Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Crafty 16.18 benchmarks on UltraSparc and Athlon

Author: Robert Hyatt

Date: 20:03:56 09/16/99

Go up one level in this thread


On September 16, 1999 at 10:42:02, walter irvin wrote:

>On September 16, 1999 at 06:35:03, Owen Lyne wrote:
>
>>Benchmarks for Crafty 16.18
>>---------------------------
>>
>>167 MHz UltraSparc (Solaris 2.6, compiled with gcc 2.8.1)
>>
>>Total nodes: 61096154
>>Raw nodes per second: 69035
>>Total elapsed time: 885
>>SMP time-to-ply measurement: 0.723164
>>
>>250 MHz UltraSparc (Solaris 2.7, compiled with gcc 2.8.1)
>>
>>Total nodes: 61096154
>>Raw nodes per second: 95612
>>Total elapsed time: 639
>>SMP time-to-ply measurement: 1.001565
>>
>>600 MHz Athlon (wcrafty-16.18.exe from Bob's ftp site)
>>
>>Total nodes: 68509284
>>Raw nodes per second: 422896
>>Total elapsed time: 162
>>SMP time-to-ply measurement: 3.950617
>>
>>Notes:
>>I'm very pleased with my new AMD Athlon 600MHz - that looks like
>>a pretty good benchmark for Crafty to me. Totally vanilla settings, just
>>downloaded the exe file and double-clicked on it, worked like a dream. Next stop
>>Winboard!
>>
>>I've also included bechmarks from the machines at work, this is a compile of my
>>own (I hope we'll get gcc2.95 soon and speed things up), as Bob says, these
>>machines don't see to perform well MHz for MHz and I know they were expensive.
>>
>>Judging by these numbers it would need a 1000MHz UltraSparc to perform as well
>>as my 600MHz Athlon.
>>
>>Owen
>cool i was thinking of getting me one but was not sure what speeds i could
>expect ? now i think i'll go ahead and get one ,looks fast enough for me .


If you want raw speed, get a 21264 alpha.  Nothing else even comes _close_...




This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.