From neuber at itp.tu-graz.ac.at Mon Dec 1 17:13:48 2003 From: neuber at itp.tu-graz.ac.at (Danilo Neuber) Date: Mon, 1 Dec 2003 17:13:48 +0100 (CET) Subject: [Pw_forum] Si band gap calculated with different functionals In-Reply-To: <200311291648.47211.giannozz@nest.sns.it> References: <200311291648.47211.giannozz@nest.sns.it> Message-ID: On Sat, 29 Nov 2003, Paolo Giannozzi wrote: > is the lattice parameter > - the same for all cases, and equal to the experimental one > - the theoretical one for each case, and they are very similar > - the theoretical one for each case, but they are not very similar > - none of the above ? Dear Mr. Giannozzi! The lattice constant was a_0 = 10.2 Bohr in all my previous calculations (according to Ashcroft & Mermin: "Solid State Physics" the experimental value is a_0 = 5.43 Angstrom = 10.26 Bohr). I did some more calculations with my PPs varying the lattice constant a_0 from 10.1 to 10.5 Bohr this time: PP: Si_blyp.mt (by DRN) ----------------------- E_cut[Ry] a_lat[Bohr] E_tot [Ry] E_gap [eV] 70.0 10.20 -15.57266012 0.855 10.30 -15.57527183 0.897 >> 10.40 -15.57643309 0.935<< 10.50 -15.57627761 0.971 PP: Si_lda.mt (by DRN) ---------------------- E_cut[Ry] a_lat[Bohr] E_tot [Ry] E_gap [eV] 30.0 10.10 -15.85440094 0.463 30.0 10.15 -15.85487606 0.488 >>30.0 10.20 -15.85500476 0.510<< 30.0 10.25 -15.85475781 0.533 30.0 10.30 -15.85413174 0.555 PP: Si_pbe.mt (by DRN) ---------------------- E_cut[Ry] a_lat[Bohr] E_tot [Ry] E_gap [eV] 30.0 10.10 -15.72409215 0.537 30.0 10.20 -15.72726160 0.589 30.0 10.30 -15.72883722 0.638 >>30.0 10.40 -15.72895958 0.684<< 30.0 10.50 -15.72773301 0.727 I modified the shell script of example 5 (pw_examples) a bit but I did not change the k-point meshes, so E_gap is not accurate to three digits - is it necessary to use a denser k-point mesh for my tests? Do you think these results are reasonable (at least for the LDA and PBE pseudopotentials)? Which further tests would you suggest? Thank you very much for your help and your comments! With kind regards, Danilo Neuber -- Dipl.-Ing. Danilo Neuber Technische Universit?t Graz Telefon: +43 / 316 / 873 - 8188 Institut f?r Theoretische Physik Telefax: +43 / 316 / 873 - 8678 Petersgasse 16, A-8010 Graz From goranka.bilalbegovic at zg.hinet.hr Tue Dec 2 09:49:26 2003 From: goranka.bilalbegovic at zg.hinet.hr (Goranka Bilalbegovic) Date: Tue, 2 Dec 2003 09:49:26 +0100 Subject: [Pw_forum] pw.1.3.0 and ifc 7.1 Message-ID: <001801c3b8b1$4dc84ac0$cd841dc3@gost> Hello. I compiled pw.1.3.0 using the latest version of ifc compiler on the Linux Mandrake 8, AMD processor. However, I am not able to compile it using the same compiler version on the Red Hat 7.2, Intel processor (mkl is not used). In this case make pw stops after the ld warning message (as previously reported in this forum by Victor Trubitsin, on Intel + mkl). I did not see any differences in the make.log files, ld is configured in the same way on both systems... There is a possibility is that some differences in the processors, operating systems, and the system libraries they use prevent compilation of pw.1.3.0 on some Linux PC's. I would like to know (at least) on which systems pw.1.3.0 + ifc 7.1 works, and on which it does not ? Up to now a very small statistics shows that it works on the AMD (one example) and it does not work on the Intel processor (two examples). Best regards, Goranka -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20031202/88dfd4a3/attachment.htm From giannozz at nest.sns.it Tue Dec 2 22:08:55 2003 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 2 Dec 2003 22:08:55 +0100 Subject: [Pw_forum] pw.1.3.0 and ifc 7.1 In-Reply-To: <001801c3b8b1$4dc84ac0$cd841dc3@gost> References: <001801c3b8b1$4dc84ac0$cd841dc3@gost> Message-ID: <200312022208.56162.giannozz@nest.sns.it> On Tuesday 02 December 2003 09:49, Goranka Bilalbegovic wrote: > I would like to know (at least) on which systems pw.1.3.0 + ifc 7.1 works on all systems I have tried (quite a few), including several different beowulf clusters, provided that the latest version of ifc is used. For instance: IBM BladeServer, 8 * 2.4 GHz dual Xeon, RedHat 9 ; 400MHz Pentium 3, SuSe 7.1 (the PC I use at work) > and on which it does not ? on none that I have tried > Up to now a very small statistics shows that it works on the AMD (one > example) and it does not work on the Intel processor (two examples). the only difference between AMD and Intel processors I am aware of, is that you cannot use the MKL library with the former. You can use ifc (i.e. it works), but you are not supposed to (i.e. it is not supported). Paolo From giannozz at nest.sns.it Tue Dec 2 22:14:43 2003 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 2 Dec 2003 22:14:43 +0100 Subject: [Pw_forum] Si band gap calculated with different functionals In-Reply-To: References: <200311291648.47211.giannozz@nest.sns.it> Message-ID: <200312022214.43442.giannozz@nest.sns.it> On Monday 01 December 2003 17:13, Danilo Neuber wrote: > I did some more calculations with my PPs varying the lattice constant a_0 > from 10.1 to 10.5 Bohr this time [...] > Do you think these results are reasonable (at least for the LDA and > PBE pseudopotentials)? Which further tests would you suggest? they look perfectly reasonable. I have no idea if it is correct that BLYP yield a larger gap, and even less why. There are millions of theoretical results for Si in the literature, maybe there is also something with BLYP. Si is anyway an "easy" element with respect to pseudopotential generation. Paolo From francesco.antoniella at aquila.infn.it Wed Dec 3 08:54:46 2003 From: francesco.antoniella at aquila.infn.it (antonief) Date: Wed, 3 Dec 2003 08:54:46 +0100 (CET) Subject: [Pw_forum] pw.1.3.0 and ifc 7.1 In-Reply-To: <200312022208.56162.giannozz@nest.sns.it> Message-ID: On Tue, 2 Dec 2003, Paolo Giannozzi wrote: > On Tuesday 02 December 2003 09:49, Goranka Bilalbegovic wrote: > > > I would like to know (at least) on which systems pw.1.3.0 + ifc 7.1 works > > on all systems I have tried (quite a few), including several different beowulf > clusters, provided that the latest version of ifc is used. For instance: > IBM BladeServer, 8 * 2.4 GHz dual Xeon, RedHat 9 ; > 400MHz Pentium 3, SuSe 7.1 (the PC I use at work) > > > and on which it does not ? > > on none that I have tried > > > Up to now a very small statistics shows that it works on the AMD (one > > example) and it does not work on the Intel processor (two examples). > > the only difference between AMD and Intel processors I am aware of, > is that you cannot use the MKL library with the former. You can use ifc In fact you can actually use mkl on AMD, using pIII mkl and no processor specific optimization. It works, I don't know how fast but it works. Whishes Francesco From ps at ned.sims.nrc.ca Tue Dec 2 22:26:28 2003 From: ps at ned.sims.nrc.ca (Serguei Patchkovskii) Date: Tue, 2 Dec 2003 16:26:28 -0500 (EST) Subject: [Pw_forum] pw.1.3.0 and ifc 7.1 In-Reply-To: <200312022208.56162.giannozz@nest.sns.it> Message-ID: On Tue, 2 Dec 2003, Paolo Giannozzi wrote: > the only difference between AMD and Intel processors I am aware of, > is that you cannot use the MKL library with the former. You most certainly can use MKL with AMD processors (we do, for pwscf as well as for many other codes). For Athlons (both XP and MP), you must choose the p3-optimized version of MKL. For Opterons, both p3 and p4 versions work fine. The resulting executable is not quite as fast as with a fully optimized library (e.g. Atlas) - but the difference is fairly small. This is not something you want to do if you only have Athlons around. However, if you have to support a heterogenous cluster, it allows a large reduction in the amount of work you have to do to build and test the binaries. Serguei --- Dr. Serguei Patchkovskii Tel: +1-(613)-990-0945 Fax: +1-(613)-947-2838 E-mail: Serguei.Patchkovskii at nrc.ca Coordinator of Modelling Software Theory and Computation Group Steacie Institute for Molecular Sciences National Research Council Canada Room 2011, 100 Sussex Drive Ottawa, Ontario K1A 0R6 Canada From tvynew at otf.pti.udm.ru Wed Dec 3 11:36:53 2003 From: tvynew at otf.pti.udm.ru (Victor Trubitsin) Date: Wed, 3 Dec 2003 14:36:53 +0400 Subject: [Pw_forum] pw.1.3.0 and ifc 7.1 In-Reply-To: <001801c3b8b1$4dc84ac0$cd841dc3@gost> References: <001801c3b8b1$4dc84ac0$cd841dc3@gost> Message-ID: Hello, Goranka You have to change of the make.sys in main PW1.3 directory (on the Red Hat 7.2 system with ifc 7.1 compiler) adding the next key for loader (ld) in LDFLAGS: -Xlinker --noinhibit-exec. For example: LDFLAGS =-Xlinker --noinhibit-exec $OSHOME)/flib/ptools.a$(OSHOME)/flib/flib.a $(OSHOME)/clib/clib.a -Vaxlib $(LIBS) WARNING: Information from: man ld ####################################################### `--noinhibit-exec' Retain the executable output file whenever it is still usable. Normally, the linker will not produce an output file if it encounters errors during the link process; it exits without writing an output file when it issues any error whatsoever. ####################################################### 2.12. 2003 12:49, You wrote > Hello. > > I compiled pw.1.3.0 using the latest version of ifc compiler on the Linux > Mandrake 8, AMD processor. However, I am not able to compile it using the > same compiler version on the Red Hat 7.2, Intel processor (mkl is not > used). In this case make pw stops after the ld warning message (as > previously reported in this forum by Victor Trubitsin, on Intel + mkl). > > I did not see any differences in the make.log files, ld is configured in > the same way on both systems... There is a possibility is that some > differences in the processors, operating systems, and the system libraries > they use prevent compilation of pw.1.3.0 on some Linux PC's. > > I would like to know (at least) on which systems pw.1.3.0 + ifc 7.1 works, > and on which it does not ? Up to now a very small statistics shows that it > works on the AMD (one example) and it does not work on the Intel processor > (two examples). > > Best regards, > Goranka From giannozz at nest.sns.it Wed Dec 3 23:15:18 2003 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Wed, 3 Dec 2003 23:15:18 +0100 Subject: [Pw_forum] pw.1.3.0 and ifc 7.1 In-Reply-To: References: Message-ID: <200312032315.18299.giannozz@nest.sns.it> On Tuesday 02 December 2003 22:26, Serguei Patchkovskii wrote: > > the only difference between AMD and Intel processors I am aware of, > > is that you cannot use the MKL library with the former. > You most certainly can use MKL with AMD processors (we do, for pwscf > as well as for many other codes). I stand corrected > For Athlons (both XP and MP), you must choose the p3-optimized version of > MKL. For Opterons, both p3 and p4 versions work fine. did you compile pwscf on Opterons ? Paolo From ps at ned.sims.nrc.ca Wed Dec 3 23:32:10 2003 From: ps at ned.sims.nrc.ca (Serguei Patchkovskii) Date: Wed, 3 Dec 2003 17:32:10 -0500 (EST) Subject: [Pw_forum] pw.1.3.0 and ifc 7.1 In-Reply-To: <200312032315.18299.giannozz@nest.sns.it> Message-ID: On Wed, 3 Dec 2003, Paolo Giannozzi wrote: > > For Athlons (both XP and MP), you must choose the p3-optimized version of > > MKL. For Opterons, both p3 and p4 versions work fine. > > did you compile pwscf on Opterons ? Yes - with pgf90 (which is AFAIK the only Fortran-90 compiler, which supports 64-bit mode on Opteron). Unfortunately, I find that pgf90 generates very slow code for pwscf, even with opteron-specific optimizations. ifc binary, running in 32-bit compatibility mode (-xW and P4 MKL) outperforms 64-bit pgf90 binary (which has twice as many registers to play with!) by a large margin, on all (small) test cases I tried. This was with 5.0-2 version of pgf90 - hopefully, things improved with 5.1 (which was released couple of weeks ago). However, I did not have time to play with it yet. Serguei --- Dr. Serguei Patchkovskii Tel: +1-(613)-990-0945 Fax: +1-(613)-947-2838 E-mail: Serguei.Patchkovskii at nrc.ca Coordinator of Modelling Software Theory and Computation Group Steacie Institute for Molecular Sciences National Research Council Canada Room 2011, 100 Sussex Drive Ottawa, Ontario K1A 0R6 Canada From zyli at sina.com Sat Dec 6 13:32:44 2003 From: zyli at sina.com (Zhenyu Li) Date: Sat, 6 Dec 2003 20:32:44 +0800 Subject: [Pw_forum] PWscf-1.3 on Compaq ALPHA clusters under Linux Message-ID: <200312061230.UAA05059@mx1.ustc.edu.cn> Dear Sergey and all PWscf users When compile pwscf-1.3.0 on an alpha machine ---------------------------------------------------- sigma-x0> uname -a OSF1 sigma-x0 V5.1 1885 alpha ----------------------------------------------------- I run ./configure alpha first, then make all, but I get the following error: ----------------------------------------------------- ar rv clib.a c_getenv.o c_today.o cpflush.o factor.o indici.o memstat.o qsort.o readocc.o round2.o cptimer.o fft_stick.o a - c_getenv.o a - c_today.o a - cpflush.o a - factor.o a - indici.o a - memstat.o a - qsort.o a - readocc.o a - round2.o a - cptimer.o a - fft_stick.o ar: Warning: creating clib.a Make: Cannot open .dependencies. Stop. *** Exit 1 Stop. ----------------------------------------------------- Can anybody give me some suggestion? Best Wishes, zyli ======= 2003-10-03 13:02:00 ????????======= >Dear PWscf authors and users, > >I think we should to separate make.sys files for DEC Alpha and Compaq Alpha machines. >I using the last machines working under Linux. I have seen the Makefile provided by script configure >does nor work on Compaq Alpha machines. I have seen the errors message and done the minor changes >in file clib/cp.h (add one underscore): > >#if defined __ALPHA && defined __LINUX > ># define FFTW_INPLACE_DRV_1D fftw_inplace_drv_1d__ ># define FFTW_INPLACE_DRV_2D fftw_inplace_drv_2d__ ># define FFTW_INPLACE_DRV_3D fftw_inplace_drv_3d__ ># define CREATE_PLAN_1D create_plan_1d__ ># define DESTROY_PLAN_1D destroy_plan_1d__ ># define CREATE_PLAN_2D create_plan_2d__ ># define CREATE_PLAN_3D create_plan_3d__ ># define DESTROY_PLAN_2D destroy_plan_2d__ ># define DESTROY_PLAN_3D destroy_plan_3d__ ># define FFT_X_STICK fft_x_stick__ ># define FFT_XY_STICK fft_xy_stick_ ># define FFT_XY fft_xy_ ># define FFT_Y_STICK fft_y_stick__ ># define FFT_Y_STICK2 fft_y_stick2_ ># define FFT_Z_STICK fft_z_stick__ ># define FFT_Z fft_z_ ># define FFT_STICK fft_stick_ ># define CP_GETENV cp_getenv_ ># define CP_DATE cp_date__ ># define CPFLUSH cpflush_ ># define CPTIMER cptimer_ ># define ELAPSED_SECONDS elapsed_seconds__ ># define CCLOCK cclock_ ># define FACTOR235 factor235_ ># define FACTOR2 factor2_ ># define LN_ALLOC ln_alloc__ ># define LN_DEALLOC ln_dealloc__ ># define LN_SET ln_set__ ># define LN_ACTIVATE ln_activate__ ># define LN_IND ln_ind__ ># define MEMSTAT memstat_ ># define READOCC readocc_ ># define ROUND2 round2_ ># define MYUNITNAME myunitname_ ># define CP_ITOA cp_itoa_ > >#endif > >and redefine CPPFLAGS : > >CPPFLAGS = -D__ALPHA -D__LINUX ..... > >The compiler is "fort", i.e. > >F90=fort >F77=fort > >Bets wishes, >Sergey >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum > > >. ??????????? From giannozz at nest.sns.it Sat Dec 6 16:54:04 2003 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Sat, 6 Dec 2003 16:54:04 +0100 Subject: [Pw_forum] PWscf-1.3 on Compaq ALPHA clusters under Linux In-Reply-To: <200312061230.UAA05059@mx1.ustc.edu.cn> References: <200312061230.UAA05059@mx1.ustc.edu.cn> Message-ID: <200312061654.04848.giannozz@nest.sns.it> On Saturday 06 December 2003 13:32, Zhenyu Li wrote: > When compile pwscf-1.3.0 on an alpha machine [...] > I get the following error: [...] > Make: Cannot open .dependencies. Stop. > *** Exit 1 > Stop. > ----------------------------------------------------- > Can anybody give me some suggestion? create an empty directory "FPMD" and re-run "./configure alpha". The missing file (flib/.dependencies) will be created this time. Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050509412 Piazza dei Cavalieri 7 Fax: +39/050509417, 050563513 I-56126 Pisa, Italy Office: Lab. NEST, Via della Faggiola 19 From zyli at sina.com Sun Dec 7 02:52:11 2003 From: zyli at sina.com (Zhenyu Li) Date: Sun, 7 Dec 2003 09:52:11 +0800 Subject: [Pw_forum] PWscf-1.3 on Compaq ALPHA clusters under Linux Message-ID: <200312070150.JAA26804@mx1.ustc.edu.cn> Dear Paolo Thanks for your advice, it works! I encounterd a small error saying "a extra comma in pwtools/kpoints.f, line 89", and by removing this extra comma, I compiled the aplpha version successfully. A strange thing in compiling alphaMPI is: when I run 'make all' the first time, I got some error by some unresolved MPI functions; but I got a successful make by just typing 'make all' again. :-) Best Wishes zyli ======= 2003-12-06 23:54:04 ????????======= >On Saturday 06 December 2003 13:32, Zhenyu Li wrote: > >> When compile pwscf-1.3.0 on an alpha machine [...] >> I get the following error: [...] >> Make: Cannot open .dependencies. Stop. >> *** Exit 1 >> Stop. >> ----------------------------------------------------- >> Can anybody give me some suggestion? > >create an empty directory "FPMD" and re-run "./configure alpha". >The missing file (flib/.dependencies) will be created this time. > >Paolo > >-- >Paolo Giannozzi e-mail: giannozz at nest.sns.it >Scuola Normale Superiore Phone: +39/050509412 >Piazza dei Cavalieri 7 Fax: +39/050509417, 050563513 >I-56126 Pisa, Italy Office: Lab. NEST, Via della Faggiola 19 > >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum > > ????????????? From eyvaz_isaev at yahoo.com Sun Dec 7 13:36:01 2003 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Sun, 7 Dec 2003 04:36:01 -0800 (PST) Subject: [Pw_forum] PWscf-1.3 on Compaq ALPHA clusters under Linux In-Reply-To: <200312070150.JAA26804@mx1.ustc.edu.cn> Message-ID: <20031207123601.73082.qmail@web60309.mail.yahoo.com> Dear Zhenyu, Could you please let me know, did you try to run a batch-file containing a number of scripts inside? I have also Linux/Alpha cluster, but I coud not manage to run this kind of scripts. It turned out that "mpirun" command works only for the first script and then it leaves batch-file. So, other scripts know nothing about the "mpirun" command. Dividing of a script makes a lot of garbage. I will be glad to receive any hint from other forum members how to fix this problem. Regards, Eyvaz. --- Zhenyu Li wrote: > Dear Paolo > > Thanks for your advice, it works! > > I encounterd a small error saying "a extra comma in > pwtools/kpoints.f, > line 89", and by removing this extra comma, I > compiled the aplpha version > successfully. > > A strange thing in compiling alphaMPI is: when I > run 'make all' the first > time, I got some error by some unresolved MPI > functions; but I got a successful > make by just typing 'make all' again. :-) > > Best Wishes > zyli > > > ======= 2003-12-06 23:54:04 ????????????????======= > > >On Saturday 06 December 2003 13:32, Zhenyu Li > wrote: > > > >> When compile pwscf-1.3.0 on an alpha machine > [...] > >> I get the following error: [...] > >> Make: Cannot open .dependencies. Stop. > >> *** Exit 1 > >> Stop. > >> > ----------------------------------------------------- > >> Can anybody give me some suggestion? > > > >create an empty directory "FPMD" and re-run > "./configure alpha". > >The missing file (flib/.dependencies) will be > created this time. > > > >Paolo > > > >-- > >Paolo Giannozzi e-mail: > giannozz at nest.sns.it > >Scuola Normale Superiore Phone: +39/050509412 > >Piazza dei Cavalieri 7 Fax: +39/050509417, > 050563513 > >I-56126 Pisa, Italy Office: Lab. NEST, Via > della Faggiola 19 > > > >_______________________________________________ > >Pw_forum mailing list > >Pw_forum at pwscf.org > >http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > ?????????????????????????? > ?????j)b� b???~�?��,q?+�m????^��+�?,�??j)fj�?�?b�?????? __________________________________ Do you Yahoo!? Protect your identity with Yahoo! Mail AddressGuard http://antispam.yahoo.com/whatsnewfree From zyli at sina.com Sun Dec 7 13:57:15 2003 From: zyli at sina.com (Zhenyu Li) Date: Sun, 7 Dec 2003 20:57:15 +0800 Subject: [Pw_forum] PWscf-1.3 on Compaq ALPHA clusters under Linux Message-ID: <200312071255.UAA14656@mx1.ustc.edu.cn> Dear Eyvaz It's not a cluster at our university, I just get a remote access to use this HP-SC45 system. And it is not even a Linux cluster, it's configurated to use prun to run MPI parallel job. I have successfully used a batch file, which contains a number of prun commands inside, to run several parallel jobs. It may be helpful to post your batch file in the list. best wishes zhenyu ======= 2003-12-07 20:36:01 ????????????????======= >Dear Zhenyu, > >Could you please let me know, did you try to run a >batch-file containing a number of scripts inside? >I have also Linux/Alpha cluster, but I coud not manage >to run this kind of scripts. >It turned out that "mpirun" command works only for the > first script and then it leaves batch-file. So, other >scripts know nothing about the "mpirun" command. >Dividing of a script makes a lot of garbage. > >I will be glad to receive any hint from other forum >members how to fix this problem. > >Regards, >Eyvaz. > > > >--- Zhenyu Li wrote: >> Dear Paolo >> >> Thanks for your advice, it works! >> >> I encounterd a small error saying "a extra comma in >> pwtools/kpoints.f, >> line 89", and by removing this extra comma, I >> compiled the aplpha version >> successfully. >> >> A strange thing in compiling alphaMPI is: when I >> run 'make all' the first >> time, I got some error by some unresolved MPI >> functions; but I got a successful >> make by just typing 'make all' again. :-) >> >> Best Wishes >> zyli >> >> >> ======= 2003-12-06 23:54:04 ????????????????======= >> >> >On Saturday 06 December 2003 13:32, Zhenyu Li >> wrote: >> > >> >> When compile pwscf-1.3.0 on an alpha machine >> [...] >> >> I get the following error: [...] >> >> Make: Cannot open .dependencies. Stop. >> >> *** Exit 1 >> >> Stop. >> >> >> >----------------------------------------------------- >> >> Can anybody give me some suggestion? >> > >> >create an empty directory "FPMD" and re-run >> "./configure alpha". >> >The missing file (flib/.dependencies) will be >> created this time. >> > >> >Paolo >> > >> >-- >> >Paolo Giannozzi e-mail: >> giannozz at nest.sns.it >> >Scuola Normale Superiore Phone: +39/050509412 >> >Piazza dei Cavalieri 7 Fax: +39/050509417, >> 050563513 >> >I-56126 Pisa, Italy Office: Lab. NEST, Via >> della Faggiola 19 >> > >> >_______________________________________________ >> >Pw_forum mailing list >> >Pw_forum at pwscf.org >> >http://www.democritos.it/mailman/listinfo/pw_forum >> > >> > >> >> ?????????????????????????? >> ?????j)b� >b???~�?#65533;�,q?�m???^��+�?�??j)fj�?#65533;?b�????? > > >__________________________________ >Do you Yahoo!? >Protect your identity with Yahoo! Mail AddressGuard >http://antispam.yahoo.com/whatsnewfree >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum > > ?????????????????????? From giannozz at nest.sns.it Sun Dec 7 17:16:49 2003 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Sun, 7 Dec 2003 17:16:49 +0100 Subject: [Pw_forum] PWscf-1.3 on Compaq ALPHA clusters under Linux In-Reply-To: <20031207123601.73082.qmail@web60309.mail.yahoo.com> References: <20031207123601.73082.qmail@web60309.mail.yahoo.com> Message-ID: <200312071716.49505.giannozz@nest.sns.it> On Sunday 07 December 2003 13:36, Eyvaz Isaev wrote: > Could you please let me know, did you try to run a > batch-file containing a number of scripts inside? I don't think you can run a batch script with "mpirun". You have to use "mpirun" to run each occurrence of the executable code Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050509412 Piazza dei Cavalieri 7 Fax: +39/050509417, 050563513 I-56126 Pisa, Italy Office: Lab. NEST, Via della Faggiola 19 From eyvaz_isaev at yahoo.com Mon Dec 8 12:03:33 2003 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Mon, 8 Dec 2003 03:03:33 -0800 (PST) Subject: [Pw_forum] PWscf-1.3 on Compaq ALPHA clusters under Linux In-Reply-To: <200312071716.49505.giannozz@nest.sns.it> Message-ID: <20031208110333.27790.qmail@web60309.mail.yahoo.com> Dear Paolo, Thank a lot for your comment. Sure, I tried to run scripts using mpirun for each executable code but I met very strange errors like " Error: Need to obtain the job magic number in GMPI_MAGIC !" (that means I have to run "mpirun job") or error reading input files (which disappeared if it run separately). Eyvaz. --- Paolo Giannozzi wrote: > On Sunday 07 December 2003 13:36, Eyvaz Isaev wrote: > > > Could you please let me know, did you try to run a > > batch-file containing a number of scripts inside? > > I don't think you can run a batch script with > "mpirun". > You have to use "mpirun" to run each occurrence of > the executable code > > Paolo > > -- > Paolo Giannozzi e-mail: > giannozz at nest.sns.it > Scuola Normale Superiore Phone: +39/050509412 > Piazza dei Cavalieri 7 Fax: +39/050509417, > 050563513 > I-56126 Pisa, Italy Office: Lab. NEST, Via > della Faggiola 19 > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum __________________________________ Do you Yahoo!? Protect your identity with Yahoo! Mail AddressGuard http://antispam.yahoo.com/whatsnewfree From pushpa at jncasr.ac.in Tue Dec 9 06:51:35 2003 From: pushpa at jncasr.ac.in (Raghani Pushpa) Date: Tue, 9 Dec 2003 11:21:35 +0530 (IST) Subject: [Pw_forum] Wannier functions Message-ID: Dear users, I have a question related to Wannier functions. Can we call the Wannier functions as the fourier transform of Bloch states (u_kn)? Thanx, Pushpa -- From sprokuda at dnk.ru Tue Dec 9 11:43:50 2003 From: sprokuda at dnk.ru (sprokuda) Date: Tue, 9 Dec 2003 13:43:50 +0300 Subject: [Pw_forum] help entreaty In-Reply-To: <200312061654.04848.giannozz@nest.sns.it> References: <200312061230.UAA05059@mx1.ustc.edu.cn> <200312061654.04848.giannozz@nest.sns.it> Message-ID: <3013510125.20031209134350@dnk.ru> Dear PWscf users! With my PWscf (1.1.2 ifc7.1 Blas-Lapack from netlib.org, AMD Duron 1300 - RedHat 7.2) already for several monthes I try to get some results on phonon dispersion on Ge. These calculations are pretty important for preparation of the neutron scattering experiments on phonon liftimes measurements in Ge. I get from time to time negative frequencies from PWscf. I have no idea what to do futher in this concern. The archive is attached to this mail - there are several scripts within. Could somebody run these scripts and send me back the results? These calculations are rather huge, and on my machine it takes 6-7hours to run one bash script. Sincerely yours, Prokudaylo Sergey. -------------- next part -------------- A non-text attachment was scrubbed... Name: Ge_bSD.rar Type: application/octet-stream Size: 13684 bytes Desc: not available Url : /pipermail/attachments/20031209/5d91e92e/attachment.obj From giannozz at nest.sns.it Tue Dec 9 14:56:07 2003 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 9 Dec 2003 14:56:07 +0100 Subject: [Pw_forum] help entreaty In-Reply-To: <3013510125.20031209134350@dnk.ru> References: <200312061230.UAA05059@mx1.ustc.edu.cn> <200312061654.04848.giannozz@nest.sns.it> <3013510125.20031209134350@dnk.ru> Message-ID: <200312091456.08099.giannozz@nest.sns.it> On Tuesday 09 December 2003 11:43, sprokuda wrote: > [...] I get from time to time negative frequencies from PWscf. > I have no idea what to do futher in this concern. from the last version of the manual: * ph.x does not yield acoustic modes with omega(q=0) = 0 This may not be an error: the Acoustic Sum Rule (ASR) is never exactly verified, because the system is never exactly translationally invariant as it should be (see the discussion above). The calculated frequency of the acoustic mode is typically less than 10 cm^{-1}, but in some cases it may be much higher, up to 100 cm^{-1}. The ultimate test is to diagonalize the dynamical matrix with program "dynmat.x", imposing the ASR. If you obtain an acoustic mode with a much smaller frequency (let us say < 1 cm^{-1}) with all other modes virtually unchanged, you can trust your results. * ph.x yields really lousy phonons, with bad or negative frequencies or wrong symmetries or gross ASR violations. Possible reasons: - Wrong data file file read. - Wrong atomic masses given in input will yield wrong frequencies (but the content of file "fildyn" should be valid, since the force constants, not the dynamical matrix, are written to file). - Convergence threshold for either SCF ("conv_thr") or phonon calculation "tr2_ph") too large (try to reduce them), - Maybe your system does have negative or strange phonon frequencies, with the approximations you used. A negative frequency signals a mechanical instability of the chosen structure. Check that the structure is reasonable, and check the following parameters: - The cutoff for wavefunctions, "ecutwfc" - For US PP: the cutoff for the charge density, "ecutrho" - The k-point grid, especially for metallic systems! Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050509412 Piazza dei Cavalieri 7 Fax: +39/050509417, 050563513 I-56126 Pisa, Italy Office: Lab. NEST, Via della Faggiola 19 From sprokuda at dnk.ru Tue Dec 9 15:10:39 2003 From: sprokuda at dnk.ru (sprokuda) Date: Tue, 9 Dec 2003 17:10:39 +0300 Subject: [Pw_forum] help entreaty In-Reply-To: <200312091456.08099.giannozz@nest.sns.it> References: <200312061230.UAA05059@mx1.ustc.edu.cn> <200312061654.04848.giannozz@nest.sns.it> <3013510125.20031209134350@dnk.ru> <200312091456.08099.giannozz@nest.sns.it> Message-ID: <16825918796.20031209171039@dnk.ru> Dear Paolo! When in one run of the same script I get positive frequency, in the other run - negative frequencies (on some points in BLZ), while other points are in good agreement with exdperiment - you think it can be explained by some features of solid states theory ? I don't think so. I strongly guess that it is a prblenm of configuration of my computer ( mostly hardware), that' why I'm asking about help. Sincerely yours, Sergey Prokudaylo. From francesco.antoniella at aquila.infn.it Thu Dec 11 16:42:04 2003 From: francesco.antoniella at aquila.infn.it (Francesco Antoniella) Date: 11 Dec 2003 16:42:04 +0100 Subject: [Pw_forum] Una domanda e una proposta In-Reply-To: <200307100952.14831.giannozz@nest.sns.it> References: <011901c3463a$37738170$a0c583c8@venus10> <200307100952.14831.giannozz@nest.sns.it> Message-ID: <1071157338.2143.5.camel@altair> Caro Paolo, ti scrivo, come si evince dal subject, per chiederti se mi puoi dare i files per lo pseudo del cromo che hai da poco generato e messo in rete. Vorrei usarlo nel mio codice di conduttanza e mi servirebbero le funzioni d'onda all-electron cosi' pensavo di ripassare l'input che hai usato nel Vanderbilt modificato (rotante e scrivente le AEWFC). Volevo inoltre proporti una modifica che ho applicato ( a martellate ) nel PW_130 per ovviare ad un piccolo inconveniente: io uso un cluster, Attila, di 7 PC con linux e LAM-MPI. La libreria LAM crea un po' di seghe nel leggere l'input da stdin cosi' ho patchato il codice per accettare un nome come argomento. Non so se serve ad altri ne' credo di averlo implementato nella maniera piu' fine ma funzia... Ciao Francesco From francesco.antoniella at aquila.infn.it Thu Dec 11 16:43:28 2003 From: francesco.antoniella at aquila.infn.it (Francesco Antoniella) Date: 11 Dec 2003 16:43:28 +0100 Subject: [Pw_forum] Seguito della mail Domanda&proposta In-Reply-To: <200307100952.14831.giannozz@nest.sns.it> References: <011901c3463a$37738170$a0c583c8@venus10> <200307100952.14831.giannozz@nest.sns.it> Message-ID: <1071157418.2143.8.camel@altair> Magari se ti allego il file e' meglio :) ciao -------------- next part -------------- A non-text attachment was scrubbed... Name: pwscf.f90.gz Type: application/x-gzip Size: 722 bytes Desc: not available Url : /pipermail/attachments/20031211/d232e33a/attachment.bin From giannozz at nest.sns.it Thu Dec 11 17:00:55 2003 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 11 Dec 2003 17:00:55 +0100 Subject: [Pw_forum] Una domanda e una proposta In-Reply-To: <1071157338.2143.5.camel@altair> References: <011901c3463a$37738170$a0c583c8@venus10> <200307100952.14831.giannozz@nest.sns.it> <1071157338.2143.5.camel@altair> Message-ID: <200312111700.55988.giannozz@nest.sns.it> Ciao ti spedisco i files usati per generare il cromo. Spero siano quelli giusti ... > La libreria LAM crea un po' di seghe nel leggere l'input da stdin cosi' > ho patchato il codice per accettare un nome come argomento. > Non so se serve ad altri altra gente mi ha segnalato problemi, ma si e' guardata bene dal segnalare soluzioni. Grazie, daro' un'occhiata alla prima occasione Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050509412 Piazza dei Cavalieri 7 Fax: +39/050509417, 050563513 I-56126 Pisa, Italy Office: Lab. NEST, Via della Faggiola 19 -------------- next part -------------- 1 0 0 0 3 ifae,ifpsp,ifprt,ifplw,ilogd (5i5) 1.80 -2.4 1.6 40 rlogd,emin,emax,nnt (3f10.5,i5) 1.0d-11 1.0d-09 0.15 thresh,tol,damp (2e10.1,f10.5) chromium title (a20) 24. 0.0 5. z,xion,exfact (f5.0,2f10.5) 280.0 6.0 59.0 rmax,aasf,bbsf (3f10.5) 8 2 ncspvs,irel (2i5) 100 2. -432.0 nnlz,wwnl,ee (i4,f7.3,f14.6) 200 2. -50.0 nnlz,wwnl,ee (i4,f7.3,f14.6) 210 6. -41.0 nnlz,wwnl,ee (i4,f7.3,f14.6) 300 2. -5.4 nnlz,wwnl,ee (i4,f7.3,f14.6) 310 6. -3.3 nnlz,wwnl,ee (i4,f7.3,f14.6) 400 1. -0.3 nnlz,wwnl,ee (i4,f7.3,f14.6) 410 0. -0.1 nnlz,wwnl,ee (i4,f7.3,f14.6) 320 5. -0.2 nnlz,wwnl,ee (i4,f7.3,f14.6) -------------- next part -------------- 0 2 1 1 3 ifae,ifpsp,ifprt,ifplw,ilogd (5i5) 1.80 -6.0 2.0 80 rlogd,emin,emax,nnt (3f10.5,i5) 1.0d-11 1.0d-09 0.15 thresh,tol,damp (2e10.1,f10.5) chromium title (a20) 3 5 3 ncores,nvales,nang (3i5) 10.0 20.0 40.0 5.0 besrmax, besemin,besemax,besde 3 0 1.0 keyps,ifpcor,rinner (2i5,f10.5) 6 1.5 nbeta,rcloc (i5,f10.5) 1.6 1.6 1.6 rc (4f10.5) 0 1 0.0 2 lll,keyee,eeread,iptype (2i5,f10.5,i5) 0 3 0.0 2 lll,keyee,eeread,iptype (2i5,f10.5,i5) 1 2 0.0 2 lll,keyee,eeread,iptype (2i5,f10.5,i5) 1 4 0.0 2 lll,keyee,eeread,iptype (2i5,f10.5,i5) 2 5 0.0 2 lll,keyee,eeread,iptype (2i5,f10.5,i5) 2 0 0.5 2 lll,keyee,eeread,iptype (2i5,f10.5,i5) 5 10.0 npf,ptryc (i5,f10.5) 3 0 0.0 1 lloc,keyee,eloc,iploctype (2i5,f10,5,i5) 3 8 10.0 ifqopt,nqf,qtryc (2i5,f10.5) From francesco.antoniella at aquila.infn.it Thu Dec 11 17:19:18 2003 From: francesco.antoniella at aquila.infn.it (Francesco Antoniella) Date: 11 Dec 2003 17:19:18 +0100 Subject: [Pw_forum] Apologies In-Reply-To: <200312111700.55988.giannozz@nest.sns.it> References: <011901c3463a$37738170$a0c583c8@venus10> <200307100952.14831.giannozz@nest.sns.it> <1071157338.2143.5.camel@altair> <200312111700.55988.giannozz@nest.sns.it> Message-ID: <1071159568.2144.15.camel@altair> I would like to apologize with the whole forum for two messages I sent in Italian that was intended for Paolo Giannozzi only, but I committed twice the error (diabolic) to send them to the forum. Thanks Francesco From giannozz at nest.sns.it Thu Dec 11 17:23:51 2003 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 11 Dec 2003 17:23:51 +0100 Subject: [Pw_forum] Apologies In-Reply-To: <1071159568.2144.15.camel@altair> References: <011901c3463a$37738170$a0c583c8@venus10> <200312111700.55988.giannozz@nest.sns.it> <1071159568.2144.15.camel@altair> Message-ID: <200312111723.51986.giannozz@nest.sns.it> On Thursday 11 December 2003 17:19, Francesco Antoniella wrote: > I would like to apologize with the whole forum for two messages > I sent in Italian that was intended for Paolo Giannozzi only, but I > committed twice the error (diabolic) to send them to the forum. ...so did I in my answer ...sorry! Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050509412 Piazza dei Cavalieri 7 Fax: +39/050509417, 050563513 I-56126 Pisa, Italy Office: Lab. NEST, Via della Faggiola 19 From breezejd at lsbu.ac.uk Thu Dec 11 18:43:02 2003 From: breezejd at lsbu.ac.uk (Jonathan Breeze) Date: Thu, 11 Dec 2003 17:43:02 +0000 Subject: [Pw_forum] Apologies References: <011901c3463a$37738170$a0c583c8@venus10> <200312111700.55988.giannozz@nest.sns.it> <1071159568.2144.15.camel@altair> <200312111723.51986.giannozz@nest.sns.it> Message-ID: <3FD8ACA6.2020002@lsbu.ac.uk> I like Italian. I tried my best to read the emails and got the general idea. pwscf is pretty much italian so .... grazie ragazze e ragazzi . Paolo Giannozzi wrote: >On Thursday 11 December 2003 17:19, Francesco Antoniella wrote: > > > >>I would like to apologize with the whole forum for two messages >>I sent in Italian that was intended for Paolo Giannozzi only, but I >>committed twice the error (diabolic) to send them to the forum. >> >> > >...so did I in my answer ...sorry! > >Paolo > > > -- Jonathan Breeze Research Fellow Centre for Physical Electronics and Materials London South Bank University 103 Borough Road London SE1 0AA Tel: +44(0)20 7815 7582 Fax: +44(0)20 7815 7599 From zyli at sina.com Fri Dec 12 16:18:27 2003 From: zyli at sina.com (Zhenyu Li) Date: Fri, 12 Dec 2003 23:18:27 +0800 Subject: [Pw_forum] problem of calculation of phonon Message-ID: <200312121517.XAA16184@mx1.ustc.edu.cn> Dear PWSCF users When doing phonon calculation, I got no problem for some q points, but I got the following error for some other q points: ----Cut from out file------------------------------------ Representation # 23 mode # 23 IOS = 36 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from davcio : error # 21 i/o error in davcio %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% It's strange that this error only exist for some specific q points. Can anybody give some suggestion? Sincerely Zhenyu From baknoo at tokyo.rist.or.jp Mon Dec 15 10:24:25 2003 From: baknoo at tokyo.rist.or.jp (PGnujeon) Date: Mon, 15 Dec 2003 18:24:25 +0900 Subject: [Pw_forum] sometimes a failure in qvan2.f90 References: <200312061230.UAA05059@mx1.ustc.edu.cn> <200312061654.04848.giannozz@nest.sns.it> <3013510125.20031209134350@dnk.ru> <200312091456.08099.giannozz@nest.sns.it> Message-ID: <004901c3c2ed$3b27d610$0d00a8c0@computat030u9q> Dear any expert at PWscf ? When I run pw.x with USPP, sometimes the program stops at qvan2.f90, reporting "invalid mpi operation in qvan2". Whenever I encounter such error message, I reduce the number of process and the code works. Sorry for this rough message. But would any body suggest any hint for this problem ? Again, "pw.x with USPP" successfully works with some number of processors. However it fails with some larger number of processors. Sincerely, Noah. ----- Original Message ----- From: "Paolo Giannozzi" To: Sent: Tuesday, December 09, 2003 10:56 PM Subject: Re: [Pw_forum] help entreaty > On Tuesday 09 December 2003 11:43, sprokuda wrote: > > > [...] I get from time to time negative frequencies from PWscf. > > I have no idea what to do futher in this concern. > > from the last version of the manual: > > * ph.x does not yield acoustic modes with omega(q=0) = 0 > This may not be an error: the Acoustic Sum Rule (ASR) is never > exactly verified, because the system is never exactly translationally > invariant as it should be (see the discussion above). > The calculated frequency of the acoustic mode is typically less > than 10 cm^{-1}, but in some cases it may be much higher, > up to 100 cm^{-1}. The ultimate test is to diagonalize the > dynamical matrix with program "dynmat.x", imposing the ASR. > If you obtain an acoustic mode with a much smaller frequency > (let us say < 1 cm^{-1}) with all other modes virtually unchanged, > you can trust your results. > * ph.x yields really lousy phonons, with bad or negative frequencies or > wrong symmetries or gross ASR violations. > Possible reasons: > - Wrong data file file read. > - Wrong atomic masses given in input will yield wrong frequencies > (but the content of file "fildyn" should be valid, since the force > constants, not the dynamical matrix, are written to file). > - Convergence threshold for either SCF ("conv_thr") or phonon > calculation "tr2_ph") too large (try to reduce them), > - Maybe your system does have negative or strange phonon > frequencies, with the approximations you used. A negative frequency > signals a mechanical instability of the chosen structure. Check that > the structure is reasonable, and check the following parameters: > - The cutoff for wavefunctions, "ecutwfc" > - For US PP: the cutoff for the charge density, "ecutrho" > - The k-point grid, especially for metallic systems! > > Paolo > -- > Paolo Giannozzi e-mail: giannozz at nest.sns.it > Scuola Normale Superiore Phone: +39/050509412 > Piazza dei Cavalieri 7 Fax: +39/050509417, 050563513 > I-56126 Pisa, Italy Office: Lab. NEST, Via della Faggiola 19 > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From giannozz at nest.sns.it Tue Dec 16 09:45:17 2003 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 16 Dec 2003 09:45:17 +0100 Subject: [Pw_forum] problem of calculation of phonon In-Reply-To: <200312121517.XAA16184@mx1.ustc.edu.cn> References: <200312121517.XAA16184@mx1.ustc.edu.cn> Message-ID: <200312160945.17155.giannozz@nest.sns.it> On Friday 12 December 2003 16:18, Zhenyu Li wrote: > When doing phonon calculation, I got no problem for some q points, > but I got the following error for some other q points: > [...] Can anybody give some suggestion? - is there enough disk space? - is the error it reproducible? - does it happen on different hardware/compilers? - does it happen in serial or parallel execution, or both? Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy *** please note my new phone number (effective december 17) *** From giannozz at nest.sns.it Tue Dec 16 09:47:13 2003 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 16 Dec 2003 09:47:13 +0100 Subject: [Pw_forum] sometimes a failure in qvan2.f90 In-Reply-To: <004901c3c2ed$3b27d610$0d00a8c0@computat030u9q> References: <200312061230.UAA05059@mx1.ustc.edu.cn> <200312091456.08099.giannozz@nest.sns.it> <004901c3c2ed$3b27d610$0d00a8c0@computat030u9q> Message-ID: <200312160947.13075.giannozz@nest.sns.it> On Monday 15 December 2003 10:24, PGnujeon wrote: > When I run pw.x with USPP, sometimes the program stops at qvan2.f90, > reporting "invalid mpi operation in qvan2". [...] would any body suggest > any hint for this problem ? if it is not reproducible, it is likely to be a MPI or hardware problem, rather than a problem of the code Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy *** please note my new phone number (after december 17) *** From zyli at sina.com Wed Dec 17 03:04:25 2003 From: zyli at sina.com (Zhenyu Li) Date: Wed, 17 Dec 2003 10:04:25 +0800 Subject: [Pw_forum] problem of calculation of phonon Message-ID: <200312170203.KAA04355@mx1.ustc.edu.cn> >On Friday 12 December 2003 16:18, Zhenyu Li wrote: > >> When doing phonon calculation, I got no problem for some q points, >> but I got the following error for some other q points: >> [...] Can anybody give some suggestion? > >- is there enough disk space? yes >- is the error it reproducible? yes >- does it happen on different hardware/compilers? yes, both pc_ifc and compaq alpha >- does it happen in serial or parallel execution, or both? both > thanks. >Paolo > >-- >Paolo Giannozzi e-mail: giannozz at nest.sns.it >Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 >Piazza dei Cavalieri 7 I-56126 Pisa, Italy >*** please note my new phone number (effective december 17) *** > >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum > > = = = = = = = = = = = = = = = = = = = = ????????? ?? ????????Zhenyu Li ????????zyli at sina.com ??????????2003-12-17 ================================================= Zhenyu LI Ph.D. Candidate Open Laboratory of Bond Selective Chemistry, University of Science and Technology of China, Hefei, Anhui, 230026, People's Republic of China Tel.: 86-551-3606428 Fax.: 86-551-3602969 http://www.bsc.ustc.edu.cn/~zyli/whoami/ ===================================================?????????????? From giannozz at nest.sns.it Wed Dec 17 10:00:59 2003 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Wed, 17 Dec 2003 10:00:59 +0100 (CET) Subject: [Pw_forum] problem of calculation of phonon In-Reply-To: <200312170203.KAA04355@mx1.ustc.edu.cn> Message-ID: On Wed, 17 Dec 2003, Zhenyu Li wrote: > >- is there enough disk space? > yes > >- is the error reproducible? > yes > >- does it happen on different hardware/compilers? > yes, both pc_ifc and compaq alpha > >- does it happen in serial or parallel execution, or both? > both could you please post your data (the jobs running pw.x and ph.x)? Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy *** please note my new phone and fax numbers *** From zyli at sina.com Wed Dec 17 14:00:34 2003 From: zyli at sina.com (Zhenyu Li) Date: Wed, 17 Dec 2003 21:00:34 +0800 Subject: [Pw_forum] problem of calculation of phonon Message-ID: <200312171259.UAA15481@mx1.ustc.edu.cn> Dear Paolo My input file is attached with this mail. I always get error in the following q points: 0.0000000,0.0000000,-.1313474 0.1250000,0.3608439,0.0000000 0.1250000,0.3608439,-.1313474 Sincerely Zhenyu ======= 2003-12-17 17:00:59 ????????======= >On Wed, 17 Dec 2003, Zhenyu Li wrote: > >> >- is there enough disk space? >> yes >> >- is the error reproducible? >> yes >> >- does it happen on different hardware/compilers? >> yes, both pc_ifc and compaq alpha >> >- does it happen in serial or parallel execution, or both? >> both > >could you please post your data (the jobs running pw.x and ph.x)? > >Paolo > >-- > >Paolo Giannozzi e-mail: giannozz at nest.sns.it >Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 >Piazza dei Cavalieri 7 I-56126 Pisa, Italy >*** please note my new phone and fax numbers *** > >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum > > = = = = = = = = = = = = = = = = = = = = ????????? ?? ????????Zhenyu Li ????????zyli at sina.com ??????????2003-12-17 ================================================= Zhenyu LI Ph.D. Candidate Open Laboratory of Bond Selective Chemistry, University of Science and Technology of China, Hefei, Anhui, 230026, People's Republic of China Tel.: 86-551-3606428 Fax.: 86-551-3602969 http://www.bsc.ustc.edu.cn/~zyli/whoami/ ===================================================?????????????? -------------- next part -------------- A non-text attachment was scrubbed... Name: bat.tar Type: application/octet-stream Size: 10240 bytes Desc: not available Url : /pipermail/attachments/20031217/b5d2ef1e/attachment.obj From pushpa at jncasr.ac.in Wed Dec 17 17:06:29 2003 From: pushpa at jncasr.ac.in (Raghani Pushpa) Date: Wed, 17 Dec 2003 21:36:29 +0530 (IST) Subject: [Pw_forum] (no subject) Message-ID: Dear users, Does anybody have numerical recipes subroutines in fortran 90. If so, could I please have it? Thanx, Pushpa -- From goranka.bilalbegovic at zg.hinet.hr Wed Dec 17 18:24:10 2003 From: goranka.bilalbegovic at zg.hinet.hr (Goranka Bilalbegovic) Date: Wed, 17 Dec 2003 18:24:10 +0100 Subject: [Pw_forum] (no subject) References: Message-ID: <001b01c3c4c2$b385dd60$bd8a1dc3@gost> > Does anybody have numerical recipes subroutines in fortran 90. If so, > could I please have it? Hi Pushpa, Numerical Recipes in Fortran 90 book On-Line is at http://www.library.cornell.edu/nr/bookf90pdf.html Best regards, Goranka From pushpa at jncasr.ac.in Thu Dec 18 04:51:06 2003 From: pushpa at jncasr.ac.in (Raghani Pushpa) Date: Thu, 18 Dec 2003 09:21:06 +0530 (IST) Subject: [Pw_forum] Absoft f90 compiler Message-ID: Dear users, Does anybody have absoft f90 compiler, I need it urgently. Could it be sent through email? If so, could you please send it to me? Thanx, Pushpa -- From whoming at sohu.com Thu Dec 18 05:51:37 2003 From: whoming at sohu.com (whoming) Date: Thu, 18 Dec 2003 12:51:37 +0800 Subject: [Pw_forum] pgf90 compile error on Linux PC References: Message-ID: <014e01c3c522$a3922610$192e77ca@whm> Dear Users, I have accounted complile errors when I try to install pwscf v1.3.0 onto my Linux PC. My compiler is PGI 5.0. According to the instruction, I do it in the following steps: 1. ./configure pc_pgi 2. make pw then this error message appears and the file Modules/cell_base.f90 perhaps wrong. make[1]: Entering directory `/home/whm/pwscf/Modules' /lib/cpp -P -traditional -D__LINUX -D__PGI -D__FFTW -D__USE_INTERNAL_FFTW -I/home/whm/pwscf/include -I./ cell_base.f90 cell_base.F90 pgf90 -fast -r8 -I/home/whm/pwscf/Modules -I/home/whm/pwscf/PW -I/home/whm/pwscf/PH -c cell_base.F90 -o cell_base.o PGF90-S-0000-Internal compiler error. size_of:bad dtype 300 (cell_base.F90: 99) PGF90-S-0000-Internal compiler error. size_of: bad dtype 0 (cell_base.F90: 99) PGF90-S-0000-Internal compiler error. scale_of:bad dtype 300 (cell_base.F90: 99) PGF90-S-0000-Internal compiler error. scale_of: bad dtype 0 (cell_base.F90: 99) 0 inform, 0 warnings, 4 severes, 0 fatal for updatecell PGF90-S-0000-Internal compiler error. size_of:bad dtype 300 (cell_base.F90: 131) PGF90-S-0000-Internal compiler error. size_of: bad dtype 0 (cell_base.F90: 131) PGF90-S-0000-Internal compiler error. scale_of:bad dtype 300 (cell_base.F90: 131) PGF90-S-0000-Internal compiler error. scale_of: bad dtype 0 (cell_base.F90: 131) 0 inform, 0 warnings, 4 severes, 0 fatal for cell_init_ht PGF90-S-0000-Internal compiler error. size_of:bad dtype 300 (cell_base.F90: 152) PGF90-S-0000-Internal compiler error. size_of: bad dtype 0 (cell_base.F90: 152) PGF90-S-0000-Internal compiler error. scale_of:bad dtype 300 (cell_base.F90: 152) PGF90-S-0000-Internal compiler error. scale_of: bad dtype 0 (cell_base.F90: 152) 0 inform, 0 warnings, 4 severes, 0 fatal for cell_init_a PGF90-S-0000-Internal compiler error. size_of:bad dtype 300 (cell_base.F90: 290) PGF90-S-0000-Internal compiler error. size_of: bad dtype 0 (cell_base.F90: 290) PGF90-S-0000-Internal compiler error. scale_of:bad dtype 300 (cell_base.F90: 290) PGF90-S-0000-Internal compiler error. scale_of: bad dtype 0 (cell_base.F90: 290) PGF90-S-0000-Internal compiler error. size_of:bad dtype 300 (cell_base.F90: 292) PGF90-S-0000-Internal compiler error. size_of: bad dtype 0 (cell_base.F90: 292) PGF90-S-0000-Internal compiler error. scale_of:bad dtype 300 (cell_base.F90: 292) PGF90-S-0000-Internal compiler error. scale_of: bad dtype 0 (cell_base.F90: 292) 0 inform, 0 warnings, 8 severes, 0 fatal for pbc make[1]: [cell_base.o] Error 2 (ignored) touch cell_base.o Any suggestion will be appreciated very much! Yours, Whoming From giannozz.nest at sns.it Thu Dec 18 09:10:38 2003 From: giannozz.nest at sns.it (Paolo Giannozzi) Date: Thu, 18 Dec 2003 09:10:38 +0100 Subject: [Pw_forum] Absoft f90 compiler In-Reply-To: References: Message-ID: <200312180910.38166.giannozz.nest@sns.it> On Thursday 18 December 2003 04:51, Raghani Pushpa wrote: > Dear users, > Does anybody have absoft f90 compiler, I need it urgently. Could it be > sent through email? If so, could you please send it to me? Pushpa, what you are asking for is illegal ... Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy *** please note my new phone and fax numbers *** From giannozz.nest at sns.it Thu Dec 18 09:16:51 2003 From: giannozz.nest at sns.it (Paolo Giannozzi) Date: Thu, 18 Dec 2003 09:16:51 +0100 Subject: [Pw_forum] pgf90 compile error on Linux PC In-Reply-To: <014e01c3c522$a3922610$192e77ca@whm> References: <014e01c3c522$a3922610$192e77ca@whm> Message-ID: <200312180916.51027.giannozz.nest@sns.it> On Thursday 18 December 2003 05:51, whoming wrote: > I have accounted complile errors when I try to install pwscf > v1.3.0 onto my Linux PC. My compiler is PGI 5.0. [...] > PGF90-S-0000-Internal compiler error. > Any suggestion will be appreciated very much! the suggestions you need are in the manual: -------------------------------------------------------------------- If you get ``internal compiler error'' or similar messages, lower the optimization level, or remove optimization, just for the routine that has problems. Some earlier versions of fortran-90 compilers were not especially reliable. If you have weird problems, try to use the most recent version of the compiler you can find, or to install patches for the compiler (most vendors release at least a few patches for free). -------------------------------------------------------------------- and: -------------------------------------------------------------------- \paragraph{Linux PCs with Portland Group compiler (pgf90)} Use the latest version of each release of the compiler, with patches if available: see the Portland Group web site, {\tt http://www.pgroup.com/faq/install.htm\#release\_info}. PWscf does not work reliably, or not at all, with some versions of the pgi compiler. It is not PWscf's fault. -------------------------------------------------------------------- Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy *** please note my new phone and fax numbers *** From baroni at sissa.it Thu Dec 18 09:35:50 2003 From: baroni at sissa.it (Stefano Baroni) Date: Thu, 18 Dec 2003 09:35:50 +0100 Subject: [Pw_forum] Absoft f90 compiler In-Reply-To: References: Message-ID: <3FE166E6.9050906@sissa.it> as Paolo already said, this is an illegal request. as far as i know, everybody can downlowad for free and legally the intel compiler from the intel web site S. Raghani Pushpa wrote: > Dear users, > Does anybody have absoft f90 compiler, I need it urgently. Could it be > sent through email? If so, could you please send it to me? > > Thanx, > Pushpa > -- Stefano Baroni SISSA and DEMOCRITOS National Simulation Center via Beirut 2-4, I34014 Trieste, Italy, [+39] 040 3787 406 From whoming at sohu.com Thu Dec 18 10:27:07 2003 From: whoming at sohu.com (whoming) Date: Thu, 18 Dec 2003 17:27:07 +0800 Subject: [Pw_forum] pgf90 compile error on Linux PC References: <014e01c3c522$a3922610$192e77ca@whm> <200312180916.51027.giannozz.nest@sns.it> Message-ID: <018201c3c549$21697630$192e77ca@whm> Dear Paolo, Thank you very much! And according to your suggestion, I have updated my compiler to PGI 5.1 and now it is ok! Best wishes! Yours, whoming ----- Original Message ----- From: "Paolo Giannozzi" To: Sent: Thursday, December 18, 2003 4:16 PM Subject: Re: [Pw_forum] pgf90 compile error on Linux PC On Thursday 18 December 2003 05:51, whoming wrote: > I have accounted complile errors when I try to install pwscf > v1.3.0 onto my Linux PC. My compiler is PGI 5.0. [...] > PGF90-S-0000-Internal compiler error. > Any suggestion will be appreciated very much! the suggestions you need are in the manual: -------------------------------------------------------------------- If you get ``internal compiler error'' or similar messages, lower the optimization level, or remove optimization, just for the routine that has problems. Some earlier versions of fortran-90 compilers were not especially reliable. If you have weird problems, try to use the most recent version of the compiler you can find, or to install patches for the compiler (most vendors release at least a few patches for free). -------------------------------------------------------------------- and: -------------------------------------------------------------------- \paragraph{Linux PCs with Portland Group compiler (pgf90)} Use the latest version of each release of the compiler, with patches if available: see the Portland Group web site, {\tt http://www.pgroup.com/faq/install.htm\#release\_info}. PWscf does not work reliably, or not at all, with some versions of the pgi compiler. It is not PWscf's fault. -------------------------------------------------------------------- Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy *** please note my new phone and fax numbers *** _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum From proffess at yandex.ru Fri Dec 19 13:31:02 2003 From: proffess at yandex.ru (Sergei Lisenkov) Date: Fri, 19 Dec 2003 15:31:02 +0300 (MSK) Subject: [Pw_forum] PWscf on ATHLON cluster In-Reply-To: <003201c387d1$ec7aeee0$ef3310ac@phys> References: <003201c387d1$ec7aeee0$ef3310ac@phys> Message-ID: <3FE2EF86.00000B.10911@soapbox.yandex.ru> Dear PWscf users, I tried to perform some calculations on ATHLON cluster (8 dual nodes) for carbon systems. I checked the diamond structure (8 atoms in unit cell). But I met the disconvergence adter 1 step: Program PWSCF v.1.3.1 starts ... Today is 19Dec2003 at 15:23: 4 Parallel version (MPI) Number of processors in use: 13 R & G space division: nprocp = 13 Ultrasoft (Vanderbilt) Pseudopotentials Current dimensions of program pwscf are: ntypx =10 npk =40000 lmax = 3 nchix = 6 ndim = 2000 nbrx = 8 nqfm = 8 noncolin = F gamma_only = F Planes per process (thick) : nr3 = 45 npp = 4 ncplane = 2025 Planes per process (smooth): nr3s= 25 npps= 2 ncplanes= 625 Proc/ planes cols G planes cols G columns G Pool (dense grid) (smooth grid) (wavefct grid) 1 4 98 2636 2 39 673 10 82 2 4 98 2636 2 39 669 10 82 3 4 98 2636 2 39 669 10 82 4 4 98 2636 2 39 669 10 80 5 4 100 2636 2 39 665 10 80 6 4 98 2636 2 39 665 10 80 7 3 98 2636 2 39 665 10 80 8 3 98 2636 2 39 661 10 80 9 3 98 2636 2 39 665 10 80 10 3 100 2636 2 40 678 10 80 11 3 99 2635 2 40 678 10 80 12 3 99 2635 2 39 677 9 79 13 3 99 2635 1 39 675 10 80 0 45 1281 34265 25 509 8709 129 1045 Title: Calculation of C-diamond, 8 atoms in unit cell, LDA bravais-lattice index = 1 lattice parameter (a_0) = 8.0000 a.u. unit-cell volume = 512.0000 (a.u.)^3 number of atoms/cell = 8 number of atomic types = 1 kinetic-energy cutoff = 25.0000 Ry charge density cutoff = 250.0000 Ry convergence threshold = 1.0E-01 beta = 0.7000 number of iterations used = 8 plain mixing Exchange-correlation = SLA PW GGX GGC (1422) iswitch = 3 nstep = **** celldm(1)= 8.00000 celldm(2)= 0.00000 celldm(3)= 0.00000 celldm(4)= 0.00000 celldm(5)= 0.00000 celldm(6)= 0.00000 crystal axes: (cart. coord. in units of a_0) a(1) = ( 1.0000 0.0000 0.0000 ) a(2) = ( 0.0000 1.0000 0.0000 ) a(3) = ( 0.0000 0.0000 1.0000 ) reciprocal axes: (cart. coord. in units 2 pi/a_0) b(1) = ( 1.0000 0.0000 0.0000 ) b(2) = ( 0.0000 1.0000 0.0000 ) b(3) = ( 0.0000 0.0000 1.0000 ) PSEUDO 1 is C (US) zval = 4.0 lmax= 1 lloc= 0 Version 0 0 0 of US pseudo code Using log mesh of 793 points The pseudopotential has 4 beta functions with: l(1) = 0 l(2) = 0 l(3) = 1 l(4) = 1 Q(r) pseudized with 8 coefficients, rinner = 0.600 0.600 0.600 atomic species valence mass pseudopotential C 4.00 12.01100 C ( 1.00) cell mass = 7.30181 UMA No symmetry! Cartesian axes site n. atom positions (a_0 units) 1 C tau( 1) = ( -0.2102321 -0.2102321 -0.2102321 ) 2 C tau( 2) = ( 0.2102321 -0.2102321 0.2102321 ) 3 C tau( 3) = ( 0.2102321 0.2102321 -0.2102321 ) 4 C tau( 4) = ( -0.2102321 0.2102321 0.2102321 ) 5 C tau( 5) = ( 0.0000000 0.0000000 0.0000000 ) 6 C tau( 6) = ( 0.0000000 0.4204642 0.4204642 ) 7 C tau( 7) = ( 0.4204642 0.0000000 0.4204642 ) 8 C tau( 8) = ( 0.4204642 0.4204642 0.0000000 ) number of k points= 1 cart. coord. in units 2pi/a_0 k( 1) = ( 0.0000000 0.0000000 0.0000000), wk = 2.0000000 G cutoff = 405.2847 ( 34265 G-vectors) FFT grid: ( 45, 45, 45) G cutoff = 162.1139 ( 8709 G-vectors) smooth grid: ( 25, 25, 25) nbndx = 32 nbnd = 16 natomwfc = 32 npwx = 82 nelec = 32.00 nkb = 64 ngl = 2636 Initial potential from superposition of free atoms Starting wfc are atomic total cpu time spent up to now is 1.52 secs iteration # 1 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization ethr = 1.00E-02, avg # of iterations = 2.8 total energy = -89.64033720 ryd estimated scf accuracy < 0.39762556 ryd total cpu time spent up to now is 2.22 secs iteration # 2 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization ethr = 1.24E-03, avg # of iterations = 4.8 total energy = -88.84743196 ryd estimated scf accuracy < 1.32539489 ryd total cpu time spent up to now is 3.21 secs iteration # 3 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization ethr = 1.24E-03, avg # of iterations = 4.1 total energy = -89.60469858 ryd estimated scf accuracy < 0.24082707 ryd total cpu time spent up to now is 4.15 secs iteration # 4 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization ethr = 7.53E-04, avg # of iterations = 3.5 ! total energy = -89.76070592 ryd estimated scf accuracy < 0.03414841 ryd convergence has been achieved Forces acting on atoms (Ry/au): atom 1 type 1 force = -0.04841235 -0.01107637 0.10650106 atom 2 type 1 force = 0.14285280 0.01861693 0.03950076 atom 3 type 1 force = 0.14141795 0.11840410 0.10854035 atom 4 type 1 force = -0.01643065 0.11076508 0.03417804 atom 5 type 1 force = -0.03470170 -0.03612651 -0.03027343 atom 6 type 1 force = 0.11134529 -0.12320537 -0.14589637 atom 7 type 1 force = -0.12834744 0.09751990 -0.16098688 atom 8 type 1 force = -0.16772389 -0.17489775 0.04843647 Total force = 0.510769 Total SCF correction = 0.277674 entering subroutine stress ... total stress (a.u.) (kbar) P= 71.14 0.00010302 -0.00000178 0.00056467 15.15 -0.26 83.07 -0.00000178 0.00039066 0.00074725 -0.26 57.47 109.92 0.00056467 0.00074725 0.00095718 83.07 109.92 140.81 Parrinello-Rahman Damped Cell-Dynamics Minimization convergence thresholds: EPSE = 0.10E-03 EPSF = 0.10E-02 EPSP = 0.50E+00 Entering Dynamics; it = 1 time = 0.00000 pico-seconds new lattice vectors (alat unit) : 1.000000083 -0.000000001 0.000000453 -0.000000001 1.000000313 0.000000599 0.000000453 0.000000599 1.000000767 new unit-cell volume = 512.0006 (a.u.)^3 new positions in cryst coord C -0.210232484 -0.210232199 -0.210231304 C 0.210233203 -0.210231973 0.210232416 C 0.210233192 0.210233017 -0.210231289 C -0.210232240 0.210232958 0.210232375 C -0.000000264 -0.000000275 -0.000000230 C 0.000000848 0.420463292 0.420463120 C 0.420463253 0.000000742 0.420463005 C 0.420462953 0.420462899 0.000000369 new positions in cart coord (alat unit) C -0.210232596 -0.210232391 -0.210231687 C 0.210233315 -0.210231914 0.210232546 C 0.210233114 0.210232956 -0.210231229 C -0.210232163 0.210233150 0.210232567 C -0.000000264 -0.000000275 -0.000000230 C 0.000001037 0.420463676 0.420463694 C 0.420463478 0.000000994 0.420463518 C 0.420462988 0.420463030 0.000000811 Ekin = 0.00000000 Ryd T = 0.0 K Etot = -89.76070592 NEW-OLD atomic charge density approx. for the potential NEW K-POINTS 0.0000000 0.0000000 0.0000000 2.0000000 total cpu time spent up to now is 7.11 secs iteration # 1 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization warning : 2 eigenvectors not converged after 4 attemps ethr = 7.53E-04, avg # of iterations = 68.3 integrated charge = 7970.93579446 total energy =*************** ryd estimated scf accuracy <*************** ryd total cpu time spent up to now is 13.69 secs iteration # 2 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization warning : 3 eigenvectors not converged after 2 attemps ethr = 1.00E-02, avg # of iterations = 31.8 integrated charge = 0.12451317 total energy =-68582.00853807 ryd estimated scf accuracy <*************** ryd total cpu time spent up to now is 17.14 secs iteration # 3 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization warning : 4 eigenvectors not converged after 1 attemps ethr = 1.00E-02, avg # of iterations = 16.4 integrated charge = 0.19686936 total energy =-62713.31117503 ryd estimated scf accuracy <*************** ryd total cpu time spent up to now is 19.27 secs iteration # 4 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization warning : 1 eigenvectors not converged after 1 attemps ethr = 1.00E-02, avg # of iterations = 9.1 integrated charge = 0.20961502 total energy =-61801.97633384 ryd estimated scf accuracy <*************** ryd total cpu time spent up to now is 20.68 secs iteration # 5 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization ethr = 1.00E-02, avg # of iterations = 3.3 integrated charge = 0.20978234 total energy =-61555.06003161 ryd estimated scf accuracy <*************** ryd total cpu time spent up to now is 21.55 secs iteration # 6 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization ethr = 1.00E-02, avg # of iterations = 9.8 integrated charge = 0.22364973 total energy =-60829.05367412 ryd estimated scf accuracy <*************** ryd total cpu time spent up to now is 23.05 secs iteration # 7 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization warning : 3 eigenvectors not converged after 1 attemps ethr = 1.00E-02, avg # of iterations = 15.9 integrated charge = 0.27674792 total energy =-57151.36271893 ryd estimated scf accuracy <*************** ryd total cpu time spent up to now is 25.13 secs iteration # 8 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization warning : 2 eigenvectors not converged after 1 attemps ethr = 1.00E-02, avg # of iterations = 14.9 integrated charge = 0.32040480 total energy =-54820.11646643 ryd estimated scf accuracy <156960.66204163 ryd total cpu time spent up to now is 27.17 secs iteration # 9 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization ethr = 1.00E-02, avg # of iterations = 5.2 integrated charge = 0.32198903 total energy =-54790.74751836 ryd estimated scf accuracy <127278.85530751 ryd total cpu time spent up to now is 28.26 secs iteration # 10 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization ethr = 1.00E-02, avg # of iterations = 3.0 integrated charge = 0.32181453 total energy =-54828.94883453 ryd estimated scf accuracy <161904.22683110 ryd total cpu time spent up to now is 29.11 secs iteration # 11 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization ethr = 1.00E-02, avg # of iterations = 5.6 integrated charge = 0.32589278 total energy =-54283.70923535 ryd estimated scf accuracy < 0.13975491 ryd total cpu time spent up to now is 30.21 secs iteration # 12 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization warning : 2 eigenvectors not converged after 1 attemps ethr = 4.37E-04, avg # of iterations = 12.7 integrated charge = 0.33026094 total energy =-54284.75510936 ryd estimated scf accuracy < 0.16240115 ryd total cpu time spent up to now is 32.02 secs iteration # 13 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization ethr = 4.37E-04, avg # of iterations = 3.1 integrated charge = 0.33024571 total energy =-54284.75980825 ryd estimated scf accuracy < 0.16311350 ryd total cpu time spent up to now is 32.88 secs iteration # 14 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization ethr = 4.37E-04, avg # of iterations = 3.0 integrated charge = 0.33023441 total energy =-54284.75675028 ryd estimated scf accuracy < 0.15355886 ryd total cpu time spent up to now is 33.78 secs iteration # 15 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization ethr = 4.37E-04, avg # of iterations = 3.0 integrated charge = 0.33023653 total energy =-54284.70372435 ryd estimated scf accuracy < 0.15288852 ryd total cpu time spent up to now is 34.63 secs iteration # 16 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization ethr = 4.37E-04, avg # of iterations = 3.0 integrated charge = 0.33022071 total energy =-54284.66090698 ryd estimated scf accuracy < 0.26644347 ryd total cpu time spent up to now is 35.47 secs iteration # 17 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization ethr = 4.37E-04, avg # of iterations = 3.0 integrated charge = 0.33022050 total energy =-54284.63327082 ryd estimated scf accuracy < 0.34142825 ryd total cpu time spent up to now is 36.38 secs iteration # 18 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization ethr = 4.37E-04, avg # of iterations = 3.0 integrated charge = 0.33020502 total energy =-54284.67953576 ryd estimated scf accuracy < 0.19171106 ryd total cpu time spent up to now is 37.26 secs iteration # 19 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization ethr = 4.37E-04, avg # of iterations = 3.0 integrated charge = 0.33020680 ! total energy =-54284.74980760 ryd estimated scf accuracy < 0.03935158 ryd convergence has been achieved Forces acting on atoms (Ry/au): atom 1 type 1 force = ****************************************** atom 2 type 1 force = ****************************************** atom 3 type 1 force = ****************************************** atom 4 type 1 force = ****************************************** atom 5 type 1 force = 39653.05856598**************************** atom 6 type 1 force = -4050.39023863**************************** atom 7 type 1 force = ****************************************** atom 8 type 1 force = ****************************************** Total force = ************ Total SCF correction = ************ entering subroutine stress ... total stress (a.u.) (kbar) P=******** *************************************** ****************************** *************************************** ****************************** *************************************** ****************************** Entering Dynamics; it = 2 time = 0.00005 pico-seconds new lattice vectors (alat unit) : ?????????????????????????????????????????? ?????????????????????????????????????????? ?????????????????????????????????????????? new unit-cell volume =???????????? (a.u.)^3 new positions in cryst coord C ?????????????????????????????????????????? C ?????????????????????????????????????????? C ?????????????????????????????????????????? C ?????????????????????????????????????????? C ?????????????????????????????????????????? C ?????????????????????????????????????????? C ?????????????????????????????????????????? C ?????????????????????????????????????????? new positions in cart coord (alat unit) C ?????????????????????????????????????????? C ?????????????????????????????????????????? C ?????????????????????????????????????????? C ?????????????????????????????????????????? C ?????????????????????????????????????????? C ?????????????????????????????????????????? C ?????????????????????????????????????????? C ?????????????????????????????????????????? Ekin = ?????????????? Ryd T = ?????? K Etot = ?????????????? NEW-OLD atomic charge density approx. for the potential Warning: negative or imaginary starting charge ???????????????????????? 1 NEW K-POINTS ???????????????????????????????????? 2.0000000 Warning: negative or imaginary starting charge ???????????????????????? 1 total cpu time spent up to now is 40.39 secs iteration # 1 ecut= 25.00 ryd beta=0.70 Conjugate-gradient style diagonalization ethr = 4.37E-04, avg # of iterations = 3.0 ! total energy =??????????????? ryd estimated scf accuracy <3FE2EF86.00000B.10911@soapbox.yandex.ru> Message-ID: <3FE2F511.9070209@sissa.it> I don't know wheter there are other problems in your calculation but your convergence threshold of 1.d-1 looks to me extremely poor... forces and stress are therefore probably completely wrong. stefano From weihe at mail.ustc.edu.cn Sat Dec 27 07:51:49 2003 From: weihe at mail.ustc.edu.cn (=?GB2312?B?utjOsA==?=) Date: Sat, 27 Dec 2003 14:51:49 +0800 Subject: [Pw_forum] a small problem Message-ID: <200312270649.OAA26446@mx1.ustc.edu.cn> Dear all: Happy new year! Now I have a small problem.When I run pw.x to make a relax calculation,in the fourth step of geometry relaxation, there is an error message : linmin: 3rd order interpolation %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from linmin : error # 2 unexpected error %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... Would somebody tell me the possible reason for this error? Thanks! ???????? Best Wishes Wei He ???????? ?????? From giannozz at nest.sns.it Sat Dec 27 14:58:23 2003 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Sat, 27 Dec 2003 14:58:23 +0100 Subject: [Pw_forum] a small problem In-Reply-To: <200312270649.OAA26446@mx1.ustc.edu.cn> References: <200312270649.OAA26446@mx1.ustc.edu.cn> Message-ID: <200312271458.23729.giannozz@nest.sns.it> On Saturday 27 December 2003 07:51, ?? wrote: > When I run pw.x to make a relax calculation, in the fourth step > of geometry relaxation, there is an error message [...] > from linmin : [...] unexpected error > Would somebody tell me the possible reason for this error? it is explained in the"Troubleshooting" section of the manual. Currently there is some ongoing work aimed at producing a more robust geometry optimizer. Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy *** please note my new phone and fax numbers *** From weihe at mail.ustc.edu.cn Sat Dec 27 15:25:15 2003 From: weihe at mail.ustc.edu.cn (=?gb2312?B?utjOsA==?=) Date: Sat, 27 Dec 2003 22:25:15 +0800 Subject: [Pw_forum] a small problem Message-ID: <200312271422.WAA03852@mx1.ustc.edu.cn> Dear Prof.Paolo Giannozzi: Thank you very much! ======= 2003-12-27 21:58:23 ????????======= >On Saturday 27 December 2003 07:51, ?? wrote: > >> When I run pw.x to make a relax calculation, in the fourth step >> of geometry relaxation, there is an error message [...] >> from linmin : [...] unexpected error >> Would somebody tell me the possible reason for this error? > >it is explained in the"Troubleshooting" section of the manual. >Currently there is some ongoing work aimed at producing a >more robust geometry optimizer. > >Paolo > >-- >Paolo Giannozzi e-mail: giannozz at nest.sns.it >Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 >Piazza dei Cavalieri 7 I-56126 Pisa, Italy >*** please note my new phone and fax numbers *** > >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum = = = = = = = = = = = = = = = = = = = = Sincerely Wei He