Romichess bug (Michael?)

[Michael Sherwin]

Thanks Miguel!

That should be easy to fix.

[Michiel Koorn]

You're probably right. When cutechess-cli receives a draw claim in a position that is not a draw, it expects that the next move from the engine draws the game (eg. Crafty does this). If the engine doesn't send any move, you get the "connection stalls" error. I'll try to fix this so that it only results in a forfeit.

Why? An error is an error.. Keep debugging simple

[ilari]

You're probably right. When cutechess-cli receives a draw claim in a position that is not a draw, it expects that the next move from the engine draws the game (eg. Crafty does this). If the engine doesn't send any move, you get the "connection stalls" error. I'll try to fix this so that it only results in a forfeit.

Why? An error is an error.. Keep debugging simple

Why what?
Why wait for a move after an invalid draw claim? Because the Winboard specs don't specify how draw claims should be handled, and some engines send the claim first and the move later.
Why change the error to a forfeit due to an invalid draw claim? Because currently cutechess-cli will terminate the engine in this situation, but there's really no need to do that.

[Michiel Koorn]

Why ''fix'' it so it is a forfeit

[ilari]

Why ''fix'' it so it is a forfeit

The error message ("Invalid draw claim") is more informative than "Connection stalls", which should only occur when the engine stops responding to input. It's also consistent with other "invalid claim" errors.

[michiguel]

Why ''fix'' it so it is a forfeit

The error message ("Invalid draw claim") is more informative than "Connection stalls", which should only occur when the engine stops responding to input. It's also consistent with other "invalid claim" errors.

After 1200 games I got another one. I think we got the right diagnostic (see the final position)

Miguel

[Event "?"]
[Site "?"]
[Date "2010.08.31"]
[Round "221"]
[White "Gaviota-hs"]
[Black "Romichess-P3K"]
[Result "1-0"]
[PlyCount "159"]
[TimeControl "40/8"]

1. e4 {book} c5 {book} 2. Nf3 {book} e6 {book} 3. d4 {book} cxd4 {book}
4. Nxd4 {book} Nc6 {book} 5. Nc3 {book} d6 {book} 6. Be3 {book} Nf6 {book}
7. f4 {book} e5 {book} 8. Nf3 {book} Ng4 {book} 9. Bc1 {book} exf4 {book}
10. Bxf4 {book} Be7 {book} 11. Nd5 {+0.14/6 1.1s} O-O {-0.21/8 0.26s}
12. Be2 {+0.11/5 0.46s} Nf6 {-0.23/8 0.23s} 13. Nxe7+ {+0.73/6 0.20s}
Nxe7 {-0.52/10 0.25s} 14. Qxd6 {+1.16/5 0.16s} Nxe4 {-0.24/10 0.60s}
15. Qb4 {+1.09/5 0.16s} Nf6 {-0.09/9 0.24s} 16. Rd1 {+0.68/4 0.16s}
Ned5 {+0.31/9 0.34s} 17. Qc5 {+0.73/5 0.17s} Be6 {+0.40/9 0.32s}
18. Qb5 {+0.14/5 0.44s} Qc8 {+0.71/9 0.21s} 19. Bd6 {-0.06/4 0.16s}
Ne3 {+0.72/9 0.30s} 20. Bxf8 {-0.13/4 0.16s} Kxf8 {+0.64/9 0.37s}
21. a3 {-0.19/4 0.16s} Nxg2+ {+0.71/9 0.30s} 22. Kf2 {+0.57/5 0.17s}
Ne4+ {+0.63/10 0.34s} 23. Kg1 {+0.56/5 0.17s} Ne3 {+0.63/10 0.21s}
24. Qb4+ {+0.78/5 0.17s} Qc5 {+0.52/11 0.27s} 25. Qxc5+ {+0.73/5 0.18s}
Nxc5 {+0.70/11 0.25s} 26. Rd2 {+0.83/6 0.18s} Rc8 {+0.72/10 0.28s}
27. c3 {+0.71/6 0.28s} Nb3 {+1.19/11 0.29s} 28. Rd3 {+0.13/7 0.18s}
Nc4 {+1.24/11 0.24s} 29. Ng5 {+0.45/6 0.18s} Nxb2 {+1.19/10 0.27s}
30. Nxe6+ {+0.25/6 0.23s} fxe6 {+0.87/11 0.46s} 31. Re3 {+0.30/7 0.62s}
Na4 {+1.04/10 0.37s} 32. Bd1 {+0.65/6 0.16s} Nbc5 {+0.94/10 0.20s}
33. Bc2 {+0.61/6 0.16s} g6 {+0.95/10 0.16s} 34. h4 {+0.56/6 0.39s}
Nb2 {+1.16/9 0.12s} 35. h5 {+0.34/6 0.14s} Nc4 {+0.84/10 0.19s}
36. Rg3 {+0.67/5 0.15s} Rc7 {+0.99/8 0.088s} 37. hxg6 {+0.73/5 0.16s}
hxg6 {+0.69/8 0.084s} 38. Bxg6 {+0.67/5 0.17s} Nxa3 {+0.88/8 0.075s}
39. Rf3+ {+0.63/5 0.20s} Ke7 {+0.86/9 0.076s} 40. Rh4 {+0.44/6 0.32s}
Kd6 {+0.67/8 0.091s} 41. Rd4+ {+0.48/5 0.20s} Ke5 {+1.08/10 0.30s}
42. Re3+ {+0.46/4 0.20s} Kf6 {+1.18/11 0.24s} 43. Rg4 {+0.32/5 0.20s}
Rg7 {+1.18/11 0.25s} 44. Rf3+ {+0.33/4 0.19s} Ke7 {+1.48/11 0.56s}
45. Rfg3 {+0.50/6 0.19s} b5 {+1.65/11 0.25s} 46. Bh5 {+0.35/6 0.18s}
Rxg4 {+2.10/11 0.16s} 47. Bxg4 {+0.09/6 0.17s} a5 {+1.95/11 0.23s}
48. Rg2 {-0.06/6 0.17s} Nc4 {+2.08/10 0.17s} 49. Be2 {-0.15/6 0.16s}
Nd6 {+1.81/11 0.18s} 50. Rg7+ {+0.07/6 0.28s} Kf6 {+1.84/13 0.27s}
51. Rg4 {-0.17/6 0.16s} a4 {+2.55/10 0.21s} 52. Rd4 {-0.30/6 0.34s}
Ke5 {+2.68/11 0.19s} 53. Rd2 {-0.46/6 0.30s} Nce4 {+2.64/12 0.20s}
54. Rc2 {-0.57/6 0.15s} Kf4 {+2.31/11 0.17s} 55. Kf1 {-0.53/6 0.22s}
e5 {+2.47/12 0.19s} 56. Ke1 {-0.61/6 0.15s} Ke3 {+2.45/13 0.27s}
57. Bf1 {-0.44/6 0.15s} Kf3 {+2.24/11 0.24s} 58. Rh2 {-0.37/6 0.15s}
Kf4 {+2.40/11 0.17s} 59. Rc2 {-0.61/6 0.16s} Kg4 {+2.01/11 0.28s}
60. c4 {-0.36/5 0.16s} bxc4 {+1.81/12 0.26s} 61. Bxc4 {-0.33/6 0.16s}
Kf4 {+1.34/11 0.26s} 62. Bf1 {-0.25/6 0.24s} Ke3 {+1.70/11 0.22s}
63. Rc6 {-0.34/6 0.16s} Nf5 {+1.77/11 0.19s} 64. Bg2 {-0.49/6 0.16s}
Nd4 {+0.30/11 0.17s} 65. Rh6 {-0.52/6 0.36s} Nf2 {+1.86/10 0.16s}
66. Rh5 {-0.50/5 0.16s} Nc2+ {+1.95/11 0.21s} 67. Kf1 Nd3 {+1.89/11 0.19s}
68. Rh7 {-0.75/6 0.32s} a3 {+2.19/11 0.21s} 69. Ra7 {-0.89/6 0.41s}
Ndb4 {+4.43/12 0.23s} 70. Re7 {-0.81/5 0.15s} Kd4 {+4.83/11 0.20s}
71. Rd7+ {-3.01/4 0.30s} Kc5 {+5.15/13 0.24s} 72. Rc7+ {-3.20/4 0.15s}
Kd6 {+5.67/14 0.20s} 73. Ra7 {-3.56/4 0.15s} a2 {+5.67/13 0.12s}
74. Ke2 {-3.89/6 0.15s} a1=Q {+5.97/13 0.17s} 75. Rxa1 {-4.11/6 0.16s}
Nxa1 {+6.21/11 0.12s} 76. Kd2 {-4.42/6 0.16s} Kc5 {+6.19/11 0.12s}
77. Ke3 {-4.13/5 0.17s} Nd5+ {+6.42/10 0.080s} 78. Kd3 {-4.41/7 0.25s}
Nf4+ {+8.16/10 0.062s} 79. Ke4 {-4.56/8 0.19s} Nxg2 {+8.79/10 0.054s}
80. Kxe5 {0.00/64 0.021s, Black's connection stalls} 1-0

Final position, Romi must have claimed a draw (it is a draw but it can't be claimed).

EDIT: I take it back, Gaviota (white) cannot claim draw, but Romi (black) can!! Even if the position cannot be automatically be adjudicated, black can claim a draw because white has no mating material. Cutechess-cli should honor the claim, even if it is for the wrong reason.

[d]8/8/8/2k1K3/8/8/6n1/n7 b - - 0 80