Author: Stefano Gemma
Date: 00:33:59 07/25/04
Go up one level in this thread
On July 24, 2004 at 15:46:01, Gerd Isenberg wrote: >On July 24, 2004 at 04:51:39, Stefano Gemma wrote: > >>On July 23, 2004 at 18:33:36, Gerd Isenberg wrote: >> >>>Is your getNextMove O(N**2)? >>[...] [...] >that sounds interesting. I guess your engine is pretty fast ;-) - starting position - Pentium 4 - 2.8MHz - only alpha-beta and iterative deepening (no hash/null/killer/quiescent... etc) - genetical algorithm disabled ply 8 nodes=65392707 msec=42281 (about 42 seconds) n/s=1546621 >I like the idea of a finite state generator too. If some concrete moves for a >node are feeded back by hash, killer-heuristic and probably hints from eval, it >is nice to check for legality only and then to make the moves, hoping for a >cutoff. It is easy to do. You just have to save the piece and the destination square. With a table, or a simple algorithm, you can get the delta that lead from the source square to destination. Then, with a move generator like mine (or more easyest with BitBoard), it is immediate to check for the move is legal. Ciao!!! Stefano Gemma
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.