New opponent, new handicaps for Komodo.

[Lyudmil Tsvetkov]

this is just a quick game I played, with SF having 1 min. for the entire game, pressed for time now, but I do not know how it is possible not to win with 2 pawns more.

You next?

the knight for pawn is easy too.

[pgn][Event "Blitz 1m"]
[Site "Microsoft"]
[Date "2017.07.03"]
[Round "?"]
[White "Stockfish 8 64 POPCNT"]
[Black "dark horse, owner"]
[Result "0-1"]
[Annotator "owner"]
[SetUp "1"]
[FEN "rnbqkbnr/ppppp1pp/8/8/8/8/PPPPPPPP/RNBQKB1R w KQkq - 0 1"]
[PlyCount "184"]
[EventDate "2017.??.??"]
[TimeControl "60"]

{512MB, OWNER-PC} 1. e4 {-2.02/17 3} e6 {2} 2. d4 {-1.60/18 2} d6 {2} 3. Bd3 {
-1.58/19 2} g6 {3} 4. c3 {-1.54/17 1} Bh6 {5} 5. O-O {-1.45/17 1} Bxc1 {11} 6.
Qxc1 {-1.55/18 2} Qh4 {12} 7. Nd2 {-1.16/16 1} Qh6 {5} 8. f4 {-1.24/22 0} Ne7 {
4} 9. Nf3 {-1.34/21 1} O-O {6} 10. g3 {-1.38/19 2} Nbc6 {5} 11. b4 {-1.25/16 1}
a6 {3} 12. a4 {-1.45/18 0} Nd8 {8} 13. Qc2 {-1.58/17 2} b6 {9} 14. a5 {-1.25/
14 1} Bb7 {23} 15. Qe2 {-1.23/19 7} b5 {7} 16. Rac1 {-1.21/21 0} d5 {19} 17. e5
{-1.04/18 2} Nf7 {45} 18. Kh1 {-0.96/28 0} Bc6 {5} 19. Rce1 {-0.70/16 1} Kh8 {3
} 20. Nd2 {-0.71/19 1} Rae8 {12} 21. Nb3 {-0.50/16 1} Bd7 {5} 22. Nc5 {-0.90/
20 3} Bc8 {2} 23. Qg2 {-0.94/19 1} g5 {46} 24. Qd2 {-1.27/25 0} g4 {126} 25.
Qa2 {-1.13/18 1} c6 {14} 26. Qf2 {-0.96/17 2} Nf5 {36} 27. Kg2 {-1.18/25 0} Rg8
{12} 28. Rg1 {-0.97/16 1} Qg6 {16} 29. Kf1 {-0.92/18 1} h5 {8} 30. Ke2 {-0.93/
20 1} N7h6 {5} 31. Kd2 {-0.81/18 1} Re7 {10} 32. Rg2 {-1.07/19 1} Reg7 {8} 33.
Kc1 {-0.98/27 0} Qe8 {3} 34. Kb2 {-1.04/19 1} Qe7 {2} 35. Rh1 {-1.02/18 1} h4 {
55} 36. Qe2 {-1.12/19 1} Rg6 {41} 37. Rgg1 {-0.97/18 4} R8g7 {5} 38. Bc2 {-0.
75/15 0} Kg8 {5} 39. Bd3 {-0.84/17 1} Kf8 {2} 40. Qf2 {-0.79/19 0} Rg8 {18} 41.
Rc1 {-1.13/14 0} R6g7 {8} 42. Ra1 {-1.56/18 0} Ke8 {43} 43. Bc2 {-1.68/21 0}
Kd8 {2} 44. Bd3 {-1.68/24 2} Kc7 {2} 45. Rae1 {-1.70/19 2} Kb8 {3} 46. Qd2 {-1.
79/17 1} Ka7 {4} 47. Rhg1 {-1.68/16 0} Rh8 {47} 48. Re2 {-1.62/20 0} Qf8 {15}
49. Reg2 {-1.62/21 0} hxg3 {28} 50. hxg3 {-1.54/16 0} Ng8 {11} 51. Bc2 {-1.53/
34 0} Nge7 {13} 52. Nd3 {-1.60/34 0} Rgh7 {16} 53. Bb3 {-1.60/19 0} Bd7 {22}
54. Bc2 {-1.60/18 0} Be8 {15} 55. Nc5 {-1.42/13 0} Qg8 {10} 56. Qf2 {-1.75/28 0
} Rh3 {14} 57. Bd1 {-1.81/26 0} R8h6 {48} 58. Kc1 {-1.47/16 0} Bh5 {19} 59. Be2
{-1.26/16 0} Rh7 {17} 60. Kb2 {-1.36/17 0} Nh6 {12} 61. Bf1 {-1.66/16 0} Bg6 {8
} 62. Ka2 {-2.06/24 0} Bf5 {11} 63. Qe1 {-2.10/16 0} Nf7 {41} 64. Rf2 {-1.97/
13 0} Rh2 {5} 65. Be2 {-1.81/14 0} Qh8 {15} 66. Rff1 {-2.08/16 0} R7h3 {17} 67.
Ka1 {-1.93/16 0} Ng8 {0} 68. Rf2 {-1.83/17 0} Rxf2 {12} 69. Qxf2 {-2.09/14 0}
Rh2 {2} 70. Qf1 {-2.66/17 0} Qh5 {11} 71. Rg2 {-2.14/14 0} Rh1 {3} 72. Rg1 {-1.
80/13 0} Rh2 {12} 73. Rg2 {0.00/27 0} Qh3 {9} 74. Rxh2 {-1.66/15 0} Qxh2 {2}
75. Qe1 {-3.05/25 0} Nfh6 {36} 76. Kb2 {-2.14/13 0} Ne7 {4} 77. Ka3 {-3.60/26 0
} Ng6 {38} 78. Bd1 {-2.08/13 0} Nh4 {60} 79. gxh4 {-2.47/13 0} g3 {1} 80. Bf3 {
-3.53/21 0} g2 {8} 81. Bxg2 {-4.19/27 0} Qxg2 {1} 82. Nb3 {-3.91/23 0} Ng4 {42}
83. Nd2 {-4.75/33 0} Qh3 {0} 84. Nf1 {-4.24/14 0} Bd3 {14} 85. Nd2 {-3.29/12 0}
Bc4 {28} 86. Nxc4 {-2.64/10 0} bxc4 {4} 87. Qd2 {-4.64/23 0} Qe3 {11} 88. Qxe3
{-1.54/13 0} Nxe3 {1} 89. h5 {-4.55/28 0} Nf5 {1} 90. Kb2 {-6.97/28 0} Kb7 {2}
91. b5 {-8.76/16 0} cxb5 {13} 92. Ka3 {-8.83/17 0} Kc6 {20} 0-1

[/pgn]

[Nordlandia]

Code: Select all

man, I still have not downloaded latest SF development.

I downloaded SF 8, and the next SF version I will download will be SF 9.

Even sesse supercomputer had to update stockfish, by massive fan requests.

Its certianly the right time to update SF to something more fresh Lyudmil Tsvetkov.

SF8 is outdated.

[Lyudmil Tsvetkov]

Code: Select all

man, I still have not downloaded latest SF development.

I downloaded SF 8, and the next SF version I will download will be SF 9.

Even sesse supercomputer had to update stockfish, by massive fan requests.

Its certianly the right time to update SF to something more fresh Lyudmil Tsvetkov.

SF8 is outdated.

nothing fresher in sight, unfortunately.
I see they are still at some +30 or so, and that does not look like very fresh to me.

in any case, I enjoy sticking to a particular engine for at least a year or so, otherwise, you will never know there is something new.

[Uri Blass]

Code: Select all

man, I still have not downloaded latest SF development.

I downloaded SF 8, and the next SF version I will download will be SF 9.

Even sesse supercomputer had to update stockfish, by massive fan requests.

Its certianly the right time to update SF to something more fresh Lyudmil Tsvetkov.

SF8 is outdated.

nothing fresher in sight, unfortunately.
I see they are still at some +30 or so, and that does not look like very fresh to me.

in any case, I enjoy sticking to a particular engine for at least a year or so, otherwise, you will never know there is something new.

+30 or even +27.41 in the last regression test is a very big improvement.

http://tests.stockfishchess.org/tests/v ... 3ea732d332

ELO: 27.41 +-1.6 (95%) LOS: 100.0%
Total: 40000 W: 6067 L: 2918 D: 31015

Number of wins is more than twice number of losses.

[Lyudmil Tsvetkov]

the position with equal material/king on e4 is a hard one, at least for me.

I get mostly draws, or, even worse, positions with some white advantage, but so complicated, that playing them at quick TC looks like an impossibility.

[pgn][Event "Blitz 1m"]
[Site "Microsoft"]
[Date "2017.07.04"]
[Round "?"]
[White "dark horse, owner"]
[Black "Stockfish 8 64 POPCNT"]
[Result "*"]
[Annotator "owner"]
[SetUp "1"]
[FEN "rnbq1bnr/ppppp1pp/5p2/8/4k3/8/PPPPPPPP/RNBQKBNR w KQ - 0 1"]
[PlyCount "62"]
[EventDate "2017.??.??"]
[TimeControl "60"]

{512MB, OWNER-PC} 1. d4 {0} Kf5 {2.42/19 3} 2. e4+ {2} Kg6 {2.33/19 0} 3. f4 {4
} Kf7 {1.97/14 1} 4. Nf3 {4} e6 {1.78/15 1} 5. Bd3 {3} d5 {1.44/19 0} 6. e5 {3}
c5 {1.42/18 1} 7. c3 {3} Nc6 {1.36/18 2} 8. O-O {2} g6 {1.27/20 0} 9. Be3 {3}
c4 {1.33/20 2} 10. Bc2 {2} f5 {1.10/17 1} 11. b4 {3} Nh6 {0.91/22 2} 12. a4 {3}
Be7 {0.81/21 1} 13. h3 {4} Rf8 {0.81/20 1} 14. Kh1 {19} Kg8 {0.81/21 1} 15. Rg1
{3} a6 {0.88/20 6} 16. g4 {9} Bd7 {0.80/19 1} 17. Ra2 {0} Qc7 {0.83/19 6} 18.
Rb2 {7} b6 {0.86/20 0} 19. Qe1 {32} Rab8 {1.10/22 0} 20. Nbd2 {27} Qc8 {1.07/
25 0} 21. Nf1 {7} Kh8 {1.07/19 1} 22. Bd1 {69} Rg8 {0.89/18 1} 23. Rbg2 {59} b5
{1.14/18 2} 24. a5 {4} Qf8 {1.45/21 1} 25. Bd2 {6} Be8 {1.36/23 0} 26. g5 {15}
Nf7 {1.16/18 1} 27. h4 {8} Nfd8 {1.09/21 1} 28. h5 {29} Nxb4 {1.05/21 1} 29.
cxb4 {5} Nc6 {0.94/24 1} 30. hxg6 {32} Bxg6 {1.05/21 1} 31. Rh2 {26} Bxb4 {1.
20/27 0} *

[/pgn]

[d]1r3qrk/7p/p1n1p1b1/Pp1pPpP1/1bpP1P2/5N2/3B3R/3BQNRK w - - 0 32

I stop the game here, as the position is simply too complicated for me to play quickly.

anyone seeing easily what the best continuations for both sides are?

[JJJ]

I think even Carslen wouldn't win this handicap game.
Human can't win positional advantage like material advantage, even if the handicap has the same rating, because positional advantage has less winning move than material advantage. Also, a blunder cost quickly more than in material advantage where you can get at least a draw after that.

[duncan]

3. Komodo plays Black with his king starting on e4 and pawn on f6, White to move!! The pawn on f6 makes the position legal. This looks like it should be a simple forced mate, but Komodo doesn't see one.

so what is the score when komodo plays against itself. ?

[lkaufman]

3. Komodo plays Black with his king starting on e4 and pawn on f6, White to move!! The pawn on f6 makes the position legal. This looks like it should be a simple forced mate, but Komodo doesn't see one.

so what is the score when komodo plays against itself. ?

I used Fritz 15 instead of Komodo to make use of the Monte Carlo option, with which I could get lots of variety very quickly. Somewhat to my surprise, White scored 84% in 700 games at ten ply; I expected more like 90%. Probably at any realistic time limit it would be close to 90%. I imagine many human GMs would say White had full compensation for a piece if we removed a White knight, at least at first glance. But apparently this is not so. When I removed the White knight from g1 from the above position, Black scored the same 84% that White scored with the knight! So we can say that the Ke4 handicap is "half a knight".

[duncan]

3. Komodo plays Black with his king starting on e4 and pawn on f6, White to move!! The pawn on f6 makes the position legal. This looks like it should be a simple forced mate, but Komodo doesn't see one.

so what is the score when komodo plays against itself. ?

I used Fritz 15 instead of Komodo to make use of the Monte Carlo option, with which I could get lots of variety very quickly. Somewhat to my surprise, White scored 84% in 700 games at ten ply; I expected more like 90%. Probably at any realistic time limit it would be close to 90%. I imagine many human GMs would say White had full compensation for a piece if we removed a White knight, at least at first glance. But apparently this is not so. When I removed the White knight from g1 from the above position, Black scored the same 84% that White scored with the knight! So we can say that the Ke4 handicap is "half a knight".

did you ever say each move is worth a third of a pawn ?

[lkaufman]

3. Komodo plays Black with his king starting on e4 and pawn on f6, White to move!! The pawn on f6 makes the position legal. This looks like it should be a simple forced mate, but Komodo doesn't see one.

so what is the score when komodo plays against itself. ?

I used Fritz 15 instead of Komodo to make use of the Monte Carlo option, with which I could get lots of variety very quickly. Somewhat to my surprise, White scored 84% in 700 games at ten ply; I expected more like 90%. Probably at any realistic time limit it would be close to 90%. I imagine many human GMs would say White had full compensation for a piece if we removed a White knight, at least at first glance. But apparently this is not so. When I removed the White knight from g1 from the above position, Black scored the same 84% that White scored with the knight! So we can say that the Ke4 handicap is "half a knight".

did you ever say each move is worth a third of a pawn ?

Yes, and since it takes four moves to bring the king home and White moves first, it's four and a half moves so 1 1/2 pawns. But Black has also lost casttling rights which brings it up to 2 pawns. But a knight is worth about 4 pawns in the opening, so it is indeed "half a knight".