Path: utzoo!utgpu!news-server.csri.toronto.edu!clyde.concordia.ca!uunet!sco!seanf
From: seanf@sco.COM (Sean Fagan)
Newsgroups: comp.arch
Subject: Re: 68040 LINPACK numbers from Microprocessor Forum
Message-ID: <6059@scolex.sco.COM>
Date: 6 May 90 10:24:05 GMT
References: <38134@mips.mips.COM> <1361@marlin.NOSC.MIL> <5913@scolex.sco.COM> <1376@marlin.NOSC.MIL>
Sender: news@sco.COM
Reply-To: seanf@sco.COM (Sean Fagan)
Distribution: comp.arch
Organization: The Santa Cruz Operation, Inc.
Lines: 21


In article <1376@marlin.NOSC.MIL> aburto@marlin.nosc.mil.UUCP (Alfred A. Aburto) writes:
>Dropping from 582 to 97 is quite a step!  It is such a giant step one
>wonders what magic is taking place.  

One realizes that Cyrix did the thing without microcode, and then one
realizes that 582 Cycles to 97 cycles isn't so incredible after all.

>Does it still yield the same 
>accuracy at 97 CC's as it did at 582 CC's ???

Yes.  I think there are only a couple of instances where the Cyrix chip got
different answers from the Intel chip, and someone mentioned that the Intel
chip gets different answers from at least one FP emulator for DOS.

-- 
-----------------+
Sean Eric Fagan  | "It's a pity the universe doesn't use [a] segmented 
seanf@sco.COM    |  architecture with a protected mode."
uunet!sco!seanf  |         -- Rich Cook, _Wizard's Bane_
(408) 458-1422   | Any opinions expressed are my own, not my employers'.