🚀 go-pugleaf

In article <aqtjo-2903960750290001@129.219.116.52>, aqtjo@imap2.asu.edu
(Ted Oliver) wrote:

> In article <stephenkDp0JpA.F6u@netcom.com>, stephenk@netcom.com (Stephen
> Knilans) wrote:
> 
> Munch, munch, munch, a simple benchmark story I ate for lunch :-)
> 
> > And the result is:
> >   
> > Hansen Floating-Point
> > Machine   OS      Processor   Raw  Normalized  Relative  Compiler  
> > --------  ------  ----------  ---  ----------  --------  --------
> > Power100  Sys7.5  PPC601/100  82s     82s          1     1,2
> > ==================================================================
> > 1: CodeWarrior 8
> > 2: optimized for speed

For completeness -- on CW8, on a PowerMac 6100/60, 24M RAM, no L2 cache:
            7.5.1 PPCc01/60     145s
and a PowerComputing PowerCurve with 256K cache
            7.5.2    PPC601/120  65s


> 1) using a single measurement
> 2) not indicating other system information (RAM, HD, whether the Power100
> had L2 cache, etc.)

I don't think all this matters for a toy benchmark that hardly hits
memory. The fact that my 120MHz figure is a little better than would be
predicted by clock speedup suggests one of 2 things -- 7.5.2 comes with a
better float library, or the original CW8 timing didn't use all possible
optimizations (which I did -- I turned on 601 scheduling, and all levels
of optimization). Since my 6100/60 time is not as fast (relatively), the
latter explanation doesn't look so good.

Enough on this though -- you can't really prove anything with a toy benchmark.
-- 
Philip Machanick             Department of Computer Science
University of the Witwatersrand,   2050 Wits,  South Africa
phone 27(11)716-3309                     fax 27(11)339-7965
http://www.cs.wits.ac.za/~philip/       philip@cs.wits.ac.za