Author: Uri Blass
Date: 10:13:49 11/17/02
Go up one level in this thread
On November 17, 2002 at 12:32:21, Russell Reagan wrote: >On November 17, 2002 at 09:31:07, Bob Durrett wrote: > >>Are there just a few different kinds of good move generators > >There are only a few that I'm aware of. Mainly 0x88 or other array based >approaches, move table, and bitboards. You should be able to find plenty about >0x88 and bitboards by searching the net or the CCC archives. You can search >Google groups for posts by Bruce Moreland explaining move table generation that >he used in his program Ferret. > >> In other words, if the move generator used is inefficient or somehow not right, >>would that mean that the chess engine would never be any good? > >Move generation is not the main task of a chess engine. It is a significant >part, but it's not absolutely critical. If you have a poorly written move >generator, but you have a very smart search and good evalution, you can still >have a strong chess program. It depends how poor is your move generator. My first move generator could make less than 10000 legal moves per seconds on 800Mhz and also had bugs of generating illegal moves or ignoring legal moves. I bet that with this move generator you are going to have big problems to get a strong program. Uri
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.