Author: Sean Mintz
Date: 19:50:06 12/30/01
Hi all,
I've been writing my own engine for a while now and lately have seen some very
strange behavior from it. Here's what it displays after being compiled with gcc
2.95.3-5 (cygwin special) and intel c 5 with no optimizations ("gcc *.c" and
"icl *.c -Qipo" respectively):
rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1
ply time nodes nps
1 0 20 20 k
2 0 400 400 k
3 0 8902 8902 k
4 94 165375 1740 k
5 1328 2474315 1861 k
6 32140 59765742 1859 k
as you can see, things mess up after ply 3.
here's the output after compiling w/ the intel c compiler using profiling ("icl
*.c -Qipo -Qprof_gen -Qprof_dir c:\profiles" then "icl *.c -Qipo -Qprof_use
-Qprof_dir c:\profiles"):
rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1
ply time nodes nps
1 0 20 20 k
2 0 400 400 k
3 0 8902 8902 k
4 63 197281 3082 k
5 1515 4865609 3209 k
6 37782 119060324 3151 k
now it gets it right. i am ok with compiling w/ profiling for myself, but i dont
feel too good about releasing the source if the majority of people can only
compile it with problems...
any ideas what's happening here?
the source is at http://stmintz.d2g.com:8080/chess/goober/0.02/
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.