*BSD News Article 54678


Return to BSD News archive

Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!simtel!news.kei.com!newsfeed.internetmci.com!swrinde!sdd.hp.com!news1.best.com!shellx.best.com!shellx.best.com!not-for-mail
From: rcarter@shellx.best.com (Russell Carter)
Newsgroups: comp.unix.bsd.freebsd.misc
Subject: Re: Pentium Pro support
Date: 13 Nov 1995 16:27:33 -0800
Organization: Best Internet Communications
Lines: 24
Message-ID: <488npl$4ge@shellx.best.com>
References: <46n3u4$256@goanna.cs.rmit.edu.au> <DHrECs.J3E@ritz.mordor.com> <30A23B7A.794BDF32@FreeBSD.org> <483ue4$po@tzlink.j51.com>
NNTP-Posting-Host: shellx.best.com

In article <483ue4$po@tzlink.j51.com>, Louis Epstein <lepslog@j51.com> wrote:
>Jordan K. Hubbard (jkh@FreeBSD.org) wrote:
>: I'm afraid you'll just have to wait and find out.. :-)
>: [Intel is being secretive about the numbers still]
>
>Intel has announced SPECmark numbers for the P6,so what's the fuss?
>
>(If the P6 were no faster than the P5,they'd be in BIG trouble!)
>

Well the interesting thing is the SPEC marks for the P5-133, which
are used to compare to the P6, are apparently taken on a non-pipelined
burst SRAM system.  The P6 then looks pretty good, especially the
200MHz ratings.  On the other hand, a pipelined burst SRAM system
is used for the Windows P5-133 benchmarks, which of course makes it
look pretty good compared to older processors (especially those pesky
486s...).  My guess is that if they really aren't using pipelined
burst SRAM in those SPEC rated 133 marks, you can do 15-20% better.
And then the P6 doesn't look so hot, at least not the 150MHz processor.

Regards,
Russell
rcarter@geli.com
http://www.geli.com