Patt-Bug, Junior 11.1a show mate and Ktulu 9.03 made patt!

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

Moderators: hgm, Harvey Williamson, bob

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Post Reply
Frank Quisinsky
Posts: 4833
Joined: Wed Nov 18, 2009 6:16 pm
Location: Trier, Germany
Contact:

Patt-Bug, Junior 11.1a show mate and Ktulu 9.03 made patt!

Post by Frank Quisinsky » Tue Mar 23, 2010 8:18 pm

Hi there,

perhaps a very nice test postion for programmers. I am sure that such a bug is in a lot of engines.

First TalkChess report about a Patt-Bug :-)

Now let us look in the position ...

Image

Junior 11.1a played 61. ... Rxd1
Very easy Ktulu 9.0x played 62. Qxg6 and the game ended with patt.
Interesting is that Junior show the mate moves before :-)

SWCR Game:

Code: Select all

[Event "ktulu-903"]
[Site "Trier-Feyen"]
[Date "2010.03.23"]
[Round "2.1"]
[White "Ktulu 9"]
[Black "Junior 11.1a"]
[Result "1/2-1/2"]
[ECO "A26"]
[Annotator "0.26;0.48"]
[PlyCount "124"]
[EventDate "2010.03.23"]
[EventType "simul"]
[Source "Quisinsky"]
[TimeControl "40/600:40/600:40/600"]

{Intel(R) Core(TM)2 CPU          6700  @ 2.66GHz 3000 MHz  W=14.9 plies; 2.
291kN/s; Strong-2010.ctg  B=17.6 plies; 2.295kN/s; Strong-2010.ctg} 1. c4 {
[%eval 0,0] [%emt 0:00:00]} e5 {[%eval 0,0] [%emt 0:00:00]} 2. Nc3 {[%eval 0,0]
[%emt 0:00:00]} Nc6 {[%eval 0,0] [%emt 0:00:00]} 3. g3 {[%eval 0,0] [%emt 0:00:
00]} g6 {[%eval 0,0] [%emt 0:00:00]} 4. Rb1 {[%eval 0,0] [%emt 0:00:00]} a5 {
[%eval 0,0] [%emt 0:00:00]} 5. a3 {[%eval 0,0] [%emt 0:00:00]} f5 {[%eval 0,0]
[%emt 0:00:00]} 6. d3 {[%eval 1,0] [%emt 0:00:00]} Nf6 {[%eval 48,14] [%emt 0:
00:11]} 7. Nf3 {[%eval 26,14] [%emt 0:00:13] (Bg2)} Bg7 {[%eval 48,15] [%emt 0:
00:19] (d6)} 8. Bg2 {[%eval 24,16] [%emt 0:00:47]} d6 {[%eval 0,0] [%emt 0:00:
00]} 9. O-O {[%eval 0,0] [%emt 0:00:00]} O-O {[%eval 0,0] [%emt 0:00:00]} 10.
b4 {[%eval 0,0] [%emt 0:00:00]} axb4 {[%eval 0,0] [%emt 0:00:00]} 11. axb4 {
[%eval 0,0] [%emt 0:00:00]} h6 {[%eval 0,0] [%emt 0:00:00]} 12. b5 {[%eval 0,0]
[%emt 0:00:00]} Ne7 {[%eval 0,0] [%emt 0:00:00]} 13. Qb3 {[%eval 0,0] [%emt 0:
00:00]} Be6 {[%eval 32,16] [%emt 0:00:14]} 14. Bb2 {[%eval 14,13] [%emt 0:00:
10] (Qb2)} Nd7 {[%eval 47,17] [%emt 0:00:27]} 15. Ra1 {[%eval 20,16] [%emt 0:
00:31]} Nc5 {[%eval 42,17] [%emt 0:00:00]} 16. Qc2 {[%eval 19,15] [%emt 0:00:
14]} f4 {[%eval 43,16] [%emt 0:00:01] (Qd7)} 17. Rxa8 {[%eval 37,14] [%emt 0:
00:25] (gxf4)} Qxa8 {[%eval 38,15] [%emt 0:00:08]} 18. Ra1 {[%eval 42,15] 
[%emt 0:00:50] (gxf4)} Qe8 {[%eval 53,17] [%emt 0:00:16] (Qc8)} 19. Nd2 {
[%eval 42,14] [%emt 0:00:24] (gxf4)} Qf7 {[%eval 64,15] [%emt 0:00:29] (Nf5)}
20. Nce4 {[%eval 38,14] [%emt 0:00:14]} Nxe4 {[%eval 60,17] [%emt 0:00:06] (b6)
} 21. Nxe4 {[%eval 42,15] [%emt 0:00:19]} Bf5 {[%eval 85,19] [%emt 0:00:48] 
(Nf5)} 22. gxf4 {[%eval 55,12] [%emt 0:00:09]} Bxe4 {[%eval 86,17] [%emt 0:00:
15] (exf4)} 23. Bxe4 {[%eval 63,14] [%emt 0:00:10] (dxe4)} Qxf4 {[%eval 46,16]
[%emt 0:00:16]} 24. e3 {[%eval 51,14] [%emt 0:00:04]} Qg5+ {[%eval 43,17] 
[%emt 0:00:16] (Qg4+)} 25. Kh1 {[%eval 67,15] [%emt 0:00:17] (Bg2)} Qh4 {
[%eval 45,17] [%emt 0:00:14] (Qf6)} 26. f3 {[%eval 68,14] [%emt 0:00:14] (Rf1)}
c6 {[%eval 48,17] [%emt 0:00:41] (Qg5)} 27. Ba3 {[%eval 73,13] [%emt 0:00:15]}
Nf5 {[%eval 44,16] [%emt 0:00:04]} 28. Rg1 {[%eval 84,13] [%emt 0:00:12] (Qg2)}
Kh7 {[%eval 36,15] [%emt 0:00:16]} 29. Qa4 {[%eval 68,14] [%emt 0:00:08]} cxb5
{[%eval 26,16] [%emt 0:00:14]} 30. Qxb5 {[%eval 95,16] [%emt 0:00:07]} Qf2 {
[%eval 32,18] [%emt 0:00:12]} 31. Qxb7 {[%eval 90,16] [%emt 0:00:14] (Qd5)}
Qxe3 {[%eval 41,19] [%emt 0:00:23]} 32. Qb2 {[%eval 93,17] [%emt 0:00:36] (c5)}
Rf6 {[%eval 47,18] [%emt 0:00:22]} 33. Qg2 {[%eval 99,17] [%emt 0:00:14]} Qa7 {
[%eval 65,18] [%emt 0:00:13] (Ne7)} 34. Bb4 {[%eval 100,16] [%emt 0:00:25] 
(Bxf5)} Qf7 {[%eval 25,18] [%emt 0:00:21]} 35. Be1 {[%eval 87,17] [%emt 0:00:
08]} Qe6 {[%eval 31,18] [%emt 0:00:12] (Ne3)} 36. Bf2 {[%eval 93,16] [%emt 0:
00:19]} Rf8 {[%eval 49,19] [%emt 0:00:11]} 37. Rb1 {[%eval 99,17] [%emt 0:00:
17]} Rd8 {[%eval 58,20] [%emt 0:01:10] (Bf6)} 38. Rb6 {[%eval 102,16] [%emt 0:
00:24]} Qf7 {[%eval 67,18] [%emt 0:00:13]} 39. Qg4 {[%eval 101,16] [%emt 0:00:
14] (Bd5)} h5 {[%eval 46,17] [%emt 0:00:37]} 40. Qg2 {[%eval 97,16] [%emt 0:00:
01] (Qg1)} Bh6 {[%eval 25,18] [%emt 0:00:33] (Bf6)} 41. Qg1 {[%eval 84,14] 
[%emt 0:00:10] (Bd5)} Bf4 {[%eval 27,17] [%emt 0:00:18] (Rd7)} 42. Ra6 {[%eval
75,16] [%emt 0:00:17] (d4)} Kh6 {[%eval 21,17] [%emt 0:00:14] (Rd7)} 43. Qb1 {
[%eval 79,14] [%emt 0:00:17]} Bg5 {[%eval 18,15] [%emt 0:00:00]} 44. Qb6 {
[%eval 78,14] [%emt 0:00:17] (Qe1)} h4 {[%eval 20,16] [%emt 0:00:14] (Qf8)} 45.
Kg2 {[%eval 89,13] [%emt 0:00:10] (Ra7)} Qf6 {[%eval 0,14] [%emt 0:00:11] (Kh5)
} 46. Ra7 {[%eval 64,12] [%emt 0:00:11] (c5)} Rf8 {[%eval 18,16] [%emt 0:00:26]
(Bf4)} 47. Ra1 {[%eval 51,12] [%emt 0:00:11] (h3)} Bc1 {[%eval -2,17] [%emt 0:
00:16] (Bf4)} 48. Kh3 {[%eval 18,13] [%emt 0:00:09]} Qg5 {[%eval 17,16] [%emt
0:00:03]} 49. c5 {[%eval 8,14] [%emt 0:00:10] (Qb1)} Qd2 {[%eval -177,16] 
[%emt 0:00:12] (Ne3)} 50. Bg1 {[%eval -104,13] [%emt 0:00:12] (cxd6)} Qe2 {
[%eval -614,16] [%emt 0:00:17] (d5)} 51. Ra2 {[%eval -534,13] [%emt 0:00:11]}
Qxa2 {[%eval -614,14] [%emt 0:00:00]} 52. cxd6 {[%eval -710,13] [%emt 0:00:20]}
Bg5 {[%eval -661,17] [%emt 0:00:00] (Nd4)} 53. Qf2 {[%eval -728,14] [%emt 0:00:
11]} Qe6 {[%eval -693,18] [%emt 0:00:17]} 54. Kg2 {[%eval -850,16] [%emt 0:00:
22]} h3+ {[%eval -724,18] [%emt 0:00:00] (Nxd6)} 55. Kh1 {[%eval -920,16] 
[%emt 0:00:25]} Nxd6 {[%eval -745,21] [%emt 0:00:00]} 56. Bc6 {[%eval -1095,16]
[%emt 0:00:18] (Qg3)} Rc8 {[%eval -813,17] [%emt 0:00:07]} 57. Ba4 {[%eval
-1361,16] [%emt 0:00:07]} Rc1 {[%eval -1453,20] [%emt 0:00:13]} 58. Bc2 {
[%eval -32747,16] [%emt 0:01:02]} Nf5 {[%eval -2116,19] [%emt 0:00:08]} 59. Qe2
{[%eval -32749,15] [%emt 0:00:08]} Nd4 {[%eval -32755,22] [%emt 0:00:12] (Be3)}
60. Qe4 {[%eval -32747,13] [%emt 0:00:39]} Qd5 {[%eval -32757,27] [%emt 0:00:
00] (Qb6)} 61. Bd1 {[%eval -32755,15] [%emt 0:00:10] (f4)} Rxd1 {[%eval -32759,
27] [%emt 0:00:17] (Qxe4)} 62. Qxg6+ {[%eval 0,25] [%emt 0:00:12] (Qg4)} Kxg6 {
[%eval -86,7] [%emt 0:00:00]} 1/2-1/2
Very, very nice test position I think, not checked with other programs, a lot to do at the moment.

Best
Frank

At the moment I check Ktulu games, now I found this one. OK, switched to Junior beta testing ...
I like computer chess!

zullil
Posts: 5117
Joined: Mon Jan 08, 2007 11:31 pm
Location: PA USA
Full name: Louis Zulli

Does your engine fall into this stalemate trap?

Post by zullil » Tue Mar 23, 2010 8:30 pm

Frank Quisinsky wrote:
Junior 11.1a played 61. ... Rxd1
Very easy Ktulu 9.0x played 62. Qxg6 and the game ended with patt.
Interesting is that Junior show the mate moves before :-)
[D]8/8/6pk/3qp1b1/3nQ3/3P1P1p/7P/2rB2BK b - - 0 1

The goal is to avoid Rxd1, which allows Qxg6 and stalemate.

By the way, Stockfish immediately claims mate-in-5 starting with Qxe4.

metax
Posts: 344
Joined: Wed Sep 23, 2009 3:56 pm
Location: Germany

Re: Does your engine fall into this stalemate trap?

Post by metax » Wed Mar 24, 2010 6:43 pm

ChessMind does not fall into the trap and I cannot imagine any other reasonable engine does (maybe except for Junior):


[D]8/8/6pk/3qp1b1/3nQ3/3P1P1p/7P/2rB2BK b - -

Engine: ChessMind 0.72
by metax

7.00 0:01 -10.04 1...Qa2 2.Qg4 Rxd1 3.Qxh3+ Kg7 4.f4 Qd5+
5.Qg2 Qxg2+ 6.Kxg2 Bxf4 (344.868) 232

7.00 0:01 -11.05++ 1...Qxe4 (351.023) 232

8.00 0:02 -M6 1...Qxe4 2.fxe4 Bf4 3.Bg4 Nf3 4.Bxf3 Be3
5.Bd1 Rxd1 6.d4 Rxg1+ (563.205) 246

best move: Qd5xe4 time: 0:02.559 min n/s: 246.000 nodes: 563.205


after playing Rxd1 on the board:



[D]8/8/6pk/3qp1b1/3nQ3/3P1P1p/7P/3r2BK w - -

Engine: ChessMind 0.72 (0 MB)
by metax

11.00 0:00 0.00 2.Qxg6+ Kxg6 (6.200) 34

12.01 0:03 0.00 2.Qxg6+ Kxg6 (842.525) 266

13.01 0:03 0.00 2.Qxg6+ Kxg6 (1.077.526) 270

14.01 0:04 0.00 2.Qxg6+ Kxg6 (1.087.553) 270

15.01 0:04 0.00 2.Qxg6+ Kxg6 (1.094.104) 270

16.01 0:04 0.00 2.Qxg6+ Kxg6 (1.101.767) 270

17.01 0:04 0.00 2.Qxg6+ Kxg6 (1.128.001) 269

18.01 0:04 0.00 2.Qxg6+ Kxg6 (1.160.035) 270

19.01 0:04 0.00 2.Qxg6+ Kxg6 (1.179.084) 269

20.01 0:04 0.00 2.Qxg6+ Kxg6 (1.220.002) 270

21.01 0:04 0.00 2.Qxg6+ Kxg6 (1.289.360) 270

22.01 0:05 0.00 2.Qxg6+ Kxg6 (1.433.375) 273

23.01 0:06 0.00 2.Qxg6+ Kxg6 (1.732.960) 277

24.01 0:07 0.00 2.Qxg6+ Kxg6 (2.044.704) 282

25.01 0:09 0.00 2.Qxg6+ Kxg6 (2.773.704) 288

best move: Qe4xg6 time: 0:10.858 min n/s: 288.000 nodes: 2.773.704


edit: By the way, the English word for 'Patt' is stalemate. ;)

Frank Quisinsky
Posts: 4833
Joined: Wed Nov 18, 2009 6:16 pm
Location: Trier, Germany
Contact:

Re: Does your engine fall into this stalemate trap?

Post by Frank Quisinsky » Wed Mar 24, 2010 7:59 pm

Hi Luca,

stalemate = Patt ...
I know, wrote the message to fast.

An other user wrote me that the stalemate problem is in the first Junior 5.32 also, means in all available Junior versions. It seems, an older bug Junior have. I checked the position today with other engines too, only Junior have this problem.

I sent Mark Uniacke the position.

The problems with Junior 11.1a I reported ... can not give mate in different endgames seems to be fixed. Also the problem that Junior lost on time. Thats great!

Thanks for testing.

Best
Frank

ChessMind?
I have to search for more information about ChessMind :-)
I like computer chess!

Volker Pittlik
Posts: 597
Joined: Wed Mar 08, 2006 8:10 pm
Location: Murten / Morat, Switzerland

Re: Patt-Bug, Junior 11.1a show mate and Ktulu 9.03 made pat

Post by Volker Pittlik » Thu Mar 25, 2010 4:12 am

Frank Quisinsky wrote:...I am sure that such a bug is in a lot of engines...

Code: Select all

Crafty v23.2 (1 cpus)

White(1): setboard 8/8/6pk/3qp1b1/3nQ3/3P1P1p/7P/2rB2BK b - - 0 1 
Black(1): go
              time surplus   0.00  time limit 30.00 (+0.00) (3:00)
              depth   time  score   variation (1)
              time=0.09  mat=-5  n=176893  fh=90%  nps=1.0M
              extensions=9K qchecks=8K reduced=17K pruned=72K
              predicted=0  evals=16K  50move=0  EGTBprobes=0  hits=0
              SMP->  splits=0  aborts=0  data=0/128  elap=0.09

mate in 5 moves.

Black(1): Qxe4
              time used:   0.09
0.09 seconds (!) on my machine from stone age.

vp

Post Reply