Author: Magoo
Date: 08:41:29 05/06/03
Go up one level in this thread
On May 06, 2003 at 02:37:07, Tony Werten wrote: >On May 05, 2003 at 20:38:50, Magoo wrote: > >>On May 05, 2003 at 19:38:47, Zach Wegner wrote: >> >>>On May 05, 2003 at 05:02:24, Magoo wrote: >>> >>>>No, thats not the problem, i have a global move stack. >>> >>>But after each make/undo, you seem to be regenerating the moves. Calling the >>>extraneous functions as much as you do would not create even close to that much >>>of a slowdown. >> >>Actually, silly me forgot to re-compile when i got those figures (i had -pg for >>profile), i'm now only about 10x slower than other "amateur" programs. >>This is what i get after compiling without profile and with -O3. >>Depth:5, Nodes:5072212, Time:136 sec, Nps:37295, Value:104 > >That speed is now something you can work with. Now for the other problem ;) Are >you doing an alpha/beta search ? Or plain minimax ? Your amount of nodes is very >high. > >Tony Well, i was doing alpha/beta before, but when i noticed my program was so much slower than other programs i had to go back, so im actually only doing minmax walks and timming to see the improvments, if you know what the perft command is, that is what im doing, perft 5.
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.