Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!husc6!cmcl2!brl-adm!brl-smoke!gwyn
From: gwyn@brl-smoke.ARPA (Doug Gwyn )
Newsgroups: comp.arch
Subject: Re: Benchmarking the 532, 68030, MIPS, 386...at a Usenix!
Message-ID: <5867@brl-smoke.ARPA>
Date: Sat, 16-May-87 23:19:24 EDT
Article-I.D.: brl-smok.5867
Posted: Sat May 16 23:19:24 1987
Date-Received: Sun, 17-May-87 11:17:55 EDT
References: <324@dumbo.UUCP> <809@killer.UUCP> <2417@homxa.UUCP>
Reply-To: gwyn@brl.arpa (Doug Gwyn (VLD/VMB) <gwyn>)
Organization: Ballistic Research Lab (BRL), APG, MD.
Lines: 14

In article <4329@nsc.nsc.com> grenley@nsc.UUCP (George Grenley) writes:
>Consider also that those of us who are chip manufacturers are primarily
>interested in CPU benchmarks, not system benchmarks.  Unix is NOT the
>entire world, yet.

As in any benchmarking, meaningful results have to be obtained in the
proper context.  Any application of a CPU chip is likely to be as part
of an operating system + application environment, perhaps with either
or both written in any of a number of programming languages other than
assembler.  I have encountered a number of CPU architectures that
simply were not properly designed to assist the software that would be
used on them.  Fortunately there are some exceptions these days; the
WE32000 family comes to mind as a typical example of an attempt to
accommodate UNIX-like operating systems and C-like languages.