Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!uwm.edu!zaphod.mps.ohio-state.edu!wuarchive!decwrl!shelby!neon!Kermit.Stanford.EDU!philip
From: philip@Kermit.Stanford.EDU (Philip Machanick)
Newsgroups: comp.sys.mac
Subject: Re: Low-Cost Macintosh
Message-ID: <1990Feb4.004822.10300@Neon.Stanford.EDU>
Date: 4 Feb 90 00:48:22 GMT
References: <20475@watdragon.waterloo.edu> <126900165@p.cs.uiuc.edu> <25c8e25c.62bb@polyslo.CalPoly.EDU> <6192@ncar.ucar.edu>
Sender: news@Neon.Stanford.EDU (USENET News System)
Reply-To: philip@pescadero.stanford.edu
Organization: Computer Science Department, Stanford University
Lines: 63

In article <20475@watdragon.waterloo.edu>, nrjwong@lion.waterloo.edu writes:
> In article <6192@ncar.ucar.edu> hpoppe@bierstadt.scd.ucar.edu (Herb
Poppe) writes:
> >In article <25c8e25c.62bb@polyslo.CalPoly.EDU>
rcfische@polyslo.CalPoly.EDU (Raymond C. Fischer) writes:
> >>
> >>In article <126900165@p.cs.uiuc.edu> gillies@p.cs.uiuc.edu writes:
> >>>...
> [Stuff Deleted about using an '030 in a low-cost Mac]
> [Stuff deleted about why the above isn't possible for Apple]
> >>...
> >What Apple and Motorola need to do to tackle this problem is to take
> >a lesson from the PC world. From a software point of view, the 80286
> >became a dead dog (listen to Bill Gates; he's telling PC users what
> >they are going to get). But it has a 16 bit external bus. A lot of
> >cheap clones could be made and sold with a chip that had the 386
> >architecture, except with a 16 bit, rather than 32 bit, external data bus.
> >So Intel came out with the 386SX.
> 
> Another reason is because Intel wanted more profits, so the 386 and 386SX
> are not second-sourced. Companies like AMD and Harris who are licensed
> to produce 286's got left out in the cold w.r.t. the 386. I think
> there were lawsuits pending about this issue. Why do you think Intel
> has been raving in the mags about the greatness of the 386SX and the
> death of the 286????
> 
> PC magazine (I believe) has shown that a 16MHz 286 and 16MHz 386SX are
> virtually identical in speed in most cases. OS/2 is still a 286-based
> operating system. There isn't that much busines software out for 386's
> especially, and PC clone makers don't have to worry about getting sued from
> IBM for copying the BIOS. They just get a BIOS from any number of
> 3rd party sources.
> 
> >
> >To a much lesser degree, the 68000 is also a dead dog, with respect to
> >future directions in Mac OS software. A "low cost", 16 bit external data
> >bus part, packaged in a cheap plastic DIP, that is software compatible
> >with the 68030 would make a great platform for a new line of Macs.
> >Make it a CMOS part and it could be used in a new Portable (it would
> >also reduce the power requirements or a non-portable, permitting a
> >cheaper power supply).
> 
> I think the main reason the 68000 seems like a dead dog is partly
> due to the Mac's (small macs) design and the lack of a decent IPC.
> Take a look at the software packages out there today. They try to
> do everything at once for everyone in one package. Why? Because
> it's a pain to switch to another program to do stuff like graphics,
> etc. from, say, a word processor. There's a Scrapbook and Clipboard
> but as the number of third-party products out there to supplement
> the functions of these two devices, they just aren't enough.
> 
[Stuff deleted about why the 68000 isn't so bad]
Performance isn't the only issue. Apple is supporting a ridiculaous
number of variations in slot standard, support (or not) for color, etc.
If all Macs had a variant on the 68030, at least Apple would be able to
standardize on some significant implementation details of the OS (e.g.,
virtual memory available on all machines, provided you have the hard
disk space). You want a better IPC? With lots of little applications
sharing memory, paging becomes a virtual (sigh) necessity.


Philip Machanick
philip@pescadero.stanford.edu