From kotelyan(+ at +)che.udel.edu  Thu Oct 24 15:31:47 1996
Received: from che.udel.edu  for kotelyan*- at -*che.udel.edu
	by www.ccl.net (8.8.0/950822.1) id OAA26145; Thu, 24 Oct 1996 14:33:08 -0400 (EDT)
Received: by che.udel.edu (AIX 3.2/UCB 5.64/4.03)
          id AA34232; Thu, 24 Oct 1996 14:32:10 -0400
Date: Thu, 24 Oct 1996 14:31:37 +22305133 (EDT)
From: Michael Kotelyanskii <kotelyan- at -che.udel.edu>
Subject: Re: CCL:Charmm24-benchmark PPro versus SGI
To: Bernd Melchers <melchers #at# chemie.fu-berlin.de>
Cc: chemistry ^at^ www.ccl.net
In-Reply-To: <m0vGOUX-008pubC |-at-| Bose.Chemie.FU-Berlin.DE>
Message-Id: <Pine.3.05.9610241435.C54920-c100000;at;che.udel.edu>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII


That's interesting ,
but why is R10000 missing from the MD ?

Michael Kotelyanskii PhD
Department of Chemical Engineering
University of Delaware                          kotelyan&$at$&che.udel.edu
Newark DE 19716                                    


On Thu, 24 Oct 1996, Bernd Melchers wrote:

> hello,
> we performed some charmm24 benchmarks with a Pentium Pro and with some
> Silicon Graphics machines:
> 
> 1. modelling (select, coor orien, hbuild) and some energy minimization
> (i.e. integer and floating point)
> a) R4600 100 MHz     24.45 MINUTES
> b) R4400 150 MHz     14.08 MINUTES
> c) R8000  75 MHz      7.21 MINUTES
> d) PPro  200 MHz      5.53 MINUTES
> e) R10000 195MHz      2.51 MINUTES
> 
> 
> 2. molecular dynamik simulation (i.e. much floating point)
> f) R4000 100 MHz   4.36 hours
> b) R4400 150 MHz   2.90 hours
> d) PPro  200 MHz   1.29 hours
> 
> 
> machine/compiler description:
> 
> a) Indy, IRIX-5.3, R4600SC, 100 MHz, 16+16/0 Cache, cc -O2
> b) Indigo, IRIX-5.3, R4400, 150 MHz, 16+16/1024 Cache, cc -O2
> c) Power Indigo^2, IRIX-6.2, R8000, 75 MHz, 16+16/2048 Cache, cc -O3 -mips4
> d) PentiumPro Asus Natoma Board, 200 MHz, 16+16(?)/256 Cache, gcc-2.7.2/
>    g77-0.5.18 -O2 running linux kernel 2.0.18
> e) R10000, IRIX-6.2, 195 MHz, 32+32/1024 Cache, cc -O3 -mips4
> f) Indigo R4000, IRIX-5.3, 100 MHz, 16+16/1024 Cache, cc -O2
> 
> the results are not too bad for the intel chips, aren't they?
> 
> Ciao
> Bernd
> -- 
> -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-+-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
> Bernd Melchers                   | melchers /at\FU-Berlin.DE
> Freie Universitaet Berlin        |
> Institut fuer Kristallographie   | Please replace battery pack,
> Takustr. 6                       | attach external power,
> D 14195 Berlin                   | or system will be
> Germany                          | shutdown in 10 seconds.
> Earth                            |
> Tel.: ++49-30-8383612            | Fax.: ++49-30-8383464
> picture and pgp key: http://www.chemie.fu-berlin.de/~melchers
>                  "We don't write software, we compose it."
> -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-+-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
> 
> -------This is added Automatically by the Software--------
> -- Original Sender Envelope Address: melchers*- at -*chemie.fu-berlin.de
> -- Original Sender From: Address: melchers |-at-| chemie.fu-berlin.de
> CHEMISTRY ":at:" www.ccl.net: Everybody | CHEMISTRY-REQUEST ":at:" www.ccl.net: Coordinator
> MAILSERV $#at#$ www.ccl.net: HELP CHEMISTRY or HELP SEARCH | Gopher: www.ccl.net 73
> Anon. ftp: www.ccl.net   | CHEMISTRY-SEARCH&$at$&www.ccl.net -- archive search
>              Web: http://www.ccl.net/chemistry.html 
>