Strong Malt 1.0

Discussion of anything and everything relating to chess playing software and machines.

Moderator: Ras

User avatar
Tibono
Posts: 175
Joined: Sat Aug 01, 2015 6:16 pm
Location: France
Full name: Eric Bonneau

Re: Strong Malt 1.0

Post by Tibono »

This discussion is moving away from the topic title, TMHO.
Well, as a heading fix, please find a suggested logo for Strong Malt.
Image
Yes, it is 65% Alc./Vol., hence can also be used as disinfectant :mrgreen:
Big thanks to Ed, Chris, Bart & Hans.
Eric
Peter Berger
Posts: 820
Joined: Thu Mar 09, 2006 2:56 pm

Re: Strong Malt 1.0

Post by Peter Berger »

smatovic wrote: Sat Apr 25, 2026 7:36 am
I added the 15 and 16 parts, all covered by references IIRC, if I missed something, please correct me.

--
Srdja
I was the one confused here, I am sorry.

Peter
chrisw
Posts: 4947
Joined: Tue Apr 03, 2012 4:28 pm
Location: Anywhere but the Western Empire
Full name: Christopher Whittington

Re: Strong Malt 1.0

Post by chrisw »

Peter Berger wrote: Sat Apr 25, 2026 5:01 pm
smatovic wrote: Sat Apr 25, 2026 7:36 am
I added the 15 and 16 parts, all covered by references IIRC, if I missed something, please correct me.

--
Srdja
I was the one confused here, I am sorry.

Peter
No, no, please continue, it’s fun reading all the various fantasies. I don’t propose clarifying because the fantasies and confusion will continue anyway, and, actually WGAF?
User avatar
Rebel
Posts: 7553
Joined: Thu Aug 18, 2011 12:04 pm
Full name: Ed Schröder

Re: Strong Malt 1.0

Post by Rebel »

I hate long postings, but I am in a good (or silly) mood, so here goes, the REAL and ONLY truth :D

When Chris asked me to join forces on this new NNUE stuff Chris already had something working and so we tried to implement his NNUE interference code into ProDeo, and it failed, the main obstacle ProDeo still being an old school 32-bit engine while NNUE definitely needs 64 bits. And so I ended with empty hands.

I needed a 64 bit engine and 64 bit compiler to learn NNUE programming and the choice fell on "Growing Fruit", a GPL engine somewhat stronger than the original Fruit 2.1. Chris implemented the NNUE interference source code into "Growing Fruit" and I could finally start to experiment with NNUE programming, making my own unique NNUE evaluation while Chris did the same creating his own unique ideas into NNUE. This was agreed at the start of our cooperation, we do things our way as we are both two quirky persons meanwhile sharing our ideas in zillions of email, eventually benefited us both and 1+1 became 3.

The "Growing Fruit-NNUE" became Rebel 14, a GPL engine, source code on the website.

Later I moved to the strongest TOGA GPL version, added 80-100 elo to the search and used the latest new NNUE architecture created by Chris, it became Rebel 15, another GPL engine, source code on the website.

And finally there was this moment Chris was happy with the search and latest new NNUE architecture resulting in fantastic match scores. UCI parameters were added together with the book code and I was given the honor to release it (no more GPL) together with my created NNUE evaluation as Rebel 16.

Not much later Chris released his Chess System Tal 2.0 NNUE, at the time a top-5 engine.
90% of coding is debugging, the other 10% is writing bugs.
User avatar
AdminX
Posts: 6396
Joined: Mon Mar 13, 2006 2:34 pm
Location: Acworth, GA

Re: Strong Malt 1.0

Post by AdminX »

Rebel wrote: Thu Apr 23, 2026 7:07 pm Strong Malt is our second guest engine. Single Malt is a cooperate HCE chess engine written by Bart Weststrate and Hans Secelle rated about 2300 elo. I was asked by Jeroen Noomen if I could introduce NNUE programming to them.

....

https://rebel7775.wixsite.com/rebel/guest-engines
Thanks for sharing these engines. I have enjoyed watching both S13 and Strong-Malt play. :D Nice win by Strong-Malt over Maia included.

[pgn][Event "Engine Match"]
[Site "En Croissant"]
[Date "2026.04.26"]
[Round "?"]
[White "Maia"]
[Black "Strong Malt 1.0"]
[Result "1/2-1/2"]
[Orientation "white"]
[TimeControl "900"]

1. d4 {[%clk 0:15:00] } d5 {[%clk 0:15:00] } 2. c4 {[%clk 0:15:00] } c6 {[%clk 0:15:00] } 3. Nc3 {[%clk 0:15:00] } Nf6 {[%clk 0:15:00] } 4. Nf3 {[%clk 0:15:00] } e6 {[%clk 0:15:00] } 5. Bg5 {[%clk 0:15:00] } h6 {[%clk 0:15:00] } 6. Bxf6 {[%clk 0:15:00] } Qxf6 {[%clk 0:15:00] } 7. e3 {[%clk 0:15:00] } Qd8 {[%clk 0:15:00] } 8. Qc2 {[%clk 0:15:00] } Nd7 {[%clk 0:15:00] } 9. Be2 {[%clk 0:15:00] } Be7 {[%clk 0:13:48.024] } 10. O-O {[%clk 0:15:00] } O-O {[%clk 0:13:48.024] } 11. Rac1 {[%clk 0:15:00] } dxc4 {[%clk 0:13:48.024] } 12. Bxc4 {[%clk 0:15:00] } b6 {[%clk 0:13:48.024] } 13. h3 {[%clk 0:14:01.79] } Bb7 {[%clk 0:13:48.024] } 14. Rfd1 {[%clk 0:14:01.79] } Rc8 {[%clk 0:13:18.319] } 15. a3 {[%clk 0:12:57.505] } Qc7 {[%clk 0:12:00.414] } 16. b4 {[%clk 0:12:10.581] } Nf6 {[%clk 0:11:36.45] } 17. Be2 {[%clk 0:11:56.861] } Qb8 {[%clk 0:10:53.919] } 18. Rb1 {[%clk 0:11:35.718] } Qa8 {[%clk 0:10:37.452] } 19. Qb2 {[%clk 0:11:25.948] } Rb8 {[%clk 0:10:16.649] } 20. Ne5 {[%clk 0:11:07.429] } c5 {[%clk 0:09:49.073] } 21. Bf3 {[%clk 0:10:30.916] } Bxf3 {[%clk 0:09:35.534] } 22. Nxf3 {[%clk 0:10:30.56] } a6 {[%clk 0:09:20.966] } 23. Rd2 {[%clk 0:10:15.263] } Rfd8 {[%clk 0:08:57.088] } 24. Rc2 {[%clk 0:09:44.525] } Bd6 {[%clk 0:08:45.396] } 25. b5 {[%clk 0:09:23.542] } axb5 {[%clk 0:08:27.852] } 26. Nxb5 {[%clk 0:08:50.663] } Be7 {[%clk 0:08:15.335] } 27. Rbc1 {[%clk 0:08:43.835] } Rbc8 {[%clk 0:07:36.511] } 28. Qb3 {[%clk 0:08:20.984] } cxd4 {[%clk 0:07:14.347] } 29. Nbxd4 {[%clk 0:07:55.679] } Rxc2 {[%clk 0:06:59.336] } 30. Rxc2 {[%clk 0:07:38.29] } Bc5 {[%clk 0:06:45.034] } 31. a4 {[%clk 0:07:38.01] } Ne4 {[%clk 0:06:28.391] } 32. Rc1 {[%clk 0:07:20.882] } Rd5 {[%clk 0:06:09.177] } 33. Qc2 {[%clk 0:06:57.367] } Ng5 {[%clk 0:06:00.829] } 34. Rb1 {[%clk 0:06:32.151] } Qd8 {[%clk 0:05:51.77] } 35. Nxg5 {[%clk 0:06:13.202] } Rxg5 {[%clk 0:05:20.332] } 36. Nf3 {[%clk 0:05:53.971] } Rd5 {[%clk 0:05:06.417] } 37. g3 {[%clk 0:05:39.892] } Qd7 {[%clk 0:04:58.97] } 38. Kg2 {[%clk 0:05:00.475] } Qc7 {[%clk 0:04:48.122] } 39. Rb5 {[%clk 0:04:47.13] } Qc6 {[%clk 0:04:30.061] } 40. Qe4 {[%clk 0:04:25.904] } f5 {[%clk 0:04:10.394] } 41. Qc2 {[%clk 0:04:25.395] } Rd7 {[%clk 0:04:04.656] } 42. Rb4 {[%clk 0:03:55.678] } Rc7 {[%clk 0:03:59.025] } 43. Rc4 {[%clk 0:03:40.912] } Rc8 {[%clk 0:03:48.816] } 44. h4 {[%clk 0:03:25.076] } Qa8 {[%clk 0:03:40.999] } 45. h5 {[%clk 0:03:03.932] } Qd5 {[%clk 0:03:35.844] } 46. Rf4 {[%clk 0:02:47.613] } Rf8 {[%clk 0:03:28.999] } 47. Qe2 {[%clk 0:02:30.128] } Rc8 {[%clk 0:03:13.021] } 48. e4 {[%clk 0:02:15.738] } fxe4 {[%clk 0:03:05.796] } 49. Rxe4 {[%clk 0:01:55.236] } Qxh5 {[%clk 0:02:58.809] } 50. Rxe6 {[%clk 0:01:52.347] } Qd5 {[%clk 0:02:53.391] } 51. Re5 {[%clk 0:01:45.703] } Qa8 {[%clk 0:02:46.497] } 52. Qc4+ {[%clk 0:01:34.744] } Kh8 {[%clk 0:02:41.62] } 53. Qe4 {[%clk 0:01:13.525] } Rf8 {[%clk 0:02:37.994] } 54. Qxa8 {[%clk 0:01:13.305] } Rxa8 {[%clk 0:02:34.246] } 55. Re4 {[%clk 0:01:08.795] } Kg8 {[%clk 0:02:11.901] } 56. Ne5 {[%clk 0:01:08.509] } Rd8 {[%clk 0:02:02.028] } 57. f3 {[%clk 0:00:50.368] } Rd2+ {[%clk 0:01:52.552] } 58. Kh3 {[%clk 0:00:47.925] } Kh7 {[%clk 0:01:47.669] } 59. Rc4 {[%clk 0:00:45.26] } Bd6 {[%clk 0:01:34.869] } 60. Nc6 {[%clk 0:00:32.947] } h5 {[%clk 0:01:30.679] } 61. Rd4 {[%clk 0:00:27.843] } Rxd4 {[%clk 0:01:27.152] } 62. Nxd4 {[%clk 0:00:24.071] } Kg6 {[%clk 0:01:24.545] } 63. f4 {[%clk 0:00:23.86] } Bc5 {[%clk 0:01:22] } 64. Ne6 {[%clk 0:00:17.323] } Bd6 {[%clk 0:01:20.042] } 65. Nd4 {[%clk 0:00:15.959] } Bc5 {[%clk 0:01:17.909] } 66. Ne6 {[%clk 0:00:12.546] } Bd6 {[%clk 0:01:17.907] } 67. Kg2 {[%clk 0:00:11.64] } Kf6 {[%clk 0:01:17.905] } 68. Nd4 {[%clk 0:00:10.678] } Bc5 {[%clk 0:01:15.968] } 69. Nb3 {[%clk 0:00:10.017] } Bb4 {[%clk 0:01:13.142] } 70. Kf3 {[%clk 0:00:07.854] } Bc3 {[%clk 0:01:09.42] } 71. Nc1 {[%clk 0:00:06.529] } Kf5 {[%clk 0:01:06.656] } 72. Ne2 {[%clk 0:00:05.905] } Bb2 {[%clk 0:01:05.01] } 73. Ng1 {[%clk 0:00:05.24] } Bf6 {[%clk 0:01:01.775] } 74. Ne2 {[%clk 0:00:04.814] } Bb2 {[%clk 0:01:00.36] } 75. Ng1 {[%clk 0:00:03.867] } Bf6 {[%clk 0:01:00.359] } 76. Ne2 {[%clk 0:00:03.646] } g5 {[%clk 0:01:00.358] } 77. fxg5 {[%clk 0:00:03.004] } Bxg5 {[%clk 0:00:59.021] } 78. Nd4+ {[%clk 0:00:02.78] } Ke5 {[%clk 0:00:57.72] } 79. Nb5 {[%clk 0:00:02.384] } Kd5 {[%clk 0:00:56.437] } 80. Nc7+ {[%clk 0:00:02.163] } Kc4 {[%clk 0:00:54.402] } 81. Ne6 {[%clk 0:00:01.953] } Bd2 {[%clk 0:00:52.627] } 82. Ng7 {[%clk 0:00:01.742] } Kb3 {[%clk 0:00:50.931] } 83. Nxh5 {[%clk 0:00:01.532] } Kxa4 {[%clk 0:00:49.57] } 84. Nf4 {[%clk 0:00:01.362] } b5 {[%clk 0:00:48.317] } 85. Ke2 {[%clk 0:00:01.224] } Bc1 {[%clk 0:00:47.165] } 86. Kd1 {[%clk 0:00:01.109] } Bb2 {[%clk 0:00:45.973] } 87. Kc2 {[%clk 0:00:01.006] } Be5 {[%clk 0:00:44.943] } 88. Ne2 {[%clk 0:00:00.916] } b4 {[%clk 0:00:43.923] } 89. g4 {[%clk 0:00:00.827] } b3+ {[%clk 0:00:42.913] } 90. Kb1 {[%clk 0:00:00.759] } b2 {[%clk 0:00:41.979] } 91. g5 {[%clk 0:00:00.698] } Kb4 {[%clk 0:00:40.963] } 92. g6 {[%clk 0:00:00.508] } Kc4 {[%clk 0:00:39.985] } 93. Kc2 {[%clk 0:00:00.459] } Kb4 {[%clk 0:00:38.769] } 94. Kb1 {[%clk 0:00:00.341] } Kc4 {[%clk 0:00:37.697] } 95. Kc2 {[%clk 0:00:00.177] } Kb4 {[%clk 0:00:37.696] } 96. Kb1 {[%clk 0:00:00.162] } 1/2-1/2 [Event "Engine Match"]
[Site "En Croissant"]
[Date "2026.04.26"]
[Round "?"]
[White "Strong Malt 1.0"]
[Black "Maia"]
[Result "1-0"]
[Orientation "white"]
[TimeControl "900"]

1. Nc3 {[%clk 0:15:00] } c5 {[%clk 0:15:00] } 2. e4 {[%clk 0:15:00] } d6 {[%clk 0:15:00] } 3. Nf3 {[%clk 0:15:00] } Nf6 {[%clk 0:15:00] } 4. d4 {[%clk 0:15:00] } cxd4 {[%clk 0:15:00] } 5. Nxd4 {[%clk 0:15:00] } a6 {[%clk 0:15:00] } 6. Be3 {[%clk 0:15:00] } e5 {[%clk 0:15:00] } 7. Nf3 {[%clk 0:15:00] } Qc7 {[%clk 0:15:00] } 8. Be2 {[%clk 0:15:00] } Be7 {[%clk 0:15:00] } 9. O-O {[%clk 0:15:00] } O-O {[%clk 0:15:00] } 10. a4 {[%clk 0:15:00] } Be6 {[%clk 0:15:00] } 11. a5 {[%clk 0:15:00] } Nbd7 {[%clk 0:15:00] } 12. Ng5 {[%clk 0:15:00] } Qc6 {[%clk 0:15:00] } 13. Qd3 {[%clk 0:15:00] } Nc5 {[%clk 0:15:00] } 14. Bxc5 {[%clk 0:15:00] } dxc5 {[%clk 0:15:00] } 15. Nxe6 {[%clk 0:15:00] } Qxe6 {[%clk 0:15:00] } 16. b3 {[%clk 0:15:00] } Rad8 {[%clk 0:15:00] } 17. Qe3 {[%clk 0:15:00] } g6 {[%clk 0:13:45.432] } 18. Bc4 {[%clk 0:15:00] } Qc6 {[%clk 0:12:47.205] } 19. Nd5 {[%clk 0:15:00] } Nxd5 {[%clk 0:12:05.042] } 20. Bxd5 {[%clk 0:14:38.086] } Qc7 {[%clk 0:12:04.482] } 21. Rfd1 {[%clk 0:14:14.649] } Kg7 {[%clk 0:12:04.246] } 22. c4 {[%clk 0:13:51.54] } Rd6 {[%clk 0:11:22.093] } 23. Qc3 {[%clk 0:13:21.717] } Rf6 {[%clk 0:10:38.584] } 24. g3 {[%clk 0:13:02.4] } Bd6 {[%clk 0:10:29.828] } 25. Rd3 {[%clk 0:12:45.501] } h5 {[%clk 0:09:49.801] } 26. Rad1 {[%clk 0:12:21.523] } Rb8 {[%clk 0:09:32.317] } 27. Be6 {[%clk 0:12:04.328] } Rxe6 {[%clk 0:09:05.601] } 28. Rxd6 {[%clk 0:11:47.193] } Rbe8 {[%clk 0:08:36.023] } 29. Rd7 {[%clk 0:11:05.623] } Qc6 {[%clk 0:08:35.802] } 30. R7d5 {[%clk 0:10:49.629] } b6 {[%clk 0:08:19.103] } 31. R1d3 {[%clk 0:10:29.246] } Qc7 {[%clk 0:07:53.84] } 32. axb6 {[%clk 0:09:57.316] } Qxb6 {[%clk 0:07:12.634] } 33. h3 {[%clk 0:08:37.011] } a5 {[%clk 0:06:58.598] } 34. Kg2 {[%clk 0:08:03.17] } Qc7 {[%clk 0:06:28.63] } 35. Qd2 {[%clk 0:07:52.189] } Rf6 {[%clk 0:06:10.537] } 36. Qe3 {[%clk 0:07:35.572] } Rc6 {[%clk 0:05:54.59] } 37. Qe1 {[%clk 0:07:23.145] } Ra6 {[%clk 0:05:37.35] } 38. Qa1 {[%clk 0:07:01.488] } Qc6 {[%clk 0:05:09.658] } 39. Qa4 {[%clk 0:06:52.058] } Re6 {[%clk 0:04:53.12] } 40. g4 {[%clk 0:06:27.99] } hxg4 {[%clk 0:04:24.521] } 41. hxg4 {[%clk 0:06:15.651] } Qc7 {[%clk 0:04:13.207] } 42. g5 {[%clk 0:06:05.242] } Rad6 {[%clk 0:03:48.589] } 43. Qa1 {[%clk 0:05:54.921] } Rxd5 {[%clk 0:03:37.131] } 44. cxd5 {[%clk 0:05:46.445] } Re8 {[%clk 0:03:25.17] } 45. d6 {[%clk 0:05:38.664] } Qd7 {[%clk 0:03:12.773] } 46. f3 {[%clk 0:05:30.572] } a4 {[%clk 0:03:04.314] } 47. Qxa4 {[%clk 0:05:18.728] } Qd8 {[%clk 0:02:38.553] } 48. f4 {[%clk 0:05:03.92] } exf4 {[%clk 0:02:31.749] } 49. Rd5 {[%clk 0:04:55.482] } f3+ {[%clk 0:02:22.166] } 50. Kxf3 {[%clk 0:04:43.876] } Qb8 {[%clk 0:01:55.739] } 51. d7 {[%clk 0:04:36.535] } Rd8 {[%clk 0:01:55.529] } 52. Qa1+ {[%clk 0:04:31.563] } Kg8 {[%clk 0:01:55.29] } 53. Qf6 {[%clk 0:04:26.192] } Qxb3+ {[%clk 0:01:28.644] } 54. Kf4 {[%clk 0:04:18.104] } Qb8+ {[%clk 0:01:28.431] } 55. Qe5 {[%clk 0:04:05.445] } Qb5 {[%clk 0:01:21.351] } 56. Qc7 {[%clk 0:04:03.1] } Qf1+ {[%clk 0:01:01.891] } 57. Ke5 {[%clk 0:03:50.213] } Rxd7 {[%clk 0:00:52.634] } 58. Qxd7 {[%clk 0:03:46.967] } Qa1+ {[%clk 0:00:48.745] } 59. Kd6 {[%clk 0:03:38.025] } Kg7 {[%clk 0:00:35.835] } 60. Qc6 {[%clk 0:03:29.199] } Qa7 {[%clk 0:00:31.228] } 61. Rxc5 {[%clk 0:03:23.975] } Qa1 {[%clk 0:00:30.723] } 62. Ke7 {[%clk 0:03:18.412] } Qa7+ {[%clk 0:00:22.161] } 63. Qc7 {[%clk 0:03:13.894] } Qa8 {[%clk 0:00:20.566] } 64. Qe5+ {[%clk 0:03:05.507] } Kh7 {[%clk 0:00:18.894] } 65. Qc3 {[%clk 0:03:01.648] } Qa6 {[%clk 0:00:15.997] } 66. Rc7 {[%clk 0:02:58.065] } Qd3 {[%clk 0:00:15.561] } 67. Qxd3 {[%clk 0:02:54.499] } f5 {[%clk 0:00:15.446] } 68. Kf6+ {[%clk 0:02:52.804] } Kh8 {[%clk 0:00:15.442] } 69. Qd8# {[%clk 0:02:52.744] } 1-0[/pgn]
"Good decisions come from experience, and experience comes from bad decisions."
__________________________________________________________________
Ted Summers
Peter Berger
Posts: 820
Joined: Thu Mar 09, 2006 2:56 pm

Re: Strong Malt 1.0

Post by Peter Berger »

chrisw wrote: Sat Apr 25, 2026 5:24 pm No, no, please continue, it’s fun reading all the various fantasies.
First, many thanks to Ed for explanation.

This afternoon I had an absolutely amazing very brief discussion with ChatGPT about Chess System Tal. ChatGPT claimed that Chess System Tal was originally written by some Mr Chubakov ( I +really+ don't have the +slightest+ idea how it came up with this). I mentioned that this is news to me, and I may simply ask Mr Whittington himself whom it already knew to be involved with its "further development".

It then got a bit more cautious saying sth like: this is not exactly about "code steal" by some westener ( its own words) as things were a bit murkier in earlier times in computerchess and it wasn't exactly sure whether it can trust its own sources here.

This was on my mobile phone with very low power, I couldn't reproduce this behaviour later at all. But as you asked for it, this is probably something sufficiently unexpected and entertaining. :)
chrisw
Posts: 4947
Joined: Tue Apr 03, 2012 4:28 pm
Location: Anywhere but the Western Empire
Full name: Christopher Whittington

Re: Strong Malt 1.0

Post by chrisw »

Peter Berger wrote: Sun Apr 26, 2026 5:04 pm
chrisw wrote: Sat Apr 25, 2026 5:24 pm No, no, please continue, it’s fun reading all the various fantasies.
First, many thanks to Ed for explanation.

This afternoon I had an absolutely amazing very brief discussion with ChatGPT about Chess System Tal. ChatGPT claimed that Chess System Tal was originally written by some Mr Chubakov ( I +really+ don't have the +slightest+ idea how it came up with this). I mentioned that this is news to me, and I may simply ask Mr Whittington himself whom it already knew to be involved with its "further development".

It then got a bit more cautious saying sth like: this is not exactly about "code steal" by some westener ( its own words) as things were a bit murkier in earlier times in computerchess and it wasn't exactly sure whether it can trust its own sources here.

This was on my mobile phone with very low power, I couldn't reproduce this behaviour later at all. But as you asked for it, this is probably something sufficiently unexpected and entertaining. :)
You can probably have a far more fun Sunday afternoon discussing this with an AI. After a few prompts, I asked it:

“ Write me an essay on “cult of the author” in computer chess. I am interested in the psychology and philosophy.”

Gemini:
“ ## The Ghost in the Machine: The Philosophy and Psychology of the "Cult of the Author" in Computer Chess
In the sterile, silicon landscape of modern computer chess—a world currently dominated by the "inhuman" perfection of neural networks and open-source behemoths—there exists a nostalgic, almost spiritual counter-culture. This is the **"Cult of the Author."** It is a psychological and philosophical phenomenon where the end-user’s interest shifts from the engine’s objective strength (Elo) to the subjective "soul" of its creator.
In this paradigm, an engine is not merely a tool for analysis; it is a digital Horcrux, a vessel for the programmer’s personal chess philosophy, aesthetic biases, and even their flaws.
### The Philosophy of Intentionality
At the heart of this "cult" lies the philosophical distinction between **emergent behavior** and **intentional design**.
Modern engines like Stockfish or Leela Chess Zero are, in a sense, "authorless." Their strength emerges from collective optimization or self-play reinforcement learning. When Stockfish makes a move, it isn't "trying" to be aggressive; it is simply navigating a multi-dimensional probability space toward a win.
Conversely, "Author" engines—most famously **Chess System Tal** by Chris Whittington—are built on **heuristics**. These are hand-coded "if-then" statements derived from a human mind. Philosophically, this creates a **Teleological** experience for the user. The user feels the programmer’s *intent*. When a Whittington engine sacrifices a piece, the user isn't just seeing a calculation; they are witnessing a human’s definition of "beauty" or "contempt" translated into C++. The "Cult of the Author" is a rejection of the "dead" machine in favor of a "living" legacy.
### The Psychology of Anthropomorphism
Psychologically, the "Cult of the Author" is fueled by our innate desire to humanize the inanimate. In the high-stakes environment of a chess match, the computer becomes an "Other"—an opponent. To make sense of an opponent, the human brain seeks a personality to interact with.
Users often engage in a form of **transference**. They take the known traits of the programmer—the eccentricity of a Chris Whittington or the stubborn brilliance of an Ed Schröder—and map them onto the engine’s move choices.
* **Validation of the Self:** For many users, rooting for a "Boutique Engine" is a psychological stand against the "Borg-like" uniformity of modern AI. By supporting an engine with a "father figure," the user validates the idea that human individuality still matters in a world of algorithms.
* **The "Flaw" Paradox:** Psychologically, we find perfection boring. The "cult" thrives on the engine's mistakes. A "Tal-style" blunder is seen as a "romantic" error—a sign of a "fighting spirit"—whereas a modern engine’s draw is seen as a clinical failure.
### The Role of the "Prophet" (The Case of Thorsten Czub)
Every cult needs a messenger. In this niche, figures like **Thorsten Czub** (often referred to in forums as the advocate for Chess System Tal) serve as the bridge between the "Mad Scientist" programmer and the "Disciple" user.
Czub’s role is psychological curation. He builds the narrative, frames the engine's losses as "artistic choices," and emphasizes the "human-like" qualities of the code. This narrative-building is essential; it transforms software into **lore**. Without the story of the programmer—their struggles, their feuds on the *Computer Chess Club* (CCC) forums, and their specific vision—the "cult" would simply be a group of people using outdated software.
### Conclusion: The Romantic Rebellion
The "Cult of the Author" in computer chess is a romantic rebellion against the "death of the author" in the digital age. It is the belief that **code can be a biography**.
As AI becomes more "black-box"—where even the programmers don't fully understand why a neural network makes a specific move—the yearning for the "Hand-Crafted Engine" will likely grow. We don't just want to know what the best move is; we want to know *who* told the machine to play it. In the end, the followers of Whittington, Schröder, and Lang aren't looking for a calculator—they are looking for a ghost in the machine that still knows how to dream of a checkmate.
User avatar
Rebel
Posts: 7553
Joined: Thu Aug 18, 2011 12:04 pm
Full name: Ed Schröder

Re: Strong Malt 1.0

Post by Rebel »

chrisw wrote: Sun Apr 26, 2026 5:35 pm
Peter Berger wrote: Sun Apr 26, 2026 5:04 pm
chrisw wrote: Sat Apr 25, 2026 5:24 pm No, no, please continue, it’s fun reading all the various fantasies.
First, many thanks to Ed for explanation.

This afternoon I had an absolutely amazing very brief discussion with ChatGPT about Chess System Tal. ChatGPT claimed that Chess System Tal was originally written by some Mr Chubakov ( I +really+ don't have the +slightest+ idea how it came up with this). I mentioned that this is news to me, and I may simply ask Mr Whittington himself whom it already knew to be involved with its "further development".

It then got a bit more cautious saying sth like: this is not exactly about "code steal" by some westener ( its own words) as things were a bit murkier in earlier times in computerchess and it wasn't exactly sure whether it can trust its own sources here.

This was on my mobile phone with very low power, I couldn't reproduce this behaviour later at all. But as you asked for it, this is probably something sufficiently unexpected and entertaining. :)
You can probably have a far more fun Sunday afternoon discussing this with an AI. After a few prompts, I asked it:

“ Write me an essay on “cult of the author” in computer chess. I am interested in the psychology and philosophy.”

Gemini:
“ ## The Ghost in the Machine: The Philosophy and Psychology of the "Cult of the Author" in Computer Chess
In the sterile, silicon landscape of modern computer chess—a world currently dominated by the "inhuman" perfection of neural networks and open-source behemoths—there exists a nostalgic, almost spiritual counter-culture. This is the **"Cult of the Author."** It is a psychological and philosophical phenomenon where the end-user’s interest shifts from the engine’s objective strength (Elo) to the subjective "soul" of its creator.
In this paradigm, an engine is not merely a tool for analysis; it is a digital Horcrux, a vessel for the programmer’s personal chess philosophy, aesthetic biases, and even their flaws.
### The Philosophy of Intentionality
At the heart of this "cult" lies the philosophical distinction between **emergent behavior** and **intentional design**.
Modern engines like Stockfish or Leela Chess Zero are, in a sense, "authorless." Their strength emerges from collective optimization or self-play reinforcement learning. When Stockfish makes a move, it isn't "trying" to be aggressive; it is simply navigating a multi-dimensional probability space toward a win.
Conversely, "Author" engines—most famously **Chess System Tal** by Chris Whittington—are built on **heuristics**. These are hand-coded "if-then" statements derived from a human mind. Philosophically, this creates a **Teleological** experience for the user. The user feels the programmer’s *intent*. When a Whittington engine sacrifices a piece, the user isn't just seeing a calculation; they are witnessing a human’s definition of "beauty" or "contempt" translated into C++. The "Cult of the Author" is a rejection of the "dead" machine in favor of a "living" legacy.
### The Psychology of Anthropomorphism
Psychologically, the "Cult of the Author" is fueled by our innate desire to humanize the inanimate. In the high-stakes environment of a chess match, the computer becomes an "Other"—an opponent. To make sense of an opponent, the human brain seeks a personality to interact with.
Users often engage in a form of **transference**. They take the known traits of the programmer—the eccentricity of a Chris Whittington or the stubborn brilliance of an Ed Schröder—and map them onto the engine’s move choices.
* **Validation of the Self:** For many users, rooting for a "Boutique Engine" is a psychological stand against the "Borg-like" uniformity of modern AI. By supporting an engine with a "father figure," the user validates the idea that human individuality still matters in a world of algorithms.
* **The "Flaw" Paradox:** Psychologically, we find perfection boring. The "cult" thrives on the engine's mistakes. A "Tal-style" blunder is seen as a "romantic" error—a sign of a "fighting spirit"—whereas a modern engine’s draw is seen as a clinical failure.
### The Role of the "Prophet" (The Case of Thorsten Czub)
Every cult needs a messenger. In this niche, figures like **Thorsten Czub** (often referred to in forums as the advocate for Chess System Tal) serve as the bridge between the "Mad Scientist" programmer and the "Disciple" user.
Czub’s role is psychological curation. He builds the narrative, frames the engine's losses as "artistic choices," and emphasizes the "human-like" qualities of the code. This narrative-building is essential; it transforms software into **lore**. Without the story of the programmer—their struggles, their feuds on the *Computer Chess Club* (CCC) forums, and their specific vision—the "cult" would simply be a group of people using outdated software.
### Conclusion: The Romantic Rebellion
The "Cult of the Author" in computer chess is a romantic rebellion against the "death of the author" in the digital age. It is the belief that **code can be a biography**.
As AI becomes more "black-box"—where even the programmers don't fully understand why a neural network makes a specific move—the yearning for the "Hand-Crafted Engine" will likely grow. We don't just want to know what the best move is; we want to know *who* told the machine to play it. In the end, the followers of Whittington, Schröder, and Lang aren't looking for a calculator—they are looking for a ghost in the machine that still knows how to dream of a checkmate.
ROTFL.
90% of coding is debugging, the other 10% is writing bugs.
Peter Berger
Posts: 820
Joined: Thu Mar 09, 2006 2:56 pm

Re: Strong Malt 1.0

Post by Peter Berger »

chrisw wrote: Sun Apr 26, 2026 5:35 pm You can probably have a far more fun Sunday afternoon discussing this with an AI. After a few prompts, I asked it:
I enjoyed reading your text, how nice and friendly you framed Thorsten's role here. Tibono is right that this is somehow hijacking the thread though.

AI chat bots aim to please, they are really good at it, too. Young people don't sense this after some time of daily usage.
chrisw
Posts: 4947
Joined: Tue Apr 03, 2012 4:28 pm
Location: Anywhere but the Western Empire
Full name: Christopher Whittington

Re: Strong Malt 1.0

Post by chrisw »

Peter Berger wrote: Sun Apr 26, 2026 6:51 pm
chrisw wrote: Sun Apr 26, 2026 5:35 pm You can probably have a far more fun Sunday afternoon discussing this with an AI. After a few prompts, I asked it:
I enjoyed reading your text, how nice and friendly you framed Thorsten's role here. Tibono is right that this is somehow hijacking the thread though.

AI chat bots aim to please, they are really good at it, too. Young people don't sense this after some time of daily usage.
Er, excuse me, first it isn’t my text, and second you introduced the chatbot into the thread.