this is dividing 180 into 30) in each machine. Each CPU core is calculating 30 z-planes. The communication is more crucial here than between K-points, so it would be good if you look it up in your output how many CPU cores you can use from a single machine to be appropriate for your mesh. My supercell is relative large, so with six K-points the binding energy for my setup is well converged. Therefore, i set npool=3. This mean, that i use the parallelization over the K-points (over three separate machines, using from each machine 6 CPU for the G-space parallelization). The Gigabit LAN is acceptable for the communication what you have between the machines for K-point parallelization. >Thanks to Axel's advices, I set the environmental variable >OMP_NUM_THREADS=1. The CPU usage in every 5 computers was always the same >case. The calculation can be achieved fast. >If I using 10 or 5 pools the CPU usage can't reach that high. Is this up to >snuff? Now please have a look on what should be the right value for your setup for G-space/CPU cores and K-points/machines. Hopefully your 'CPU usage' gonna get better. >After testing the lattice optimizations, another questions rises. I need to >calculate the surface structure with gamma point only, because of the system >composed of ~80 atoms ( scientists always calculate gamma point optimization >in my area of researching). This depends on your system again (and that quantity what you are interested in). You have to keep in mind, that doing calculations with proper K-point sampling can be several times more expensive than a gamma point only calculations. If you get good enough results with Gamma point only with a larger supercell, than it is a happy case. If your system does not have a wide-band gap (like metallic systems and so on) the Gamma point only results are just far off, and you are forced to do K-point sampling. >But when I calculate the surface structure with >gamma point only, I couldn't use many pools. Therefore the cpu usage for >gamma point calculation is coming down, about ~20% again. How could I >calculation with a high cpu useage? If you do Gamma point only, than you can exploit the parallelization over G-space only. If you wanna use the G-space parallelization over the whole cluster than as i mentioned you need a fast interconnect (infiniband, myrinet, SCI, quadrics and so on). Gigabit LAN is just too slow for that (except for insanely large supercells). Even if you do have a fast interconnect, you still better keep an eye on the right number of CPUs for the mesh. All the best, Janos. ================================================================== Janos Kiss e-mail: janos.kiss at theochem.ruhr-uni-bochum.de Lehrstuhl fuer Theoretische Chemie Phone: +49 (0)234/32-26485 NC 03/297 +49 (0)234 32 26754 Ruhr-Universitaet Bochum Fax: +49 (0)234/32-14045 D-44780 Bochum http://www.theochem.ruhr-uni-bochum.de ================================================================== From vegalew at hotmail.com Sun Jun 29 15:21:43 2008 From: vegalew at hotmail.com (vega lew) Date: Sun, 29 Jun 2008 21:21:43 +0800 Subject: [Pw_forum] How to using gamma point calculation with high efficiency In-Reply-To: <200806291340.51526.janos.kiss@theochem.ruhr-uni-bochum.de> References: <200806291340.51526.janos.kiss@theochem.ruhr-uni-bochum.de> Message-ID: > Does this mean, that you have single socket quad core CPU/machines, and you > have 5 nodes in your cluster? What is the interconnect between the machines? Yes, my mashines is a single socket on system board. My cpu is intel Core TM II Q6600. The network between my cluster is only 1Gbit ethernet using TCP/IP. > Now please have a look on what should be the right value for your setup for > G-space/CPU cores and K-points/machines. Hopefully your 'CPU usage' gonna get > better. It's very nice of you saying so. Do your have ever run calculation with the gamma point only via 1Gbit LAN? What about the efficiency? What about the CPU time ? Wall time? How large the system was? > This depends on your system again (and that quantity what you are interested > in). You have to keep in mind, that doing calculations with proper K-point > sampling can be several times more expensive than a gamma point only > calculations. > If you get good enough results with Gamma point only with a larger supercell, > than it is a happy case. If your system does not have a wide-band gap (like > metallic systems and so on) the Gamma point only results are just far off, > and you are forced to do K-point sampling. My most interested system is the Titanium Oxide surface with 1~3 noble metal atom (too many is more interested, but DFT can't handle this. I used to calculate from atomic simulation but the force field is not accurate enough) I think K-point is enough for my ~80 atoms system. But Stefano Baroni tell me 'never trust a truth, unless you can give by yourself a correct answer to a legitimate question'. I think this is quite right. So I'll try 2X2X1 and make a energy and geometry comparison with the gamma point only. > If you do Gamma point only, than you can exploit the parallelization over > G-space only. If you wanna use the G-space parallelization over the whole > cluster than as i mentioned you need a fast interconnect (infiniband, > myrinet, SCI, quadrics and so on). Gigabit LAN is just too slow for that > (except for insanely large supercells). How could I exploit the parallelization over G-space? Only by use the the proper number of CPU? no pools can be used? What's the exact meaning of Gigabit LAN is just too slow for that (except for insanely large supercells)? Do you mean I could calculate with relative high efficiency by using large systems? How many atoms do you call it 'large' supercells? It's very kind of you to share your experience of PWscf. Thank you so much for the detail discussion. By the way, are you a German? I like German football team. Let's expect the final of Euro 2008 and the German's excellent show, even though, Ballack maybe absence. Come on~ best wishes. Vega Lew PH.D Candidate in Chemical Engineering State Key Laboratory of Materials-oriented Chemical Engineering College of Chemistry and Chemical Engineering Nanjing University of Technology, 210009, Nanjing, Jiangsu, China _________________________________________________________________ News, entertainment and everything you care about at Live.com. Get it now! http://www.live.com/getstarted.aspx -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080629/c6f3faea/attachment.htm From giannozz at democritos.it Mon Jun 30 09:27:30 2008 From: giannozz at democritos.it (Paolo Giannozzi) Date: Mon, 30 Jun 2008 09:27:30 +0200 Subject: [Pw_forum] Pseudopotentials In-Reply-To: <4867561C.80607@unifi.it> References: <4867561C.80607@unifi.it> Message-ID: <84C0B1D0-45AA-477C-A036-C8327A005339@democritos.it> On Jun 29, 2008, at 11:30 , Alessandro Bencini wrote: > Has anyone a working example of input for ld1.x to generate a > pseudopotential for copper? in espresso/atomic_doc/pseudo_library there isn't unfortunately any example for Copper, but there are several for Ag that you could use as a starting point > Have you also some comments on the importance of including l=3 > wavefunctions for transition metals? if you do not need it, don't include it 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 lathiot at googlemail.com Mon Jun 30 11:28:38 2008 From: lathiot at googlemail.com (Nektarios Lathiotakis) Date: Mon, 30 Jun 2008 12:28:38 +0300 Subject: [Pw_forum] Small U for Co in ZnO for LDA+U Message-ID: <65c086de0806300228r141ca064t3b589c3f3a699ea9@mail.gmail.com> Hi pwscf guys, I have done some calculations on Co impurities in ZnO. I tried to fit U_Co in such a way that the caluclated Co-d projected DOS to fit experiment (J. Appl. Phys. 103 07D130 (2008)). In this paper, they also fit U_Co. They use VASP. The problem is that the optimal U for pwscf and VASP differ quite a lot. I found a projected DOS similar to the one in the above paper for U_Co = 0.5eV which is much smaller than their value (5.5eV with VASP). If I use their value, I get a Co-d projected DOS far deep into the valence (completely non sense). Any idea why the optimal U values for Co in this case are so different? Unfortunately, their value is more common (most transition metals around 5eV). Is it the different implementations for VASP and pwscf? Pseudopotential? I have already tried to publish our results but the referees can not accept the publication based on such a small U_Co value. Interestingly, my fitted value is very similar to the one determined self consistently by Hu et al, PRB 73, 245205 (2006) for the same system (0.35 eV). However, a referee thinks that these small U numbers are "eroneous". I include a typical input file with just 8 k points. Thanks Nektarios --------INPUT------------------------------------ &control title = 'zno', calculation = 'scf', nstep = 200 restart_mode = 'from_scratch', prefix = 'zno', pseudo_dir = '/usr/espresso-4.0/pseudo/', outdir = '/home/lat/tmp/', etot_conv_thr = 1.d-5, wf_collect = .true. / &system ibrav = 0, celldm(1) = 12.4172 , nat = 16, ntyp = 3, ecutwfc = 50.0, nosym = .true. , nbnd = 80 , occupations = 'smearing', degauss = 1.d-3, starting_magnetization(3)=1, nspin=2 lda_plus_u=.true. Hubbard_U(1)=6.5 Hubbard_U(2)=0.0 Hubbard_U(3)=0.5 / &electrons electron_maxstep = 400, diagonalization = 'david', mixing_mode = 'plain', mixing_beta = 0.3, conv_thr = 1.0d-6 / &IONS pot_extrapolation = "second_order", wfc_extrapolation = "second_order", / ATOMIC_SPECIES Zn 65.38000 Zn.pbe-van.UPF O 15.99900 O.pbe-rrkjus.UPF Co 58.93320 Co.pbe-nd-rrkjus.UPF CELL_PARAMETERS 0.500000 0.866025 0.000000 .5000000 -.8660250 0.000000 0.000000 0.000000 .8050500 ATOMIC_POSITIONS crystal Zn .1666666 .3333333 .0000000 0 0 0 O .1666666 .3333333 .3792000 0 0 0 Zn .3333333 .1666666 .5000000 0 0 0 O .3333333 .1666666 .8792000 0 0 0 Zn .6666666 .3333333 .0000000 0 0 0 O .6666666 .3333333 .3792000 0 0 0 Zn .8333333 .1666666 .5000000 0 0 0 O .8333333 .1666666 .8792000 0 0 0 Zn .1666666 .8333333 .0000000 0 0 0 O .1666666 .8333333 .3792000 0 0 0 Co .3333333 .6666666 .5000000 0 0 0 O .3333333 .6666666 .8792000 0 0 0 Zn .6666666 .8333333 .0000000 0 0 0 O .6666666 .8333333 .3792000 0 0 0 Zn .8333333 .6666666 .5000000 0 0 0 O .8333333 .6666666 .8792000 0 0 0 K_POINTS automatic 2 2 2 0 0 0 -- -------------------------------------------------------------------- Dr. Nektarios N. Lathiotakis -------------------------------------------------------------------- Theoretical and Physical Chemistry Institute National Hellenic Research Foundation Vass. Constantinou 48, GR-11635, Athens, Greece -------------------------------------------------------------------- Tel: +30 210 7273805 FAX: +30 210 7273794 Inst. secretary: +30 210 7273792, +30 210 7273795 -------------------------------------------------------------------- Web: http:www.eie.gr/nhrf/institutes/tpci/cvs/cv-lathiotakis-en.html -------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080630/cfed6070/attachment.htm From marzari at MIT.EDU Mon Jun 30 12:18:00 2008 From: marzari at MIT.EDU (Nicola Marzari) Date: Mon, 30 Jun 2008 12:18:00 +0200 Subject: [Pw_forum] Small U for Co in ZnO for LDA+U In-Reply-To: <65c086de0806300228r141ca064t3b589c3f3a699ea9@mail.gmail.com> References: <65c086de0806300228r141ca064t3b589c3f3a699ea9@mail.gmail.com> Message-ID: <4868B2D8.6090401@mit.edu> Dear Nektarios, two comments: 1) U is indeed pseudopotential dependent, especially if the pseudos have been generated in different oxidation states. I.e. there is not a unique U value - our tests in molecules show that different pseudos give different U but the same chemistry - in solids the situations is less clear cut, and not fully clear to me at this stage. Note also that the U of pwscf is a U-J. 2) a U of 0.5 eV seems indeed small - i.e. does it have a significant effect with respect to U=0 ? If it has, there must be something very subtle going on. 3) tell the referee to get lost. nicola Nektarios Lathiotakis wrote: > Hi pwscf guys, > > I have done some calculations on Co impurities in ZnO. I tried to fit > U_Co in such a > way that the caluclated Co-d projected DOS to fit experiment > (J. Appl. Phys. 103 07D130 (2008)). In this paper, they also fit U_Co. > They use VASP. The problem is that the optimal U for pwscf and VASP differ > quite a lot. > > I found a projected DOS similar to the one in the above paper for > U_Co = 0.5eV which is much smaller than their value (5.5eV with VASP). > If I use their value, I get a Co-d projected DOS far deep into the valence > (completely non sense). > > Any idea why the optimal U values for Co in this case are so different? > Unfortunately, their value is more common (most transition metals around > 5eV). > Is it the different implementations for VASP and pwscf? Pseudopotential? > > I have already tried to publish our results but the referees can not > accept the publication > based on such a small U_Co value. > > Interestingly, my fitted value is very similar to the one determined > self consistently by > Hu et al, PRB 73, 245205 (2006) for the same system (0.35 eV). However, > a referee thinks that > these small U numbers are "eroneous". > > I include a typical input file with just 8 k points. > > Thanks > Nektarios -- --------------------------------------------------------------------- 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 lathiot at googlemail.com Mon Jun 30 19:27:58 2008 From: lathiot at googlemail.com (Nektarios Lathiotakis) Date: Mon, 30 Jun 2008 20:27:58 +0300 Subject: [Pw_forum] Small U for Co in ZnO for LDA+U (update) Message-ID: <65c086de0806301027v4db15423p30f02e8e4b223055@mail.gmail.com> Hi again, As I wrote previously I have a Co impurity atom in ZnO and trying to find the right Hubbard-U for Co by fitting to experimental Co-d projected band from J. Appl. Phys. 103 07D130 (2008). A typical input file is in my previous e-mail. I played with U_projection_type parameter and: 1) 'ortho-atomic' and 'norm-atomic' agree and give the same optimal U for Co that is around 4eV. Very reasonable value and no referee would ever object. 2) 'atomic' (the default) gives as optimal U 0.5eV which is much smaller. In other words the Co-d projected band you get with U_projection_type='ortho-atomic' and U_Co = 4 eV is the very close to the band you get with U_projection_type='atomic' and U_Co = 0.5 eV. Why is that difference? The problem with changing the U_projection_type is that only atomic supports forces. Thanks in advance and also to Nicola for his previous reply. Nektarios -- -------------------------------------------------------------------- Dr. Nektarios N. Lathiotakis -------------------------------------------------------------------- Theoretical and Physical Chemistry Institute National Hellenic Research Foundation Vass. Constantinou 48, GR-11635, Athens, Greece -------------------------------------------------------------------- Tel: +30 210 7273805 FAX: +30 210 7273794 Inst. secretary: +30 210 7273792, +30 210 7273795 -------------------------------------------------------------------- Web: http:www.eie.gr/nhrf/institutes/tpci/cvs/cv-lathiotakis-en.html -------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080630/63f42bcd/attachment.htm From marzari at MIT.EDU Mon Jun 30 19:37:58 2008 From: marzari at MIT.EDU (Nicola Marzari) Date: Mon, 30 Jun 2008 19:37:58 +0200 Subject: [Pw_forum] Small U for Co in ZnO for LDA+U (update) In-Reply-To: <65c086de0806301027v4db15423p30f02e8e4b223055@mail.gmail.com> References: <65c086de0806301027v4db15423p30f02e8e4b223055@mail.gmail.com> Message-ID: <486919F6.1030500@mit.edu> Dear Nektarios, thanks for the update. Maybe Matteo Cococcioni or other could chip in - could it be that the atomic orbital in the Co psp was not normalized, so only when ortho-atomic or norm-atomic are applied the normalization constant is taken into account ? If the case, is it correct to say that norm-atomic and atomic are doing the same thing, but norm-atomic automatically normalizes the projection (something that should have been there in the Co psp to begin with) ? Best, Nicola Nektarios Lathiotakis wrote: > Hi again, > > As I wrote previously I have a Co impurity atom in ZnO and > trying to find the right Hubbard-U for Co by fitting to experimental Co-d > projected band from J. Appl. Phys. 103 07D130 (2008). A typical input > file is in my > previous e-mail. > > I played with U_projection_type parameter and: > > 1) 'ortho-atomic' and 'norm-atomic' agree and give the same optimal U for Co > that is around 4eV. Very reasonable value and no referee would ever object. > > 2) 'atomic' (the default) gives as optimal U 0.5eV which is much smaller. > > In other words the Co-d projected band you get with > U_projection_type='ortho-atomic' > and U_Co = 4 eV is the very close to the band you get with > U_projection_type='atomic' > and U_Co = 0.5 eV. > > Why is that difference? The problem with changing the U_projection_type > is that only > atomic supports forces. > > Thanks in advance and also to Nicola for his previous reply. > Nektarios -- --------------------------------------------------------------------- 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 matteo at umn.edu Mon Jun 30 20:04:47 2008 From: matteo at umn.edu (Matteo Cococcioni) Date: 30 Jun 2008 13:04:47 -0500 Subject: [Pw_forum] Small U for Co in ZnO for LDA+U (update) In-Reply-To: <486919F6.1030500@mit.edu> References: <65c086de0806301027v4db15423p30f02e8e4b223055@mail.gmail.com> <486919F6.1030500@mit.edu> Message-ID: Dear Nektarios, I was going to write but Nicola has been faster than me. Nicola is right: probably the reason why "ortho-atomic" and "norm-atomic" produce similar results is that the atomic wavefunctions in the Co PP are not normalized. Try to check the occupations in the "atomic" case. are they much bigger than 1? Ortho-atomic: atomic wfcs are orthogonalized with those belonging to different atoms. atomic: wfcs are orthogonal (by construction) only to those belonging to the same atom (there are overlap regions in between atoms) norm-atomic: quick and dirty fix added to use PPs whose wfcs are not normalized. when these wfcs are normalized it gives the same results as "atomic". In general I would suggest to not look at the value of U itself but rather to the results it produces. I just found a paper: PRL 100, 256401 (2008) where they use U=2 for Co and 5eV for Zn (with vasp). Why don't you actually try to compute the U (PRB 71, 35105 (2005))? hope this helps. Matteo On Jun 30 2008, Nicola Marzari wrote: > >Dear Nektarios, > > >thanks for the update. Maybe Matteo Cococcioni or other could chip in - >could it be that the atomic orbital in the Co psp was not normalized, >so only when ortho-atomic or norm-atomic are applied the normalization >constant is taken into account ? > >If the case, is it correct to say that norm-atomic and atomic are doing >the same thing, but norm-atomic automatically normalizes the projection >(something that should have been there in the Co psp to begin with) ? > > >Best, > > Nicola > > >Nektarios Lathiotakis wrote: >> Hi again, >> >> As I wrote previously I have a Co impurity atom in ZnO and >> trying to find the right Hubbard-U for Co by fitting to experimental Co-d >> projected band from J. Appl. Phys. 103 07D130 (2008). A typical input >> file is in my >> previous e-mail. >> >> I played with U_projection_type parameter and: >> >> 1) 'ortho-atomic' and 'norm-atomic' agree and give the same optimal U >> for Co that is around 4eV. Very reasonable value and no referee would >> ever object. >> >> 2) 'atomic' (the default) gives as optimal U 0.5eV which is much smaller. >> >> In other words the Co-d projected band you get with >> U_projection_type='ortho-atomic' >> and U_Co = 4 eV is the very close to the band you get with >> U_projection_type='atomic' >> and U_Co = 0.5 eV. >> >> Why is that difference? The problem with changing the U_projection_type >> is that only >> atomic supports forces. >> >> Thanks in advance and also to Nicola for his previous reply. >> Nektarios > > -- %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Matteo Cococcioni Department of Chemical Engineering and Materials Science University of Minnesota 151 Amundson Hall 421 Washington Av. SE Minneapolis, MN 55455 Tel. +1 612 624 9056 Fax +1 612 626 7246 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% From matteo at umn.edu Mon Jun 30 19:53:59 2008 From: matteo at umn.edu (Matteo Cococcioni) Date: Mon, 30 Jun 2008 12:53:59 -0500 Subject: [Pw_forum] Small U for Co in ZnO for LDA+U (update) In-Reply-To: <65c086de0806301027v4db15423p30f02e8e4b223055@mail.gmail.com> References: <65c086de0806301027v4db15423p30f02e8e4b223055@mail.gmail.com> Message-ID: <48691DB7.6000101@umn.edu> Dear Nektarios, I was going to write but Nicola has been faster than me. Nicola is right: probably the reason why "ortho-atomic" and "norm-atomic" produce similar results is that the atomic wavefunctions in the Co PP are not normalized. Try to check the occupations in the "atomic" case. are they much bigger than 1? Ortho-atomic: atomic wfcs are orthogonalized with those belonging to different atoms. atomic: wfcs are orthogonal only to those belonging to the same atom (there are overlap regions in between atoms) norm-atomic: quick and dirty fix added to use PPs whose wfcs are not normalized. when these wfcs are normalized it gives the same results as "atomic". In general I would suggest to not look at the value of U itself but rather to the results it produces. I just found a paper: PRL 100, 256401 (2008) where they use U=2 for Co and 5eV for Zn (with Vasp). Why don't you actually try to compute it (PRB 71, 35105 (2005))? hope this helps. Matteo Nektarios Lathiotakis wrote: > Hi again, > > As I wrote previously I have a Co impurity atom in ZnO and > trying to find the right Hubbard-U for Co by fitting to experimental Co-d > projected band from J. Appl. Phys. 103 07D130 (2008). A typical input > file is in my > previous e-mail. > > I played with U_projection_type parameter and: > > 1) 'ortho-atomic' and 'norm-atomic' agree and give the same optimal U > for Co > that is around 4eV. Very reasonable value and no referee would ever > object. > > 2) 'atomic' (the default) gives as optimal U 0.5eV which is much smaller. > > In other words the Co-d projected band you get with > U_projection_type='ortho-atomic' > and U_Co = 4 eV is the very close to the band you get with > U_projection_type='atomic' > and U_Co = 0.5 eV. > > Why is that difference? The problem with changing the > U_projection_type is that only > atomic supports forces. > > Thanks in advance and also to Nicola for his previous reply. > Nektarios > > > > -- > -------------------------------------------------------------------- > Dr. Nektarios N. Lathiotakis > -------------------------------------------------------------------- > Theoretical and Physical Chemistry Institute > National Hellenic Research Foundation > Vass. Constantinou 48, GR-11635, > Athens, Greece > -------------------------------------------------------------------- > Tel: +30 210 7273805 FAX: +30 210 7273794 > Inst. secretary: +30 210 7273792, +30 210 7273795 > -------------------------------------------------------------------- > Web: http:www.eie.gr/nhrf/institutes/tpci/cvs/cv-lathiotakis-en.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: matteo.vcf Type: text/x-vcard Size: 294 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20080630/551db06f/attachment.vcf From akohlmey at cmm.chem.upenn.edu Mon Jun 30 21:22:19 2008 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Mon, 30 Jun 2008 15:22:19 -0400 (EDT) Subject: [Pw_forum] How to using gamma point calculation with high efficiency In-Reply-To: References: <200806291340.51526.janos.kiss@theochem.ruhr-uni-bochum.de> Message-ID: On Sun, 29 Jun 2008, vega lew wrote: VL> VL> > Does this mean, that you have single socket quad core CPU/machines, and you VL> > have 5 nodes in your cluster? What is the interconnect between the machines? VL> VL> Yes, my mashines is a single socket on system board. My cpu is intel Core TM II Q6600. VL> The network between my cluster is only 1Gbit ethernet using TCP/IP. just for the sake of completeness. it is not the bandwidth of GigE networking that is limiting, but the overhead of TCP/IP networking. for g-space parallelization you have a lot of messages (k-point parallelization is more of the embarrassingly parallel type of problem) and the high latency time of sending TCP/IP encoded data is killing the performance. you would see a similar behavior with with running TCP/IP over myrinet or infiniband. OTOH, there are attempts to implement MPI libraries that use GigE hardware, but bypass the TCP/IP encoding (you need a second set of network card and a second switch for that), but those are generally only working for a specific hardware/software combination due to lack of manpower. none of the (big) companies is supporting those efforts, since that would impact their business of selling you the more expensive high performance hardware. as in (interesting?) aside. the myrinet MX hardware is actually 10Gigabit ethernet. the magic is all in the software... ;-). [...] VL> My most interested system is the Titanium Oxide surface with 1~3 VL> noble metal atom (too many is more interested, but DFT can't handle what do you mean by DFT cannot handle this? you should have learned by now that making general statements like this without any backing can be very "dangerous". ;-) [...] VL> How could I exploit the parallelization over G-space? Only by use VL> the the proper number of CPU? no pools can be used? please _read_ the documentation that you have been asked to read now several times. g-space parallelization is the default for pw.x. you don't have to turn it on. VL> What's the exact meaning of Gigabit LAN is just too slow for that VL> (except for insanely large supercells)? Do you mean I could VL> calculate with relative high efficiency by using large systems? How VL> many atoms do you call it 'large' supercells? before discussing any of these issues any further, you should inform yourself about parallel efficiency in particular and parallelization in general. it is pretty pointless to discuss details, if you don't know the basics well. the following is a "for dummies"-style description that will hopefully get you started looking into the right direction. TCP/IP networking incurs communication latencies. the g-space parallelization distributes data across the MPI tasks, i.e. the more nodes you use, the less work you have to do on each node. at the same time the average wait time for a parallel message across the machine goes up (since you have to contact more nodes). when this wait time is more than the time spent on processing the data, you have reached the limit of scaling. if you make your problem bigger (more data), you reach the limit of scaling later, but since the physics of the problem does not scale linearly with the problem set size, your calculation goes in total slower. you only waste less time. in summary, if your hardware doesn't support good scaling, you don't get it. at least not unless you invent a new algorithm that needs less communication or can hide the communication latencies. this is the reason why you have to make some benchmarks first to evaluate the scaling for your problem on your machine. cheers, axel. -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From lathiot at googlemail.com Mon Jun 30 22:04:15 2008 From: lathiot at googlemail.com (Nektarios Lathiotakis) Date: Mon, 30 Jun 2008 23:04:15 +0300 Subject: [Pw_forum] Small U for Co in ZnO for LDA+U (update) In-Reply-To: <48691DB7.6000101@umn.edu> References: <65c086de0806301027v4db15423p30f02e8e4b223055@mail.gmail.com> <48691DB7.6000101@umn.edu> Message-ID: <65c086de0806301304y7a94ff6an54253c8e24688298@mail.gmail.com> Hi again, Mateo is right, the occupations for Co are 0.4 instead of 1. A last question that is a bit different... The gap of pure ZnO (no Co impurities) can not be fixed by pure lda+u because the Zn-d states are fully occupied. However, (I guess) the gap must be sensitive to a hubbard U for oxygen p-states and could be fixed by fitting U_oxygen. However, I am not aware of anyone using Hubbard U for p states... and a quick search in bibliography didnt find anything. On the other hand Oxygen is in elements list in tabd.f90. That makes me suspect that 1) It might not be completely non sense 2) there might be other attempts to do that. Any comment? Nektarios On Mon, Jun 30, 2008 at 8:53 PM, Matteo Cococcioni wrote: > > Dear Nektarios, > > I was going to write but Nicola has been faster than me. Nicola is right: > probably the reason why > "ortho-atomic" and "norm-atomic" produce similar results is that the atomic > wavefunctions in the Co PP > are not normalized. Try to check the occupations in the "atomic" case. are > they much bigger than 1? > > Ortho-atomic: atomic wfcs are orthogonalized with those belonging to > different atoms. > atomic: wfcs are orthogonal only to those belonging to the same atom (there > are overlap regions in between atoms) > norm-atomic: quick and dirty fix added to use PPs whose wfcs are not > normalized. when these wfcs are normalized it gives the same results as > "atomic". > > In general I would suggest to not look at the value of U itself but rather > to the results it produces. I just found a paper: PRL 100, 256401 (2008) > where they use U=2 for Co and 5eV for Zn (with Vasp). Why don't you > actually try to compute it (PRB 71, 35105 (2005))? > > hope this helps. > > Matteo > > > Nektarios Lathiotakis wrote: > >> Hi again, >> >> As I wrote previously I have a Co impurity atom in ZnO and >> trying to find the right Hubbard-U for Co by fitting to experimental Co-d >> projected band from J. Appl. Phys. 103 07D130 (2008). A typical input >> file is in my >> previous e-mail. >> >> I played with U_projection_type parameter and: >> >> 1) 'ortho-atomic' and 'norm-atomic' agree and give the same optimal U for >> Co >> that is around 4eV. Very reasonable value and no referee would ever >> object. >> >> 2) 'atomic' (the default) gives as optimal U 0.5eV which is much smaller. >> >> In other words the Co-d projected band you get with >> U_projection_type='ortho-atomic' >> and U_Co = 4 eV is the very close to the band you get with >> U_projection_type='atomic' >> and U_Co = 0.5 eV. >> >> Why is that difference? The problem with changing the U_projection_type is >> that only >> atomic supports forces. >> >> Thanks in advance and also to Nicola for his previous reply. >> Nektarios >> >> >> >> -- >> -------------------------------------------------------------------- >> Dr. Nektarios N. Lathiotakis >> -------------------------------------------------------------------- >> Theoretical and Physical Chemistry Institute >> National Hellenic Research Foundation >> Vass. Constantinou 48, GR-11635, >> Athens, Greece >> -------------------------------------------------------------------- >> Tel: +30 210 7273805 FAX: +30 210 7273794 >> Inst. secretary: +30 210 7273792, +30 210 7273795 >> -------------------------------------------------------------------- >> Web: http:www.eie.gr/nhrf/institutes/tpci/cvs/cv-lathiotakis-en.html < >> http://www.eie.gr/nhrf/institutes/tpci/cvs/cv-lathiotakis-en.html> >> -------------------------------------------------------------------- >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> 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 > > -- -------------------------------------------------------------------- Dr. Nektarios N. Lathiotakis -------------------------------------------------------------------- Theoretical and Physical Chemistry Institute National Hellenic Research Foundation Vass. Constantinou 48, GR-11635, Athens, Greece -------------------------------------------------------------------- Tel: +30 210 7273805 FAX: +30 210 7273794 Inst. secretary: +30 210 7273792, +30 210 7273795 -------------------------------------------------------------------- Web: http:www.eie.gr/nhrf/institutes/tpci/cvs/cv-lathiotakis-en.html -------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20080630/9ae09cf9/attachment.htm From ceresoli at MIT.EDU Mon Jun 30 22:17:23 2008 From: ceresoli at MIT.EDU (Davide Ceresoli) Date: Mon, 30 Jun 2008 16:17:23 -0400 Subject: [Pw_forum] Small U for Co in ZnO for LDA+U (update) In-Reply-To: <65c086de0806301304y7a94ff6an54253c8e24688298@mail.gmail.com> References: <65c086de0806301027v4db15423p30f02e8e4b223055@mail.gmail.com> <48691DB7.6000101@umn.edu> <65c086de0806301304y7a94ff6an54253c8e24688298@mail.gmail.com> Message-ID: <48693F53.6030900@mit.edu> Nektarios Lathiotakis wrote: > The gap of pure ZnO (no Co impurities) can not be fixed by pure lda+u > because the Zn-d states are fully occupied. However, (I guess) the gap must > be sensitive to a hubbard U for oxygen p-states and could be fixed by > fitting U_oxygen. > However, I am not aware of anyone using Hubbard U for p states... and a > quick search > in bibliography didnt find anything I know only one reference: Phys. Rev. B 74, 035112 (2006). -- +----------------------------------------------------------+ Davide Ceresoli DMSE 13-4084 Massachusetts Institute of Technology 77 Massachusetts Avenue Cambridge, MA 02139-4307 Phone: (617) 253-6026 Mobile: +39-347-1001570 Skype: dceresoli +----------------------------------------------------------+