Question to CCRL - Dumb 1.9 evals etc.

[Guenther]

While being curious for finding traces of a bug discovered in Blunder recently I downloaded all commented games
from the 40/15 list for Blunder 7.6.0 (8.0.0 wasn't there yet) and noticed some glitch with the Dumb 1.9 games in it.

I don't know if it is the used GUI or the PGN converting script (it is obvious a script is used because spurious
[%eval remnants remain there if they contain mate symbols), but in only 16 Dumb 1.9 games there are
297 entries of a 310.00 eval which makes no sense? See game below.
A little test in Cute/WB shows normal evals for the positions when analysing, so I don't know where they
come from?
Regrettably this breaks some pgn tools/scripts.

BTW you should be also able to fix the spurios %eval pgn output in your script.
They look like this:

Code: Select all

{(Qd2) -12.92/18 100s} 45. Rxg7+ {(Rxd2) +319.78/15 0s} Kxg7 {[%eval -#7,17]
87s} 46. Rg1+ {(Rg1) +319.78/15 0s} Kh8 {[%eval -#6,16] 50s} 47. Ng6+
{+319.78/15 0s} Kg7 {[%eval -#5,16] 16s} 1-0

[pgn][Event "CCRL 40/15"]
[Site "CCRL"]
[Date "2022.03.29"]
[Round "817.3.261"]
[White "Dumb 1.9 64-bit"]
[Black "Blunder 7.6.0 64-bit"]
[Result "1-0"]
[ECO "D37"]
[Opening "QGD"]
[Variation "classical variation (5.Bf4)"]
[PlyCount "93"]
[WhiteElo "2724"]
[BlackElo "2636"]

1. Nf3 {+0.00/1 0s} d5 {+0.00/1 0s} 2. d4 {+0.00/1 0s} Nf6 {+0.00/1 0s} 3. c4
{+0.00/1 0s} e6 {+0.00/1 0s} 4. Nc3 {+0.00/1 0s} Be7 {+0.00/1 0s} 5. Bf4
{+0.00/1 0s} O-O {+0.00/1 0s} 6. e3 {+0.00/1 0s} c5 {+0.00/1 0s} 7. dxc5
{+0.00/1 0s} Bxc5 {+0.00/1 0s} 8. Qc2 {+0.00/1 0s} Nc6 {+0.00/1 0s} 9. a3
{+0.24/22 17s} Qe7 {(Bd6) -0.33/16 26s} 10. O-O-O {(O-O-O) +310.00/25 19s} Rd8
{-0.80/18 34s} 11. Bg5 {(Bg5) +310.00/26 19s} dxc4 {-0.64/18 25s} 12. Bxc4
{(Bxc4) +310.00/23 20s} h6 {-1.07/15 33s} 13. Rxd8+ {(Rxd8) +0.45/24 17s} Nxd8
{(Qxd8) -0.78/20 25s} 14. Bh4 {(Bxf6) +0.46/25 20s} Bd7 {-0.88/19 24s} 15. Ne4
{(Ne4) +0.60/26 14s} g5 {(g5) -0.44/19 18s} 16. Nfxg5 {(Nfxg5) +0.24/26 20s}
Nxe4 {(Nxe4) -0.74/20 24s} 17. Nxe4 {(Nxe4) +0.20/29 18s} Qxh4 {(Qxh4) -0.54/21
31s} 18. Nxc5 {(Nxc5) +0.23/28 15s} Bc6 {(Bc6) -0.48/20 18s} 19. Rd1 {(Rd1)
+310.00/26 20s} Qxh2 {-0.44/19 18s} 20. Ne4 {+0.46/22 17s} Qxg2 {(Qxg2)
-0.25/20 23s} 21. Nf6+ {(Nf6) +0.42/23 16s} Kf8 {(Kg7) -0.25/19 17s} 22. Bf1
{(Qc3) +0.38/25 19s} Qf3 {(Qg6) -0.25/17 17s} 23. Qc5+ {(Qc3) +310.00/23 21s}
Kg7 {-1.07/17 17s} 24. Nh5+ {(Nh5) +310.00/25 21s} Kg8 {-2.85/21 37s} 25. Qe5
{(Qe5) +2.14/25 15s} Kf8 {(Kf8) -2.72/21 33s} 26. Rd4 {+310.00/25 21s} Rc8
{-2.89/21 26s} 27. Qd6+ {(Qd6) +2.22/25 22s} Kg8 {-3.09/23 14s} 28. Rg4+ {(Rg4)
+2.46/25 21s} Qxg4 {(Qxg4) -2.97/24 14s} 29. Nf6+ {(Nf6) +2.31/27 16s} Kg7
{(Kg7) -3.00/24 14s} 30. Nxg4 {(Nxg4) +310.00/27 22s} Bb5+ {-3.09/23 20s} 31.
Kd2 {(Kd2) +2.34/26 23s} Bxf1 {-3.09/24 17s} 32. Qf4 {(Qd7) +310.00/26 23s} Kf8
{-3.07/20 15s} 33. Qxh6+ {(Qxh6) +310.00/25 23s} Ke8 {-3.17/21 13s} 34. Qh8+
{(Qh8) +310.00/25 23s} Ke7 {-3.10/21 13s} 35. Qf6+ {+310.00/25 23s} Ke8
{-3.15/21 13s} 36. Qd4 {(Qd4) +3.13/24 23s} Kf8 {(Ra8) -3.15/22 8s} 37. Qd7
{(Qd6) +4.36/26 18s} Ra8 {(Ra8) -3.66/18 7s} 38. Nf6 {(Nf6) +310.00/26 25s} Nc6
{-3.82/19 6s} 39. Qxb7 {(Nh7) +4.44/26 22s} Rd8+ {(Rd8) -5.44/16 5s} 40. Kc1
{(Kc1) +310.00/26 30s} Ne7 {-5.74/19 4s} 41. Qh1 {(Qh1) +310.00/25 15s} Nd5
{-5.92/22 38s} 42. Nxd5 {(Nxd5) +4.63/26 14s} Rxd5 {(Rxd5) -5.99/21 22s} 43.
Qxf1 {(Qxf1) +4.63/26 13s} Ke7 {(a5) -6.13/21 28s} 44. e4 {(Qa6) +4.98/25 12s}
Rd7 {(Rd7) -6.04/19 21s} 45. Qb5 {(Qb5) +310.00/26 16s} Rc7+ {-6.30/21 21s} 46.
Kd2 {(Kd2) +5.27/27 14s} f6 {(f6) -6.42/21 16s} 47. Ke3 {(e5) +310.00/28 16s}
1-0[/pgn]

[Modern Times]

That comes from a ChessGUI game which outputs pgns in a chessbase format. Seems to be an engine issue, Graham can advise.


[Event "Dumb Gauntlet"]
[Site "ChessGUI11"]
[Date "2022.03.29"]
[Round "1.9"]
[White "Dumb 1.9 64-bit"]
[Black "Blunder 7.6.0 64-bit"]
[Result "1-0"]
[Time "5:14:40 pm"]
[ECO "A06"]
[Opening "A06"]
[TimeControl "40/660:40/660:40/660"]
[PlyCount "93"]
[Number "9"]
[Termination "GUI adjudication"]
[WhiteType "program"]
[BlackType "program"]
[Variant "normal"]

{ AMD Ryzen 9 5950X 16-Core Processor 3.40 GHz }
1.Nf3 {[%eval 0,1] [%emt 00:00:00]} d5 {[%eval 0,1] [%emt 00:00:00]}
2.d4 {[%eval 0,1] [%emt 00:00:00]} Nf6 {[%eval 0,1] [%emt 00:00:00]}
3.c4 {[%eval 0,1] [%emt 00:00:00]} e6 {[%eval 0,1] [%emt 00:00:00]}
4.Nc3 {[%eval 0,1] [%emt 00:00:00]} Be7 {[%eval 0,1] [%emt 00:00:00]}
5.Bf4 {[%eval 0,1] [%emt 00:00:00]} O-O {[%eval 0,1] [%emt 00:00:00]}
6.e3 {[%eval 0,1] [%emt 00:00:00]} c5 {[%eval 0,1] [%emt 00:00:00]}
7.dxc5 {[%eval 0,1] [%emt 00:00:00]} Bxc5 {[%eval 0,1] [%emt 00:00:00]}
8.Qc2 {[%eval 0,1] [%emt 00:00:00]} Nc6 {[%eval 0,1] [%emt 00:00:00]}
9.a3 {[%eval 24,22] [%emt 00:00:17]} Qe7 {(Bd6) [%eval -33,16] [%emt 00:00:26]}
10.O-O-O {(O-O-O) [%eval 31000,25] [%emt 00:00:19]} Rd8 {[%eval -80,18] [%emt 00:00:34]}
11.Bg5 {(Bg5) [%eval 31000,26] [%emt 00:00:19]} dxc4 {[%eval -64,18] [%emt 00:00:25]}
12.Bxc4 {(Bxc4) [%eval 31000,23] [%emt 00:00:20]} h6 {[%eval -107,15] [%emt 00:00:33]}
13.Rxd8 {(Rxd8) [%eval 45,24] [%emt 00:00:17]} Nxd8 {(Qxd8) [%eval -78,20] [%emt 00:00:25]}
14.Bh4 {(Bxf6) [%eval 46,25] [%emt 00:00:20]} Bd7 {[%eval -88,19] [%emt 00:00:24]}
15.Ne4 {(Ne4) [%eval 60,26] [%emt 00:00:14]} g5 {(g5) [%eval -44,19] [%emt 00:00:18]}
16.Nfxg5 {(Nfxg5) [%eval 24,26] [%emt 00:00:20]} Nxe4 {(Nxe4) [%eval -74,20] [%emt 00:00:24]}
17.Nxe4 {(Nxe4) [%eval 20,29] [%emt 00:00:18]} Qxh4 {(Qxh4) [%eval -54,21] [%emt 00:00:31]}
18.Nxc5 {(Nxc5) [%eval 23,28] [%emt 00:00:15]} Bc6 {(Bc6) [%eval -48,20] [%emt 00:00:18]}
19.Rd1 {(Rd1) [%eval 31000,26] [%emt 00:00:20]} Qxh2 {[%eval -44,19] [%emt 00:00:18]}
20.Ne4 {[%eval 46,22] [%emt 00:00:17]} Qxg2 {(Qxg2) [%eval -25,20] [%emt 00:00:23]}
21.Nf6 {(Nf6) [%eval 42,23] [%emt 00:00:16]} Kf8 {(Kg7) [%eval -25,19] [%emt 00:00:17]}
22.Bf1 {(Qc3) [%eval 38,25] [%emt 00:00:19]} Qf3 {(Qg6) [%eval -25,17] [%emt 00:00:17]}
23.Qc5 {(Qc3) [%eval 31000,23] [%emt 00:00:21]} Kg7 {[%eval -107,17] [%emt 00:00:17]}
24.Nh5 {(Nh5) [%eval 31000,25] [%emt 00:00:21]} Kg8 {[%eval -285,21] [%emt 00:00:37]}
25.Qe5 {(Qe5) [%eval 214,25] [%emt 00:00:15]} Kf8 {(Kf8) [%eval -272,21] [%emt 00:00:33]}
26.Rd4 {[%eval 31000,25] [%emt 00:00:21]} Rc8 {[%eval -289,21] [%emt 00:00:26]}
27.Qd6 {(Qd6) [%eval 222,25] [%emt 00:00:22]} Kg8 {[%eval -309,23] [%emt 00:00:14]}
28.Rg4 {(Rg4) [%eval 246,25] [%emt 00:00:21]} Qxg4 {(Qxg4) [%eval -297,24] [%emt 00:00:14]}
29.Nf6 {(Nf6) [%eval 231,27] [%emt 00:00:16]} Kg7 {(Kg7) [%eval -300,24] [%emt 00:00:14]}
30.Nxg4 {(Nxg4) [%eval 31000,27] [%emt 00:00:22]} Bb5 {[%eval -309,23] [%emt 00:00:20]}
31.Kd2 {(Kd2) [%eval 234,26] [%emt 00:00:23]} Bxf1 {[%eval -309,24] [%emt 00:00:17]}
32.Qf4 {(Qd7) [%eval 31000,26] [%emt 00:00:23]} Kf8 {[%eval -307,20] [%emt 00:00:15]}
33.Qxh6 {(Qxh6) [%eval 31000,25] [%emt 00:00:23]} Ke8 {[%eval -317,21] [%emt 00:00:13]}
34.Qh8 {(Qh8) [%eval 31000,25] [%emt 00:00:23]} Ke7 {[%eval -310,21] [%emt 00:00:13]}
35.Qf6 {[%eval 31000,25] [%emt 00:00:23]} Ke8 {[%eval -315,21] [%emt 00:00:13]}
36.Qd4 {(Qd4) [%eval 313,24] [%emt 00:00:23]} Kf8 {(Ra8) [%eval -315,22] [%emt 00:00:08]}
37.Qd7 {(Qd6) [%eval 436,26] [%emt 00:00:18]} Ra8 {(Ra8) [%eval -366,18] [%emt 00:00:07]}
38.Nf6 {(Nf6) [%eval 31000,26] [%emt 00:00:25]} Nc6 {[%eval -382,19] [%emt 00:00:06]}
39.Qxb7 {(Nh7) [%eval 444,26] [%emt 00:00:22]} Rd8 {(Rd8) [%eval -544,16] [%emt 00:00:05]}
40.Kc1 {(Kc1) [%eval 31000,26] [%emt 00:00:30]} Ne7 {[%eval -574,19] [%emt 00:00:04]}
41.Qh1 {(Qh1) [%eval 31000,25] [%emt 00:00:15]} Nd5 {[%eval -592,22] [%emt 00:00:38]}
42.Nxd5 {(Nxd5) [%eval 463,26] [%emt 00:00:14]} Rxd5 {(Rxd5) [%eval -599,21] [%emt 00:00:22]}
43.Qxf1 {(Qxf1) [%eval 463,26] [%emt 00:00:13]} Ke7 {(a5) [%eval -613,21] [%emt 00:00:28]}
44.e4 {(Qa6) [%eval 498,25] [%emt 00:00:12]} Rd7 {(Rd7) [%eval -604,19] [%emt 00:00:21]}
45.Qb5 {(Qb5) [%eval 31000,26] [%emt 00:00:16]} Rc7 {[%eval -630,21] [%emt 00:00:21]}
46.Kd2 {(Kd2) [%eval 527,27] [%emt 00:00:14]} f6 {(f6) [%eval -642,21] [%emt 00:00:16]}
47.Ke3 {(e5) [%eval 31000,28] [%emt 00:00:16]}
1-0

[Guenther]

I checked it now with debug window open in the CuteChessGUI from the position after move 10...Rd8 against itself.

[fen]r1br2k1/pp2qppp/2n1pn2/2bp4/2P2B2/P1N1PN2/1PQ2PPP/2KR1B1R w - - 3 11[/fen]


It is a mixture of an output bug and an unfortunate reaction of the ChessGUI.

Obviously quite often Dumb 1.9 reports a score of 'mate 0' (??) and ChessGUI translates it to 310.00,
while e.g. CuteChess translates it to 0.00, which makes a bit more sense, but even better would be
not to record any score, just the time then.
The best would be of course to fix it in Dumb :)

Code: Select all

<Dumb_19-64(4): info depth 24 score cp 28 nodes 78324011 time 42800 nps 1829999 pv f1d3 d5c4 d3c4 c8d7 f4c7 d8c8 c7g3 c5b6 c4d3 h7h6 c1b1 e6e5 d3e4 d7g4 h1f1 f6e4 c2e4 f7f5 e4c4 g8h7 c3d5 e7c5 c4c5 b6c5 f1e1 
<Dumb_19-64(4): info depth 25 score mate 0 nodes 138511863 time 74000 nps 1871774 pv 
<Dumb_19-64(4): bestmove f4g5

Code: Select all

1. Bg5 0.00/25 74.00s d4 -0.33/25 58.32s 
2. Ne4 +0.36/25 61.12s e5 0.00/27 73.29s 

I guess it was already there in Dumb 1.3!
But another GUI was used and translated it to 0.00 always.
There are lots of moves recorded with 0.00 when eval
should be completely different.

[pgn][Event "CCRL 40/15"]
[Site "CCRL"]
[Date "2019.05.16"]
[Round "668.3.180"]
[White "Dumb 1.3 64-bit"]
[Black "Absolute Zero 2.4.7.2 64-bit"]
[Result "1-0"]
[ECO "A17"]
[Opening "English opening"]
[PlyCount "122"]
[WhiteElo "2293"]
[BlackElo "2298"]

1. c4 {book} Nf6 {book} 2. Nc3 {book} e6 {book} 3. Nf3 {book} c5 {book} 4. g3
{book} b6 {book} 5. e4 {book} Bb7 {book} 6. d3 {book} Be7 {book} 7. Bh3
{+0.20/11 39s} Nc6 {-0.20/19 29s} 8. O-O {+0.19/11 31s} d6 {-0.19/21 23s} 9.
Be3 {+0.11/11 39s} O-O {-0.17/20 23s} 10. Qe2 {-0.02/11 39s} d5 {+0.17/21 40s}
11. e5 {-0.14/13 39s} d4 {+0.14/20 23s} 12. exf6 {-0.19/14 39s} Bxf6 {+0.14/21
23s} 13. Ne4 {-0.22/14 39s} dxe3 {+0.10/21 23s} 14. Nxf6+ {+0.00/14 39s} Qxf6
{+0.10/21 23s} 15. fxe3 {+0.00/13 39s} Rad8 {+0.13/20 23s} 16. Bg2 {-0.15/12
39s} Rd7 {+0.11/19 22s} 17. Rad1 {-0.07/12 39s} Qd8 {+0.10/19 22s} 18. a3
{-0.01/11 37s} Re8 {+0.11/20 21s} 19. g4 {+0.05/12 39s} f5 {+0.23/18 20s} 20.
g5 {+0.01/13 39s} Rf8 {+0.30/18 19s} 21. h4 {+0.00/13 39s} Kh8 {+0.25/18 20s}
22. b3 {+0.38/12 39s} a5 {+0.41/15 21s} 23. d4 {+0.28/12 39s} cxd4 {+0.75/20
36s} 24. exd4 {+0.00/13 39s} Nxd4 {+0.28/19 37s} 25. Nxd4 {+0.00/16 39s} Bxg2
{+0.23/20 19s} 26. Nxe6 {+0.79/16 39s} Bxf1 {-0.30/21 38s} 27. Nxd8 {+0.88/17
39s} Bxe2 {-0.30/23 19s} 28. Rxd7 {+0.00/18 39s} a4 {-0.44/22 21s} 29. Kf2
{+0.88/17 31s} Bg4 {-0.48/20 18s} 30. bxa4 {+0.00/17 40s} Re8 {-1.12/21 38s}
31. Nf7+ {+1.75/15 40s} Kg8 32. Nd6 {+1.74/16 40s} Re2+ {-1.13/19 18s} 33. Kg3
{+2.08/15 30s} g6 {-0.72/20 30s} 34. Kf4 {+0.00/16 41s} Re1 {-1.82/19 54s} 35.
Rb7 {+2.35/15 40s} Bd1 {-1.96/19 18s} 36. Rxb6 {+2.85/15 34s} Bxa4 {-2.27/18
23s} 37. c5 {+0.00/15 43s} Kg7 {-2.38/18 33s} 38. c6 {+3.59/15 43s} Rf1+
{-2.37/16 15s} 39. Ke5 {+3.89/14 33s} Re1+ {-2.82/18 27s} 40. Kd5 {+0.00/16
54s} Rc1 {-2.66/19 24s} 41. Nc4 {+4.08/15 33s} f4 {-3.77/20 128s} 42. Ne5
{+0.00/15 33s} f3 {-4.65/19 108s} 43. Nxf3 {+0.00/15 33s} Ra1 {-5.21/20 87s}
44. Rb7+ {+5.85/15 30s} Kg8 {-5.25/23 34s} 45. Ne5 {+5.74/15 33s} Rd1+
{-5.39/22 48s} 46. Kc5 {+5.90/16 26s} Rc1+ {-5.68/23 52s} 47. Kd6 {+6.00/17
33s} Rd1+ {-5.85/23 71s} 48. Ke7 {+0.00/18 33s} Bxc6 {-7.00/27 102s} 49. Nxc6
{+5.98/17 33s} Re1+ {-6.87/26 27s} 50. Kd7 {+0.00/17 33s} Ra1 {-7.06/23 62s}
51. Ne7+ {+6.73/17 33s} Kf8 {-7.17/24 27s} 52. Nd5 {+6.93/17 26s} Ra2 {-7.38/24
59s} 53. Ra7 {+7.07/17 34s} Rh2 {-7.45/20 34s} 54. Nf6 {+7.43/17 34s} Rd2+
{-7.63/22 36s} 55. Ke6 {+7.45/17 34s} Re2+ {-8.17/24 62s} 56. Kd5 {+0.00/18
34s} Rd2+ {-8.57/23 55s} 57. Ke4 {+7.69/17 24s} Re2+ {-8.96/22 49s} 58. Kf3
{+9.63/17 31s} Re6 {-9.97/25 67s} 59. Kf4 {+10.43/17 34s} Re7 {-16.37/22 86s}
60. Nxh7+ {+17.50/18 34s} Kf7 {-#26/26 75s} 61. Rxe7+ {+0.00/19 34s} Kxe7
{adjudication -M24/24 20s, White wins by adjudication} 1-0[/pgn]

Edit:
Filed an issue now at Dumbs github repo

[abulmo2]

Thanks for reporting this bug.

I just released Dumb 1.10 that should display only valid scores & pv. Play & strength should be the same as version 1.9.

https://github.com/abulmo/Dumb/releases/tag/v1.10