Author: Robert Hyatt
Date: 14:44:10 02/15/00
Go up one level in this thread
On February 14, 2000 at 06:57:38, Ulrich Tuerke wrote: >On February 10, 2000 at 23:11:02, Robert Hyatt wrote: > >>On February 10, 2000 at 22:22:02, Howard Exner wrote: >> >>>[D]8/4k3/2bq4/1p6/p6P/P5P1/1P2QP1K/8 b - - id Comet - Diep; >>> >>>Is this a tough choice for a computer to make? Diep played Qe6 and Comet went on >>>to win. >>>What do other programs prefer here? Seems that in this case leaving the Queens >>>on would benefit black or at least exchanging them under more favourable >>>circumstances. >> >> >>This game was a dead draw until Diep played Kf7 rather than Kg6 a few moves from >>this point. Crafty saw that turning into a KPP vs KB ending that the tablebases >>said was 'drawn'. I think vincent's program thought (for reasons unknown) that >>KB vs KPP was lost for the KB, so he wouldn't go for that. >> >>I had crafty analyzing during the game, and until he played Kf7 the score was >>zero. After Kf7 my score jumped to +2.5 and kept going. within 5 moves it >>was at +10 or so... > > >I have also had a closer look to this. It seems that already the queens exchange >lost the game. >I think that also Kg6 looses though not as obvious as Kf7. In my analysis, the >white king will capture the queen side pawns of black in this case. > >Uli You are correct. It ends up a kpp vs kb ending. Fortunately, that is a draw in this position (tablebases work wonders). Crafty was seeing the total evaporation of the kingside pawns, while the queenside pawns of black disappear. But black has the B, which makes this just barely a draw, assuming my search didn't totally screw this up. I'll try to find the Kg6 position and post the search analysis for it... Bob
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.