Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!akgua!mcnc!ecsvax!hes
From: hes@ecsvax.UUCP
Newsgroups: net.micro,net.micro.pc
Subject: Re: PC vs. VAX/unix
Message-ID: <2115@ecsvax.UUCP>
Date: Tue, 6-Mar-84 08:30:33 EST
Article-I.D.: ecsvax.2115
Posted: Tue Mar  6 08:30:33 1984
Date-Received: Wed, 7-Mar-84 08:04:35 EST
Lines: 19

<> There is one major consideration missing in the generic question:
         Is  a micro = 1/n'th of a larger computer?
That is consideration of what job you are trying to do!  For a job
which fits comfortably on a micro, I believe that that is the best
(cheapest) way to do that job.  E.g., word processing of the typical
1-50 or so page letter/memo can be done very cheaply  (including the
user's time & effort) on a micro.
   A big number crunching type job will probably require such things
on a micro as extensive reprogramming to partition/overlay/etc. so it
fits the constraints of memory; a wall-clock execution time exceeding
the MTBF of the micro and/or the users patience/life-span; I/O of such
magnitude that the user must spend much time changing disks, etc.
   For cheap convenient running of small jobs one needs a micro; for
acceptable performance in larger jobs one needs larger machines
(maybe even larger than VAX/unix!) which are easily accessable and
ideally upwards compatible.  The best of all worlds is a micro per user
with good communications to an array of larger and larger computers.
(That's the hardware-then one needs suitable software.)
--henry schaffer  ncsu genetics