Author: Robert Hyatt
Date: 15:31:51 12/03/03
Go up one level in this thread
On December 03, 2003 at 16:35:46, Slater Wold wrote: >On December 03, 2003 at 16:25:59, Robert Hyatt wrote: > >>On December 03, 2003 at 15:08:59, Matthew Hull wrote: >> >>>On December 03, 2003 at 14:59:03, Robert Hyatt wrote: >>> >>>>The other day, someone was discussing WAC. As I have been working on the >>>>quad-opteron machine at AMD, I took some time to run WAC three times, one >>>>for 1 second per position, one for 5 and one for 10. The results: >>>> >>>>===================== 1 seconds per position======================== >>>>test results summary: >>>> >>>>total positions searched.......... 300 >>>>number right...................... 297 >>>>number wrong...................... 3 >>>>percentage right.................. 99 >>>>percentage wrong.................. 1 >>>>total nodes searched.............. 111851199 >>>>average search depth.............. 4.5 >>>>nodes per second.................. 6072269 >>>> >>>>===================== 5 seconds per position======================== >>>>test results summary: >>>> >>>>total positions searched.......... 300 >>>>number right...................... 298 >>>>number wrong...................... 2 >>>>percentage right.................. 99 >>>>percentage wrong.................. 0 >>>>total nodes searched.............. 320786849 >>>>average search depth.............. 5.6 >>>>nodes per second.................. 6299702 >>>> >>>>=====================10 seconds per position======================== >>>>test results summary: >>>> >>>>total positions searched.......... 300 >>>>number right...................... 299 >>>>number wrong...................... 1 >>>>percentage right.................. 99 >>>>percentage wrong.................. 0 >>>>total nodes searched.............. 259379471 >>>>average search depth.............. 4.6 >>>>nodes per second.................. 6369720 >>>> >>>>Benchmark: >>>> >>>>Crafty v19.7 (4 cpus) >>>> >>>>White(1): mt=4 >>>>max threads set to 4 >>>>White(1): bench >>>>Running benchmark. . . >>>>...... >>>>Total nodes: 109241860 >>>>Raw nodes per second: 6068992 >>>>Total elapsed time: 18 >>>>SMP time-to-ply measurement: 35.555556 >>>>White(1): >>>> >>>>That now includes the inline FirstOne()/LastOne()/PopCnt() 64 bit code I >>>>wrote. It is about 4-5% faster. I have not written the attack stuff yet >>>>but I suppose I might bite the bullet to see what happens... >>> >>> >>>Holy smokes. Is this still gcc and no profiling? >>> >>>MH >> >> >>yes. gcc + profiling dies an ugly (and noisy) death, complaining about >>corrupted branch probability files. I've given up temporarily on getting >>that to work... > >What happens if you try to profile with 'mt 1'? There is not a chance in hell of profiling with mt > 1. I've looked at the gcc code a while back and it is not thread-safe. So I profile with mt=0 but even that fails. Works fine with my intel compiler of course, but not with gcc... > >A while back I had the Intel compiler bitch at me because of SMP profiling. So >I just profiled with 'mt 1' and everything worked fine. Still got a kick ass >speedup too. > >>I am looking at other optimizations however, so it might go a bit faster >>if I am lucky. >> >>remember that this is a quad 1.8ghz opteron. 2.2's are around. And there >>are also 8-way and beyond boxes as well. :) > >What's the speedup between 1, 2, and 4 CPUs? Any idea on the speedup of going >to 64-bit? Just note the raw speeds. about 1.5M nps on a single 1.8ghz opteron. I get 2.5M on a dual 2.8 xeon. That's not bad. the speedup for 1-4 is still about 3.1X as always... > >I know a 2x2.0Ghz Opteron system, 32-bit does around 2.2M on 19.4.
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.