From STINCH%IMISIAM.BITNET@phem3.acs.ohio-state.edu Thu Jul 15 10:47:45 1993 Date: Thu, 15 Jul 93 15:45:03 ITA From: Davide Proserpio Subject: Tektronik Emulator To: CHEMISTRY@ccl.net Message-Id: <01H0KGATRYDU9H1Y7T@phem3.acs.ohio-state.edu> Dear Netters I am sure that my inquire has been posted in the past, but I hope someone can help me anyway. I am looking for commercial and/or public domain programs to emulate a tektron ik terminal on a PC and on SiliconGraphics. I have hear about EMU-TEK. Any suggestion with addresses and indicative price will be very welcome thank you all Davide P.S. please reply directly to me at (do not use reply, please) davide@stinch6.csrsrc.mi.cnr.it **************************************************************** Davide M. Proserpio UNIVERSITA' DI MILANO ISTITUTO DI CHIMICA STRUTTURISTICA INORGANICA VIA VENEZIAN 21, 20133 MILANO - ITALY PHONE : 39-2-70635120 FAX : 39-2-70635288 E-MAIL: DAVIDE@STINCH6.CSRSRC.MI.CNR.IT **************************************************************** From paolo@gandalf.ciam.unibo.it Thu Jul 15 17:23:18 1993 Date: Thu, 15 Jul 1993 16:23:18 +0100 From: paolo@gandalf.ciam.unibo.it (Paolo Celani) Message-Id: <9307151523.AA13961@gandalf.ciam.unibo.it> To: chemistry@ccl.net Subject: Link_prg_with_GAUSSIAN_subr I have a problem with GAUSSIAN 92. I write a little program for my use ( ict.f ) which calls the GAUSSIAN routine 'ZtoC' in util.a . I build executable like this: # xlf -oict.exe ict.f util.a Computer answers: # xlf -oict.exe ict.f util.a ** ict === End of Compilation 1 === 1501-510 Compilation successful for file ict.f. 0706-317 ERROR: Unresolved or undefined symbols detected: Symbols in error (followed by references) are dumped to the load map. The -bloadmap: option will create a load map. .ztoc # What is wrong ? How can i link GAUSSIAN routines with my programs? I'm running GAUSSIAN92 revB on IBM RISC6000 (580) with AIX 3.2.3 extended. Reply directly to me. Thank you in advance Paolo Paolo Celani Universita' di Bologna Italia e-mail: paolo@gandalf.ciam.unibo.it Load map ------------------------------------------ (ld): halt 4 (ld): insert /lib/crt0.o (ld): insert ict.o (ld): insert util.a (ld): insert /usr/lib/libxlf.a 0706-406 IMPORT_FILE: 236 symbols imported. (ld): insert /usr/lib/libm.a (ld): insert /usr/lib/libc.a 0706-406 IMPORT_FILE: 14 symbols imported. 0706-406 IMPORT_FILE: 2569 symbols imported. 0706-406 IMPORT_FILE: 2 symbols imported. (ld): resolve rpolx.f from util.a[rpolx.o] Replaced by symbol from: bdrys.f from util.a[bdrys.o] rpolx.f from util.a[rpolx.o] Replaced by symbol from: bdrys.f from util.a[bdrys.o] rysset.f from util.a[rysset.o] Replaced by symbol from: bdrys.f from util.a[bdrys.o] 0706-224 WARNING: Duplicate symbol 'times_'. trani1.f from util.a[trani1.o] Replaced by symbol from: rs6kdum.f from util.a[rs6kdum.o] vrpolx.f from util.a[vrpolx.o] Replaced by symbol from: bdrys.f from util.a[bdrys.o] vrpolx.f from util.a[vrpolx.o] Replaced by symbol from: bdrys.f from util.a[bdrys.o] _qgamma.f from /usr/lib/libm.a[_qgamma.o] Replaced by symbol from: lgamma.c from /usr/lib/libm.a[lgamma.o] 0706-224 WARNING: Duplicate symbol '._xlqadd'. _xlqadd.f from /usr/lib/libc.a[_xlqadd.o] Replaced by symbol from: _xlqadd.f from /usr/lib/libxlf.a[_xlqadd.o] 0706-224 WARNING: Duplicate symbol '._xlqdiv'. _xlqdiv.f from /usr/lib/libc.a[_xlqdiv.o] Replaced by symbol from: _xlqdiv.f from /usr/lib/libxlf.a[_xlqdiv.o] 0706-224 WARNING: Duplicate symbol '._xlqmul'. _xlqmul.f from /usr/lib/libc.a[_xlqmul.o] Replaced by symbol from: _xlqmul.f from /usr/lib/libxlf.a[_xlqmul.o] 0706-224 WARNING: Duplicate symbol '._xlqsub'. _xlqsub.f from /usr/lib/libc.a[_xlqsub.o] Replaced by symbol from: _xlqsub.f from /usr/lib/libxlf.a[_xlqsub.o] 0706-224 WARNING: Duplicate symbol '.logb'. *NO TAG* from /usr/lib/libc.a[logb.o] Replaced by symbol from: *NO TAG* from /usr/lib/libm.a[logb.o] 0706-224 WARNING: Duplicate symbol '.scalb'. scalb.c from /usr/lib/libc.a[scalb.o] Replaced by symbol from: scalb.c from /usr/lib/libm.a[scalb.o] 0706-224 WARNING: Duplicate symbol '.finite'. *NO TAG* from /usr/lib/libc.a[finite.o] Replaced by symbol from: *NO TAG* from /usr/lib/libm.a[finite.o] 0706-222 WARNING: Import version of 'abort' replaced by import definition. .main from GENERATED SYMBOL Replaced by symbol from: .main from GENERATED SYMBOL 0706-222 WARNING: Import version of '.abort' replaced by import definition. .main from GENERATED SYMBOL Replaced by symbol from: .main from GENERATED SYMBOL 0706-196 RESOLVE: Kept 15171 of 27215 symbols. (ld): gc 0706-188 Number of symbols garbage collected: 15141 of 15171 (ld): comprld 0706-180 COMPRLD: Kept 55 of 55 relocation entries. (ld): mismatch 0706-294 No type mismatches detected. (ld): er full 0706-317 ERROR: Unresolved or undefined symbols detected: Symbols in error (followed by references) are dumped to the load map. The -bloadmap: option will create a load map. 0706-290 The following symbols were left unresolved: 0706-350 Name Address Type Stg Information Filename Hash Value Imp/Exp/Ent [10].ztoc 0000031E ER PR ict.f from ict.o [11]<.main> 000007BA AL.26( [10].ztoc) ict.f from ict.o 0706-120 The return code is 8 From dvsg23424@ggr.co.uk Thu Jul 15 11:47:33 1993 From: dvsg23424@ggr.co.uk Message-Id: <9307151526.AA29487@mailhub.ggr.co.uk> Date: 15 Jul 93 16:17:00 BST Subject: MOPAC MEP charges To: chemistry@ccl.net Dear All, I am having trouble with the ESP routines in MOPAC6.0. I am studying some anionic heterocycles, but the MEP charges I obtain appear to be random numbers. SPARTAN gives much more "realistic" results. I have all the corrections w.r.t. MULLIKEN and ESP keywords etc. This problem is not related to MULLIKEN at all. Rather, it appears to be related to the CHARGE keyword(?) Neutral species are fine. I have tried varying the surface parameters etc, but the results are still garbage. Can anyone put me out of my misery?? With thanks, Darren Green. From jjp1@tams.chem.buffalo.edu Thu Jul 15 07:50:31 1993 Date: Thu, 15 Jul 93 11:50:31 EDT From: jjp1@tams.chem.buffalo.edu (Joe Pavelites) Message-Id: <9307151550.AA12854@tams.chem.buffalo.edu> To: CHEMISTRY@ccl.net Subject: zmat->cart. coord wrap up. Dear Netters: Here is a summary of a summary that I received from FREDVC@esds01.es.dupont.com. Thanks for all responses! >from Mark A. Zottola: Gaussian 92 has a utility Newzmat that converts PDB files to z-matrix form. >from Leslie Glasser (009LGZS@WITSVMA.WITS.AC.ZA): Prof. Glasser has written a MathCAD file that converts crystal to cartesian coordinates or internal coordinates. You need MathCAD (either DOS or Windows) to use the script. >from Gregory L. Durst: QCPE has a program EXTOIN (QCMP070) for PC's that converts MM2 external cartesian coordinates into a z-matrix. It is in Fortran. >from James Stewart: MOPAC has a subroutine XYZINT that can form the core of a program to perform this conversion. Feel free to use it. in addition, Zoran Zdravkovski reponds: MOPAC does have an option to get cartesian coordinates from Z- matrix, by using the keywords XYZ 0SCF. AMPAC probably has the same option, and you could probaly easily find the source of earlier versions of either somewhere on the network. Try using, for example, archie -c MOPAC (or AMPAC) to locate an FTP site with the programs. Joe Pavelites SUNYAB Chem From KUCHARSKI@itots1.ch-pwr.wroc.edu.pl Thu Jul 15 13:47:30 1993 To: chemistry@ccl.net From: "Stanislaw Kucharski" Date: 15 Jul 93 19:14:04 MET Subject: info Message-Id: <1F4E5B7883@itots1.ch-pwr.wroc.edu.pl> help info Prof.Dr.Stanislaw Kucharski Institute of Organic and Polymer Technology Wroclaw Technical University 50-370 Wroclaw,Poland Tel. (048) 71 202862; (048) 71 202426 Fax (048) 71 203503 From blonski@alumina.rutgers.edu Thu Jul 15 09:21:43 1993 Date: Thu, 15 Jul 93 13:21:43 EDT From: blonski@alumina.rutgers.edu (Slawomir Blonski) Message-Id: <9307151721.AA11269@alumina.rutgers.edu> To: chemistry@ccl.net Subject: IBM vs SGI vs HP There was recently a discussion about differences in performance between various platforms. Therefore, I think that I can post this question and ask for help. I have compiled and run the same molecular dynamics program on three machines. What is surprising to me, is the weak performance of the IBM workstation. I thought that it should be comparable with the HP. Does somebody have any idea what is the reason for that ? Here is the test and the results: time knitesp.x < datak > outk & HP 9000/750 54.1u 0.2s 1:48 50% 54.1u 0.2s 1:48 50% 54.1u 0.2s 1:47 50% SGI Indigo 2 (100 MHz, 32 MB) 78.6u 14.4s 1:34 98% 77.8u 14.3s 1:33 98% 77.8u 14.8s 1:33 98% IBM RS/6000 550 (64 MB) 112.7u 0.50s 1:55 97% 155+1829k 0+0io 93pf+0w 111.93u 0.54s 1:54 97% 155+1828k 0+0io 92pf+0w 112.11u 0.48s 1:54 98% 155+1829k 0+0io 92pf+0w Slawomir Blonski From fant@hungryjoe.plk.af.mil Thu Jul 15 05:54:30 1993 From: fant@hungryjoe.plk.af.mil (Andrew Fant) Message-Id: <9307151854.AA23887@hungryjoe.plk.af.mil> Subject: Re: Link_prg_with_GAUSSIAN_subr To: paolo@gandalf.ciam.unibo.it (Paolo Celani) Date: Thu, 15 Jul 1993 12:54:30 -0700 (MDT) Paolo: You don't want to just put util.a on the command line. Try compiling by doing: xlf -oict.exe -L/usr/local/g92/util -l util.a ict.f You will want to replace /usr/local/g92/util with the path to the util.a library on your system. I hope this helps. Andy Andrew D. Fant Computational Chemistry Systems Analyst TAI/UTS PLSC-Kirtland AFB fant@hungryjoe.plk.af.mil (505)266-1957 Humiston's Law: When you are surrounded by alligators, it is hard to remember your original intention was to drain the swamp. From modesto@doctor.chem.yale.edu Thu Jul 15 11:58:08 1993 Date: Thu, 15 Jul 93 15:58:08 EDT From: Modesto Orozco Message-Id: <9307151958.AA10147@doctor.chem.yale.edu> To: chemistry@ccl.net Subject: Mathematica Hi I would be very happy to know the disk and memory requirements to use Mathematica on a HP-9000 workstation. Thanks,... Dr. Modesto Orozco Department of Chemistry Yale University Modesto@doctor.chem.yale.edu (203)432-6288 From m10!frisch@uunet.UU.NET Thu Jul 15 14:29:33 1993 Message-Id: <9307152231.AA15208@relay1.UU.NET> Date: Thu, 15 Jul 93 18:29:33 EDT From: m10!frisch@uunet.UU.NET (Michael Frisch) Subject: Linking with Gaussian subroutines To: chemistry@ccl.net On the RS/6000, Gaussian 92 is built using the -qextname switch, which causes the link-time names of Fortran subroutines to include a trailing underscore. Thus "Subroutine ZToC" looks like ztoc_ to the linker. To link with Gaussian routines, you should either: a) Use the same switch: xlf -qextname -o myexe mysource.f util.a or b) include the underscore in the file: Program MyProgram ... Call ZToC_ (...) ... xlf -o myexe mysource.f util.a (Of course, you probably want a path name in front of util.a, if you aren't in the g92 directory. If your account is set up to run Gaussian 92 in the usual way, $GAUSS_EXEDIR/util.a will work.) Mike Frisch ------- From rec@california.sandia.gov Thu Jul 15 10:21:18 1993 Date: Thu, 15 Jul 93 17:21:18 -0700 From: rec@ca.sandia.gov (Ray Cline) Message-Id: <9307160021.AA27917@california.sandia.gov> To: blonski@alumina.rutgers.edu Subject: IBM vs SGI vs HP We have recently completed a first cut of testing a distributed MD application on sets of HP, SGI, and IBM workstations. As part of that test we ran single processor jobs and our results were similar to those posted by you. I have also run linpack, slalom, and livermore loops benchmarks on all of these machines. In the case of these "standard" benchmarks the IBM (350H) out performs the SGI (R4K Indigo) by about 50% (eg mflops ratio IBM:SGI 1.5:1). The only explanation that I can resonably justify is that the "standard" benchmarks tend to be relatively "easy" on memory caches while the MD codes tend to be "cache busters" (long loops). The IBM are notorious for dying when they run "out of cache" while the MIPS chips tend to degrade gracefully (and have a 512kb cache). So, it just shows that the "standard" becnhmarks and vendors numbers are not representative of all applications. Things also get complicated when you go distributed and the SGI FDDI speed blow everyone away (so 4 hp 735's are only a little faster than 4 SGI R4K indigos, and 8 hp 735's are slower than 8 SGI's), but that is another story.... (and our results are only preliminary). Ray Raymond E. Cline, Jr. Department Manager, Distributed Computing Phone: (510) 294-1395 Sandia National Laboratories FAX: (510) 294-1225 Box 969, Org 1952 E-mail: rec@ca.sandia.gov Livermore, CA 94551 *******************************************************************************