Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!husc6!rutgers!labrea!rocky!stergios
From: stergios@rocky.STANFORD.EDU (Stergios Marinopoul)
Newsgroups: comp.sys.mac
Subject: Re: 68030 in MacII
Message-ID: <577@rocky.STANFORD.EDU>
Date: Fri, 11-Sep-87 19:47:49 EDT
Article-I.D.: rocky.577
Posted: Fri Sep 11 19:47:49 1987
Date-Received: Sat, 12-Sep-87 20:33:18 EDT
References: <2549@okstate.UUCP>
Reply-To: stergios@rocky.UUCP (Stergios Marinopoul)
Organization: Stanford University Computer Science Department
Lines: 30
Keywords: 68030 MacII

In article <2549@okstate.UUCP> norman@a.cs.okstate.edu (Norman Graham) writes:
>According to Rand Miller's artical in the September issue of Macazine, the
>"Mac II is designed so this new chip [the 68030] will pop right in to
>replace the 68020."
> {edited}
>Are the pinouts the same (implying that
>the "Harvard style parallel architecture" is not seen off chip)?
>
>Norman Graham (norman @ okstate)

I'll speculate on this a little even though I have not read the article.
Even though the 020 has 32 address lines and the 000 has 24, you can still
pop in the former in the latter with a little glue simulating the requests
for lower or upper word on data transfers.  You simply map the higher data
lines on the first 16, and let the 020 run at half the bandwith on long data
moves. ( for a complete description on the how to's see EDN Jan. 23, 86)


Now this probably not the problem you are refering to, however I suspect
similiar gluing can be acomplished to make the two (020 & 030) compatable.
In any event, you propbably wont get the full performance increase 
available, but half is still pretty good.

Stergios Marinopoulos 
S&M Engineering

-- 
% UUCP:         !decwrl!rocky.stanford.edu!stergios			%
% ARPA:		f.flex@othello.stanford.edu				%
% USnail:	Crothers Memorial #690, Stanford, CA. 94305		%
% Pa Bell:	(415) 326-9051						%