Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watnot!watmath!clyde!cbatt!ihnp4!homxb!houxm!mhuxt!mhuxm!mhuxo!ulysses!allegra!alice!adb
From: adb@alice.UUCP
Newsgroups: comp.arch
Subject: Re: register window machine questions
Message-ID: <6699@alice.uUCp>
Date: Tue, 10-Mar-87 10:13:22 EST
Article-I.D.: alice.6699
Posted: Tue Mar 10 10:13:22 1987
Date-Received: Thu, 12-Mar-87 22:39:42 EST
References: <4376@columbia.UUCP> <448@cpocd2.UUCP> <495@apple.UUCP> <14@winchester.mips.UUCP>
Organization: AT&T Bell Laboratories, Murray Hill NJ
Lines: 17
Summary: What CRISP really does

In article <14@winchester.mips.UUCP>, mash@mips.UUCP writes:
> >Ditzel et. al. have finally published details of the CRISP processor,
> >which is a chip implements a stack cache with 32 registers....
> >.... They claim exceptional high hit
> >ratios on the stack cache, and performance of about 9x a VAX780
> >running at 16Mhz (for some number of benchmarks including PCC--
> >smaller benchmarks show 30x!). There are a number of other
> I do believe that was X a 750, not a 780 ...
Sorry, John, it really was 9 times a 780, not a 750.  I suspected that posting
a slide with 750 numbers would confuse people.
> 	13X	(750) Dhrystone	(11,000? hard to know what their 750s do?)
The Dhrystone number we published was 13,560, and the 750 we used ran about
996.  The precise numbers for wc and Ackerman's are 19.5 and 33.6 times
a 750, respectively.

	Alan Berenbaum, AT&T-Something
	..!ihnp4!homxa!adb   ..!research!adb