Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/5/84; site ssc-vax.UUCP
Path: utzoo!watmath!clyde!cbosgd!ihnp4!houxm!vax135!cornell!uw-beaver!ssc-vax!gml
From: gml@ssc-vax.UUCP (Gregory M Lobdell)
Newsgroups: net.ai
Subject: Do you really need an AI work station
Message-ID: <284@ssc-vax.UUCP>
Date: Fri, 11-Oct-85 21:15:54 EDT
Article-I.D.: ssc-vax.284
Posted: Fri Oct 11 21:15:54 1985
Date-Received: Mon, 14-Oct-85 04:24:33 EDT
Organization: Boeing Aerospace Co., Seattle, WA
Lines: 24

While I really don't know your situation, here are some thoughts:

In a previous article someone ranted and raved about their 3600 and
their vax that weren't connect, etc...

All that I can say to you is, for shame.  That 3600 should be
connected and you should be using it for ALL your files.  Granted, a
LispM is a single user machine and thus if you have less machines
than people you have problems, but people don't work on LispMs 100%
of the time.  Also you don't have to boot the machine every time you
walk up to it.  I'd be willing to bet that it takes longer for the
vax to boot than the LispM.  And you can't beat the development
environment of a LispM!!!  

On the topic of performace, here at Boeing, we developed and expert
system on a vax and it ran twice as slow as real time, that is it
had an internal clock and that clock was only able to run and half
real time.  The whole thing was ported to a Lisp Machine and it ran
18 times faster than real time.  That is an improvement of 36 times.
This is without any attempts at making it really fly.

Gregg Lobdell
Boeing Aerospace
>> Does a corporation have opinions??? <<