Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: Notesfiles $Revision: 1.7.0.10 $; site uiucdcsb
Path: utzoo!watmath!clyde!cbosgd!ihnp4!inuxc!pur-ee!uiucdcsb!jabusch
From: jabusch@uiucdcsb.CS.UIUC.EDU
Newsgroups: net.micro
Subject: Re: What if IBM Had chosen the 68000? N
Message-ID: <4400129@uiucdcsb>
Date: Sat, 23-Nov-85 15:51:00 EST
Article-I.D.: uiucdcsb.4400129
Posted: Sat Nov 23 15:51:00 1985
Date-Received: Sun, 24-Nov-85 11:12:33 EST
References: <456@looking.UUCP>
Lines: 37
Nf-ID: #R:looking.UUCP:456:uiucdcsb:4400129:000:2025
Nf-From: uiucdcsb.CS.UIUC.EDU!jabusch    Nov 23 14:51:00 1985



	Yes, IBM does make the CS-9000, which is 68k-based.  I hate it!
I've used it for classwork, with Xenix.  It's incredibly slow.  We have
people around here who use them (the 9000's) a lot, and I think that
most of them would agree.  A lot of this is due to the fact that they
use the older hard disks, which are slower than those of an AT.  Not 
all the speed difference is due to this, though.  I've compiled programs
on my PC faster than programs compiled on the 9000, and the 9000 was
not loaded down by other terminals or users at the time.
	My experience with the 9000's left me very disappointed.  They
were (are still?) very unreliable, although some argued that this was
because of the lab conditions.  I've seen them go down in other areas,
also, and stick to my guns.
	The AT runs several times faster under Xenix, and is far more
preferable in terms of reliability, excluding the problems with the 
hard disks that were so common at release time.  I have never seen an
AT running Xenix at clock speeds faster than 6Mhz, and don't know what
the clock speed is for the 9000, so perhaps someone else around here
with more insight to the relative merits of each machine can lend a 
hand.  On the whole, though, the AT is a nice machine, especially for
the user who does little or no programming.  
	I like to program when I have the time, and prefer to stick
to higher level languages anyway, so the segmentation does not really
bother me that much.  I rely on the compiler to solve those problems
for me.  I have done little machine-language programming, as most of
my programs did not need the extra bit of tweeking that the compiler
couldn't provide.  Of course, there are those of you who do like that 
sort of thing a lot, and all power to you.

John W. Jabusch
	Department of Computer Science
	University of Illinois at Urbana-Champaign
        CSNET:	jabusch%uiuc@csnet-relay.ARPA
	UUCP:	{ihnp4,convex,pur-ee}!uiucdcs!jabusch
        USENET:	...!{pur-ee,ihnp4}!uiucdcs!jabusch
        ARPA:	jabusch@uiuc.arpa