Author: Rafael Andrist
Date: 08:11:29 11/18/01
Go up one level in this thread
On November 18, 2001 at 10:34:12, Uri Blass wrote: >On November 18, 2001 at 10:14:09, Rafael Andrist wrote: > >>On November 18, 2001 at 10:03:15, piet de hoop wrote: >> >>>The next position occured in Diep versus Rebel Century 4 round 6 of the Dutch >>>Open Computer Chess 2001 >>> >>>[D] 8/6p1/3k3p/8/3K4/7P/8/8 b - - 0 43 >> >>If the program doesn't know it is a draw, then Black won't give his extra pawn >>away in the search. To recognize a draw by 50 move rule, several hundreds of >>plys are necessary. > >I think that it is possible to prove a draw by repetition in less plies. >I did not check it but my intuition tells me that the black king has not too >many squares to defend the pawns so it seems that less than 100 plies are needed >for a repetition or losing a pawn for black even after considering the fact that >the black pawns may go forward every few plies to avoid repetition. > >Uri Your intuition seems to be right, Uri. I experimentally told my chessprogram Wilhelm to look at it as a co-ordinated squares problem. To hold the draw score which I immediately get, Wilhelm starts to advance pawns. So probably less than 100 plys are needed, but I guess it will be around this depth. Rafael B. Andrist
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.