Author: Uri Blass
Date: 10:31:27 06/29/03
Go up one level in this thread
On June 29, 2003 at 04:41:06, Vincent Lejeune wrote: >On June 28, 2003 at 21:37:18, Jeremiah Penery wrote: > >>On June 28, 2003 at 20:25:49, Slater Wold wrote: >> >>>1.Ce1 Rb3 2.Cd3 Rc3 3.Cc1 Rb2 4.Ca2 Rxa2 >>> +- (#24) Profondeur: 43 06:08:09 8531825kN, tb=47573427 <//==- >>> >>>I'd say he did. :) >>>Might not be a complete set or something though. >> >>The position is most certainly NOT a mate in 24. > >should be a bug in Chesstiger 15 :) You are probably right but in theory it may be not a bug. programs may be too optimistic about a position and I see no reason not to allow it when the score is a mate score. It is possible to have 2 numbers when one number is a bound and one number is an evaluation. If a program see mate in at most 9 moves after 700000 nodes then it can give an evaluation of mate in 7.2 based on statistics that says that the average number of moves to mate in similiar situations is 7.2 The statistic may be calculated by using chest to calculate the real distance to mate in a lot of positions and use some mathematical model to get an estimate for the real distance to mate. Uri
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.