Author: Gerd Isenberg
Date: 11:06:16 09/13/04
Go up one level in this thread
On September 13, 2004 at 12:55:05, Alessandro Scotti wrote: >On September 13, 2004 at 05:27:57, Gerd Isenberg wrote: > >>Further node definitions are root-nodes, and minimal game tree node kinds, like >>PV-nodes, cut-nodes and all-nodes. I also use most left nodes after each root >>move, PV-node or refutation line node. > >BTW, I see that GNUChess keeps the node type available during a search... is it >useful to have? Except pv-nodes i don't determine cut- and all-nodes yet. Believing in a rather perfectly ordered tree it might be usefull for several reasons and heuristics. Pruning decisions and nullmove observation, extensions. Tree-Splitting in a multi-threading environment is an issue too i guess. And all the nice ideas of Yngvi Björnsson, Tony Marsland and others: http://www.cs.ualberta.ca/~yngvi/ http://www.cs.ualberta.ca/~yngvi/html/publications.html http://www.cs.ualberta.ca/~yngvi/Papers/tcs2001.pdf btw. while googling Yngvi, i found following german paper: http://www.ke.informatik.tu-darmstadt.de/lehre/ss04/se-spiele/LearningSearchControl.pdf Gerd
This page took 0.01 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.