From cdy2000 at vip.sina.com Mon Nov 3 02:58:04 2003 From: cdy2000 at vip.sina.com (cdy2000 at vip.sina.com) Date: Mon, 03 Nov 2003 09:58:04 +0800 Subject: [CPMD-list] about VIBRATIONAL ANALYSIS Message-ID: <20031103015804.8323.qmail@vip.sina.com.cn> Hello, everyone! I have used the keyword VIBRATIONAL ANALYSIS LR and get some output files, such as FINDIF, MOLVIB, VIBEIGVEC, lots of data. Who could tell me the meaning of these files and data in detail? Thanks for any help. Sincerely Cao Dayong ______________________________________ =================================================================== From phsorkin at techunix.technion.ac.il Mon Nov 3 10:37:23 2003 From: phsorkin at techunix.technion.ac.il (Sorkin Slava) Date: Mon, 3 Nov 2003 11:37:23 +0200 Subject: [CPMD-list] helium Message-ID: <20031103093723.7A9E1882B6@phjoan3.technion.ac.il> Hello, everyone! I wander if anybody know where one can find pseudopotential for helium (solid 4He)? Thanks for any help. Sincerely , S.Sorkin From J.C.Wojdel at tnw.tudelft.nl Mon Nov 3 12:17:29 2003 From: J.C.Wojdel at tnw.tudelft.nl (Jacek Wojdel) Date: Mon, 3 Nov 2003 12:17:29 +0100 Subject: [CPMD-list] CPMD having problems with electronic spectra. Message-ID: <20031103121729.C13107@dct2918.dct.tudelft.nl> Dear CPMD users, Has anyone experienced problems with calculating electronic spectra for transition metal complexes? I'm using LSD and Vanderbilt pseudopotentials. When presented with the following input: &CPMD ELECTRONIC SPECTRA RESTART WAVEFUNCTION COORDINATES LATEST COMPRESS WRITE32 LSD MIRROR &END &SYSTEM SYMMETRY 2 ANGSTROM SCALE CARTESIAN CELL ABSOLUTE DEGREE 10.147408880 10.147408880 10.147408880 90.00 90.00 90.00 CUTOFF 40.0 MULTIPLICITY 2 &END &ATOMS *Co3_ps.uspp BINARY NEWF LMAX=2 1 0.00 0.00 0.00 *Fetu3_ps.uspp BINARY NEWF LMAX=2 1 0.50 0.00 0.00 *C_ps.uspp BINARY NEWF LMAX=2 6 0.1807169772 0.0000000000 0.0000000000 -0.1807169772 0.0000000000 0.0000000000 0.0000000000 0.1807169772 0.0000000000 0.0000000000 -0.1807169772 0.0000000000 0.0000000000 0.0000000000 0.1807169772 0.0000000000 0.0000000000 -0.1807169772 *N_ps.uspp BINARY NEWF LMAX=2 6 0.2970532813 0.0000000000 0.0000000000 -0.2970532813 0.0000000000 0.0000000000 0.0000000000 0.2970532813 0.0000000000 0.0000000000 -0.2970532813 0.0000000000 0.0000000000 0.0000000000 0.2970532813 0.0000000000 0.0000000000 -0.2970532813 &END &TDDFT &END &DFT FUNCTIONAL PW91 GC-CUTOFF 1.D-5 &END CPMD dies with "Address error" at this point of the output: NFI GEMAX CNORM ETOT DETOT TCPU 1 5.222E-06 2.217E-07 -255.584304 0.000E+00 32.97 RESTART INFORMATION WRITTEN ON FILE ./RESTART.1 TIME FOR MINIMUM STRUCTURE : 47.510 SECONDS *** SPECTRA| THE NEW SIZE OF THE PROGRAM IS 157920 kBYTES *** ================================================================ == END OF REFERENCE CALCULATION == == GENERATE INITIAL GUESS VECTORS == ================================================================ *** SPECTRA| THE NEW SIZE OF THE PROGRAM IS 157968 kBYTES *** Using Lanczos diagonalisation (as suggested in the example from the manual) does add a line "<<<< LANCZOS DIAGONALIZATION <<<<" to the output, but does not go any further. Is there something wrong in the input, or is it just a case of TDDFT not yet completely functional? Thanks in advance for any help, Jacek -- +-------------------------------------+ |from: J.C.Wojdel | | J.C.Wojdel at tnw.tudelft.nl | +-------------------------------------+ From hutter at pci.unizh.ch Mon Nov 3 14:09:48 2003 From: hutter at pci.unizh.ch (Juerg Hutter) Date: Mon, 3 Nov 2003 14:09:48 +0100 (MET) Subject: [CPMD-list] CPMD having problems with electronic spectra. In-Reply-To: <20031103121729.C13107@dct2918.dct.tudelft.nl> References: <20031103121729.C13107@dct2918.dct.tudelft.nl> Message-ID: Hi all linear response and TDDFT options in CPMD are not implemented for Vanderbilt PP. Sorry. 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 Mon, 3 Nov 2003, Jacek Wojdel wrote: > Dear CPMD users, > Has anyone experienced problems with calculating electronic spectra > for transition metal complexes? I'm using LSD and Vanderbilt > pseudopotentials. > When presented with the following input: > &CPMD > ELECTRONIC SPECTRA > RESTART WAVEFUNCTION COORDINATES LATEST > COMPRESS WRITE32 > LSD > MIRROR > &END > &SYSTEM > SYMMETRY > 2 > ANGSTROM > SCALE CARTESIAN > CELL ABSOLUTE DEGREE > 10.147408880 10.147408880 10.147408880 90.00 90.00 90.00 > CUTOFF > 40.0 > MULTIPLICITY > 2 > &END > &ATOMS > *Co3_ps.uspp BINARY NEWF > LMAX=2 > 1 > 0.00 0.00 0.00 > > *Fetu3_ps.uspp BINARY NEWF > LMAX=2 > 1 > 0.50 0.00 0.00 > > *C_ps.uspp BINARY NEWF > LMAX=2 > 6 > 0.1807169772 0.0000000000 0.0000000000 > -0.1807169772 0.0000000000 0.0000000000 > 0.0000000000 0.1807169772 0.0000000000 > 0.0000000000 -0.1807169772 0.0000000000 > 0.0000000000 0.0000000000 0.1807169772 > 0.0000000000 0.0000000000 -0.1807169772 > > *N_ps.uspp BINARY NEWF > LMAX=2 > 6 > 0.2970532813 0.0000000000 0.0000000000 > -0.2970532813 0.0000000000 0.0000000000 > 0.0000000000 0.2970532813 0.0000000000 > 0.0000000000 -0.2970532813 0.0000000000 > 0.0000000000 0.0000000000 0.2970532813 > 0.0000000000 0.0000000000 -0.2970532813 > > &END > &TDDFT > &END > &DFT > FUNCTIONAL PW91 > GC-CUTOFF > 1.D-5 > &END > > CPMD dies with "Address error" at this point of the output: > > NFI GEMAX CNORM ETOT DETOT TCPU > 1 5.222E-06 2.217E-07 -255.584304 0.000E+00 32.97 > > RESTART INFORMATION WRITTEN ON FILE ./RESTART.1 > > TIME FOR MINIMUM STRUCTURE : 47.510 SECONDS > *** SPECTRA| THE NEW SIZE OF THE PROGRAM IS 157920 kBYTES *** > ================================================================ > == END OF REFERENCE CALCULATION == > == GENERATE INITIAL GUESS VECTORS == > ================================================================ > > *** SPECTRA| THE NEW SIZE OF THE PROGRAM IS 157968 kBYTES *** > > Using Lanczos diagonalisation (as suggested in the example from the > manual) does add a line "<<<< LANCZOS DIAGONALIZATION <<<<" to the > output, but does not go any further. > Is there something wrong in the input, or is it just a case of > TDDFT not yet completely functional? > Thanks in advance for any help, > Jacek > -- > +-------------------------------------+ > |from: J.C.Wojdel | > | J.C.Wojdel at tnw.tudelft.nl | > +-------------------------------------+ > > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > From ekasciut at unity.ncsu.edu Wed Nov 5 17:49:08 2003 From: ekasciut at unity.ncsu.edu (Eliana Asciutto) Date: Wed, 05 Nov 2003 11:49:08 -0500 Subject: [CPMD-list] path integral calculation Message-ID: <3FA92A04.7030206@unity.ncsu.edu> Dear CPMD users, I'm running a 4 beads path integral calculation on a Linux Cluster 32 processors. The wavefunction optimization is ok, the calculation ended with 4 Restart files. When I try to do Molecular Dynamics, it only reads the first restart file (Restart_1.1) and obviously the program stop. This is the output file: RESTART INFORMATION READ ON FILE ./RESTART_1.1 DEGREES OF FREEDOM FOR SYSTEM: 30 QMDOF| NUMBER OF QUANTUM DEGREES OF FREEDOM FOR IP=1 : 27 QMDOF| NUMBER OF QUANTUM DEGREES OF FREEDOM FOR IP>1 : 30 *** PHFAC| THE NEW SIZE OF THE PROGRAM IS 48608 kBYTES *** p3_15275: p4_error: net_recv read: probable EOF on socket: 1 p1_18204: (54.982920) net_recv failed for fd = 3 p1_18204: p4_error: net_recv read, errno = : 104 p2_15256: (54.693325) net_recv failed for fd = 3 p2_15256: p4_error: net_recv read, errno = : 104 bm_list_18184: (55.305640) wakeup_slave: unable to interrupt slave 0 pid 18183 bm_list_18184: (55.311794) wakeup_slave: unable to interrupt slave 0 pid 18183 bm_list_18184: (55.318922) wakeup_slave: unable to interrupt slave 0 pid 18183 Broken pipe Broken pipe Broken pipe Could anyone help me with this problem? Thanks in advance, Eliana Asciutto ekasciut at unity.ncsu.edu From cdy2000 at vip.sina.com Tue Nov 11 03:14:49 2003 From: cdy2000 at vip.sina.com (cdy2000 at vip.sina.com) Date: Tue, 11 Nov 2003 10:14:49 +0800 Subject: [CPMD-list] about VIBRATIONAL ANALYSIS Message-ID: <20031111021449.10909.qmail@vip.sina.com.cn> Hello, everyone! I have used the keyword VIBRATIONAL ANALYSIS LR and get some output files, such as FINDIF, MOLVIB, VIBEIGVEC, lots of data. Who could tell me the meaning of these files and data in detail? Thanks for any help. Sincerely Cao Dayong ______________________________________ =================================================================== From hutter at pci.unizh.ch Tue Nov 11 11:38:29 2003 From: hutter at pci.unizh.ch (Juerg Hutter) Date: Tue, 11 Nov 2003 11:38:29 +0100 (MET) Subject: [CPMD-list] about VIBRATIONAL ANALYSIS In-Reply-To: <20031111021449.10909.qmail@vip.sina.com.cn> References: <20031111021449.10909.qmail@vip.sina.com.cn> Message-ID: Hi files are MOLVIB : information needed by the molvib analysis program -> no longer available VIBEIGVEC : Eigenvectors of the second derivative matrix -> these are the normal modes -> see the program for the format FINDIF : information on the individual perturbations -> needed for restarts 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 Tue, 11 Nov 2003 cdy2000 at vip.sina.com wrote: > Hello, everyone! > > I have used the keyword VIBRATIONAL ANALYSIS LR and get some output files, such as FINDIF, MOLVIB, VIBEIGVEC, lots of data. Who could tell me the meaning of these files and data in detail? > Thanks for any help. > > Sincerely Cao Dayong > > ______________________________________ > > =================================================================== > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > From arodrig at cscs.ch Thu Nov 13 11:49:05 2003 From: arodrig at cscs.ch (Antonio Rodriguez Fortea) Date: 13 Nov 2003 11:49:05 +0100 Subject: [CPMD-list] order of atoms and Vandebilt PP-s Message-ID: <1068720545.11057.21.camel@rancate.usilu.net> Dear CPMD-users, For my system,I would like to use both ultrasoft and norm-conserving PP-s. I have noticed however, that depending on where I put in the input file the atoms having norm-conserving PP-s, I obtain different results. In the following example only the sulfur atoms are norm-conserving, while the others are ultrasoft PP-s. I performed wavefunction minimization in 10 steps. In the first case, the sulfur atoms are the last ones, while in the second case the sulfur atoms are placed after the iron atoms. INITIALIZE WAVEFUNCTION MAXSTEP 10 1.) VANDERBILT AUGMENTATION CHARGES (MEAN VALUE PER ATOM) ATOM TYPE NR. OF ATOMS CHARGE Fe 12 8.698 N 1 0.285 O 3 0.653 O 17 0.672 H 34 0.040 S 4 0.000 S 20 0.000 (K+E1+L+N+X) TOTAL ENERGY = -2098.77200681 A.U. (K) KINETIC ENERGY = 394.31861310 A.U. (E1=A-S+R) ELECTROSTATIC ENERGY = -1421.97144431 A.U. (S) ESELF = 1567.17302900 A.U. (R) ESR = 15.02526143 A.U. (L) LOCAL PSEUDOPOTENTIAL ENERGY = -1125.18996136 A.U. (N) N-L PSEUDOPOTENTIAL ENERGY = 350.21600136 A.U. (X) EXCHANGE-CORRELATION ENERGY = -296.14521560 A.U. GRADIENT CORRECTION ENERGY = -11.90686486 A.U. 2.) VANDERBILT AUGMENTATION CHARGES (MEAN VALUE PER ATOM) ATOM TYPE NR. OF ATOMS CHARGE Fe 12 8.698 S 4 0.000 S 20 0.000 N 1 0.000 O 3 0.000 O 17 0.000 H 34 0.414 (K+E1+L+N+X) TOTAL ENERGY = -2098.77242167 A.U. (K) KINETIC ENERGY = 394.31776465 A.U. (E1=A-S+R) ELECTROSTATIC ENERGY = -1421.97104142 A.U. (S) ESELF = 1567.17302900 A.U. (R) ESR = 15.02526143 A.U. (L) LOCAL PSEUDOPOTENTIAL ENERGY = -1125.19160611 A.U. (N) N-L PSEUDOPOTENTIAL ENERGY = 350.21809011 A.U. (X) EXCHANGE-CORRELATION ENERGY = -296.14562889 A.U. GRADIENT CORRECTION ENERGY = -11.90688997 A.U. It seems to me, that in the second case the N and O atoms are not calculated correctly as Vanderbilt pseudos. Nevertheless it might be that the first case is not correct either. I would appreciate any comments and ideas about this problem. Andras Stirling & Antonio Rodriguez-Fortea ETH/Lugano, Switzerland -- Antonio Rodriguez Fortea Physical Chemistry Department Phone: +41 (0)91 913 8809 ETH Zurich/USI Campus Fax: Not available yet Via Giuseppe Buffi, 13 e-mail: antonio.rodriguez at ethz.ch CH-6900 Lugano (Switzerland) From proffess at mail.ru Sat Nov 15 09:29:28 2003 From: proffess at mail.ru (=?koi8-r?Q?=22?=Sergey =?koi8-r?Q?=22=20?=) Date: Sat, 15 Nov 2003 11:29:28 +0300 Subject: [CPMD-list] OpenMP support with IFC Message-ID: Dear CPM authors and users, It seems that IFC cannot compile CPMD with OpenMP support. I met: mpif77 -c -r8 -w90 -w95 -O3 -pc64 -tpp7 -w -ipo -ipo_obj -parallel -cpp -openmp -D_REENTRANT ./util.f -o ./util.o external subroutine MY_FLUSH external subroutine AZZERO C$OMP 1 DO PRIVATE(I) SHARED(II,N,A) ^ Error 22 at (37:util.f) : invalid real constant ^ Error 24 at (37:util.f) : syntax error C$OMP 1 DO PRIVATE(I) SHARED(II,N,A) ^ Error 24 at (37:util.f) : syntax error C$OMP 1 DO PRIVATE(I) SHARED(II,N,A) ^ Error 24 at (37:util.f) : syntax error C$OMP 1 DO PRIVATE(I) SHARED(II,N,A) ^ Error 24 at (37:util.f) : syntax error C$OMP 1 DO PRIVATE(I) SHARED(II,N,A) ^ Error 24 at (37:util.f) : syntax error external subroutine IAZZERO C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) ^ Error 22 at (69:util.f) : invalid real constant ^ Error 24 at (69:util.f) : syntax error C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) ^ Error 24 at (69:util.f) : syntax error C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) ^ Error 24 at (69:util.f) : syntax error C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) ^ Error 24 at (69:util.f) : syntax error C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) ^ Error 24 at (69:util.f) : syntax error external subroutine I8AZZERO C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) ^ Error 22 at (101:util.f) : invalid real constant ^ Error 24 at (101:util.f) : syntax error C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) ^ Error 24 at (101:util.f) : syntax error C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) ^ Error 24 at (101:util.f) : syntax error C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) ^ Error 24 at (101:util.f) : syntax error C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) ^ Error 24 at (101:util.f) : syntax error external function XRANF external subroutine MPRAND external subroutine UNITMX C$OMP 1 DO PRIVATE(I) ^ Error 22 at (156:util.f) : invalid real constant ^ Error 24 at (156:util.f) : syntax error C$OMP 1 DO PRIVATE(I) ^ Error 24 at (156:util.f) : syntax error external subroutine SIMPSN external subroutine ZCLEAN C$OMP 1 DO PRIVATE(I) ^ Error 22 at (202:util.f) : invalid real constant ^ Error 24 at (202:util.f) : syntax error C$OMP 1 DO PRIVATE(I) ^ Error 24 at (202:util.f) : syntax error external subroutine ICOPY C$OMP 1 DO PRIVATE(I,IAX,IBX) ^ Error 22 at (220:util.f) : invalid real constant ^ Error 24 at (220:util.f) : syntax error C$OMP 1 DO PRIVATE(I,IAX,IBX) ^ Error 24 at (220:util.f) : syntax error C$OMP 1 DO PRIVATE(I,IAX,IBX) ^ Error 24 at (220:util.f) : syntax error C$OMP 1 DO PRIVATE(I,IAX,IBX) ^ Error 24 at (220:util.f) : syntax error external subroutine SYMMA C$OMP 1 DO PRIVATE(I,J) ^ Error 22 at (242:util.f) : invalid real constant ^ Error 24 at (242:util.f) : syntax error C$OMP 1 DO PRIVATE(I,J) ^ Error 24 at (242:util.f) : syntax error C$OMP 1 DO PRIVATE(I,J) ^ Error 24 at (242:util.f) : syntax error external subroutine MATMOV C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) ^ Error 22 at (263:util.f) : invalid real constant ^ Error 24 at (263:util.f) : syntax error C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) ^ Error 24 at (263:util.f) : syntax error C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) ^ Error 24 at (263:util.f) : syntax error C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) ^ Error 24 at (263:util.f) : syntax error C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) ^ Error 24 at (263:util.f) : syntax error C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) ^ Error 24 at (263:util.f) : syntax error C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) ^ Error 24 at (263:util.f) : syntax error C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) ^ Error 24 at (263:util.f) : syntax error external subroutine CGESUM C$OMP 1 DO PRIVATE(I,J) ^ Error 22 at (295:util.f) : invalid real constant ^ Error 24 at (295:util.f) : syntax error C$OMP 1 DO PRIVATE(I,J) ^ Error 24 at (295:util.f) : syntax error C$OMP 1 DO PRIVATE(I,J) ^ Error 24 at (295:util.f) : syntax error ELSE ^ Error 394 at (301:util.f) : syntax error - misplaced executable statement C$OMP 1 DO PRIVATE(I,J) ^ Error 22 at (303:util.f) : invalid real constant ^ Error 24 at (303:util.f) : syntax error C$OMP 1 DO PRIVATE(I,J) ^ Error 24 at (303:util.f) : syntax error C$OMP 1 DO PRIVATE(I,J) ^ Error 24 at (303:util.f) : syntax error ENDIF ^ Error 394 at (309:util.f) : syntax error - misplaced executable statement external subroutine DATUM external subroutine NUMCPUS external subroutine INVMAT external subroutine INVERSEMAT external subroutine DSPEVY external subroutine ZGETMO external subroutine ZGTHR external subroutine ZSCTR external subroutine DURAND external subroutine DAXPYI external function IZAMAX external function IDAMIN external subroutine MY_GETARG external subroutine MY_GETENV external subroutine MY_GETLOG external subroutine ZCLEAN_K external subroutine SETKWF external subroutine FSKIP external function DGIVE external function ZGIVE external subroutine ROUNDOFF external subroutine DEBROT external subroutine STORE external subroutine DETERM external function NXXFUN external function SBES0 external function FC4 external function SB1 external function SB2 external function SB3 external function SB4 external function SB5 52 Errors compilation aborted for ./util.f (code 1) make: *** [util.o] Error 1 How to solve it? Thanks, Sergey From axel.kohlmeyer at theochem.ruhr-uni-bochum.de Sat Nov 15 18:08:40 2003 From: axel.kohlmeyer at theochem.ruhr-uni-bochum.de (Axel Kohlmeyer) Date: Sat, 15 Nov 2003 18:08:40 +0100 Subject: [CPMD-list] OpenMP support with IFC In-Reply-To: Message-ID: <200311151708.hAFH8ec20274@yello.theochem.ruhr-uni-bochum.de> On Sat, 15 Nov 2003 11:29:28 +0300 =?koi8-r?Q?=22?=Sergey =?koi8-r?Q?=22=20?= wrote: > Dear CPM authors and users, > > It seems that IFC cannot compile CPMD with OpenMP support. I met: hi sergey, for this kind of problems, it usually helps A LOT, if you also would state which version of CPMD and which version of the intel compiler you are using. luckily in this specific case the problem has not much to do with the compiler (or the cpmd version). > > > mpif77 -c -r8 -w90 -w95 -O3 -pc64 -tpp7 -w -ipo -ipo_obj -parallel -cpp -openmp -D_REENTRANT ./util.f -o ./util.o > external subroutine MY_FLUSH > external subroutine AZZERO > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,A) > ^ ^^^^^^ here should be a PARALLEL (compare with the file util.F) this is due to the preprocessing stage of the cpmd compile. since you need to define PARALLEL if you want to build for MPI, _every_ occurrance of 'PARALLEL' is replaced with a '1'. just replace all 'C$OMP PARALLEL' statements with 'C$OMP parallel' and it should compile again. unfortunately, i have some bad news for you: i tried various versions of the intel compilers (and cpmd), but so far the cpmd executable did not work (address error). only with the PGI compiler i got a working OpenMP/MPI executable. and to add insult to injury, i had to find out, that these executables were in most cases on pentium 4 xeon cpus even slower than not using the second cpu at all. dual amd athlons were a little faster, but not much. i can't wait to test this on a cluster of SMP-opteron machines though, should be a real screamer. if you get it to work, and get some speedup i (and probably many more on the mailing list) would really like to hear from you. there are quite a few of us using pentium-4 xeon machines and this could get them a serious performance boost. good luck, axel. > Error 22 at (37:util.f) : invalid real constant > ^ > Error 24 at (37:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,A) > ^ > Error 24 at (37:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,A) > ^ > Error 24 at (37:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,A) > ^ > Error 24 at (37:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,A) > ^ > Error 24 at (37:util.f) : syntax error > external subroutine IAZZERO > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) > ^ > Error 22 at (69:util.f) : invalid real constant > ^ > Error 24 at (69:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) > ^ > Error 24 at (69:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) > ^ > Error 24 at (69:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) > ^ > Error 24 at (69:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) > ^ > Error 24 at (69:util.f) : syntax error > external subroutine I8AZZERO > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) > ^ > Error 22 at (101:util.f) : invalid real constant > ^ > Error 24 at (101:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) > ^ > Error 24 at (101:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) > ^ > Error 24 at (101:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) > ^ > Error 24 at (101:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) > ^ > Error 24 at (101:util.f) : syntax error > external function XRANF > external subroutine MPRAND > external subroutine UNITMX > > C$OMP 1 DO PRIVATE(I) > ^ > Error 22 at (156:util.f) : invalid real constant > ^ > Error 24 at (156:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I) > ^ > Error 24 at (156:util.f) : syntax error > external subroutine SIMPSN > external subroutine ZCLEAN > > C$OMP 1 DO PRIVATE(I) > ^ > Error 22 at (202:util.f) : invalid real constant > ^ > Error 24 at (202:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I) > ^ > Error 24 at (202:util.f) : syntax error > external subroutine ICOPY > > C$OMP 1 DO PRIVATE(I,IAX,IBX) > ^ > Error 22 at (220:util.f) : invalid real constant > ^ > Error 24 at (220:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I,IAX,IBX) > ^ > Error 24 at (220:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I,IAX,IBX) > ^ > Error 24 at (220:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I,IAX,IBX) > ^ > Error 24 at (220:util.f) : syntax error > external subroutine SYMMA > > C$OMP 1 DO PRIVATE(I,J) > ^ > Error 22 at (242:util.f) : invalid real constant > ^ > Error 24 at (242:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I,J) > ^ > Error 24 at (242:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I,J) > ^ > Error 24 at (242:util.f) : syntax error > external subroutine MATMOV > > C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) > ^ > Error 22 at (263:util.f) : invalid real constant > ^ > Error 24 at (263:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) > ^ > Error 24 at (263:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) > ^ > Error 24 at (263:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) > ^ > Error 24 at (263:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) > ^ > Error 24 at (263:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) > ^ > Error 24 at (263:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) > ^ > Error 24 at (263:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) > ^ > Error 24 at (263:util.f) : syntax error > external subroutine CGESUM > > C$OMP 1 DO PRIVATE(I,J) > ^ > Error 22 at (295:util.f) : invalid real constant > ^ > Error 24 at (295:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I,J) > ^ > Error 24 at (295:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I,J) > ^ > Error 24 at (295:util.f) : syntax error > > ELSE > ^ > Error 394 at (301:util.f) : syntax error - misplaced executable statement > > C$OMP 1 DO PRIVATE(I,J) > ^ > Error 22 at (303:util.f) : invalid real constant > ^ > Error 24 at (303:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I,J) > ^ > Error 24 at (303:util.f) : syntax error > > C$OMP 1 DO PRIVATE(I,J) > ^ > Error 24 at (303:util.f) : syntax error > > ENDIF > ^ > Error 394 at (309:util.f) : syntax error - misplaced executable statement > external subroutine DATUM > external subroutine NUMCPUS > external subroutine INVMAT > external subroutine INVERSEMAT > external subroutine DSPEVY > external subroutine ZGETMO > external subroutine ZGTHR > external subroutine ZSCTR > external subroutine DURAND > external subroutine DAXPYI > external function IZAMAX > external function IDAMIN > external subroutine MY_GETARG > external subroutine MY_GETENV > external subroutine MY_GETLOG > external subroutine ZCLEAN_K > external subroutine SETKWF > external subroutine FSKIP > external function DGIVE > external function ZGIVE > external subroutine ROUNDOFF > external subroutine DEBROT > external subroutine STORE > external subroutine DETERM > external function NXXFUN > external function SBES0 > external function FC4 > external function SB1 > external function SB2 > external function SB3 > external function SB4 > external function SB5 > > 52 Errors > compilation aborted for ./util.f (code 1) > make: *** [util.o] Error 1 > > How to solve it? > Thanks, > Sergey > > _______________________________________________ > 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 hutter at pci.unizh.ch Sat Nov 15 19:15:25 2003 From: hutter at pci.unizh.ch (Juerg Hutter) Date: Sat, 15 Nov 2003 19:15:25 +0100 (MET) Subject: [CPMD-list] OpenMP support with IFC In-Reply-To: <200311151708.hAFH8ec20274@yello.theochem.ruhr-uni-bochum.de> References: <200311151708.hAFH8ec20274@yello.theochem.ruhr-uni-bochum.de> Message-ID: Hi changing the preprocessor flags from -DPARALLEL to -DPARALLEL=PARALLEL should also help. 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 Sat, 15 Nov 2003, Axel Kohlmeyer wrote: > On Sat, 15 Nov 2003 11:29:28 +0300 =?koi8-r?Q?=22?=Sergey =?koi8-r?Q?=22=20?= wrote: > > > Dear CPM authors and users, > > > > It seems that IFC cannot compile CPMD with OpenMP support. I met: > > hi sergey, > > for this kind of problems, it usually helps A LOT, if you also would > state which version of CPMD and which version of the intel compiler > you are using. luckily in this specific case the problem has not > much to do with the compiler (or the cpmd version). > > > > > > > mpif77 -c -r8 -w90 -w95 -O3 -pc64 -tpp7 -w -ipo -ipo_obj -parallel -cpp > -openmp -D_REENTRANT ./util.f -o ./util.o > > external subroutine MY_FLUSH > > external subroutine AZZERO > > > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,A) > > ^ > ^^^^^^ here should be a PARALLEL (compare with the file util.F) > > this is due to the preprocessing stage of the cpmd compile. since you > need to define PARALLEL if you want to build for MPI, _every_ occurrance > of 'PARALLEL' is replaced with a '1'. just replace all 'C$OMP PARALLEL' > statements with 'C$OMP parallel' and it should compile again. > > unfortunately, i have some bad news for you: i tried various versions > of the intel compilers (and cpmd), but so far the cpmd executable did not > work (address error). only with the PGI compiler i got a working OpenMP/MPI > executable. and to add insult to injury, i had to find out, that these > executables were in most cases on pentium 4 xeon cpus even slower than > not using the second cpu at all. dual amd athlons were a little faster, > but not much. i can't wait to test this on a cluster of SMP-opteron > machines though, should be a real screamer. > > if you get it to work, and get some speedup i (and probably > many more on the mailing list) would really like to hear from you. > there are quite a few of us using pentium-4 xeon machines and this > could get them a serious performance boost. > > good luck, > axel. > > > > Error 22 at (37:util.f) : invalid real constant > > ^ > > Error 24 at (37:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,A) > > ^ > > Error 24 at (37:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,A) > > ^ > > Error 24 at (37:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,A) > > ^ > > Error 24 at (37:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,A) > > ^ > > Error 24 at (37:util.f) : syntax error > > external subroutine IAZZERO > > > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) > > ^ > > Error 22 at (69:util.f) : invalid real constant > > ^ > > Error 24 at (69:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) > > ^ > > Error 24 at (69:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) > > ^ > > Error 24 at (69:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) > > ^ > > Error 24 at (69:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) > > ^ > > Error 24 at (69:util.f) : syntax error > > external subroutine I8AZZERO > > > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) > > ^ > > Error 22 at (101:util.f) : invalid real constant > > ^ > > Error 24 at (101:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) > > ^ > > Error 24 at (101:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) > > ^ > > Error 24 at (101:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) > > ^ > > Error 24 at (101:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I) SHARED(II,N,M) > > ^ > > Error 24 at (101:util.f) : syntax error > > external function XRANF > > external subroutine MPRAND > > external subroutine UNITMX > > > > C$OMP 1 DO PRIVATE(I) > > ^ > > Error 22 at (156:util.f) : invalid real constant > > ^ > > Error 24 at (156:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I) > > ^ > > Error 24 at (156:util.f) : syntax error > > external subroutine SIMPSN > > external subroutine ZCLEAN > > > > C$OMP 1 DO PRIVATE(I) > > ^ > > Error 22 at (202:util.f) : invalid real constant > > ^ > > Error 24 at (202:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I) > > ^ > > Error 24 at (202:util.f) : syntax error > > external subroutine ICOPY > > > > C$OMP 1 DO PRIVATE(I,IAX,IBX) > > ^ > > Error 22 at (220:util.f) : invalid real constant > > ^ > > Error 24 at (220:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I,IAX,IBX) > > ^ > > Error 24 at (220:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I,IAX,IBX) > > ^ > > Error 24 at (220:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I,IAX,IBX) > > ^ > > Error 24 at (220:util.f) : syntax error > > external subroutine SYMMA > > > > C$OMP 1 DO PRIVATE(I,J) > > ^ > > Error 22 at (242:util.f) : invalid real constant > > ^ > > Error 24 at (242:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I,J) > > ^ > > Error 24 at (242:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I,J) > > ^ > > Error 24 at (242:util.f) : syntax error > > external subroutine MATMOV > > > > C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) > > ^ > > Error 22 at (263:util.f) : invalid real constant > > ^ > > Error 24 at (263:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) > > ^ > > Error 24 at (263:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) > > ^ > > Error 24 at (263:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) > > ^ > > Error 24 at (263:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) > > ^ > > Error 24 at (263:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) > > ^ > > Error 24 at (263:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) > > ^ > > Error 24 at (263:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I,J,JJ) SHARED(M,N,A,B) > > ^ > > Error 24 at (263:util.f) : syntax error > > external subroutine CGESUM > > > > C$OMP 1 DO PRIVATE(I,J) > > ^ > > Error 22 at (295:util.f) : invalid real constant > > ^ > > Error 24 at (295:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I,J) > > ^ > > Error 24 at (295:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I,J) > > ^ > > Error 24 at (295:util.f) : syntax error > > > > ELSE > > ^ > > Error 394 at (301:util.f) : syntax error - misplaced executable statement > > > > C$OMP 1 DO PRIVATE(I,J) > > ^ > > Error 22 at (303:util.f) : invalid real constant > > ^ > > Error 24 at (303:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I,J) > > ^ > > Error 24 at (303:util.f) : syntax error > > > > C$OMP 1 DO PRIVATE(I,J) > > ^ > > Error 24 at (303:util.f) : syntax error > > > > ENDIF > > ^ > > Error 394 at (309:util.f) : syntax error - misplaced executable statement > > external subroutine DATUM > > external subroutine NUMCPUS > > external subroutine INVMAT > > external subroutine INVERSEMAT > > external subroutine DSPEVY > > external subroutine ZGETMO > > external subroutine ZGTHR > > external subroutine ZSCTR > > external subroutine DURAND > > external subroutine DAXPYI > > external function IZAMAX > > external function IDAMIN > > external subroutine MY_GETARG > > external subroutine MY_GETENV > > external subroutine MY_GETLOG > > external subroutine ZCLEAN_K > > external subroutine SETKWF > > external subroutine FSKIP > > external function DGIVE > > external function ZGIVE > > external subroutine ROUNDOFF > > external subroutine DEBROT > > external subroutine STORE > > external subroutine DETERM > > external function NXXFUN > > external function SBES0 > > external function FC4 > > external function SB1 > > external function SB2 > > external function SB3 > > external function SB4 > > external function SB5 > > > > 52 Errors > > compilation aborted for ./util.f (code 1) > > make: *** [util.o] Error 1 > > > > How to solve it? > > Thanks, > > Sergey > > > > _______________________________________________ > > 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. > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > From axel.kohlmeyer at theochem.ruhr-uni-bochum.de Sun Nov 16 14:02:52 2003 From: axel.kohlmeyer at theochem.ruhr-uni-bochum.de (Axel Kohlmeyer) Date: Sun, 16 Nov 2003 14:02:52 +0100 Subject: [CPMD-list] OpenMP support with IFC In-Reply-To: Your message of "Sat, 15 Nov 2003 19:15:25 +0100." Message-ID: <200311161302.hAGD2qR05000@yello.theochem.ruhr-uni-bochum.de> >>> "JH" == Juerg Hutter writes: JH> Hi JH> changing the preprocessor flags from JH> -DPARALLEL JH> to JH> -DPARALLEL=PARALLEL JH> should also help. nope. won't work. c preprocessor macros are expanded recursively and this would create an infinite recursion. OTOH, using -DPARALLEL=parallel should do the trick, _if_ all of the prepocessor logic is always using '#ifdef PARALLEL' or '#if defined(PARALLEL)'. it _should_ and AFAICT it is ok in the 3.7 and the current cvs version. axel. p.s.: proving the one-who-knows(tm) wrong feels a bit like blasphemy. hope you don't mind. ;-) JH> Juerg JH> ---------------------------------------------------------- JH> Juerg Hutter Phone : ++41 1 635 4491 JH> Physical Chemistry Institute FAX : ++41 1 635 6838 JH> University of Zurich E-mail: hutter at pci.unizh.ch JH> Winterthurerstrasse 190 JH> CH-8057 Zurich, Switzerland JH> ---------------------------------------------------------- -- ======================================================================= 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/~axel.kohlmeyer/ ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From proffess at mail.ru Sun Nov 16 14:23:35 2003 From: proffess at mail.ru (=?koi8-r?Q?=22?=Sergey=?koi8-r?Q?=22=20?=) Date: Sun, 16 Nov 2003 16:23:35 +0300 Subject: [CPMD-list] OpenMP support with IFC Message-ID: Dear Axel and Juerg, Thanks for your help. Really, Using -DPARALLEL=parallel I can compile the CPMD. Thanks a lot, Best wishes, Sergey From apdipt at yahoo.co.in Mon Nov 3 20:50:37 2003 From: apdipt at yahoo.co.in (=?iso-8859-1?q?ABHIJIT=20PODDAR?=) Date: Mon, 3 Nov 2003 19:50:37 +0000 (GMT) Subject: [CPMD-list] installation query ?? Message-ID: <20031103195037.70838.qmail@web8103.mail.in.yahoo.com> I want to install the latest CPMD ver 3.7.2 onto my Intel Pentium 4 2.4 Ghz Hyper Threading machine having Red Hat 9 OS. When I run the shell script ./Configure from within the SOURCE directory I am given a list of architectures but I cannot find any "Intel 32 bit for Linux" option. Which architecture should I chose? In anticipation ... Sincerely, Abhijit. ________________________________________________________________________ Yahoo! India Matrimony: Find your partner online. Go to http://yahoo.shaadi.com From dkutlan at freemail.hu Tue Nov 4 23:33:33 2003 From: dkutlan at freemail.hu (=?ISO-8859-2?Q?Kutl=E1n_Diana?=) Date: Tue, 4 Nov 2003 23:33:33 +0100 (CET) Subject: [CPMD-list] Re: Troubleshooting with installation Message-ID: > > Dear CPMD group! > > I got the license of CPMD program (two weeks before) and I tried to > install this to the SGI R12000 computer, but during the translation a > problem occurred. I followed the instruction in manual: > > - at first I extracted the tar file (cpmdv37_2.tar) with tar > > - secondly I made the Makefile: configure SGI-ORIGIN (and SGI- ORIGIN- > MPI) > Makefile > > - thirdly: make -f Makefile > > During the translation a problem occurred, but I did not understand it. I > supposed, that some files are missing. > > I listed all the *.o, *.f and *.c file and I compared these files. I am not > an > expert but I recognized that two .c files (fdate.c and timef.c) has not > got > a pair among the *.o files, namely this two C++ file have not got > equivalent object files. Moreover the fdata.c file proved to be an empty > file and the timef.c contains only comments. > > Is it possible, that the SGI-ORIGIN platform name is wrong for the SGI > OKTANE? > or can you send me the correct fdate.c and timef.c file? > > In order to illustrate the problem I copy here the end of my make.log > file, which is the output file containing the errors. > > Please help me, because I cannot use this program at this time. > > On the other hand I'd like to ask you, if it is possible, please send me > some examples about the input file, because some parts of the manual > is not so clear for me and I hope that your examples can clarify every > details. > > Regards > > Zoltan Mucsi > > > > kdp_stress_kin.o kdpoints.o rhoofr_kdp.o mm_detsp.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 - > lcomplib.sgimath -lblas -lmpi > *** Error code 2 (bu21) > > ld64: FATAL 9 : I/O error (-lblas): No such file or directory > *** Error code 2 (bu21) > > > > > > From astirlin at cscs.ch Fri Nov 7 14:56:43 2003 From: astirlin at cscs.ch (Andras Stirling (CSCS)) Date: 07 Nov 2003 14:56:43 +0100 Subject: [CPMD-list] order of atoms and Vandebilt PP-s Message-ID: <1068213404.29246.15.camel@novazzano.usilu.net> Dear CPMD-users, For my system,I would like to use both ultrasoft and norm-conserving PP-s. I have noticed however, that depending on where I put in the input file the atoms having norm-conserving PP-s, I obtain different results. In the following example only the sulfur atoms are norm-conserving, while the others are ultrasoft PP-s. I performed wavefunction minimization in 10 steps. In the first case, the sulfur atoms are the last ones, while in the second case the sulfur atoms are placed after the iron atoms. INITIALIZE WAVEFUNCTION MAXSTEP 10 1.) VANDERBILT AUGMENTATION CHARGES (MEAN VALUE PER ATOM) ATOM TYPE NR. OF ATOMS CHARGE Fe 12 8.698 N 1 0.285 O 3 0.653 O 17 0.672 H 34 0.040 S 4 0.000 S 20 0.000 (K+E1+L+N+X) TOTAL ENERGY = -2098.77200681 A.U. (K) KINETIC ENERGY = 394.31861310 A.U. (E1=A-S+R) ELECTROSTATIC ENERGY = -1421.97144431 A.U. (S) ESELF = 1567.17302900 A.U. (R) ESR = 15.02526143 A.U. (L) LOCAL PSEUDOPOTENTIAL ENERGY = -1125.18996136 A.U. (N) N-L PSEUDOPOTENTIAL ENERGY = 350.21600136 A.U. (X) EXCHANGE-CORRELATION ENERGY = -296.14521560 A.U. GRADIENT CORRECTION ENERGY = -11.90686486 A.U. 2.) VANDERBILT AUGMENTATION CHARGES (MEAN VALUE PER ATOM) ATOM TYPE NR. OF ATOMS CHARGE Fe 12 8.698 S 4 0.000 S 20 0.000 N 1 0.000 O 3 0.000 O 17 0.000 H 34 0.414 (K+E1+L+N+X) TOTAL ENERGY = -2098.77242167 A.U. (K) KINETIC ENERGY = 394.31776465 A.U. (E1=A-S+R) ELECTROSTATIC ENERGY = -1421.97104142 A.U. (S) ESELF = 1567.17302900 A.U. (R) ESR = 15.02526143 A.U. (L) LOCAL PSEUDOPOTENTIAL ENERGY = -1125.19160611 A.U. (N) N-L PSEUDOPOTENTIAL ENERGY = 350.21809011 A.U. (X) EXCHANGE-CORRELATION ENERGY = -296.14562889 A.U. GRADIENT CORRECTION ENERGY = -11.90688997 A.U. It seems to me, that in the second case the N and O atoms are not calculated correctly as Vanderbilt pseudos. Nevertheless it might be that the first case is not correct either. I would appreciate any comments and ideas about this problem. Andras Stirling ETH/Lugano, Switzerland From YGMu at ntu.edu.sg Tue Nov 11 11:29:19 2003 From: YGMu at ntu.edu.sg (Mu Yuguang (Dr)) Date: Tue, 11 Nov 2003 18:29:19 +0800 Subject: [CPMD-list] cell symmetry Message-ID: <1613B0F4C9A9B1479B477BE021204F3C0160DC08@exchange06.staff.main.ntu.edu.sg> Is it correct to using the truncated octahedron cell as following: Cell 1.55 1.0 1.0 -0.33333333 -0.33333333 -0.33333333 combined with symmetry 5 (trigonal) If so, the volumn should be 1.55**3/2 (in Angstrom ) , right ? Regards, Yuguang ? Dr. Yuguang Mu School of Biological Sciences Nanyang Technological?University Singapore 637616 Tel: 0065-67906489 ? From alex.msilva at uol.com.br Tue Nov 11 13:16:37 2003 From: alex.msilva at uol.com.br (Alexander Martins) Date: Tue, 11 Nov 2003 10:16:37 -0200 Subject: [CPMD-list] Problems in the installation of CPMD-3.7.2 with IFC Message-ID: <3FB0D325.6090609@uol.com.br> HI, I tried to compile the CPMD-3.7.2 code in my RH 7.3 Linux machine with Intel Fortran Compliler. Before that, I've compiled the Lapack and Blas libraries with this same compiler. I put these libraries, lapack-ifc.a and f77ifc-blas.a, in the SORUCE directory. After the compilation of the obejct files, i've got this error: .... 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 -llapack-ifc -L. -lf77blas-ifc L. -latlas-ifc -Vaxlib ld: cannot find -llapack-ifc make: *** [cpmd.x] Error 1 What's wrong? The Makefile is the following: SHELL = /bin/sh # #--------------- Default Configuration for PC-IFC --------------- SRC = . DEST = . BIN = . #QMMM_FLAGS = -D__QMECHCOUPL #QMMM_LIBS = -L. -lmm FFLAGS = -c -r8 -w90 -w95 -O3 -pc64 -axM -ip -tpp6 LFLAGS = -llapack-ifc -L. -lf77blas-ifc L. -latlas-ifc -Vaxlib $(QMMM_LIBS) CFLAGS = -c -O2 -Wall CPP = /lib/cpp -P -C -traditional CPPFLAGS = -D__Linux -D__PGI -DLAPACK -DFFT_DEFAULT -DLINUX_IFC NOOPT_FLAG = CC = gcc FC = ifc LD = ifc AR = Regards, Alexander. -- ************************************************************ + Alexander Martins da Silva 2nd email: alex at iq.ufrj.br + + Rio de Janeiro, Brazil 22,80 s 43,43 w + ************************************************************ From jim_arken at yahoo.com Fri Nov 14 03:58:16 2003 From: jim_arken at yahoo.com (Jim Arken) Date: Thu, 13 Nov 2003 18:58:16 -0800 (PST) Subject: [CPMD-list] CPMD tests Message-ID: <20031114025816.57653.qmail@web20709.mail.yahoo.com> Dear CPMD community, I would like to obtain and begin using CPMD to study a number of different ionic systems(NaCL,LiF,LiBr,KCl). Since there aredifferent approaches to ab initio MD I am reading about these different options. Car-Parrinello appears to be the most popular and the fastest approach(is this true ?) but I am having some difficulty finding tests of this method. My understanding is that the theoretical basis of CPMD is less well founded than pure ground-state approaches and so I would like to find where its accuracy has been demonstrated. All I can find is tests on silicon which may not be relevant for my purposes. There are many simulations of these kinds of ionic materials which show some agreement with experiment but I assume that some kind of non-empirical evidence also exists. Somebody must have demonstrated (either theoretically or with simulation) that ground state forces are reproduced in practice. Can anybody point me in the right direction ? Thanks in advance, Jim Arken __________________________________ Do you Yahoo!? Protect your identity with Yahoo! Mail AddressGuard http://antispam.yahoo.com/whatsnewfree From proffess at yandex.ru Sat Nov 15 15:14:58 2003 From: proffess at yandex.ru (Sergei Lisenkov) Date: Sat, 15 Nov 2003 17:14:58 +0300 (MSK) Subject: [CPMD-list] Error:GORDER| PROGRAMING ERROR. INFORM THE PROGRAMER Message-ID: <3FB634E2.00000D.16134@colgate.yandex.ru> Dear CPMD authors, I met the next message (see below) on Compaq ALpha. ..... PARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARA NCPU NGW NHG PLANES GXRAYS HXRAYS ORBITALS Z-PLANES 0 49149 393279 12 826 3314 12 1 1 49145 393289 12 826 3314 13 1 2 49143 393289 12 826 3314 12 1 3 49144 393301 12 828 3314 12 1 4 49152 393269 12 828 3314 12 1 5 49150 393285 12 828 3314 13 1 6 49152 393303 12 828 3314 12 1 7 49144 393289 12 828 3314 12 1 8 49140 393287 12 828 3314 12 1 9 49138 393291 12 828 3314 13 1 10 49140 393281 12 828 3314 12 1 11 49142 393289 12 828 3314 12 1 12 49136 393303 12 828 3314 12 1 13 49142 393273 12 828 3314 13 1 14 49142 393291 12 828 3314 12 1 15 49148 393288 12 828 3316 12 1 16 49146 393284 12 828 3316 13 1 17 49146 393280 12 828 3316 12 1 18 49154 393296 12 828 3316 12 1 19 49154 393282 12 828 3316 12 1 20 49156 393278 12 828 3316 13 1 21 49154 393286 12 828 3316 12 1 22 49156 393280 12 828 3316 12 1 23 49160 393278 12 828 3316 12 1 24 49154 393151 12 827 3315 13 1 25 49156 393292 12 828 3316 12 1 26 49164 393286 12 828 3316 12 1 27 49166 393284 12 828 3316 12 1 28 49170 393296 12 828 3316 13 1 29 49170 393290 12 828 3316 12 1 GORDER| PROGRAMING ERROR. INFORM THE PROGRAMER G=0 COMPONENT ON PROCESSOR : 24 PARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARA What does it mean? Thanks, Sergey From sh at stchem.bham.ac.uk Mon Nov 17 15:14:23 2003 From: sh at stchem.bham.ac.uk (Scott Habershon) Date: Mon, 17 Nov 2003 14:14:23 +0000 Subject: [CPMD-list] Geometry optimization troubles.... Message-ID: <03111714142303.12107@schem09> Hi everyone, I'm trying to use CPMD to optimize the geometry of a molecular crystal but I'm having some problems. I've tried a number of different calculations, but I just can't get the geometry optimization calculation to converge. It seems OK for the first optimization iteration (ie. DETOT and GEMAX converge) but at the beginning of the second iteration I always see this message: WARNING| SPLINE REGION SMALLER THAN MAXIMUM G-VALUE and the following optimization doesn't go anywhere. I see the following periodically too: ODIIS| Insufficient progress for K = 1; reset! I guess this means that the minimization isn't working, but I'm baffled as to why......I've tried a large number of calculations with different minimization types, cutoffs, etc., but I've had no luck. Any useful suggestions would be greatly appreciated. I've included a typical input file that I'm using below, in case I'm making a stupid mistake...... Best wishes, Scott Habershon --------------------------------------------------------- ----- START OF INPUT FILE &CPMD CONVERGENCE 1.D-5 1.D-4 BFGS MAXITER 1000 MAXSTEP 1000 OPTIMIZE GEOMETRY WAVEFUNCTION MEMORY BIG STEEPEST DESCENT CELL STRUCTURE BONDS ANGLES &END &SYSTEM SYMMETRY 14 POINT GROUP AUTO ANGSTROM CELL ABSOLUTE 5.0822 5.0822 5.5513 0.0 0.0 0.0 KPOINTS MONKHORST-PACK FULL 2 2 2 CUTOFF 50 TESR 8 &END &DFT FUNCTIONAL BLYP &END &ATOMS BLAH....some coordinates........ &END ---- END OF INPUT FILE -- ----------------- Scott Habershon, School of Chemistry, The University of Birmingham, UK, B15 2TT. E-mail: sh at stchem.bham.ac.uk From audrius.alkauskas at unibas.ch Wed Nov 26 11:51:56 2003 From: audrius.alkauskas at unibas.ch (Audrius Alkauskas) Date: Wed, 26 Nov 2003 11:51:56 +0100 Subject: [CPMD-list] Copper, the difficult element Message-ID: <3FC485CC.8050905@unibas.ch> Hello all I have problems with copper when used with norm-conserving PPs. Does anyone have experience? With silver I had no problems, 50 Ry cutoff is enough to reproduce bulk parameters and surface parameters from the literature. But copper is painful. Other people use USPPs, but since k-points are not implemented with USPPs in CPMD, I use it with norm-conserving PPs. I use cutoff 80 Ry with PBE functional and PP generated by fhi98PP package, but cannot reproduce values from the literature - GGA seems to overbind more than LDA (reported in the literature). I get lattice constant 3.45 of bulk copper as compared to 3.65 GGA/USPP and 3.55 LDA/USPP and 3.50 experimental. Those, who have experience with copper, do you have any suggestions? How to optimize PP, what cutoff to use, etc, etc? Sincerely Audrius Alkausas =========================================================== Audrius Alkauskas, PhD student Condensed Matter Theory Group & NCCR for Nanoscale Science Institute of Physics, University of Basel Switzerland Homepage: http://monet.physik.unibas.ch/~audrius/ Tel.: +41-61-267-3708 =========================================================== From hutter at pci.unizh.ch Sat Nov 29 16:05:12 2003 From: hutter at pci.unizh.ch (Juerg Hutter) Date: Sat, 29 Nov 2003 16:05:12 +0100 (MET) Subject: [CPMD-list] Error:GORDER| PROGRAMING ERROR. INFORM THE PROGRAMER In-Reply-To: <3FB634E2.00000D.16134@colgate.yandex.ru> References: <3FB634E2.00000D.16134@colgate.yandex.ru> Message-ID: Hi this is an un-solved problem in CPMD. There are two possible workarounds 1) slightly change you computational box in one dimension eg from 10.000000 to 10.000001 this helps some times 2) go to loadpa.F and comment out the two lines with CALL STOPGM... in subroutine GORDER this just ignores the problem, but usually doesn't harm the calculation. However, be sure to check that the results are reasonable regards Juerg ---------------------------------------------------------- Juerg Hutter Phone : ++41 1 635 4491 Physical Chemistry Institute FAX : ++41 1 635 6838 University of Zurich E-mail: hutter at pci.unizh.ch Winterthurerstrasse 190 CH-8057 Zurich, Switzerland ---------------------------------------------------------- On Sat, 15 Nov 2003, Sergei Lisenkov wrote: > Dear CPMD authors, > > I met the next message (see below) on Compaq ALpha. > ..... > PARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARA > NCPU NGW NHG PLANES GXRAYS HXRAYS ORBITALS Z-PLANES > 0 49149 393279 12 826 3314 12 1 > 1 49145 393289 12 826 3314 13 1 > 2 49143 393289 12 826 3314 12 1 > 3 49144 393301 12 828 3314 12 1 > 4 49152 393269 12 828 3314 12 1 > 5 49150 393285 12 828 3314 13 1 > 6 49152 393303 12 828 3314 12 1 > 7 49144 393289 12 828 3314 12 1 > 8 49140 393287 12 828 3314 12 1 > 9 49138 393291 12 828 3314 13 1 > 10 49140 393281 12 828 3314 12 1 > 11 49142 393289 12 828 3314 12 1 > 12 49136 393303 12 828 3314 12 1 > 13 49142 393273 12 828 3314 13 1 > 14 49142 393291 12 828 3314 12 1 > 15 49148 393288 12 828 3316 12 1 > 16 49146 393284 12 828 3316 13 1 > 17 49146 393280 12 828 3316 12 1 > 18 49154 393296 12 828 3316 12 1 > 19 49154 393282 12 828 3316 12 1 > 20 49156 393278 12 828 3316 13 1 > 21 49154 393286 12 828 3316 12 1 > 22 49156 393280 12 828 3316 12 1 > 23 49160 393278 12 828 3316 12 1 > 24 49154 393151 12 827 3315 13 1 > 25 49156 393292 12 828 3316 12 1 > 26 49164 393286 12 828 3316 12 1 > 27 49166 393284 12 828 3316 12 1 > 28 49170 393296 12 828 3316 13 1 > 29 49170 393290 12 828 3316 12 1 > GORDER| PROGRAMING ERROR. INFORM THE PROGRAMER > G=0 COMPONENT ON PROCESSOR : 24 > PARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARA > > What does it mean? > > Thanks, > Sergey > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > From hutter at pci.unizh.ch Sat Nov 29 16:00:10 2003 From: hutter at pci.unizh.ch (Juerg Hutter) Date: Sat, 29 Nov 2003 16:00:10 +0100 (MET) Subject: [CPMD-list] CPMD tests In-Reply-To: <20031114025816.57653.qmail@web20709.mail.yahoo.com> References: <20031114025816.57653.qmail@web20709.mail.yahoo.com> Message-ID: Hi > Dear CPMD community, > > I would like to obtain and begin using CPMD to study > a number of different ionic > systems(NaCL,LiF,LiBr,KCl). > Since there aredifferent approaches to ab initio MD I > am reading about these different options. > Car-Parrinello appears to be the most popular and the > fastest approach(is this true ?) but I am having some > difficulty finding tests of this method. My You may start from the papers cited in chapter 5.4 in Ab Initio Molecular Dynamics: Theory and Implementation D. Marx and J. Hutter published in Modern Methods and Algorithms of Quantum Chemistry, Forschungszentrum J?lich, NIC Series, Vol. 1 (2000), 301-449 see http://www.fz-juelich.de/nic-series/Volume1/ > understanding is that the > theoretical basis of CPMD is less well founded than > pure > ground-state approaches and so I would like to find This is discussed in chapter 2.4 of the same reference. There are also references to papers that give mathematical proof for the CP method. > where its accuracy has been demonstrated. All I can > find is tests on silicon which may not be relevant for > my purposes. There are many simulations of these kinds > of ionic materials which show some agreement with > experiment but I assume that some kind of > non-empirical > evidence also exists. Somebody must have demonstrated > (either theoretically or with simulation) that ground > state forces are reproduced in practice. See again chapter 2.4. regards Juerg Hutter > Can anybody point me in the right direction ? > > Thanks in advance, > > > Jim Arken > > > > > __________________________________ > Do you Yahoo!? > Protect your identity with Yahoo! Mail AddressGuard > http://antispam.yahoo.com/whatsnewfree > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > From M.A.Zwijnenburg at tnw.tudelft.nl Fri Nov 28 18:42:05 2003 From: M.A.Zwijnenburg at tnw.tudelft.nl (Martijn Zwijnenburg) Date: Fri, 28 Nov 2003 18:42:05 +0100 Subject: [CPMD-list] Geometry optimization troubles.... In-Reply-To: <03111714142303.12107@schem09> Message-ID: <3FC796F1.10865.3A215D55@localhost> Hi Scott, I asked a similar question some time ago and this was Juerg Hutters answer (although I think in 3.7.x the dfault number of spline poins has been increased): Hi this in fact touches several points. 1) Pseudopotential functions in CPMD are calculated on a radial grid in G-space and then used in spline interpolations. This speeds up variable cell calculations considerably. The maximum grid point is given by the cutoff. With the keyword SPLINE RANGE x.xx you can enlarge the grid to x.xx times the cutoff. BTW. the default number of spline points is 501. This is at the lower limit for accuracy. Especially if you have high cutoffs it is better to increase this value, e.g. SPLINE POINTS 2500 2) In a variable cell calculation, CPMD uses a constant number of plane waves. Therefore if your cell contracts the cutoff increases, if the cell gets larger the cutoff decreases. So if you have the first case the spline interpolation needs points above the original cutoff and you get a warning. Depending on the amount of change in the cell you expect a value for the range of 2 - 5 is needed. 3) Comming back to the cunstant number of plane waves. If your cell gets larger the effective cutoff decreases. This may have very undesirable effects and it is better to define the plane waves on a box larger than any box you anticipate the simulation will reach. In order not to have to start with an unreasonable cell you can define the plane waves not with the actual box but with a reference cell, use the keyword REFERENCE CELL a b c xa xb xc 4) However, you really want to do a constant cutoff calculation, not a constant number of plane waves. For technical reasons this is not possible and in principle you should do the calculation at a high enough cutoff in order that the calculation is converged all along the simulation path (with the slight changes in cutoff). To avoid this very high cutoffs the group in Trieste came up with a method that allows to perform pseudo constant cutoff calculations. This method is implemented in CPMD and explained in this paper M. Bernasconi et al. J. Phys. Chem Solids 56 pp 501-505 (1995) Hope this explains some of the issues of variable cell calculations with CPMD regards Juerg ---------------------------------------------------------- Juerg Hutter Phone : ++41 1 635 4491 Physical Chemistry Institute FAX : ++41 1 635 6838 University of Zurich E-mail: hutter at pci.unizh.ch Winterthurerstrasse 190 CH-8057 Zurich, Switzerland ---------------------------------------------------------- On Mon, 3 Jun 2002, Martijn Zwijnenburg wrote: > Hi, > > I'm optimizing quartz (both ion positions and cell volume) using > CPMD and get the warning message: Warning! Spline region > smaller then maximum G-value. The optimization seems to > converge nicely but what does this warning mean / imply? > > Cheers, > Martijn Hope this helps. Cheers, Martijn ---------------------------------------------------------------------- --- Martijn Zwijnenburg Lab. of Applied Organic Chemistry and Catalysis Delft University of Technology Julianalaan 136 2628 BL Delft The Netherlands Tel: 0031-(0)152782691 Fax: 0031-(0)152784700 e-mail: M.A.Zwijnenburg at tnw.tudelft.nl web page: http://come.to/tock From Ari.P.Seitsonen at PCI.unizh.ch Fri Nov 28 17:59:50 2003 From: Ari.P.Seitsonen at PCI.unizh.ch (Ari P Seitsonen) Date: Fri, 28 Nov 2003 17:59:50 +0100 (CET) Subject: [CPMD-list] Problems in the installation of CPMD-3.7.2 with IFC In-Reply-To: <3FB0D325.6090609@uol.com.br> References: <3FB0D325.6090609@uol.com.br> Message-ID: Dear Alexander, The libraries should be called "liblapack-ifc.a" etc. 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 On Tue, 11 Nov 2003, Alexander Martins wrote: > > HI, > > I tried to compile the CPMD-3.7.2 code in my RH 7.3 Linux > machine with Intel Fortran Compliler. Before that, I've compiled the > Lapack and Blas libraries with this same compiler. I put these > libraries, lapack-ifc.a and f77ifc-blas.a, in the SORUCE directory. > After the compilation of the obejct files, i've got this error: > > .... > > 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 > -llapack-ifc -L. -lf77blas-ifc L. -latlas-ifc -Vaxlib > > ld: cannot find -llapack-ifc > > make: *** [cpmd.x] Error 1 > > > What's wrong? The Makefile is the following: > > SHELL = /bin/sh > > # > > #--------------- Default Configuration for PC-IFC --------------- > > SRC = . > > DEST = . > > BIN = . > > #QMMM_FLAGS = -D__QMECHCOUPL > > #QMMM_LIBS = -L. -lmm > > FFLAGS = -c -r8 -w90 -w95 -O3 -pc64 -axM -ip -tpp6 > > LFLAGS = -llapack-ifc -L. -lf77blas-ifc L. -latlas-ifc -Vaxlib $(QMMM_LIBS) > > CFLAGS = -c -O2 -Wall > > CPP = /lib/cpp -P -C -traditional > > CPPFLAGS = -D__Linux -D__PGI -DLAPACK -DFFT_DEFAULT -DLINUX_IFC > > NOOPT_FLAG = > > CC = gcc > > FC = ifc > > LD = ifc > > AR = > > > Regards, > > Alexander. > > -- > ************************************************************ > + Alexander Martins da Silva 2nd email: alex at iq.ufrj.br + > + Rio de Janeiro, Brazil 22,80 s 43,43 w + > ************************************************************ > > > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > > From Ari.P.Seitsonen at PCI.unizh.ch Fri Nov 28 17:57:59 2003 From: Ari.P.Seitsonen at PCI.unizh.ch (Ari P Seitsonen) Date: Fri, 28 Nov 2003 17:57:59 +0100 (CET) Subject: [CPMD-list] installation query ?? In-Reply-To: <20031103195037.70838.qmail@web8103.mail.in.yahoo.com> References: <20031103195037.70838.qmail@web8103.mail.in.yahoo.com> Message-ID: Dear Abhijit, Which compiler do you use? There are plenty of options for 32 bit linux: PC-ABSOFT PC-ABSOFT-MPI PC-IFC PC-IFC-MPI PC-PGI PC-PGI-MPI (Maybe the "PC" is misleading, for us "PC" stands for linux, not for this funny products from a company in Redmond) 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 On Mon, 3 Nov 2003, ABHIJIT PODDAR wrote: > I want to install the latest CPMD ver 3.7.2 onto my > Intel Pentium 4 2.4 Ghz Hyper Threading machine having > Red Hat 9 OS. > > When I run the shell script ./Configure from within > the SOURCE directory I am given a list of > architectures but I cannot find any "Intel 32 bit for > Linux" option. > > Which architecture should I chose? > > In anticipation ... > Sincerely, > Abhijit. > > ________________________________________________________________________ > Yahoo! India Matrimony: Find your partner online. > Go to http://yahoo.shaadi.com > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > From Ari.P.Seitsonen at PCI.unizh.ch Fri Nov 28 17:48:16 2003 From: Ari.P.Seitsonen at PCI.unizh.ch (Ari P Seitsonen) Date: Fri, 28 Nov 2003 17:48:16 +0100 (CET) Subject: [CPMD-list] about RAMAN In-Reply-To: <3F926BAD.7080503@status-m.ru> References: <20031009085434.19418.qmail@vip.sina.com.cn> <3F926BAD.7080503@status-m.ru> Message-ID: Dear V., Please see the paper # Phys. Rev. Lett. 88, 176401 (2002) # Anharmonic Raman Spectra in High-Pressure Ice from Ab Initio Simulations # Anna Putrino and Michele Parrinello # http://link.aps.org/abstract/PRL/v88/e176401 Unfortunately I didn't find Anna's thesis on-line. Basically what you to do is the auto-correlation function of the polarisability. Greetings from the foggy 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 On Sun, 19 Oct 2003, V.Sokolov wrote: > cdy2000 at vip.sina.com wrote: > > >I have computed RAMAN by LINEAR RESPONSE, and get two files: > >APT and POLARIZABILITY, lots of data in these files. I want > >to know the meaning of the data, please give me some answer in detail. > > > > > > > Dear CPMD experts, > Wouldn't you give an advice how to get Raman intensities from these data? > > Thanking you in advance, > -- > V.Sokolov > > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > From Ari.P.Seitsonen at PCI.unizh.ch Fri Nov 28 18:03:45 2003 From: Ari.P.Seitsonen at PCI.unizh.ch (Ari P Seitsonen) Date: Fri, 28 Nov 2003 18:03:45 +0100 (CET) Subject: [CPMD-list] Error:GORDER| PROGRAMING ERROR. INFORM THE PROGRAMER In-Reply-To: <3FB634E2.00000D.16134@colgate.yandex.ru> References: <3FB634E2.00000D.16134@colgate.yandex.ru> Message-ID: Ciao Sergey, It's a warning from a test, which probes whether the G vectors are in a "safe" order, namely such that after a restart with a different number of processes or on a different machine the results agree. Usually this error only occurs in large systems with a high cut-off energy, i.e. where one gets lots of close-lying G vectors. Then it might be that on some architectures it becomes difficult to decide whether the length of two G vectors should be "equal" (i.e. they belong to the same star of vectors) or not. That's what I vaguely remember... Some one please correct my biggest mistakes. 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 On Sat, 15 Nov 2003, Sergei Lisenkov wrote: > Dear CPMD authors, > > I met the next message (see below) on Compaq ALpha. > ..... > PARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARA > NCPU NGW NHG PLANES GXRAYS HXRAYS ORBITALS Z-PLANES > 0 49149 393279 12 826 3314 12 1 > 1 49145 393289 12 826 3314 13 1 > 2 49143 393289 12 826 3314 12 1 > 3 49144 393301 12 828 3314 12 1 > 4 49152 393269 12 828 3314 12 1 > 5 49150 393285 12 828 3314 13 1 > 6 49152 393303 12 828 3314 12 1 > 7 49144 393289 12 828 3314 12 1 > 8 49140 393287 12 828 3314 12 1 > 9 49138 393291 12 828 3314 13 1 > 10 49140 393281 12 828 3314 12 1 > 11 49142 393289 12 828 3314 12 1 > 12 49136 393303 12 828 3314 12 1 > 13 49142 393273 12 828 3314 13 1 > 14 49142 393291 12 828 3314 12 1 > 15 49148 393288 12 828 3316 12 1 > 16 49146 393284 12 828 3316 13 1 > 17 49146 393280 12 828 3316 12 1 > 18 49154 393296 12 828 3316 12 1 > 19 49154 393282 12 828 3316 12 1 > 20 49156 393278 12 828 3316 13 1 > 21 49154 393286 12 828 3316 12 1 > 22 49156 393280 12 828 3316 12 1 > 23 49160 393278 12 828 3316 12 1 > 24 49154 393151 12 827 3315 13 1 > 25 49156 393292 12 828 3316 12 1 > 26 49164 393286 12 828 3316 12 1 > 27 49166 393284 12 828 3316 12 1 > 28 49170 393296 12 828 3316 13 1 > 29 49170 393290 12 828 3316 12 1 > GORDER| PROGRAMING ERROR. INFORM THE PROGRAMER > G=0 COMPONENT ON PROCESSOR : 24 > PARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARAPARA > > What does it mean? > > Thanks, > Sergey > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > > From Ari.P.Seitsonen at PCI.unizh.ch Fri Nov 28 18:11:43 2003 From: Ari.P.Seitsonen at PCI.unizh.ch (Ari P Seitsonen) Date: Fri, 28 Nov 2003 18:11:43 +0100 (CET) Subject: [CPMD-list] Geometry optimization troubles.... In-Reply-To: <03111714142303.12107@schem09> References: <03111714142303.12107@schem09> Message-ID: Dear Scott, Have you tried the option "SPLINE RANGE"? (Also if the cell changes a lot, you might get into problems, in that case you might want to check the initial guess first) Terveiset Z?richst?, apsi PS To the developers: Does the calculation of stress work with k points now? Great if it does! -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=- 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 On Mon, 17 Nov 2003, Scott Habershon wrote: > Hi everyone, > > I'm trying to use CPMD to optimize the geometry of a molecular crystal but > I'm having some problems. > I've tried a number of different calculations, but I just can't get the > geometry optimization calculation to converge. > It seems OK for the first optimization iteration (ie. DETOT and GEMAX > converge) but at the beginning of the second iteration I always see this > message: > > WARNING| SPLINE REGION SMALLER THAN MAXIMUM G-VALUE > > and the following optimization doesn't go anywhere. > I see the following periodically too: > > ODIIS| Insufficient progress for K = 1; reset! > > I guess this means that the minimization isn't working, but I'm baffled as to > why......I've tried a large number of calculations with different > minimization types, cutoffs, etc., but I've had no luck. > > Any useful suggestions would be greatly appreciated. I've included a typical > input file that I'm using below, in case I'm making a stupid mistake...... > > Best wishes, > > Scott Habershon > --------------------------------------------------------- > > ----- START OF INPUT FILE > &CPMD > CONVERGENCE > 1.D-5 1.D-4 > BFGS > MAXITER > 1000 > MAXSTEP > 1000 > OPTIMIZE GEOMETRY WAVEFUNCTION > MEMORY BIG > STEEPEST DESCENT CELL > STRUCTURE BONDS ANGLES > &END > > &SYSTEM > SYMMETRY > 14 > POINT GROUP > AUTO > ANGSTROM > CELL ABSOLUTE > 5.0822 5.0822 5.5513 0.0 0.0 0.0 > KPOINTS MONKHORST-PACK FULL > 2 2 2 > CUTOFF > 50 > TESR > 8 > &END > > &DFT > FUNCTIONAL BLYP > &END > > &ATOMS > BLAH....some coordinates........ > &END > ---- END OF INPUT FILE > > -- > ----------------- > Scott Habershon, > School of Chemistry, > The University of Birmingham, > UK, B15 2TT. > E-mail: sh at stchem.bham.ac.uk > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list > From aulucfph at iitr.ernet.in Sun Nov 30 13:27:22 2003 From: aulucfph at iitr.ernet.in (sushil auluck) Date: Sun, 30 Nov 2003 17:57:22 +0530 Subject: [CPMD-list] Re: compilation troubles.... References: <03111714142303.12107@schem09> Message-ID: <3FC9E22A.6030102@iitr.ernet.in> hi, as a lot of new users of cpmd have problems in compiling, it might be a good idea if someone OR cpmd.org put up the compiled versions on the net.at least for the linux version this ought not to be a big problem.also compiled versions for other OS's should also be posted on the cpmd.org site........ s.auluck -- **************************************************************************** Prof. Sushil Auluck Phone: +91-1332-274075(Home) Department of Physics +91-1332-285010(Home) Indian Institute of Technology +91-1332-285745(Work) Roorkee 247 667 Uttranchal Fax: +91-1332-273560 India e-mail: aulucfph at iitr.ernet.in sushilauluck at yahoo.co.in http://www.iitr.ernet.in/acads/depts/physics/faculty/facthtml/aulucfph.shtml **************************************************************************** From cur at zurich.ibm.com Sat Nov 29 17:06:47 2003 From: cur at zurich.ibm.com (Alessandro Curioni) Date: Sat, 29 Nov 2003 17:06:47 +0100 Subject: [CPMD-list] cpmd-list problem - solved Message-ID: Dear all, we had a problem with the cpmd mailing list that prevented it to operate normally in the last few days (thousands of spam mails). Now everything is back to normality, sorry for any problem this has created. Best Regards, A. Curioni From Ari.P.Seitsonen at PCI.unizh.ch Fri Nov 28 18:21:23 2003 From: Ari.P.Seitsonen at PCI.unizh.ch (Ari P Seitsonen) Date: Fri, 28 Nov 2003 18:21:23 +0100 (CET) Subject: [CPMD-list] Re: Troubleshooting with installation In-Reply-To: References: Message-ID: Dear Diana, The problem is related to the BLAS library; on the SGI origin, in which I have an account, it is located in the standard path: # mymachine (~) : find /usr/lib* -name '*blas*' -print # /usr/lib/images/blast.icon # /usr/lib/filetype/system/iconlib/blast.fti # /usr/lib/desktop/iconcatalog/pages/C/Demos/blast # /usr/lib/libblas.so # /usr/lib/libblas_mp.so # /usr/lib32/mips3/libblas.so # /usr/lib32/mips3/libblas_mp.so # /usr/lib32/mips4/r8000/libblas.so # /usr/lib32/mips4/r8000/libblas_mp.so # /usr/lib32/mips4/libblas_mp.so # /usr/lib32/mips4/libblas.so # /usr/lib32/libblas.so # /usr/lib32/libblas_mp.so # /usr/lib64/mips3/libblas.so # /usr/lib64/mips3/libblas_mp.so # /usr/lib64/mips4/r8000/libblas_mp.so # /usr/lib64/mips4/r8000/libblas.so # /usr/lib64/mips4/libblas.so # /usr/lib64/mips4/libblas_mp.so # /usr/lib64/libblas.so # /usr/lib64/libblas_mp.so If your system doesn't have blas installed you can easily compile it yourself (the source is available under http://www.netlib.org/). Saluti da Zurigo, 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 On Tue, 4 Nov 2003, [ISO-8859-2] Kutl?n Diana wrote: > > > > > Dear CPMD group! > > > > I got the license of CPMD program (two weeks before) and I tried to > > install this to the SGI R12000 computer, but during the translation a > > problem occurred. I followed the instruction in manual: > > > > - at first I extracted the tar file (cpmdv37_2.tar) with tar > > > > - secondly I made the Makefile: configure SGI-ORIGIN (and SGI- > ORIGIN- > > MPI) > Makefile > > > > - thirdly: make -f Makefile > > > > During the translation a problem occurred, but I did not understand > it. I > > supposed, that some files are missing. > > > > I listed all the *.o, *.f and *.c file and I compared these files. I am not > > an > > expert but I recognized that two .c files (fdate.c and timef.c) has not > > got > > a pair among the *.o files, namely this two C++ file have not got > > equivalent object files. Moreover the fdata.c file proved to be an > empty > > file and the timef.c contains only comments. > > > > Is it possible, that the SGI-ORIGIN platform name is wrong for the > SGI > > OKTANE? > > or can you send me the correct fdate.c and timef.c file? > > > > In order to illustrate the problem I copy here the end of my make.log > > file, which is the output file containing the errors. > > > > Please help me, because I cannot use this program at this time. > > > > On the other hand I'd like to ask you, if it is possible, please send me > > some examples about the input file, because some parts of the > manual > > is not so clear for me and I hope that your examples can clarify every > > details. > > > > Regards > > > > Zoltan Mucsi > > > > > > > > kdp_stress_kin.o kdpoints.o rhoofr_kdp.o mm_detsp.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 - > > lcomplib.sgimath -lblas -lmpi > > *** Error code 2 (bu21) > > > > ld64: FATAL 9 : I/O error (-lblas): No such file or directory > > *** Error code 2 (bu21) > > _______________________________________________ > CPMD-list mailing list > CPMD-list at cpmd.org > http://www.cpmd.org/mailman/listinfo/cpmd-list >