Subject: Re: mindless boredom, speed and compiling kernels
To: Chris Jones <cjones@honors.montana.edu>
From: Simon Burge <simonb@telstra.com.au>
List: port-pmax
Date: 05/21/1998 21:44:16
On Tue, 19 May 1998 14:56:14 -0600 (MDT)  Chris Jones wrote:

> On Tue, 19 May 1998, Simon Burge wrote:
> 
> [lots of interesting stuff deleted]
> 
> > On an entirely different note, here's some timings for the "crafty" chess
> > program calculating it's own performance (over three runs):
> > 
> > 	DECstation 5000/260 NetBSD 1.3.1 - NetBSD binary (gcc 2.7.2.2+myc1 -O2):
> > 		generated 480769 moves per second
> > 		generated 479616 moves per second
> > 		generated 480769 moves per second
> > 
> > 	DECstation 5000/260 NetBSD 1.3.1 - Ultrix binary (gcc 2.7.2.2 -O2):
> > 		generated 614088 moves per second
> > 		generated 614576 moves per second
> > 		generated 615244 moves per second
> > 
> > 	DECstation 5000/260 Ultrix 4.5 - Ultrix binary (gcc 2.7.2.2 -O2):
> > 		generated 587867 moves per second
> > 		generated 589220 moves per second
> > 		generated 587868 moves per second
> 
> Hmm.  So, does NetBSD run the Ultrix binary faster than the native one
> because the Ultrix one isn't ELF?  Or is there some other reason I'm
> missing?  I know I've seen NetBSD/i386 run simple benchmarks faster than
> Linux, and I assumed that was because of the a.out/ELF difference, but I
> don't really know...

Not sure.  Certainly the NetBSD binary (which compiled with -static)
is larger than the Ultrix binary.  Maybe it's a cache thing.  FWIW, a
-static NetBSD native binary got around 5200000 moves per second.

	vlad:chess/crafty/crafty-15.7 15> size crafty-pmax
	text    data    bss     dec     hex     filename
	764704  32708   318696  1116108 1107cc  crafty-pmax

	balrog:chess/crafty/crafty-15.7 89> size crafty-ultrix
	text    data    bss     dec     hex
	405504  98304   309280  813088  c6820

Both were compiled with gcc -O2.

Simon.