Testing K11.2.2

Discussion of computer chess matches and engine tournaments.

Moderators: hgm, Rebel, chrisw

mjlef
Posts: 1494
Joined: Thu Mar 30, 2006 2:08 pm

Re: Testing K11.2.2

Post by mjlef »

Nordlandia wrote:Galaxy S8+
5-men Syzygy
512 MB hash per engine
Komodo TM: 128 MB
Komodo Contempt: 0
Stockfish: default settings

[pgn][Event "Chess for Android Game"]
[Site "Norge"]
[Date "2017.08.05"]
[Round "?"]
[White "Komodo 11.2.2"]
[Black "Stockfish 02.08.17 Arm8"]
[Result "*"]
[PlyCount "168"]
[TimeControl "1800+30"]
[Device "SM-G955F"]

1. e4 c5
2. Nf3 d6
3. d4 cxd4
4. Nxd4 Nf6
5. Nc3 a6
6. h4 e5
7. Nb3 Be7
8. Be3 Be6
9. f3 O-O
10. Qd2 Nbd7
11. g4 b5
12. O-O-O Nb6
13. Qf2 Nfd7
14. Kb1 Qc7
15. Nd5 Bxd5
16. exd5 Nc4
17. Bd3 a5
18. Nd2 a4
19. Ne4 Nxe3
20. Qxe3 b4
21. Bb5 a3
22. Bxd7 Qxd7
23. b3 f5
24. gxf5 Qxf5
25. Rd3 Qh5
26. Ng3 Qg6
27. Qe1 Rab8
28. h5 Qe8
29. Qg1 Qd7
30. h6 g6
31. Nh5 Bf6
32. Ng3 Rf7
33. Qe1 Be7
34. Ne4 Rf4
35. Qe2 Qb7
36. Rc1 Bf8
37. Qg2 Qd7
38. Rh1 Be7
39. Rc1 Qf5
40. Qf2 Rb7
41. Qf1 Rh4
42. c4 Rxh6
43. c5 dxc5
44. d6 Bd8
45. Qf2 Rh3
46. Rf1 Kg7
47. Qxc5 Rh2
48. Rfd1 h5
49. Qc4 h4
50. Nc5 Rb8
51. Ne6+ Kh6
52. Nxd8 Rxd8
53. Qxb4 g5
54. Qxa3 Rh1
55. Kb2 e4
56. Rxh1 exd3
57. Qb4 Kh5
58. Qd4 Qxf3
59. Re1 Rxd6
60. Qxd6 Qf2+
61. Ka3 Qxe1
62. Qxd3 Qa5+
63. Kb2 Qe5+
64. Ka3 g4
65. Qh7+ Kg5
66. Qg8+ Kf4
67. Qc4+ Kg3
68. Qd3+ Kf2
69. Qd2+ Qe2
70. Qd4+ Kg2
71. Qa7 g3
72. Qb7+ Qf3
73. Qb4 h3
74. Qd2+ Qf2
75. Qd5+ Kh2
76. Qe5 Kh1
77. Qa1+ Qg1
78. Qc3 Qa7+
79. Kb2 Qf2+
80. Ka3 Qf4
81. b4 Kh2
82. Qc2+ g2
83. b5 Qf3+
84. Kb2 Kg3 *[/pgn]

Game 2:

Komodo sacrifice a pawn for obtaining pair of bishops. Komodo's king activity is proving ample compensation for the pawn. However Komodo is not able to break through.

Question for Mark: Is King walk heuristics adjusted in Komodo recently? My impression is that Komodo is more keen in letting the king taking a stroll, over the board. Maybe Komodo is inspired by Stenitz.

[pgn][Event "Chess for Android Game"]
[Site "Norge"]
[Date "2017.08.05"]
[Round "?"]
[White "Stockfish 02.08.17"]
[Black "Komodo 11.2.2"]
[Result "1/2-1/2"]
[PlyCount "153"]
[EventDate "2017.??.??"]

1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. h4 Nc6 7. Bg5 e6 8. f3
Be7 9. Qd2 Nxd4 10. Qxd4 O-O 11. O-O-O b5 12. Bxf6 Bxf6 13. Qxd6 Qxd6 14. Rxd6
Bb7 15. Kb1 Rfb8 16. h5 h6 17. a3 Be5 18. Rd2 Bc6 19. Na2 a5 20. c3 Kf8 21. Bd3
Ke7 22. Bc2 Be8 23. Rdd1 Rb6 24. Bb3 Rbb8 25. Nc1 Rc8 26. g4 Rd8 27. Rxd8 Rxd8
28. Kc2 Bc7 29. Nd3 Kf6 30. Kd2 Kg5 31. Rd1 a4 32. Ba2 Kh4 33. Ke2 Kg3 34. Rg1+
Kh2 35. Rf1 Bb6 36. Bb1 Bc7 37. Ne1 Bf4 38. Rf2+ Kg3 39. Nd3 Bc7 40. Rf1 Bb6
41. Nb4 Kf4 42. Rh1 Kg3 43. Bd3 Bc5 44. Nc2 Bb6 45. Rc1 Kf4 46. Nb4 Bc5 47. Nc2
Kg3 48. Rd1 Kf4 49. Rb1 Kg3 50. Rh1 Bb6 51. Nb4 Kf4 52. Bb1 Kg3 53. Rf1 Kf4 54.
Bd3 Bc5 55. Ra1 Bxb4 56. cxb4 Rc8 57. Rd1 e5 58. Bb1 f6 59. Kf2 Bf7 60. Bd3 Be8
61. Rf1 Rd8 62. Rd1 Rc8 63. Ke2 Bd7 64. Ke1 Be8 65. Kf2 Rc7 66. Bb1 Bf7 67. Bd3
Be8 68. Re1 Rc8 69. Ke2 Bd7 70. Rg1 Be6 71. Bxb5 Rc2+ 72. Kd3 Bb3 73. Rf1 Rc7
74. Kd2 Rc2+ 75. Kd3 Rc7 76. Kd2 Rc2+ 77. Kd3 1/2-1/2



[/pgn]

https://lichess.org/1pCPCCDF


https://lichess.org/hfaQL95E
We worked on King Safety lately and some is in K 11.2.2, but thinking much involving king "trolling". In a development version we did another change for king eval which is more likely to encourage the king to go toward the opponent's side of the board, but it is a pretty small change. Komodo in the past has been known to take king strolls, like this one in a past TCEC: https://www.youtube.com/watch?v=HoUDv08wXRI
leavenfish
Posts: 282
Joined: Mon Sep 02, 2013 8:23 am

Re: Testing K11.2.2

Post by leavenfish »

mjlef wrote: Komodo in the past has been known to take king strolls, like this one in a past TCEC: https://www.youtube.com/watch?v=HoUDv08wXRI


When the opponents Queen has 'self stalemated' itself, I guess the King can wonder all he wants! :)
User avatar
Nordlandia
Posts: 2821
Joined: Fri Sep 25, 2015 9:38 pm
Location: Sortland, Norway

Re: Testing K11.2.2

Post by Nordlandia »

2x i5-5200U | 2-Core | Ponder | Probing 5-men | Fritz 15 GUI | 4096 Mb Hash | 120m+30s |

[pgn][Event "Lenova-Laptop #2, Slow 120m+30s"]
[Site "Lenova-Laptop #2"]
[Date "2017.08.22"]
[Round "1"]
[White "Stockfish 200817 64 BMI2"]
[Black "Komodo 11.2.2 64-bit"]
[Result "0-1"]
[Annotator "-0.43;-0.91"]
[SetUp "1"]
[FEN "1n2kbnr/1p1b1pp1/1P2p3/1p1pP2p/3p1Q2/5N2/1PPBKPPP/r4B1R b k - 0 1"]
[PlyCount "59"]
[TimeControl "7200+30"]

1... Nc6 {-0.91/31 226 Both last book move} 2. Nxd4 {-0.43/33 291} Ra4 {
-0.82/32 0} 3. b4 {-0.32/35 221 (c3)} Nxb4 {-0.94/30 137} 4. Qg3 {-0.30/35 0
(Qf3)} Na6 {-0.88/29 145} 5. Nb3 {-0.38/35 117 (Qd3)} Ne7 {-1.11/29 136} 6. Qc3
{-0.41/34 89 (Kd1)} Ng6 {-1.21/30 194 (Nf5)} 7. Kd1 {-0.58/33 161} Be7 {
-1.22/31 0} 8. g3 {-0.60/32 38} O-O {-1.38/31 165} 9. f4 {-0.79/36 124} Rc8 {
-1.52/32 372} 10. Qb2 {-1.02/38 34 (Qf3)} Nb4 {-1.51/30 99 (Ba3)} 11. Bd3 {
-1.33/34 321 (c3)} Nxd3 {-2.23/31 145} 12. cxd3 {-1.49/40 126} Ba3 {-2.33/33 78
} 13. Qb1 {-1.50/35 34} Ne7 {-2.41/33 131} 14. Na5 {-2.35/38 624} Nf5 {-2.40/
37 0 (Bc6)} 15. Nxb7 {-1.84/34 98} Rb8 {-2.85/35 292} 16. Na5 {-2.02/40 1} Bc5
{-3.08/32 111} 17. b7 {-2.33/40 191} Bb6 {-3.17/34 0} 18. Nb3 {-2.42/36 44}
Rxb7 {-3.16/32 121} 19. Ke2 {-2.49/38 201 (Bc3)} Ra3 {-3.39/33 132 (b4)} 20.
Kf1 {-3.59/38 596} b4 {-3.60/36 0} 21. Bc1 {-3.77/34 40} Ra8 {-3.96/32 120} 22.
Bd2 {-3.97/40 450 (Kg2)} Ba4 {-4.37/32 203 (Nd4)} 23. Kg2 {-4.37/35 300} Nd4 {
-4.77/33 0} 24. Na1 {-4.62/31 29 (Nc1)} b3 {-5.01/31 126} 25. Bb4 {-5.37/38
634 (Qb2)} Nc2 {-5.48/33 259 (Nf5)} 26. Bd2 {-5.34/34 150} Ne3+ {-6.10/32 16
(Nd4)} 27. Kh3 {-6.23/38 377} Nf5 {-6.47/36 0} 28. Nxb3 {-6.60/39 306} Bd8 {
-6.92/35 4 (Bd4)} 29. Qa2 {-7.01/39 403} Rxb3 {-7.19/37 0} 30. Ra1 {-7.62/40
271} Rbb8 {-7.59/37 113} 0-1[/pgn]

>------------------------------------------------------<

Lichess analysis - https://lichess.org/Ga8slUwU

Chessbase viewer http://view.chessbase.com/cbreader/2017 ... 32640.html