Author: Bruce Moreland
Date: 13:21:20 12/19/02
Go up one level in this thread
On December 19, 2002 at 08:24:30, Rolf Tueschen wrote: >How about giving just one single good position to explain what you were talking >about. I think one diagram is better than thousand words. I'm surprised that you >see so many difficulties in special for the middle game, Bruce. I was >extrapolating from my average chess knowledge, but couldn't make such >conclusions of course. I don't have a diagram that would clear anything up. >Is nullmove what I said or is it the key for the future of computer chess? I >said that nullmove is more PR for the average amateurs, but the super GM have >another source to exploit typical computer weaknesses. What do you think here? Null-move forward pruning is just a performance optimization. It may be possible to tweak it and fiddle with it so that it is even more significant. That is what Omid is trying to do with his VR=3 optimization. The hardware will always carry us forward, so it doesn't really matter if anyone invents anything new, everyone will still get better. It may be possible to discard these shared search techniques in favor of something stronger, but if I knew what that might be, I'd do it. bruce
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.