From rjsa at mail.shcnc.ac.cn Thu May 1 08:10:59 2003 From: rjsa at mail.shcnc.ac.cn (Rongjian Sa) Date: Thu, 1 May 2003 14:10:59 +0800 Subject: [CPMD-list] help on constrained MD Message-ID: <200305010608.IAA06658@internet-fence.zurich.ihost.com> Dear everyone, I wander if who can tell me more about constrained MD calculation. For a system contain ions A and B, if I just set the distance r= rAB as my reaction coordinate. And I discretize my reaction coordinate in 5 points, from 1.5 to 3 A (1.5, 2.0, 2.3, 2.5,3.0A) After this procedure, I will perform MD simulation at these five points. Then I have get five CONSTRAIN files. From every CONSTRAIN files, I would obtain the average force for corresponding reaction coordinate. According the formula deltaF=integral Force(x)dx, I should have two values as my integration area. But I do not how to decide these two values. Also I want to know if I need integrate these five points. Then I can get five energies and graph free energy vs distance. At last, I will obtain a curve and from it I will get the free energy of systems. I do know whether my procedure is correct. So I want to get your kind help! Thank you! Sincerely yours, Rongjian Sa -------------- next part -------------- An HTML attachment was scrubbed... URL: http://cpmd.org/pipermail/cpmd-list/attachments/20030501/45dbff1f/attachment.html From proffess at mail.ru Mon May 5 10:47:26 2003 From: proffess at mail.ru (Sergey) Date: Mon, 05 May 2003 12:47:26 +0400 Subject: [CPMD-list] Keyword POLYMER and SURFACE Message-ID: Dear CPMD users, What is input format for POLYMER and SURFACE keywords? Thank you, Sergey From lghiring at science.uva.nl Thu May 8 14:58:01 2003 From: lghiring at science.uva.nl (luca) Date: Thu, 08 May 2003 14:58:01 +0200 Subject: [CPMD-list] Constraining Center of Mass Message-ID: <3EBA5459.4C277A29@science.uva.nl> Hi, I need to constraint the CofM in a MD simulation. These because, with nose thermostats for ions and wavefunctions swithched on, for a system I am studying there are huge drifts in the CofMass, that is absorbing a consistent amount of the total kinetik energy, so that the real temperature of the system is far from the target. It seems that the keyword FIX COM (reading the code) does work only during a geometry optimization, and not the MD run. Is that true (I am using version 3.4.3)? Is there a way to constraint the center of mass, better than restart every now and then to impose a new set of velocities? Regards, Luca Ghiringhelli. From veronique.vanspeybroeck at rug.ac.be Fri May 9 10:25:13 2003 From: veronique.vanspeybroeck at rug.ac.be (Veronique Van Speybroeck) Date: Fri, 09 May 2003 10:25:13 +0200 Subject: [CPMD-list] Temperature simulations Message-ID: <3EBB65E9.9090907@rug.ac.be> Dear All, I have some problems simulating the KCL lattice at elevated temperature. First I did an optimizations on the lattice. Next I tried to heat up the crystal with the inputfile as given below. Normally I would expect only small oscillations of the atoms around their equilibrium positions, due to the low temperature. Strangely the atoms undergo very large variations and I get a totally distorted system. The temperature in the ENERGIES file raises also drastically. Can somebody tell me what I do wrong? The energies file is given in attachment. When I do a MD simulations microcanonically, everything goes fine and the system stays in the minimuum. Thanks in advance veronique &CPMD RESTART LATEST WAVEFUNCTIONS MOLECULAR DYNAMICS CP TIMESTEP 5 MAXSTEP 2000 PRINT 10 TEMPERATURE 5 QUENCH BO STRUCTURE BONDS ANGLES MOVIE SAMPLE 1 &END &SYSTEM ANGSTROM SYMMETRY 1 CELL 6.3 1.0 1.0 0.0 0.0 0.0 CUTOFF 25. &END &ATOMS *Cl_VDB_U BINARY NEWF LMAX=P 4 0.000000 0.000000 0.000000 3.150000 0.000000 3.150000 0.000000 3.150000 3.150000 3.150000 3.150000 0.000000 *019-K-ca-sp-vgrp.uspp BINARY NEWF LMAX=P 4 3.150000 3.150000 3.150000 0.000000 0.000000 3.150000 0.000000 3.150000 0.000000 3.150000 0.000000 0.000000 &END &DFT LDA CORRELATION PZ FUNCTIONAL BP GC-CUTOFF 5.D-5 &END -- ----------------------------------------------------------------------- Dr. ir. Van Speybroeck Veronique Laboratorium voor Theoretische Fysica Universiteit Gent Proeftuinstraat 86 9000 Gent Tel +32-9-264.65.58 GSM : +32/474/259767 Fax +32-9-264.65.60 email : veronique.vanspeybroeck at rug.ac.be http://inwfaxp2.rug.ac.be/~web/onderzoek/proj_3.html ----------------------------------------------------------------------- -------------- next part -------------- A non-text attachment was scrubbed... Name: Energies Type: application/x-java-applet Size: 314999 bytes Desc: not available Url : http://cpmd.org/pipermail/cpmd-list/attachments/20030509/0d632010/attachment.bin From WANGYI at engr.sc.edu Thu May 8 17:30:04 2003 From: WANGYI at engr.sc.edu (WANG, YIXUAN ) Date: Thu, 8 May 2003 11:30:04 -0400 Subject: [CPMD-list] multiplicity Message-ID: Dear Colleagues, If the periodic boundary condition is applied to an isolated cluster system with spin=1/2 (multiplicity=2), should we still use the same multiplicity or we need other macroscopic information to determine it? Additionally, if multiplicity=2 is specified for spin=1/2 system in an input LSD approximation has to be added, however if multiplicity is absent ( since default multiplicity is the smallest possible one, it should also be 2 for odd electron system ), LSD is not necessary, which is much faster than the former. The question is if the latter is reliable. I appreciate your comments. Yixuan Wang Department of Chem. Eng. University South Carolina Columbia, SC 29208 From cdy2000 at vip.sina.com Sun May 11 15:18:15 2003 From: cdy2000 at vip.sina.com (cdy2000 at vip.sina.com) Date: Sun, 11 May 2003 21:18:15 +0800 Subject: [CPMD-list] I cant use ODIIS NO_RESET Message-ID: <20030511131815.20687.qmail@vip.sina.com> Hello, all CPMDer! My CPMD 3.7 programme is upgraded from 3.5, this is a part of my inputfile: &CPMD ODIIS NO_RESET OPTIMIZE GEOMETRY XYZ ISOLATED MOLECULE &END &SYSTEM ANGSTROM SYMMETRY 0 CELL 25.0000 0.7170 0.6561 0 0 0 CUTOFF 12.5 &END when I run cpmd.x, the info is PGFIO/stdio: No such file or directory PGFIO-F-/list-directed write/unit=6/error code returned by host stdio -2. File name = stdout formatted, sequential access record = 4 In source file ./control.f, at line number 1952 I want to know why? Are there some syntax wrong in my input file, or no some origin file in my CPMD programme? Thanks for any help Sincerely Cao Dayong ______________________________________ =================================================================== From Ari.P.Seitsonen at iki.fi Sun May 11 16:24:36 2003 From: Ari.P.Seitsonen at iki.fi (Ari.P.Seitsonen at iki.fi) Date: Sun, 11 May 2003 16:24:36 +0200 Subject: [CPMD-list] I cant use ODIIS NO_RESET In-Reply-To: <20030511131815.20687.qmail@vip.sina.com> (cdy2000@vip.sina.com) References: <20030511131815.20687.qmail@vip.sina.com> Message-ID: <200305111424.h4BEOaT27876@magadino.cscs.ch> Dear Cao Dayong, > My CPMD 3.7 programme is upgraded from 3.5, this is a part of my inputfile: > > &CPMD > ODIIS NO_RESET > OPTIMIZE GEOMETRY XYZ > ISOLATED MOLECULE > &END You need to provide a value for the option "NO_RESET", for example &CPMD ODIIS NO_RESET=-1 10 ... &END That is, resets are switched off and the DIIS algorithm uses 10 vectors. Greetings from Zurich, apsi PS There's also a way to give 'NO_RESET' in the input without the keyword 'ODIIS', please read the file 'control.F' in the source code if you are interested in this -- -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- Ari Paavo Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/ Tel +41 1 635 44 97 / Fax +41 1 635 68 38 / GSM +41 79 719 09 35 Anschrift: Physikalisch Chemisches Institut (PCI), Universit?t Z?rich (UniZh) Indirizzo: Winterthurerstra?e 190, CH-8057 Z?rich Address: Schweiz / Svizzera / Suisse / Svizra / Switzerland From Ari.P.Seitsonen at iki.fi Sun May 11 16:32:12 2003 From: Ari.P.Seitsonen at iki.fi (Ari.P.Seitsonen at iki.fi) Date: Sun, 11 May 2003 16:32:12 +0200 Subject: [CPMD-list] Temperature simulations In-Reply-To: <3EBB65E9.9090907@rug.ac.be> (message from Veronique Van Speybroeck on Fri, 09 May 2003 10:25:13 +0200) References: <3EBB65E9.9090907@rug.ac.be> Message-ID: <200305111432.h4BEWCE27921@magadino.cscs.ch> Dear Veronique, Some suggestions: - The system is too small for Gamma point only, it might be that the ionic configuration is unstable; you could test this by giving the ions a small displacement from the equilibrium positions (by the way, for such crystals I find the keyword 'SCALED' very useful) and then relax the geometry and see, if the atoms return to the lattice positions. The natural solution, if the structure is indeed unstable, is to use a bigger cell (2x2x2, i.e. 32 atoms of each species) - You don't specify the electronic mass, the default might be unsuitable for this system (also the electrons heat up slowly in your simulation) Please tell if this doesn't help, then we can try something else. Greetings, apsi PS I guess that you use the option 'MOVIE SAMPLE \n 1' only for the debugging, right? > I have some problems simulating the KCL lattice at elevated temperature. > First I did an optimizations on the lattice. Next I tried to heat up > the crystal with the inputfile as given below. Normally I would expect > only small oscillations of the atoms around their equilibrium positions, > due to the low temperature. Strangely the atoms undergo very large > variations and I get a totally distorted system. The temperature in the > ENERGIES file raises also drastically. Can somebody tell me what I do > wrong? The energies file is given in attachment. > When I do a MD simulations microcanonically, everything goes fine and > the system stays in the minimuum. > > Thanks in advance > > veronique > > > &CPMD > RESTART LATEST WAVEFUNCTIONS > MOLECULAR DYNAMICS CP > TIMESTEP > 5 > MAXSTEP > 2000 > PRINT > 10 > TEMPERATURE > 5 > QUENCH BO > STRUCTURE BONDS ANGLES > MOVIE SAMPLE > 1 > &END > > &SYSTEM > ANGSTROM > SYMMETRY > 1 > CELL > 6.3 1.0 1.0 0.0 0.0 0.0 > CUTOFF > 25. > &END > > &ATOMS > *Cl_VDB_U BINARY NEWF > LMAX=P > 4 > 0.000000 0.000000 0.000000 > 3.150000 0.000000 3.150000 > 0.000000 3.150000 3.150000 > 3.150000 3.150000 0.000000 > *019-K-ca-sp-vgrp.uspp BINARY NEWF > LMAX=P > 4 > 3.150000 3.150000 3.150000 > 0.000000 0.000000 3.150000 > 0.000000 3.150000 0.000000 > 3.150000 0.000000 0.000000 > &END > > &DFT > LDA CORRELATION PZ > FUNCTIONAL BP > GC-CUTOFF > 5.D-5 > &END -- -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- Ari Paavo Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/ Tel +41 1 635 44 97 / Fax +41 1 635 68 38 / GSM +41 79 719 09 35 Anschrift: Physikalisch Chemisches Institut (PCI), Universit?t Z?rich (UniZh) Indirizzo: Winterthurerstra?e 190, CH-8057 Z?rich Address: Schweiz / Svizzera / Suisse / Svizra / Switzerland From Ari.P.Seitsonen at iki.fi Sun May 11 16:41:21 2003 From: Ari.P.Seitsonen at iki.fi (Ari.P.Seitsonen at iki.fi) Date: Sun, 11 May 2003 16:41:21 +0200 Subject: [CPMD-list] multiplicity In-Reply-To: (WANGYI@engr.sc.edu) References: Message-ID: <200305111441.h4BEfLL27964@magadino.cscs.ch> Dear Yixuan Wang, > If the periodic boundary condition is applied to an isolated cluster system > with spin=1/2 (multiplicity=2), should we still use the same multiplicity or > we need other macroscopic information to determine it? > > Additionally, if multiplicity=2 is specified for spin=1/2 system in an > input LSD approximation has to be added, however if multiplicity is absent ( > since default multiplicity is the smallest possible one, it should also be 2 > for odd electron system ), LSD is not necessary, which is much faster than > the former. The question is if the latter is reliable. I appreciate your > comments. Sorry, I didn't quite get your question: Do you still want to simulate the system as an isolated cluster, or a crystal? In the first case you should always use LSD, otherwise you will get unphysical results. In the latter case it depends on the system you are studying; also please be careful that your system is large enough in order to justify using the Gamma point only. Is the band structure of the extended system known? Since your system is radical your molecules might dimerise etc in order to get rid off the unpaired electron. (Just a thought, I'm not a chemist :v) Greetings, apsi -- -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- Ari Paavo Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/ Tel +41 1 635 44 97 / Fax +41 1 635 68 38 / GSM +41 79 719 09 35 Anschrift: Physikalisch Chemisches Institut (PCI), Universit?t Z?rich (UniZh) Indirizzo: Winterthurerstra?e 190, CH-8057 Z?rich Address: Schweiz / Svizzera / Suisse / Svizra / Switzerland From Ari.P.Seitsonen at iki.fi Sun May 11 16:47:44 2003 From: Ari.P.Seitsonen at iki.fi (Ari.P.Seitsonen at iki.fi) Date: Sun, 11 May 2003 16:47:44 +0200 Subject: [CPMD-list] Constraining Center of Mass In-Reply-To: <3EBA5459.4C277A29@science.uva.nl> (message from luca on Thu, 08 May 2003 14:58:01 +0200) References: <3EBA5459.4C277A29@science.uva.nl> Message-ID: <200305111447.h4BEliX28002@magadino.cscs.ch> Dear Luca, Yes, it seems that in the current version (I don't know about 3.4.3) the 'FIX COM' only works with the optimisation of the geometry. However you could remove the COM motion from the velocities easily after the calculation of the new velocities (with call a to 'FIXCOM()'? It's not necessary to do this every iteration), or you could fix one atom. Hopefully this helps you further. Greetings, apsi > I need to constraint the CofM in a MD simulation. These because, > with nose thermostats for ions and wavefunctions swithched on, for a > system I am studying there are huge drifts in the CofMass, that is > absorbing a consistent amount of the total kinetik energy, so that the > real temperature of the system is far from the target. It seems that the > keyword FIX COM (reading the code) does work only during a geometry > optimization, and not the MD run. Is that true (I am using version > 3.4.3)? > Is there a way to constraint the center of mass, better than restart > every now and then to impose a new set of velocities? > > Regards, > Luca Ghiringhelli. -- -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- Ari Paavo Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/ Tel +41 1 635 44 97 / Fax +41 1 635 68 38 / GSM +41 79 719 09 35 Anschrift: Physikalisch Chemisches Institut (PCI), Universit?t Z?rich (UniZh) Indirizzo: Winterthurerstra?e 190, CH-8057 Z?rich Address: Schweiz / Svizzera / Suisse / Svizra / Switzerland From ZRLSRB at ch.ibm.com Mon May 12 09:35:15 2003 From: ZRLSRB at ch.ibm.com (Salomon Billeter) Date: Mon, 12 May 2003 09:35:15 +0200 Subject: [CPMD-list] I cant use ODIIS NO_RESET Message-ID: Hi Cao Dayong, > &CPMD > ODIIS NO_RESET > OPTIMIZE GEOMETRY XYZ > ISOLATED MOLECULE > &END both ODIIS and NO_RESET need a value. At the moment, NO_RESET is implemented as a keyword on its own, i.e. it must not appear on the same line as e.g. ODIIS. An example is here: &CPMD ODIIS 5 NO_RESET 20 OPTIMIZE GEOMETRY XYZ ISOLATED MOLECULE &END This means that you use 5 steps for DIIS and reset DIIS after 20 steps if the progress is not good enough (or -1 if you want to switch off resetting DIIS altogether). For a geometry optimization, you can also try adapting the tolerance of the wavefunction to the nuclear gradient. You are going to save wavefunction cycles, and maybe don't even need NO_RESET. Here is an example: &CPMD ... CONVERGENCE ADAPT 0.02 CONVERGENCE RELAX 25 &END In addition, the lines CONVERGENCE RELAX and 25 make the wavefunction convergence criteria temporarily relax by a factor of two every 20 wavefunction cycles until the next geometry step which resets them. For details of adaptive tolerance, see the manual. Best regards, Salomon From veronique.vanspeybroeck at rug.ac.be Mon May 12 11:43:17 2003 From: veronique.vanspeybroeck at rug.ac.be (Veronique Van Speybroeck) Date: Mon, 12 May 2003 11:43:17 +0200 Subject: [CPMD-list] Temperature simulations References: <3EBB65E9.9090907@rug.ac.be> <200305111432.h4BEWCE27921@magadino.cscs.ch> Message-ID: <3EBF6CB5.8020407@rug.ac.be> Dear Ari, thanks for your suggestions. First of all I gave the atoms a random displacement of 0.05 by the option randomize in the inputfile and then did again a geometry optimization. The programs ends normally, but the atoms do not return to their original position, the energy however is converged. I must conclude then that the system is indeed unstable. The input file is given below of this simulations. By the way I did not optimize the cell parameters, is this necessary? Now some simulations run on the larger system. this takes longer so I do not know the result yet. Is it an option to use the nose keyword for the electrons to keep them close to the BO surface? Then the question remains how to choose this electronic mass? Thanks in advance veronique PS : the thing movie sample is for some conversion programs that I have written. They transform the MOVIE file into car and arc files which can be read in by CERIUS. input file for geometry optimization : &CPMD RESTART LATEST WAVEFUNCTION OPTIMIZE GEOMETRY TIMESTEP 5 MAXSTEP 3000 PRINT 10 RANDOMIZE COORDINATES 0.05 STRUCTURE BONDS ANGLES MOVIE SAMPLE 1 &END &SYSTEM ANGSTROM SYMMETRY 1 CELL 6.3 1.0 1.0 0.0 0.0 0.0 CUTOFF 25. &END &ATOMS *Cl_VDB_U BINARY NEWF LMAX=P 4 0.000000 0.000000 0.000000 3.150000 0.000000 3.150000 0.000000 3.150000 3.150000 3.150000 3.150000 0.000000 *019-K-ca-sp-vgrp.uspp BINARY NEWF LMAX=P 4 3.150000 3.150000 3.150000 0.000000 0.000000 3.150000 0.000000 3.150000 0.000000 3.150000 0.000000 0.000000 &END &DFT LDA CORRELATION PZ FUNCTIONAL BP GC-CUTOFF 5.D-5 &END Ari.P.Seitsonen at iki.fi wrote: >Dear Veronique, > > Some suggestions: > > - The system is too small for Gamma point only, it might be that the > ionic configuration is unstable; you could test this by giving the > ions a small displacement from the equilibrium positions (by the > way, for such crystals I find the keyword 'SCALED' very useful) and > then relax the geometry and see, if the atoms return to the lattice > positions. The natural solution, if the structure is indeed > unstable, is to use a bigger cell (2x2x2, i.e. 32 atoms of each > species) > > - You don't specify the electronic mass, the default might be > unsuitable for this system (also the electrons heat up slowly in > your simulation) > > Please tell if this doesn't help, then we can try something else. > > Greetings, > > apsi > >PS I guess that you use the option 'MOVIE SAMPLE \n 1' only for the > debugging, right? > > > >>I have some problems simulating the KCL lattice at elevated temperature. >> First I did an optimizations on the lattice. Next I tried to heat up >>the crystal with the inputfile as given below. Normally I would expect >>only small oscillations of the atoms around their equilibrium positions, >>due to the low temperature. Strangely the atoms undergo very large >>variations and I get a totally distorted system. The temperature in the >>ENERGIES file raises also drastically. Can somebody tell me what I do >>wrong? The energies file is given in attachment. >>When I do a MD simulations microcanonically, everything goes fine and >>the system stays in the minimuum. >> >>Thanks in advance >> >>veronique >> >> >>&CPMD >> RESTART LATEST WAVEFUNCTIONS >> MOLECULAR DYNAMICS CP >> TIMESTEP >> 5 >> MAXSTEP >> 2000 >> PRINT >> 10 >> TEMPERATURE >> 5 >> QUENCH BO >> STRUCTURE BONDS ANGLES >> MOVIE SAMPLE >> 1 >>&END >> >>&SYSTEM >> ANGSTROM >> SYMMETRY >> 1 >> CELL >> 6.3 1.0 1.0 0.0 0.0 0.0 >> CUTOFF >> 25. >>&END >> >>&ATOMS >>*Cl_VDB_U BINARY NEWF >> LMAX=P >> 4 >> 0.000000 0.000000 0.000000 >> 3.150000 0.000000 3.150000 >> 0.000000 3.150000 3.150000 >> 3.150000 3.150000 0.000000 >>*019-K-ca-sp-vgrp.uspp BINARY NEWF >> LMAX=P >> 4 >> 3.150000 3.150000 3.150000 >> 0.000000 0.000000 3.150000 >> 0.000000 3.150000 0.000000 >> 3.150000 0.000000 0.000000 >>&END >> >>&DFT >> LDA CORRELATION PZ >> FUNCTIONAL BP >> GC-CUTOFF >> 5.D-5 >>&END >> >> > > > -- ----------------------------------------------------------------------- Dr. ir. Van Speybroeck Veronique Laboratorium voor Theoretische Fysica Universiteit Gent Proeftuinstraat 86 9000 Gent Tel +32-9-264.65.58 GSM : +32/474/259767 Fax +32-9-264.65.60 email : veronique.vanspeybroeck at rug.ac.be http://inwfaxp2.rug.ac.be/~web/onderzoek/proj_3.html ----------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://cpmd.org/pipermail/cpmd-list/attachments/20030512/11a13bf9/attachment.html From veronique.vanspeybroeck at rug.ac.be Mon May 12 11:46:21 2003 From: veronique.vanspeybroeck at rug.ac.be (Veronique Van Speybroeck) Date: Mon, 12 May 2003 11:46:21 +0200 Subject: [CPMD-list] Temperature simulations References: <3EBB65E9.9090907@rug.ac.be> <00a501c317cb$0fc11d60$e92dfec8@frota8uvg7i4db> Message-ID: <3EBF6D6D.8080306@rug.ac.be> Dear , my inputfile is given below : &CPMD RESTART LATEST WAVEFUNCTION GEOFILE OPTIMIZE GEOMETRY TIMESTEP 5 MAXSTEP 3000 PRINT 10 STRUCTURE BONDS ANGLES MOVIE SAMPLE 1 &END &SYSTEM ANGSTROM SYMMETRY 1 CELL 6.3 1.0 1.0 0.0 0.0 0.0 CUTOFF 25. &END &ATOMS *Cl_VDB_U BINARY NEWF LMAX=P 4 0.000000 0.000000 0.000000 3.150000 0.000000 3.150000 0.000000 3.150000 3.150000 3.150000 3.150000 0.000000 *019-K-ca-sp-vgrp.uspp BINARY NEWF LMAX=P 4 3.150000 3.150000 3.150000 0.000000 0.000000 3.150000 0.000000 3.150000 0.000000 3.150000 0.000000 0.000000 &END &DFT LDA CORRELATION PZ FUNCTIONAL BP GC-CUTOFF 5.D-5 &END It is important to set the GC-cutoff to a larger value than for hard pseudopotentials. I used to have problems when I used values of 1D-7 but with the larger value I never get problems anymore. Neither for lattices as for simple organic systems Hope this helps. veronique Von Braun Nascimento wrote: >Dear Veronique, > >I'm having some problems with geometry optimizations when I use Vanserbilt >USPP pseudo-potentials. My total static energy versus lattice parameter >doesn't even presents a minimum. With utilization of NCPP no problems >occurr. >You are successfully using VDB USPP, at least in the geometry optimization >step. Have you had any trouble with VDB USSPP ? Is it possible for you to >send me the input files you have used in the geometry optimization ? >Thanks in advance. >Best regards, >V.B. Nascimento > >----- Original Message ----- >From: "Veronique Van Speybroeck" >To: >Sent: Friday, May 09, 2003 5:25 AM >Subject: [CPMD-list] Temperature simulations > > > > >>Dear All, >> >>I have some problems simulating the KCL lattice at elevated temperature. >> First I did an optimizations on the lattice. Next I tried to heat up >>the crystal with the inputfile as given below. Normally I would expect >>only small oscillations of the atoms around their equilibrium positions, >>due to the low temperature. Strangely the atoms undergo very large >>variations and I get a totally distorted system. The temperature in the >>ENERGIES file raises also drastically. Can somebody tell me what I do >>wrong? The energies file is given in attachment. >>When I do a MD simulations microcanonically, everything goes fine and >>the system stays in the minimuum. >> >>Thanks in advance >> >>veronique >> >> >>&CPMD >> RESTART LATEST WAVEFUNCTIONS >> MOLECULAR DYNAMICS CP >> TIMESTEP >> 5 >> MAXSTEP >> 2000 >> PRINT >> 10 >> TEMPERATURE >> 5 >> QUENCH BO >> STRUCTURE BONDS ANGLES >> MOVIE SAMPLE >> 1 >>&END >> >>&SYSTEM >> ANGSTROM >> SYMMETRY >> 1 >> CELL >> 6.3 1.0 1.0 0.0 0.0 0.0 >> CUTOFF >> 25. >>&END >> >>&ATOMS >>*Cl_VDB_U BINARY NEWF >> LMAX=P >> 4 >> 0.000000 0.000000 0.000000 >> 3.150000 0.000000 3.150000 >> 0.000000 3.150000 3.150000 >> 3.150000 3.150000 0.000000 >>*019-K-ca-sp-vgrp.uspp BINARY NEWF >> LMAX=P >> 4 >> 3.150000 3.150000 3.150000 >> 0.000000 0.000000 3.150000 >> 0.000000 3.150000 0.000000 >> 3.150000 0.000000 0.000000 >>&END >> >>&DFT >> LDA CORRELATION PZ >> FUNCTIONAL BP >> GC-CUTOFF >> 5.D-5 >>&END >> >>-- >>----------------------------------------------------------------------- >>Dr. ir. Van Speybroeck Veronique >>Laboratorium voor Theoretische Fysica >>Universiteit Gent >>Proeftuinstraat 86 >>9000 Gent >>Tel +32-9-264.65.58 GSM : +32/474/259767 >>Fax +32-9-264.65.60 >>email : veronique.vanspeybroeck at rug.ac.be >>http://inwfaxp2.rug.ac.be/~web/onderzoek/proj_3.html >>----------------------------------------------------------------------- >> >> >> >> > > > > -- ----------------------------------------------------------------------- Dr. ir. Van Speybroeck Veronique Laboratorium voor Theoretische Fysica Universiteit Gent Proeftuinstraat 86 9000 Gent Tel +32-9-264.65.58 GSM : +32/474/259767 Fax +32-9-264.65.60 email : veronique.vanspeybroeck at rug.ac.be http://inwfaxp2.rug.ac.be/~web/onderzoek/proj_3.html ----------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://cpmd.org/pipermail/cpmd-list/attachments/20030512/0270e4d4/attachment.html From Ari.P.Seitsonen at iki.fi Mon May 12 13:57:55 2003 From: Ari.P.Seitsonen at iki.fi (Ari.P.Seitsonen at iki.fi) Date: Mon, 12 May 2003 13:57:55 +0200 Subject: [CPMD-list] Temperature simulations In-Reply-To: <3EBF6CB5.8020407@rug.ac.be> (message from Veronique Van Speybroeck on Mon, 12 May 2003 11:43:17 +0200) References: <3EBB65E9.9090907@rug.ac.be> <200305111432.h4BEWCE27921@magadino.cscs.ch> <3EBF6CB5.8020407@rug.ac.be> Message-ID: <200305121157.h4CBvtH01444@magadino.cscs.ch> Dear Veronique, > thanks for your suggestions. First of all I gave the atoms a random > displacement of 0.05 by the option randomize in the inputfile and then > did again a geometry optimization. The programs ends normally, but the > atoms do not return to their original position, the energy however is > converged. I must conclude then that the system is indeed unstable. So indeed the 4+4 atom cell is too small for the Gamma point only. The next bigger (simple cubic) cell is 32+32 atoms. There are also intermediate cells but they have either BCC or FCC basis vectors. > The input file is given below of this simulations. By the way I did > not optimize the cell parameters, is this necessary? The lattice parameters are not important - as long as e.g. the pseudos are good and the basis set more or less converged. Usually one should be within about 2 % of the experimental lattice constant, and in such a small range there are no competing structures; even if there would be one completely different lattice strcture there are too few atoms for a phase transition to occur (and the basis vectors are anyway fixed). > Is it an option to use the nose keyword for the electrons to keep them > close to the BO surface? I think that KCl should be an insulator with a large electronic gap, right? Then the electrons should not deviate far from the Born-Oppenheimer surface unless the time step is too large or the electronic mass is inapproriate. > Then the question remains how to choose this electronic mass? Well, if one already has a value for the maximal time step you can try a couple of values for the \mu and take the "average" one which conserves the total energy (KS+ionic kinetic+electron kinetic). If you are not sure about the time step you can take a smaller one, run the tests for the electronic mass and then increase them so that dt^2/\mu is a constant, again as far as the total energy is well preserved. (How to choose the time step: about 1/6th ... 1/10th of the fastest ionic period unless the fastest electronic oscillations = E_cut limit the value) Greetings, apsi -- -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- Ari Paavo Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/ Tel +41 1 635 44 97 / Fax +41 1 635 68 38 / GSM +41 79 719 09 35 Anschrift: Physikalisch Chemisches Institut (PCI), Universit?t Z?rich (UniZh) Indirizzo: Winterthurerstra?e 190, CH-8057 Z?rich Address: Schweiz / Svizzera / Suisse / Svizra / Switzerland From s.meloni at caspur.it Mon May 12 19:11:47 2003 From: s.meloni at caspur.it (Simone Meloni) Date: Mon, 12 May 2003 17:11:47 +0000 (UTC) Subject: [CPMD-list] Pseudo for Ni or Pd Message-ID: Dear CPMDer, does anybody have a working potential for either Ni or Pd? we would appreciate help. Ciao Simone & Galina -- +------------------------------------------------------------------------+ PhD Simone Meloni | Achab: Lascio una scia Bianca e inquieta, CASPUR | acque pallide, facce piu' pallide Via Dei Tizii 6b | dovunque passo. Le onde invidiose 00185 Rome (Italy) | si gonfiano ai lati per sommergere Phone: +39-06-44486-711 | la mia traccia: facciano, ma prima Fax: +39-06-4957083 | io passo. From c00ycl00 at nchc.gov.tw Tue May 13 04:02:32 2003 From: c00ycl00 at nchc.gov.tw (c00ycl00) Date: Tue, 13 May 2003 10:02:32 +0800 Subject: [CPMD-list] Some questions about MD in CPMD Message-ID: <3EC052381F4.EA2BC00YCL00@smtp.nchc.gov.tw> Dear all, I am trying some exercises of CPMD. But when I do the exercise 2.4 of the CECAM tutorial ( MD simulation for NaCl), I got some errors in the TRAJECTORY and ENERGIES output file. I use the Born-Oppenheimer MD with EMASS of 1200 a.u. and timestep of 10 a.u.. The program can stop normally and there is no any error message in the log file. But in the TRAJECTORY file, all of the position and gradient of the unconstrianed Na atom appear NaNQ sign ( but the finial results of the position and gradient in log file is normal). I also use the Car-Parrinello MD with EMASS of 1200 a.u. and timestep of 5 a.u., 10 a.u., 50 a.u., the TRAJECTORIES and ENERGIES files are normal. But when I use the timestep of 100 a.u., the TRAJECTORIES and ENERGIES appear NaNQ again. Can anyone tell me what's wrong in this simulation? Thank you for your help. Best regards, Yung-Chiang Lan The input file I used is listed below: ----------------------------------------------------------------- &CPMD RESTART WAVEFUNCTION COORDINATES LATEST MOLECULAR DYNAMICS BO MAXSTEP 3310 TIMESTEP 10.0 EMASS 1200 TRAJECTORY SAMPLE 1 STORE 500 &END &DFT NEWCODE FUNCTIONAL BLYP GC-CUTOFF 0.1E-04 &END &SYSTEM ANGSTROMS SYMMETRY 1 CELL 5.6352 1.0 1.0 0 0 0 CUTOFF 30.0 &END &ATOMS *Na_SG_BLYP KLEINMAN-BYLANDER LMAX=P 4 2.8176 2.8176 2.6176 2.8176 0.0 0.0 0.0 2.8176 0.0 0.0 0.0 2.8176 *Cl_SG_BLYP KLEINMAN-BYLANDER LMAX=P 4 0.0 0.0 0.0 2.8176 2.8176 0.0 2.8176 0.0 2.8176 0.0 2.8176 2.8176 CONSTRAINTS FIX ATOMS 7 2 3 4 5 6 7 8 END CONSTRAINTS &END ------------------------------------------------------------ From axel.kohlmeyer at theochem.ruhr-uni-bochum.de Thu May 15 13:05:37 2003 From: axel.kohlmeyer at theochem.ruhr-uni-bochum.de (Axel Kohlmeyer) Date: Thu, 15 May 2003 13:05:37 +0200 Subject: [CPMD-list] Install with Redhat 7.3 and ABSOFT compiler In-Reply-To: <07906A56-66B4-11D7-A649-000393488768@sci.hokudai.ac.jp> Message-ID: <200305151105.h4FB5bh17502@yello.theochem.ruhr-uni-bochum.de> On Sat, 5 Apr 2003 00:42:23 +0900 Yamaki Takayoshi wrote: > Dear all, > > I tried to install CPMD 3.7 to PC with Redhat 7.3, ABSOFT. > I ran "Configure" as following: > Configure PC-ABSOFT > Makefile > Then, Makefile was modified in order to be refer to the libraries > correctly. > However, I got a error message, > > ........... > .o mm_init.o mm_dim.o mm_qmmm_forcedr.o mm_mdmain.o mm_coninp.o > mm_cpmd_add_MM_forces_f77.o mm_cpmd_esp_charges_f77.o > mm_cpmd_ext_pot_f77.o vdwin.o vdw.o sysdepend.o shmemory.o > -L/home/private2/prog/ABSOFT/opt/absoft/lib -llibf90math -llibimsl > -llibimslblas -llibimslf90 -lblas -llibU77 > /usr/bin/ld: cannot find -llibf90math > collect2: ld returned 1 exit status > > make: *** [cpmd.x] Error 1 > ............ > > and compile is not successful. > > I can not understand this message. Please help me solving this problem. well. you have to know, how library names are resolved under linux/unix: the linker is looking for a library named liblibf90math.a or liblibf90math.so. if my guess right, the library is named libf90math.a and you have to change the linker command from: > -L/home/private2/prog/ABSOFT/opt/absoft/lib -llibf90math -llibimsl > -llibimslblas -llibimslf90 -lblas -llibU77 to something like: > -L/home/private2/prog/ABSOFT/opt/absoft/lib -lf90math -limsl > -limslblas -limslf90 -lU77 to explain: the 'lib' in the library filename and the extension '.a' or '.so' is already implied by the '-l' flag. hope this helps, axel kohlmeyer. > > Modified makefile is the following. > > #----------------------------------------------------------------------- > ----- > # Makefile for cpmd.x (plane wave electronic calculation) > # Configuration: PC-ABSOFT > # Creation of Makefile: Apr 4 2003 > # on Linux soranuma 2.4.18-3smp #1 SMP Thu Apr 18 07:27:31 EDT 2002 > i686 unknown > # Author: soranuma!yamaki > #----------------------------------------------------------------------- > ----- > # > SHELL = /bin/sh > # > #--------------- Default Configuration for PC-ABSOFT ---------------SRC > = . > DEST = .BIN = . > #QMMM_FLAGS = -D__QMECHCOUPL > #QMMM_LIBS = -L. -lmm > FFLAGS = > #LFLAGS = -L/usr/absoft/lib -llapack -lblas -lU77 $(QMMM_LIBS) > LFLAGS = -L/home/private2/prog/ABSOFT/opt/absoft/lib -llibf90math > -llibimsl -llibimslblas -llibimslf90 -lblas -llibU77 $(QMMM_LIBS) > #LFLAGS = -L/home/private2/prog/ABSOFT/opt/absoft/lib -llibf77math > -llibimsl -llibimslblas -llibimslf90 -lblas -llibU77 $(QMMM_LIBS) > CFLAGS = > CPP = /lib/cpp -P -C > CPPFLAGS = -D__NOINT8 -D__Linux -D__ABSOFT -DLAPACK -DFFT_DEFAULT > NOOPT_FLAG = > CC = cc -O > FC = f90 -c -B108 -B100 -YEXT_NAMES=LCS -O > LD = f90 -O -X-Bstatic > AR = > #----------------------------------------------------------------------- > ----- > # > # LIST OF FILES > # > > > Thank you in advance. > > Sincerely yours. > T.Yamaki > > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > -- ======================================================================= Axel Kohlmeyer e-mail: axel.kohlmeyer at theochem.ruhr-uni-bochum.de Lehrstuhl fuer Theoretische Chemie Phone: ++49 (0)234/32-26673 Ruhr-Universitaet Bochum - NC 03/53 Fax: ++49 (0)234/32-14045 D-44780 Bochum http://www.theochem.ruhr-uni-bochum.de ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From axel.kohlmeyer at theochem.ruhr-uni-bochum.de Thu May 15 13:21:07 2003 From: axel.kohlmeyer at theochem.ruhr-uni-bochum.de (Axel Kohlmeyer) Date: Thu, 15 May 2003 13:21:07 +0200 Subject: [CPMD-list] help In-Reply-To: <20030401131138.3241.qmail@vip.sina.com> Message-ID: <200305151121.h4FBL7E18072@yello.theochem.ruhr-uni-bochum.de> On Tue, 01 Apr 2003 21:11:38 +0800 cdy2000 at vip.sina.com wrote: > Hello, every CPMDer! > > Thank Isaev's help. > > Another problem, when I optimize the geometry and the wavefunction of a cluster, error output, the info is: > > PROGRAM STOP IN SUBROUTINE RGMOPT| SIMUL NOT WORKING > > what's meaning? what is RGMOPT? How should I do? in this case you have to 'use the source, luke': e.g.: 'grep -i rgmopt *.F' yields: gmopts.F: CALL RGMOPT(C0,C1,C2,CM,SC0,PME,GDE,VPP,EIGV) inr_dr.F:c ... changing tmixgdiis for hessup in rgmopt rgmopt.F: SUBROUTINE RGMOPT(C0,C1,C2,CM,SC0,PME,GDE,VPP,EIGV) rgmopt.F: CALL STOPGM("RGMOPT","TDDFT.AND.TDIAG NOT POSSIBLE") rgmopt.F: CALL GIVE_SCR_RGMOPT(LSCR,TAG) rgmopt.F: CALL STOPGM("RGMOPT","TDDFT NOT POSSIBLE") rgmopt.F: CALL STOPGM('RGMOPT','K POINTS NOT IMPLEMENTED') rgmopt.F: CALL STOPGM('RGMOPT','qmmm + steepest not tested') rgmopt.F: CALL STOPGM('RGMOPT','SIMUL NOT WORKING') rgmopt.F: & ' RGMOPT| THE MAXIMUM NUMBER OF STEP IS REACHED', rgmopt.F: SUBROUTINE GIVE_SCR_RGMOPT(LRGMOPT,TAG) rgmopt.F: INTEGER LRGMOPT rgmopt.F: LRGMOPT=MAX(LINITRUN,LCOPOT,LORTHO,LRNLSM,LRHOOFR, rprfo.F:C == INITIALIZE FLAGS RETURNED TO RGMOPT IF RESTART == rprfo.F:C == MAP_P(i_in_coord) = i_in_xpar (from RGMOPT) == rprfo.F: LALLHS = .TRUE. ! Allocate Hessian in RGMOPT rprfo.F:C Set pointers within work space and return codes to RGMOPT looking into rgmopt.F line 537ff gives you the source of the message: ELSEIF(SIMUL) THEN CALL STOPGM('RGMOPT','SIMUL NOT WORKING') so the boolean SIMUL triggers it. now repeat the search for SIMUL. 'grep SIMUL *.F' yields: control_def.F: SIMUL=.FALSE. control.F: SIMUL=.TRUE. control_pri.F: * ' SIMULATED ANNEALING OF IONS WITH ','ANNERI =',ANNERI control_pri.F: * ' SIMULATED ANNEALING OF ELECTRONS WITH ','ANNERE =',ANNERE control_pri.F: * ' SIMULATED ANNEALING OF CELL WITH ','ANNERC =',ANNERC control_pri.F: IF(SIMUL) WRITE(*,'(A)') gmopts.F: IF(SIMUL) NCM=NC0 mddiag.F: & 'MOLECULAR DYNAMICS SIMULATION' mddiag.F:C CALCULATE PROPERTIES DURING SIMULATION. mddiag.F: & 'END OF MOLECULAR DYNAMICS SIMULATION' proppt.F:C == THIS ROUTINE CALCULATES PROPERTIES DURING THE SIMULATION == rgmopt.F: ELSEIF(SIMUL) THEN rgmopt.F: CALL STOPGM('RGMOPT','SIMUL NOT WORKING') scratch.F: * CALL STOPGM('SCRSET','TOO MANY SCRATCH SPACES SIMULTANIOUS') scratch.F: CALL STOPGM('SCRSET','TOO MANY SCRATCH SPACES SIMULTANIOUS') so SIMUL is set to .true. in control.F. looking at line 375ff shows that this turns on the combined optimization scheme, for which the manual claims, that it is not yet fully operational. so you have to either complete the implementation of that code or try something else. :-( regards, axel kohlmeyer. > > Thanks for any help. > > Sincerely Cao Dayong > ______________________________________ > > =================================================================== > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > -- ======================================================================= Axel Kohlmeyer e-mail: axel.kohlmeyer at theochem.ruhr-uni-bochum.de Lehrstuhl fuer Theoretische Chemie Phone: ++49 (0)234/32-26673 Ruhr-Universitaet Bochum - NC 03/53 Fax: ++49 (0)234/32-14045 D-44780 Bochum http://www.theochem.ruhr-uni-bochum.de ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From t.laino at sns.it Thu May 15 13:34:34 2003 From: t.laino at sns.it (Teodoro Laino) Date: Thu, 15 May 2003 13:34:34 +0200 (CEST) Subject: [CPMD-list] Install with Redhat 7.3 and ABSOFT compiler In-Reply-To: <07906A56-66B4-11D7-A649-000393488768@sci.hokudai.ac.jp> Message-ID: >>> " YT " == Yamaki Takayoshi on Sat, 5 Apr 2003 : YT> Dear all, YT> YT> I tried to install CPMD 3.7 to PC with Redhat 7.3, ABSOFT. YT> I ran "Configure" as following: YT> Configure PC-ABSOFT > Makefile YT> Then, Makefile was modified in order to be refer to the libraries YT> correctly. YT> However, I got a error message, YT> YT> ........... YT> .o mm_init.o mm_dim.o mm_qmmm_forcedr.o mm_mdmain.o mm_coninp.o YT> mm_cpmd_add_MM_forces_f77.o mm_cpmd_esp_charges_f77.o YT> mm_cpmd_ext_pot_f77.o vdwin.o vdw.o sysdepend.o shmemory.o YT> -L/home/private2/prog/ABSOFT/opt/absoft/lib -llibf90math -llibimsl YT> -llibimslblas -llibimslf90 -lblas -llibU77 First of all be sure that: -) in /home/private2/prog/ABSOFT/opt/absoft/lib you have the following file: -) libf90math.a then change the Makefile line : -llibf90math with the following: -lf90math It's just a problem of library linking.. Let me know whether this helps.. cheers teo YT> /usr/bin/ld: cannot find -llibf90math YT> collect2: ld returned 1 exit status YT> YT> make: *** [cpmd.x] Error 1 YT> ............ YT> YT> and compile is not successful. YT> YT> I can not understand this message. Please help me solving this problem. YT> YT> Modified makefile is the following. YT> YT> #----------------------------------------------------------------------- YT> ----- YT> # Makefile for cpmd.x (plane wave electronic calculation) YT> # Configuration: PC-ABSOFT YT> # Creation of Makefile: Apr 4 2003 YT> # on Linux soranuma 2.4.18-3smp #1 SMP Thu Apr 18 07:27:31 EDT 2002 YT> i686 unknown YT> # Author: soranuma!yamaki YT> #----------------------------------------------------------------------- YT> ----- YT> # YT> SHELL = /bin/sh YT> # YT> #--------------- Default Configuration for PC-ABSOFT ---------------SRC YT> = . YT> DEST = .BIN = . YT> #QMMM_FLAGS = -D__QMECHCOUPL YT> #QMMM_LIBS = -L. -lmm YT> FFLAGS = YT> #LFLAGS = -L/usr/absoft/lib -llapack -lblas -lU77 $(QMMM_LIBS) YT> LFLAGS = -L/home/private2/prog/ABSOFT/opt/absoft/lib -llibf90math YT> -llibimsl -llibimslblas -llibimslf90 -lblas -llibU77 $(QMMM_LIBS) YT> #LFLAGS = -L/home/private2/prog/ABSOFT/opt/absoft/lib -llibf77math YT> -llibimsl -llibimslblas -llibimslf90 -lblas -llibU77 $(QMMM_LIBS) YT> CFLAGS = YT> CPP = /lib/cpp -P -C YT> CPPFLAGS = -D__NOINT8 -D__Linux -D__ABSOFT -DLAPACK -DFFT_DEFAULT YT> NOOPT_FLAG = YT> CC = cc -O YT> FC = f90 -c -B108 -B100 -YEXT_NAMES=LCS -O YT> LD = f90 -O -X-Bstatic YT> AR = YT> #----------------------------------------------------------------------- YT> ----- YT> # YT> # LIST OF FILES YT> # YT> YT> YT> Thank you in advance. YT> YT> Sincerely yours. YT> T.Yamaki YT> YT> _______________________________________________ YT> CPMD-list mailing list YT> CPMD-list at cpmd.org YT> http://www.cpmd.org/mailman/listinfo/cpmd-list YT> , , %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% /( )` %% \ \___ / | %% /- _ `-/ ' %% (/\/ \ \ /\ Teodoro Laino %% / / | ` \ %% O O ) / | PhD. in Molecular Biophysics %% `-^--'`< ' NEST - Physics Labs %% (_.) _ ) / Scuola Normale Superiore di Pisa %% `.___/` / Piazza dei Cavalieri 7 %% `-----' / 56100 Pisa (Italy) %% <----. __ / __ \ %% <----|====O)))==) \) /==== %% <----' `--' `.__,' \ Tel. +39-329-9624004 (Mobile) %% | | Tel. +39-050-509412 (Office) %% \ / %% ______( (_ / \______ %% ,' ,-----' | \ http://www.nest.sns.it/~teo %% `--{__________) \/ %% %% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% From ct at chemie.hu-berlin.de Fri May 16 08:46:06 2003 From: ct at chemie.hu-berlin.de (Christian Tuma) Date: Fri, 16 May 2003 08:46:06 +0200 Subject: [CPMD-list] Problem doing transition state search Message-ID: <20030516064606.GA21373@uranus.chemie.hu-berlin.de> Dear CPMD developers, before CPMD release 3.7 I used an external optimiser to locate transition states with CPMD. Now I was trying to do this using version 3.7 and its recently implemented internal transition search algorithm. The system under study was a small zeolite system (37 atoms per unit cell). First I did a frequency calculation, the structure was already VERY close to the transition state: -1273.4964 -31.8808 34.0303 45.3821 82.6421 84.8253 94.0645 105.0102 ... translation projected: -1272.7988 .0000 .0000 .0000 80.7846 82.4328 89.3480 104.7284 ... Then I tried a transition state search using the calculated Hessian: &CPMD OPTIMIZE GEOMETRY CONVERGENCE ORBITALS 1.0E-06 CONVERGENCE GEOMETRY 1.0E-04 RESTART HESSIAN LATEST RFO ORDER=1 &END ... The first relaxation steps looked like this: *** GNMAX= 1.132939E-04 ETOT= -432.246618 *** *** GNMAX= 2.066742E-04 [9.24E-03] ETOT= -432.246615 *** *** GNMAX= 3.516007E-04 [9.36E-03] ETOT= -432.246610 *** *** GNMAX= 5.045955E-04 [9.47E-03] ETOT= -432.246602 *** *** GNMAX= 6.773491E-04 [1.13E-02] ETOT= -432.246591 *** *** GNMAX= 1.041622E-03 [2.61E-02] ETOT= -432.246554 *** *** GNMAX= 1.026987E-03 [5.18E-03] ETOT= -432.246553 *** *** GNMAX= 9.818658E-04 [6.93E-03] ETOT= -432.246550 *** *** GNMAX= 9.562766E-04 [4.51E-03] ETOT= -432.246550 *** *** GNMAX= 9.000179E-04 [4.27E-03] ETOT= -432.246547 *** *** GNMAX= 8.270440E-04 [7.04E-03] ETOT= -432.246550 *** *** GNMAX= 8.178444E-04 [1.79E-03] ETOT= -432.246550 *** *** GNMAX= 7.637884E-04 [4.46E-03] ETOT= -432.246546 *** *** GNMAX= 7.829667E-04 [7.01E-03] ETOT= -432.246533 *** *** GNMAX= 8.246068E-04 [8.47E-03] ETOT= -432.246510 *** ... Then it continued with almost the same values for GNMAX and ETOT. But I expected that it would take only a few steps to get from GNMAX= 1.132939E-04 to GNMAX < 1.0E-04 to fulfil my given convergence criterion that I usually use with ordinary geometry optimisations. Instead GNMAX got bigger. Did I make something wrong in my input file? Any recommendations for changes? One more thing that puzzled me were statements in the output like the following (example after first step): RFO| MODES: NEGATIVE= 1 ZERO= 9 POSITIVE= 101 LOWEST HESSIAN EIGENVALUES: -.095510 -.000729 .000858 .001499 .004318 .004675 .005900 .007038 Why ZERO= 9 ? From the source code (rrfo.f) I got the answer, every eigenvalue between -0.01 and 0.01 is treated as 0. I do not know the details of the RFO scheme, but after projection of the translation there are only 3 eigenvalues exactly 0. Does this somehow influence the performance of the transition state search? Thank you, any advice is welcome. Christian. -- Christian Tuma Humboldt-Universitaet Berlin ct at chemie.hu-berlin.de Arbeitsgruppe Quantenchemie (Prof. Sauer) phone: +49-30-20937140 Brook-Taylor-Str. 2, 12489 Berlin, GERMANY fax: +49-30-20937136 http://www.chemie.hu-berlin.de/ag_sauer From cambounet.m at euro.apple.com Fri May 16 08:56:27 2003 From: cambounet.m at euro.apple.com (Matthieu Cambounet) Date: Fri, 16 May 2003 08:56:27 +0200 Subject: [CPMD-list] Mac OS X support In-Reply-To: <20030516064606.GA21373@uranus.chemie.hu-berlin.de> Message-ID: <83DE106E-876B-11D7-913D-000393B3A834@euro.apple.com> Good Morning, I'm desperately looking for a version of CPMD working on the Apple unix system Mac OS X. Also I have been trying to get register on the CPMD web side in order to get the source code for trying to re compile it but without luck (nobody never sent me the username pawd) Can someone can help me ? thank you, Matthieu Cambounet Apple Computer France Tel : +33 1 69 86 32 29 Gsm: + 33 6 73 19 75 03 From cur at zurich.ibm.com Fri May 16 10:39:40 2003 From: cur at zurich.ibm.com (Alessandro Curioni) Date: Fri, 16 May 2003 10:39:40 +0200 Subject: [CPMD-list] Mac OS X support In-Reply-To: <83DE106E-876B-11D7-913D-000393B3A834@euro.apple.com> Message-ID: Dear Mr. Cambounet, we sent an answer to you some weeks ago. In case you have not received it, we are re-sending it now. Best Regards, A.Curioni From jwh at science.uva.nl Fri May 16 11:16:30 2003 From: jwh at science.uva.nl (Jan-Willem Handgraaf) Date: Fri, 16 May 2003 11:16:30 +0200 Subject: [CPMD-list] Problem doing transition state search In-Reply-To: <20030516064606.GA21373@uranus.chemie.hu-berlin.de> References: <20030516064606.GA21373@uranus.chemie.hu-berlin.de> Message-ID: <200305161116.30516.jwh@ct-pc70.chem.uva.nl> Dear Tuma, Due to strong anharmonicities near a transition state, it is often much harder to compute than a ordinary minimum. In my experience, the rational function optimization algorithm as implemented in CPMD is a somewhat crude optimizer and is rather dependent on the choice of the empirical initial Hessian. However, your values of the gradients on the ions are already quite small? I suspect that you are at the limit of what is possible with this algorithm. Hope this helps, Jan-Willem Handgraaf P.S. I've tested CPMD against the atomic-orbital based ADF package a got a very good agreement with respect to transition-state geometries of transition metal complexes. -- Jan-Willem Handgraaf Department of Chemical Engineering University of Amsterdam Nieuwe Achtergracht 166 1018 WV Amsterdam The Netherlands Room: C6.18 Tel: +31-20-5256492 Fax: +31-20-5255604 Email: jwh at science.uva.nl Web: http://molsim.chem.uva.nl/ From ZRLSRB at ch.ibm.com Sun May 18 21:45:52 2003 From: ZRLSRB at ch.ibm.com (Salomon Billeter) Date: Sun, 18 May 2003 21:45:52 +0200 Subject: [CPMD-list] Problem doing transition state search Message-ID: Dear Christian, you can try the linear-scaling geometry optimizer which is new in CPMD. It can do both linear-scaling geometry optimization and transition-state search (P-RFO, cubic scaling, and microiterative based on P-RFO and L-BFGS, linear scaling). You'll find details in Comput. Mat. Sci. 27, 437-445, 2003. The manual tells you the options. Please see also the hints section. P-RFO partitions your configurational space into two subspaces, one spanned by the Hessian eigenvector along which you want to maximize the energy and the orthogonal space for which the energy is to be minimized. You can either follow an eigenmode (based on overlap), or you can select a fixed mode. The P-RFO implementation in CPMD has a dynamic trust radius scheme based on energy prediction which helps ensure the stability of the optimization. However, it requires you to choose the right mode at the beginning. Best regards, Salomon From jb376 at cam.ac.uk Mon May 19 12:11:27 2003 From: jb376 at cam.ac.uk (Jochen Blumberger) Date: Mon, 19 May 2003 11:11:27 +0100 Subject: [CPMD-list] LSD - KS diag. Message-ID: <3EC8ADCF.3040907@cam.ac.uk> Dear CPMD developers, I encountered a problem for the Kohn-Sham diagonlization with LSD in V3.7.1: The wavefucntion optimization for a sample of 32 waters gives ELECTRONIC GRADIENT: MAX. COMPONENT = 5.10300E-06 NORM = 7.29372E-08 NUCLEAR GRADIENT: MAX. COMPONENT = 4.18221E-02 NORM = 1.19544E-02 TOTAL INTEGRATED ELECTRONIC DENSITY IN G-SPACE = 256.000000 IN R-SPACE = 256.000000 TOTAL INTEGRATED SPIN DENSITY .000026 (K+E1+L+N+X) TOTAL ENERGY = -548.80201537 A.U. (K) KINETIC ENERGY = 412.08735410 A.U. (E1=A-S+R) ELECTROSTATIC ENERGY = -367.57551500 A.U. (S) ESELF = 404.26149474 A.U. (R) ESR = 24.02130066 A.U. (L) LOCAL PSEUDOPOTENTIAL ENERGY = -513.72725987 A.U. (N) N-L PSEUDOPOTENTIAL ENERGY = 54.97660217 A.U. (X) EXCHANGE-CORRELATION ENERGY = -134.56319678 A.U. GRADIENT CORRECTION ENERGY = -7.53087044 A.U. **************************************************************** but the diagonalization using the Restart file of the optimization run gives: TOTAL INTEGRATED ELECTRONIC DENSITY IN G-SPACE = 256.000000 IN R-SPACE = 256.000000 TOTAL INTEGRATED SPIN DENSITY .000026 (B+E2+X-V) TOTAL ENERGY = -549.88465526 A.U. (B) BAND ENERGY = -88.88410491 A.U. (E2=I-H-S+R) ELECTROSTATIC ENERGY = -499.36486171 A.U. (S) ESELF = 404.26149474 A.U. (R) ESR = 24.02130066 A.U. (X) EXCHANGE-CORRELATION ENERGY = -134.56319678 A.U. (V) EXCHANGE-CORRELATION POTEN. = -172.92750821 A.U. GRADIENT CORRECTION ENERGY = -7.53087044 A.U. ==------------------------------------------------------------== == NFI= 1 ETOT= -549.884655 TCPU= 14.14 == == DRHOMAX= 3.550E-09 DETOT= 0.000E+00 THL= 0.000E+00 == ==------------------------------------------------------------== In the LDA case wavefunction optimization and diagonlization gives the same total energy which also coincides with the total energy of the LSD wavefunction optimization. Here's the output for LDA diagonalization: TOTAL INTEGRATED ELECTRONIC DENSITY IN G-SPACE = 256.000000 IN R-SPACE = 256.000000 (B+E2+X-V) TOTAL ENERGY = -548.80201716 A.U. (B) BAND ENERGY = -88.88413317 A.U. (E2=I-H-S+R) ELECTROSTATIC ENERGY = -499.36483919 A.U. (S) ESELF = 404.26149474 A.U. (R) ESR = 24.02130066 A.U. (X) EXCHANGE-CORRELATION ENERGY = -134.56321558 A.U. (V) EXCHANGE-CORRELATION POTEN. = -174.01017083 A.U. GRADIENT CORRECTION ENERGY = -7.53087727 A.U. ==------------------------------------------------------------== == NFI= 1 ETOT= -548.802017 TCPU= 7.11 == == DRHOMAX= 4.225E-09 DETOT= 0.000E+00 THL= 0.000E+00 == ==------------------------------------------------------------== The KS energy levels of the LSD and LDA diagonalization are ~identical (as they should be) but the total energies are 549.88465526 for LSD and 548.80201716 for LDA. The difference comes from the EXCHANGE-CORRELATION POTENTIAL: -172.92750821 for LSD, -174.01017083 for LDA. Could there be a bug in the calculation of the total exch.corr.Potential in the LSD case ? (In previous versions LSD diag. worked well) many thanks best regards Jochen Blumberger From mkosmows at mailbox.syr.edu Mon May 19 19:24:02 2003 From: mkosmows at mailbox.syr.edu (mkosmows) Date: Mon, 19 May 2003 13:24:02 -0400 Subject: [CPMD-list] Thinking about buying an AMD Opteron system... Message-ID: <3ED3301E@OrangeMail> Dear CPMD community: I am thinking about buying an AMD dual Opteron machine to run CPMD on and was wondering if anyone has any experience or opinions to share. Most importantly, I am hoping to find out for sure if an Opteron based machine will be able to allocate more than 1 Gb of memory to a single process. If I do make the purchase, I would use the 64 bit version of Mandrake Linux 9.0 and the 64 bit version of the PGI compiler. Thank you for any advice you all might have, Mark Kosmowski Chemistry Department Syracuse University mkosmows at syr.edu From axel.kohlmeyer at theochem.ruhr-uni-bochum.de Mon May 19 22:10:05 2003 From: axel.kohlmeyer at theochem.ruhr-uni-bochum.de (Axel Kohlmeyer) Date: Mon, 19 May 2003 22:10:05 +0200 Subject: [CPMD-list] Thinking about buying an AMD Opteron system... In-Reply-To: <3ED3301E@OrangeMail> Message-ID: <200305192010.h4JKA5B15622@yello.theochem.ruhr-uni-bochum.de> On Mon, 19 May 2003 13:24:02 -0400 mkosmows wrote: > Dear CPMD community: > > I am thinking about buying an AMD dual Opteron machine to run CPMD on > and was wondering if anyone has any experience or opinions to share. Most well, if you want to run in 64-bit mode, you probably have to 'port' cpmd to that platform (which probably is not a big deal, since you can 'steal' most of the flags from the alpha-linux configuration). > importantly, I am hoping to find out for sure if an Opteron based machine will > be able to allocate more than 1 Gb of memory to a single process. If I do > make the purchase, I would use the 64 bit version of Mandrake Linux 9.0 and > the 64 bit version of the PGI compiler. well, you also need a very well optimized LAPACK/BLAS implementation. e.g. for the amd athlon platform this can mean a 2x speedup for CPMD runs (up to 8x for DGEMM). AFAIK, there is no such optimized LAPACK/BLAS available (yet). you should check the ATLAS homepage to be sure. > > Thank you for any advice you all might have, since cpmd does distributed memory parallization, you should also check, how the opteron machine compares to a small dual-athlon (with not too fast FSB266 cpus) cluster. in combination with dolphin/SCI interconnect we have found nearly ideal scaling and extremely high reliability. the latter cannot be expected from a new platform (once you got the code actually working). on the other hand: _if_ you get it to work (and don't have to worry about wasted money), it probably will be a very nice machine. the technical specs indicate, that the opteron cpu should be much better suited for cpmd than current ia32 cpus. but then again: if money really is not an issue, better get a power4 or an alpha-cpu based workstation. those are known to work, well optimized and also 64-bit. regards, axel kohlmeyer. > > Mark Kosmowski > > Chemistry Department > Syracuse University > mkosmows at syr.edu > > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > -- ======================================================================= Axel Kohlmeyer e-mail: axel.kohlmeyer at theochem.ruhr-uni-bochum.de Lehrstuhl fuer Theoretische Chemie Phone: ++49 (0)234/32-26673 Ruhr-Universitaet Bochum - NC 03/53 Fax: ++49 (0)234/32-14045 D-44780 Bochum http://www.theochem.ruhr-uni-bochum.de ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From cdy2000 at vip.sina.com Wed May 21 13:46:43 2003 From: cdy2000 at vip.sina.com (cdy2000 at vip.sina.com) Date: Wed, 21 May 2003 19:46:43 +0800 Subject: [CPMD-list] please Re to the top 10 :-) Message-ID: <20030521114643.4440.qmail@vip.sina.com> Hello,every CPMDer! send a mail to support our mail-list (ps re my mail to the top 10) :-) Two questions: 1. I use the fhi98pp to generate the pp, use "./psgen -o atomname atomname:h.ini", other atom can be generated but hydrogen, the info is segmentation fault rm: No match foreach: No match I find someone ask the same question in mail-list last year, but the answer was not clear. 2. the programme generates a set of files, which should I use in CPMD? only *.cpi, or all of? how to denote in &ATOMS ...&END??use *PPNAME KLEINMAN-BYLANDER, or add some other keywords? Thx for any help! Sincerely Cao Dayong ______________________________________ =================================================================== From hutter at pci.unizh.ch Fri May 23 10:53:16 2003 From: hutter at pci.unizh.ch (Juerg Hutter) Date: Fri, 23 May 2003 10:53:16 +0200 (MEST) Subject: [CPMD-list] LSD - KS diag. In-Reply-To: <3EC8ADCF.3040907@cam.ac.uk> References: <3EC8ADCF.3040907@cam.ac.uk> Message-ID: Hi Jochen I had some troubles finding this bug but finaly I was able to track it down. The fixed file vofrhob.F is attached. best wishes 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, 19 May 2003, Jochen Blumberger wrote: > Dear CPMD developers, > > > I encountered a problem for the Kohn-Sham diagonlization with LSD in V3.7.1: > > The wavefucntion optimization for a sample of 32 waters gives > > ELECTRONIC GRADIENT: > MAX. COMPONENT = 5.10300E-06 NORM = 7.29372E-08 > NUCLEAR GRADIENT: > MAX. COMPONENT = 4.18221E-02 NORM = 1.19544E-02 > > > TOTAL INTEGRATED ELECTRONIC DENSITY > IN G-SPACE = 256.000000 > IN R-SPACE = 256.000000 > TOTAL INTEGRATED SPIN DENSITY .000026 > > (K+E1+L+N+X) TOTAL ENERGY = -548.80201537 A.U. > (K) KINETIC ENERGY = 412.08735410 A.U. > (E1=A-S+R) ELECTROSTATIC ENERGY = -367.57551500 A.U. > (S) ESELF = 404.26149474 A.U. > (R) ESR = 24.02130066 A.U. > (L) LOCAL PSEUDOPOTENTIAL ENERGY = -513.72725987 A.U. > (N) N-L PSEUDOPOTENTIAL ENERGY = 54.97660217 A.U. > (X) EXCHANGE-CORRELATION ENERGY = -134.56319678 A.U. > GRADIENT CORRECTION ENERGY = -7.53087044 A.U. > > **************************************************************** > > > but the diagonalization using the Restart file of the optimization run > gives: > > TOTAL INTEGRATED ELECTRONIC DENSITY > IN G-SPACE = 256.000000 > IN R-SPACE = 256.000000 > TOTAL INTEGRATED SPIN DENSITY .000026 > > (B+E2+X-V) TOTAL ENERGY = -549.88465526 A.U. > (B) BAND ENERGY = -88.88410491 A.U. > (E2=I-H-S+R) ELECTROSTATIC ENERGY = -499.36486171 A.U. > (S) ESELF = 404.26149474 A.U. > (R) ESR = 24.02130066 A.U. > (X) EXCHANGE-CORRELATION ENERGY = -134.56319678 A.U. > (V) EXCHANGE-CORRELATION POTEN. = -172.92750821 A.U. > GRADIENT CORRECTION ENERGY = -7.53087044 A.U. > > ==------------------------------------------------------------== > == NFI= 1 ETOT= -549.884655 TCPU= 14.14 == > == DRHOMAX= 3.550E-09 DETOT= 0.000E+00 THL= 0.000E+00 == > ==------------------------------------------------------------== > > > In the LDA case wavefunction optimization and diagonlization gives the > same total energy which also coincides with the total energy of the LSD > wavefunction optimization. > Here's the output for LDA diagonalization: > > TOTAL INTEGRATED ELECTRONIC DENSITY > IN G-SPACE = 256.000000 > IN R-SPACE = 256.000000 > > (B+E2+X-V) TOTAL ENERGY = -548.80201716 A.U. > (B) BAND ENERGY = -88.88413317 A.U. > (E2=I-H-S+R) ELECTROSTATIC ENERGY = -499.36483919 A.U. > (S) ESELF = 404.26149474 A.U. > (R) ESR = 24.02130066 A.U. > (X) EXCHANGE-CORRELATION ENERGY = -134.56321558 A.U. > (V) EXCHANGE-CORRELATION POTEN. = -174.01017083 A.U. > GRADIENT CORRECTION ENERGY = -7.53087727 A.U. > > ==------------------------------------------------------------== > == NFI= 1 ETOT= -548.802017 TCPU= 7.11 == > == DRHOMAX= 4.225E-09 DETOT= 0.000E+00 THL= 0.000E+00 == > ==------------------------------------------------------------== > > > The KS energy levels of the LSD and LDA diagonalization are ~identical > (as they should be) but the total energies are 549.88465526 for LSD and > 548.80201716 for LDA. The difference comes from the > EXCHANGE-CORRELATION POTENTIAL: -172.92750821 for LSD, -174.01017083 for > LDA. Could there be a bug in the calculation of the total > exch.corr.Potential in the LSD case ? (In previous versions LSD diag. > worked well) > > many thanks > > best regards > > Jochen Blumberger > > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > -------------- next part -------------- C ================================================================== SUBROUTINE VOFRHOB(FION,RHOE,V,VTEMP,SCR,LSCR,TFOR,TSTRESS) C ==--------------------------------------------------------------== C == COMPUTES == C == THE ONE-PARTICLE POTENTIAL V IN REAL SPACE == C == THE TOTAL ENERGY ETOT == C == THE FORCES FION ACTING ON THE IONS == C ==--------------------------------------------------------------== C == RHOE: in electronic density in real space == C == out total electronic potential in real space == C ==--------------------------------------------------------------== IMPLICIT NONE INCLUDE 'system.h' INCLUDE 'spin.inc' INCLUDE 'ions.inc' INCLUDE 'ener.inc' INCLUDE 'elct.inc' INCLUDE 'cppt.inc' INCLUDE 'pslo.inc' INCLUDE 'simul.inc' INCLUDE 'fft.inc' INCLUDE 'geq0.inc' INCLUDE 'cvan.inc' INCLUDE 'nlps.inc' INCLUDE 'sfac.inc' INCLUDE 'nlcc.inc' INCLUDE 'strs.inc' INCLUDE 'str2.inc' INCLUDE 'efld.inc' INCLUDE 'extpot.inc' C Arguments INTEGER LSCR REAL*8 FION(3,NAX,NSX), & RHOE(NNR1,NLSD),SCR(LSCR) COMPLEX*16 V(MAXFFT,NLSD),VTEMP(NHG,NLSD) LOGICAL TFOR,TSTRESS C Variables DIMENSION F(N),DQG(*) COMPLEX*16 VTMP(NHG,NLSX) REAL*8 RHOVAL(NNR1,NLSD),EXTPOT(*),EXTF(*),GRAD(NNR1,NLSD*4) POINTER (IP_VTMP,VTMP),(IP_GRAD,GRAD),(IP_RHOVAL,RHOVAL) REAL*8 FNL(IMAGP,LDF1,NHXS,*),DFNL(*),SXC,SGCX,SGCC,VGC INTEGER ISUB,IG,IR,NNRS,KK, & IL_VTMP,IL_GRAD,IL_RHOVAL,LENGTH LOGICAL DEBUG C ==--------------------------------------------------------------== CALL TISET(' VOFRHOB',ISUB) C SCR partition. DEBUG=.FALSE. CALL SCRSET(SCR,LSCR,DEBUG) IL_VTMP = MAX(NNR1,NHG*2)*NLSX !GCENER and COREC IF(TGC) THEN IL_GRAD=NNR1*NLSD*4 !GRADEN ELSE IL_GRAD=0 ENDIF IF(TINLC) THEN IL_RHOVAL=NNR1*NLSD !XCENER ELSE IL_RHOVAL=0 ENDIF CALL SCRPTR(SCR,LSCR,IP_VTMP,IL_VTMP,'VTMP',DEBUG) CALL SCRPTR(SCR,LSCR,IP_GRAD,IL_GRAD,'GRAD',DEBUG) CALL SCRPTR(SCR,LSCR,IP_RHOVAL,IL_RHOVAL,'RHOVAL',DEBUG) C ==--------------------------------------------------------------== C == ADD CORE CHARGE TO RHOE == C ==--------------------------------------------------------------== IF(TINLC) THEN CALL DCOPY(NNR1*NLSD,RHOE(1,1),1,RHOVAL(1,1),1) CALL COREC(RHOE,VTMP,V) CALL DCOPY(2*NHG,VTEMP(1,1),1,VNLT(1),1) ENDIF C ==-------------------------------------------------------------== C == VTEMP (Potential in G-Space) -FFT-> V(R) == C ==--------------------------------------------------------------== CALL AZZERO(V,2*MAXFFT) #ifdef __NEC !CDIR NODEP #endif DO IG=1,NHG V(INDZ(IG),1) = DCONJG(VTEMP(IG,1)) V(NZH(IG),1) = VTEMP(IG,1) ENDDO IF(GEQ0) V(NZH(1),1) = VTEMP(1,1) CALL INVFFT(V) C ==--------------------------------------------------------------== C == ADD EXTERNAL POTENTIAL TO V == C ==--------------------------------------------------------------== IF(TEXTFLD) THEN C..potential from classical interface DO IR=1,NNR1 V(IR,1)=V(IR,1)+DCMPLX(EXTF(IR),0.D0) ENDDO ENDIF IF(TEXPOT) THEN C..static external potential DO IR=1,NNR1 V(IR,1)=V(IR,1)+DCMPLX(EXTPOT(IR),0.D0) ENDDO ENDIF IF((TEXPOT.OR.TEXTFLD).AND.TINLC) THEN CALL FWFFT(V) DO IG=1,NHG VNLT(IG) = V(NZH(IG),1) ENDDO CALL INVFFT(V) ENDIF C ==--------------------------------------------------------------== C == COMPUTE EXCHANGE AND CORRELATION ENERGY (EXC) == C ==--------------------------------------------------------------== IF(TINLC) THEN CALL XCENER(SXC,VXC,RHOVAL,RHOE,V) ELSE CALL XCENER(SXC,VXC,RHOE,RHOE,V) ENDIF SGCX = 0.0D0 SGCC = 0.0D0 VGC = 0.0D0 IF(TGC) THEN C ==------------------------------------------------------------== C == CALCULATE THE GRADIENT OF THE DENSITY == C ==------------------------------------------------------------== IF(TLSD) THEN IF(TSTRESS.OR.TDIAG) THEN CALL DCOPY(2*NNR1,V(1,1),1,DQG(1),1) CALL DCOPY(2*NNR1,V(1,2),1,DQG(NNR1+1),1) ENDIF CALL FWFFT(V(1,1)) CALL ZGTHR(NHG,V(1,1),VTEMP(1,1),NZH) CALL FWFFT(V(1,2)) CALL ZGTHR(NHG,V(1,2),VTEMP(1,2),NZH) DO IR=1,NNR1 RHOE(IR,1)=RHOE(IR,1)-RHOE(IR,2) ENDDO CALL GRADEN(RHOE(1,1),V,GRAD(1,1),VTMP) CALL GRADEN(RHOE(1,2),V,GRAD(1,5),VTMP) ELSE IF(TSTRESS.OR.TDIAG) CALL DCOPY(2*NNR1,V(1,1),1,DQG(1),1) CALL FWFFT(V) CALL ZGTHR(NHG,V,VTEMP,NZH) CALL GRADEN(RHOE,V,GRAD,VTMP) ENDIF C ==--------------------------------------------------------------== C == GRADIENT CORRECTION TO THE EXCHANGE ENERGY (EGCX) == C ==--------------------------------------------------------------== IF(TLSD) THEN IF(TSTRESS) CALL STOPGM('VOFRHOB','NO STRESS WITH LSD AND GC') CALL GCLSD(SGCX,SGCC,RHOE,V,VTEMP,VTMP,GRAD) IF(TSTRESS.OR.TDIAG) THEN CALL DAXPY(2*NNR1,-1.0D0,V(1,1),1,DQG(1),1) CALL DAXPY(2*NNR1,-1.0D0,V(1,2),1,DQG(NNR1+1),1) CALL DSCAL(4*NNR1,-1.0D0,DQG(1),1) VGC=0.0D0 IF(TINLC) THEN DO IR=1,NNR1 VGC=VGC+(RHOVAL(IR,1)-RHOVAL(IR,2))*DREAL(DQG(IR)) VGC=VGC+RHOVAL(IR,2)*DREAL(DQG(NNR1+IR)) ENDDO ELSE DO IR=1,NNR1 VGC=VGC+RHOE(IR,1)*DREAL(DQG(IR)) VGC=VGC+RHOE(IR,2)*DREAL(DQG(NNR1+IR)) ENDDO ENDIF ENDIF ELSE CALL GCENER(SGCX,SGCC,RHOE,V,VTEMP,VTMP,GRAD,TSTRESS) IF(TSTRESS.OR.TDIAG) THEN CALL DAXPY(2*NNR1,-1.0D0,V(1,1),1,DQG(1),1) CALL DSCAL(2*NNR1,-1.0D0,DQG(1),1) VGC=0.0D0 IF(TINLC) THEN DO IR=1,NNR1 VGC=VGC+RHOVAL(IR,1)*DREAL(DQG(IR)) ENDDO ELSE DO IR=1,NNR1 VGC=VGC+RHOE(IR,1)*DREAL(DQG(IR)) ENDDO ENDIF ENDIF ENDIF ENDIF C ==--------------------------------------------------------------== C == XC CONTRIBUTION TO STRESS (VANDERBILT CHARGE) == C ==--------------------------------------------------------------== IF(TIVAN.AND.TSTRESS) * CALL XCSTR(DRHOVG,RHOE,DQG,DQG(MAXFFT+1)) C ==--------------------------------------------------------------== C == NLCC CONTRIBUTION TO STRESS == C ==--------------------------------------------------------------== IF(TINLC.AND.TSTRESS) CALL NLCCSTR(DQG,DQG(NLSD*MAXFFT+1)) C ==--------------------------------------------------------------== C == RELEASE TEMPORARY ARRAYS FROM SCRATCH SPACE == C ==--------------------------------------------------------------== CALL SCRREL(SCR,LSCR,DEBUG) C ==--------------------------------------------------------------== C == V CONTAINS THE TOTAL POTENTIAL IN R-SPACE == C == MOVE IT TO RHOE == C ==--------------------------------------------------------------== DO IR=1,NNR1 RHOE(IR,1)=DREAL(V(IR,1)) ENDDO IF(TLSD) THEN DO IR=1,NNR1 RHOE(IR,2)=DREAL(V(IR,2)) ENDDO ENDIF C ==--------------------------------------------------------------== C == CORRECT ENERGY FOR CORE CHARGE == C ==--------------------------------------------------------------== IF(TINLC) THEN SXC=SXC-EXCCO SGCX=SGCX-EGCXCO SGCC=SGCC-EGCCCO ENDIF IF(TIVAN.OR.(TINLC.AND.TFOR)) THEN C ==--------------------------------------------------------------== C == TRANSF. TOTAL POTENTIAL IN G-SPACE == C == PUT THE TOTAL POTENTIAL IN G-SPACE INTO VTEMP == C ==--------------------------------------------------------------== CALL FWFFT(V) CALL ZGTHR(NHG,V,VTEMP,NZH) IF(TLSD) THEN CALL FWFFT(V(1,2)) CALL ZGTHR(NHG,V(1,2),VTEMP(1,2),NZH) ENDIF ENDIF IF(TIVAN) THEN IF(IMAGP.EQ.2)CALL STOPGM('VOFRHOB','K-POINT NOT IMPLEMENTED') C ==--------------------------------------------------------------== C == CALCULATE DEEQ FOR VANDERBILT PP == C == FORCE ON IONS DUE TO THE "VANDERBILT CHARGE" == C ==--------------------------------------------------------------== IF(TLSD) THEN CALL NEWD(FNL(1,1,1,1),DEEQ(1,1,1,1),F(1), * VTEMP(1,1),FION,SCR,LSCR,NSUP,TFOR) CALL NEWD(FNL(1,1,1,NSUP+1),DEEQ(1,1,1,2),F(NSUP+1), & VTEMP(1,2),FION,SCR,LSCR,NSDOWN,TFOR) ELSE CALL NEWD(FNL,DEEQ,F,VTEMP,FION,SCR,LSCR,N,TFOR) ENDIF ENDIF C ==--------------------------------------------------------------== C == FORCE ON IONS DUE TO NLCC == C ==--------------------------------------------------------------== IF(TINLC.AND.TFOR) CALL COFOR(FION,VTEMP) C ==--------------------------------------------------------------== C == SINGLE PIECES OF THE ENERGY: == C ==--------------------------------------------------------------== NNRS = NR1S*NR2S*NR3S EGC=(SGCC+SGCX)*OMEGA/DFLOAT(NNRS) EXC=SXC*OMEGA/DFLOAT(NNRS) + EGC VXC=(VXC + VGC)*OMEGA/DFLOAT(NNRS) IF(TSTRESS) THEN DO KK=1,6 DEXC(KK) = DEXC(KK) + (EXC-VXC)*DELTA(ALPHA(KK),BETA(KK)) ENDDO ENDIF CALL TIHALT(' VOFRHOB',ISUB) C ==--------------------------------------------------------------== RETURN END C ================================================================== SUBROUTINE GIVE_SCR_VOFRHOB(LVOFRHOB,TAG) C ==--------------------------------------------------------------== IMPLICIT NONE INCLUDE 'system.h' !NHG TGC INCLUDE 'spin.inc' !NLSD NLSX INCLUDE 'pslo.inc' !TIVAN INCLUDE 'nlcc.inc' !TINLC C Arguments INTEGER LVOFRHOB CHARACTER TAG*30 C Variables INTEGER LTGC,LTINLC C ==--------------------------------------------------------------== IF(TGC) THEN LTGC=NNR1*NLSD*4 !GRADEN ELSE LTGC=0 ENDIF IF(TINLC) THEN !XCENER LTINLC=NNR1*NLSD ELSE LTINLC=0 ENDIF LVOFRHOB = MAX(NNR1,NHG*2)*NLSX+LTGC+LTINLC !GCENER and COREC IF(TIVAN) LVOFRHOB=MAX(LVOFRHOB,NHG*4+NAX) !NEWD LVOFRHOB = LVOFRHOB + 100 !For boundary checks in SCRPTR TAG='MAX(NNR1,NHG*2)*NLSX+NNR1*NLSD*4' C ==--------------------------------------------------------------== RETURN END C ================================================================== From arodrig at cscs.ch Fri May 23 14:44:31 2003 From: arodrig at cscs.ch (Antonio Rodriguez Fortea) Date: 23 May 2003 14:44:31 +0200 Subject: [CPMD-list] Project Wavefunction on Atomic Basis Set of Gaussians Message-ID: <1053693871.2533.34.camel@rancate.cscs.ch> Dear CPMD users, I am trying to project the wavefunction to an atomic basis set (Gaussians) for the small test system NO2. I want to use an atomic basis set of double-zeta quality, i.e. two functions for each l-channel. Following the instructions in the Users Manual v. 3.7.0 (Basis Set), I have created the corresponding files in the directory where are the pseudopotentials. Checking in the source (setbasis.f) where the Atomic Basis is read: ************************************************************************* IF(INDEX(LINE,'GAUSS').NE.0) THEN C Gaussians READ(IFNUM,*) READ(IFNUM,*) LGMAX DO IL=1,LGMAX READ(IFNUM,*) READ(IFNUM,*) NFMAX(IL),NALP(IL) READ(IFNUM,*) READ(IFNUM,*) (GEXP(I,IL),I=1,NALP(IL)) DO J=1,NFMAX(IL) READ(IFNUM,*) (GCOE(I,J,IL),I=1,NALP(IL)) ENDDO ENDDO MESHAT(IS)=256 DCLOG=1.050 CLOGAT(IS)=DLOG(DCLOG) ATRG(1,IS)=.7142857E-03 DO IR=2,MESHAT(IS) ATRG(IR,IS)=DCLOG*ATRG(IR-1,IS) ENDDO DO I=1,4 NTT(I)=0 ENDDO DO I=1,NSHELL(IS) L=LSHELL(I,IS)+1 NTT(L)=NTT(L)+1 CALL GTORPS(ATRG(1,IS),ATWFR(1,I,IS),MESHAT(IS), * L-1,GEXP(1,L),GCOE(1,NTT(L),L),NALP(L)) ENDDO ELSE . . . **************************************************************************** the files for the basis set should be like this: **************************************************************************** Double-zeta basis Ahlrich et al. 2 Functions for l=0 (s) 2 2 0.587318566 0.187645923 1.000000000 0.000000000 0.000000000 1.000000000 Functions for l=1 (p) 2 4 13.571470233 2.925737287 0.799277508 0.219543480 -0.040072399 -0.218070450 -0.512944660 0.000000000 0.000000000 0.000000000 0.000000000 1.000000000 **************************************************************************** But in the output of a PROPERTIES CALCULATION (PROJECT WAVEFUNCTION) I get: **************************************************************************** WAVEFUNCTIONS IN ATOMIC ORBITAL BASIS ****** ALPHA SPIN ****** ORBITAL 1 2 3 4 5 6 7 8 COMPLETNESS 0.984 0.987 0.955 0.958 0.889 0.888 0.907 0.908 OCCUPATION 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1 N S -0.444 0.000 -0.596 0.000 -0.119 0.000 0.000 0.000 Px -0.087 0.000 0.065 0.000 0.590 0.000 0.000 0.000 Pz 0.000 0.000 0.000 0.000 0.000 -0.628 0.000 0.000 Py 0.000 -0.260 0.000 0.452 0.000 0.000 0.000 -0.019 2 O S -0.493 -0.614 0.472 -0.284 -0.081 0.000 0.000 0.008 Px 0.057 0.043 0.173 -0.288 0.344 0.000 0.000 0.543 Pz 0.000 0.000 0.000 0.000 0.000 -0.450 0.674 0.000 Py 0.114 0.064 0.269 -0.393 -0.307 0.000 0.000 -0.402 3 O S -0.493 0.614 0.472 0.284 -0.081 0.000 0.000 -0.008 Px 0.057 -0.043 0.173 0.288 0.344 0.000 0.000 -0.543 Pz 0.000 0.000 0.000 0.000 0.000 -0.450 -0.674 0.000 Py -0.114 0.064 -0.269 -0.393 0.307 0.000 0.000 -0.402 **************************************************************************** that is, only ONE function for each l-channel. So my question is: does anyone know if is it actually possible to project the wavefunction on a double-zeta atomic basis set? And if this is the case, could anyone please tell me what I am doing wrong or tell me any suggestions? Thank you very much in advance, Antonio Rodriguez-Fortea -- Antonio Rodriguez Fortea Phone: +41 (0)91 610 8258 ETH Zurich/CSCS Fax: +41 (0)91 610 8282 Via Cantonale, Galleria 2 e-mail: arodrig at cscs.ch CH-6928 Manno (Switzerland) From adrainzhou at yahoo.com.cn Sat May 24 17:26:15 2003 From: adrainzhou at yahoo.com.cn (=?gb2312?q?Adrain=20Zhou?=) Date: Sat, 24 May 2003 23:26:15 +0800 (CST) Subject: [CPMD-list] Problem with MD calculation! Message-ID: <20030524152615.48884.qmail@web15307.mail.bjs.yahoo.com> Dear all, I met such a problem when I performed a MD calculation on SP3. DSPEV : 2538-2028 The job option argument (ARG NO. 1) must be 0, 1, 20, or 21. DSPEV : 2538-2028 The job option argument (ARG NO. 1) must be 0, 1, 20, or 21. DSPEV : 2538-2028 The job option argument (ARG NO. 1) must be 0, 1, 20, or 21. DSPEV : 2538-2099 End of input argument error reporting. For more information, refer to Engineering and Scientific Subroutine Library Guide and Reference (SA22-7272). DSPEV : 2538-2099 End of input argument error reporting. For more information, refer to Engineering and Scientific Subroutine Library Guide and Reference (SA22-7272). DSPEV : 2538-2099 End of input argument error reporting. For more information, refer to Engineering and Scientific Subroutine Library Guide and Reference (SA22-7272). Can anyone tell me what's wrong with it? Thanks in advance! With regards, Zhou --------------------------------- Do You Yahoo!? "????????????!??????????????????????????????????????~???????????????????????????????? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://cpmd.org/pipermail/cpmd-list/attachments/20030524/2226943b/attachment.html From hutter at pci.unizh.ch Sun May 25 11:17:02 2003 From: hutter at pci.unizh.ch (Juerg Hutter) Date: Sun, 25 May 2003 11:17:02 +0200 (MEST) Subject: [CPMD-list] Problem with MD calculation! In-Reply-To: <20030524152615.48884.qmail@web15307.mail.bjs.yahoo.com> References: <20030524152615.48884.qmail@web15307.mail.bjs.yahoo.com> Message-ID: Hi the problem is related to the fact that there is a name conflict between LAPACK and ESSL. Both libraries have routines called DSPEV with different calling sequence but performing the same task. In CPMD we use the LAPACK routine DSPEV (unless you have changed the file util.F or used -DESSL as a preprocessor option, both not recommended). To be sure to call the correct library you have to use the linker option -llapack -lessl -llapack This should be the default when you use the Configure script. However, the libraries might be called differently on your system. Juerg Hutter ---------------------------------------------------------- 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 Sat, 24 May 2003, [gb2312] Adrain Zhou wrote: > Dear all, > > I met such a problem when I performed a MD calculation on SP3. > > > DSPEV : 2538-2028 > > The job option argument (ARG NO. 1) must be 0, 1, 20, or 21. > > > > DSPEV : 2538-2028 > > The job option argument (ARG NO. 1) must be 0, 1, 20, or 21. > > > > DSPEV : 2538-2028 > > The job option argument (ARG NO. 1) must be 0, 1, 20, or 21. > > > > DSPEV : 2538-2099 > > End of input argument error reporting. For more information, refer to > > Engineering and Scientific Subroutine Library Guide and Reference (SA22-7272). > > DSPEV : 2538-2099 > > End of input argument error reporting. For more information, refer to > > Engineering and Scientific Subroutine Library Guide and Reference (SA22-7272). > > DSPEV : 2538-2099 > > End of input argument error reporting. For more information, refer to > > Engineering and Scientific Subroutine Library Guide and Reference (SA22-7272). > > > > Can anyone tell me what's wrong with it? > > > > Thanks in advance! > > With regards, > > Zhou > > > > > > > --------------------------------- > Do You Yahoo!? > "????????????!??????????????????????????????????????~???????????????????????????????? From hutter at pci.unizh.ch Sun May 25 11:59:25 2003 From: hutter at pci.unizh.ch (Juerg Hutter) Date: Sun, 25 May 2003 11:59:25 +0200 (MEST) Subject: [CPMD-list] Project Wavefunction on Atomic Basis Set of Gaussians In-Reply-To: <1053693871.2533.34.camel@rancate.cscs.ch> References: <1053693871.2533.34.camel@rancate.cscs.ch> Message-ID: Hi it works for me. I'm using an input of this type (adapt it for the number of atomic species you have). &BASIS *filename 4 GAUSSIAN 0 0 1 1 &END Juerg Hutter ---------------------------------------------------------- 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 Fri, 23 May 2003, Antonio Rodriguez Fortea wrote: > > Dear CPMD users, > > I am trying to project the wavefunction to an atomic basis set > (Gaussians) for the small test system NO2. I want to use an atomic basis > set of double-zeta quality, i.e. two functions for each l-channel. > Following the instructions in the Users Manual v. 3.7.0 (Basis Set), I > have created the corresponding files in the directory where are the > pseudopotentials. Checking in the source (setbasis.f) where the Atomic > Basis is read: > > ************************************************************************* > IF(INDEX(LINE,'GAUSS').NE.0) THEN > C Gaussians > READ(IFNUM,*) > READ(IFNUM,*) LGMAX > DO IL=1,LGMAX > READ(IFNUM,*) > READ(IFNUM,*) NFMAX(IL),NALP(IL) > READ(IFNUM,*) > READ(IFNUM,*) (GEXP(I,IL),I=1,NALP(IL)) > DO J=1,NFMAX(IL) > READ(IFNUM,*) (GCOE(I,J,IL),I=1,NALP(IL)) > ENDDO > ENDDO > MESHAT(IS)=256 > DCLOG=1.050 > CLOGAT(IS)=DLOG(DCLOG) > ATRG(1,IS)=.7142857E-03 > DO IR=2,MESHAT(IS) > ATRG(IR,IS)=DCLOG*ATRG(IR-1,IS) > ENDDO > DO I=1,4 > NTT(I)=0 > ENDDO > DO I=1,NSHELL(IS) > L=LSHELL(I,IS)+1 > NTT(L)=NTT(L)+1 > CALL GTORPS(ATRG(1,IS),ATWFR(1,I,IS),MESHAT(IS), > * L-1,GEXP(1,L),GCOE(1,NTT(L),L),NALP(L)) > ENDDO > ELSE > . > . > . > **************************************************************************** > > the files for the basis set should be like this: > > **************************************************************************** > Double-zeta basis Ahlrich et al. > 2 > Functions for l=0 (s) > 2 2 > > 0.587318566 0.187645923 > 1.000000000 0.000000000 > 0.000000000 1.000000000 > Functions for l=1 (p) > 2 4 > > 13.571470233 2.925737287 0.799277508 0.219543480 > -0.040072399 -0.218070450 -0.512944660 0.000000000 > 0.000000000 0.000000000 0.000000000 1.000000000 > **************************************************************************** > > But in the output of a PROPERTIES CALCULATION (PROJECT WAVEFUNCTION) I > get: > > **************************************************************************** > > WAVEFUNCTIONS IN ATOMIC ORBITAL BASIS > > ****** ALPHA SPIN ****** > > ORBITAL 1 2 3 4 5 6 > 7 8 > COMPLETNESS 0.984 0.987 0.955 0.958 0.889 0.888 0.907 > 0.908 > OCCUPATION 1.000 1.000 1.000 1.000 1.000 1.000 1.000 > 1.000 > > 1 N S -0.444 0.000 -0.596 0.000 -0.119 0.000 0.000 > 0.000 > Px -0.087 0.000 0.065 0.000 0.590 0.000 0.000 > 0.000 > Pz 0.000 0.000 0.000 0.000 0.000 -0.628 0.000 > 0.000 > Py 0.000 -0.260 0.000 0.452 0.000 0.000 0.000 > -0.019 > 2 O S -0.493 -0.614 0.472 -0.284 -0.081 0.000 0.000 > 0.008 > Px 0.057 0.043 0.173 -0.288 0.344 0.000 0.000 > 0.543 > Pz 0.000 0.000 0.000 0.000 0.000 -0.450 0.674 > 0.000 > Py 0.114 0.064 0.269 -0.393 -0.307 0.000 0.000 > -0.402 > 3 O S -0.493 0.614 0.472 0.284 -0.081 0.000 0.000 > -0.008 > Px 0.057 -0.043 0.173 0.288 0.344 0.000 0.000 > -0.543 > Pz 0.000 0.000 0.000 0.000 0.000 -0.450 -0.674 > 0.000 > Py -0.114 0.064 -0.269 -0.393 0.307 0.000 0.000 > -0.402 > **************************************************************************** > > that is, only ONE function for each l-channel. > > So my question is: does anyone know if is it actually possible to > project the wavefunction on a double-zeta atomic basis set? And if this > is the case, could anyone please tell me what I am doing wrong or tell > me any suggestions? > > Thank you very much in advance, > > Antonio Rodriguez-Fortea > > -- > Antonio Rodriguez Fortea Phone: +41 (0)91 610 8258 > ETH Zurich/CSCS Fax: +41 (0)91 610 8282 > Via Cantonale, Galleria 2 e-mail: arodrig at cscs.ch > CH-6928 Manno (Switzerland) > > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > From Ari.P.Seitsonen at iki.fi Sun May 25 12:37:50 2003 From: Ari.P.Seitsonen at iki.fi (Ari.P.Seitsonen at iki.fi) Date: Sun, 25 May 2003 12:37:50 +0200 Subject: [CPMD-list] Problem with MD calculation! In-Reply-To: (message from Juerg Hutter on Sun, 25 May 2003 11:17:02 +0200 (MEST)) References: <20030524152615.48884.qmail@web15307.mail.bjs.yahoo.com> Message-ID: <200305251037.h4PAboX01790@magadino.cscs.ch> Dear Zhou, To complete the Boss'es answer: The problem could also be that you are using a version prior to 3.7, the call wasn't appropriate back then. Greetings from Zurich, apsi > From: Juerg Hutter > Cc: CPMD-list at cpmd.org > Sender: cpmd-list-admin at cpmd.org > List-Help: > List-Post: > List-Subscribe: , > > List-Id: the mailing list of the CPMD program > List-Unsubscribe: , > > List-Archive: > Date: Sun, 25 May 2003 11:17:02 +0200 (MEST) > > Hi > > the problem is related to the fact that there is a name > conflict between LAPACK and ESSL. Both libraries have > routines called DSPEV with different calling sequence > but performing the same task. > > In CPMD we use the LAPACK routine DSPEV (unless you have > changed the file util.F or used -DESSL as a preprocessor > option, both not recommended). > To be sure to call the correct library you have to use > the linker option > > -llapack -lessl -llapack > > This should be the default when you use the Configure script. > However, the libraries might be called differently on your system. > > Juerg Hutter > > > ---------------------------------------------------------- > 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 Sat, 24 May 2003, [gb2312] Adrain Zhou wrote: > > > Dear all, > > > > I met such a problem when I performed a MD calculation on SP3. > > > > > > DSPEV : 2538-2028 > > > > The job option argument (ARG NO. 1) must be 0, 1, 20, or 21. > > > > > > > > DSPEV : 2538-2028 > > > > The job option argument (ARG NO. 1) must be 0, 1, 20, or 21. > > > > > > > > DSPEV : 2538-2028 > > > > The job option argument (ARG NO. 1) must be 0, 1, 20, or 21. > > > > > > > > DSPEV : 2538-2099 > > > > End of input argument error reporting. For more information, refer to > > > > Engineering and Scientific Subroutine Library Guide and Reference (SA22-7272). > > > > DSPEV : 2538-2099 > > > > End of input argument error reporting. For more information, refer to > > > > Engineering and Scientific Subroutine Library Guide and Reference (SA22-7272). > > > > DSPEV : 2538-2099 > > > > End of input argument error reporting. For more information, refer to > > > > Engineering and Scientific Subroutine Library Guide and Reference (SA22-7272). > > > > > > > > Can anyone tell me what's wrong with it? > > > > > > > > Thanks in advance! > > > > With regards, > > > > Zhou > > > > > > > > > > > > > > --------------------------------- > > Do You Yahoo!? > > "????????????!??????????????????????????????????????~???????????????????????????????? > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > -- -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- Ari Paavo Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/ Tel +41 1 635 44 97 / Fax +41 1 635 68 38 / GSM +41 79 719 09 35 Anschrift: Physikalisch Chemisches Institut (PCI), Universit?t Z?rich (UniZh) Indirizzo: Winterthurerstra?e 190, CH-8057 Z?rich Address: Schweiz / Svizzera / Suisse / Svizra / Switzerland From cdy2000 at vip.sina.com Sun May 25 13:29:55 2003 From: cdy2000 at vip.sina.com (cdy2000 at vip.sina.com) Date: Sun, 25 May 2003 19:29:55 +0800 Subject: [CPMD-list] two questions about FHIPP Message-ID: <20030525112955.8789.qmail@vip.sina.com> Hello! I use FHIPP to generate the pp, other atoms can be generated but H, someone said there are some bugs in FHIPP package, I want to know how to generate the pp of H? another question, what keyword should I use in &atom ... &end, I can use the output file "*.cpi" in CPMD? could comeone give me a example? Thanks for any help. Sincerely Cao Dayong ______________________________________ =================================================================== From sttp_komin at hotmail.com Sun May 25 19:02:36 2003 From: sttp_komin at hotmail.com (sttp komin) Date: Mon, 26 May 2003 00:02:36 +0700 Subject: [CPMD-list] Time for compute NMR chemical shifts Message-ID: Dear All, I run cpmd to calculate NMR in C6H6 from input file example. The lenght time which run on linux os with athlon XP 1.33 Mhz. and ram 256 MB free disk 2 Gbyte and IFC compiler was shown that CPU TIME : 0 HOURS 32 MINUTES 5.77 SECONDS ELAPSED TIME : 0 HOURS 50 MINUTES 11.97 SECONDS when compare with the output file example shown that CPU TIME : 0 HOURS 1 MINUTES 48.95 SECONDS ELAPSED TIME : 0 HOURS 1 MINUTES 52.90 SECONDS So i would like to ask that which computer which you use to test the output file?(parallel?) and how i can improve my calculation time? and the last problem is i can not run file input example to test full calculation on NMR chemical shifts. The result is shown below. **************************************************************** * RESPONSE CALCULATIONS * **************************************************************** ********************************************************RESPONSE* * RESPONSE TYPE: NMR chemical shifts * * PRECONDITIONER TYPE: smooth (ds) * * PRECONDITIONER CUTOFF: 0.50000 * * WFN CONVERGENCE: 0.100E-04 * * GND STATE WFN: reciprocal space only * * ALGORITHM: preconditioned conj gradient * * Fletcher-Reeves variant * * CG: ANALYTIC STEPS: 3 * * CG: LENGTH PREFACTOR: 0.60000 * ********************************************************RESPONSE* *** DO_PERTU| THE NEW SIZE OF THE PROGRAM IS 170996 kBYTES *** **************************************************************** PROCESSOR 0 ALLOCATION OF 51260040 WORDS OF MEMORY FAILED **************************************************************** *** MEMORY| THE NEW SIZE OF THE PROGRAM IS 171060 kBYTES *** ================================================================ BIG MEMORY ALLOCATIONS scr 12974052 PME 4746380 psi 2162342 YF 2162342 PSI 2162342 XF 2162342 CATOM 1898520 SCR 1587897 GDE 1186595 rhoe 1081171 ---------------------------------------------------------------- [PEAK NUMBER 80] PEAK MEMORY 29294612 = 234.4 MBytes ================================================================ STOPGM! STACK OF MAIN CALLS: STOPGM! CALL NMR PROGRAM STOPS IN SUBROUTINE MEMORY| ALLOCATION FAILED (C1) what should i do? or any one have any suggestion. Thank you and with best wishes! Sincerely yours, Sittipong Komin Physics Dept. Science Fact. Ubonratchatani University 34190 Thailand _________________________________________________________________ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/featuredemail From axel.kohlmeyer at theochem.ruhr-uni-bochum.de Mon May 26 10:27:24 2003 From: axel.kohlmeyer at theochem.ruhr-uni-bochum.de (Axel Kohlmeyer) Date: Mon, 26 May 2003 10:27:24 +0200 Subject: [CPMD-list] Time for compute NMR chemical shifts In-Reply-To: Your message of "Mon, 26 May 2003 00:02:36 +0700." Message-ID: <200305260827.h4Q8ROc01755@yello.theochem.ruhr-uni-bochum.de> >>> "ST" == sttp komin writes: ST> Dear All, ST> I run cpmd to calculate NMR in C6H6 from input file example. ST> The lenght time which run on linux os with athlon XP 1.33 Mhz. and ram 256 ST> MB free disk 2 Gbyte and IFC compiler was shown that ST> CPU TIME : 0 HOURS 32 MINUTES 5.77 SECONDS ST> ELAPSED TIME : 0 HOURS 50 MINUTES 11.97 SECONDS ST> when compare with the output file example shown that ST> CPU TIME : 0 HOURS 1 MINUTES 48.95 SECONDS ST> ELAPSED TIME : 0 HOURS 1 MINUTES 52.90 SECONDS ST> So i would like to ask that which computer which you use to test the ST> output file?(parallel?) and how i can improve ST> my calculation time? hi. the example run _is_ a parallel run, as you can see for yourself if you look for the string 'PARAPARAPARA' in the output file. the following section shows how the job is distributed over the parallel nodes. seems to be a 32-cpu parallel run, btw. given the size of the box and the cutoff in the example the job will have needed more memory than the ~200MB available on your machine and i assume the machine has been swapping happily during the run. see the difference between cpu time and elapsed time. of course i assume you have not run anything else on the machine at the same time. so to improve (a bit) you need more memory. if you have not done so already, you should use an optimized LAPACK/BLAS library. this can be either build from the ATLAS package or you can try the intel math kernel library (pentium-III) version. an athlon-xp optimized binary of a combined LAPACK/BLAS/ATLAS library can be found at: to get into the neighborhood of 2 minutes you either need a large multiprocessor smp-box (16 cpu or more) or a cluster of pcs with very fast interconnect (dolphin/SCI or myrinet). ST> and the last problem is i can not run file input example to test full ST> calculation on NMR chemical shifts. ST> The result is shown below. ST> **************************************************************** ST> * RESPONSE CALCULATIONS * ST> **************************************************************** ST> ********************************************************RESPONSE* ST> * RESPONSE TYPE: NMR chemical shifts * ST> * PRECONDITIONER TYPE: smooth (ds) * ST> * PRECONDITIONER CUTOFF: 0.50000 * ST> * WFN CONVERGENCE: 0.100E-04 * ST> * GND STATE WFN: reciprocal space only * ST> * ALGORITHM: preconditioned conj gradient * ST> * Fletcher-Reeves variant * ST> * CG: ANALYTIC STEPS: 3 * ST> * CG: LENGTH PREFACTOR: 0.60000 * ST> ********************************************************RESPONSE* ST> *** DO_PERTU| THE NEW SIZE OF THE PROGRAM IS 170996 kBYTES *** ST> **************************************************************** ST> PROCESSOR 0 ALLOCATION OF 51260040 WORDS OF MEMORY FAILED ST> **************************************************************** ST> *** MEMORY| THE NEW SIZE OF THE PROGRAM IS 171060 kBYTES *** ST> ================================================================ ST> BIG MEMORY ALLOCATIONS ST> scr 12974052 PME 4746380 ST> psi 2162342 YF 2162342 ST> PSI 2162342 XF 2162342 ST> CATOM 1898520 SCR 1587897 ST> GDE 1186595 rhoe 1081171 ST> ---------------------------------------------------------------- ST> [PEAK NUMBER 80] PEAK MEMORY 29294612 = 234.4 MBytes ST> ================================================================ ST> STOPGM! STACK OF MAIN CALLS: ST> STOPGM! CALL NMR ST> PROGRAM STOPS IN SUBROUTINE MEMORY| ALLOCATION FAILED (C1) ST> what should i do? or any one have any suggestion. well, the error message says it: you need more memory! the program tried to request 51260040 words (= ~400MB) and the request failed. regards, axel kohlmeyer. ST> Thank you and with best wishes! ST> Sincerely yours, ST> Sittipong Komin ST> Physics Dept. ST> Science Fact. ST> Ubonratchatani University ST> 34190 ST> Thailand -- ======================================================================= Dr. Axel Kohlmeyer e-mail: axel.kohlmeyer at theochem.ruhr-uni-bochum.de Lehrstuhl fuer Theoretische Chemie Phone: ++49 (0)234/32-26673 Ruhr-Universitaet Bochum - NC 03/53 Fax: ++49 (0)234/32-14045 D-44780 Bochum http://www.theochem.ruhr-uni-bochum.de ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From tdeutsch at cea.fr Tue May 27 13:58:33 2003 From: tdeutsch at cea.fr (Thierry Deutsch) Date: Tue, 27 May 2003 13:58:33 +0200 Subject: [CPMD-list] ovlap2_c Message-ID: <200305271358.33080.tdeutsch@cea.fr> Hello, For an aluminum interface (174 atoms, cutoff of 10Ry 310 states, free energy functional), I have the following behavior on a Fujitsu VPP 5000: SUBROUTINE CALLS CPU TIME ELAPSED TIME OVLAP2_C 9060 30066.36 30067.66 EHPSI_C 8457 9696.90 9740.52 FFT-G/S 3038922 4757.54 4842.34 FRIESNER_C 240 4010.71 4021.63 S_FWFFT 725161 3545.42 3609.48 S_INVFFT 794300 3022.41 3091.97 OVLAP_H 881 2686.38 2686.41 CALC_BILN 8457 2379.40 2379.68 EVPSI 515741 1011.01 1033.20 RGS_C 639 638.15 638.18 VBETA 240 526.16 538.59 You can see that ovalp2_C is used a lot. In this routine ZGEMM is used. I think it is not possible to optimize this routine. May I wrong? Au revoir, Thierry From carlospi at estudiantes.fisica.unal.edu.co Wed May 28 18:09:51 2003 From: carlospi at estudiantes.fisica.unal.edu.co (Pinilla Castellanos Carlos) Date: Wed, 28 May 2003 11:09:51 -0500 (COT) Subject: [CPMD-list] question Message-ID: Hi all I have one question about theory of CPMD: Not understanding what is GNMAX, GNORM and CNSTR in geometry optimization. I find information about its? Thanks for your help Carlos From aijun_du at hotmail.com Thu May 29 11:03:58 2003 From: aijun_du at hotmail.com (Du Aijun) Date: Thu, 29 May 2003 17:03:58 +0800 Subject: [CPMD-list] parameter for VDW calculation! Message-ID: Dear all, I want to perform a calculation on a system involving Benzene molecule based on the VDW correction. But I can't find any parameter from the reference (J.Phys.Chem, 88(1984)4272 and J.Chem.Phys.,86(1987)1485) cited by CPMD manual. It seems the form is not the same as that in the manual. Does anyeone have some experiences on it? If so, Can you tell me how to use it and the relative parameters for hydrocarbon molecule? Many thanks in advance! With regards, Aijun Du _________________________________________________________________ ???????????????????????????? MSN Messenger: http://messenger.msn.com/cn From francesco.filippone at ism.cnr.it Tue May 20 14:10:08 2003 From: francesco.filippone at ism.cnr.it (Francesco Filippone) Date: Tue, 20 May 2003 14:10:08 +0200 (CEST) Subject: [CPMD-list] Thinking about buying an AMD Opteron system... In-Reply-To: <200305192010.h4JKA5B15622@yello.theochem.ruhr-uni-bochum.de> Message-ID: On Mon, 19 May 2003, Axel Kohlmeyer wrote: > > > importantly, I am hoping to find out for sure if an Opteron based machine will > > be able to allocate more than 1 Gb of memory to a single process. If I do > > make the purchase, I would use the 64 bit version of Mandrake Linux 9.0 and > > the 64 bit version of the PGI compiler. > > well, you also need a very well optimized LAPACK/BLAS implementation. > e.g. for the amd athlon platform this can mean a 2x speedup for CPMD runs > (up to 8x for DGEMM). AFAIK, there is no such optimized LAPACK/BLAS > available (yet). you should check the ATLAS homepage > to be sure. > Dear Mark & Axel, only a note. There is a library developed by AMD, AMD Core Math Library (ACML), that should be the analogous of Intel MKL. I just checked for its existence, since I've not an AMD box yet, but I'm going to buy one. The library is designed for Opteron processors and incorporates support for BLAS, LAPACK and FFT routines. Seems to be free, but I've not downloaded it yet. Maybe it's worth to be downloaded and tested by those of the CPMD community who have already an AMD box available. They don't mention any possibility of using the library with 32-bit processors, but they give a 32-bit mode library. The link for the ACML page is http://www.developwithamd.com/acml/welcome/ Hope this helps, ciao, F. -- ---------------------------------------------------------------------------- Francesco Filippone | Tanto gentile e CNR - ISTITUTO DI STRUTTURA DELLA MATERIA | tant'onesta pare v. Salaria Km 29,300 - C.P. 10 | la donna mia quand'ella I 00016 - Monterotondo Stazione (RM) | altrui saluta... Tel + 39 06 90672342 - Fax +39 06 90672316 | e-mail: | Dante ---------------------------------------------------------------------------- From luca at duke.dichi.unina.it Tue May 27 15:21:13 2003 From: luca at duke.dichi.unina.it (Luca Minieri) Date: Tue, 27 May 2003 13:21:13 +0000 (GMT) Subject: [CPMD-list] CPMD and itanium Message-ID: Hi, I am trying to compile 3.5.4 on a itanium. I used ITANIUM-EFC from Configure, and it seems that cpmd is compiled properly. However, when I run it I ahve a lot of cpmd.x(8896): unaligned access to 0x6000000000477aa8, ip=0x40000000000fd2f0 The input is read in correctly, and the very first iteration (WF opt) is performed properly (compared to the out on a different and working arch). From veronique.vanspeybroeck at ugent.be Wed May 28 10:28:32 2003 From: veronique.vanspeybroeck at ugent.be (Veronique Van Speybroeck) Date: Wed, 28 May 2003 10:28:32 +0200 Subject: [CPMD-list] MEMORY ALLOCATION Message-ID: <3ED47330.1040803@ugent.be> Dear cpmd users, I have a problem allocating memory. The input file of the job is given below and is run on a IBM dual processor machine with two Xeon processors, which has 4 GB RAM memory. I get the following output message : *** WFOPTS| THE NEW SIZE OF THE PROGRAM IS 137968 kBYTES *** *** PHFAC| THE NEW SIZE OF THE PROGRAM IS 138160 kBYTES *** **************************************************************** PROCESSOR 0 ALLOCATION OF 395928 WORDS OF MEMORY FAILED **************************************************************** *** MEMORY| THE NEW SIZE OF THE PROGRAM IS 184032 kBYTES *** ================================================================ BIG MEMORY ALLOCATIONS PME 55100060 GDE 13775015 C0 11020004 C2 11020004 SCR 7018210 YLMB 6600250 EIGR 5740956 QRL 5421600 QRAD 3589920 TWNL 3563352 ---------------------------------------------------------------- [PEAK NUMBER 108] PEAK MEMORY 135398590 = 1083.2 MBytes ================================================================ PROGRAM STOPS IN SUBROUTINE MEMORY| ALLOCATION FAILED (CX) [PROC= 0] I tested both with the parallel and unparallel version of the CPMD code. How can I solve this problem? Has it something to do with compilation flags? Thanks in advance veronique inputfile : &CPMD OPTIMIZE WAVEFUNCTION TIMESTEP 20 PRINT 10 ISOLATED MOLECULE STRUCTURE BONDS ANGLES &END &SYSTEM ANGSTROM CHARGE +1 SYMMETRY 1 CELL 21.0 1.0 1.0 0.0 0.0 0.0 CUTOFF 25. &END &ATOMS *045-Rh-gpw-n-campos.uspp BINARY NEWF LMAX=D 1 -0.377245 -0.101850 0.178396 *015-P-ca--vgrp.uspp BINARY NEWF MAX=P 4 -0.377245 -0.101850 2.474729 0.620693 1.744877 -0.468820 0.641089 -2.029866 -0.507751 -2.513948 -0.169410 -0.625693 *007-N-ca--bm.uspp BINARY NEWF LMAX=P 4 1.014140 2.235974 -2.153407 0.383196 -3.566781 0.408488 .... -- ----------------------------------------------------------------------- Dr. ir. Van Speybroeck Veronique Laboratorium voor Theoretische Fysica Universiteit Gent Proeftuinstraat 86 9000 Gent Tel +32-9-264.65.58 GSM : +32/474/259767 Fax +32-9-264.65.60 email : veronique.vanspeybroeck at rug.ac.be http://inwfaxp2.rug.ac.be/~web/onderzoek/proj_3.html ----------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://cpmd.org/pipermail/cpmd-list/attachments/20030528/c00a8492/attachment.html From massimo.celino at casaccia.enea.it Thu May 29 18:20:00 2003 From: massimo.celino at casaccia.enea.it (Massimo Celino) Date: Thu, 29 May 2003 18:20:00 +0200 Subject: [CPMD-list] SSM_CSS_INIT error on IBM computers Message-ID: <4.3.2.7.0.20030529175457.00b25f00@casaccia.enea.it> Dear all, I compiled cpmd 3.7 on IBM SP3 with 16 processors shared memory (aix 5.1, xlf 8.1, poe 3.2). No problems running with 4 and 5 processors. On the contrary when I allocate 6 processors or more the job does not start and the error is : ERROR: 0031-332 SSM_CSS_INIT expected and it was not received. I find the same error when I compile with both IBM-SP3 and IBM-SP3-SMP options of the Configure script. Can anyone tell me what is wrong in my compilation or in our parallel environment ?? Thank you Massimo Celino -------------------------------------------------------------- Massimo Celino ENEA - CAMO C.R. Casaccia - SP103 Via Anguillarese 301, 00060 S.Maria di Galeria (Roma), Italy Tel.: +39-06-30483871 Fax: +39-06-30484230 From Ari.P.Seitsonen at iki.fi Thu May 29 18:37:37 2003 From: Ari.P.Seitsonen at iki.fi (Ari.P.Seitsonen at iki.fi) Date: Thu, 29 May 2003 18:37:37 +0200 Subject: [CPMD-list] please Re to the top 10 :-) In-Reply-To: <20030521114643.4440.qmail@vip.sina.com> (cdy2000@vip.sina.com) References: <20030521114643.4440.qmail@vip.sina.com> Message-ID: <200305291637.h4TGbb630693@magadino.cscs.ch> Dear Cao, > Hello,every CPMDer! > > send a mail to support our mail-list (ps re my mail to the top 10) :-) > > Two questions: > 1. > I use the fhi98pp to generate the pp, use "./psgen -o atomname > atomname:h.ini", other atom can be generated but hydrogen, the info > is > segmentation fault > rm: No match > foreach: No match > I find someone ask the same question in mail-list last year, but the > answer was not clear. > > 2. > the programme generates a set of files, which should I use in CPMD? > only *.cpi, or all of? how to denote in &ATOMS ...&END??use *PPNAME > KLEINMAN-BYLANDER, or add some other keywords? > > Thx for any help! > > ______________________________________ > Hello! > > I use FHIPP to generate the pp, other atoms can be generated but H, > someone said there are some bugs in FHIPP package, I want to know > how to generate the pp of H? > > another question, what keyword should I use in &atom ... &end, I can > use the output file "*.cpi" in CPMD? could comeone give me a > example? > > Thanks for any help. > > Sincerely Cao Dayong > ______________________________________ Could you please send me the input file which you used for the FHIPP package? Second, you should use the *.pp file generated by 'psgen' (I presume that you have the enhanced version of 'psgen', please see e.g. http://www.cpmd.org/pipermail/cpmd-list/2002-February/000025.html) Greetings from Zurich, apsi -- -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- Ari Paavo Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/ Tel +41 1 635 44 97 / Fax +41 1 635 68 38 / GSM +41 79 719 09 35 Anschrift: Physikalisch Chemisches Institut (PCI), Universit?t Z?rich (UniZh) Indirizzo: Winterthurerstra?e 190, CH-8057 Z?rich Address: Schweiz / Svizzera / Suisse / Svizra / Switzerland From Ari.P.Seitsonen at iki.fi Thu May 29 18:40:25 2003 From: Ari.P.Seitsonen at iki.fi (Ari.P.Seitsonen at iki.fi) Date: Thu, 29 May 2003 18:40:25 +0200 Subject: [CPMD-list] Pseudo for Ni or Pd In-Reply-To: (message from Simone Meloni on Mon, 12 May 2003 17:11:47 +0000 (UTC)) References: Message-ID: <200305291640.h4TGePH30726@magadino.cscs.ch> Grueziwohl Simone & Galina & tutti, Have you already obtained the potentials (or instructions how to generate them)? For Ni I would recommend VDB pseudo (unless impossible with the present version of CPMD, e.g. FEMD/k points). For Pd a Troullier-Martins should do fine. Saluti da Zurigo, apsi > From: Simone Meloni > List-Archive: > Date: Mon, 12 May 2003 17:11:47 +0000 (UTC) > > Dear CPMDer, > does anybody have a working potential for either Ni or Pd? > we would appreciate help. > Ciao > > Simone & Galina > -- > +------------------------------------------------------------------------+ > PhD Simone Meloni | Achab: Lascio una scia Bianca e inquieta, > CASPUR | acque pallide, facce piu' pallide > Via Dei Tizii 6b | dovunque passo. Le onde invidiose > 00185 Rome (Italy) | si gonfiano ai lati per sommergere > Phone: +39-06-44486-711 | la mia traccia: facciano, ma prima > Fax: +39-06-4957083 | io passo. > > > > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > -- -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- Ari Paavo Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/ Tel +41 1 635 44 97 / Fax +41 1 635 68 38 / GSM +41 79 719 09 35 Anschrift: Physikalisch Chemisches Institut (PCI), Universit?t Z?rich (UniZh) Indirizzo: Winterthurerstra?e 190, CH-8057 Z?rich Address: Schweiz / Svizzera / Suisse / Svizra / Switzerland From Ari.P.Seitsonen at iki.fi Thu May 29 19:14:48 2003 From: Ari.P.Seitsonen at iki.fi (Ari.P.Seitsonen at iki.fi) Date: Thu, 29 May 2003 19:14:48 +0200 Subject: [CPMD-list] question In-Reply-To: (message from Pinilla Castellanos Carlos on Wed, 28 May 2003 11:09:51 -0500 (COT)) References: Message-ID: <200305291714.h4THEmZ31007@magadino.cscs.ch> Dear Carlos, > I have one question about theory of CPMD: > > Not understanding what is GNMAX, GNORM and CNSTR in geometry optimization. > I find information about its? Reading the code a bit (gsize.F,cnstfc.F): GNMAX = max_{I,a} { F_Ia } = largest absolute component of force on any atom GNORM = Sum_I (F_Ia)^2 / Sum_I = average force on the ions CNSTR = max_{I,a} { F^constr_Ia } = largest absolute component of force due to constraints on any atom Greetings, apsi PS The constraints during GEOMETRY OPTIMISATION are not reliable anyway, if someone wants to optimise geometries with constraints please consider doing molecular dynamics annealing the ions (and electrons), with already a low initial temperature -- -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- Ari Paavo Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/ Tel +41 1 635 44 97 / Fax +41 1 635 68 38 / GSM +41 79 719 09 35 Anschrift: Physikalisch Chemisches Institut (PCI), Universit?t Z?rich (UniZh) Indirizzo: Winterthurerstra?e 190, CH-8057 Z?rich Address: Schweiz / Svizzera / Suisse / Svizra / Switzerland From hutter at pci.unizh.ch Fri May 30 10:50:45 2003 From: hutter at pci.unizh.ch (Juerg Hutter) Date: Fri, 30 May 2003 10:50:45 +0200 (MEST) Subject: [CPMD-list] ovlap2_c In-Reply-To: <200305271358.33080.tdeutsch@cea.fr> References: <200305271358.33080.tdeutsch@cea.fr> Message-ID: Hi Thierry usually xGEMM routines perform at the highest possible rate on all computers. However, there might be the following problems. A bad implementation of ZGEMM on this computer. You could test this by comparing to the DGEMM performance. The other possibility is that in the FEMD code you call ZGEMM very often with small dimensions. The VPP 5000 is a vector machine that might not achieve peak performance in this case. A possible solution in this case would be to add special code for N=1,2,... . There is already a special code for the case of Matrix*Vector multiplication instead of Matrix*Matrix (N=1). greetings 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 Tue, 27 May 2003, Thierry Deutsch wrote: > Hello, > > For an aluminum interface (174 atoms, cutoff of 10Ry 310 states, free energy > functional), I have the following behavior on a Fujitsu VPP 5000: > > SUBROUTINE CALLS CPU TIME ELAPSED TIME > OVLAP2_C 9060 30066.36 30067.66 > EHPSI_C 8457 9696.90 9740.52 > FFT-G/S 3038922 4757.54 4842.34 > FRIESNER_C 240 4010.71 4021.63 > S_FWFFT 725161 3545.42 3609.48 > S_INVFFT 794300 3022.41 3091.97 > OVLAP_H 881 2686.38 2686.41 > CALC_BILN 8457 2379.40 2379.68 > EVPSI 515741 1011.01 1033.20 > RGS_C 639 638.15 638.18 > VBETA 240 526.16 538.59 > > You can see that ovalp2_C is used a lot. > In this routine ZGEMM is used. I think it is not possible to optimize this > routine. > May I wrong? > > Au revoir, > > Thierry > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > From jarek at elrond.chem.uni.wroc.pl Fri May 30 11:41:17 2003 From: jarek at elrond.chem.uni.wroc.pl (Jaroslaw Panek) Date: Fri, 30 May 2003 11:41:17 +0200 (CEST) Subject: [CPMD-list] negative USPP augmentation charges Message-ID: Dear CPMD'ers, while generating and testing various Vanderbilt pseudopotentials in CPMD3.7, we have encountered (very rarely) negative values of mean augmentation charges on halogen atoms (ranging from -0.2 to -0.5). While this does not seem to me to contradict rules along which USPPs are generated, and moreover uspp7.3.4 diagnostics seem all OK, I would like to know your opinion if this behaviour can affect some parts of CPMD calculations? We have observed no abnormalities, but are there certain functionals or other parts of code that can misbehave encountering negative USPP augmentation charges? With greetings from sunny Wroclaw, Jaroslaw Panek Faculty of Chemistry, University of Wroclaw ul. F. Joliot-Curie 14, 50-383 Wroclaw, Poland