WAC 100

Discussion of chess software programming and technical issues.

Moderators: hgm, Rebel, chrisw

flok

Re: WAC 100

Post by flok »

embla ref 4043:

Code: Select all

info depth 1 seldepth 2 score cp 189 time 1 nodes 11 pv a4a3
info depth 2 seldepth 4 score cp 187 time 1 nodes 81 pv a4a3 a7b8
info depth 3 seldepth 6 score cp 193 ebf 3.12 time 1 nodes 188 pv a4a3 a7b8 a3a2
info depth 4 seldepth 8 score cp 191 ebf 1.82 time 1 nodes 419 pv a4a3 a7b8 a3a2 c7b6
info depth 5 seldepth 8 score cp 193 ebf 2.23 time 2 nodes 949 pv a4a3 a7b8 a3a2 c7b6 a2b1
info depth 6 seldepth 10 score cp 196 ebf 1.90 time 3 nodes 1787 pv a4a3 a7b8 a3a2 c7b6 a2b1 b8a7
info depth 7 seldepth 11 score cp 198 ebf 1.88 time 5 nodes 3667 pv a4a3 a7b8 a3a2 c7b6 a2b1 b8a7 d2c3
info depth 8 seldepth 13 score cp 196 ebf 1.38 time 7 nodes 5265 pv a4a3 a7b8 a3a2 c7b6 a2b1 b6c7 d2e3 c7b6
info depth 9 seldepth 14 score cp 193 ebf 1.61 time 13 nodes 10162 pv a4a3 a7b8 a3a2 c7b6 a2b1 b6c7 d2e3 c7b6 b1a1
info depth 10 seldepth 14 score cp 196 ebf 1.69 time 21 nodes 14752 pv a4a3 a7b8 a3a2 c7b6 a2b1 b6c7 d2e3 c7b6 b1a1 b6c7
info depth 11 seldepth 16 score cp 196 ebf 1.32 time 35 nodes 23344 pv a4a3 a7b8 a3a2 c7b6 a2b1 b6c7 d2e3 c7b6 b1a1 b6c7 a1a2
info depth 12 seldepth 17 score cp 193 ebf 1.82 time 57 nodes 38566 pv a4a3 a7b8 a3a2 c7b6 a2b1 b6c7 d2e3 c7b6 b1a1 b6c7 a1a2
info depth 13 seldepth 18 score cp 196 ebf 1.52 time 82 nodes 58350 pv a4a3 a7b8 a3a2 c7b6 a2b1 b6c7 d2e3 c7b6 b1a1 b8c7 a1a2 c7b8
info depth 14 seldepth 20 score cp 198 ebf 2.15 time 189 nodes 128505 pv a4a3 a7b8 a3a2 c7b6 d2e1 b6c7 e1f2 c7b6 a2b1 b8c7 f2g3 c7b8
info depth 15 seldepth 20 score cp 198 ebf 1.53 time 248 nodes 175071 pv a4a3 a7b8 a3a2 c7b6 d2e1 b6c7 e1f2 c7b6 a2b1 b8c7 f2g3 c7b8
info depth 16 seldepth 21 score cp 198 ebf 1.02 time 311 nodes 247510 pv a4a3 a7b8 a3a2 c7b6 d2e1 b6c7 e1f2 c7b6 a2b1 b8c7 f2g3 c7b8 b1a1 b6c7 a1b1 c7b6
info depth 17 seldepth 22 score cp 198 ebf 1.61 time 415 nodes 367698 pv a4a3 a7b8 a3a2 c7b6 d2e1 b6c7 e1f2 c7b6 a2b1 b8c7 f2g3 c7b8 g3e1 b6c7 e1g3 c7b6
info depth 18 seldepth 23 score cp 198 ebf 1.74 time 605 nodes 586026 pv a4a3 a7b8 a3a2 c7b6 d2e1 b6c7 e1f2 c7b6 a2b1 b8c7 f2g3 c7b8 g3e1 b6c7 e1g3 c7b6
info depth 19 seldepth 24 score cp 196 ebf 1.78 time 930 nodes 966364 pv a4a3 a7b8 a3a2 c7b6 d2e1 b6c7 e1f2 c7b6 a2b1 b8c7 f2g3 c7b8 g3e1 b6c7 e1f2 c7b6 f2g3
info depth 20 seldepth 25 score cp 196 ebf 1.57 time 1516 nodes 1503299 pv a4a3 a7b8 a3a2 c7b6 d2e1 b6c7 e1f2 c7b6 a2b1 b8c7 f2g3 c7b8 g3e1
info depth 21 seldepth 27 score cp 196 ebf 1.44 time 2349 nodes 2289533 pv a4a3 a7b8 a3a2 c7b6 d2e1 b6c7 e1f2 c7b6 a2b1 b8c7 f2e3 c7b8 b1a1 b8c7 e3d2 c7b8 d2c3 b6c7 a1a2 b8a7 c3d2
info depth 22 seldepth 28 score cp 196 ebf 1.63 time 3740 nodes 3723186 pv a4a3 a7b8 a3a2 c7b6 d2e1 b6c7 e1f2 c7b6 a2b1 b8c7 f2e3 c7b8 b1a1 b8c7 e3d2 c7b8 d2e1 b6c7 e1g3 c7b6
info depth 23 seldepth 31 score cp 196 ebf 1.56 time 5836 nodes 5646765 pv a4a3 a7b8 a3a2 c7b6 d2e1 b6c7 e1f2 c7b6 a2b1 b8c7 f2e3 c7b8 b1a1 b8c7 e3d2 c7b8 d2e1 b6c7 e1g3
info depth 24 seldepth 35 score cp 196 ebf 1.56 time 8966 nodes 9121775 pv a4a3 a7b8 a3a2 c7b6 d2e1 b6c7 e1f2 c7b6 a2b1 b8c7 f2e3 c7b8 b1a1 b8c7 e3d2 c7b8 d2e1 b6c7 e1g3
info depth 25 seldepth 34 score cp 196 ebf 1.64 time 13675 nodes 14312547 pv a4a3 a7b8 a3a2 c7b6 d2e1 b6c7 e1f2 c7b6 a2b1 b8c7 f2e3 c7b8 b1a1 b8c7 e3d2 c7b8 d2e1 b6c7 e1g3 c7b6
info depth 26 seldepth 37 score cp 196 ebf 1.48 time 21035 nodes 21945873 pv a4a3 a7b8 a3a2 c7b6 d2e1 b6c7 e1f2 c7b6 a2b1 b8c7 f2e3 c7b8 b1a1 b8c7 e3d2 c7b8 d2e1 b6c7 e1g3 c7b6
info depth 27 seldepth 39 score cp 271 ebf 2.10 time 42161 nodes 44759525 pv b5b6
bob
Posts: 20943
Joined: Mon Feb 27, 2006 7:30 pm
Location: Birmingham, AL

Re: WAC 100

Post by bob »

jwes wrote:
Dann Corbit wrote:Texel using MPV:3 found that Be3 is no worse than Mate in 42.

Code: Select all

     25	06:30	4,445,666,578	11,393,389	+M42	Be3 Bb6 Bd2 Bc7 b6+ Bxb6 Kb5 Bd8 Be3 Ka8 Kxc5 Ba5 Kc6 Bc3 c5 Bb2 Kb5 Ba3 c6 Bd6 Kc4 Kb8 Kd5 Bc7 Ke6 Ka8 Kd7 Ba5 a7 Bb6 Bc1 Kxa7 c7 Bxc7 Kxc7 Ka8 Bd2 Ka7 Be3+ Ka6 Kc6 Ka5 Kd7 Kb4 Bd2+ Kc4 Ke6 Kd3 Kxf5 Kxd2 Kxf6 Kd3 Kg6 Ke4 Kxh5 Kf5 Kh6 Ke4 Kg5 Kd5 Kg6 Kc4 f5 Kb3 h5 Ka2 f6 Ka1 h6 Kb1 f7 Ka1 f8Q Kb2 Qf3 Ka1 Qe2 Kb1 h7 Ka1 h8Q+ Kb1 Qeb2+
The PV shows that it repeats the start position and then plays b6+. I'm not sure that counts as an alternate solution.
It is actually OK. This was discussed a long time back with GM participation. It depends on what you are searching for. The perfectly shortest solution, or a solution that wins. In the Win At Chess book, the goal was stated as finding the move that wins by force. Not necessarily the shortest win by force. Used to be lots of arguments as the 300 positions were first subjected to good computer analysis.
petero2
Posts: 688
Joined: Mon Apr 19, 2010 7:07 pm
Location: Sweden
Full name: Peter Osterlund

Re: WAC 100

Post by petero2 »

bob wrote:
jwes wrote:
Dann Corbit wrote:Texel using MPV:3 found that Be3 is no worse than Mate in 42.

Code: Select all

     25	06:30	4,445,666,578	11,393,389	+M42	Be3 Bb6 Bd2 Bc7 b6+ Bxb6 Kb5 Bd8 Be3 Ka8 Kxc5 Ba5 Kc6 Bc3 c5 Bb2 Kb5 Ba3 c6 Bd6 Kc4 Kb8 Kd5 Bc7 Ke6 Ka8 Kd7 Ba5 a7 Bb6 Bc1 Kxa7 c7 Bxc7 Kxc7 Ka8 Bd2 Ka7 Be3+ Ka6 Kc6 Ka5 Kd7 Kb4 Bd2+ Kc4 Ke6 Kd3 Kxf5 Kxd2 Kxf6 Kd3 Kg6 Ke4 Kxh5 Kf5 Kh6 Ke4 Kg5 Kd5 Kg6 Kc4 f5 Kb3 h5 Ka2 f6 Ka1 h6 Kb1 f7 Ka1 f8Q Kb2 Qf3 Ka1 Qe2 Kb1 h7 Ka1 h8Q+ Kb1 Qeb2+
The PV shows that it repeats the start position and then plays b6+. I'm not sure that counts as an alternate solution.
It is actually OK. This was discussed a long time back with GM participation. It depends on what you are searching for. The perfectly shortest solution, or a solution that wins. In the Win At Chess book, the goal was stated as finding the move that wins by force. Not necessarily the shortest win by force. Used to be lots of arguments as the 300 positions were first subjected to good computer analysis.
I let texel analyze this position in multi-PV mode for about an hour and it claims there are 8 moves that lead to a forced mate. From a human point of view there does not seem to be 8 different winning plans though, only 2. The first plan is to push b6, play Kb5 and capture on c5. The second plan is to let or force black to take on f4, then invade the king side with the king. It seems like the second plan is not enforceable though, so in those lines black can force white to revert to plan 1.

So it seems there is only one "real" winning plan (depending on how you define "real" and "plan"), but from a game theoretic point of view there are still 8 winnings moves (unless the texel analysis is buggy).
[pgn]
[Event "WAC 100 analysis"]
[Site "?"]
[Date "2017.03.31"]
[Round "?"]
[White "?"]
[Black "?"]
[Result "1-0"]
[FEN "8/k1b5/P4p2/1Pp2p1p/K1P2P1P/8/3B4/8 w - - 0 1"]
[SetUp "1"]

{[37] m24:} 1. b6+ ({[37] m25:} 1. Be3 Bd6 2. Ka5 Bc7+ 3. b6+ Bxb6+ 4. Kb5 Bc7
5. Bxc5+ Ka8 6. Be3 Bb8 7. c5 Ka7 8. c6+ Ka8 9. Kc4 Bd6 10. Kd5 Bc7 11. Ke6 Bd8
12. Kd7 Bc7 13. Bc5 Ba5 14. Be7 Ka7 15. Bd8 Bc7 16. Bxc7 Kxa6 17. Bb8 Kb5 18.
c7 Kc4 19. c8=Q+ Kd4 20. Ba7+ Kd3 21. Ke6 Kd2 22. Qc4 Kd1 23. Be3 Ke1 24. Qc2
Kf1 25. Qf2#) ({[37] m41:} 1. Kb3 Bd6 2. Ka4 Kb6 3. Ba5+ Ka7 4. Bd8 Bxf4 5. Kb3
Bb8 6. Kc2 Bh2 7. Be7 f4 8. Bxc5+ Kb8 9. Kd3 f3 10. Ke4) ({[36] m39:} 1. Bc1
Kb6 2. Bd2 Bb8 3. Ba5+ Ka7 4. Bd8 Bxf4 5. Kb3 Bd6 6. Kc2 Bf8 7. Kd3 Bd6 8. Bxf6
Kb6 9. Bd8+ Ka7 10. Ke3 Bb8 11. b6+ Kxa6 12. Bc7 Bxc7 13. bxc7 Kb7 14. Kf4 Kxc7
15. Kxf5 Kd7 16. Kg6 Ke6 17. Kxh5 Kf5 18. Kh6 Kf6 19. h5 Kf7 20. Kg5 Kg7 21.
Kf4 Kf6 22. Ke4 Kg5 23. Kd5 Kxh5 24. Kxc5 Kg6 25. Kd5 Kf6 26. c5 Kf7 27. Ke5
Ke7 28. c6 Ke8 29. Ke6 Kd8 30. Kd6 Kc8 31. c7 Kb7 32. Kd7 Ka6 33. c8=Q+ Kb5 34.
Qc2 Kb4 35. Kc6 Ka3 36. Qb1 Ka4 37. Qb2 Ka5 38. Qa3#) ({[36] m41:} 1. Ka3 Bd6
2. Ka4 Kb6 3. Ba5+ Ka7 4. Bd8 Bxf4 5. Kb3 Bd6 6. Kc2 Bf8 7. Kd3 Bd6 8. Kc2)
({[36] m43:} 1. Bc3 Bxf4 2. Be1 Bc7 3. Kb3 Be5 4. Bf2 Bd6 5. Kc3 Kb6 6. Kd3 Bb8
7. Be1 Ka7 8. Ba5 f4 9. Ke2 f5 10. Bd2 Bd6 11. Kf3 Be7 12. Bxf4 Kb6 13. Bg5 Bd6
14. Bd8+ Ka7 15. Ke3 Bb8 16. b6+ Kxa6 17. Bc7 Bxc7 18. bxc7 Kb7 19. Kf4 Kxc7
20. Kxf5 Kd7 21. Kg6 Ke6 22. Kxh5 Kf5 23. Kh6 Kf6 24. h5 Kf7 25. Kg5 Kg7 26.
Kf4 Kf6 27. Ke4 Kg5 28. Kd5 Kxh5 29. Kxc5 Kg6 30. Kd5 Kf6 31. c5 Kf7 32. Ke5
Ke7 33. c6 Ke8 34. Ke6 Kd8 35. Kd6 Kc8 36. c7 Kb7 37. Kd7 Ka6 38. c8=Q+ Kb5 39.
Qc2 Kb4 40. Kc6 Ka3 41. Qb1 Ka4 42. Qb2 Ka5 43. Qa3#) ({[36] m43:} 1. Ba5 Bxf4
2. Be1 Bc7 3. Kb3 Be5 4. Bf2 Bd6 5. Kc3 Kb6 6. Kd3 Bb8 7. Be1 Ka7 8. Ba5 f4 9.
Ke2 f5 10. Bd2 Bd6 11. Kf3 Be7 12. Bxf4 Kb6 13. Bg5 Bd6 14. Bd8+ Ka7 15. Ke3
Bb8 16. b6+ Kxa6 17. Bc7 Bxc7 18. bxc7 Kb7 19. Kf4 Kxc7 20. Kxf5 Kd7 21. Kg6
Ke6 22. Kxh5 Kf5 23. Kh6 Kf6 24. h5 Kf7 25. Kg5 Kg7 26. Kf4 Kf6 27. Ke4 Kg5 28.
Kd5 Kxh5 29. Kxc5 Kg6 30. Kd5 Kf6 31. c5 Kf7 32. Ke5 Ke7 33. c6 Ke8 34. Ke6 Kd8
35. Kd6 Kc8 36. c7 Kb7 37. Kd7 Ka6 38. c8=Q+ Kb5 39. Qc2 Kb4 40. Kc6 Ka3 41.
Qb1 Ka4 42. Qb2 Ka5 43. Qa3#) ({[36] m951:} 1. Be1 Bxf4 2. Bf2 Bd6 3. Kb3 f4 4.
Kc2 Kb6 5. Kb3 Ka7) ({[36] 0.00:} 1. Bb4 cxb4 2. c5 b3 3. b6+ Bxb6 4. cxb6+
Kxa6 5. Kxb3 Kxb6 6. Kb4 Kc6 7. Kc4 Kd6 8. Kd4 Ke7 9. Kc5 Kd7 10. Kd5 Ke7 11.
Kc5) 1... Bxb6 2. Kb5 Bc7 3. Be3 Ka8 4. Kxc5 Kb8 5. Kb5 Ka8 6. c5 Bb8 7. c6 Bc7
8. Kc4 Bd6 9. Kd5 Bc7 10. Ke6 Bd8 11. Kd7 Bc7 12. Bc5 Ba5 13. Be7 Ka7 14. Bd8
Bc7 15. Bxc7 Kxa6 16. Bb8 Kb5 17. c7 Kc4 18. c8=Q+ Kd4 19. Ba7+ Kd3 20. Ke6 Kd2
21. Qc4 Kd1 22. Be3 Ke1 23. Qc2 Kf1 24. Qf2# 1-0
[/pgn]