Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!bronze!lothario!tagreen
From: tagreen@lothario.ucs.indiana.edu (Todd Green)
Newsgroups: comp.sys.next
Subject: Re: RAM and performance
Message-ID: <1991Apr22.185224.10680@bronze.ucs.indiana.edu>
Date: 22 Apr 91 18:52:24 GMT
References: <1991Apr22.135539.2058@ulrik.uio.no>
Sender: news@bronze.ucs.indiana.edu (USENET News System)
Organization: Indiana University Computing Services
Lines: 23

In article <1991Apr22.135539.2058@ulrik.uio.no> me@spirea.UiB.NO (My Account) writes:

>
>Cosider 16MB a realistic minimum on a NeXT.  

I agree with at least 16mb.  The diff in performance between 8 and 16
is considerable.  (At least it was on my 030 cube).


> I can't say anything about 32MB (not yet anyway ;-) but I would expect a linear improvement over 16 (i.e 32 is to 16 as 16 is to 8).  If I'm wrong I'm confident someone who knows better will point out my error.  

I would tend to disagree on this.  I would doubt that it'd be linear.
I say this because most of the performace increase seemed to be due
less swapping.  As long as everthing is still in memory I don't think
that adding 16 more meg for a non color workstation would make THAT
much of a difference.  Now if you're talking about a server or a color
workstation that is being used to run memory intensize programs then
yes it probably would help, but for the average user I don't think so.

-- 
Internet: tagreen@bronze.ucs.indiana.edu
NeXTMail: tagreen@lothario.ucs.indiana.edu
BitNet:	  tagreen@iubacs.bitnet