Author: Tom Kerrigan
Date: 01:06:45 01/01/00
Go up one level in this thread
On December 31, 1999 at 14:18:02, James Robertson wrote: >I think I solved the problem neatly without multithreading or anything fancy. > >Every time my program starts a search, it uses two boards; one for the game, and >one to search with (initialized every search). After the search is completed and >a move is found, my program updates the game board. > >To ponder, I copy the game board to the search board, make the 'expected' >opponent response on the search board, and search as if the opponent had already >made the move. If the opponent does not reply as expected, I simply terminate >the search, make the opponent's move on both boards, and start searching again. >If the opponent _does_ make the expected move, I make it on the game board, >notify the searching classes we are no longer pondering, and when the search >finishes I make the new search move on the game board. > >James This is also how I used to do it. I think this approach is the most common. -Tom
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.