Author: Aivaras Juzvikas
Date: 01:39:38 08/03/04
Go up one level in this thread
On August 03, 2004 at 00:56:48, Scott Gasch wrote: >On August 02, 2004 at 16:46:23, Gian-Carlo Pascutto wrote: > >>On August 02, 2004 at 16:44:36, Cesar Contreras wrote: >> >>> >>>Is supose changing: >>> >>>value_capture=(piece_value[victim_piece_type]-piece_value[attacker_piece_type]); >>> >>>To >>> >>>value_capture=(piece_value[victim_piece_type]-piece_value[attacker_piece_type]/10); >>> >>>will do the trick. Supossing PAWN_VALUE = 100 >>>I'm doing this, because i need bigger diferences between capture values because >>>i use piece square table as a (small )factor to order captures. Is it wrong to >>>use piece square tables on ordering captures? >> >>Sounds like a very good idea to me. Capturing a well placed piece with >>a badly placed one should on average be better than the reverse. >> > >Every time I have messed around with using psqts to order moves I've found that >my trees get bigger. I read on Ed's Rebel page that they work well for him, >though. Does anyone here have similar good luck using psqt in move ordering? > >Scott piece square tables for move ordering work better for me than history heuristics
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.