NetNews Usenet Archive 1992 #19/NN_1992_19.iso/spool/comp/sys/sgi/12792

Text File | 1992-08-23 | 1.7 KB | 34 lines

Newsgroups: comp.sys.sgi
Path: sparky!uunet!usc!sol.ctr.columbia.edu!avogadro.barnard.columbia.edu!shenkin
From: [e-mail] (Peter S. Shenkin)
Subject: Machine Performance Ratio Query
Organization: Dept. of Chem, Barnard College, Columbia U, New York
Originator: [e-mail]
Message-ID: <[e-mail]>
Sender: [e-mail] (The Daily Lose)
Date: Sun, 23 Aug 1992 19:16:38 GMT
X-Posted-From: avogadro.barnard.columbia.edu
X-Posted-Through: sol.ctr.columbia.edu
Lines: 20

Can someone remind me of the relative speeds of the 4d25 and 4d35?  I'm
compiling some performance statistics for a program of mine, using data sets
of different sizes.  I did some runs on a 25 and some on a 35, and I'd like
to put them all on a common basis.

A harder question:  what would be a reasonable number expressing the
ratio of the speed of a 4d25 to that of the gold-plated 1-MIP VAX 11/780fpa
at the NBS?  :-)   I know that there is no single answer -- in particular,
the answer will differ depending on the relative amounts of integer, fp and
io performed by the application -- but I'd like to have some approximate
figure to use, guessing that it could be off by perhaps a factor of two
either way.  A combination of intuition and vague recollection leads me
to believe that "5" would not be a bad number;  any more informed guesses?

    -P.
--
************************f*u*cn*rd*ths*u*cn*gt*a*gd*jb*************************
Peter S. Shenkin, Department of Chemistry, Barnard College, New York, NY 10027
[phone]    [e-mail]   [e-mail]
******************** The singular of "media" is "medium". ********************