From hhy74 at hotmail.com Sat Jun 1 16:09:30 2002 From: hhy74 at hotmail.com (Hou Huaiyu) Date: Sat, 01 Jun 2002 22:09:30 +0800 Subject: [CPMD-list] ask for help - compiling question Message-ID: Hello, I cannot compile the CPMDV3.5 code on DEC-alpha (500au personal workstation). The error message is: " can't locate files for -ldxml." Could anyone help me, please? Hou Huaiyu _________________________________________________________________ ???????? MSN Explorer??http://explorer.msn.com/lccn/intl.asp From tdeutsch at cea.fr Mon Jun 3 11:18:28 2002 From: tdeutsch at cea.fr (Thierry Deutsch) Date: Mon, 3 Jun 2002 11:18:28 +0200 Subject: [CPMD-list] ask for help - compiling question In-Reply-To: References: Message-ID: <15611.13412.962649.404095@absinthe.ceng.cea.fr> > Hou Huaiyu ?crivit le 1/6/2002 : >\ > | Hello, > | I cannot compile the CPMDV3.5 code on DEC-alpha (500au personal > | workstation). The error message is: > | " can't locate files for -ldxml." > | Could anyone help me, please? > | > | Hou Huaiyu > | > | _________________________________________________________________ > | ???????? MSN Explorer??http://explorer.msn.com/lccn/intl.asp > | > | > | Hi Hou, dxml is the library that contains LAPACK and another routines (FFT). Normally if you use the fortran from Compaq (which bought DEC and which is bought by HP) you have only to specify -ldxml. In a DEC-alpha Workstation 500, the location is: /usr/shlib (/usr/shlib/libdxml.so) I do not know where the static library libdxml is. Thierry Deutsch -- ----------------------------------------------------------------------- Thierry Deutsch Laboratoire de simulation atomistique (L_Sim) DRFMC/SP2M T?l:(33) 04 38 78 34 06 C.E.A.Grenoble Fax:(33) 04 38 78 51 97 17, Avenue des Martyrs mailto:TDeutsch at cea.fr 38054 GRENOBLE CEDEX 9 FRANCE http://www-drfmc.cea.fr/SP2M ----------------------------------------------------------------------- From M.A.Zwijnenburg at tnw.tudelft.nl Mon Jun 3 13:26:58 2002 From: M.A.Zwijnenburg at tnw.tudelft.nl (Martijn Zwijnenburg) Date: Mon, 03 Jun 2002 13:26:58 +0200 Subject: [CPMD-list] Warning Message-ID: <3CFB6E83.30295.8ADE34C8@localhost> Hi, I'm optimizing quartz (both ion positions and cell volume) using CPMD and get the warning message: Warning! Spline region smaller then maximum G-value. The optimization seems to converge nicely but what does this warning mean / imply? Cheers, Martijn ------------------------------------------------------------------------- Martijn Zwijnenburg Lab. of Applied Organic Chemistry and Catalysis Delft University of Technology Julianalaan 136 2628 BL Delft The Netherlands Tel: 0031-(0)152782691 Fax: 0031-(0)152784700 e-mail: M.A.Zwijnenburg at tnw.tudelft.nl web page: http://come.to/tock From hutter at pci.unizh.ch Mon Jun 3 20:48:45 2002 From: hutter at pci.unizh.ch (Juerg Hutter) Date: Mon, 3 Jun 2002 20:48:45 +0200 (MEST) Subject: [CPMD-list] Warning In-Reply-To: <3CFB6E83.30295.8ADE34C8@localhost> Message-ID: Hi this in fact touches several points. 1) Pseudopotential functions in CPMD are calculated on a radial grid in G-space and then used in spline interpolations. This speeds up variable cell calculations considerably. The maximum grid point is given by the cutoff. With the keyword SPLINE RANGE x.xx you can enlarge the grid to x.xx times the cutoff. BTW. the default number of spline points is 501. This is at the lower limit for accuracy. Especially if you have high cutoffs it is better to increase this value, e.g. SPLINE POINTS 2500 2) In a variable cell calculation, CPMD uses a constant number of plane waves. Therefore if your cell contracts the cutoff increases, if the cell gets larger the cutoff decreases. So if you have the first case the spline interpolation needs points above the original cutoff and you get a warning. Depending on the amount of change in the cell you expect a value for the range of 2 - 5 is needed. 3) Comming back to the cunstant number of plane waves. If your cell gets larger the effective cutoff decreases. This may have very undesirable effects and it is better to define the plane waves on a box larger than any box you anticipate the simulation will reach. In order not to have to start with an unreasonable cell you can define the plane waves not with the actual box but with a reference cell, use the keyword REFERENCE CELL a b c xa xb xc 4) However, you really want to do a constant cutoff calculation, not a constant number of plane waves. For technical reasons this is not possible and in principle you should do the calculation at a high enough cutoff in order that the calculation is converged all along the simulation path (with the slight changes in cutoff). To avoid this very high cutoffs the group in Trieste came up with a method that allows to perform pseudo constant cutoff calculations. This method is implemented in CPMD and explained in this paper M. Bernasconi et al. J. Phys. Chem Solids 56 pp 501-505 (1995) Hope this explains some of the issues of variable cell calculations with CPMD regards Juerg ---------------------------------------------------------- Juerg Hutter Phone : ++41 1 635 4491 Physical Chemistry Institute FAX : ++41 1 635 6838 University of Zurich E-mail: hutter at pci.unizh.ch Winterthurerstrasse 190 CH-8057 Zurich, Switzerland ---------------------------------------------------------- On Mon, 3 Jun 2002, Martijn Zwijnenburg wrote: > Hi, > > I'm optimizing quartz (both ion positions and cell volume) using > CPMD and get the warning message: Warning! Spline region > smaller then maximum G-value. The optimization seems to > converge nicely but what does this warning mean / imply? > > Cheers, > Martijn > ------------------------------------------------------------------------- > Martijn Zwijnenburg > Lab. of Applied Organic Chemistry and Catalysis > Delft University of Technology > Julianalaan 136 > 2628 BL Delft > The Netherlands > Tel: 0031-(0)152782691 > Fax: 0031-(0)152784700 > e-mail: M.A.Zwijnenburg at tnw.tudelft.nl > web page: http://come.to/tock > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > From hwsheng at jhu.edu Thu Jun 6 05:17:18 2002 From: hwsheng at jhu.edu (HW Sheng) Date: Wed, 05 Jun 2002 23:17:18 -0400 Subject: [CPMD-list] Lanthanides References: Message-ID: <000701c20d08$aa9cd6e0$a567dc80@mrs.jhu.edu> Hi, all; In an early dialogue between Juerg Hutter and Gabriele balducci (May, 2), I noticed that the limitation of Vanderbilt type pseudopotential to handle lanthanides in CPMD. I then chose Stefan-Godecker type pp to work on f electrons, specifically, lanthanum. It turned out that there is no problem running Wavefunction Optimization, etc. But when I switched to Molecular Dynamics employing Parrinello-Rahman algorithm (NPT), it prompted that "PROGRAM STOPS IN SUBROUTINE DYLMR | L NOT PROGRAMMED". After checking the source code, I found there is some problem in putbet.f, where the BETAA array is derived. So, CPMD can not handle f electrons after all. am I correct about this? Thanks, Howard Sheng From hutter at pci.unizh.ch Thu Jun 6 12:06:13 2002 From: hutter at pci.unizh.ch (Juerg Hutter) Date: Thu, 6 Jun 2002 12:06:13 +0200 (MEST) Subject: [CPMD-list] Lanthanides In-Reply-To: <000701c20d08$aa9cd6e0$a567dc80@mrs.jhu.edu> Message-ID: Hi you are right. The derivatives of f spherical harmonics needed in the calculation of the stress tensor are not implemented in CPMD. You can still perform normal CP simulations with pseudopotentials having f nonlocal components. regards Juerg PS: It's not very difficult to extend the routine dylmr to include f functions. ---------------------------------------------------------- Juerg Hutter Phone : ++41 1 635 4491 Physical Chemistry Institute FAX : ++41 1 635 6838 University of Zurich E-mail: hutter at pci.unizh.ch Winterthurerstrasse 190 CH-8057 Zurich, Switzerland ---------------------------------------------------------- On Wed, 5 Jun 2002, HW Sheng wrote: > Hi, all; > > In an early dialogue between Juerg Hutter and Gabriele balducci (May, 2), I > noticed that the limitation of Vanderbilt type pseudopotential to handle > lanthanides in CPMD. I then chose Stefan-Godecker type pp to work on f > electrons, specifically, lanthanum. It turned out that there is no problem > running Wavefunction Optimization, etc. But when I switched to Molecular > Dynamics employing Parrinello-Rahman algorithm (NPT), it prompted that > "PROGRAM STOPS IN SUBROUTINE DYLMR | L NOT PROGRAMMED". After checking the > source code, I found there is some problem in putbet.f, where the BETAA > array is derived. > > So, CPMD can not handle f electrons after all. am I correct about this? > > Thanks, > > Howard Sheng > > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > From carloni at sissa.it Thu Jun 6 19:05:44 2002 From: carloni at sissa.it (Paolo Carloni) Date: Thu, 06 Jun 2002 19:05:44 +0200 Subject: [CPMD-list] Poistion at the INFM-DEMOCRITOS Center, Trieste, Italy Message-ID: <3CFF9668.2080001@sissa.it> Dear Collegue, can you please advertise the following position. Thank you Paolo Carloni ------------------------------------------------------------------------------------------ Italian Institute for the Physics of Matter (INFM, www.infm.it) DEMOCRITOS CENTER FOR NUMERICAL SIMULATIONS (Trieste, Italy) An opening for a JUNIOR RESEARCHER, tenure-track position in Computational Chemistry and Theoretical Biophysics is available starting from January 2003. The successful candidate will have a sizeable record in Car-Parrinello Molecular Dynamics (CPMD) and/or hybrid CPMD/Molecular Mechanics calculations on biomimetic compounds and/or metal complexes with catalytic properties as well as on biological systems. He/she is expected to have a PhD in Computational Chemistry or Computational (Bio)Physics, and a postdoctoral experience in the above fields. The Center encourages young scientists (less than 30) to apply for this position. He/she is expected to interact actively with computational physicists/chemists in a highly interdisciplinary environment. Please send CV and contact information with (at least) two references to: Prof. Paolo Carloni INFM-DEMOCRITOS and International School for Advanced Studies (SISSA/ISAS) Via Beirut 2-4 34100 Trieste, Italy Email: carloni at sissa.it The DEMOCRITOS center has been very recently established by INFM (Italian Institute for the Physics of Matter). The aim of the center is to investigate properties of matter by using computer simulation methods. DEMOCRITOS, which is located at the International School for Advanced Studies (SISSA, Trieste, Italy), is a highly interdisciplinary center constituted by about 40 researchers. -------------- next part -------------- A non-text attachment was scrubbed... Name: ad.doc Type: application/msword Size: 20992 bytes Desc: not available Url : http://cpmd.org/pipermail/cpmd-list/attachments/20020606/dbab5bba/attachment.doc From kresslists at kressworks.com Thu Jun 6 19:31:19 2002 From: kresslists at kressworks.com (Jim Kress) Date: Thu, 6 Jun 2002 13:31:19 -0400 Subject: [CPMD-list] Making CPMD with only Portland Group compilers Message-ID: <000f01c20d7f$f895cdd0$0401a8c0@quantumnt> Does anyone have a CPMD makefile that uses ONLY the Portland Group Fortran, C, and C++ compilers to make CPMD? If yes, could I please get a copy? I will summarize responses to the list. Thanks. Jim From ccdiazt at kaliman.iimatercu.unam.mx Fri Jun 7 17:38:55 2002 From: ccdiazt at kaliman.iimatercu.unam.mx (=?ISO-8859-1?Q?C=E9sar?= Carlos =?ISO-8859-1?Q?D=EDaz_Torrej=F3n?=) Date: 07 Jun 2002 10:38:55 -0500 Subject: [CPMD-list] Re: CPMD-list digest, Vol 1 #63 - 1 msg In-Reply-To: <200206071001.MAA17742@cpmd.org> References: <200206071001.MAA17742@cpmd.org> Message-ID: <1023464335.24882.3.camel@kaliman.iimatercu.unam.mx> Hello All: This Makefile was I use for compiling using Portlan Group Compilers # Author: cluster.iimatercu.unam.mx!ccdiazt #---------------------------------------------------------------------------- # SHELL = /bin/sh # #--------------- Default Configuration for PC-PGI-MPI --------------- SRC = . DEST = . BIN = . #QMMM_FLAGS = -D__QMECHCOUPL #QMMM_LIBS = -L. -lmm FFLAGS = -Mr8 -pc64 -byteswapio -Msignextend -Msecond_underscore LFLAGS = -L/usr/pgi/linux86/lib -llapack \ -L/usr/local/lib/ATLAS -lf77blas -latlas -byteswapio CFLAGS = CPP = /lib/cpp -P -C -traditional CPPFLAGS = -D__Linux -D__PGI -DLAPACK -DFFT_DEFAULT \ -DPARALLEL -DMP_LIBRARY=__MPI CC = mpicc -O2 -Wall FC = mpif77 -c -fast LD = mpif77 -fast AR = #---------------------------------------------------------------------------- The rest of the file is the same. mpif77 using pgf90 for compiling, I export the next environment variable in my .bashrc export LAMHF77=pgf90 Best Regards C.C.D?az On Fri, 2002-06-07 at 05:01, cpmd-list-request at cpmd.org wrote: > Send CPMD-list mailing list submissions to > cpmd-list at cpmd.org > > To subscribe or unsubscribe via the World Wide Web, visit > http://www.cpmd.org/mailman/listinfo/cpmd-list > or, via email, send a message with subject or body 'help' to > cpmd-list-request at cpmd.org > > You can reach the person managing the list at > cpmd-list-admin at cpmd.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of CPMD-list digest..." > > ---- > > Today's Topics: > > 1. Making CPMD with only Portland Group compilers (Jim Kress) > ---- > > From: Jim Kress > To: CPMD-list at cpmd.org > Subject: [CPMD-list] Making CPMD with only Portland Group compilers > Date: 06 Jun 2002 13:31:19 -0400 > > Does anyone have a CPMD makefile that uses ONLY the Portland Group Fortran, > C, and C++ compilers to make CPMD? If yes, could I please get a copy? > > I will summarize responses to the list. > > Thanks. > > Jim > > > > ---- > > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > -- _________________________________________________________ F?s. C?sar Carlos D?az Torrej?n Administrador de Sistemas Unix Instituto de Investigaciones en Materiales, UNAM Oficina: A-007 e-mail: ccdiazt at kaliman.iimatercu.unam.mx tel: (5)6224611 (oficina) ________________________________________________________ From sdecker at uwo.ca Mon Jun 10 16:54:53 2002 From: sdecker at uwo.ca (sdecker) Date: Mon, 10 Jun 2002 10:54:53 -0400 Subject: [CPMD-list] Monitoring Pressure Message-ID: <033501c2108e$c7c349a0$293d6481@woolabpcvsj70k> Hi, I am currently using CPMD to examine the properties of compressed liquids and I would like to know how to monitor the pressure during the NVT simulations? In a previous version of CPMD, 3.3a, you could use the STRESS keyword to print out the entire stress tensor every nth MD step, however, it appears that this keyword has been disabled in newer releases of CPMD. Any help would be greatly appreciated. Thanks in advance, Stephen Decker _________________________________________________ Dr. Stephen A. Decker, Ph.D. NSERC Postdoctoral Fellow Department of Chemistry University of Western Ontario London, Ontario, Canada N6A 5B7 phone: (519)-661-2111 ext. 86353 fax: (519)-661-3022 email: sdecker at uwo.ca _________________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://cpmd.org/pipermail/cpmd-list/attachments/20020610/20b477ce/attachment.html From roland at gatubela.physics.ncsu.edu Sun Jun 2 21:49:50 2002 From: roland at gatubela.physics.ncsu.edu (Christopher Roland) Date: Sun, 2 Jun 2002 15:49:50 -0400 Subject: [CPMD-list] compiling cpmd with intel mkl Message-ID: <200206021949.g52Jnom03026@gatubela.physics.ncsu.edu> Hi Everyone, I am relatively new to the CPMD code, and have been working to install it a Linux PC running Red Hat7.3 (dual processor PIII machine). I am running the intel fortran compiler (IFC), and have been running into the compilation problems already noted in the mail list. Based on the advice given, I have installed intel's MKL package. However, I still am unable to get anything to work. Can anyone who has gone through this process send me or post the first part of the Makefile? That would really help me a lot. Thanks! chris roland (roland at gatubela.physics.ncsu.edu) From perand at foi.se Mon Jun 3 12:11:30 2002 From: perand at foi.se (Per Andersson) Date: Mon, 03 Jun 2002 12:11:30 +0200 Subject: [CPMD-list] ask for help - compiling question Message-ID: <3CFB40D2.10806@foi.se> Hou Huaiyu ?crivit le 1/6/2002 : >\ > | Hello, > | I cannot compile the CPMDV3.5 code on DEC-alpha (500au personal > | workstation). The error message is: > | " can't locate files for -ldxml." > | Could anyone help me, please? > | > | Hou Huaiyu > | > | _________________________________________________________________ > | ???????? MSN Explorer??http://explorer.msn.com/lccn/intl.asp > | > | > | >Hi Hou, > dxml is the library that contains LAPACK and another >routines (FFT). Normally if you use the fortran from Compaq >(which bought DEC and which is bought by HP) you have only >to specify -ldxml. >In a DEC-alpha Workstation 500, the location is: >/usr/shlib >(/usr/shlib/libdxml.so) >I do not know where the static library libdxml is. >Thierry Deutsch Please notice that the new name of the library is cxml (Digtal eXtended Math Library -> Compaq eXtended Math Library) If you have a Fortran compiler from Compaq you can link using -lcxml without specifing the path unless you have moved things around yourself. /Per Andersson Swedish Defense Research Agency From h33990l7 at rz.hu-berlin.de Wed Jun 5 16:49:28 2002 From: h33990l7 at rz.hu-berlin.de (Veronika Brazdova) Date: Wed, 5 Jun 2002 16:49:28 +0200 (MET DST) Subject: [CPMD-list] Dipole moment for periodic systems in CPMD Message-ID: Dear all, I am trying to get the dipole moment calculated for each atom displacement in a finite differences run, so as to be able to calculate the intensities of the lines in a vibrational spectrum of a periodic system. I'll be glad if anyone can point me in the direction of references that would help me to understand the theory behind the routines dipo and ddipo. Is it a good idea to start with the routine ddipo and modify it so that it can be called from the secder routine? Regards Veronika Brazdova _______________________________________________________________________________ Veronika Brazdova Humboldt University, Institute of Chemistry, Brook-Taylor-Strasse 2, 12489, Berlin vb at chemie.hu-berlin.de http://atrey.karlin.mff.cuni.cz/~verunka phone: +49 30 2093 7140 fax: +49 03 2093 7136 _______________________________________________________________________________ From b.zarychta at phys.keele.ac.uk Wed Jun 5 17:48:36 2002 From: b.zarychta at phys.keele.ac.uk (Zak) Date: Wed, 5 Jun 2002 16:48:36 +0100 Subject: [CPMD-list] Scratch array too small Message-ID: <003201c20ca8$749a3f20$ae5005a0@phys.keele.ac.uk> Dear all, I am trying too run a geometry optimisation on solid benzene. The super cell consists of a tetramoleculkar unit of orthohombic dimensions I get the following error. PROGRAM STOPS IN SUBROUTINE RGDIIS | SCRATCH ARRAY TOO SMALL. Can anyone tell me how I might avoid this. any help would be greatly appreciated, Zak -------------- next part -------------- An HTML attachment was scrubbed... URL: http://cpmd.org/pipermail/cpmd-list/attachments/20020605/6a9579e2/attachment.html From natasa.mateljevic at yale.edu Tue Jun 11 18:12:44 2002 From: natasa.mateljevic at yale.edu (Natasa Mateljevic) Date: Tue, 11 Jun 2002 12:12:44 -0400 (EDT) Subject: [CPMD-list] optimization on graphite Message-ID: Hi, I am having problems optimizing the geometry of graphite, cpmd totally reshuffles things around and the optimizied structure is not graphite. I am wondering if anybody has done that successfully and could help me out. Thanks, Natasa From sdecker at uwo.ca Tue Jun 11 22:39:47 2002 From: sdecker at uwo.ca (sdecker) Date: Tue, 11 Jun 2002 16:39:47 -0400 Subject: [CPMD-list] Free Programs to Calculate Pair Correlation Functions Message-ID: <011d01c21188$207de060$293d6481@woolabpcvsj70k> Hello All, I was wondering if there are any freely available programs which will calculate pair correlation functions from the TRAJECTORY file created with CPMD? Thanks in advance for your assistance. Sincerely, Stephen Decker _________________________________________________ Dr. Stephen A. Decker, Ph.D. NSERC Postdoctoral Fellow Department of Chemistry University of Western Ontario London, Ontario, Canada N6A 5B7 phone: (519)-661-2111 ext. 86353 fax: (519)-661-3022 email: sdecker at uwo.ca _________________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://cpmd.org/pipermail/cpmd-list/attachments/20020611/fb229629/attachment.html From hwsheng at jhu.edu Thu Jun 13 20:43:17 2002 From: hwsheng at jhu.edu (HW Sheng) Date: Thu, 13 Jun 2002 14:43:17 -0400 Subject: [CPMD-list] Temperature Control References: <011d01c21188$207de060$293d6481@woolabpcvsj70k> Message-ID: <000b01c2130a$2f413b00$a567dc80@mrs.jhu.edu> Hi there, I encountered a temperature control problem in performing NPT molecular dynamics simulation. It seems that the cell "temperature" (cell dynamics) can not be well controlled via velocity scaling. In other words, the keyword TEMPCONTROL CELL is invalid if not coupled with the use of "TEMPCONTROL ELECTRONS". Any reasons behind? It's not difficult to modify the source code to validate TEMPCONTROL CELL without invoking TEMPCONTROL ELECTRONS. But, if doing so, are there any hidden problems? Any suggestions would be greatly appreciated. Howard Sheng -------------- next part -------------- An HTML attachment was scrubbed... URL: http://cpmd.org/pipermail/cpmd-list/attachments/20020613/f3539b81/attachment.html From sdecker at uwo.ca Thu Jun 13 21:29:45 2002 From: sdecker at uwo.ca (sdecker) Date: Thu, 13 Jun 2002 15:29:45 -0400 Subject: [CPMD-list] Problem with Calculating Stress Tensor Message-ID: <004d01c21310$ac8d6fb0$293d6481@woolabpcvsj70k> Hi All, I am having some problems calculating the stress tensor during an NVE simulation of a liquid with CPMD. I am able to calculate the stress tensors for a test system of liquid H2O with 32 molecules in the simulation cell with CPMD versions 3.4.3 and 3.5.1. However, when I try to calculate the stress tensor for my system of interest: liquid nitromethane (CH3NO2) with 32 molecules in the simulation cell I cannot do so. With CPMD version 3.4.3 my job exits with an error stating that more programming is required, and with CPMD version 3.5.1 my job becomes frozen during the initialization right after generating the atomic basis set and just before it goes through some sort of initialization routine for the stress tensor. Has anyone encountered a similar problem. Since I am able to calculate the stress tensor for the smaller liquid H2O system it seems like this may be due to a restriction on the system size. Your help would be greatly appreciated. Sincerely, Stephen Decker _________________________________________________ Dr. Stephen A. Decker, Ph.D. NSERC Postdoctoral Fellow Department of Chemistry University of Western Ontario London, Ontario, Canada N6A 5B7 phone: (519)-661-2111 ext. 86353 fax: (519)-661-3022 email: sdecker at uwo.ca _________________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://cpmd.org/pipermail/cpmd-list/attachments/20020613/39b5d612/attachment.html From tdeutsch at cea.fr Fri Jun 14 11:48:56 2002 From: tdeutsch at cea.fr (Thierry Deutsch) Date: Fri, 14 Jun 2002 11:48:56 +0200 Subject: [CPMD-list] compiling cpmd with intel mkl In-Reply-To: <200206021949.g52Jnom03026@gatubela.physics.ncsu.edu> References: <200206021949.g52Jnom03026@gatubela.physics.ncsu.edu> Message-ID: <15625.48136.847579.766765@absinthe.ceng.cea.fr> Hi Christopher, I send few weeks ago some lines to include in the Configure file in order to compile with efc and mkl. If you modify the "e" of "efc" and replace it by "i", you have the following lines: echo 'We use the Intel Library Kernel.' >&2 if [ "${MKL_LIB}" = "" ]; then echo "The variable MKL_LIB gives the location of this library." >&2 echo "The variable MKL_LIB does not exist or is empty." >&2 exit 2 fi IRAT=2 CFLAGS='' CPP='/lib/cpp -P -C -traditional' CPPFLAGS='-D__Linux -D__PGI -DLAPACK -DFFT_DEFAULT -DLINUX_IFC' FFLAGS=' -pc64 -w90 -w95' LFLAGS='-L${MKL_LIB} -lmkl_lapack -lmkl_itp -lpthread -Vaxlib' FFLAGS_GROMOS='-Dgood_luck' if [ $debug ]; then FC='ifc -c -g' CC='gcc -g -Wall' LD='ifc -g' else FC='ifc -c -O3 -tp p6 -axiM' CC='gcc -O2 -Wall' LD='ifc -O3 -tp p6 -axiM' fi By default the location of mkl is: /opt/intel/ml/lib/32. I hope that this message is helpfull. Thierry Deutsch -- ----------------------------------------------------------------------- Thierry Deutsch Laboratoire de simulation atomistique (L_Sim) DRFMC/SP2M T?l:(33) 04 38 78 34 06 C.E.A.Grenoble Fax:(33) 04 38 78 51 97 17, Avenue des Martyrs mailto:TDeutsch at cea.fr 38054 GRENOBLE CEDEX 9 FRANCE http://www-drfmc.cea.fr/SP2M ----------------------------------------------------------------------- From R.Jones at fz-juelich.de Fri Jun 14 15:02:21 2002 From: R.Jones at fz-juelich.de (R.Jones) Date: Fri, 14 Jun 2002 15:02:21 +0200 Subject: [CPMD-list] sdlinres.F Message-ID: <1020614150221.ZM24182@iff856.iff.kfa-juelich.de> The distributed version of sdlinresp.F does not allow the program to restart if a linear response calculation of vibration frequencies is incomplete. Finite difference calculations can be restarted. The linear response calculation does restart if two lines are copied from the corresponding place in secder.F. If, after 300 CONTINUE RVIB=.FALSE. we add: CLOSE(24) CALL FILEOPEN(24,FILEN,'OLD') things work better. Cheers, RJ -- Robert O. Jones Forschungszentrum J"ulich, IFF D-52425 J"ulich, Germany. e-mail : r.jones at fz-juelich.de Tel. : (+49) 2461-614202 Fax : (+49) 2461-612850 WWW : http://www.fz-juelich.de/iff/personen/R.Jones From andrea.fortini at mi.infn.it Wed Jun 19 16:36:04 2002 From: andrea.fortini at mi.infn.it (Andrea Fortini) Date: Wed, 19 Jun 2002 16:36:04 +0200 Subject: [CPMD-list] Cpmd on HP In-Reply-To: <200206031001.MAA19596@cpmd.org> References: <200206031001.MAA19596@cpmd.org> Message-ID: <200206191636.04575.andrea.fortini@mi.infn.it> Hello, i want report some problems i had compiling CPMD on HP machine, and the way i solved them. I would know if some one others had same problems and if he found a best way to solve them. 1) It seems that there are no timing functions for HP and the preprocessor directive in file timec.F lead to a file timec.f that is empty and the F90 compile don't like this. I solved returning zero in functions TIMEF and TIMEC. 2) the default option DFFT_HP in the Makefile leads to a compilation error, i have to use the DFFT_DEFAULT. 3) problem with the function trim(), solved as alredy reported on the mailing list. 4) problem with timetag.F, i had to add this line #elif defined(__HP) Andrea Fortini From edulsa at quimica.ufpr.br Wed Jun 19 17:19:14 2002 From: edulsa at quimica.ufpr.br (Eduardo Lemos de Sa) Date: Wed, 19 Jun 2002 12:19:14 -0300 (BRT) Subject: [CPMD-list] (no subject) Message-ID: Dears Has anyone get to compile CPMD in a i386 plataform running Linux/FreeBSD using g77? Cheers Eduardo From hutter at pci.unizh.ch Mon Jun 24 13:37:33 2002 From: hutter at pci.unizh.ch (Juerg Hutter) Date: Mon, 24 Jun 2002 13:37:33 +0200 (MEST) Subject: [CPMD-list] (no subject) In-Reply-To: Message-ID: Hi CPMD makes extensive use of the CRAY Pointer extension to F77. This extension is not included in g77 (it's on the todo list since years) and there is no way that you can get CPMD working without this. Juerg ---------------------------------------------------------- Juerg Hutter Phone : ++41 1 635 4491 Physical Chemistry Institute FAX : ++41 1 635 6838 University of Zurich E-mail: hutter at pci.unizh.ch Winterthurerstrasse 190 CH-8057 Zurich, Switzerland ---------------------------------------------------------- On Wed, 19 Jun 2002, Eduardo Lemos de Sa wrote: > Dears > > Has anyone get to compile CPMD in a i386 plataform running > Linux/FreeBSD using g77? > > Cheers > > > Eduardo > > > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > From hutter at pci.unizh.ch Mon Jun 24 13:40:55 2002 From: hutter at pci.unizh.ch (Juerg Hutter) Date: Mon, 24 Jun 2002 13:40:55 +0200 (MEST) Subject: [CPMD-list] Cpmd on HP In-Reply-To: <200206191636.04575.andrea.fortini@mi.infn.it> Message-ID: Hi thanks for the information on the compile problems of CPMD on HP platforms. We do not have access to a HP and therefore HP support was always weak. However, we are getting help from HP Germany and all features should be working on all HP platforms in the next major release of CPMD. Juerg ---------------------------------------------------------- Juerg Hutter Phone : ++41 1 635 4491 Physical Chemistry Institute FAX : ++41 1 635 6838 University of Zurich E-mail: hutter at pci.unizh.ch Winterthurerstrasse 190 CH-8057 Zurich, Switzerland ---------------------------------------------------------- On Wed, 19 Jun 2002, Andrea Fortini wrote: > Hello, > i want report some problems i had compiling CPMD on HP machine, and the way > i solved them. > I would know if some one others had same problems and if he found a best way > to solve them. > > 1) It seems that there are no timing functions for HP and the preprocessor > directive in file timec.F lead to a file timec.f that is empty and the F90 > compile don't like this. I solved returning zero in functions TIMEF and > TIMEC. > > 2) the default option DFFT_HP in the Makefile leads to a compilation error, i > have to use the DFFT_DEFAULT. > > 3) problem with the function trim(), solved as alredy reported on the mailing > list. > > 4) problem with timetag.F, i had to add this line > #elif defined(__HP) > > Andrea Fortini > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > From cur at zurich.ibm.com Thu Jun 27 10:11:04 2002 From: cur at zurich.ibm.com (Alessandro Curioni) Date: Thu, 27 Jun 2002 10:11:04 +0200 Subject: [CPMD-list] CPMD V3.5.2 code Message-ID: Announcement: The updated version 3.5.2 of the CPMD program (ab-initio Molecular Dynamics program, developed by J. Hutter, M. Parrinello et al, copyright IBM Corp and MPI Stuttgart) is now available at www.cpmd.org. New Features included in the update are: -Interface to the new Vanderbilt pseudopotentials generation code. -Non Linear Core Correction (NLCC) with Vanderbilt pseudopotentials -Vanderbilt pseudopotentials bugs fixed -NEC optimization -General Fixes Any non-profit organization can obtain a free licence there. Thank you to everyone contributed to these new release!!! Your cpmd.org team. (www.cpmd.org) From troya at mercury.chem.northwestern.edu Mon Jun 17 17:37:01 2002 From: troya at mercury.chem.northwestern.edu (Diego Troya) Date: Mon, 17 Jun 2002 10:37:01 -0500 (CDT) Subject: [CPMD-list] Energies of molecular dynamics. Message-ID: Dear list, In a molecular dynamics simulation, CPMD prints out a list of energies for each integration step. The heading for this list is: NFI EKINC TEMPP EKS ECLASSIC EHAM DIS Except for EKS, that seems to be the Kohn-Sham energy, the meaning of the rest of energies is unknown to me. Does anyone know the meaning of EKINC, ECLASSIC, EHAM and DIS? Likewise, which of this is the total energy that has to be conserved along the molecular dynamics simulation? Which is the criterion usually required to decide whether total energy has been conserved? Thanks in advance, Diego.