Xref: utzoo comp.arch:16126 alt.folklore.computers:3487
Path: utzoo!censor!isgtec!bmw
From: bmw@isgtec.UUCP (Bruce Walker)
Newsgroups: comp.arch,alt.folklore.computers
Subject: Re: Why IBM went for the 8088 -- The Untold Story (8086 clones)
Keywords: S100 IEEE-696 SC-DOS Seattle MS-DOS IBM_PC
Message-ID: <457@isgtec.UUCP>
Date: 28 May 90 13:14:48 GMT
References: <1WWjqN#0SBCWN5sLFYH9X6tz009zLSK=eric@snark.uu.net> <1990May23.154706.16476@sq.sq.com> <4790@pegasus.ATT.COM>
Reply-To: bmw@isgtec.UUCP (Bruce Walker)
Organization: ISG Technologies Inc. Mississauga Ont. Canada
Lines: 27

In article <4790@pegasus.ATT.COM> dmt@pegasus1.ATT.COM (Dave Tutelman) writes:
> 	Does anyone know of any other 8086-based XT clones (not made by
> 	Olivetti), or is this the only counter-example to Ian's
> 	statement?

Well, yes: the IBM PS/2 Model 30 (and 25) is an XT clone with an 8MHz 8086
that accepts 8 bit ISA boards.

A few others exist; predating the IBM entry (and possibly the Olivetti) was
the BEST Mark III from a small Canadian manufacturer (Exceltronics).

The only really tricky part about achieving compatibility with the 8088 XT
is the "bus bridging" circuit.  The bus bridge decides how to handle various
bus requests from the CPU to different parts of the address space.  It has
to "know" where there are 8 and 16 bit peripherals and how to turn a 16 bit
request to an 8 bit "area" into two back-to-back 8 bit requests with proper
wait-states and all.*

The solution to this same problem made the AT design messier.

[* I know this problem well:  I helped design a single-chip gate-array for
8086 XT-clone motherboards for LSI Logic.]

-- 
Bruce Walker         ...uunet!utai!lsuc!isgtec!bmw         bmw@isgtec.uucp
"Remember Rule Number 79: When the tough get going, the weak get screwed."
ISG Technologies Inc.  3030 Orlando Dr. Mississauga.   Ont.  Can.  L4V 1S8