Rebel 16.2 release

Add a history to your list, see my old gambit rating list - http://rebel13.nl/grl-best-40-2.html

CCRL has such an option too.

Dann has a point.

Crafty is still struggling for its first draw against Rebel 16.2. In the following game it came pretty close.
Crafty chose a questionable Rauser line, and Rebel 16.2 saw a chance to immediately liquidate into what it thought to be a winning endgame. Once the endgame was reached, it turned out to be a draw.
Rebel kept trying though, and Crafty blundered the game eventually, though it is not completely obvious at which exact move.
If you like endgames, this is a game to enjoy IMHO.
The game made me think about the „Rep2_Draw_Trolling“ feature some more. It has a human touch anyway (it is a technique endgame specialists have suggested for ages), but it seems to be unique in the engine world. It felt very effective in  this game, always gaining some free minutes on the clock.
I am curious how many ELO points it is worth. Has someone tested this?

[Event "Lang 120min+10sek"]
[Site "Berlin"]
[Date "2023.07.11"]
[Round "?"]
[White "Rebel-16.2"]
[Black "Crafty 25.6"]
[Result "1-0"]
[ECO "B66"]
[PlyCount "153"]
[TimeControl "7200+10"]

{4096MB, LAPTOP-NCDN8BTK} 1. e4 {[%eval 1,0] [%emt 0:00:00]} c5 {[%emt 0:00:06]
} 2. Nf3 {[%eval 1,0] [%emt 0:00:00]} Nc6 {[%emt 0:00:05]} 3. d4 {[%eval 1,0]
[%emt 0:00:00]} cxd4 {[%emt 0:00:05]} 4. Nxd4 {[%eval 65,35] [%emt 0:05:05]}
Nf6 {[%emt 0:00:05]} 5. Nc3 {[%eval 1,0] [%emt 0:00:00]} d6 {[%emt 0:00:05]} 6.
Bg5 {[%eval 1,0] [%emt 0:00:00]} e6 {[%emt 0:00:06]} 7. Qd2 {[%eval 1,0] [%emt
0:00:00]} a6 {[%emt 0:00:11]} 8. O-O-O {[%eval 1,0] [%emt 0:00:01]} h6 {
[%emt 0:00:07]} 9. Be3 {[%eval 70,34] [%emt 0:04:12]} Be7 {[%emt 0:00:07]} 10.
f3 {[%eval 57,33] [%emt 0:03:58]} Nxd4 {[%emt 0:00:06]} 11. Bxd4 {[%eval 71,35]
[%emt 0:04:41]} e5 {[%emt 0:00:07] (b5)} 12. Be3 {[%eval 149,37] [%emt 0:03:20]
} Be6 {[%emt 0:00:07]} 13. g4 {[%eval 91,43] [%emt 0:06:45]} Rc8 {[%emt 0:00:
06] (b5)} 14. Kb1 {[%eval 185,33] [%emt 0:02:13]} b5 {[%emt 0:03:49] (Da5)} 15.
Nd5 {[%eval 189,40] [%emt 0:05:48]} Nxd5 {[%emt 0:04:02] (Lxd5)} 16. exd5 {
[%eval 211,39] [%emt 0:02:12]} Bd7 {[%emt 0:00:34]} 17. h4 {[%eval 198,39]
[%emt 0:04:27]} Bxh4 {[%emt 0:02:59]} 18. g5 {[%eval 214,41] [%emt 0:07:08]}
hxg5 {[%emt 0:00:14]} 19. Bxg5 {[%eval 213,41] [%emt 0:03:20]} Qxg5 {[%emt 0:
00:08]} 20. Qxg5 {[%eval 238,41] [%emt 0:03:07]} Bxg5 {[%emt 0:00:08]} 21.
Rxh8+ {[%eval 226,37] [%emt 0:01:50]} Ke7 {[%emt 0:00:07]} 22. Rxc8 {[%eval
203,41] [%emt 0:04:23]} Bxc8 {[%emt 0:00:07]} 23. c4 {[%eval 228,41] [%emt 0:
05:04]} bxc4 {[%emt 0:00:06] (b4)} 24. Bxc4 {[%eval 203,40] [%emt 0:01:51]} a5
{[%emt 0:00:51] (Le3)} 25. Kc2 {[%eval 203,41] [%emt 0:02:24]} Be3 {[%emt 0:00:
14]} 26. Rh1 {[%eval 217,41] [%emt 0:01:41]} g5 {[%emt 0:00:58] (Ld7)} 27. Rh8
{[%eval 291,34] [%emt 0:01:08]} Bd7 {[%emt 0:01:29]} 28. Be2 {[%eval 280,37]
[%emt 0:01:49]} Be8 {[%emt 0:01:15] (Lf5+)} 29. Kd3 {[%eval 239,40] [%emt 0:03:
08]} Bd4 {[%emt 0:02:45]} 30. b3 {[%eval 234,42] [%emt 0:01:09]} Bb5+ {[%emt 0:
01:20]} 31. Kd2 {[%eval 238,42] [%emt 0:01:35]} Be8 {[%emt 0:00:51]} 32. Kd3 {
[%eval 0,40] [%emt 0:00:03]} Bb5+ {[%emt 0:02:50]} 33. Kd2 {[%eval 0,4] [%emt
0:00:01]} Be8 {[%emt 0:02:54]} 34. Rh5 {[%eval 184,38] [%emt 0:01:05]} Kf6 {
[%emt 0:01:15]} 35. Rh6+ {[%eval 182,42] [%emt 0:01:24]} Ke7 {[%emt 0:00:49]}
36. Rh5 {[%eval 0,43] [%emt 0:00:25]} Kf6 {[%emt 0:02:48]} 37. Rh6+ {[%eval 0,
4] [%emt 0:00:01]} Ke7 {[%emt 0:02:39]} 38. Rh7 {[%eval 185,45] [%emt 0:03:01]}
Kf6 {[%emt 0:00:45]} 39. Rh5 {[%eval 182,46] [%emt 0:01:22]} Bd7 {[%emt 0:01:
10]} 40. Rh8 {[%eval 185,45] [%emt 0:02:04]} Bf5 {[%emt 0:00:37] (Lc5)} 41. Bd3
{[%eval 260,41] [%emt 0:00:42]} Bd7 {[%emt 0:09:34]} 42. Be2 {[%eval 0,48]
[%emt 0:00:05]} Bf5 {[%emt 0:02:00]} 43. Bd3 {[%eval 0,4] [%emt 0:00:01]} Bd7 {
[%emt 0:01:50]} 44. Ra8 {[%eval 258,43] [%emt 0:01:12]} Bc5 {[%emt 0:00:23]}
45. Kc3 {[%eval 207,47] [%emt 0:04:03]} Bh3 {[%emt 0:01:07]} 46. Rxa5 {[%eval
413,33] [%emt 0:00:53]} Bg2 {[%emt 0:01:36]} 47. Be2 {[%eval 410,36] [%emt 0:
01:05]} e4 {[%emt 0:00:32]} 48. fxe4 {[%eval 385,36] [%emt 0:00:48]} Bxe4 {
[%emt 0:00:45]} 49. b4 {[%eval 431,35] [%emt 0:01:10]} Bf2 {[%emt 0:00:19]} 50.
Ra8 {[%eval 431,36] [%emt 0:00:50]} Ke5 {[%emt 0:00:38] (Lxd5)} 51. Re8+ {
[%eval 504,31] [%emt 0:00:34]} Kf4 {[%emt 0:01:28]} 52. Kc4 {[%eval 593,34]
[%emt 0:00:38]} Bg2 {[%emt 0:04:48]} 53. Rf8 {[%eval 615,41] [%emt 0:02:25]}
Ke5 {[%emt 0:01:56] (f5)} 54. a4 {[%eval 640,31] [%emt 0:00:37]} Bxd5+ {
[%emt 0:00:46]} 55. Kb5 {[%eval 696,32] [%emt 0:01:00]} f5 {[%emt 0:08:31]} 56.
Ka6 {[%eval 726,41] [%emt 0:02:21]} g4 {[%emt 0:01:06]} 57. Bxg4 {[%eval 748,
36] [%emt 0:00:49]} Be6 {[%emt 0:07:55] (fxg4)} 58. Be2 {[%eval 1496,26] [%emt
0:00:44]} Bd5 {[%emt 0:03:11] (Ld7)} 59. b5 {[%eval 1671,28] [%emt 0:00:45]}
Be3 {[%emt 0:04:15] (Lg3)} 60. b6 {[%eval 2024,29] [%emt 0:01:19]} Bh6 {
[%emt 0:02:41] (Kd4)} 61. Re8+ {[%eval 2232,27] [%emt 0:00:28]} Kf6 {[%emt 0:
02:52]} 62. a5 {[%eval 2863,31] [%emt 0:01:29]} Kf7 {[%emt 0:00:23] (Ld2)} 63.
Bh5+ {[%eval 3121,23] [%emt 0:00:18]} Kf6 {[%emt 0:00:12]} 64. b7 {[%eval 3801,
25] [%emt 0:00:26]} Bxb7+ {[%emt 0:01:04] (Le3)} 65. Kxb7 {[%eval 4308,24]
[%emt 0:00:19]} Kg5 {[%emt 0:00:17]} 66. Be2 {[%eval 32736,37] [%emt 0:00:18]}
Bg7 {[%emt 0:00:30] (Kh4)} 67. Rg8 {[%eval 32738,56] [%emt 0:00:20]} Kh6 {
[%emt 0:00:24]} 68. Rxg7 {[%eval 32742,59] [%emt 0:00:44]} Kxg7 {[%emt 0:00:32]
} 69. a6 {[%eval 32744,62] [%emt 0:00:23]} Kf6 {[%emt 0:00:08]} 70. a7 {
[%eval 32746,64] [%emt 0:00:24]} Ke6 {[%emt 0:00:07] (d5)} 71. Kc6 {[%eval
32750,59] [%emt 0:00:17]} d5 {[%emt 0:00:07] (Ke5)} 72. a8=Q {[%eval 32752,60]
[%emt 0:00:20]} Ke5 {[%emt 0:00:08]} 73. Qa3 {[%eval 32754,64] [%emt 0:00:19]}
Kf6 {[%emt 0:00:22] (d4)} 74. Kxd5 {[%eval 32756,60] [%emt 0:00:17]} Kg5 {
[%emt 0:00:07]} 75. Qd6 {[%eval 32758,86] [%emt 0:00:15]} Kh4 {[%emt 0:00:06]
(f4)} 76. Qf4+ {[%eval 32764,116] [%emt 0:00:01]} Kh3 {[%emt 0:00:06]} 77. Bf1#
{[%eval 32766,116] [%emt 0:00:01]} 1-0

Peter you are downgrading Ed's level of programming by comparing his engine to Crafty's level.

Damir Desevac wrote:

Peter you are downgrading Ed's level of programming by comparing his engine to Crafty's level.

I am surprised you think this way and feel the need to tell it to me more than once. I can look at rating lists, too - Rebel 16.2 is about 100 points behind latest Stockfish. and so it is about 400 points stronger than Crafty.

Crafty has been playing in the very same setup against Stockfish and LC0, and it has been able to draw some games against them, some even very easily. I think it is pretty amazing that it doesn't seem to be able to do the same against Rebel.

I also thought that the game was pretty interesting itself. Anyway, you can relax. I have decided that it is time to find a new opponent for the strong engines, so this will be my last post about Rebel's games against Crafty.

Rebel creates chances by taking chances.. SF does not!
Hence SF is liable to draw more...

I think it's more that sf and Leela are perfectionist and also sf is trained on a lot of Leela data. Rebel is a worse engine trained on more imbalanced data. This results in the engine playing for more of a win vs a weaker engine. Similarly with Dragon which, likely trained on some Leela data, has to have some heavily imbalanced DFRC data because it excels at beating weaker engines, much better than SF

Those are perfectionist engines playing super solid chess, almost not to lose, making waste of their enormous strength when facing weaker opposition.

Nezhman wrote:

Those are perfectionist engines playing super solid chess, almost not to lose, making waste of their enormous strength when facing weaker opposition.

exactly. dragon has built in uci option to specify an opponent. which is probably a small list of top engines. but they others have no such thing, since they don't have any idea of when or how to apply risk-taking because we don't know how to do it without losing strength overall.

texium wrote:

Nezhman wrote:

Those are perfectionist engines playing super solid chess, almost not to lose, making waste of their enormous strength when facing weaker opposition.

exactly. dragon has built in uci option to specify an opponent. which is probably a small list of top engines. but they others have no such thing, since they don't have any idea of when or how to apply risk-taking because we don't know how to do it without losing strength overall.

Crafty had opponent recognition. Dr. Hyatt wanted to know when he was playing a strong engine or a GM. it was just a list to search, and he would use the xboard api to decipher it.

Peter Berger wrote:

. I have decided that it is time to find a new opponent for the strong engines, so this will be my last post about Rebel's games against Crafty.  

I lied - and secretly continued doing these games . It took some time:

[Event "Lang 120min+10sek"]
[Site "Berlin"]
[Date "2023.11.24"]
[Round "?"]
[White "Rebel-16.2"]
[Black "Crafty 25.6"]
[Result "1/2-1/2"]
[ECO "B66"]
[PlyCount "343"]
[TimeControl "7200+10"]

{4096MB, LAPTOP-NCDN8BTK} 1. e4 {[%eval 1,0] [%emt 0:00:00]} c5 {[%emt 0:00:05]
} 2. Nf3 {[%eval 1,0] [%emt 0:00:00]} Nc6 {[%emt 0:00:06]} 3. d4 {[%eval 1,0]
[%emt 0:00:00]} cxd4 {[%emt 0:00:05]} 4. Nxd4 {[%eval 78,34] [%emt 0:03:39]}
Nf6 {[%emt 0:00:07] (e6)} 5. Nc3 {[%eval 1,0] [%emt 0:00:01]} d6 {[%emt 0:00:
06]} 6. Bg5 {[%eval 1,0] [%emt 0:00:01]} e6 {[%emt 0:00:06]} 7. Qd2 {[%eval 1,
0] [%emt 0:00:00]} a6 {[%emt 0:00:05]} 8. O-O-O {[%eval 1,0] [%emt 0:00:00]}
Bd7 {[%emt 0:00:06]} 9. f3 {[%eval 1,0] [%emt 0:00:00]} h6 {[%emt 0:00:05]} 10.
Be3 {[%eval 1,0] [%emt 0:00:00]} b5 {[%emt 0:00:18]} 11. Kb1 {[%eval 110,33]
[%emt 0:03:25]} Ne5 {[%emt 0:00:08]} 12. f4 {[%eval 116,34] [%emt 0:03:21]} Nc6
{[%emt 0:06:54] (b4)} 13. Bd3 {[%eval 119,33] [%emt 0:03:26]} Be7 {[%emt 0:04:
24]} 14. Rhe1 {[%eval 81,34] [%emt 0:05:21]} Qc7 {[%emt 0:04:20]} 15. Bxb5 {
[%eval 137,37] [%emt 0:06:42]} axb5 {[%emt 0:02:40]} 16. Ndxb5 {[%eval 133,37]
[%emt 0:06:34]} Qb8 {[%emt 0:00:09] (Da5)} 17. Nxd6+ {[%eval 133,34] [%emt 0:
04:29]} Bxd6 {[%emt 0:00:08]} 18. Qxd6 {[%eval 130,35] [%emt 0:03:06]} Qxd6 {
[%emt 0:00:07] (Se7)} 19. Rxd6 {[%eval 142,35] [%emt 0:02:59]} Na5 {[%emt 0:00:
09] (Tc8)} 20. b3 {[%eval 94,35] [%emt 0:03:52]} Rc8 {[%emt 0:03:31]} 21. Rd3 {
[%eval 162,38] [%emt 0:05:16]} Bc6 {[%emt 0:01:07] (Sb7)} 22. Bb6 {[%eval 98,
35] [%emt 0:05:14]} Ra8 {[%emt 0:00:55] (Sb7)} 23. Nd5 {[%eval 217,33] [%emt 0:
02:42]} exd5 {[%emt 0:02:31]} 24. exd5+ {[%eval 223,36] [%emt 0:03:09]} Kd7 {
[%emt 0:00:07]} 25. dxc6+ {[%eval 193,37] [%emt 0:02:19]} Kxc6 {[%emt 0:00:08]}
26. Bd4 {[%eval 198,38] [%emt 0:05:16]} Rhe8 {[%emt 0:00:27] (Tae8)} 27. Red1 {
[%eval 232,35] [%emt 0:01:58]} Ne4 {[%emt 0:02:28] (Te6)} 28. Bxg7 {[%eval 182,
32] [%emt 0:03:47]} Rg8 {[%emt 0:14:14] (Sb7)} 29. Be5 {[%eval 202,34] [%emt 0:
02:00]} Rxg2 {[%emt 0:00:13]} 30. Rh3 {[%eval 201,33] [%emt 0:01:12]} Rag8 {
[%emt 0:00:33]} 31. Rxh6+ {[%eval 192,35] [%emt 0:01:27]} R8g6 {[%emt 0:00:16]}
32. Rxg6+ {[%eval 215,34] [%emt 0:01:07]} Rxg6 {[%emt 0:02:55] (fxg6)} 33. Rd4
{[%eval 204,35] [%emt 0:01:35]} f5 {[%emt 0:01:10]} 34. Rd8 {[%eval 205,36]
[%emt 0:01:39]} Kb6 {[%emt 0:00:29]} 35. Rb8+ {[%eval 199,36] [%emt 0:03:46]}
Ka7 {[%emt 0:00:07] (Sb7)} 36. Rf8 {[%eval 197,36] [%emt 0:01:07]} Nc6 {
[%emt 0:01:54]} 37. h4 {[%eval 187,37] [%emt 0:01:01]} Rg1+ {[%emt 0:01:19]
(Th6)} 38. Kb2 {[%eval 225,20] [%emt 0:00:01]} Rh1 {[%emt 0:11:23]} 39. Rxf5 {
[%eval 183,38] [%emt 0:01:43]} Rxh4 {[%emt 0:00:08]} 40. Rf8 {[%eval 182,37]
[%emt 0:01:03]} Rh2 {[%emt 0:01:19]} 41. a3 {[%eval 181,38] [%emt 0:03:09]} Kb6
{[%emt 0:00:09] (Sxe5)} 42. Rh8 {[%eval 163,35] [%emt 0:02:46]} Re2 {[%emt 0:
01:26] (Tf2)} 43. Rh5 {[%eval 165,33] [%emt 0:01:07]} Ng3 {[%emt 0:01:14] (Tf2)
} 44. Rh6 {[%eval 167,35] [%emt 0:00:51]} Kb7 {[%emt 0:00:19]} 45. Rh7+ {
[%eval 158,35] [%emt 0:01:37]} Kb6 {[%emt 0:01:11] (Ka6)} 46. Rh6 {[%eval 0,4]
[%emt 0:00:01]} Kb7 {[%emt 0:01:20]} 47. Rh7+ {[%eval 0,4] [%emt 0:00:01]} Kb6
{[%emt 0:01:12]} 48. Bc7+ {[%eval 165,32] [%emt 0:01:26]} Kc5 {[%emt 0:01:06]}
49. Rh3 {[%eval 176,36] [%emt 0:01:12]} Rg2 {[%emt 0:01:09]} 50. Be5 {[%eval
173,37] [%emt 0:02:10]} Kd5 {[%emt 0:01:02] (Se4)} 51. Bc3 {[%eval 173,34]
[%emt 0:00:46]} Ne2 {[%emt 0:01:11] (Se4)} 52. Rd3+ {[%eval 148,33] [%emt 0:00:
53]} Ke4 {[%emt 0:01:17]} 53. f5 {[%eval 141,35] [%emt 0:02:13]} Ned4 {[%emt 0:
00:57] (Sxc3)} 54. Bxd4 {[%eval 132,37] [%emt 0:01:05]} Nxd4 {[%emt 0:00:55]}
55. Rc3 {[%eval 128,39] [%emt 0:00:59]} Nxf5 {[%emt 0:00:07] (Kxf5)} 56. b4 {
[%eval 124,36] [%emt 0:00:47]} Ne3 {[%emt 0:00:52] (Sd6)} 57. Rc7 {[%eval 116,
33] [%emt 0:00:21]} Rg3 {[%emt 0:00:54] (Tg1)} 58. c3 {[%eval 74,30] [%emt 0:
00:39]} Rg1 {[%emt 0:00:51] (Sf1)} 59. a4 {[%eval 116,32] [%emt 0:01:18]} Rd1 {
[%emt 0:00:07]} 60. Rh7 {[%eval 24,35] [%emt 0:00:20]} Kd3 {[%emt 0:00:49]} 61.
Rh3 {[%eval 36,37] [%emt 0:00:22]} Re1 {[%emt 0:00:49]} 62. Rf3 {[%eval 39,40]
[%emt 0:00:30]} Ke4 {[%emt 0:00:10] (Te2+)} 63. Rf8 {[%eval 31,35] [%emt 0:00:
33]} Rd1 {[%emt 0:00:11]} 64. Rf7 {[%eval 31,35] [%emt 0:01:12]} Kd5 {[%emt 0:
01:18]} 65. Rh7 {[%eval 0,39] [%emt 0:00:57]} Nc4+ {[%emt 0:00:45]} 66. Kc2 {
[%eval 0,43] [%emt 0:01:01]} Rd2+ {[%emt 0:00:08]} 67. Kc1 {[%eval 0,44] [%emt
0:00:32]} Ra2 {[%emt 0:00:08] (Tg2)} 68. Rh4 {[%eval 0,36] [%emt 0:00:19]} Ne3
{[%emt 0:01:06]} 69. Kb1 {[%eval 0,42] [%emt 0:00:54]} Rg2 {[%emt 0:00:49]
(Txa4)} 70. Rd4+ {[%eval 0,38] [%emt 0:00:29]} Kc6 {[%emt 0:00:50] (Ke5)} 71.
Rf4 {[%eval 0,35] [%emt 0:00:20]} Rh2 {[%emt 0:00:52] (Sc2)} 72. Re4 {[%eval 0,
38] [%emt 0:00:22]} Nd1 {[%emt 0:00:10] (Sc2)} 73. Rc4+ {[%eval 0,39] [%emt 0:
00:28]} Kb7 {[%emt 0:00:07] (Kb6)} 74. Rc5 {[%eval 0,39] [%emt 0:00:25]} Nb2 {
[%emt 0:00:08]} 75. Rb5+ {[%eval 0,40] [%emt 0:00:19]} Ka7 {[%emt 0:00:19]
(Kc6)} 76. Kc1 {[%eval 0,38] [%emt 0:00:18]} Nd3+ {[%emt 0:00:41]} 77. Kb1 {
[%eval 0,43] [%emt 0:00:21]} Rb2+ {[%emt 0:00:40] (Ka6)} 78. Ka1 {[%eval 0,34]
[%emt 0:00:01]} Rb3 {[%emt 0:00:29]} 79. Ra5+ {[%eval 0,43] [%emt 0:00:22]} Kb8
{[%emt 0:00:37]} 80. Rb5+ {[%eval 0,44] [%emt 0:00:44]} Kc7 {[%emt 0:00:08]
(Ka8)} 81. a5 {[%eval 0,40] [%emt 0:00:18]} Kd6 {[%emt 0:00:11] (Txc3)} 82. a6
{[%eval 0,39] [%emt 0:00:21]} Ra3+ {[%emt 0:00:36] (Txc3)} 83. Kb1 {[%eval 0,
29] [%emt 0:00:01]} Rxa6 {[%emt 0:00:26]} 84. Rh5 {[%eval 0,41] [%emt 0:00:35]}
Ra8 {[%emt 0:01:07] (Tc6)} 85. Rh6+ {[%eval 0,39] [%emt 0:00:26]} Kd5 {[%emt 0:
00:32] (Kc7)} 86. Rh4 {[%eval 0,38] [%emt 0:00:20]} Ne5 {[%emt 0:00:32]} 87.
Kc2 {[%eval 0,40] [%emt 0:00:26]} Rg8 {[%emt 0:00:48]} 88. Kb3 {[%eval 0,40]
[%emt 0:00:28]} Rg3 {[%emt 0:00:31]} 89. Rh5 {[%eval 0,39] [%emt 0:00:27]} Rf3
{[%emt 0:00:19] (Tg1)} 90. Rh1 {[%eval 0,36] [%emt 0:00:26]} Kc6 {[%emt 0:00:
29] (Tg3)} 91. Rh6+ {[%eval 0,36] [%emt 0:00:24]} Kb5 {[%emt 0:00:28] (Kc7)}
92. Rh5 {[%eval 0,39] [%emt 0:00:26]} Re3 {[%emt 0:00:08]} 93. Rh7 {[%eval 0,
41] [%emt 0:00:22]} Kc6 {[%emt 0:00:44] (Sc4)} 94. Rh6+ {[%eval 0,40] [%emt 0:
00:17]} Kd5 {[%emt 0:00:08] (Kc7)} 95. Rh7 {[%eval 0,38] [%emt 0:00:14]} Ng4 {
[%emt 0:00:37] (Kd6)} 96. Rd7+ {[%eval 0,37] [%emt 0:00:16]} Kc6 {[%emt 0:00:
26]} 97. Rh7 {[%eval 0,36] [%emt 0:00:21]} Rf3 {[%emt 0:00:25]} 98. Rh8 {
[%eval 0,35] [%emt 0:00:16]} Ne3 {[%emt 0:00:24] (Kc7)} 99. Ka4 {[%eval 0,34]
[%emt 0:00:26]} Rf4 {[%emt 0:00:42] (Sd1)} 100. Kb3 {[%eval 0,37] [%emt 0:00:
14]} Kb7 {[%emt 0:00:13]} 101. Rh7+ {[%eval 0,39] [%emt 0:00:11]} Kb6 {[%emt 0:
00:10] (Ka8)} 102. Rh6+ {[%eval 0,41] [%emt 0:00:13]} Kb5 {[%emt 0:00:55] (Ka7)
} 103. Rh5+ {[%eval 0,41] [%emt 0:00:17]} Nf5 {[%emt 0:00:21] (Kb6)} 104. c4+ {
[%eval 0,37] [%emt 0:00:12]} Kc6 {[%emt 0:00:12] (Kb6)} 105. Rh3 {[%eval 0,37]
[%emt 0:00:15]} Re4 {[%emt 0:00:21] (Kd6)} 106. Rh5 {[%eval 0,39] [%emt 0:00:
11]} Re3+ {[%emt 0:00:23]} 107. Ka4 {[%eval 0,39] [%emt 0:00:15]} Re5 {[%emt 0:
00:16] (Tf3)} 108. Rg5 {[%eval 0,36] [%emt 0:00:08]} Kd6 {[%emt 0:00:20]} 109.
c5+ {[%eval 0,39] [%emt 0:00:09]} Ke6 {[%emt 0:00:18] (Kc6)} 110. Ka5 {[%eval
0,35] [%emt 0:00:10]} Kf6 {[%emt 0:00:18] (Se7)} 111. Rh5 {[%eval 0,29] [%emt
0:00:08]} Nd4 {[%emt 0:00:19] (Se3)} 112. Rh4 {[%eval 0,29] [%emt 0:00:10]} Rd5
{[%emt 0:00:20] (Sf5)} 113. Kb6 {[%eval 41,31] [%emt 0:00:08]} Ke6 {[%emt 0:00:
07] (Ke5)} 114. Re4+ {[%eval 41,31] [%emt 0:00:19]} Kd7 {[%emt 0:00:18] (Kf5)}
115. Rg4 {[%eval 41,32] [%emt 0:00:17]} Ke7 {[%emt 0:00:18] (Ke6)} 116. Rg7+ {
[%eval 0,31] [%emt 0:00:16]} Kf6 {[%emt 0:00:17] (Ke6)} 117. Rg8 {[%eval 49,33]
[%emt 0:00:14]} Re5 {[%emt 0:00:16] (Ke6)} 118. c6 {[%eval 51,37] [%emt 0:00:
12]} Rb5+ {[%emt 0:00:17]} 119. Ka6 {[%eval 51,41] [%emt 0:00:11]} Rxb4 {
[%emt 0:00:07]} 120. c7 {[%eval 51,40] [%emt 0:00:13]} Rc4 {[%emt 0:00:16]}
121. c8=Q {[%eval 52,44] [%emt 0:00:12]} Rxc8 {[%emt 0:00:09]} 122. Rxc8 {
[%eval 52,46] [%emt 0:00:09]} Kg6 {[%emt 0:00:07] (Sf5)} 123. Rf8 {[%eval 77,
34] [%emt 0:00:06]} Kh6 {[%emt 0:00:06] (Kg5)} 124. Rg8 {[%eval 76,35] [%emt 0:
00:07]} Nc6 {[%emt 0:00:12] (Kh7)} 125. Kb5 {[%eval 76,43] [%emt 0:00:09]} Kh7
{[%emt 0:00:07] (Se7)} 126. Rg3 {[%eval 140,40] [%emt 0:00:14]} Nd4+ {[%emt 0:
00:08] (Se7)} 127. Kc5 {[%eval 136,44] [%emt 0:00:07]} Nf5 {[%emt 0:00:08]}
128. Rg5 {[%eval 144,49] [%emt 0:00:13]} Ne3 {[%emt 0:00:07] (Sh6)} 129. Kd4 {
[%eval 142,42] [%emt 0:00:06]} Kh6 {[%emt 0:00:06]} 130. Rg8 {[%eval 142,46]
[%emt 0:00:13]} Kh7 {[%emt 0:00:07] (Sf5+)} 131. Rg5 {[%eval 0,4] [%emt 0:00:
01]} Kh6 {[%emt 0:00:06]} 132. Rg8 {[%eval 0,4] [%emt 0:00:01]} Kh7 {[%emt 0:
00:09]} 133. Rg1 {[%eval 194,36] [%emt 0:00:11]} Nf5+ {[%emt 0:00:07]} 134. Ke5
{[%eval 171,32] [%emt 0:00:08]} Nh6 {[%emt 0:00:07] (Se7)} 135. Rg2 {[%eval
157,33] [%emt 0:00:14]} Ng8 {[%emt 0:00:07]} 136. Kd6 {[%eval 147,34] [%emt 0:
00:13]} Nf6 {[%emt 0:00:07] (Sh6)} 137. Ke6 {[%eval 166,30] [%emt 0:00:16]} Ng8
{[%emt 0:00:06]} 138. Kd6 {[%eval 0,29] [%emt 0:00:04]} Nf6 {[%emt 0:00:10]}
139. Ke6 {[%eval 0,4] [%emt 0:00:01]} Ng8 {[%emt 0:00:06]} 140. Kf7 {[%eval
140,41] [%emt 0:00:06]} Nh6+ {[%emt 0:00:07]} 141. Ke7 {[%eval 138,45] [%emt 0:
00:06]} Ng8+ {[%emt 0:00:07]} 142. Kf7 {[%eval 0,36] [%emt 0:00:10]} Nh6+ {
[%emt 0:00:07]} 143. Ke7 {[%eval 0,4] [%emt 0:00:01]} Ng8+ {[%emt 0:00:06]}
144. Ke6 {[%eval 142,37] [%emt 0:00:15]} Nh6 {[%emt 0:00:07]} 145. Ke5 {
[%eval 0,29] [%emt 0:00:03]} Ng8 {[%emt 0:00:09]} 146. Ke6 {[%eval 0,4] [%emt
0:00:01]} Nh6 {[%emt 0:00:06]} 147. Rg1 {[%eval 140,27] [%emt 0:00:14]} Ng8 {
[%emt 0:00:08]} 148. Rg4 {[%eval 141,35] [%emt 0:00:18]} Kh8 {[%emt 0:00:06]}
149. Rg6 {[%eval 139,35] [%emt 0:00:16]} Kh7 {[%emt 0:00:07]} 150. Rg1 {
[%eval 136,34] [%emt 0:00:16]} Nh6 {[%emt 0:00:07]} 151. Rg5 {[%eval 135,33]
[%emt 0:00:06]} Ng8 {[%emt 0:00:06]} 152. Rg4 {[%eval 0,37] [%emt 0:00:01]} Kh8
{[%emt 0:00:08]} 153. Rg6 {[%eval 0,4] [%emt 0:00:01]} Kh7 {[%emt 0:00:05]}
154. Rg1 {[%eval 0,4] [%emt 0:00:00]} Nh6 {[%emt 0:00:06]} 155. Rg5 {[%eval 0,
4] [%emt 0:00:01]} Ng8 {[%emt 0:00:06]} 156. Kd6 {[%eval 117,32] [%emt 0:00:21]
} Nf6 {[%emt 0:00:07]} 157. Ke5 {[%eval 76,32] [%emt 0:00:16]} Ne8 {[%emt 0:00:
07] (Sg8)} 158. Kf5 {[%eval 134,29] [%emt 0:00:22]} Nd6+ {[%emt 0:00:07]} 159.
Kf4 {[%eval 119,28] [%emt 0:00:18]} Nf7 {[%emt 0:00:07] (Kh6)} 160. Rg1 {
[%eval 70,27] [%emt 0:00:14]} Kh6 {[%emt 0:00:06]} 161. Kf5 {[%eval 70,24]
[%emt 0:00:20]} Nd6+ {[%emt 0:00:07]} 162. Ke6 {[%eval 63,25] [%emt 0:00:14]}
Ne4 {[%emt 0:00:06] (Se8)} 163. Rf1 {[%eval 22,22] [%emt 0:00:14]} Ng3 {
[%emt 0:00:07] (Kg5)} 164. Rf3 {[%eval 0,23] [%emt 0:00:14]} Ne4 {[%emt 0:00:
06] (Sh5)} 165. Rf1 {[%eval 0,4] [%emt 0:00:00]} Ng3 {[%emt 0:00:06]} 166. Rf3
{[%eval 0,4] [%emt 0:00:00]} Ne4 {[%emt 0:00:05]} 167. Kd5 {[%eval 0,65] [%emt
0:00:06]} Ng5 {[%emt 0:00:10]} 168. Rg3 {[%eval 0,89] [%emt 0:00:08]} Kh5 {
[%emt 0:00:07] (Sh7)} 169. Rg2 {[%eval 0,91] [%emt 0:00:16]} Kh4 {[%emt 0:00:
06] (Sf7)} 170. Ke5 {[%eval 0,97] [%emt 0:00:07]} Nh3 {[%emt 0:00:07] (Kh5)}
171. Kf5 {[%eval 0,116] [%emt 0:00:04]} Kh5 {[%emt 0:00:06]} 172. Rg8 {[%eval
0,116] [%emt 0:00:00]} 1/2-1/2

lol when is a scotch not a draw

texium wrote:

lol when is a scotch not a draw

When it's single malt. Then it's a winner.

@Dann, just curious, did you ever try to make an opening book from your large epd base ?

Admin wrote:

@Dann, just curious, did you ever try to make an opening book from your large epd base ?

Well, my database is terabytes in size, but I have made a SQL link to it. It is impractical as a book in some sense, but on the other hand, Bojun Guo's database is probably more uniform and it may be of similar size now. That one is available online. I don't have a static IP address and I feel that if I did obtain one and give out the address of my database I would have to deal with unwanted attacks.

My database is in constant flux, and it enlarges all the time. I have given out bands of data on request to several people over time.

A significant volume of data is also more or less curiosities. For instance, I have many millions of unusual checkmates that would not come up in real games. I also have a checkmate multiplier, which can take a single checkmate and turn it into hundreds of variants.

I have a tool to take data from chess contests like TCEC, CEGT, CCRL, etc. and atomize the data into decorated EPD data, which I push into my database. So I have hundreds of machines world over which are slaving to fill my database for no cost to me.

Ed, I finally have a laptop with AVX2. Where can I download Rebel 16.2? The provided links don't work, it seems.

matejst wrote:

Ed, I finally have a laptop with AVX2. Where can I download Rebel 16.2? The provided links don't work, it seems.

Wow.........

https://rebel7775.wixsite.com/rebel
https://rebel7775.wixsite.com/rebel/blank-1 = chess system tal

Ed, I already downloaded Rebel EAS. But I would also like to have the previous version. I guess that now, with NNs, it could be pointless, but nonetheless, I like sometimes old versions better.

Hello Matejst, here is my copy of Rebel 16.2 Date is 20-02-2023

https://www.dropbox.com/s/k2y709fot30kp5c/Rebel-16.2.7z?dl=0


This was the version before that as far as I can see:

https://www.dropbox.com/scl/fi/0kex3ls6m00pn172g5ibp/REBEL-16.7z?rlkey=t46fi4x8yznx8a8wujpl2ugyx&dl=0

Thanks, Eelco!

I test Rebel16.2 with no book against other weak opponents like stockfish15 10k or stockfish15 30k

stockfish15 10k is supposed to be significantly weaker than crafty based on the following rating list and I am surprised by stockfish's good results because stockfish15 10k got one draw out of 2 against Rebel16.2 100M(I see in the pgn less than 100M but I guess it does not report the line when it stopped the search).

Rebel16.2 100M opponents are Stockfsih15 10k,30k,100k and stockfish16.1 10k 30k,100k(2 games against every opponent)

Results:
1.5-0.5 against Stockfish15 10k,stockfish15 30k,stockfish15 100k stockfish16.1 100k
2-0 against stockfish16.1 10k stockfish 16.1 30k

Here is the game when Rebel16.1 100M failed to win with black against stockfish15 10k

https://lichess.org/1uLsJa6V#62