Author: Stuart Cracraft
Date: 07:18:52 10/17/04
Go up one level in this thread
On October 16, 2004 at 22:31:45, TEERAPONG TOVIRAT wrote: >On October 16, 2004 at 10:58:17, Ron Peterson wrote: > >> I have seen some messeages about the use of bitboards for move generation. >>I use a offset approch for my move generator and I am pleased with the speed. I >>have enclosed a copy of my "gen_white_knight" code. >> My question is what type of speed increase wuold you expect to get in a >>move generator using bitboard? 50%? 100%? > >Hi, > >I think bitboards generate move faster especially capture moves. But it's slower >in makemove() and unmakemove() because it has to update a lot of data. >However, you can do a lot of tricks with bitboards in eval(). So,if you want to >compare the two methods, you 've to compare the performance not only speed of >movegen(). > >Regards, >Teerapong Movegen is not the issue with bitboards. In fact, it is a drawback. Much harder to implement movegen with bitboards than with 0x88 or mailbox. The joy is in bitboard eval-ing. You should make a hybrid program at first using mailbox or 0x88 movegen, regular incheck/attack routines, and bitboards for movegen. Later, when you have more time or want bigger challenges, convert to bitboarding using examples like GNU Chess and Crafty. I think the next challenge above that is to do what Slate and Atkin did and that is to maintain incremental attack tables that are updated after each makemove and unmakemove. Stuart
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.