Author: Uri Blass
Date: 14:48:19 09/28/01
Go up one level in this thread
On September 28, 2001 at 17:23:37, Dieter Buerssner wrote:
>On September 28, 2001 at 14:27:32, Simon Finn wrote:
>
>>On September 28, 2001 at 12:36:55, Dieter Buerssner wrote:
>>
>>[snip]
>>
>>>I think, often in this type of positions (only few pawns), chess programs are
>>>usually very good. There are only few pawn moves possible, and programs reach a
>>>high search depth fast, and of course, because of the TBs. Without TBs I get:
>>
>>I would be very impressed if any program can solve the following
>>innocent-looking position (Grigoriev 1933) in a reasonable time. By "solve" I
>>mean find most of the PV (up to 13. Ka8, say) with a winning score.
>
>I think, I did not expresss myself clearly. In this very interesting position,
>there are 6 pawns that can move. In the position, that I meant by "this type of
>position", there could only 2 pawns move (and both would be captured
>immediately). I looked at it from a chess-engine search point of view. I may be
>of course wrong, but I have the feeling, that chess-engines are often rather
>good in pawn endgames with few pawns, that are blocked.
>
>>[D]k7/8/pp6/2p4K/8/PPP5/8/8 w - - bm c4;c0 "Grigoriev 1933";pv "1. c4 Kb7 2. a4
>>Kc6 3. Kg5 Kc7 4. Kf6 Kd7 5. a5 Kd6 6. Kf5 Kc7 7. Ke6 Kc6 8. Ke7 Kc7 9. Ke8 Kf8
>>10. axb6 Kb7 11. Kd7 Kxb6 12. Kc8 Kb6 13. Ka8 a5 14. Kb8 a4 15. bxa4 Ka5 16. Kb7
>>Kxa4 17. Kc6 Kb4 18. Kd5";
>
>Yace, AMD K6-2 475, 20M hash, KPPKP as only relevant 5-men TB, all 4-men TBs.
>Unfortunately, the PV is hidden in the hash. Naturally, when I follow the line,
>after some moves, a winning score is seen. Sure, I cannot show the high scores,
>and the whole line, but it does at least not look too bad.
>
> nodes time score depth
> 9104109 2:15.9 0.03 18. 1.c4 Kb7 2.Kg5 a5 3.a4 Kc7 4.Kg6 Kc6 5.Kf7 Kd6
> 6.Kf6 Kd7 7.Kf7 Kc8 8.Ke6 Kc7 9.Ke7 Kc6 10.Ke6
> Kc7 11.Ke7 Kc6 12.Ke6 Kc7 {0}
> 11544446 2:57.9 0.06 19t 1.c4 Kb7 2.a4 a5 3.Kh6 Kc7 4.Kg7 Kd6H 5.Kf6H
> Kd7H 6.Kf7H b5H 7.cxb5H Kd6H 8.b6H Ke5H 9.b7H
> Kd6H {HT}
> 13006082 3:30.0 0.06 19. 1.c4 Kb7 2.a4 a5 3.Kh6 Kc7 4.Kg7 Kd6 5.Kf6 Kd7
> 6.Kf7 b5 7.cxb5 Kd6 8.b6 Ke5 9.b7 Kd6
> 14649431 3:50.5 0.06 20t 1.c4 Kb7 2.a4 a5 3.Kh6 Kc7 4.Kg7 Kd6 5.Kf6 Kd7
> 6.Kf7 Kd6 7.Ke8 Ke6H 8.Kd8H Kd6H 9.Kc8H Kc6H
> 10.Kb8H Kd7H 11.Kb7H Ke6H 12.Kxb6H Ke5H 13.Kb7H
> Kd4H 14.Ka7H Kc3H 15.Kb6H Kb4H 16.Kc6H {HT} {80}
> 16762214 4:33.0 0.06 20. 1.c4 Kb7 2.a4 a5 3.Kh6 Kc7 4.Kg7 Kd6 5.Kf6 Kd7
> 6.Kf7 Kd6 7.Ke8 Ke6 8.Kd8 Kd6 9.Kc8 Kc6 10.Kb8
> Kd7 11.Kb7 Ke6 12.Kxb6 Ke5 13.Kb7 Kd4 14.Ka7
> Kc3 15.Kb6 Kb4 16.Kc6 {HT} {80}
> 19858667 5:16.1 0.06 21t 1.c4 Kb7 2.a4 a5 3.Kh6 Kc7 4.Kg7 Kd6 5.Kf6 Kd7
> 6.Kf7 Kd6H 7.Ke8H Ke6H 8.Kd8H b5H 9.axb5H a4H
> 10.bxa4H Kd6H 11.b6H Kc6H 12.b7H Kxb7H 13.a5H
> Kb8H {HT} {80}
> 23100946 6:47.0 0.06 21. 1.c4 Kb7 2.a4 a5 3.Kh6 Kc7 4.Kg7 Kd6 5.Kf6 Kd7
> 6.Kf7 Kd6 7.Ke8 Ke6 8.Kd8 b5 9.axb5 a4 10.bxa4
> Kd6 11.b6 Kc6 12.b7 Kxb7 13.a5 Kb8 {HT} {80}
> 29924869 8:35.2 0.45 22t 1.c4 Kb7 2.a4 Kc6 3.Kg5 Kc7 4.Kg6H Kd8H 5.Kf5H
> b5H {HT}
> 34894188 10:28.1 0.45 22. 1.c4 Kb7 2.a4 Kc6 3.Kg5 Kc7 4.Kg6 Kd8 5.Kf5 b5
> 45047790 12:50.4 0.47 23t 1.c4 Kb7 2.a4 Kc6 3.Kg5 Kc7 4.Kg6 Kc6 5.Kf6 Kd7
> 6.a5 Kd6 7.Kf5 Ke7 8.Ke5 Kd7 9.Kd5 Kc7 10.Ke4
> Kc6 11.Kd3 Kb7H 12.axb6H Kxb6H 13.Kc3H {HT} {0}
> 52677230 16:13.9 0.47 23. 1.c4 Kb7 2.a4 Kc6 3.Kg5 Kc7 4.Kg6 Kc6 5.Kf6 Kd7
> 6.a5 Kd6 7.Kf5 Ke7 8.Ke5 Kd7 9.Kd5 Kc7 10.Ke4
> Kc6 11.Kd3 Kb7 12.axb6 Kxb6 13.Kc3 {HT} {0}
> 67742584 19:56.0 0.47 24t 1.c4 Kb7 2.a4 Kc6 3.Kg5 Kc7 4.Kf6 Kd7 5.a5 Kd6H
> 6.Kf5H Kc7H 7.Ke6H Kc6H 8.Ke7H Kc7H {HT} {0}
> 79835250 23:46.9 0.47 24. 1.c4 Kb7 2.a4 Kc6 3.Kg5 Kc7 4.Kf6 Kd7 5.a5 Kd6
> 6.Kf5 Kc7 7.Ke6 Kc6 8.Ke7 Kc7 {HT} {0}
>
>>Pawn endings are difficult!
>
>I won't disagree at all.
>
>Regards,
>Dieter
Note that I suspect that yace may find a forced mate after 8...Kc7
After 9.Ke8 Kc8 the top programs have no problem to see a forced mate.
I expect yace to be sligthly better and to see a forced mate even without 9.Ke8
Kc8
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.