[lug] Various Arch/Compiler Binaries living together

Ed Hill ed at eh3.com
Wed Jun 26 12:15:50 MDT 2002


On Wed, 2002-06-26 at 11:52, John Karns wrote:
>
> Results are very surprising:
> 
> PIII/850:
> 26.160u 0.000s 0:26.08 100.3%   0+0k 0+0io 115pf+0w
> Xeon/1.7:
> 24.300u 0.000s 0:24.30 100.0%   0+0k 0+0io 114pf+0w
> 
> Does it make any sense? Both CPUs have the same size cache (256kB)  CPU
> clock is two times faster on Xeon. The memory is four times faster on Xeon
> (400MHz RAMBUS vs. 100MHz bus in PIII). So what could be the possible
> reason for such strange results. I can't imagine that Xeon's floating
> point operations take twice as many clock ticks to complete as on PIII


Thanks for posting that!

If you care about floating point and have been following the information
on sites like spec.org, aceshardware.com, and tomshardware.com you
wouldn't be so surprised.  The P4 has *rotten* legacy x87 support.  The
only way you get good floating point is by having a compiler that can
generate the SIMD (MMX/SSE/SSE2) instructions.  And gcc, AFAIK, has
pretty poor support in that area.  So what you're measuring is the
ability of the chips to execute x87 code.

Heres a sample from a couple of my machines:

  PIII/900 (ThinkPad A22p):
  34.580u 0.020s 0:34.60 100.0%   0+0k 0+0io 112pf+0w

  Athlon "XP 1700+" (1.5 GHz)
  17.660u 0.020s 0:17.76 99.5%    0+0k 0+0io 112pf+0w

Notice how the Athlons *really* shine on x87 code.  They have more x87
add/multiply pipelines and they totally trash the P4s.


> On other, not-so-stupid applications, the Xeon performs roughly 1.7-2
> times faster than PIII. Still, I'm very curious why is it so bad on this
> simple program.

Yeah, thats integer performance.  The P4 does have pretty good integer
performance in some areas.

Ed


-- 
Edward H. Hill III, PhD 
Post-Doctoral Researcher   |  Emails:      ed at eh3.com, ehill at mines.edu
Division of ESE            |  URL:         http://www.eh3.com
Colorado School of Mines   |  Phone:       303-273-3483
Golden, CO  80401          |  Fax:         303-273-3311
Key fingerprint = 5BDE 4DA1 66BE 4F7B BC17  3A0C 932B 7266 1E76 F123



More information about the LUG mailing list