Complex Puzzle

[peter]

At least the "solution variant" isn't correct, after 4...Ne7+ instead of ...Bc8? White with 5.Nxe7 is mate in 52, with everything else faster

Did put it in these words not being sure 4...Ne7+ to be the shortest way to win even, neither as for the moves from starting position.
But after Backward of one of the won 7somes we get such output from komodo with full Syzygys:


[d]6nn/2N5/B5b1/1K6/7b/6N1/8/3k4 w - - 0 1

Analysis by Komodo 9.42 64-bit:

1.Nh1 Be4 2.Kb6 Bxh1 3.Nb5 Nf7 4.Bc8 Nf6 5.Be6 Bf2+ 6.Kc7 Ne8+ 7.Kb8 Ne5 8.Nc3+ Kc2 9.Na4 Bc6 10.Bf5+ Kd2 11.Nb2 Nd6 12.Be6 Bb6 13.Nc4+ Ndxc4 14.Kc8 Nd7 15.Bxc4
-+ (-250.00) Depth: 42 00:00:17 534MN, tb=16507420


and after 1.Nf1:

6nn/2N5/B5b1/1K6/7b/8/8/3k1N2 b - - 0 1

Analysis by Komodo 9.42 64-bit:

1...Bd3+ 2.Kb4 Be1+ 3.Kb3 Bxa6 4.Ne3+ Kd2 5.Nf5 Bc8 6.Nd4 Bg3 7.Ncb5 Kd3 8.Nf3 Nf7 9.Kb2 Bg4 10.Ng1 Bf2 11.Na3 Bxg1 12.Nc2 Nh8 13.Kb1 Bc5 14.Kb2 Kd2 15.Ka1 Kxc2
-+ (-250.00) Depth: 45 00:00:05 174MN, tb=3023237

Seems to be cooked, because even if 50 moves wouldn't be enough to reach 6some after 7some, (but I'd be rather sure they would, having seen the one Lomonosov- line after 4...Nxe7), that wouldn't matter as for a study.
Studies don't take 50 moves rule into acount

[zullil]

[d]6nn/2N5/B5b1/1K6/7b/6N1/8/3k4 w - -

Possible Cook?

Stockfish says Black wins:

Code: Select all

info depth 67 seldepth 110 multipv 1 score cp -12340 nodes 1352200816200 nps 45259556 hashfull 999 tbhits 34282108674 time 29876581 pv g3f1 g6d3 b5b4 h4e1 b4b3 d3a6 f1e3 d1d2 e3f5 a6c8 f5d4 d2d3 d4c2 e1g3 c2b4 d3d2 c7b5 c8e6 b3b2 g3e5 b2b1 g8e7 b5a3 e5d6 b1b2 d6b4 a3b5 b4a3 b2b1 e6a2 b1a1 a3b2 a1a2

[peter]

Stockfish says Black wins

Hi Louis!
Seem to be little less than 50 moves to mate.

[d]6nn/2N5/B5b1/1K6/7b/6N1/8/3k4 w - - 0 1

1. Nf1 Bd3+ 2. Kc6 Bxa6 3. Ne3+ Ke2 4. Nf5 Bd3 5. Nxh4 Kf2 6. Nd5 Bb1 7. Nf4 Nf6 8. Nhg2 Ng4 9. Kc5 Kg3 10. Kc4 Nf7 11. Kb4 Nfh6 12. Ka4 Nf5 13. Ka3 Be4 14. Kb3 Nh4 15. Kc3 Nxg2 16. Ne2+ Kf2 17. Kd4 Bg6 18. Nc1 Ne1 19. Nb3 Nf3+ 20. Kc5 Ke3 21. Na5 Nd2 22. Nc4+ Nxc4 23. Kxc4 Bf7+ 24. Kc3 Be6 25. Kc2 Kd4 26. Kb2 Ne3 27. Kb1 Kc3 28. Ka1 Nc2+ 29. Kb1 Bd5 30. Kc1 Ba2 31. Kd1 Nd4 32. Ke1 Bd5 33. Kd1 Be4 34. Kc1 Ne2+ 35. Kd1 Kd3 36. Ke1 Ke3 37. Kd1 Bd3 38. Ke1 Bc2 39. Kf1 Nf4 40. Ke1 Ng2+ 41. Kf1 Kf3 42. Kg1 Ne3 43. Kh1 Bf5 44. Kg1 Kg3 45. Kh1 Kf2 46. Kh2 Nf1+ 47. Kh1 Be4#

Built line with SF MateFinder and Lomonosovs but can't take #-output backward further than to 4th move, spitting there with 4...Ne7+, which would be -#53.

[peter]

, spitting there with 4...Ne7+

Too late to edit: read splitting for spitting please!

[Nordlandia]

No. 1331: P. Rossi. As far as I have been able to discover, the solution
has not been published in the magazine. Presumably (AJR) 1. Kgl
Rh6 2. c7 Kh5 3. c8S g6 4. Sb6 ab 5. a7 ba 6. a8Q(B) a4 7, Qd5 ed
8. e6 d4 9. e7 d3 10. e8S and mates.

The intended solution:

[pgn][Event "?"]
[Site "?"]
[Date "????.??.??"]
[Round "?"]
[White "New game"]
[Black "?"]
[Result "1-0"]
[SetUp "1"]
[FEN "8/p5pp/P1P1p1k1/P3P1p1/6p1/P5P1/6Pr/5K2 w - - 0 0"]
[PlyCount "21"]

1. Kg1 Rh6 2. c7 Kh5 3. c8=N g6 4. Nb6 axb6 5. a7 bxa5 6. a8=Q a4 7. Qd5 exd5
8. e6 d4 9. e7 d3 10. e8=N d2 11. Nf6# 1-0[/pgn]

But the author overlooked 3...Kg6!

[d]2N5/p5pp/P3p1kr/P3P1p1/6p1/P5P1/6P1/6K1 w - - 0 3

Conclusion: it's draw after all, although the main variation is not forced.

[Nordlandia]

[d]kB5K/2P4B/P4rP1/4r3/2N1n3/1RP4N/3b4/6RQ b - - 0 0

Stockfish is not certain whether is's draw or win.

Desperado rook!

[zullil]

[d]kB5K/2P4B/P4rP1/4r3/2N1n3/1RP4N/3b4/6RQ b - - 0 0
Stockfish is not certain whether is's draw or win.

Desperado rook!

My Stockfish "thinks" draw:

Code: Select all

info depth 124 seldepth 128 multipv 1 score cp 0 nodes 17510846883 nps 44876708 hashfull 999 tbhits 0 time 390199 pv f6f8 h8g7 d2h6 g7h6 e5h5 h6h5 f8f5 h3g5 f5g5 h5h6 g5g6 h6h5 g6g5 h5h4 g5g4 h4h3 g4g3 h3h2 g3h3 h2g2 h3g3 g2f1 g3f3 f1e2 f3e3 e2d1 e3d3 d1c1 d3d1 c1b2 d1d2 b2a3 d2a2 a3b4 a2a4 b4b5 a4a5 b5c6 a5c5 c6d7 c5d5 d7e8 d5e5 e8f7 e5f5 f7g8 f5g5 h7g6 g5g6 g8h7 g6g7 h7h6 g7g6 h6h5 g6g5 h5h4 g5g4 h4h3 g4g3 h3h2 g3h3 h2g2 h3g3 g2f1 g3f3 f1e2 f3e3 e2d1 e3d3 d1c1 d3d1 c1b2 d1d2 b2a3 d2a2 a3b4 a2a4 b4b5 a4a5 b5c6 a5c5 c6d7 c5d5 c4d6 d5d6 d7e8 d6e6 e8f8 e6f6 f8g7 f6g6 g7h7 g6g7 h7h6 g7g6 h6h5 g6g5 h5h4 g5g4 h4h3 g4g3 h3h2 g3h3 h2g2 h3g3 g2f1 g3f3 f1e1 f3e3 e1d1 e3d3 d1c1 d3d1 c1b2 d1d2 b2a3 d2a2 a3b4 a2a4 b4b5 a4b4 b5c6 b4c4 c6d7 c4d4 d7e6 d4d6

[Nordlandia]

Stockfish evaluates it as 0.00 because of "Horizon Effect"

It quickly change it view whether it's draw or won.

https://en.wikipedia.org/wiki/Horizon_effect

https://chessprogramming.wikispaces.com/Horizon+Effect

[peter]

But the author overlooked 3...Kg6!

[d]2N5/p5pp/P3p1kr/P3P1p1/6p1/P5P1/6P1/6K1 w - - 0 3

Conclusion: it's draw after all, although the main variation is not forced.

After 3...Kg6 it's mate in 24.

2N5/p5pp/P3p1kr/P3P1p1/6p1/P5P1/6P1/6K1 w - - 0 1

Analysis by SF MF110416 64 sse:

4.a4 Kh5 5.Ne7 Rg6 6.Kf2 Rh6 7.Ke3 Rg6 8.Kd4 Rh6 9.Kc5 Rf6 10.exf6 gxf6 11.Nc6 f5 12.Nxa7 f4 13.Kd4 f3 14.gxf3 gxf3 15.Ke3 Kg4 16.Nb5 Kxg3 17.a7 f2 18.Ke2 g4 19.a8Q h6 20.a6 f1N 21.Kxf1 Kh4 22.a7 Kg5 23.Qf8 g3 24.a8Q e5 25.Qe4 g2+ 26.Kxg2 h5 27.Qg7#
+- (#24) Depth: 51/50 00:01:40 3445MN

Author's line is not the longest way to mate but the one he thinks to be clearest and most beautiful, so a ? at 3...g6 maybe is just unnecessary for a study too, even if there's only a not quite half as long way to mate.

That's a Zugzwang- study, not an mate-in-x-problem, so it's not about the number of moves to mate but about some main thematic lines.
Yet the longer ways should be shown too, at least a few moves long to get the point more easily without engines' help.

[Event "?"]
[Site "?"]
[Date "????.??.??"]
[Round "?"]
[White "Rossi, P,."]
[Black "?"]
[Result "1-0"]
[Annotator "Martan,Peter"]
[SetUp "1"]
[FEN "8/p5pp/P1P1p1k1/P3P1p1/6p1/P5P1/6Pr/5K2 w - - 0 0"]
[PlyCount "53"]

1. Kg1 Rh6 (1... Kh5 2. c7 h6 3. c8=N g6 4. Nb6 axb6 5. a7 bxa5 6. a8=Q a4 7.
Qd5 Rh4 8. gxh4 exd5 9. e6 gxh4 10. e7 Kg5 11. e8=Q Kf5 12. Qxa4 h3 13. gxh3
gxh3 14. Qd7+ Ke5 15. Qxh3 Kd4 16. a4 Kc4 17. a5 Kb5 18. Qa3 Kc4 19. a6 Kd4 20.
a7 Ke4 21. a8=Q g5 22. Qd6 Kf3 23. Qaxd5+ Kg4 24. Qg2+ Kf5 25. Qf3#) 2. c7 Kh5
3. c8=N Kg6 (3... g6 $2 4. Nb6 axb6 5. a7 bxa5 6. a8=Q a4 7. Qd5 exd5 8. e6 d4
9. e7 d3 10. e8=N d2 11. Nf6#) 4. a4 (4. Ne7+ $2 Kf7) 4... Kh5 (4... Kf5 5.
Nxa7 Kxe5 6. Nc6+ Kd6 7. a7 Rf6 8. a8=Q h5 9. a6 Rf7 10. Qb8+ Kc5 11. Qe5+ Kc4
12. Qxe6+ Kc3 13. Qxf7 h4 14. a7 h3 15. a8=Q h2+ 16. Kxh2 Kb2 17. Qf2+ Kc3 18.
Qd4+ Kc2 19. Qb7 g6 20. Qbb2#) (4... Kf7 5. Nxa7 Rg6 6. Nb5 Rf6 7. exf6 Kxf6 8.
a7 Ke5 9. a8=Q h5 10. a6 h4 11. a7 h3 12. Qb8+ Kf6 13. a8=Q h2+ 14. Kxh2 Kf5
15. Qf8+ Kg6 16. Qae8+ Kh7 17. Qh8#) 5. Ne7 Rg6 (5... g6 6. Nd5 exd5 7. e6 d4
8. e7 d3 9. e8=N d2 10. Ng7#) 6. Kf2 Kh6 7. Ke3 Kh5 8. Kd4 Rh6 9. Kc5 Rf6 10.
exf6 gxf6 11. Nc8 f5 12. Nxa7 f4 13. Kd4 f3 14. Ke3 fxg2 15. Kf2 e5 16. Nb5 e4
17. a7 e3+ 18. Kxg2 e2 19. Kf2 e1=B+ 20. Kxe1 Kh6 21. a8=Q Kg7 22. Qe4 Kf6 23.
a6 Kg7 24. a7 h5 25. a8=Q h4 26. Qa7+ Kf6 27. Qae7# 1-0

[Ignacio]

kB5K/2P4B/P4rP1/4r3/2N1n3/1RP4N/3b4/6RQ b - - bm #-67; Defence;

It's mate in 67.