Author: blass uri
Date: 12:29:35 04/04/98
Go up one level in this thread
my idea is that a programmer may do 2 programs the first is the original chess program and the second is a program that generates a tree ( after it gets an input of an evaluation function of a position(generated by the first program) it decides what position to analyze and how much time it should be analyzed by the originial program) the program that generate the tree can decide to analyze a position that was analyzed before for longer time or to analyze the next best move in a position that was not analyzed. a good idea is to analyze the main line for a longer time and to analyze forced moves that can be identified by the numbers in the tree.
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.