Author: José Carlos
Date: 17:00:15 08/03/04
Go up one level in this thread
On August 03, 2004 at 15:04:54, Cesar Contreras wrote: >Hi > >I'm working on extended futility prunning and it's not working, it seems to make >a node reduction, but it plays worse than without it. > >My question is, do any of you have seen good improvement using extended futility >prunning? > >what normal futility margin do you use? >what extended futility margin do you use? > >Thank you. If my memory is correct, most programmers report it doesn't work for them. For me, it seems to work not so bad, but the improvement, if any, is very small (I don't have statistical proof to support it; I just observed it). But I don't do extended futility pruning as described by Heinz. I use a margin of 300 both in depth = 1 and depth = 2. My idea, which I haven't been able to prove wrong so far, is this: when you enter quiesce, you're allowed to try captures or stand pat. You can't stand pat before, so nodes with depth = 2 means that I can move now, then my opponent will move and it's me who will have the right to stand pat or keep searching. So going from depth = 1 to depth = 2 doesn't give my opponent an advantage. It might be terrible to the rest of the programs, but for me it seems to work. But futility is one of those pruning techniques that work very differently depending of the rest of the program (which is not the case of null move, for example, that works fine for 99% of the programs). You shouldn't be surprised if it doesn't work for you. José C.
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.