Author: Rémi Coulom
Date: 09:25:38 11/23/01
Go up one level in this thread
On November 22, 2001 at 17:56:07, Colin Frayn wrote:
>I've been looking at static endgame evals again (i.e. no searching).
>
>I'm not sure how 18.12 manages, but Crafty 18.10 completely misevaluates this
>simple endgame position. So does the current build of Beowulf. It's obviously
>a win for white, but yet both engines evaluate this position as a massive
>advantage to black.
>
>[D]8/1KPk4/8/6p1/p7/8/8/8 b - -
>
>I really have some work to do. Surely this *has* to be easy to code, right?
>You wouldn't believe the number of trivial positions that no engine seems to
>understand.
My program does not understand many position, but it does understand this kind
of thing with its "generalized square of the king".
http://www.it.ro/ccc_search/ccc.php?art_id=65487
>
>The position comes from the following endgame test;
>[D]8/8/5p2/pK6/2Pk2P1/8/8/8 w - -
>
>Best move is g5!
>
>Cheers,
>Colin
Output by TCB (no tablebase):
(@n)EngineControl|Tournament>setgear i
00:00:00,02 0 ------ Depth = 1
00:00:00,02 2 -11 1. Kxa5
00:00:00,02 9 ------ Depth = 2
00:00:00,02 18 -11 1. Kxa5 Kxc4
00:00:00,02 42 ------ Depth = 3
00:00:00,02 64 -16 1. Kxa5 Kxc4 2. Kb6
00:00:00,02 104 -11 1. c5
00:00:00,02 129 -7 1. c5 a4 2. Kc6
00:00:00,02 145 ------ Depth = 4
00:00:00,02 249 -10 1. c5 a4 2. Kc6 a3
00:00:00,02 323 ------ Depth = 5
00:00:00,02 701 -4 1. c5 a4 2. c6 a3 3. c7 a2
00:00:00,03 884 ------ Depth = 6
00:00:00,03 1555 -10 1. c5 a4 2. c6 a3 3. c7 a2 4. Kc6
00:00:00,03 1867 ------ Depth = 7
00:00:00,04 3538 -10 1. c5 a4 2. c6 a3 3. c7 a2 4. Kc6 a1=Q
00:00:00,04 4117 ------ Depth = 8
00:00:00,05 7553 -10 1. c5 a4 2. c6 a3 3. c7 a2 4. Kc6 a1=Q 5. c8=Q
00:00:00,06 8084 -7 1. Kxa5
00:00:00,06 8365 -4 1. Kxa5 Kxc4 2. Kb6 Kd5 3. Kb5 Kd4 4. Kc6 Ke4
00:00:00,06 8880 ------ Depth = 9
00:00:00,08 11597 -9 1. Kxa5 Kxc4 2. Kb6 Kd5 3. Kc7 Kc5 4. Kd7 Kd4 5.
Ke6
00:00:00,09 14786 ------ Depth = 10
00:00:00,11 18591 -3 1. Kxa5 Kxc4 2. Kb6 Kd5 3. Kc7 Ke5 4. Kd7 f5 5.
gxf5 Kxf5 6. Kd6
00:00:00,19 32416 -2 1. g5
00:00:01,04 218443 +391 1. g5 fxg5 2. c5 Ke5 3. c6 Kd6 4. Kb6 Ke6 5. Ka6
Kf5
00:00:01,09 228449 ------ Depth = 11
00:00:01,22 252641 +643 1. g5 fxg5 2. c5 Ke5 3. c6 Kd6 4. Kb6 a4 5. c7 g4
6. c8=Q g3
00:00:01,30 269825 ------ Depth = 12
00:00:01,50 308036 +654 1. g5 fxg5 2. c5 Ke5 3. c6 Kd6 4. Kb6 a4 5. c7
Kd7 6. Kb7 Kd6 7. c8=Q
00:00:01,71 349316 ------ Depth = 13
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.