Path: utzoo!attcan!uunet!mcsun!ukc!slxsys!jpp
From: jpp@specialix.co.uk (John Pettitt)
Newsgroups: comp.sys.ibm.pc.hardware
Subject: Re: Whis is fastest 386/33 or 486/25 ?
Message-ID: <1990Jul13.104507.3078@specialix.co.uk>
Date: 13 Jul 90 10:45:07 GMT
References: <2418@mindlink.UUCP>
Organization: Specialix International, London
Lines: 35

a1499@mindlink.UUCP (Andrew Harmsworth) writes:
>> laughner@news.nd.edu writes:
>> 
>> Msg-ID: <217@news.nd.edu>
>> Posted: 12 Jul 90 14:37:06 GMT
>> 
>> Org.  : Univ. of Notre Dame
>> Person: Tom laughner
>> There would be no difference in speed between a 386 with a math
>> coprocessor and a 486.  The 486 chip is a 386 + the math coprocessor in
>> one.  Intel considers the 486 as a part of the 386 family.

>I'm afraid that is not correct, essentially it is true, but a number of
>optimizations were carried out in combining the two chips.  I don't know the
>details, what it affects, or by how much, but there are differences.

As a rough guide a 25Mhz 486 is about twice a fast as a 25Mhz 386
(both cached) running SCO UNIX (musbus and dhrystone benchmarks).

I did a quick check here (bc 512 ^ 512) and got the folowing:
33 Mhz ALR 386		6.7 seconds user time
25 Mhz ALR 486 (mca)	5.0 seconds user time

This means that the 486 is 25% faster on this test, which equates
quite well to the dhrystone numbers.

FYI A 25 Mhz MIPS 3240 will do the same test in 3.3 seconds !




-- 
John Pettitt, Specialix International, 
Email: jpp@specialix.com Tel +44 (0) 9323 54254 Fax +44 (0) 9323 52781
Disclaimer: Me, say that ?  Never, it's a forged posting !