Current Position Game #3 WCC = Stockfish sees mate

[MikeB]

After move 60 the timecontrol was 30sec/move +15min for the rest of the game.

Did Stockfish see the mate under this conditions?

No, SF was in analysis mode. Many of the moves were forced , i.e, there was only one move leading to advantage and when I got the mate score and went backwards to the position reference above , the mate score had stuck.

[AdminX]

[d]8/4R3/5k2/4NP1K/1p5p/1P5r/8/8 w - - 1 4

<< info depth 60 seldepth 89 multipv 1 score cp 12820 nodes 3017950030 nps 32803091 hashfull 990 tbhits 49113782 time 92002 pv e7e8 f6f5 e5c6 h3h1 c6d4 f5f4 e8e2 h4h3 h5h4 h3h2 d4e6 f4f3 e2b2 f3e4 e6c5 e4d5 c5a4 h1d1 b2h2 d1d4 h4g5 d4d3 h2b2 d5c6 g5f4 c6b5 f4e4 d3h3 e4d4 h3h4 d4d3 h4h3 d3c2 h3h1 b2b1 h1h2 c2c1 b5c6 a4b2 c6d5 b2d1 h2h3 b1b2 d5c6 c1d2 c6b5 d1f2 h3g3 f2d3 g3h3 b2b1 h3h2 d2e3 h2h3 e3e4 h3h4 d3f4 h4h7 f4e6 h7e7 e4e5 e7h7 e6d4 b5c5 b1c1 c5b6 c1a1 h7h5 e5d6 h5h6 d6d5 h6h7 d5c4 h7d7 a1a4 d7e7 a4b4

[AdminX]

[mar]

[d]8/4R3/5k2/4NP1K/1p5p/1P5r/8/8 w - - 1 4

<< info depth 60 seldepth 89 multipv 1 score cp 12820 nodes 3017950030 nps 32803091 hashfull 990 tbhits 49113782 time 92002 pv e7e8 f6f5 e5c6 h3h1 c6d4 f5f4 e8e2 h4h3 h5h4 h3h2 d4e6 f4f3 e2b2 f3e4 e6c5 e4d5 c5a4 h1d1 b2h2 d1d4 h4g5 d4d3 h2b2 d5c6 g5f4 c6b5 f4e4 d3h3 e4d4 h3h4 d4d3 h4h3 d3c2 h3h1 b2b1 h1h2 c2c1 b5c6 a4b2 c6d5 b2d1 h2h3 b1b2 d5c6 c1d2 c6b5 d1f2 h3g3 f2d3 g3h3 b2b1 h3h2 d2e3 h2h3 e3e4 h3h4 d3f4 h4h7 f4e6 h7e7 e4e5 e7h7 e6d4 b5c5 b1c1 c5b6 c1a1 h7h5 e5d6 h5h6 d6d5 h6h7 d5c4 h7d7 a1a4 d7e7 a4b4

Yes, stupid Carlsen cannot even search 3 billion nodes with 6-man tablebases...
An exciting game though.

[MikeB]

[d]8/4R3/5k2/4NP1K/1p5p/1P5r/8/8 w - - 1 4

<< info depth 60 seldepth 89 multipv 1 score cp 12820 nodes 3017950030 nps 32803091 hashfull 990 tbhits 49113782 time 92002 pv e7e8 f6f5 e5c6 h3h1 c6d4 f5f4 e8e2 h4h3 h5h4 h3h2 d4e6 f4f3 e2b2 f3e4 e6c5 e4d5 c5a4 h1d1 b2h2 d1d4 h4g5 d4d3 h2b2 d5c6 g5f4 c6b5 f4e4 d3h3 e4d4 h3h4 d4d3 h4h3 d3c2 h3h1 b2b1 h1h2 c2c1 b5c6 a4b2 c6d5 b2d1 h2h3 b1b2 d5c6 c1d2 c6b5 d1f2 h3g3 f2d3 g3h3 b2b1 h3h2 d2e3 h2h3 e3e4 h3h4 d3f4 h4h7 f4e6 h7e7 e4e5 e7h7 e6d4 b5c5 b1c1 c5b6 c1a1 h7h5 e5d6 h5h6 d6d5 h6h7 d5c4 h7d7 a1a4 d7e7 a4b4

Yes, stupid Carlsen cannot even search 3 billion nodes with 6-man tablebases...
An exciting game though.

Yes very nice game , if one looks at the line , the 13th move for white, e2b2, is so counter-intuitive , I'm not sure anyone could find that move without computer help and it is apparently the only move that preserves the win at that point.

[mar]

Yes very nice game , if one looks at the line , the 13th move for white, e2b2, is so counter-intuitive , I'm not sure anyone could find that move without computer help and it is apparently the only move that preserves the win at that point.

Do you have the pgn? I'm too lazy to google it
I also think it would be nice to have a thread here dedicated to WCC.

[AdminX]

Yes very nice game , if one looks at the line , the 13th move for white, e2b2, is so counter-intuitive , I'm not sure anyone could find that move without computer help and it is apparently the only move that preserves the win at that point.

Do you have the pgn? I'm too lazy to google it
I also think it would be nice to have a thread here dedicated to WCC.

[pgn][Event "AGON FWCM 2016"]
[Site "New York"]
[Date "2016.11.14"]
[Round "3"]
[White "Carlsen, Magnus"]
[Black "Karjakin, Sergey"]
[Result "1/2-1/2"]
[ECO "C67"]
[WhiteElo "2857"]
[BlackElo "2769"]
[Annotator "ChessBase"]
[PlyCount "156"]
[EventDate "2016.??.??"]
[EventType "match"]
[EventCountry "USA"]
[SourceTitle "playchess.com"]
[Source "ChessBase"]
[TimeControl "40/6000+30:20/3000+30:900+30"]

1. e4 {0} e5 {9} 2. Nf3 {0} Nc6 {3} 3. Bb5 {0} Nf6 {10} 4. O-O {0} Nxe4 {5} 5.
Re1 {5} Nd6 {7} 6. Nxe5 {11} Be7 {6} 7. Bf1 {6} Nxe5 {37} 8. Rxe5 {4} O-O {7}
9. d4 {11} Bf6 {26} 10. Re2 {7} b6 {1508} 11. Re1 {11} Re8 {251} 12. Bf4 {122}
Rxe1 {13} 13. Qxe1 {2} Qe7 {366} 14. Nc3 {908} Bb7 {33} 15. Qxe7 {136} Bxe7 {14
} 16. a4 {26} a6 {254} 17. g3 {308} g5 {540} 18. Bxd6 {363} Bxd6 {13} 19. Bg2 {
740} Bxg2 {8} 20. Kxg2 {2} f5 {624} 21. Nd5 {463} Kf7 {16} 22. Ne3 {483} Kf6 {
39} 23. Nc4 {206} Bf8 {119} 24. Re1 {227} Rd8 {125} 25. f4 {343} gxf4 {1046}
26. gxf4 {2} b5 {11} 27. axb5 {14} axb5 {1} 28. Ne3 {91} c6 {76} 29. Kf3 {54}
Ra8 {122} 30. Rg1 {415} Ra2 {293} 31. b3 {82} c5 {157} 32. Rg8 {347} Kf7 {283}
33. Rg2 {59} cxd4 {27} 34. Nxf5 {34} d3 {7} 35. cxd3 {7} Ra1 {79} 36. Nd4 {278}
b4 {9} 37. Rg5 {227} Rb1 {606} 38. Rf5+ {115} Ke8 {25} 39. Rb5 {166} Rf1+ {124}
40. Ke4 {0} Re1+ {139} 41. Kf5 {134} Rd1 {112} 42. Re5+ {659} Kf7 {117} 43. Rd5
{65} Rxd3 {386} 44. Rxd7+ {132} Ke8 {115} 45. Rd5 {33} Rh3 {30} 46. Re5+ {261}
Kf7 {76} 47. Re2 {55} Bg7 {785} 48. Nc6 {347} Rh5+ {448} 49. Kg4 {404} Rc5 {14}
50. Nd8+ {245} Kg6 {8} 51. Ne6 {131} h5+ {121} 52. Kf3 {7} Rc3+ {5} 53. Ke4 {
111} Bf6 {18} 54. Re3 {323} h4 {186} 55. h3 {60} Rc1 {268} 56. Nf8+ {656} Kf7 {
20} 57. Nd7 {5} Ke6 {145} 58. Nb6 {46} Rd1 {56} 59. f5+ {8} Kf7 {60} 60. Nc4 {0
} Rd4+ {787} 61. Kf3 {75} Bg5 {18} 62. Re4 {415} Rd3+ {14} 63. Kg4 {6} Rg3+ {6}
64. Kh5 {4} Be7 {127} 65. Ne5+ {88} Kf6 {74} 66. Ng4+ {10} Kf7 {205} 67. Re6 {
320} Rxh3 {340} 68. Ne5+ {69} Kg7 {2} 69. Rxe7+ {221} Kf6 {2} 70. Nc6 {94} Kxf5
{145} 71. Na5 {230} Rh1 {10} 72. Rb7 {30} Ra1 {216} 73. Rb5+ {29} Kf4 {4} 74.
Rxb4+ {22} Kg3 {20} 75. Rg4+ {230} Kf2 {2} 76. Nc4 {122} h3 {9} 77. Rh4 {30}
Kg3 {8} 78. Rg4+ {8} Kf2 {6} 1/2-1/2
[/pgn]

http://talkchess.com/forum/viewtopic.php?t=62073

[mar]

Thanks Ted!

[Rochester]

I say "no". The stockfish programme see the win but he not see the mate. You can check the uci codes.

[tpoppins]

Indeed, a 128.xx eval stands for a tablebase win, not mate.
And even at d=61 44 billion nodes (TBs off) SF still doesn't see mate. This position has already been analyzed on Let's Check, too:



SF 121116, d=74 - the mate remains elusive.


BTW, can anyone refresh my memory on what the ".xx" part of the tablebase win eval represents? I recall seeing Ronald explaining it on here a couple of years ago but cannot find the post now.