Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!samsung!rex!uflorida!gatech!mcnc!uvaarpa!murdoch!astsun.astro.Virginia.EDU!gl8f
From: gl8f@astsun.astro.Virginia.EDU (Greg Lindahl)
Newsgroups: comp.benchmarks
Subject: Re: BYTE and benchmarks (anecdote)
Message-ID: <1991Jan14.202646.5677@murdoch.acc.Virginia.EDU>
Date: 14 Jan 91 20:26:46 GMT
References: <EACHUS.91Jan8114647@aries.linus.mitre.org> <1991Jan09.171216.21923@convex.com> <josef.663497368@ugum01>
Sender: news@murdoch.acc.Virginia.EDU
Distribution: comp
Organization: Department of Astronomy, University of Virginia
Lines: 12

In article <josef.663497368@ugum01> josef@nixpbe.nixdorf.de (josef Moellers) writes:

>losing out of sight the purpose of a
>benchmark: not to produce the numbers in the least amount of time, but
>rather to keep Your machine busy for some time.

Gee, and here I thought the purpose of a benchmark was to predict the
relative performance of various machines on the applications that I
use them for.

Silly me. I propose we use the "infinite loop" benchmark instead of
SPEC.