Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Intel 8 vs MSVC 6 /w Crafty

Author: Robert Hyatt

Date: 10:44:43 02/16/04

Go up one level in this thread


On February 16, 2004 at 10:43:53, Bryan Hofmann wrote:

>On February 16, 2004 at 09:48:33, Robert Hyatt wrote:
>
>>On February 16, 2004 at 06:29:09, Bryan Hofmann wrote:
>>
>>>On February 16, 2004 at 04:18:38, Peter Skinner wrote:
>>>
>>>>I have made all 4 builds available for download for people to test, and possiby
>>>>post their results. I would like to see a few:
>>>>
>>>>http://members.shaw.ca/craftychess/wcrafty-test.zip
>>>>
>>>>Peter.
>>>
>>>
>>>Here are the results for a AMD 3000+. The strange thing is the total raw node
>>>searched for the wcrafty-MSVC6-nonsmp.exe is 90002615 instead of the normal
>>>reported of all of the other compiles.
>>
>>If you didn't change hashing sizes, something is broken in the compiler
>>options...  apparently a bad optimization somewhere.
>>
>>node counts can not vary...
>
>
>The hash sizes were not changed, so it would appear that something is broken
>with the wcrafty-MSVC6-nonsmp.exe compile.

It is possible.  Of course it is possible that a bug in Crafty can cause this,
such as an unitialized local variable that gets moved around on the stack as the
optimizer plays games...


>
>
>
>>
>>>
>>>Peter Skinner's New Compiles
>>>
>>>*wcrafty-MSVC6-smp.exe
>>>Total nodes: 89942714
>>>Raw nodes per second: 1232091
>>>Total elapsed time: 73
>>>SMP time-to-ply measurement: 8.767123
>>>
>>>*wcrafty-MSVC6-nonsmp.exe
>>>Total nodes: 90002615
>>>Raw nodes per second: 1071459
>>>Total elapsed time: 84
>>>SMP time-to-ply measurement: 7.619048
>>>
>>>*wcrafty-I8-nonsmp.exe
>>>Total nodes: 89942714
>>>Raw nodes per second: 1070746
>>>Total elapsed time: 84
>>>SMP time-to-ply measurement: 7.619048
>>>
>>>*wcrafty-I8-SMP.exe
>>>Total nodes: 89942714
>>>Raw nodes per second: 1033824
>>>Total elapsed time: 87
>>>SMP time-to-ply measurement: 7.356322
>>>
>>>Other compiles
>>>
>>>*Crafty 19.10 Russell Reagan's Compile (with -DFUTILITY)
>>>Total nodes: 89942714
>>>Raw nodes per second: 1303517
>>>Total elapsed time: 69
>>>SMP time-to-ply measurement: 9.275362
>>>
>>>*Crafty 19.10 Russell Reagan's Compile (with out -DFUTILITY)
>>>Total nodes: 102625951
>>>Raw nodes per second: 1315717
>>>Total elapsed time: 78
>>>SMP time-to-ply measurement: 8.205128
>>>
>>>*Crafty 19.10 Aaron Gordon's Compile (with -DFUTILITY)
>>>Total nodes: 89942714
>>>Raw nodes per second: 1249204
>>>Total elapsed time: 72
>>>SMP time-to-ply measurement: 8.888889
>>>
>>>*Crafty 19.10 Aaron Gordon's Compile (with out -DFUTILITY)
>>>Total nodes: 102625951
>>>Raw nodes per second: 1282824
>>>Total elapsed time: 80
>>>SMP time-to-ply measurement: 8.000000
>>>
>>>*Crafty SE 19.10 Normal Mike Byrnes' Compile
>>>Total nodes: 89942714
>>>Raw nodes per second: 1096862
>>>Total elapsed time: 82
>>>SMP time-to-ply measurement: 7.804878
>>>
>>>*Crafty SE 19.10 PGO Mike Byrnes' Compile
>>>Total nodes: 89942714
>>>Raw nodes per second: 1232091
>>>Total elapsed time: 73
>>>SMP time-to-ply measurement: 8.767123
>>>
>>>*Crafty 19.10 Peter Skinner's Compile
>>>Total nodes: 89942714
>>>Raw nodes per second: 1232091
>>>Total elapsed time: 73
>>>SMP time-to-ply measurement: 8.767123
>>>
>>>*Crafty 19.10 Mingw/GCC 3.2.3 Compile (2 stage compile)
>>>Total nodes: 89942714
>>>Raw nodes per second: 1124283
>>>Total elapsed time: 80
>>>SMP time-to-ply measurement: 8.000000



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.