deeds_chess_system

[deeds]

I added the previous 90 previous games to Stockfish_15_1.exp.

At the moment, Stockfish_15_1.exp is here :

Code: Select all

   # PLAYER                            :  RATING  ERROR  POINTS  PLAYED   (%)    W    D    L  D(%)  OppAvg  OppN
   1 Stockfish_15_1.exp (d.c.s. v2)    :       8     16   408.5     800  51.1   52  713   35  89.1       0     1
   2 Stockfish 15.1                    :       0   ----   391.5     800  48.9   35  713   52  89.1       8     1

White advantage = 36.53 +/- 8.83
Draw rate (equal opponents) = 50.00 % +/- 0.00

Games : +6 wins and -10 losses



Exp. data : +1.5M moves

[deeds]

I added the previous 90 previous games to Stockfish_15_1.exp.

At the moment, Stockfish_15_1.exp is here :

Code: Select all

   # PLAYER                            :  RATING  ERROR  POINTS  PLAYED   (%)    W    D    L  D(%)  OppAvg  OppN
   1 Stockfish_15_1.exp (d.c.s. v2)    :       8     16   408.5     800  51.1   52  713   35  89.1       0     1
   2 Stockfish 15.1                    :       0   ----   391.5     800  48.9   35  713   52  89.1       8     1

White advantage = 36.53 +/- 8.83
Draw rate (equal opponents) = 50.00 % +/- 0.00

Games : +6 wins and -10 losses



Exp. data : +1.5M moves

[deeds]

I added the previous 90 previous games to Stockfish_15_1.exp.

At the moment, Stockfish_15_1.exp is here :

Code: Select all

   # PLAYER                            :  RATING  ERROR  POINTS  PLAYED   (%)    W    D    L  D(%)  OppAvg  OppN
   1 Stockfish_15_1.exp (d.c.s. v2)    :       8     16   408.5     800  51.1   52  713   35  89.1       0     1
   2 Stockfish 15.1                    :       0   ----   391.5     800  48.9   35  713   52  89.1       8     1

White advantage = 36.53 +/- 8.83
Draw rate (equal opponents) = 50.00 % +/- 0.00

Games : +6 wins and -10 losses



Exp. data : +1.5M moves

[deeds]

What "d.c.s. v2" means ?

It is the new release of the private tool called "DeeDs Chess System".

It drives 3 engines in realtime, example :
- playing_engine = E:\Stockfish 15.1 x64 BMI2.exe
- experience_engine = E:\Eman 8.50 64-bit BMI2.exe
- book_engine = E:\asmFish 291118 x64 BMI2.exe

For the given position,
- the book_engine requests a polyglot bin book in order to get the pre-configured move.
- the experience_engine requests an experience file in order to get the experience data.
- if no move/exp. data was found, the playing_engine searchs for the bestmove.

[deeds]

hardware : dual xeon e5-2660v3
software : cutechess 1.2.0
settings : TC 60s+1s, 40 threads, hash 4096 MB, partial syzygy 7men

Opening list : GHPPN's 15 most popular openings

Code: Select all

   # PLAYER                :  RATING  ERROR  POINTS  PLAYED   (%)    W    D    L   D(%)  OppAvg  OppN
   1 deeds_chess_system    :       0   ----    15.0      30  50.0    0   30    0  100.0       0     1
   2 stockfish 15.1        :       0     92    15.0      30  50.0    0   30    0  100.0       0     1

White advantage = 0.00 +/- 46.89
Draw rate (equal opponents) = 50.00 % +/- 0.00

Code: Select all

   # PLAYER                :  RATING  ERROR  POINTS  PLAYED   (%)    W    D    L   D(%)  OppAvg  OppN
   1 deeds_chess_system    :       0   ----    15.0      30  50.0    0   30    0  100.0       0     1
   2 stockfish 16.1        :       0     92    15.0      30  50.0    0   30    0  100.0       0     1

White advantage = 0.00 +/- 46.89
Draw rate (equal opponents) = 50.00 % +/- 0.00

[deeds]

hardware : dual xeon e5-2660v3
software : cutechess 1.2.0
settings : TC 60s+1s, 40 threads, hash 4096 MB, partial syzygy 7men

Opening list : 20 first legal moves

Code: Select all

   # PLAYER                :  RATING  ERROR  POINTS  PLAYED   (%)    W    D    L  D(%)  OppAvg  OppN
   1 deeds_chess_system    :       0   ----    21.0      40  52.5    2   38    0  95.0     -18     1
   2 stockfish 15.1        :     -18     81    19.0      40  47.5    0   38    2  95.0       0     1

White advantage = -17.58 +/- 41.39
Draw rate (equal opponents) = 50.00 % +/- 0.00

Code: Select all

   # PLAYER                :  RATING  ERROR  POINTS  PLAYED   (%)    W    D    L  D(%)  OppAvg  OppN
   1 deeds_chess_system    :       0   ----    20.5      40  51.3    1   39    0  97.5      -9     1
   2 stockfish 16.1        :      -9     81    19.5      40  48.8    0   39    1  97.5       0     1

White advantage = -8.77 +/- 41.60
Draw rate (equal opponents) = 50.00 % +/- 0.00

I only reveal these games as a courtesy :

[Event "deeds_chess_system"]
[Site "dual xeon e5-2660v3"]
[Date "2024.08.25"]
[Round "1"]
[White "deeds_chess_system"]
[Black "stockfish 16.1"]
[Result "1/2-1/2"]
[ECO "A00"]
[GameDuration "00:03:58"]
[Opening "Grob"]
[PlyCount "146"]
[Termination "adjudication"]
[TimeControl "60+1"]
[Variation "Fritz gambit"]

1. g4 {book} d5 {-1.30/21 1.4s} 2. Bg2 {-1.75/245 0.002s} Bxg4 {-1.55/18 0.62s}
3. c4 {-1.49/245 0.001s} c6 {-1.42/24 1.5s} 4. Qb3 {-1.81/245 0.002s}
e6 {-1.35/25 1.8s} 5. Qxb7 {-1.63/245 0.002s} Nd7 {-1.38/20 0.79s}
6. Nc3 {-1.56/245 0.002s} Ne7 {-1.42/25 1.8s} 7. cxd5 {-1.45/245 0.003s}
exd5 {-1.45/22 0.58s} 8. d4 {-1.50/245 0.002s} Rb8 {-1.48/24 1.1s}
9. Qa6 {-1.70/245 0.002s} Rb6 {-1.50/20 0.83s} 10. Qd3 {-1.75/62 0.002s}
Ng6 {-1.47/24 1.5s} 11. h3 {-1.32/219 0.001s} Be6 {-1.34/22 1.1s}
12. Nf3 {-1.48/245 0.001s} h6 {-1.25/25 8.3s} 13. h4 {-2.01/40 0.001s}
Bd6 {-1.24/20 0.61s} 14. Bh3 {-2.29/40 0.001s} Bxh3 {-1.25/21 0.64s}
15. Rxh3 {-1.78/186 0.001s} O-O {-1.24/23 1.9s} 16. Rh1 {-2.03/40 0.001s}
Nf6 {-1.31/23 2.5s} 17. h5 {-1.25/27 0.001s} Ne7 {-1.32/21 0.40s}
18. b3 {-2.01/28 0.001s} Rb7 {-1.36/21 1.1s} 19. Bb2 {-1.14/35 0.001s}
Bb4 {-1.23/23 3.7s} 20. Nd2 {-1.21/26 0.001s} a5 {-1.18/22 1.8s}
21. O-O-O {-1.78/26 0.001s} Bxc3 {-1.18/21 1.1s} 22. Bxc3 {-1.12/28 0.001s}
a4 {-1.13/23 1.0s} 23. b4 {-1.91/29 0.001s} Nc8 {-1.03/26 9.5s}
24. Rdg1 {-1.80/30 24s} Nd6 {-1.08/23 2.1s} 25. Rh2 {-1.67/29 5.4s}
Qc8 {-0.94/23 4.4s} 26. Nf3 {-1.59/21 2.0s} Re8 {-1.07/22 1.1s}
27. Bd2 {-1.62/24 2.7s} Kh8 {-1.29/17 0.28s} 28. Rhg2 {-1.75/26 4.8s}
Nf5 {-1.10/23 1.2s} 29. a3 {-2.32/22 2.9s} Qe6 {-1.12/25 1.4s}
30. Ne5 {-2.11/29 16s} c5 {-1.18/21 0.41s} 31. bxc5 {-2.24/24 1.2s}
Nxh5 {-1.12/22 1.2s} 32. e3 {-2.14/26 1.9s} Rb3 {-1.14/21 0.63s}
33. Qe2 {-2.11/24 1.5s} Nf6 {-1.17/21 0.85s} 34. Bb4 {-2.23/25 2.1s}
g5 {-1.16/24 1.0s} 35. Rd1 {-2.20/26 2.9s} Nh4 {-1.12/21 1.3s}
36. Rh2 {-1.96/25 1.8s} Kg7 {-1.03/22 1.3s} 37. Rd3 {-1.87/25 1.9s}
Rxd3 {-1.08/21 0.47s} 38. Qxd3 {-2.05/25 3.6s} Ng4 {-1.07/23 0.65s}
39. Nxg4 {-1.79/24 1.6s} Qxg4 {-1.14/26 1.4s} 40. Qd1 {-1.52/26 2.2s}
Nf3 {-0.94/21 1.5s} 41. Rh1 {-1.53/25 2.7s} Qg2 {-0.84/30 6.3s}
42. Qf1 {-1.51/22 1.9s} Qxf1+ {-0.72/25 1.2s} 43. Rxf1 {-1.17/26 1.7s}
h5 {-0.70/23 0.94s} 44. Ba5 {-1.23/24 1.6s} Rc8 {-0.76/23 0.52s}
45. Kd1 {-1.20/24 1.5s} g4 {-0.69/27 7.8s} 46. Ke2 {-0.97/23 1.3s}
f6 {-0.79/19 0.21s} 47. Rc1 {-1.15/29 6.7s} Rc6 {-0.82/21 0.46s}
48. Kf1 {-1.09/26 1.2s} Kf7 {-0.85/22 0.36s} 49. Rb1 {-0.69/27 5.4s}
h4 {-0.58/27 2.2s} 50. Kg2 {-0.62/24 1.5s} Rc8 {-0.54/23 0.73s}
51. Rc1 {-0.58/25 2.0s} Rc6 {-0.28/26 6.4s} 52. Rb1 {-0.68/22 0.69s}
Rc8 {-0.33/24 0.99s} 53. Rc1 {-0.54/25 0.52s} f5 {-0.20/27 2.0s}
54. c6 {-0.43/27 0.76s} h3+ {-0.34/26 0.58s} 55. Kg3 {-0.44/29 1.4s}
Ng5 {-0.13/25 0.98s} 56. Kf4 {-0.38/27 0.69s} Ne4 {-0.10/26 0.45s}
57. f3 {-0.37/30 0.77s} gxf3 {-0.10/25 0.40s} 58. Kxf3 {-0.32/27 0.62s}
Ke7 {-0.08/26 0.45s} 59. Bb4+ {-0.35/26 0.74s} Ke6 {-0.08/28 0.44s}
60. Be1 {-0.25/28 0.74s} Rh8 {-0.06/26 0.37s} 61. c7 {-0.30/30 0.86s}
Kd7 {-0.07/26 0.34s} 62. Bg3 {-0.25/31 1.2s} Nd2+ {-0.08/26 0.32s}
63. Kf4 {-0.23/31 1.3s} Nc4 {-0.08/26 0.32s} 64. Kxf5 {-0.17/29 0.79s}
h2 {-0.09/25 0.36s} 65. Rh1 {-0.14/28 0.71s} Nxe3+ {-0.08/27 0.34s}
66. Kf6 {-0.10/29 0.84s} Rf8+ {-0.08/29 0.73s} 67. Kg5 {-0.14/35 1.0s}
Rg8+ {-0.08/27 0.38s} 68. Kf4 {-0.10/30 0.85s} Kxc7 {-0.06/35 2.0s}
69. Rxh2 {-0.09/32 1.6s} Nc4 {-0.05/29 0.36s} 70. Rh7+ {0.00/31 2.3s}
Kc6 {-0.04/30 1.5s} 71. Ra7 {0.00/31 0.72s} Kb5 {-0.04/37 1.6s}
72. Rb7+ {0.00/30 0.64s} Kc6 {-0.03/29 0.25s} 73. Ra7 {0.00/36 0.96s}
Rf8+ {-0.03/32 0.55s, Draw by adjudication} 1/2-1/2

[Event "deeds_chess_system"]
[Site "dual xeon e5-2660v3"]
[Date "2024.08.25"]
[Round "1"]
[White "stockfish 16.1"]
[Black "deeds_chess_system"]
[Result "0-1"]
[ECO "A00"]
[GameDuration "00:03:56"]
[Opening "Grob's attack"]
[PlyCount "142"]
[TimeControl "60+1"]

1. g4 {book} d5 {-1.68/245 0.002s} 2. e3 {-1.25/23 4.5s} e5 {-1.75/245 0.002s}
3. d4 {-1.22/19 0.32s} Nc6 {-1.80/245 0.002s} 4. Nc3 {-1.20/19 0.50s}
Be6 {-1.58/245 0.002s} 5. dxe5 {-1.28/22 1.8s} Nxe5 {-1.79/245 0.002s}
6. h3 {-1.30/25 2.4s} h5 {-1.63/245 0.002s} 7. Nf3 {-1.25/23 1.9s}
Nxf3+ {-2.00/245 0.002s} 8. Qxf3 {-1.27/22 0.57s} hxg4 {-2.03/245 0.001s}
9. hxg4 {-1.28/22 0.59s} Bxg4 {-1.67/245 0.001s} 10. Qg2 {-1.25/21 0.67s}
Rxh1 {-2.03/245 0.001s} 11. Qxh1 {-1.30/25 3.0s} Nf6 {-1.94/245 0.001s}
12. Bd2 {-1.42/24 2.0s} Qd6 {-2.00/245 0.001s} 13. Qg2 {-1.42/23 2.9s}
O-O-O {-1.80/245 0.001s} 14. f3 {-1.43/22 2.6s} Bd7 {-2.07/245 0.001s}
15. O-O-O {-1.48/19 0.58s} g6 {-2.72/40 0.001s} 16. Qf2 {-1.42/23 4.2s}
Qb6 {-3.02/40 0.001s} 17. Kb1 {-1.47/19 0.46s} Re8 {-3.10/40 0.001s}
18. Bc1 {-1.45/23 3.0s} Bc5 {-3.01/36 5.7s} 19. Rd3 {-1.47/19 0.52s}
Qd6 {-3.13/30 1.8s} 20. Qh4 {-1.53/26 2.0s} Qe7 {-3.08/29 3.6s}
21. Bh3 {-1.53/23 0.48s} Bxh3 {-3.01/28 3.5s} 22. Qxh3+ {-1.50/22 0.59s}
Qe6 {-2.94/30 6.1s} 23. Qh1 {-1.49/25 1.4s} a6 {-3.00/30 6.5s}
24. Na4 {-1.51/24 2.3s} Ba7 {-3.03/26 1.6s} 25. Nc3 {-1.53/23 1.6s}
c6 {-3.16/31 6.4s} 26. a3 {-1.57/25 2.7s} Qf5 {-3.15/29 4.0s}
27. Qh6 {-1.55/25 0.57s} Bc5 {-3.11/28 1.6s} 28. Na4 {-1.53/25 4.0s}
Be7 {-3.15/24 1.8s} 29. Qh1 {-1.50/21 1.3s} Bd8 {-3.03/25 2.1s}
30. b3 {-1.53/20 0.45s} Qh5 {-3.14/22 1.9s} 31. Qd1 {-1.53/22 1.1s}
Bc7 {-3.24/22 1.6s} 32. Rd4 {-1.79/23 4.3s} g5 {-3.68/25 7.1s}
33. Rd3 {-1.77/28 7.8s} Re5 {-4.12/21 1.6s} 34. Nc5 {-1.95/22 1.2s}
Rf5 {-4.49/24 2.7s} 35. Qe1 {-2.25/28 5.8s} Bd6 {-4.67/22 2.0s}
36. Qb4 {-2.28/25 3.4s} Bxc5 {-4.78/23 2.0s} 37. Qxc5 {-2.34/20 0.53s}
Nd7 {-5.02/21 1.7s} 38. Qc3 {-2.37/21 0.87s} Qxf3 {-5.02/21 1.9s}
39. Qh8+ {-2.30/18 0.55s} Kc7 {-5.09/22 2.6s} 40. Bd2 {-2.56/30 5.9s}
Qe4 {-5.36/21 2.1s} 41. Qa8 {-2.55/19 0.27s} Nb6 {-5.66/22 2.0s}
42. Qg8 {-2.71/24 3.8s} Nc8 {-5.97/23 2.9s} 43. Bc3 {-2.75/21 2.2s}
c5 {-6.23/22 2.7s} 44. Qf8 {-2.81/20 1.1s} b6 {-6.50/21 2.7s}
45. a4 {-2.71/17 0.32s} a5 {-7.25/20 2.8s} 46. Kb2 {-3.20/24 3.5s}
Nd6 {-7.60/20 3.4s} 47. Qa8 {-3.24/18 0.62s} Rf2 {-8.07/20 1.8s}
48. Be1 {-3.58/20 1.1s} Rg2 {-8.37/22 1.7s} 49. Qa6 {-3.85/22 2.9s}
Nb7 {-9.24/22 2.0s} 50. Rd2 {-4.04/17 0.85s} Rxd2 {-9.66/20 2.0s}
51. Bxd2 {-4.24/19 0.54s} g4 {-10.27/21 3.0s} 52. Qf1 {-4.64/22 2.3s}
Nd6 {-10.32/22 1.9s} 53. Qg1 {-4.68/20 2.0s} Nf5 {-10.95/20 1.6s}
54. Be1 {-4.65/20 1.5s} Nxe3 {-11.26/22 3.5s} 55. Qh2+ {-4.58/15 0.12s}
Kd7 {-11.72/20 1.1s} 56. Bg3 {-5.04/23 0.48s} Nf1 {-11.87/24 1.4s}
57. Qh5 {-5.07/20 0.15s} Nxg3 {-13.39/18 1.4s} 58. Qxf7+ {-5.14/23 0.34s}
Kd6 {-14.27/18 1.4s} 59. Ka2 {-6.38/21 3.0s} Qxc2+ {-M41/34 1.1s}
60. Ka3 {-199.34/18 0.35s} Qc1+ {-M29/47 1.1s} 61. Ka2 {-M28/40 0.40s}
Qd2+ {-M27/50 1.1s} 62. Ka3 {-M26/39 0.23s} Ne4 {-M25/52 1.3s}
63. Qf8+ {-M24/43 0.16s} Ke5 {-M23/55 4.9s} 64. Qh8+ {-M22/40 0.18s}
Kf4 {-M21/54 0.57s} 65. Qf8+ {-M18/42 0.20s} Kg3 {-M13/88 0.46s}
66. Qf1 {-M12/57 0.19s} b5 {-M11/105 0.42s} 67. axb5 {-M10/126 0.21s}
Nc3 {-M9/155 0.45s} 68. Qg1+ {-M8/245 0.031s} Kf4 {-M7/245 0.041s}
69. Qf1+ {-M6/245 0.020s} Kg5 {-M5/245 0.033s} 70. Qa1 {-M4/245 0.021s}
Nxb5+ {-M3/245 0.044s} 71. Ka4 {-M2/245 0.006s}
Qb4# {-M1/245 0.043s, Black mates} 0-1

[deeds]

hardware : dual xeon e5-2660v3
software : cutechess 1.2.0
settings : TC 60s+1s, 40 threads, hash 4096 MB, partial syzygy 7men

Opening list : Albert Silver's 25 gambit suit

Code: Select all

   # PLAYER                :  RATING  ERROR  POINTS  PLAYED   (%)    W    D    L  D(%)  OppAvg  OppN
   1 deeds_chess_system    :       0   ----    27.5      50  55.0   11   33    6  66.0     -35     1
   2 stockfish 15.1        :     -35     69    22.5      50  45.0    6   33   11  66.0       0     1

White advantage = 7.08 +/- 35.50
Draw rate (equal opponents) = 50.00 % +/- 0.00

Code: Select all

   # PLAYER                :  RATING  ERROR  POINTS  PLAYED   (%)    W    D    L  D(%)  OppAvg  OppN
   1 deeds_chess_system    :       0   ----    27.0      50  54.0    9   36    5  72.0     -28     1
   2 stockfish 16.1        :     -28     69    23.0      50  46.0    5   36    9  72.0       0     1

White advantage = 0.00 +/- 35.52
Draw rate (equal opponents) = 50.00 % +/- 0.00

[deeds]

hardware : dual xeon e5-2660v3
software : cutechess 1.2.0
settings : TC 60s+1s, 40 threads, hash 4096 MB, partial syzygy 7men

Opening list : Jeroen Noomen's 50 tcec season 22 superfinal openings

Code: Select all

   # PLAYER                :  RATING  ERROR  POINTS  PLAYED   (%)    W    D    L  D(%)  OppAvg  OppN
   1 deeds_chess_system    :       0   ----    71.0     100  71.0   43   56    1  56.0    -214     1
   2 stockfish 15.1        :    -214     75    29.0     100  29.0    1   56   43  56.0       0     1

White advantage = 214.31 +/- 38.27
Draw rate (equal opponents) = 50.00 % +/- 0.00

Code: Select all

   # PLAYER                :  RATING  ERROR  POINTS  PLAYED   (%)    W    D    L  D(%)  OppAvg  OppN
   1 deeds_chess_system    :       0   ----    70.5     100  70.5   41   59    0  59.0    -185     1
   2 stockfish 16.1        :    -185     68    29.5     100  29.5    0   59   41  59.0       0     1

White advantage = 164.34 +/- 34.73
Draw rate (equal opponents) = 50.00 % +/- 0.00

[deeds]

As expected, Stockfish releases are getting harder and harder...

Previously, against Stockfish 15.1, d.c.s needed only Eman 8.40 training data, at worst it needed to reinforce (at D40) only 2-3 moves from key positions for the given openings.

Now, against Stockfish 16.1, when Eman 8.40 training data is not enough (I mean: when playing 2000 games/opening is not enough to really learn the opening = when the average first novelty is less than 50%), it is necessary to reinforce all moves from key positions for the given openings.

[deeds]

Realtime learning features added to a non-learning engine without updating/modifying its source code :

And any playing_engine (ex : Stockfish, Dragon, Houdini, Torch, etc.) can use any experience file thanks to the experience_engine (ex : Eman, HypnoS, BrainLearn, etc.)