Path: utzoo!attcan!uunet!cbmvax!brim
From: brim@cbmvax.commodore.com (Mike Brim - Product Assurance)
Newsgroups: comp.sys.ibm.pc.misc
Subject: Re: 80386-33 and "double sigma rule" ??
Message-ID: <15882@cbmvax.commodore.com>
Date: 15 Nov 90 14:22:38 GMT
References: <3720001@hpsgwp.sgp.hp.com>
Reply-To: brim@cbmvax.commodore.com (Mike Brim - Product Assurance)
Organization: Commodore, West Chester, PA
Lines: 27

In article <3720001@hpsgwp.sgp.hp.com> plim@hpsgwp.sgp.hp.com (Peter Lim) writes:

> When I was shopping for 20 to 25 MHz 80386 CPU's, I was told that the
> bug free version is those with double sigma marking. I did saw them.
> Now I am interested in 80386-33. It trouble me that I don't see the
> same double sigma marking on the CPUs I have looked at so far.
> 
> Does the double sigma rule apply to 33 MHz 80386 or is it only to
> the slower parts ? In another word, how can I tell if a 33 MHz 80386
> is a bug free version (may be they all are, but I hope someone
> knowledgeable can tell me if it is so).
> 


If you are referring to the "16-bit only operation" bug, that occurred only 
with early 16MHz 80386 CPUs (and the 1st 20,000 of them).  20MHz and greater
do not have the problem.  The buggy 16MHZ 80386's are also suppose to have 
markings on it stating that the chip is for 16-bit operation only.  
Unfortunately many chips went out before this marking was added.
-- 
********************************************************************************
Disclaimer: My company knows not what I say (or do).

Mike Brim			     |	Commodore Electronics Limited
PC Analyst - System Evaluation Group | 	West Chester, PA 19380
Product Assurance		     |	InterNet: brim@cbmvax.commodore.com
********************************************************************************