Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1exp 10/6/83; site ihlts.UUCP
Path: utzoo!linus!security!genrad!decvax!harpo!eagle!mhuxi!houxm!ihnp4!ihlts!rjnoe
From: rjnoe@ihlts.UUCP
Newsgroups: net.chess
Subject: Re: Fourth World Computer Chess Championship
Message-ID: <228@ihlts.UUCP>
Date: Fri, 14-Oct-83 16:01:04 EDT
Article-I.D.: ihlts.228
Posted: Fri Oct 14 16:01:04 1983
Date-Received: Sat, 15-Oct-83 03:51:56 EDT
References: <217@ihlts.UUCP> <2463@utcsrgv.UUCP>
Organization: AT&T Bell Labs, Naperville, Il
Lines: 19

I really think David Slate et al. from Northwestern University have taken
their design about as far as it will go.  Meanwhile, the impression I have
is that Belle still has a definite edge over the other machines.  Cray
Blitz is frequently underestimated and Kaissa has usually been strong in
the World Computer Chess Championships, winning the first back in 1974.
I recognize that I may be a little biased toward Belle (being an AT&T Bell
Laboratories employee) but I really think that Ken Thompson and Joe Condon
have done a truly outstanding job and with much less pre-packaged computing
hardware to work with than other teams have had.  (Cray Blitz of course has
a Cray 1 behind it and the Soviets have apparently spent many rubles for
their chess machine.)  This to me indicates superior design and execution,
not to mention more economical.

Belle, last I heard, was doing about a 2150-2200 rating.  Cray Blitz can't
be over 2000, I'm fairly sure.  I don't know a rating estimate for Kaissa.
And Nuchess will end up in fourth place (or thereabouts) I would think.

-- 
		Roger Noe		...ihnp4!ihlts!rjnoe