From harry@ccl.net Mon Nov 29 02:51:31 1993 Received: from wrzx01.rz.uni-wuerzburg.de for harry@ccl.net by www.ccl.net (8.6.4/930601.1506) id CAA14079; Mon, 29 Nov 1993 02:37:52 -0500 Received: from wpcs01.phys-chemie.uni-wuerzburg.dbp.de (wpcx11.phys-chemie.uni-wuerzburg.de) by wrzx01.rz.uni-wuerzburg.de (4.1/uniwue-M-3.1) id AA12273; Mon, 29 Nov 93 08:37:42 +0100 Received: by wpcs01.phys-chemie.uni-wuerzburg.dbp.de (4.1/uniwue-C-1.7) id AA19267; Mon, 29 Nov 93 08:37:41 +0100 Date: Mon, 29 Nov 93 08:37:41 +0100 From: Harald Lanig Message-Id: <9311290737.AA19267@wpcs01.phys-chemie.uni-wuerzburg.dbp.de> To: chemistry@ccl.net Subject: domain error on IBM-PC Dear netters! I tried to compile react2 by Mike Whitbeck, a program to simulate chemical reaction kinetics on a 486DX33 using Borland C 2.0 The package is available on our kekule-server. The compilation is done, but the program runs into a floating-point domain error. A sample program to test the lsoda integrator (also in the package) fails with the same error. I think the problem has todo with machine dependent precision in variable-type specification. Has anyone out there solved this problem? Possibly someone is using the lsoda routines on a IBM PC. (React2 runs perfectly on a SUN Sparc 2!) Perhaps, DOS executables are alrady available? Thanks in advance for any help or comment! -Harry --------------------------------------------------------------- Harald Lanig Institut fuer Physikalische Chemie Marcusstr. 9-11 phone: +49(0)931-31-510 97070 Wuerzburg fax: +49(0)931-57-849 Germany email: harry@phys-chemie.uni-wuerzburg.de --------------------------------------------------------------- From NMUELLER@EDVZ.UNI-LINZ.AC.AT Mon Nov 29 06:51:38 1993 Received: from ALIJKU11.EDVZ.Uni-Linz.AC.AT for NMUELLER@EDVZ.UNI-LINZ.AC.AT by www.ccl.net (8.6.4/930601.1506) id GAA15026; Mon, 29 Nov 1993 06:29:11 -0500 Message-Id: <199311291129.GAA15026@www.ccl.net> Received: from [140.78.98.5] by ALIJKU11.EDVZ.Uni-Linz.AC.AT (IBM VM SMTP V2R2) with TCP; Mon, 29 Nov 93 12:29:00 CET X-Sender: K360171@ALIJKU11.EDVZ.UNI-LINZ.AC.AT Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Mon, 29 Nov 1993 12:29:08 +0100 To: chemistry@ccl.net From: NMUELLER@EDVZ.UNI-LINZ.AC.AT (Norbert Mueller) Subject: Summary: HyperChem (SGI) replacements Approximately 3 weeks ago I posted a question about possible replacements for similarily priced HyperChem on the SGI platform for use in a teaching lab. This has become necessary for us as AutoDesk has no (ooficial) plans for upgrading HyperChem 2.0 for SGI on that platform. Here comes the promised summary of responses: Square brackets [] indicate comments inserted by myself. Dear Sir, I think that AMPAC 4.5 might be a good choice. It does not do Molecular Mechanics, but is an excellent semiempirical package. The graphical user interface is easy and intuitive, based on frag- mentary construction. I'll be happy to forward a brochure and other information if you would like. Andy Holder =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= DR. ANDREW HOLDER Assistant Professor of Computational/Organic Chemistry Department of Chemistry || BITNET Addr: AHOLDER@UMKCVAX1 University of Missouri - Kansas City || Internet Addr: aholder@vax1.umkc.edu Spencer Chemistry, Room 315 || Phone Number: (816) 235-2293 Kansas City, Missouri 64110 || FAX Number: (816) 235-1717 =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= We are currently using SYBYL from Tripos Associates in the teaching role that you described in your message. There are various options available for this package which make it quite flexible for use both in and out of the classroom. In our case, SYBYL is currently installed on an IRIS INDIGOII system which is then linked to lecture stations and student workstations using both the NITRO terminal package (Mac II and PC)and the XDEV X11 interface. Our favorable experience with this arrangement has stimulated our administration into funding a broader acquisition of SYBYL for use on our entire campus network of over 600 RS/6000 and Sun Sparc student workstations. While SYBYL is somewhat more expensive than Hyperchem, Tripos offers quite attractive academic pricing. Curt Breneman Asst. Professor of Chemistry Rensselaer Polytechnic Institute Troy, NY 12180. breneman@xray.chem.rpi.edu _______________________________________________________________________ I suggest that you look into Sybyl. It does quite well with interfaces to standard QCPE programs and visualization. I have no clue as to the pricing structure. Yvonne Martin Abbott Laboratories _______________________________________________________________________ While I don't know the price or HC for the SGI, the academic price for Spartan (single copy) is $1500 US plus shipping - there are smaller costs for additional copies up to a site license for, I think, $5000 US plus shipping (single tape for all SGI). We're finishing up a pure-X version that will support Macs and PCs using 8-bit color X-servers (with ethernet/TCP). We have various modules (QM, MM, graphics, etc), and connect to G92 (V3.0) and Allinger's MM3 (V3.1). If you want, I'll have somebody send you additional information (or I'll try to answer what questions you might raise). Joe Leonard jle@world.std.com ________________________________________________________________________ From: Jeffrey J Gosper We use ChemX (by Chemical design Limited) for this purpose. ________________________________________________________________________ From: km%biosym.uucp@Germany.EU.net [Insight Express (Biosym) is suggested by Klaus May, original mail was in German] ________________________________________________________________________ For the problems you describe (molecular mecahnics and semi-empirical) I suggest Spartan. It has a good graphics front end, gives good choice of surfaces (HOMO's, LUMO's, esp, etc), and can also do ab initio studies. It is weak on the biomolecules. Although they pormise to improve that, HC has more options there. I also like the user interface on Spartan better. LabVision from Tripos does a reasonable job with biomolecules, but you are pretty much limited to the SYBYL force field. I do not like the user interface there as well either. Phil Bays Saint Mary's College Notre Dame IN 46556 pbays@saintmarys.edu bays@indigo.saintmarys.edu ________________________________________________________________________ From: fagerbur@Kodak.COM Herr Dr. Mueller! I am sure you will get several responses like mine for Hyperchem replacement on the SGI. The program we have that runs very well is SPARTAN from Warren Hehre at Wavefunction in Irvine Calif. The graphical interface is really well done and the builder "knows" lots of organic chemistry. We have found it easy to use. It was also inexpensive - I think it is $1500 for academics. Note: the above is the personal opinion of the author and not that of Eastman Chemical Company. ________________________________________________________________________ From: shenkin@still3.chem.columbia.edu (Peter Shenkin) I don't know what Hyperchem used to cost for SGI, but here is the blurb on MacroModel: ========================================================================== INTRODUCTION TO MACROMODEL 4.0 Sep 20 1993 [[ You may obtain this document remotely by means of "finger mmod@still3.chem.columbia.edu". ]] ________________________________________________________________________ Norbert- An excellent choice for a replacement for Hyperchem on the SGI platform would be MacroModel, written by Clark Still at Columbia University. The porgram is available for Academics for the low cost of $750. For information on ordering, constact Joan HOrgan at 212-854-2577 or e-mail sl1joan@cuchmc.chem.columbia.edu. The program supports MM2, MM#, AMBER and OPLS force fields; Metropolis\ MOte Carlo structure searching and has a very nice graphical user interface. Rachelle Bienstock (rachelle@picard.niehs.nih.gov) ________________________________________________________________________ From: "CBAS25 ::P_BLADON ::CBAS25" Dear Dr. Mueller, I have seen your recent posting on the network regarding the demise of Hyperchem. I can offer you the package INTERCHEM which might file you needs. I have sent a brief description of the program(s) in a second mail message. I can send you a more detailed description if you are interested. The package is available either through QCPE or from me direct for the price of $400 (US) or the equivalent in other currencies. This is for a site license. Because the package is distributed as source code, you would neeed to have available (on one machine at least) Fortran and C compilers. Yours sincerely Peter Bladon ________________________________________________________________________ The program you should get is SPARTAN: Wavefunction Inc. 18401 Von Karman Suite 370 Irvine, CA 92715 USA (714)955-2120, FAX (714)955-2118 or jle@world.std.com (Joe Lennard) SPARTAN is an INTEGRATED MMX, semiemp, AND ab initio package with excellent graphics. elewars@trentu.ca Errol Lewars Chem Dept Trent Uni Peterborough Ontario Canada K9J 7B8 ============================================= _______________________________________________________________________ I have recently purchased a SGI Iris Indigo 4000, with Entry graphics, and am running a superb computational package on it. The program is far more powerful than HC, and it's considerably easier to use. The software is Spartan, version 3.0, available from Wavefunction, Inc. 18401 Von Karman, Suite 370 Irvine, CA 92715 USA Phone (714) 955-2120 Academic pricing of the program is $1500 (US), so it's probably more expensive than HyperChem, but it is well worth the money. Spartan can perform MM calculations using MM2, MM3, and SYBYL force fields; Semiempirical calculations using MNDO, AM1, and PM3. It can also do some CI. SPARTAN also can do ab initio calculations. It has STO-3g, 3-21G, 6-31G, and 6-311G basis sets, as well as heavy atom polarization functions (*). You could also input your own basis set functions. SPARTAN can calculate UV spectra. It can also calculate the normal vibrational modes of a molecule, and display each one, animated! This should be especially useful in teaching IR spectroscopy. I calculated the vibrational modes of phenol in less than five minutes. This included building the molecule, performing the semiempirical calculation, and calculating all of the vibrations. Spartan can also calculate transition-state geometries and the energies of those states. This is my particular interest, but I haven't found the time to perform one of those calculations; it's a very busy semester. The program is very easy to use. I'd guess students would be performing rather high-level calculations within thirty minutes. They'll be able to build molecules within five minutes. It has a superb graphically-based interface, which builds molecules and performs some preliminary geometry optimization. It produces genuine three-dimensional plots of MOs. electron density, spin density, etc., and will display one of those properties mapped on another. I have been very pleased with SPARTAN, and with the service Wavefunction has given. They are a very professional, personable company. I hope this has been useful to you. By the way, in spite of the tone of this message, I have no financial stake in SPARTAN; I'm just an enthusiastic fan of the package. Regards, Francis M. Klein Creighton University Omaha, NE, 68178 USA fklein@creighton.edu _______________________________________________________________________ Check out SPARTAN, from Wavefunction, Inc., 18401 Von Karman Avenue, Suite 370, Irvine, CA 92715. Telephone: (714)955-2120. The software is as easy to use as HyperChem and more versatile. They supply a several hundred page manual of experiments for undergraduates in organic chemistry using Spartan. Susan C. Jackels Department of Chemistry Wake Forest University Winston-Salem, NC 27109 Phone: (919)759-5514 FAX: (919)759-4656 Internet: sjackels@ac.wfunet.wfu.edu On sabbatical for 1993-1994 at: Department of Medicinal Chemistry 308 Harvard Street S.E. University of Minnesota Minneapolis, MN 55455 Phone: (612)626-4429 _______________________________________________________________________ _______________________________________________________________________ [FRom what I have read here I will gather furtehr info on Spartan and MacroModel. InterChem sounds very good price/performance wise but its requirement of a 24-bit z-buffer eliminates our entry level graphics machines] -- Norbert Mueller Institut fuer Chemie, Johannes Kepler Universitaet A-4040 Linz, Austria >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> IMPORTANT NOTICE <<<<<<<<<<<<<<<<<<<<<<<<<< Superior powers have decided to change our e-mail addresses again. So please take note that some time between now and March 1994 my e-mail address changes to NMUELLER@JKU.UNI-LINZ.AC.AT Sorry for any inconvenience and delays this may cause. >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> IMPORTANT NOTICE <<<<<<<<<<<<<<<<<<<<<<<<<< e-mail: NMUELLER@edvz.uni-linz.ac.at - expires in 1994 NMUELLER@jk.uni-linz.ac.at - use starting 1994 (preferred) Norbert@soft.uni-linz.ac.at NorbertM,ALLinz - AmdaLink NorbertM@AMDA.UNI-LINZ.AC.AT - FirstClass AmdaLink Internet Gateway From sauer@organik.uni-erlangen.de Mon Nov 29 07:51:39 1993 Received: from faui45.informatik.uni-erlangen.de for sauer@organik.uni-erlangen.de by www.ccl.net (8.6.4/930601.1506) id GAA15122; Mon, 29 Nov 1993 06:54:17 -0500 Received: from derioc1.organik.uni-erlangen.de by uni-erlangen.de with SMTP; id AA16968 (5.65c-6/7.3v-FAU); Mon, 29 Nov 1993 12:53:59 +0100 Received: by organik.uni-erlangen.de; id AA10076 (5.64/7.3n-FAU); Mon, 29 Nov 93 12:53:41 +0100 From: Wolfgang Sauer Message-Id: <9311291153.AA10076@derioc1.organik.uni-erlangen.de> Subject: FELIX + JEOL ? To: CHEMISTRY@ccl.net (CompChem MailExploder) Date: Mon, 29 Nov 93 12:53:39 MET X-Mailer: ELM [version 2.3 PL11] Dear netters, I'd like to get in touch with people using FELIX on data produced on JEOL hardware (esp. 2D {3D ?}), as this does not seem to be trivial. Any hints, tricks, recommendations ??? Thanks in advance. Wolfgang Sauer -- +============================================================================+ ! Please note the change of all German postcodes since 1 July 1993 ! +======================================+=====================================+ | Wolfgang Sauer | | | Institut fuer Organische Chemie I | "You have to be three standard | | Henkestr. 42, D-91054 Erlangen, FRG | deviations away from a normal | | | personality to like UNIX." | | sauer@organik.uni-erlangen.de | | | Tel.: 49/0 - 9131 - 85 - 2952 | Morris Jones (C&T) | | Fax: - 9132 | | +======================================+=====================================+ From BOYD_DONALD_B@Lilly.com Mon Nov 29 10:52:52 1993 Received: from Lilly.com for BOYD_DONALD_B@Lilly.com by www.ccl.net (8.6.4/930601.1506) id KAA16726; Mon, 29 Nov 1993 10:06:42 -0500 From: Received: from DECNET-MAIL (QSAR@MCVAX0) by INET.D48.LILLY.COM (PMDF V4.2-13 #2311) id <01H5VTUG76RK0000XX@INET.D48.LILLY.COM>; Mon, 29 Nov 1993 10:06:07 EST Date: Mon, 29 Nov 1993 10:06:07 -0500 (EST) Subject: * book announcement * To: chemistry@ccl.net Message-id: <01H5VTUG7Q1U0000XX@INET.D48.LILLY.COM> X-VMS-To: INT::"chemistry@ccl.net" X-VMS-Cc: ME MIME-version: 1.0 Content-transfer-encoding: 7BIT A new volume in the book series REVIEWS IN COMPUTATIONAL CHEMISTRY has just been published. Volume V features 458 pages of tutorials and reviews on the following topics: The Development of Computational Chemistry in the United States John D. Bolcer and Robert B. Hermann Applications of Post-Hartree-Fock Methods: A Tutorial Rodney J. Bartlett and John F. Stanton Population Analysis and Electron Densities from Quantum Mechanics Steven M. Bachrach Biological Applications of Electrostatic Calculations and Brownian Dynamics Simulations Jeffry D. Madura, Malcolm E. Davis, Michael K. Gilson, Rebecca C. Wade, Brock A. Luty, and J. Andrew McCammon Computer Simulation of Lipid Systems K. V. Damodaran and Kenneth M. Merz Jr. Distance Geometry in Molecular Modeling Jeffrey M. Blaney and J. Scott Dixon A Perspective of Modern Methods in Computer-Aided Drug Design Lisa M. Balbes, S. Wayne Mascarella, and Donald B. Boyd Compendium of Software Donald B. Boyd Balancing academic and industrial interests, REVIEWS IN COMPUTATIONAL CHEMISTRY features definitive treatments of the latest advances in computational methods for chemists and scientists in allied fields. A valuable resource for novices and practitioners alike, many of the chapters are written as tutorials aimed at the level of first- or second-year graduate students. All aspects of computer-aided chemistry are within the scope of the series, including molecular modeling, molecular mechanics and dynamics, quantum chemistry, computer-assisted molecular design, quantitative structure-activity relationships, and structural databases. Included in each volume is a continually updated, comprehensive software compendium which serves as a valuable source guide. Listed are hundreds of programs, services, suppliers, and other helpful information. Kenny Lipkowitz and Donald B. Boyd, Co-Editors. -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Information about Volume V (ISBN 1-56081-658-9) can be obtained from VCH Publishers, 303 NW 12th Avenue, Deerfield Beach, FL 33442, tel. 800-367-8249, fax: 1-800-367-8247. In Europe, contact VCH, POB 101161, 69451 Weinheim, Germany, tel. 49-6201-6060, fax: 49-6201-606328. Price is $110, but with a standing order for the series, it is reduced to $90. -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- From: BOYD DONALD B (MCVAX0::QSAR) To: VMS MAIL ADDRESSEE (INT::"chemistry@ccl.net") cc: BOYD DONALD B (MCVAX0::QSAR) From ackerman@theochem.tu-muenchen.de Mon Nov 29 10:54:48 1993 Received: from theo1.theochem.tu-muenchen.de for ackerman@theochem.tu-muenchen.de by www.ccl.net (8.6.4/930601.1506) id KAA17085; Mon, 29 Nov 1993 10:48:33 -0500 Received: from theo80.theochem.tu-muenchen.de by theo1.theochem.tu-muenchen.de (5.64/1.34) id AA00542; Mon, 29 Nov 93 16:03:37 +0100 Received: by theo80.theochem.tu-muenchen.de (4.1/SMI-4.1) id AA00283; Mon, 29 Nov 93 16:03:35 +0100 Date: Mon, 29 Nov 93 16:03:35 +0100 From: ackerman@theochem.tu-muenchen.de (Lutz Ackermann) Message-Id: <9311291503.AA00283@theo80.theochem.tu-muenchen.de> To: chemistry@ccl.net Subject: Summary: Comp Chem Course Dear netters, About two weeks ago I asked for suggestions concerning textbooks and other educational material for a Comp. Chem. course. Here is a short summary of quotations from various responses so far (thanks a lot ye'all!): -------------------------------------------------------------------------------- -> A list of books has been on CCL on August 18 this year (I won't repeat that) -> "Reviews in Computational Chemistry Vol 4" eds Boyd & Lipkowitz VCH, 1993. Chapter 4 was written by professors from 3 different Universities on their experiences teaching comp chem to undergraduates. There are also more references at the end of the chapter. -> Even if you don't have GAUSSIAN for Windows the textbook that comes with it is sold separately and is pretty good. Foresman and Frisch Exploring Chemistry etc isbn 0-9636769-0-3 20 dollars (?) or so direct from GAUSSIAN Inc -> You might look at Chapter 15 of Levine's Quantum Chemistry text (4th ed). He has revised the material and it has some useful material. I also has some stuff that Warren Hehre presented at an ACS short course that evaluates the performance of various ab initio, semi-emiprical, and mechanics methods. Much of it I believe has been published under his name, which narrows a literature search considerably. I will be using these materials for the last three weeks of my Quantum Chemistry course, as I do not have a full semester to teach the material. -> We are also planning to introduce at least one experiment on the applications of Artificial Intelligence to Chemistry. If you are interested in this area, the text we shall use as a reference is a straightforward and short undergraduate text called "Applications of Artificial Intelligence in Chemistry". This text is part of the 'Oxford Chemistry Primers' series, published by Oxford University Press. I am the author (Dr Hugh Cartwright, Physical Chemistry Laboratory, Oxford University, UK) and the ISBN number is ISBN 0-19-855736-1; it should be published in about 4 weeks, and cost about 12-15DM (5 pounds UK). It is very much an introductory text, assuming no knowledge of AI (and almost no knowledge of computing as well!) It was written for scientific undergraduates in general, not just chemists. -> One book I have found useful, at the undergraduate level, is: _Molecular Dynamics Simulation Elementary Methods_ by: J.M. Haile Published by: Wiley Interscience ISBN number:0-471-81966-2 -------------------------------------------------------------------------------- I have not yet evaluated all af the references cited above, so I will not make any comment on those. It seems to me, that quite a lot of people are interested in this subject (please, excuse me, that I did not answer individually, but I assume that this summary will be helpful, too). I would encourage, that those, who have worked with textbooks or other materials apt for teaching computational chemistry give reviews of those via CCL-mail. Thanks a lot, once again! L. Ackermann ***************************************************** * * * Dr. Lutz Ackermann * * * * Lehrstuhl fuer Theoretische Chemie (TU Munich) * * Lichtenbergstr. 4 * * 85747 Garching * * Germany * * * * room Nr. 63103 * * Tel. (+49 +89) 3209 3602 * * Fax (+49 +89) 3209 3622 * * e-mail ackerman@theo1.theochem.tu-muenchen.de * * * ***************************************************** From CHAMANKHAH@sask.usask.ca Mon Nov 29 14:51:44 1993 Received: from skycat.usask.ca for CHAMANKHAH@sask.usask.ca by www.ccl.net (8.6.4/930601.1506) id OAA19115; Mon, 29 Nov 1993 14:20:06 -0500 From: Received: from SKYCAT.USask.CA by SKYCAT.USask.CA (PMDF V4.2-14 #3676) id <01H5W0I7ZQ688WWVCS@SKYCAT.USask.CA>; Mon, 29 Nov 1993 13:19:05 CST Date: Mon, 29 Nov 1993 13:19:05 -0600 (CST) Subject: Program for Elemental Composition To: CHEMISTRY@ccl.net Message-id: <01H5W0I85MCY8WWVCS@SKYCAT.USask.CA> X-Envelope-to: CHEMISTRY@ccl.net X-VMS-To: IN::"CHEMISTRY@ccl.net" MIME-version: 1.0 Content-transfer-encoding: 7BIT Hi Netters Does anybody know if there is a computer program capable of calculating the elemental composition of a compound knowing the exact mass by high resolution mass spectrometry? Thus the input is exact mass (high resolution) and the expected output will be the general molecular formula or elemental composition of the compound. Thanks for your help. M.Chamankhah College of Pharmacy University of Saskatchewan Saskatoon, Canada From pbays@saintmarys.edu Mon Nov 29 15:52:43 1993 Received: from jade.saintmarys.edu for pbays@saintmarys.edu by www.ccl.net (8.6.4/930601.1506) id PAA19524; Mon, 29 Nov 1993 15:19:00 -0500 Received: by jade.saintmarys.edu (1.37.109.8/16.2) id AA11079; Mon, 29 Nov 1993 15:21:52 -0500 Date: Mon, 29 Nov 1993 15:19:44 -0500 (EST) From: Phil Bays Subject: Comp. Chem. Seats To: CHEMISTRY@ccl.net Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I have funding for setting up a computational chemistry lab here at Saint Mary's College. This will be used by undergraduate students. The college size is 1500 students, with about 50 taking organic, and 5-10 taking physical chemistry. I give that background because it is important for the following questions: What has been your experience in setting up such a lab in terms of the ratio of students to workstation? I can buy 5 IndyPC's, 4 Indy SC's or 2 Indigo^2's. I will be running Spartan and Labvision, possibly MacroModel and Gaussian. SGI sales people tell me to forget the IndyPC, and go at least to the IndySC. On the other hand, they also tell me that the Indigo^2 is the only one that has future expandability and upgradability. My own observation is that SGI pricing is such that upgrading is frequently the same price as a new machine. Any advice?? Phil Bays Saint Mary's College Notre Dame IN 46556 pbays@saintmarys.edu From CHM001@marshall.wvnet.edu Mon Nov 29 20:51:43 1993 Received: from muvms3.mu.wvnet.edu for CHM001@marshall.wvnet.edu by www.ccl.net (8.6.4/930601.1506) id UAA21861; Mon, 29 Nov 1993 20:38:31 -0500 Received: from marshall.wvnet.edu by marshall.wvnet.edu (PMDF #2995 ) id <01H5WG048FZK9AMJK6@marshall.wvnet.edu>; Mon, 29 Nov 1993 20:39:15 EST Date: 29 Nov 1993 20:39:15 -0500 (EST) From: "DR. LAWRENCE R. SCHMITZ" Subject: Cray Time - Dollar value To: chemistry@ccl.net Message-id: <01H5WG048FZM9AMJK6@marshall.wvnet.edu> X-VMS-To: _IN%"chemistry@ccl.net" MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII Content-transfer-encoding: 7BIT I recently received a small grant of Cray C90 time at the Pittsburgh Supercomputing Center. The University propaganda machine requires that I estimate the dollar value of the grant. Therefore, I would very much appreciate it if those of you that are familiar with commercial rates for Cray C90 time could enlighten me. The grant is given in units of Service Units. A service unit is define as: SU = 0.75*CPU+1/64*MemInt+0.01*Con where CPU = CPU hours MemInt = a memory integral, in megaword hours. The memory integral is calculated by adding increments based on changes in the amount of memory your program is using: increment = memory size * (total time when program was connected to at least one CPU since last memory size change) Con = connect time hours For the way I use the Cray, the charges will be almost entirely based on CPU time. (For an hour of cpu time, I will use about 12 MB (1.5 megawords) of RAM and perhaps a half hour of connect time.) Therefore, a service unit is about 1.3 hours of CPU time. My question then is: If you were going to buy about 100 hours of CPU time on a C90, how much would you expect to pay per hour? Thanks for any help you can offer! Sincerely, Larry Schmitz PS In April or May of 92 I asked a similar question about time on a Cray Y-MP/832. The rates I was quoted varied widely. I finally decided that about $600/hour was a fair market value at that time. The C90 is faster than the Y-MP/832 but I suspect that rates have declined.