From news.funet.fi!fuug!mcsun!uunet!kddlab!cs.titech!wnoc-tyo-news!sh.wide!astemgw!icspub!chorus.mei!panasonic.com!mips!zaphod.mps.ohio-state.edu!qt.cs.utexas.edu!cs.utexas.edu!swrinde!gatech!purdue!bu2.bu.edu!bu.edu!olivea!apple!apple.com!kevina Wed Feb 19 08:23:30 EET 1992
Article: 8832 of comp.lang.postscript
Path: uwasa.fi!news.funet.fi!fuug!mcsun!uunet!kddlab!cs.titech!wnoc-tyo-news!sh.wide!astemgw!icspub!chorus.mei!panasonic.com!mips!zaphod.mps.ohio-state.edu!qt.cs.utexas.edu!cs.utexas.edu!swrinde!gatech!purdue!bu2.bu.edu!bu.edu!olivea!apple!apple.com!kevina
From: kevina@apple.com (Kevin Andresen)
Newsgroups: comp.lang.postscript
Subject: Re: Results of PPST, please!!!
Message-ID: <20305@goofy.Apple.COM>
Date: 13 Feb 92 23:58:34 GMT
Sender: usenet@Apple.COM
Organization: Apple Computer, Inc.
Lines: 24

In article <1992Feb11.181100.21680@adobe.com> caro@adobe.com (Perry A. 
Caro) writes:
> Benchmarking is a tricky business.  It's quite difficult to do in a way
> that yields reliable results.  There are a number of variables that
> contribute to the performance of a particular interpreter.  Because of
> the scope and complexity of these variables, benchmarks can often be
> misleading without taking these variables into consideration.

Perry raises a good point.  Another issue to consider is that the test 
assumes that usertime returns accurate results... this isn't always the 
case, especially in tight loops!

But for what it's worth, here are the results for the Apple LaserWriter 
IIg, PS version 2010.113:

Speed results (index):

Graphics:   8.5000 (6.3)
RAM/fonts:  2.5918 (25.2)
Math:      11.6797 (3.3)
Total:     22.7695 (6.9)

--Kevin Andresen [kevina@apple.com]
"You took the wind right out of my sails/It must be luff/It must be luff"