Path: utzoo!attcan!uunet!lll-winken!lll-lcc!unisoft!gethen!farren
From: farren@gethen.UUCP (Michael J. Farren)
Newsgroups: comp.sys.ibm.pc
Subject: Re: GATHER and say NO to MCA!
Message-ID: <978@gethen.UUCP>
Date: 30 Jun 88 09:56:14 GMT
References: <42900016@uicsrd.csrd.uiuc.edu> <257@octopus.UUCP>
Reply-To: farren@gethen.UUCP (Michael J. Farren)
Organization: There's Unix there in Oakland
Lines: 36

In article <257@octopus.UUCP> pete@octopus.UUCP (Pete Holzmann) writes:

[Before I start, one question:  Mr. Holzmann, have you ever designed a
 high-speed digital bus?  I have.  Now, on with the show...]

>1) MCA is cleaner, newer, nicer, etc etc:
>	All true, all irrelevant. On a bus, what works is what counts!

This is true.  However, what is NOT true is the claim that the AT bus
"works".  No 32-bit data/address path, insufficient DMA and interrupt
support, insufficient attention to loading and timing details essential
to truly high-speed operation, and a god-awful electrical emissions
characteristic.  I don't blame IBM one bit for getting rid of the damned
thing.

>3) MCA handles multiple CPU's.
>	So does the AT bus. A kludge, true,

Not a kludge.  A disaster.  True coprocessing is damn near impossible.

>My conclusion: the clone-makers need to pick a 32-bit AT bus extension
>	standard. There is little engineering reason (right now) to go to the
>	MCA.

Unless you count greater reliability, better support for advanced architecture,
better support for I/O, greater noise resistance, less emissions problems
(therefore easier to get FCC type approval, therefore cheaper), etc....

Not that I think that MCA is the BEST bus around, mind, but anyone who
claims that the AT bus is sufficient is talking through his engineering
hat.

-- 
Michael J. Farren             | "INVESTIGATE your point of view, don't just 
{ucbvax, uunet, hoptoad}!     | dogmatize it!  Reflect on it and re-evaluate
        unisoft!gethen!farren | it.  You may want to change your mind someday."
gethen!farren@lll-winken.llnl.gov ----- Tom Reingold, from alt.flame