From chemistry-request@server.ccl.net Thu May 1 12:27:24 2003 Received: from nature.Berkeley.EDU ([128.32.175.19]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id h41GROV02614 for ; Thu, 1 May 2003 12:27:24 -0400 Received: from localhost (localhost [127.0.0.1]) by nature.Berkeley.EDU (Postfix) with ESMTP id 36E0426135D for ; Thu, 1 May 2003 09:27:24 -0700 (PDT) Received: from nature.Berkeley.EDU ([127.0.0.1]) by localhost (nature.Berkeley.EDU [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 28957-02 for ; Thu, 1 May 2003 09:27:23 -0700 (PDT) Received: from nature.berkeley.edu (localhost [127.0.0.1]) by nature.Berkeley.EDU (Postfix) with SMTP id 50D36261351 for ; Thu, 1 May 2003 09:27:23 -0700 (PDT) Received: from 128.32.27.14 (SquirrelMail authenticated user moconnor) by nature.Berkeley.EDU with HTTP; Thu, 1 May 2003 09:27:23 -0700 (PDT) Message-ID: <1075.128.32.27.14.1051806443.squirrel@nature.Berkeley.EDU> Date: Thu, 1 May 2003 09:27:23 -0700 (PDT) Subject: re:VASP From: " Mary V. O'Connor" To: "computational chemistry" X-Priority: 3 Importance: Normal X-Mailer: SquirrelMail (version 1.2.11) MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Virus-Scanned: by amavisd-new at nature.berkeley.edu Hello: To everyone using the Vienna ab-initio software package (VASP)-could I get opinions on the strengths and weaknesses of the software, as well as all of its capabilities? Thanks, Mary Mary V. O'Connor, Ph.D. Rm. 235 Hilgard Hall University of California Berkeley, CA 94720 From chemistry-request@server.ccl.net Fri May 2 15:12:09 2003 Received: from col-msxproto2.col.missouri.edu ([128.206.3.152]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id h42JC9V15652 for ; Fri, 2 May 2003 15:12:09 -0400 Received: from COL-EMAIL2 ([128.206.3.132]) by col-msxproto2.col.missouri.edu with Microsoft SMTPSVC(5.0.2195.6673); Fri, 2 May 2003 14:12:00 -0500 x-mimeole: Produced By Microsoft Exchange V6.0.6375.0 content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="Windows-1252" Subject: dummy/pseuo atoms for NOE in CHARMM Date: Fri, 2 May 2003 14:12:00 -0500 Message-ID: <3393392D7484804AAC174AA1D24F08551DF3BF@COL-EMAIL2.col.missouri.edu> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: dummy/pseuo atoms for NOE in CHARMM Thread-Index: AcMQ3rVHhqypO5KNT5ChedJm5b3Z6A== From: "Hu, Xiaohua" To: X-OriginalArrivalTime: 02 May 2003 19:12:00.0894 (UTC) FILETIME=[B56A69E0:01C310DE] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by server.ccl.net id h42JC9V15657 Dear Friends Does any one have a procedure of setting up dummy/pseuo atoms for NOE distance constraints invloving equivalent CH2 and CH3 protons using CHARMM? Thank you for your help Shawn Hu Chemistry Depatment University of Missouri, Columbia From chemistry-request@server.ccl.net Fri May 2 07:53:52 2003 Received: from richardson.uni2.net ([130.227.52.104]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id h42BrpV26557 for ; Fri, 2 May 2003 07:53:52 -0400 Received: from HLUSF26A.lundbeck.com (mail.lundbeck.com [130.227.160.31]) by richardson.uni2.net (8.11.6/8.11.6) with SMTP id h42BrpH11942 for ; Fri, 2 May 2003 13:53:52 +0200 Received: from ([172.16.0.163]) by HLUSF26A.lundbeck.com; Fri, 02 May 2003 13:55:23 +0200 (CEST) Received: FROM ludk0006.lundbeck.com BY ludk0163.lundbeck.com ; Fri May 02 13:53:48 2003 +0200 Received: by ludk0006.lundbeck.com with Internet Mail Service (5.5.2653.19) id ; Fri, 2 May 2003 13:53:49 +0200 Message-ID: From: Jan Torleif Pedersen To: "'chemistry@ccl.net'" Subject: Glycosphingolipid analogs Date: Fri, 2 May 2003 13:53:47 +0200 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2653.19) Content-Type: text/plain; charset="iso-8859-1" Greetings CClers, The list looks heavy on the QM stuff these days ! I have a more practical questions - has anybody worked on glycosphingolipid analogs especially GM3 analogs. I am trying to identify a lab that has/is working on the synthesis of GM3 analogs. What is the appropriate newsgroup/mailing community/mailing list to address with such a query. Many regards -- Jan Dr. Jan Torleif Pedersen Section head, Bioinformatics Department of Molecular Genetics H. Lundbeck A/S Ottiliavej 9, 2500 DK-Valby Denmark phone: +45 36 43 28 87 FAX: 36438271 email: jatp@lundbeck.com From chemistry-request@server.ccl.net Thu May 1 12:49:59 2003 Received: from epsilon.ccqc.uga.edu ([128.192.2.119]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id h41GnxV06228 for ; Thu, 1 May 2003 12:49:59 -0400 Received: from epsilon.ccqc.uga.edu (loopback [127.0.0.1]) by epsilon.ccqc.uga.edu (8.12.5/8.12.5) with ESMTP id h417Mlon000969; Thu, 1 May 2003 03:22:47 -0400 Received: from localhost (ericyan@localhost) by epsilon.ccqc.uga.edu (8.12.5/8.12.5/Submit) with ESMTP id h417MlAS000965; Thu, 1 May 2003 03:22:47 -0400 Date: Thu, 1 May 2003 03:22:47 -0400 (EDT) From: ericyan@epsilon.ccqc.uga.edu To: jmmckel@attglobal.net cc: chemistry@ccl.net Subject: Re: CCL:Intel Linux compiler 7.0 In-Reply-To: <3EAE58C0.7722C2EB@attglobal.net> Message-ID: References: <3EAE58C0.7722C2EB@attglobal.net> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII That seems to be a problem when linking both C and Fortran object files. My way (though not the best one) of solving this problem is by also linking with GCC's g2c library: -L/usr/lib/gcc-lib/i386-redhat-linux/2.96 -lg2c The actual linking path may be different in your machine Good luck. Eric -------------------------------------------------------------------- Eric Yan | phone: 706 542-4022 Ctr. for Comp. Quantum Chem. | fax: 706 542-0406 University of Georgia | e-mail: ericyan@ccqc.uga.edu Athens, GA 30602-2556 | http://epsilon.ccqc.uga.edu -------------------------------------------------------------------- On Tue, 29 Apr 2003 jmmckel@attglobal.net wrote: > Greetings! > > I have successfully compiled a version of MOPAC using the unsupported > Intel Linux fortran compiler Version7.0 [I need F90 for dynamic memory > allocation.] Upon linking the code against the Intel libraries as well > as libf77blas,libatlas, liblapack and lapack_linux libraries from > netlib.org I get the following complaint: > > for dlamch.o s_wsfe, do_fio, and e_wsfe are not resolved. > > Comments and suggestions are most welcome... > > many thanks! > > John McKelvey > > > > > > > -= This is automatically added to each message by mailing script =- > CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- To Admins > Ftp: ftp.ccl.net | WWW: http://www.ccl.net/chemistry/ | Jan: jkl@ccl.net > > > > > From chemistry-request@server.ccl.net Thu May 1 18:26:58 2003 Received: from nature.Berkeley.EDU ([128.32.175.19]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id h41MQwV12544 for ; Thu, 1 May 2003 18:26:58 -0400 Received: from localhost (localhost [127.0.0.1]) by nature.Berkeley.EDU (Postfix) with ESMTP id 64E1526134A for ; Thu, 1 May 2003 15:26:58 -0700 (PDT) Received: from nature.Berkeley.EDU ([127.0.0.1]) by localhost (nature.Berkeley.EDU [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 05280-08 for ; Thu, 1 May 2003 15:26:57 -0700 (PDT) Received: from nature.berkeley.edu (localhost [127.0.0.1]) by nature.Berkeley.EDU (Postfix) with SMTP id A9E4A26132D for ; Thu, 1 May 2003 15:26:57 -0700 (PDT) Received: from 128.32.27.14 (SquirrelMail authenticated user moconnor) by nature.Berkeley.EDU with HTTP; Thu, 1 May 2003 15:26:57 -0700 (PDT) Message-ID: <1066.128.32.27.14.1051828017.squirrel@nature.Berkeley.EDU> Date: Thu, 1 May 2003 15:26:57 -0700 (PDT) Subject: DFT and H bonding From: " Mary V. O'Connor" To: "computational chemistry" X-Priority: 3 Importance: Normal X-Mailer: SquirrelMail (version 1.2.11) MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Virus-Scanned: by amavisd-new at nature.berkeley.edu Hello again: Sorry for 2 questions in one day. Can anyone give me reviews, papers or even personal opinions about how various DFT functionals handle hydrogen bonding? I am interested in both GTO and plane wave basis sets. Thanks, Mary Mary V. O'Connor, Ph.D. Rm. 235 Hilgard Hall University of California Berkeley, CA 94720 From chemistry-request@server.ccl.net Sat May 3 22:26:44 2003 Received: from localhost.localdomain ([162.105.153.224]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id h442QhV14868 for ; Sat, 3 May 2003 22:26:44 -0400 Received: from localhost.localdomain (lab [127.0.0.1]) by localhost.localdomain (8.12.8/8.12.8) with ESMTP id h442ST7x023155 for ; Sun, 4 May 2003 10:28:29 +0800 Received: (from nichloas@localhost) by localhost.localdomain (8.12.8/8.12.8/Submit) id h442SSov023153 for chemistry@ccl.net; Sun, 4 May 2003 10:28:28 +0800 Date: Sun, 4 May 2003 10:28:28 +0800 From: Li Daobing To: chemistry@ccl.net Subject: What's the expression of D5 shell and F shell? Message-ID: <20030504022828.GA23140@lab> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.4.1i Hello, As I know, 6-31G basis set use D6 shell and 6-311G use D5 shell. I know the expression of D6 shell: Dxy = N * exp(-ar^2) * x * y; etc. But I don't know the expressions of D5 shell or F shell, Who can tell me the expressions or where can I find them? Regards, Li Daobing From chemistry-request@server.ccl.net Sat May 3 15:50:04 2003 Received: from ned.sims.nrc.ca ([132.246.109.100]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id h43Jo4V06005 for ; Sat, 3 May 2003 15:50:04 -0400 Received: from ned.sims.nrc.ca (localhost [127.0.0.1]) by ned.sims.nrc.ca (8.12.6/8.12.6/SuSE Linux 0.6) with ESMTP id h43Jnvbb008685; Sat, 3 May 2003 15:49:58 -0400 Received: from localhost (ps@localhost) by ned.sims.nrc.ca (8.12.6/8.12.6/Submit) with ESMTP id h43Jnvdb008682; Sat, 3 May 2003 15:49:57 -0400 Date: Sat, 3 May 2003 15:49:57 -0400 (EDT) From: Serguei Patchkovskii Reply-To: Serguei.Patchkovskii@nrc.ca To: jmmckel@attglobal.net cc: Serguei.Patchkovskii@nrc.ca, , Subject: Re: CCL:Intel Linux compiler 7.0 In-Reply-To: <3EB28027.763C0394@attglobal.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Virus-Scanned: by AMaViS - amavis-milter (http://www.amavis.org/) On Fri, 2 May 2003 jmmckel@attglobal.net wrote: > O1 even when using -malign-double, to give executables that work properly. Using > -malign-double gets you to O1, but no further... > > Compiling with the Intel7.0 compiler with -align and no optimization and > the linking with the proper libraries does not give a properly functioning > executable. "Proper libraries" is a bit vague - getting BLAS and LAPACK libraries, which work correctly for all corner cases, is tricky. Quite a few of the precompiled libraries, floating out there on the net (or even distributed with professional Fortran compilers) care more about the speed, than about getting correct results. Just for fun, try running the test suite(s), distributed with LAPACK 3.0 source, and see how many failures you'll get (I am yet to see any LAPACK/BLAS combo, including those I compile myself, which passes all the tests). MKL 5.2SP1 is one of the better ones in this respect. Some things to try, in the order of increasing speed impact: -pc64 -prec_div -fp_port -mp1 -mp Needing one of these flags does not necessarily mean that there is anything wrong with the program (or the compiler) - only that it is numerically sensitive. You can also try: -zero -save -nopad -common_args -r8 If either one of these flags is needed to get program to work, the code is likely buggy, and should be fixed if possible/practical. It may also be a good idea to run the program with: -O0 -C -d2 Unless the code is -extremely- well written (or has been through ifc -C before), be prepared for many, many surprizes. Best regards, Serguei --- Dr. Serguei Patchkovskii Tel: +1-(613)-990-0945 Fax: +1-(613)-947-2838 E-mail: Serguei.Patchkovskii@nrc.ca Coordinator of Modelling Software Theory and Computation Group Steacie Institute for Molecular Sciences National Research Council Canada Room 2011, 100 Sussex Drive Ottawa, Ontario K1A 0R6 Canada From chemistry-request@server.ccl.net Sat May 3 09:46:44 2003 Received: from ned.sims.nrc.ca ([132.246.109.100]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id h43DkiV00872 for ; Sat, 3 May 2003 09:46:44 -0400 Received: from ned.sims.nrc.ca (localhost [127.0.0.1]) by ned.sims.nrc.ca (8.12.6/8.12.6/SuSE Linux 0.6) with ESMTP id h43Djcbb006267; Sat, 3 May 2003 09:45:38 -0400 Received: from localhost (ps@localhost) by ned.sims.nrc.ca (8.12.6/8.12.6/Submit) with ESMTP id h43Djc6A006264; Sat, 3 May 2003 09:45:38 -0400 Date: Sat, 3 May 2003 09:45:38 -0400 (EDT) From: Serguei Patchkovskii Reply-To: Serguei.Patchkovskii@nrc.ca To: ericyan@epsilon.ccqc.uga.edu cc: jmmckel@attglobal.net, Subject: Re: CCL:Intel Linux compiler 7.0 In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Virus-Scanned: by AMaViS - amavis-milter (http://www.amavis.org/) Actually, the problem arises because John is trying to link against lapack and blas libraries, compiled with g77 (GNU Fortran-77 compiler). Because some of these subroutines use Fortran I/O statements, they need to be linked with g77 I/O library - which is provided by g2c. An even better solution would be to link with Intel MKL (Math kernel Library), which provides highly optimized versions of BLAS and LAPACK (and a few other things). MKL can be downloaded from: http://developer.intel.com/software/products/mkl/mkl52/ Serguei On Thu, 1 May 2003 ericyan@epsilon.ccqc.uga.edu wrote: > That seems to be a problem when linking both C and Fortran object files. > My way (though not the best one) of solving this problem is by also > linking with GCC's g2c library: > -L/usr/lib/gcc-lib/i386-redhat-linux/2.96 -lg2c > The actual linking path may be different in your machine > Good luck. > > Eric > > -------------------------------------------------------------------- > Eric Yan | phone: 706 542-4022 > Ctr. for Comp. Quantum Chem. | fax: 706 542-0406 > University of Georgia | e-mail: ericyan@ccqc.uga.edu > Athens, GA 30602-2556 | http://epsilon.ccqc.uga.edu > -------------------------------------------------------------------- > > > On Tue, 29 Apr 2003 jmmckel@attglobal.net wrote: > > > Greetings! > > > > I have successfully compiled a version of MOPAC using the unsupported > > Intel Linux fortran compiler Version7.0 [I need F90 for dynamic memory > > allocation.] Upon linking the code against the Intel libraries as well > > as libf77blas,libatlas, liblapack and lapack_linux libraries from > > netlib.org I get the following complaint: > > > > for dlamch.o s_wsfe, do_fio, and e_wsfe are not resolved. > > > > Comments and suggestions are most welcome... > > > > many thanks! > > > > John McKelvey > > --- Dr. Serguei Patchkovskii Tel: +1-(613)-990-0945 Fax: +1-(613)-947-2838 E-mail: Serguei.Patchkovskii@nrc.ca Coordinator of Modelling Software Theory and Computation Group Steacie Institute for Molecular Sciences National Research Council Canada Room 2011, 100 Sussex Drive Ottawa, Ontario K1A 0R6 Canada From chemistry-request@server.ccl.net Sat May 3 15:17:28 2003 Received: from prserv.net ([32.97.166.51]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id h43JHSV00657 for ; Sat, 3 May 2003 15:17:28 -0400 Received: from attglobal.net (dhcp065-029-075-111.indy.rr.com[65.29.75.111]) by prserv.net (asmtp1) with SMTP id <2003050319171725101516i6e> (Authid: jmmckel); Sat, 3 May 2003 19:17:18 +0000 Message-ID: <3EB28027.763C0394@attglobal.net> Date: Fri, 02 May 2003 14:26:47 +0000 From: jmmckel@attglobal.net Reply-To: jmmckel@attglobal.net X-Mailer: Mozilla 4.51 [en]C-CCK-MCD (WinNT; I) X-Accept-Language: en MIME-Version: 1.0 To: Serguei.Patchkovskii@nrc.ca CC: ericyan@epsilon.ccqc.uga.edu, chemistry@ccl.net Subject: Re: CCL:Intel Linux compiler 7.0 References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Folks, I agree with Serguei. But even with proper linking problems do not go away. On working with a good version of MOPAC93 directly from Jimmy Stewart and the compiler furnished with RH7.1 it is not possible to get optimizations past O1 even when using -malign-double, to give executables that work properly. Using -malign-double gets you to O1, but no further... Compiling with the Intel7.0 compiler with -align and no optimization and the linking with the proper libraries does not give a properly functioning executable. With both compilers I have found that it is one thing to compile routines seperately, and a major headache to compile all the subroutines in a single file..... BTW, I have MOPAC93 and MOPAC97 working just fine using the Digital Visual compilers with all possible optimization turned on under Windows. Seems to me there's something to be said for running MPI-PRO under Windows for coarse grain parallel work... Suggestions are welcome... Thanks! John McKelvey Serguei Patchkovskii wrote: > Actually, the problem arises because John is trying to link against lapack > and blas libraries, compiled with g77 (GNU Fortran-77 compiler). Because > some of these subroutines use Fortran I/O statements, they need to be linked > with g77 I/O library - which is provided by g2c. > > An even better solution would be to link with Intel MKL (Math kernel Library), > which provides highly optimized versions of BLAS and LAPACK (and a few other > things). MKL can be downloaded from: > > http://developer.intel.com/software/products/mkl/mkl52/ > > Serguei > > On Thu, 1 May 2003 ericyan@epsilon.ccqc.uga.edu wrote: > > > That seems to be a problem when linking both C and Fortran object files. > > My way (though not the best one) of solving this problem is by also > > linking with GCC's g2c library: > > -L/usr/lib/gcc-lib/i386-redhat-linux/2.96 -lg2c > > The actual linking path may be different in your machine > > Good luck. > > > > Eric > > > > -------------------------------------------------------------------- > > Eric Yan | phone: 706 542-4022 > > Ctr. for Comp. Quantum Chem. | fax: 706 542-0406 > > University of Georgia | e-mail: ericyan@ccqc.uga.edu > > Athens, GA 30602-2556 | http://epsilon.ccqc.uga.edu > > -------------------------------------------------------------------- > > > > > > On Tue, 29 Apr 2003 jmmckel@attglobal.net wrote: > > > > > Greetings! > > > > > > I have successfully compiled a version of MOPAC using the unsupported > > > Intel Linux fortran compiler Version7.0 [I need F90 for dynamic memory > > > allocation.] Upon linking the code against the Intel libraries as well > > > as libf77blas,libatlas, liblapack and lapack_linux libraries from > > > netlib.org I get the following complaint: > > > > > > for dlamch.o s_wsfe, do_fio, and e_wsfe are not resolved. > > > > > > Comments and suggestions are most welcome... > > > > > > many thanks! > > > > > > John McKelvey > > > > > --- > Dr. Serguei Patchkovskii > > Tel: +1-(613)-990-0945 > Fax: +1-(613)-947-2838 > E-mail: Serguei.Patchkovskii@nrc.ca > > Coordinator of Modelling Software > Theory and Computation Group > Steacie Institute for Molecular Sciences > National Research Council Canada > Room 2011, 100 Sussex Drive > Ottawa, Ontario > K1A 0R6 Canada