Xref: utzoo comp.sys.att:11613 unix-pc.general:7434
Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!fernwood!portal!cup.portal.com!thad
From: thad@cup.portal.com (Thad P Floryan)
Newsgroups: comp.sys.att,unix-pc.general
Subject: Re: 3B1 Rising Popularity
Message-ID: <38506@cup.portal.com>
Date: 27 Jan 91 01:44:44 GMT
References: <1699@madnix.UUCP> <42@morwyn.UUCP>
Organization: The Portal System (TM)
Lines: 37

forrie@morwyn.UUCP (Forrie Aldrich) in <42@morwyn.UUCP> writes:

	Yes it would be NICE if someone could figure out a way to use the 
	68030 on the UNIX-PC... believe me, it would be incredible.  But 
	unfortunately, I think it's impossible at this point.  As doing such
	would require all new hardware (mostly) AND I believe a new
	KERNEL --- who's gonna write it?  Because of different address space,
	and all that complicated stuff.  BUT if someone DOES it, let ME know!

Welllll, a number of companies mfg. 68020/68881 daughterboards that plug in
directly to a 68000 or 68010 socket.  That trend started about 6 years ago
when Motorola wanted to upgrade their 68000/68010 trainers, and has since been
expanded to upgrading Amigas, Macs, and other 68000-family machines.

The Amiga automatically identifies a 68000, 68010, 68020, 68030 or 68040 and
thus has NO problems with any CPU in the 680x0 family (i.e. part of the CPU ID
interrogation includes adjusting its kernel for stack frame differences and
cognizance of special modes of the other chips during its boot procedure).

The only software problems on the Amiga (with other CPU chips) concerns brain
damaged games from Europe with their hideous self-modifying-code copy
protection (which screws the data and instruction caches on 68020, 68030 and
68040 processors).

I plugged one of the Amiga cards into the 3B1 and it worked for all of a few
milliseconds; I suspect the problem was due to the stack frame on interrupts.

In a Motorola Application Note it is claimed to require just a few hours to
"fix" any kernel to handle the different CPU.  HOWEVER, this requires kernel
source (or a DAMN GOOD disassembler and a willingness to experiment).

Note also the daughterboards (that I've seen) will not permit the top-level
metal cover (the floppy/HD/power-supply plate) to be closed, so a tower-like
case with abundant room seems a necccessity.   Hmmmm, now that I have one 3B1
in a tower case with external monitor, hmmm .... :-)

Thad Floryan [ thad@cup.portal.com ]