Author: Daniel Clausen
Date: 00:10:20 06/04/02
Go up one level in this thread
On June 04, 2002 at 01:55:45, Russell Reagan wrote: >How much faster would a chess engine be if it didn't have to deal with the >special cases such as en passant, castling, promotion, etc.? In other words, >only having to deal with moving one piece from one square to another. The chess game tree grows exponentially with depth. Why bother with constant factors? :) >Would it be twice as fast? Or would more costly things such as the evaluation >function make the gain in speed less than 2x? It probably depends on a whole lot of things, like board representation. But did you already profile your engine and saw how much time you spend in things like movegen/domove/undomove compared to other things? I think most people overestimate that. Sargon
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.