Author: Serge Desmarais
Date: 22:18:50 09/27/98
Go up one level in this thread
On September 27, 1998 at 20:23:09, John Coffey wrote: >On September 27, 1998 at 19:49:42, Serge Desmarais wrote: > >So is it not the >>pruning of the lines that allow the program to reach a depth of 11 plies (if not >>even more)? >> >>Serge Desmarais > > >Based upon my limited understanding, Null move pruning is cheap way to get >a huge savings because if your opponent can't improve his position by >making two moves in a row, then you can eliminate the whole branch of the >tree. This might lead to some inaccuracies, and Hyatt did say it is >potentially dangerous, but the benefits seems to outweigh the risk by far. > >It is my understanding that a null move search allows the opponent to make >2 moves in a row, and is done to depth -1, so the search takes about 25 >times less than what the search would have taken otherwise. So if the >null move search fails, we have only added about 4% to our computation time. >But if on the other hand the null move search succeeds, then we save 96% of >the time on that particular branch of the tree. > >I hope that I got the facts right. If not, I hope somebody sets me straight. >:-) > >John Coffey I have been told by Bruce, once in ICC, that Ferret uses the Null Move, and he thought Fritz probably do, but he also said the not all the programs use it. Mostly, the "fast searchers" use it while those based on "knowledge" more than on the search could not be using it. What about HIARCS or MChess, for example? I have been told that MChess doesn't reach a profound depth in tournament time? I own Fritz 5 and it reaches astonishing depths. I also know that in some kind of positions, the null movers are very weak/unable to find a/THE solution, whatever time given... I do not think Genius uses the null move. Serge Desmarais
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.