Author: Heiner Marxen
Date: 19:01:17 03/26/01
Go up one level in this thread
On March 26, 2001 at 19:11:34, Paul wrote:
>On March 26, 2001 at 14:14:57, Heiner Marxen wrote:
>
>>On March 23, 2001 at 01:54:29, Heinz-Josef Schumacher wrote:
>>
>>>The follow study is not soluble for progrms today!?
>>>
>>>5nr1/2Pp2pk/3Pp1p1/4P1P1/6P1/5K2/8/7n w - -
>>>
>>>http://homepages.compuserve.de/hjsxy/2.gif
>>>
>>>M.Judowitsch- Study 1976
>>>
>>>1.c8N!! [1.Kg2? Rh8 2.Kxh1 Kg8+ 3.Kg2 Rh4=). 1...Rh8 2.Ne7 Nf2 3.Kxf2 Rg8 4.Ke3
>>>Kh8 5.Kd4 Kh7 6.Kc5 Kh8 7.Kb6 Kh7 8.Kc7 Rh8 9.Kd8 Rg8 10.Ke8 Rh8 11.Kf7 Rg8
>>>12.Nxg8 Kh8 13.Kxf8 1-0
>>
>>My program Chest confirms, that after 8...Rh8 there is a mate in 12:
>>
>>9.Kd8 Rg8 10.Ke8 Kh8 11.Kf7 Kh7 12.Nxg8 Kh8 13.Ne7 Kh7 14.Kxf8 Kh8 15.Ke8 Kh7
>>16.Kxd7 Kh8 17.Kc7 Kh7 18.d7 Kh8 19.d8=Q+ Kh7 20.Qg8#
>>
>>When I went back one more move of the intended solution, Chest refutes
>>8.Kc7 with Kh8 instead of Rh8, with the following line:
>>8.Kc7 Kh8 9.Kd8? Nh7+!, avoiding the mate with Nxg6.
>>
>>Neither I nor Chest have found a solution here, so it may be a refutation
>>of the study, but I doubt it. Does anyone know better?
>
>Hi Heiner!
>
>I've also played around a bit with this position and the longest line Pretz
>finds is a Mate in 41:
>
>1. c8N!! Rh8 2. Ne7 Nf2 3. Kxf2 Rg8 4. Ke3 Kh8 5. Kd4 Kh7 6. Kc5 Kh8 7. Kb6 Kh7
>8. Kc7 Kh8 9. Kc8 Nh7+ 10. Nxg8 Kxg8 11. Kxd7 Kf8 12. Kc8 Nxg5 13. d7 Nf7 14.
>d8=Q+ Nxd8 15. Kxd8 Kf7 16. Kd7 Kf8 17. Kxe6 Ke8 18. g5 Kf8 19. Kd7 Kf7 20. e6+
>Kg8 21. e7 Kh8 22. e8=N Kh7 23. Nxg7 Kxg7 24. Ke7 Kg8 25. Kf6 Kh7 26. Kf7 Kh8
>27. Kxg6 Kg8 28. Kh6 Kf7 29. Kh7 Ke7 30. g6 Ke6 31. g7 Ke5 32. g8=Q Kd4 33. Qg4+
>Kd3 34. Kg6 Ke3 35. Kg5 Kd3 36. Qb4 Ke3 37. Kg4 Ke2 38. Qd4 Kf1 39. Qd2 Kg1 40.
>Kg3 Kh1 41. Qg2#
After 15...Kf7 Crafty 17.7 with 3&4 piece tables finds a mate in 17:
Black(1): setboard [D]3K4/5kp1/4p1p1/4P3/6P1/8/8/8 w - -
White(1): go
end-game phase
time surplus 8.57 time limit 1:00 (1:00)
nss depth time score variation (1)
(2) 19-> 0.30 11.52 1. g5 Kf8 2. Kd7 Kf7 3. Kd6 Kf8 4.
Kxe6 Ke8 5. Kd6 Kf8 6. e6 Ke8 7. e7
Kf7 8. Kd7 Kg8
20 0.30 ++ 1. g5!!
(2) 20-> 0.57 11.91 1. g5 Kf8 2. Kd7 Kf7 3. Kd6 Kf8 4.
Kxe6 Ke8 5. Kd6 Kf8 6. e6 Ke8 7. e7
Kf7 8. Kd7 Kg8
21 0.57 ++ 1. g5!!
21 0.92 Mat17 1. g5 Kf8 2. Kd7 Kf7 3. Kd6 Ke8 4.
Kxe6 Kd8 5. Kf7 <HT>
21-> 1.13 Mat17 1. g5 Kf8 2. Kd7 Kf7 3. Kd6 Ke8 4.
Kxe6 Kd8 5. Kf7 <HT>
22 1.23 Mat17 1. g5 Kf8 2. Kd7 Kf7 3. Kd6 Ke8 4.
Kxe6 Kd8 5. Kf7 <HT>
(2) 22-> 1.80 Mat17 1. g5 Kf8 2. Kd7 Kf7 3. Kd6 Ke8 4.
Kxe6 Kd8 5. Kf7 <HT>
23 5.51 Mat17 1. g5 Kf8 2. Kd7 Kf7 3. Kd6 Ke8 4.
Kxe6 Kd8 5. Kf7 <HT>
23-> 7.17 Mat17 1. g5 Kf8 2. Kd7 Kf7 3. Kd6 Ke8 4.
Kxe6 Kd8 5. Kf7 <HT>
24 11.44 Mat17 1. g5 Kf8 2. Kd7 Kf7 3. Kd6 Ke8 4.
Kxe6 Kd8 5. Kf7 <HT>
(2) 24-> 15.59 Mat17 1. g5 Kf8 2. Kd7 Kf7 3. Kd6 Ke8 4.
Kxe6 Kd8 5. Kf7 <HT>
25 25.18 Mat17 1. g5 Kf8 2. Kd7 Kf7 3. Kd6 Ke8 4.
Kxe6 Kd8 5. Kf7 <HT>
(2) 25-> 34.03 Mat17 1. g5 Kf8 2. Kd7 Kf7 3. Kd6 Ke8 4.
Kxe6 Kd8 5. Kf7 <HT>
time=1:00 cpu=99% mat=-1 n=27043469 fh=99% nps=448333
ext-> checks=4457892 recaps=1137 pawns=18865 1rep=53171 thrt:0
predicted=0 nodes=27043469 evals=156698
endgame tablebase-> probes done=62059 successful=62059
SMP-> split=0 stop=0 data=0/64 cpu=59.91 elap=1:00
which results in #32, much shorter than yours.
One move earlier in your line crafty plays different:
White(2): setboard [D]2Kn1k2/6p1/4p1p1/4P3/6P1/8/8/8 w - -
White(1): go
end-game phase
clearing hash tables
time surplus 0.00 time limit 1:00 (1:00) [easy move]
nss depth time score variation (1)
(3) 18-> 0.34 2.77 1. Kxd8 g5 2. Kd7 Kf7 3. Kd6 Kf8 4.
Kxe6 Ke8 5. Kf5 Kf7 6. Kxg5 Ke6 7.
Kf4 Kf7 8. Kf5 <HT>
(2) 19 0.36 2.77 1. Kxd8 g5 2. Kd7 Kf7 3. Kd6 Kf8 4.
Kxe6 Ke8 5. Kf5 Kf7 6. Kxg5 Ke6 7.
Kf4 Kf7 8. Kf5 <HT>
(3) 19-> 0.51 2.77 1. Kxd8 g5 2. Kd7 Kf7 3. Kd6 Kf8 4.
Kxe6 Ke8 5. Kf5 Kf7 6. Kxg5 Ke6 7.
Kf4 Kf7 8. Kf5 <HT>
(2) 20 0.56 3.01 1. Kxd8 g5 2. Kd7 Kf7 3. Kd6 Kf8 4.
Kxe6 Ke8 5. Kf5 Kf7 6. Kxg5 Ke6 7.
Kf4 Kf7 8. Kf5 Ke7 9. g5 Kf7 10. g6+
Ke7 11. e6
(3) 20-> 0.66 3.01 1. Kxd8 g5 2. Kd7 Kf7 3. Kd6 Kf8 4.
Kxe6 Ke8 5. Kf5 Kf7 6. Kxg5 Ke6 7.
Kf4 Kf7 8. Kf5 Ke7 9. g5 Kf7 10. g6+
Ke7 11. e6
(2) 21 0.72 3.35 1. Kxd8 g5 2. Kd7 Kf7 3. Kd6 Kf8 4.
Kxe6 Ke8 5. Kf5 Kf7 6. Kxg5 Ke6 7.
Kf4 Kf7 8. Kf5 Kf8 9. e6 Ke8 10. Kg6
Kf8 11. Kf5
(3) 21-> 0.98 3.35 1. Kxd8 g5 2. Kd7 Kf7 3. Kd6 Kf8 4.
Kxe6 Ke8 5. Kf5 Kf7 6. Kxg5 Ke6 7.
Kf4 Kf7 8. Kf5 Kf8 9. e6 Ke8 10. Kg6
Kf8 11. Kf5
(2) 22 1.05 3.15 1. Kxd8 g5 2. Kd7 Kf7 3. Kd6 Kf8 4.
Kxe6 Ke8 5. Kf5 Kf7 6. Kxg5 Ke6 7.
Kf4 Kf7 8. Kf5 Kf8 9. Ke6 Kg8 10. Kd6
Kh7 11. e6 Kg6
(4) 22-> 1.35 3.15 1. Kxd8 g5 2. Kd7 Kf7 3. Kd6 Kf8 4.
Kxe6 Ke8 5. Kf5 Kf7 6. Kxg5 Ke6 7.
Kf4 Kf7 8. Kf5 Kf8 9. Ke6 Kg8 10. Kd6
Kh7 11. e6 Kg6
23 1.40 ++ 1. Kxd8!!
(3) 23 20.53 Mat28 1. Kxd8 g5 2. Kd7 Kf7 3. Kd6 Kf8 4.
Kxe6 Ke8 5. Kf5 Kf7 6. Kxg5 Ke6 7.
Kf4 Ke7 8. Kf5 Kf7 9. e6+ Ke7 10. Kg6
Kf8 11. Kf5 <HT>
23-> 20.88 Mat28 1. Kxd8 g5 2. Kd7 Kf7 3. Kd6 Kf8 4.
Kxe6 Ke8 5. Kf5 Kf7 6. Kxg5 Ke6 7.
Kf4 Ke7 8. Kf5 Kf7 9. e6+ Ke7 10. Kg6
Kf8 11. Kf5 <HT>
24 22.52 Mat25 1. Kxd8 g5 2. Kd7 Kf7 3. Kd6 Kf8 4.
Kxe6 Ke8 5. Kf5 Kf7 6. Kxg5 Ke6 7.
Kf4 Ke7 8. Kf5 Kf7 9. e6+ Ke7 10. Kg6
Kf8 11. g5 Ke8 12. Kxg7 <EGTB>
(2) 24-> 24.01 Mat25 1. Kxd8 g5 2. Kd7 Kf7 3. Kd6 Kf8 4.
Kxe6 Ke8 5. Kf5 Kf7 6. Kxg5 Ke6 7.
Kf4 Ke7 8. Kf5 Kf7 9. e6+ Ke7 10. Kg6
Kf8 11. g5 Ke8 12. Kxg7 <EGTB>
25 25.19 Mat25 1. Kxd8 g5 2. Kd7 Kf7 3. Kd6 Kf8 4.
Kxe6 Ke8 5. Kf5 Kf7 6. Kxg5 Ke6 7.
Kf4 Ke7 8. Kf5 Kf7 9. e6+ Ke7 10. Kg6
Kf8 11. g5 Ke8 12. Kxg7 <EGTB>
(2) 25-> 27.44 Mat25 1. Kxd8 g5 2. Kd7 Kf7 3. Kd6 Kf8 4.
Kxe6 Ke8 5. Kf5 Kf7 6. Kxg5 Ke6 7.
Kf4 Ke7 8. Kf5 Kf7 9. e6+ Ke7 10. Kg6
Kf8 11. g5 Ke8 12. Kxg7 <EGTB>
26 31.56 Mat25 1. Kxd8 g5 2. Kd7 Kf7 3. Kd6 Kf8 4.
Kxe6 Ke8 5. Kf5 Kf7 6. Kxg5 Ke6 7.
Kf4 Ke7 8. Kf5 Kf7 9. e6+ Ke7 10. Kg6
Kf8 11. g5 Ke8 12. Kxg7 <EGTB>
(2) 26-> 42.70 Mat25 1. Kxd8 g5 2. Kd7 Kf7 3. Kd6 Kf8 4.
Kxe6 Ke8 5. Kf5 Kf7 6. Kxg5 Ke6 7.
Kf4 Ke7 8. Kf5 Kf7 9. e6+ Ke7 10. Kg6
Kf8 11. g5 Ke8 12. Kxg7 <EGTB>
27 44.82 Mat25 1. Kxd8 g5 2. Kd7 Kf7 3. Kd6 Kf8 4.
Kxe6 Ke8 5. Kf5 Kf7 6. Kxg5 Ke6 7.
Kf4 Ke7 8. Kf5 Kf7 9. e6+ Ke7 10. Kg6
Kf8 11. g5 Ke8 12. Kxg7 <EGTB>
time=1:00 cpu=98% mat=-4 n=25159353 fh=97% nps=415788
ext-> checks=4130069 recaps=448 pawns=74283 1rep=175540 thrt:11484
predicted=0 nodes=25159353 evals=517568
endgame tablebase-> probes done=101635 successful=101635
SMP-> split=0 stop=0 data=0/64 cpu=59.35 elap=1:00
This is a total depth of 14+25=39
So, my impression is that white really wins, but the win is much deeper
than the initially given line, and also too deep for our programs to find
the shortest mate.
>If instead of 11... Kf8 black plays 11... Nxg5 a much shorter mate follows:
>
>11... Nxg5 12. Ke7 Nf7 13. d7 g5 14. d8=Q+ Nxd8 15. Kxd8 Kf8 16. Kd7 Kf7 17. Kd6
>Kf8 18. Kxe6 Ke8 19. Kf5 Ke7 20. Kxg5 Ke6 21. Kf4 Ke7 22. Kf5 Kf7 23. e6+ Ke7
>24. Kg6 Kxe6 25. Kxg7 Ke5 26. g5 Kf5 27. Kh6 Kf4 28. g6 Kf3 29. g7 Ke3 30. g8=Q
>Kd3 31. Qg6+ Ke3 32. Qg4 Kd3 33. Qb4 Ke3 34. Kg5 Ke2 35. Kf4 Kd3 36. Kf3 Kc2 37.
>Ke3 Kc1 38. Kd3 Kd1 39. Qb1# :)
>
>Of course, as usual, I'm not totally convinced this is correct, but I think it's
>close :) Anyway, 8... Kh8 is a much tougher defence than 8... Rh8.
Yes, I agree. Aren't we smart guys? :-)
>Groetjes,
>Paul
Have a nice night, and a good sleep ;-)
Heiner
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.