Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!apple!bloom-beacon!mit-eddie!uw-beaver!rice!sun-spots-request
From: news%crltrx.DEC@decwrl.dec.com (USENET News System)
Newsgroups: comp.sys.sun
Subject: Re: SUN SPARCstation 1 vs. DEC 3100
Keywords: Hardware
Message-ID: <3530@kalliope.rice.edu>
Date: 24 May 89 17:48:31 GMT
Sender: usenet@rice.edu
Organization: Sun-Spots
Lines: 18
Approved: Sun-Spots@rice.edu
X-Sun-Spots-Digest: Volume 8, Issue 6, message 2 of 12

Yup.  And the DECstation (PMAX) is alot faster than the SPARCstation 1,
particularly for floating point.

Personally, I don't use or want tape drives on all machines; and my
personal opinion of the TK50 is hardly printable in public.  We have 20
PMAXes in our lab, and we may have one of those tape drives, and in fact
generally use the tapes off our file servers.  So if you knock of the tape
drive (which does increase our entry price for your first machine if you
don't have any tape drives on any server somewhere; as I said, I'm not a
fan of it...) more realistic comparisons are alot closer (given your
numbers; I've not studied Sun's price lists...)  I certainly don't buy a
tape drive with each machine....

So I suspect we're on similar price / performance curves for integer
applications, and well ahead of the Sparcstation for floating point
applications. 

- Jim Gettys