Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Simple position - no understanding for many chess programs

Author: Uri Blass

Date: 14:20:41 02/13/02

Go up one level in this thread


On February 13, 2002 at 15:38:58, Kurt Utzinger wrote:

>Black to move. The position is a draw, even if White could manage to win both
>Black pawns. But quite a lot of [top] programs do not at all understand this and
>show completely wrong evaluations.
>
>A good example where a 1500-ELO-player does better than the so called
>2700-ELO-silicon-monsters!!
>Kurt
>
>[D] 8/8/5k1p/6pP/6K1/8/8/3B4 b - - 0 1


Junior7 and shredder5.32 have no problem(analysis at the end of this post).
I wonder which programs have problem here

Note that I think that it is not a good test position for playing because a
programmer may decide not to use the knowledge for evaluation of the leaves if
the root is evaluated as a draw by the same knowledge.

It may be a good idea for Junior that can do stupid errors because it stops the
search when it finds a draw by insufficient material and returns a draw.

stopping to search may cause problems also when the root position is more
complicated but experience with Junior showed that many of it's stupid mistakes
about it(I believe almost all of them) happened when the root position could be
evaluated as a draw.

an example when Junior lost against Fritz in a drawn position was posted
recently by James Walker if I remember correctly.


New position
8/8/5k1p/6pP/6K1/8/8/3B4 b - - 0 1

Analysis by Junior 7:

1...Ke7 2.Bb3
  +-  (1.63)   Depth: 3   00:00:00
1...Ke5 2.Bb3
  +-  (1.51)   Depth: 3   00:00:00
1...Ke5 2.Bb3 Kf6
  +-  (1.50)   Depth: 6   00:00:00
1...Kg7!
  ±  (1.21)   Depth: 6   00:00:00
1...Kg7! 2.Bb3
  =  (0.00)   Depth: 6   00:00:00
1...Kg7 2.Bb3
  =  (0.00)   Depth: 9   00:00:00
1...Kg7 2.Bb3
  =  (0.00)   Depth: 12   00:00:00  2kN
1...Kg7 2.Bb3
  =  (0.00)   Depth: 15   00:00:01  4kN
1...Kg7 2.Bb3
  =  (0.00)   Depth: 18   00:00:01  13kN
1...Kg7 2.Bb3
  =  (0.00)   Depth: 21   00:00:01  30kN
1...Kg7 2.Bb3
  =  (0.00)   Depth: 24   00:00:01  65kN
1...Kg7 2.Bb3
  =  (0.00)   Depth: 27   00:00:02  146kN
1...Kg7 2.Bb3
  =  (0.00)   Depth: 30   00:00:04  305kN
1...Kg7 2.Bb3
  =  (0.00)   Depth: 33   00:00:08  599kN
1...Kg7 2.Bb3
  =  (0.00)   Depth: 35   00:00:14  1044kN
1...Kg7 2.Bb3
  =  (0.00)   Depth: 37   00:00:24  1695kN
1...Kg7 2.Bb3
  =  (0.00)   Depth: 39   00:00:39  2740kN
1...Kg7 2.Bb3
  =  (0.00)   Depth: 41   00:01:04  4403kN
1...Kg7 2.Bb3
  =  (0.00)   Depth: 42   00:01:28  5836kN
1...Kg7 2.Bb3
  =  (0.00)   Depth: 43   00:01:57  7655kN
1...Kg7 2.Bb3
  =  (0.00)   Depth: 44   00:02:34  9865kN
1...Kg7 2.Bb3
  =  (0.00)   Depth: 45   00:03:20  12699kN
1...Kg7 2.Bb3
  =  (0.00)   Depth: 46   00:04:16  16180kN

New position
8/8/5k1p/6pP/6K1/8/8/3B4 b - - 0 1

Analysis by Shredder 5.32:

1...Kg7
  =  (0.00)   Depth: 1/2   00:00:00
1...Kg7
  =  (0.00)   Depth: 1/2   00:00:00
1...Kg7 2.Bf3
  =  (0.00)   Depth: 2/4   00:00:00
1...Kg7 2.Bf3 Kg8
  =  (0.00)   Depth: 3/6   00:00:00
1...Kg7 2.Bf3 Kg8 3.Bd1
  =  (0.00)   Depth: 4/8   00:00:00
1...Kg7 2.Bf3 Kg8 3.Bd1 Kh8
  =  (0.00)   Depth: 5/10   00:00:00
1...Kg7 2.Bf3 Kg8 3.Bd1 Kh8 4.Bf3 Kh7
  =  (0.00)   Depth: 6/12   00:00:00
1...Kg7 2.Bf3 Kg8 3.Bd1 Kh8 4.Bf3
  =  (0.00)   Depth: 7/14   00:00:00
1...Kg7 2.Bf3 Kg8 3.Bd1 Kh8 4.Bf3
  =  (0.00)   Depth: 8/16   00:00:00  1kN
1...Kg7 2.Kf5 g4 3.Bb3 Kh8 4.Bd1
  =  (0.01)   Depth: 9/18   00:00:00  4kN
1...Kf7 2.Bf3 Kg8 3.Bd1 Kh8 4.Bf3
  =  (0.00)   Depth: 9/18   00:00:00  5kN
1...Kf7 2.Bf3 Kg8 3.Bd1 Kh8 4.Bf3
  =  (0.00)   Depth: 9/18   00:00:00  5kN
1...Kf7 2.Bf3 Kg8 3.Bd1 Kh8 4.Bf3
  =  (0.00)   Depth: 10/20   00:00:00  7kN
1...Kf7 2.Bf3 Kg8 3.Bd1 Kh8 4.Bf3
  =  (0.00)   Depth: 11/22   00:00:00  9kN
1...Kf7 2.Bf3 Kg8 3.Bd1 Kh8 4.Bf3
  =  (0.00)   Depth: 12/24   00:00:00  11kN
1...Kg7 2.Bf3 Kh8 3.Bd1
  =  (-0.01)   Depth: 12/24   00:00:00  12kN
1...Kg7 2.Bf3 Kh8 3.Bd1 Kh7
  =  (0.00)   Depth: 12/24   00:00:00  12kN
1...Kg7 2.Kf5 g4 3.Bf3
  =  (0.01)   Depth: 13/25   00:00:01  21kN
1...Kf7 2.Bf3 Kg7 3.Bd1
  =  (0.00)   Depth: 13/25   00:00:01  24kN
1...Kf7 2.Bf3 Kg7 3.Bd1
  =  (0.00)   Depth: 13/25   00:00:01  24kN
1...Kf7 2.Bf3 Kg7 3.Bd1
  =  (0.00)   Depth: 14/26   00:00:01  28kN
1...Kf7 2.Bb3+ Kg7 3.Kf5 g4 4.Be6
  =  (0.02)   Depth: 15/27   00:00:02  81kN
1...Kf7 2.Kf5 Kg7 3.Bc2 g4 4.Kxg4 Kh8 5.Bb3 Kh7
  =  (0.01)   Depth: 16/28   00:00:02  145kN
1...Kf7 2.Bb3+ Kg7 3.Kf5 g4 4.Be6
  =  (0.02)   Depth: 17/29   00:00:03  195kN
1...Kf7 2.Kf5 Kg8 3.Kg6 Kh8 4.Kxh6 Kg8 5.Bf3 Kh8
  ²  (0.27)   Depth: 18/30   00:00:04  277kN
1...Kf7 2.Kf5 Kg8 3.Kg6 Kh8 4.Kf6 g4 5.Bxg4 Kh7 6.Bf5+ Kg8 7.Be6+ Kh8 8.Kf5 Kg7
  ²  (0.36)   Depth: 18/30   00:00:05  351kN
1...Kg7 2.Kf5 Kf7 3.Bc2 Kg7 4.Ba4 g4 5.Ke5
  ²  (0.35)   Depth: 18/30   00:00:05  370kN
1...Kg7 2.Kf5 Kf7 3.Bc2 Kg7 4.Bb3 g4 5.Be6 g3 6.Bd5 g2 7.Bxg2 Kf7 8.Bd5+ Kg7
9.Be6 Kh8
  ²  (0.35)   Depth: 18/30   00:00:05  390kN
1...Ke6 2.Bb3+ Ke5 3.Ba4 Kf6 4.Kf3
  ²  (0.34)   Depth: 18/30   00:00:06  403kN
1...Ke6 2.Ba4 Kf6 3.Bc2 Kf7 4.Bd1 Kg8 5.Kf5
  =  (0.01)   Depth: 18/30   00:00:10  780kN
1...Ke5 2.Bb3 Ke4 3.Bc2+ Ke5 4.Bb3 Ke4
  =  (0.00)   Depth: 18/30   00:00:11  852kN
1...Ke5 2.Bb3 Ke4 3.Bc2+ Ke5 4.Bb3 Ke4
  =  (0.00)   Depth: 18/30   00:00:11  853kN
1...Ke5 2.Bb3 Ke4 3.Bc2+ Ke5 4.Bb3 Ke4
  =  (0.00)   Depth: 19/31   00:00:12  919kN
1...Ke5 2.Bb3 Ke4 3.Bc2+ Ke5 4.Bb3 Ke4
  =  (0.00)   Depth: 20/32   00:00:13  1035kN
1...Ke5 2.Bb3 Ke4 3.Bc2+ Ke5 4.Bd1 Ke6 5.Ba4 Ke5 6.Bb3 Ke4
  =  (0.01)   Depth: 21/33   00:00:15  1204kN


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.