🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Article View: comp.sys.powerpc
Article #3494

UseMarks -- our own benchmarks

#3494
From: Gregory L Hansen
Date: Fri, 29 Mar 1996 18:59
53 lines
2809 bytes
Instead of arguing about which platform is faster,  let's prove it with
our own benchmarks, the UseMarks.  We'll publish standardized programs
that everyone will run on their own machines and publish publicly,
with system info.  There will be a raw score, a normalized score, and
a relative score.  The raw score is the number of seconds taken to
complete the task.  (Don't worry about tenths of a second; I'm assuming
the difference between 60 seconds and 60.4 seconds is negligible and
nobody really cares.)  The normalized score is how long it would have
taken your machine to complete the task at 100MHz, for an accurate
per megahertz comparison.  So if you're running a 75MHz machine, you
can find the relative score by raw*75/100, to get a score 1/4 lower
than if you had run on the faster chip.  The relative score is for
a quick feeling of how the chips compare.  Since I'm making this up,
I reserve the right to choose the standard, which will be the 100MHz
PPC601, since that's what I have.  Divide your relative score by the
standard score to find out how many times faster, per megahertz, your
machine was than the PPC601/100.  Add references to compiler notes
and put explanations at the bottom for compiler type and optimizations.
Assume no optimizations were used that aren't explicitly stated.
PowerPC, Intels, Alphas, SPARCS, and all other processors welcome.
    
Someone will likely come up with some better tests, but I'll start with
the Hansen Floating-Point test.  As you can probably guess, it's just
a bunch of floating-point operations kludged together to take up
processor time.
 
/*** begin ***/  
#include <stdio.h>
#include <math.h>
 
void main(void)
{
	double	sum, i;
	
	sum = 0;
	for(i=1;i<10000000;i++)
	{
		sum = sum + sqrt(sin(i/4.2)*sin(i/4.2) + log(i));
	}
	printf("total: %f", sum);
}
/*** end ***/
  
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
Technical Details
Message-ID: <4jhbus$dga@nuke.csu.net>
Path: arc1-ca.pugleaf.net!newshub.csu.net!ghansen
References: <nevin-0602962253080001@150.135.1.79> <4j7ek8$vst@news.erie.net> <jragosta-2703960830420001@ppp-2016.d <danpop.827971541@rscernix>

Show Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!newshub.csu.net!ghansen
From: ghansen@isc.sjsu.edu (Gregory L Hansen)
Newsgroups: comp.sys.powerpc,comp.os.linux.advocacy,comp.sys.mac.advocacy,comp.os.os2.advocacy,comp.sys.amiga.advocacy,comp.sys.intel
Subject: UseMarks -- our own benchmarks
Date: 29 Mar 1996 18:59:40 GMT
Organization: San Jose State University, San Jose, California
Message-ID: <4jhbus$dga@nuke.csu.net>
References: <nevin-0602962253080001@150.135.1.79> <4j7ek8$vst@news.erie.net> <jragosta-2703960830420001@ppp-2016.d <danpop.827971541@rscernix>
NNTP-Posting-Host: athens.sjsu.edu