Path: utzoo!attcan!uunet!lll-winken!lll-tis!helios.ee.lbl.gov!nosc!ucsd!ucbvax!decwrl!purdue!bu-cs!bloom-beacon!mit-eddie!uw-beaver!cornell!rochester!srs!dan
From: dan@srs.UUCP (Dan Kegel)
Newsgroups: comp.sys.ibm.pc
Subject: Re: GATHER and say NO to MCA!
Summary: Choose the Nu-bus!
Message-ID: <925@srs.UUCP>
Date: 27 Jun 88 23:52:12 GMT
References: <42900016@uicsrd.csrd.uiuc.edu> <257@octopus.UUCP>
Reply-To: srs!dan@cs.rochester.edu (Dan Kegel)
Organization: S.R.Systems
Lines: 33

pete@octopus.UUCP (Pete Holzmann) writes:
> leein@uicsrd.csrd.uiuc.edu writes:
>>Dear PC clone makers;
>>   Gather as OSF members did, and choose a new advanced bus like NuBus 
>>   (which is in the public domain) ...
>
> [ summary: ]
> The AT bus is plenty fast; all it needs is a 32-bit address extension.  
> Most applications don't need multiple CPU's.
> Let's extend the AT bus- there is no good engineering reason to switch 
> to MCA or anything else.

The AT-bus is sufficient for here and now; however, some people will
eventually need machines with large shared address spaces that the AT
bus just can't provide, even with a 32-bit extension.
Let's make a clean break, rather than add a kludge atop a kludge atop a
poor bus.

Choosing a standard bus will have the benefits that boards developed for
it will be usable on other computers, and vice versa.
The Nu-Bus is a logical choice because it is low-cost compared to busses
like VME or Multibus, it is certainly up to the job, and there are already
many very nifty boards available for it (e.g. RISC coprocessor boards).
It is also more reliable than the AT bus; its sockets are pin-and-plug,
rather than edge-card.

As a hardware and software engineer who deals with workstations, I am very 
excited by the prospect of a single bus for the major low-cost 32-bit 
workstations (the PC 386 and the Mac II); with only one bus to target, 
hardware designers would have an easier time getting to market, and buyers 
would have a better variety to choose from.
-- 
  Dan Kegel   "Take this job..."
  srs!dan@cs.rochester.edu  rochester!srs!dan dan%srs.uucp@harvard.harvard.edu