Author: Uri Blass
Date: 01:08:59 04/06/01
Go up one level in this thread
On April 05, 2001 at 23:52:19, Alex Szabo wrote: >On April 04, 2001 at 23:15:05, Robert Hyatt wrote: > >>On April 04, 2001 at 20:26:17, Stephane Daniel wrote: >> >>> >>>Position 230 of the well known "Win At Chess" test suite. >>> >>>2b5/1r6/2kBp1p1/p2pP1P1/2pP4/1pP3K1/1R3P2/8 b - - 0 1 >>> >>>Right move is supposed to be Rb4 >>> >>> >>>My program just can't solve this one. Is there an error in this particular test? >>>What do your programs find for this position? Mine keeps telling me to play Rh7 >>>with a score of -1.71 >> >> >>It is the right move. But you have to evaluate passed pawns very highly to >>get this one, as you really will not search it to a queen in any finite amount >>of time, due to the checks. I have seen older versions of crafty solve it. >>Someone reported that if you do "eval ppscale 200" then it will get it pretty >>quickly... > >Rb4 does NOT win for black; it's a draw. Can you post the drawing line? 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.