Author: Dieter Buerssner
Date: 15:02:36 09/09/00
Go up one level in this thread
On September 09, 2000 at 16:08:09, Howard Exner wrote:
>I wonder if this below position is solvable by some programs?
>bm is Kb7 with the King march over to the kingside.
>
>[D] 3k4/2pP4/2Kp2P1/3P2P1/B2b4/8/8/8 w - -
>
>It does illustrate nicely the difference between human and machine
>in their approach to solving certain positions. A human will in a sense work
>backwards, visualizing that the king must penetrate to the f5,e6,g6 area
>in order to win. A machine starts from the initial position and starts winding
>its way through a maze of variations, perhaps hitting upon a winning line.
Yace finds Kb7 in a fraction of a second, and soon has the king march in the PV.
But the score doesn't advance for quite some time, until it fails high after
1600 seconds (On a AMD K6-2 475, that is also doing some other tasks at the
same time). 6 minutes later, it comes up with the following PV. (Then I lost
patience)
247798547 1966.1 390 24t : Kc6b7 Bg7e5 Ba4c6 Be5g7 Kb7a6 Bg7d4 Ka6b5 Bd4g7
Kb5c4 Bg7e5 Kc4d3 Kd8e7 Kd3e4 Be5g7 Ke4f5 Bg7e5
d7d8=Q+ Ke7xd8 Kf5e6 Kd8c8 Ke6f7 Kc8b8 g6g7
Be5xg7 Kf7xg7 Kb8a7 <HT> [330]
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.