*BSD News Article 20934


Return to BSD News archive

Xref: sserve comp.unix.misc:9102 comp.os.linux:54770 comp.os.386bsd.misc:994
Path: sserve!newshost.anu.edu.au!munnari.oz.au!network.ucsd.edu!news.cerf.net!pagesat!olivea!spool.mu.edu!howland.reston.ans.net!gatech!swrinde!cs.utexas.edu!uunet!pipex!uknet!mucs!m1!qinyi
From: qinyi@cs.man.ac.uk (Yi Qin (BCW PhD))
Newsgroups: comp.unix.misc,comp.os.linux,comp.os.386bsd.misc
Subject: Re: What are the various PC bassed Unix box OS?
Message-ID: <QINYI.93Sep14111941@r6i.cs.man.ac.uk>
Date: 14 Sep 93 10:19:41 GMT
References: <1993Sep02.184251.23903@engr.engr.uark.edu>
	<2C8E9252.3016@news.service.uci.edu>
	<michaelv.747547179@ponderous.cc.iastate.edu>
	<1993Sep12.235912.32501@unixland.ci.net>
	<1993Sep13.084852.1164@black.ox.ac.uk>
	<1993Sep13.121546.19533@taylor.uucp>
Sender: news@cs.man.ac.uk
Followup-To: comp.unix.misc
Organization: Department of Computer Science, University of Manchester
Lines: 16
In-reply-to: mark@taylor.uucp's message of 13 Sep 93 12:15:46 GMT

Are we missing anything here when talking about performance?
Besides the factors of CPU speed, local bus and memory size,
the performance is also very much dependent on the memory
access patterns in the benchmark/application programs. Unless
we use the same code to test different machines, the result
would not hold in every sense. That's often the trick played
by the dealers and manufacturers, I believe.

So you guys may pass on the test programs to each other and
run it on different machines. Then the result may tell you
a bit more. 


Yi Qin
Oxford
UK