Path: utzoo!attcan!uunet!wuarchive!usc!elroy.jpl.nasa.gov!ncar!midway!quads.uchicago.edu!rtp1
From: rtp1@quads.uchicago.edu (raymond thomas pierrehumbert)
Newsgroups: comp.sys.apollo
Subject: Re: ADUS conference news
Message-ID: <1990Oct18.155345.3622@midway.uchicago.edu>
Date: 18 Oct 90 15:53:45 GMT
References: <901015.16343798.030626@RMC.CP6> <1990Oct16.064450.2850@alphalpha.com> <1990Oct17.181313.7970@quintro.uucp>
Sender: news@midway.uchicago.edu (News Administrator)
Organization: University of Chicago
Lines: 25

What is the reason for buying a new DNxx machine?  Well, for one the
DN10000 is still a better number cruncher than the new line, especially
if it is programmed effectively (and I hope the new compilers do
better than the pitiful performance of 10.7p)  As for upgradability,
the DN10k is fully upgradable at a reasonable cost (for universities
a board swap to the new double speed processor is about $6000, I think,
which puts it cheaper than getting the performance increase by buying
two to four SparcStations).  (Memory and disk drives for the 10k are
ruinously expensive, and there is little or no third party market.
Wake up HP/Ap! This is the main sticking point in further sales, at
least to me).  Anyway, if I were buying a machine right now,
maybe I'd look at Stardent or the IBM RISC series, but it still looks
to me like the 10k is a strong competitor.  And consider this:  You
can get in on the ground floor of a 10k at a university for about
$50k, and for purchases of around $15k per year (which you can
get relatively easily on individual grants) you can double to
quadruple your throughput each year by buying additional processor
boards or upgrading to faster boards.  This is why I bought the
machine.
    Despite my grumbling, basically the 10k has been very good for
my work, and I see some real performance improvements coming down
the line which make its performance advantages clearer.  Thus,
I am very chagrined to hear the rumblings that HP may not support
further OS releases on the DN series.
.