DiscoCheck 4.2

[lucasart]

Here's a new, and hopefully better, version of DiscoCheck.

Contempt option added. The default value of 25cp has the good property of significantly reducing the draw rate, while not losing any elo. Thanks to contempt, DiscoCheck will avoid draws and would rather fight to the death than accept a 3-repetition draw in a +/- equal position.

UCI stop command implemented. This means that infinite analysis is finally possible, as well as stopping the engine while it's searching in game play.

No skill level yet, but the recommend way to do that is to limit the number of nodes searched. Now even small values will never crash. For example, you can try 256 nodes, 512 nodes, 1024 nodes, etc. until you find the suitable strength opponent for you.

Just click the link in my signature, and you will find:
* source code in 'src' folder
* compiles for Linux and Windows (64-bit, SSE 4.2) in the 'bin' folder

Regarding ELO, I reckon about +20 ELO compared to DiscoCheck 4.1.

Enjoy!

[lucasart]

Once I could program in sevral languages. Now I am not even able to download a file from the internet. Total collapse.

LOL

The file is called
* discocheck_4.2.exe for Windows
* discocheck_4.2 for Linux

EDIT:
Actually, I've had the same problem trying it myself. It is a stupid behaviour from GitHub, that tries to "view" the files, instead of offering to savce them on disk.
What you can do is click on the "ZIP" button, which will download the entire repo. Then you have a ZIP file with a bin folder in it, and all the compiles there.
Let me know if that doesn't work.

[lucasart]

Once I could program in sevral languages. Now I am not even able to download a file from the internet. Total collapse.

LOL

The file is called
* discocheck_4.2.exe for Windows
* discocheck_4.2 for Linux

Am I correct in assuming that the program does not run in a 32-bit environment?

Yes. The compiles provided are 64-bit only.

[lucasart]

OK, here's the download link to make it simple:
https://github.com/lucasart/chess/archive/master.zip

[noctiferus]

Actually. I too had troubles, before figuring out that the download was under the label "Raw"... Blame me

[lucasart]

No skill level yet, but the recommend way to do that is to limit the number of nodes searched. Now even small values will never crash. For example, you can try 256 nodes, 512 nodes, 1024 nodes, etc. until you find the suitable strength opponent for you.

Here's a game I just played against DiscoCheck, limiting the search to 1024 nodes. I played relatively fast, and allowed myself no takebacks. The engine had no opening book, and blundered a piece on move #3...
[pgn]
[Date "2013.05.21"]
[White "lucas"]
[Black "DiscoCheck"]
[Result "1-0"]
[PlyCount "73"]
[TimeControl "inf"]

1. e4 Nf6 {+0.22/3 0.002s} 2. e5 {2.3s} Ne4 {+0.18/3 0.002s} 3. d4 {5.2s}
Na6 {+0.17/2 0.002s} 4. f3 {4.4s} d5 {-1.44/3 0.002s} 5. fxe4 {2.3s}
dxe4 {-2.01/4 0.002s} 6. Be3 {12s} Nb4 {-1.48/3 0.002s} 7. c3 {4.4s}
Nd5 {-1.02/3 0.001s} 8. Qd2 {14s} Nxe3 {-0.81/3 0.001s} 9. Qxe3 {1.8s}
Qd5 {-1.09/3 0.001s} 10. Nd2 {2.7s} Bf5 {-1.42/3 0.001s} 11. Bc4 {5.8s}
Qc6 {-2.05/3 0.001s} 12. Ne2 {13s} Qh6 {-2.22/3 0.002s} 13. Qxh6 {4.5s}
gxh6 {-3.24/4 0.001s} 14. Ng3 {11s} e6 {-3.23/3 0.001s} 15. Nxf5 {1.6s}
exf5 {-2.88/4 0.002s} 16. O-O {5.7s} e3 {-3.26/4 0.002s} 17. Nb3 {14s}
Rg8 {-3.26/2 0.002s} 18. Rae1 {5.0s} Be7 {-3.90/2 0.002s} 19. Rxf5 {4.3s}
Rg7 {-4.33/3 0.002s} 20. Rxe3 {5.2s} Rd8 {-4.51/2 0.002s} 21. Ref3 {2.9s}
Kd7 {-5.17/3 0.002s} 22. Rxf7 {6.1s} Rxf7 {-5.57/4 0.002s} 23. e6+ {11s}
Kd6 {-5.45/3 0.001s} 24. exf7 {1.9s} Rf8 {-5.38/3 0.001s} 25. Nc5 {20s}
Bh4 {-5.28/2 0.001s} 26. Ne6 {15s} Be7 {-6.62/3 0.001s} 27. Nxf8 {2.6s}
Bxf8 {-6.99/4 0.001s} 28. Re3 {12s} b5 {-6.91/2 0.001s} 29. Bb3 {4.0s}
Kc6 {-6.83/2 0.002s} 30. Re8 {2.8s} Bg7 {-9.61/3 0.002s} 31. Rg8 {13s}
Kd6 {-13.34/5 0.002s} 32. Rxg7 {1.6s} Ke7 {-21.16/3 0.001s} 33. Rg8 {9.8s}
Kd7 {-22.13/3 0.002s} 34. f8=Q {2.6s} Kc6 {-M6/3 0.001s} 35. Qc5+ {7.9s}
Kd7 {-M4/8 0.001s} 36. Rg7+ {4.4s} Ke8 {-M2/52 0.002s}
37. Qe7# {2.8s, White mates} 1-0
[/pgn]
So, at this very weak level, it can easily be beaten, even by a weak player like me (~1600 ELO).

Maybe 1024 nodes is too weak for you, so feel free to increase it, until the level suits you!

[lucasart]

Actually. I too had troubles, before figuring out that the download was under the label "Raw"... Blame me

Ah yes, that works too. Thank you for the useful tip. Now I can finally provide some direct download links:
* Linux: https://github.com/lucasart/chess/raw/m ... ocheck_4.2
* Windows: https://github.com/lucasart/chess/raw/m ... ck_4.2.exe

[noctiferus]

Glad to help.

BTW, I was running a blitz, and found a very bad endgame eval in a simple position (-3 instead of a short mate. Dunno if it reproducible and if can be useful:

[pgn]
[Event "discocheck tex"]
[Site "UTENTE-PC"]
[Date "2013.05.21"]
[Round "1"]
[White "DiscoCheck 4.2"]
[Black "Texel 1.01 JA 64-bit"]
[Result "0-1"]
[ECO "E09"]
[WhiteElo "2800"]
[BlackElo "2800"]
[Annotator "-0.13;0.19"]
[PlyCount "134"]
[TimeControl "60+2"]

{Intel(R) Core(TM) i7 CPU Q 720 @ 1.60GHz 1600 MHz W=23.9 ply; 626kN/s;
HS-Variety.ctg B=16.0 ply; 1.236kN/s; HS-Variety.ctg} 1. Nf3 {
[%eval 0,0] [%emt 0:00:00]} Nf6 {[%eval 0,0] [%emt 0:00:00]} 2. d4 {
[%eval 0,0] [%emt 0:00:00]} d5 {[%eval 0,0] [%emt 0:00:00]} 3. c4 {
[%eval 0,0] [%emt 0:00:00]} e6 {[%eval 0,0] [%emt 0:00:00]} 4. g3 {
[%eval 0,0] [%emt 0:00:00]} c6 {[%eval 0,0] [%emt 0:00:00]} 5. Nbd2 {
[%eval 0,0] [%emt 0:00:00]} Nbd7 {[%eval 0,0] [%emt 0:00:00]} 6. Bg2 {
[%eval 0,0] [%emt 0:00:00]} Be7 {[%eval 0,0] [%emt 0:00:00]} 7. O-O {
[%eval 0,0] [%emt 0:00:00]} b6 {[%eval 0,0] [%emt 0:00:00]} 8. Qc2 {
[%eval 0,0] [%emt 0:00:00]} Bb7 {[%eval 0,0] [%emt 0:00:00]} 9. e4 {
[%eval 0,0] [%emt 0:00:00]} O-O {[%eval 0,0] [%emt 0:00:00]} 10. Rd1 {
[%eval 0,0] [%emt 0:00:00]} c5 {Both last book move [%eval 0,0] [%emt 0:00:00]}
11. cxd5 {[%eval -13,15] [%emt 0:00:05]} exd5 {[%eval 19,14] [%emt 0:00:03]}
12. e5 {[%eval -13,15] [%emt 0:00:05]} Ne8 {(Nxe5) [%eval 40,14] [%emt 0:00:04]
} 13. Nb3 {(Bh3) [%eval 15,15] [%emt 0:00:05]} a5 {
(Nc7) [%eval 39,14] [%emt 0:00:03]} 14. Bd2 {(a4) [%eval -3,14] [%emt 0:00:05]}
a4 {[%eval 16,14] [%emt 0:00:03]} 15. Nc1 {[%eval 4,16] [%emt 0:00:04]} Nc7 {
(a3) [%eval 3,13] [%emt 0:00:03]} 16. Ne2 {[%eval -4,14] [%emt 0:00:04]} f6 {
(cxd4) [%eval 17,12] [%emt 0:00:04]} 17. Rac1 {
(Be3) [%eval 18,13] [%emt 0:00:04]} fxe5 {(Kh8) [%eval 18,13] [%emt 0:00:03]}
18. dxe5 {(Nxe5) [%eval 23,13] [%emt 0:00:02]} g5 {
(d4) [%eval 6,13] [%emt 0:00:04]} 19. g4 {[%eval 28,14] [%emt 0:00:04]} Ne6 {
[%eval 0,14] [%emt 0:00:03]} 20. Re1 {[%eval -5,15] [%emt 0:00:04]} c4 {
[%eval -9,14] [%emt 0:00:03]} 21. Ned4 {(Nfd4) [%eval -4,15] [%emt 0:00:05]}
Ndc5 {[%eval -3,13] [%emt 0:00:03]} 22. Nxe6 {
(Rcd1) [%eval -4,14] [%emt 0:00:06]} Nxe6 {[%eval -52,14] [%emt 0:00:03]} 23.
Red1 {(Qd1) [%eval -31,13] [%emt 0:00:04]} Nf4 {[%eval -79,14] [%emt 0:00:04]}
24. Bxf4 {[%eval -52,15] [%emt 0:00:04]} Rxf4 {[%eval -54,16] [%emt 0:00:03]}
25. h3 {[%eval -74,14] [%emt 0:00:04]} Bc5 {(Qf8) [%eval -78,14] [%emt 0:00:06]
} 26. Rd2 {(a3) [%eval -67,12] [%emt 0:00:04]} Qe7 {
(a3) [%eval -106,14] [%emt 0:00:03]} 27. Qd1 {[%eval -88,13] [%emt 0:00:04]}
Kh8 {[%eval -147,14] [%emt 0:00:10]} 28. Rxd5 {
(a3) [%eval -109,14] [%emt 0:00:05]} Bxd5 {[%eval -211,14] [%emt 0:00:03]} 29.
Qxd5 {[%eval -118,16] [%emt 0:00:05]} Qd8 {(Rd8) [%eval -196,15] [%emt 0:00:03]
} 30. Qxd8+ {[%eval -130,15] [%emt 0:00:03]} Rxd8 {
[%eval -192,17] [%emt 0:00:03]} 31. Rc2 {[%eval -151,15] [%emt 0:00:04]} h6 {
[%eval -235,16] [%emt 0:00:03]} 32. Kh2 {[%eval -147,15] [%emt 0:00:03]} a3 {
[%eval -260,18] [%emt 0:00:03]} 33. bxa3 {[%eval -156,17] [%emt 0:00:03]} Rd3 {
[%eval -280,18] [%emt 0:00:03]} 34. Kg3 {[%eval -171,16] [%emt 0:00:03]} c3 {
[%eval -301,18] [%emt 0:00:03]} 35. h4 {(Re2) [%eval -178,16] [%emt 0:00:04]}
Bd4 {(Bxa3) [%eval -360,16] [%emt 0:00:03]} 36. Re2 {
[%eval -280,16] [%emt 0:00:03]} c2 {[%eval -466,16] [%emt 0:00:03]} 37. Rxc2 {
[%eval -281,16] [%emt 0:00:02]} Bxe5 {[%eval -499,17] [%emt 0:00:03]} 38. hxg5
{[%eval -314,15] [%emt 0:00:02]} Rfxf3+ {[%eval -500,16] [%emt 0:00:03]} 39.
Kh4 {[%eval 1,0] [%emt 0:00:00]} hxg5+ {[%eval -510,16] [%emt 0:00:03]} 40.
Kxg5 {[%eval -320,14] [%emt 0:00:02]} Rf8 {[%eval -508,16] [%emt 0:00:02]} 41.
Rc6 {[%eval -316,15] [%emt 0:00:03]} Rd4 {(Rxf2) [%eval -515,14] [%emt 0:00:04]
} 42. Rxb6 {[%eval -231,12] [%emt 0:00:02]} Rxf2 {
(Bf4+) [%eval -637,15] [%emt 0:00:02]} 43. Re6 {[%eval -258,13] [%emt 0:00:04]}
Rxg2 {[%eval -766,16] [%emt 0:00:02]} 44. Rxe5 {[%eval -271,13] [%emt 0:00:02]}
Rdxg4+ {[%eval -900,15] [%emt 0:00:02]} 45. Kf6 {[%eval -280,15] [%emt 0:00:03]
} Ra4 {(Rg6+) [%eval -902,13] [%emt 0:00:02]} 46. Re8+ {
(Re7) [%eval -287,15] [%emt 0:00:03]} Kh7 {[%eval -908,15] [%emt 0:00:02]} 47.
Re7+ {(Re3) [%eval -291,15] [%emt 0:00:03]} Kh6 {[%eval -916,15] [%emt 0:00:02]
} 48. Re3 {[%eval -291,16] [%emt 0:00:03]} Ra6+ {[%eval -924,15] [%emt 0:00:02]
} 49. Ke5 {[%eval -296,16] [%emt 0:00:03]} Rg5+ {
(Ra5+) [%eval -940,18] [%emt 0:00:02]} 50. Ke4 {
(Kf4) [%eval -299,19] [%emt 0:00:03]} Re6+ {[%eval -955,18] [%emt 0:00:03]} 51.
Kd3 {[%eval -301,18] [%emt 0:00:03]} Rxe3+ {[%eval -948,17] [%emt 0:00:03]} 52.
Kxe3 {[%eval -303,18] [%emt 0:00:01]} Rg3+ {[%eval -956,19] [%emt 0:00:02]} 53.
Ke2 {(Kd4) [%eval -306,18] [%emt 0:00:03]} Rxa3 {[%eval -956,20] [%emt 0:00:02]
} 54. Kd2 {[%eval -314,19] [%emt 0:00:03]} Rxa2+ {
[%eval -964,21] [%emt 0:00:02]} 55. Ke3 {(Kd3) [%eval -321,21] [%emt 0:00:03]}
Kg5 {[%eval -32688,22] [%emt 0:00:02]} 56. Kf3 {
(Kd3) [%eval -337,17] [%emt 0:00:03]} Kf5 {
(Ra3+) [%eval -32754,25] [%emt 0:00:02]} 57. Ke3 {
[%eval -339,17] [%emt 0:00:03]} Ra4 {(Ra3+) [%eval -32755,22] [%emt 0:00:02]}
58. Kd3 {[%eval -338,18] [%emt 0:00:03]} Kf4 {
(Kg4) [%eval -32756,21] [%emt 0:00:02]} 59. Kc2 {
(Kd2) [%eval -34745,17] [%emt 0:00:03]} Ke3 {[%eval -32758,18] [%emt 0:00:01]}
60. Kb3 {[%eval -34743,19] [%emt 0:00:03]} Rg4 {
(Re4) [%eval -32759,16] [%emt 0:00:00]} 61. Kc3 {
[%eval -34745,18] [%emt 0:00:03]} Ke2 {(Rf4) [%eval -32760,14] [%emt 0:00:00]}
62. Kc2 {[%eval -34753,19] [%emt 0:00:03]} Rg3 {
[%eval -32761,12] [%emt 0:00:00]} 63. Kc1 {[%eval -34757,81] [%emt 0:00:03]}
Kd3 {[%eval -32762,10] [%emt 0:00:00]} 64. Kb2 {
[%eval -34759,127] [%emt 0:00:00]} Kd2 {[%eval -32763,8] [%emt 0:00:00]} 65.
Kb1 {[%eval -34761,127] [%emt 0:00:00]} Kc3 {[%eval -32764,6] [%emt 0:00:00]}
66. Ka2 {[%eval -34763,127] [%emt 0:00:00]} Rg1 {
(Kc2) [%eval -32765,4] [%emt 0:00:00]} 67. Ka3 {[%eval 1,0] [%emt 0:00:00]}
Ra1# {[%eval -32766,2] [%emt 0:00:00]} 0-1 [/pgn]

[lucasart]

Glad to help.

BTW, I was running a blitz, and found a very bad endgame eval in a simple position (-3 instead of a short mate. Dunno if it reproducible and if can be useful:

[pgn]
[Event "discocheck tex"]
[Site "UTENTE-PC"]
[Date "2013.05.21"]
[Round "1"]
[White "DiscoCheck 4.2"]
[Black "Texel 1.01 JA 64-bit"]
[Result "0-1"]
[ECO "E09"]
[WhiteElo "2800"]
[BlackElo "2800"]
[Annotator "-0.13;0.19"]
[PlyCount "134"]
[TimeControl "60+2"]

{Intel(R) Core(TM) i7 CPU Q 720 @ 1.60GHz 1600 MHz W=23.9 ply; 626kN/s;
HS-Variety.ctg B=16.0 ply; 1.236kN/s; HS-Variety.ctg} 1. Nf3 {
[%eval 0,0] [%emt 0:00:00]} Nf6 {[%eval 0,0] [%emt 0:00:00]} 2. d4 {
[%eval 0,0] [%emt 0:00:00]} d5 {[%eval 0,0] [%emt 0:00:00]} 3. c4 {
[%eval 0,0] [%emt 0:00:00]} e6 {[%eval 0,0] [%emt 0:00:00]} 4. g3 {
[%eval 0,0] [%emt 0:00:00]} c6 {[%eval 0,0] [%emt 0:00:00]} 5. Nbd2 {
[%eval 0,0] [%emt 0:00:00]} Nbd7 {[%eval 0,0] [%emt 0:00:00]} 6. Bg2 {
[%eval 0,0] [%emt 0:00:00]} Be7 {[%eval 0,0] [%emt 0:00:00]} 7. O-O {
[%eval 0,0] [%emt 0:00:00]} b6 {[%eval 0,0] [%emt 0:00:00]} 8. Qc2 {
[%eval 0,0] [%emt 0:00:00]} Bb7 {[%eval 0,0] [%emt 0:00:00]} 9. e4 {
[%eval 0,0] [%emt 0:00:00]} O-O {[%eval 0,0] [%emt 0:00:00]} 10. Rd1 {
[%eval 0,0] [%emt 0:00:00]} c5 {Both last book move [%eval 0,0] [%emt 0:00:00]}
11. cxd5 {[%eval -13,15] [%emt 0:00:05]} exd5 {[%eval 19,14] [%emt 0:00:03]}
12. e5 {[%eval -13,15] [%emt 0:00:05]} Ne8 {(Nxe5) [%eval 40,14] [%emt 0:00:04]
} 13. Nb3 {(Bh3) [%eval 15,15] [%emt 0:00:05]} a5 {
(Nc7) [%eval 39,14] [%emt 0:00:03]} 14. Bd2 {(a4) [%eval -3,14] [%emt 0:00:05]}
a4 {[%eval 16,14] [%emt 0:00:03]} 15. Nc1 {[%eval 4,16] [%emt 0:00:04]} Nc7 {
(a3) [%eval 3,13] [%emt 0:00:03]} 16. Ne2 {[%eval -4,14] [%emt 0:00:04]} f6 {
(cxd4) [%eval 17,12] [%emt 0:00:04]} 17. Rac1 {
(Be3) [%eval 18,13] [%emt 0:00:04]} fxe5 {(Kh8) [%eval 18,13] [%emt 0:00:03]}
18. dxe5 {(Nxe5) [%eval 23,13] [%emt 0:00:02]} g5 {
(d4) [%eval 6,13] [%emt 0:00:04]} 19. g4 {[%eval 28,14] [%emt 0:00:04]} Ne6 {
[%eval 0,14] [%emt 0:00:03]} 20. Re1 {[%eval -5,15] [%emt 0:00:04]} c4 {
[%eval -9,14] [%emt 0:00:03]} 21. Ned4 {(Nfd4) [%eval -4,15] [%emt 0:00:05]}
Ndc5 {[%eval -3,13] [%emt 0:00:03]} 22. Nxe6 {
(Rcd1) [%eval -4,14] [%emt 0:00:06]} Nxe6 {[%eval -52,14] [%emt 0:00:03]} 23.
Red1 {(Qd1) [%eval -31,13] [%emt 0:00:04]} Nf4 {[%eval -79,14] [%emt 0:00:04]}
24. Bxf4 {[%eval -52,15] [%emt 0:00:04]} Rxf4 {[%eval -54,16] [%emt 0:00:03]}
25. h3 {[%eval -74,14] [%emt 0:00:04]} Bc5 {(Qf8) [%eval -78,14] [%emt 0:00:06]
} 26. Rd2 {(a3) [%eval -67,12] [%emt 0:00:04]} Qe7 {
(a3) [%eval -106,14] [%emt 0:00:03]} 27. Qd1 {[%eval -88,13] [%emt 0:00:04]}
Kh8 {[%eval -147,14] [%emt 0:00:10]} 28. Rxd5 {
(a3) [%eval -109,14] [%emt 0:00:05]} Bxd5 {[%eval -211,14] [%emt 0:00:03]} 29.
Qxd5 {[%eval -118,16] [%emt 0:00:05]} Qd8 {(Rd8) [%eval -196,15] [%emt 0:00:03]
} 30. Qxd8+ {[%eval -130,15] [%emt 0:00:03]} Rxd8 {
[%eval -192,17] [%emt 0:00:03]} 31. Rc2 {[%eval -151,15] [%emt 0:00:04]} h6 {
[%eval -235,16] [%emt 0:00:03]} 32. Kh2 {[%eval -147,15] [%emt 0:00:03]} a3 {
[%eval -260,18] [%emt 0:00:03]} 33. bxa3 {[%eval -156,17] [%emt 0:00:03]} Rd3 {
[%eval -280,18] [%emt 0:00:03]} 34. Kg3 {[%eval -171,16] [%emt 0:00:03]} c3 {
[%eval -301,18] [%emt 0:00:03]} 35. h4 {(Re2) [%eval -178,16] [%emt 0:00:04]}
Bd4 {(Bxa3) [%eval -360,16] [%emt 0:00:03]} 36. Re2 {
[%eval -280,16] [%emt 0:00:03]} c2 {[%eval -466,16] [%emt 0:00:03]} 37. Rxc2 {
[%eval -281,16] [%emt 0:00:02]} Bxe5 {[%eval -499,17] [%emt 0:00:03]} 38. hxg5
{[%eval -314,15] [%emt 0:00:02]} Rfxf3+ {[%eval -500,16] [%emt 0:00:03]} 39.
Kh4 {[%eval 1,0] [%emt 0:00:00]} hxg5+ {[%eval -510,16] [%emt 0:00:03]} 40.
Kxg5 {[%eval -320,14] [%emt 0:00:02]} Rf8 {[%eval -508,16] [%emt 0:00:02]} 41.
Rc6 {[%eval -316,15] [%emt 0:00:03]} Rd4 {(Rxf2) [%eval -515,14] [%emt 0:00:04]
} 42. Rxb6 {[%eval -231,12] [%emt 0:00:02]} Rxf2 {
(Bf4+) [%eval -637,15] [%emt 0:00:02]} 43. Re6 {[%eval -258,13] [%emt 0:00:04]}
Rxg2 {[%eval -766,16] [%emt 0:00:02]} 44. Rxe5 {[%eval -271,13] [%emt 0:00:02]}
Rdxg4+ {[%eval -900,15] [%emt 0:00:02]} 45. Kf6 {[%eval -280,15] [%emt 0:00:03]
} Ra4 {(Rg6+) [%eval -902,13] [%emt 0:00:02]} 46. Re8+ {
(Re7) [%eval -287,15] [%emt 0:00:03]} Kh7 {[%eval -908,15] [%emt 0:00:02]} 47.
Re7+ {(Re3) [%eval -291,15] [%emt 0:00:03]} Kh6 {[%eval -916,15] [%emt 0:00:02]
} 48. Re3 {[%eval -291,16] [%emt 0:00:03]} Ra6+ {[%eval -924,15] [%emt 0:00:02]
} 49. Ke5 {[%eval -296,16] [%emt 0:00:03]} Rg5+ {
(Ra5+) [%eval -940,18] [%emt 0:00:02]} 50. Ke4 {
(Kf4) [%eval -299,19] [%emt 0:00:03]} Re6+ {[%eval -955,18] [%emt 0:00:03]} 51.
Kd3 {[%eval -301,18] [%emt 0:00:03]} Rxe3+ {[%eval -948,17] [%emt 0:00:03]} 52.
Kxe3 {[%eval -303,18] [%emt 0:00:01]} Rg3+ {[%eval -956,19] [%emt 0:00:02]} 53.
Ke2 {(Kd4) [%eval -306,18] [%emt 0:00:03]} Rxa3 {[%eval -956,20] [%emt 0:00:02]
} 54. Kd2 {[%eval -314,19] [%emt 0:00:03]} Rxa2+ {
[%eval -964,21] [%emt 0:00:02]} 55. Ke3 {(Kd3) [%eval -321,21] [%emt 0:00:03]}
Kg5 {[%eval -32688,22] [%emt 0:00:02]} 56. Kf3 {
(Kd3) [%eval -337,17] [%emt 0:00:03]} Kf5 {
(Ra3+) [%eval -32754,25] [%emt 0:00:02]} 57. Ke3 {
[%eval -339,17] [%emt 0:00:03]} Ra4 {(Ra3+) [%eval -32755,22] [%emt 0:00:02]}
58. Kd3 {[%eval -338,18] [%emt 0:00:03]} Kf4 {
(Kg4) [%eval -32756,21] [%emt 0:00:02]} 59. Kc2 {
(Kd2) [%eval -34745,17] [%emt 0:00:03]} Ke3 {[%eval -32758,18] [%emt 0:00:01]}
60. Kb3 {[%eval -34743,19] [%emt 0:00:03]} Rg4 {
(Re4) [%eval -32759,16] [%emt 0:00:00]} 61. Kc3 {
[%eval -34745,18] [%emt 0:00:03]} Ke2 {(Rf4) [%eval -32760,14] [%emt 0:00:00]}
62. Kc2 {[%eval -34753,19] [%emt 0:00:03]} Rg3 {
[%eval -32761,12] [%emt 0:00:00]} 63. Kc1 {[%eval -34757,81] [%emt 0:00:03]}
Kd3 {[%eval -32762,10] [%emt 0:00:00]} 64. Kb2 {
[%eval -34759,127] [%emt 0:00:00]} Kd2 {[%eval -32763,8] [%emt 0:00:00]} 65.
Kb1 {[%eval -34761,127] [%emt 0:00:00]} Kc3 {[%eval -32764,6] [%emt 0:00:00]}
66. Ka2 {[%eval -34763,127] [%emt 0:00:00]} Rg1 {
(Kc2) [%eval -32765,4] [%emt 0:00:00]} 67. Ka3 {[%eval 1,0] [%emt 0:00:00]}
Ra1# {[%eval -32766,2] [%emt 0:00:00]} 0-1 [/pgn]

Interesting. I suppose you are refering to 55. Ke3, right ?
The short answer is: it's not a bug, it's a feature

In a KRK endgame like this, DiscoCheck will evaluate the materual difference to 1 rook for texel, plus some other things like mobility, king safety etc. But it will also see that black has no pawns, so it will half the eval. This is a general rule that works quite well as far as game results is concerned. It may result in some strange evals in these endgames, but that's inconsequential as the result of the game is the same regardless of what score is displayed.

[noctiferus]

Thx for explanation.
I'm unable to reproduce the game, and, moreover, cannot see how to post here the evals, if you were interested. In this case, tell me if you want to see them via PM.
ciao
e