Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!samsung!noose.ecn.purdue.edu!mentor.cc.purdue.edu!mace.cc.purdue.edu!abe
From: abe@mace.cc.purdue.edu (Vic Abell)
Newsgroups: comp.unix.aix
Subject: Re: SPECmarks for RS/6000 systems - lies???
Summary: probably not
Message-ID: <5709@mace.cc.purdue.edu>
Date: 6 Oct 90 00:49:31 GMT
References: <4733@lure.latrobe.edu.au>
Organization: Purdue University
Lines: 26

In article <4733@lure.latrobe.edu.au>, CCHD@lure.latrobe.edu.au (Huw Davies - La Trobe University Computer Centre) writes:
> I have just got a copy of the September 1990 SPECwatch and I am
> a bit concerned about the following paragraph:
> 
> "There also seems to be a problem with replicating IBM's RS6000
> SPECmark results, and with achieving the expected levels of
> performance with other code.

While I haven't had a chance today to run the entire SPEC 1.0 suite on my
RS/6000-520 (GA code), I did run the following benchmarks from the suite.

		008.espresso
		013.spice2g6
		023.eqntott
		030.matrix300
		047.tomcatv

The SPEC ratios I got differ from IBM's by no more than 0.4.  I got the
same result for one test; one of my results was higher by 0.3, another by
0.4; and two were lower by 0.3.  (I think SPEC reporting rules keep me
from publishing an incomplete report.)

Vic Abell
Assistant Director
Purdue University Computing Center
SPEC license 310