Chiron 1.1 bug?

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
User avatar
rvida
Posts: 481
Joined: Thu Apr 16, 2009 10:00 am
Location: Slovakia, EU

Chiron 1.1 bug?

Post by rvida » Sat Dec 24, 2011 12:17 am

Hi,

This is from a Critter beta tester and I am not able to verify because I do not own Chiron. He submitted some logs when he tested Critter's pawn blockage detection (and compared with Chiron).

Starting from this position:
[d]8/6pp/1k2p3/p1p1P1P1/P1R2r1P/3P4/1P2P1K1/8 b - - 0 1

... both engines saw the blockage, and gave 0.00 evaluation. However, Chiron was not able to hold the draw...

[d]8/7p/1k2p1p1/p1p1P1P1/P1P4P/5K2/1P2P3/8 b - - 1 3
Here Chiron played 3. ...h5?? and lost the game. It's most probably caused by ignoring en passant in the pawn blockage detection code. Hope this helps...

pgn:

Code: Select all

[Event "Computer chess game"]
[Site "far_away_galaxy"]
[Date "2011.12.21"]
[Round "?"]
[White "Critter dev 64bit SSE"]
[Black "Chiron-1.1-64bit"]
[Result "*"]
[WhiteElo "2200"]
[TimeControl "60+1"]
[SetUp "1"]
[FEN "8/6pp/1k2p3/p1p1P1P1/P1R2r1P/3P4/1P2P1K1/8 b - - 0 1"]
[PlyCount "15"]

1. ... Rxc4 {(Rxc4 dxc4 g6 h5) 0.00/32 2} 2. dxc4 {(dxc4 g6 Kf3) 0.00/33 3}
g6 {(g6 h5 gxh5 e4) 0.00/34 2} 3. Kf3 {(Kf3) 0.00/64 0} h5 {(h5 gxh6)
0.00/42 2} 4. gxh6 {(gxh6 Kc6 h7 Kd7 h8Q Ke7 Qf6+ Kd7 Ke4 Kc7 Qxe6 Kd8 Qd6+
Kc8 e6 Kb7 e7 g5 e8Q gxh4 Qeb8+) +M11/22 2} Kc6 {(Kc6 h7 Kc7 h8Q Kb6 Qb8+
Kc6 Qe8+ Kc7 Qxg6 Kc6 h5 Kd7 h6 Kc7 h7 Kb7 h8Q Ka6 Qhg7 Kb6 Qxe6+) -M12/14
0} 5. h7 {(h7 Kd7 h8Q Ke7 Qf6+ Kd7 Ke4 Kc7 Qxe6 Kd8 Qd6+ Kc8 e6 Kb7 e7 g5
e8Q gxh4 Qeb8+) +M10/21 0} Kb6 {(Kb6 h8Q Kc7 Qg7+ Kc8 Qg8+ Kd7 Qxg6 Kc7 h5
Kd7 h6 Kc7 h7 Kb7 h8Q Ka6 Qhg7 Kb6 Qxe6+) -M11/9 0} 6. h8=Q {(h8Q Kb7 Qe8
Kb6 Qxe6+ Kc7 Qd6+ Kb7 e6 g5 Ke4 gxh4 e7 h3 e8Q Ka7 Qeb8+) +M9/21 2} Kb7
{(Kb7 Qg7+ Kc8 Qg8+ Kd7 Qxg6 Kc7 h5 Kd7 h6 Kc7 h7 Kb7 h8Q Ka6 Qhg7 Kb6
Qxe6+) -M10/6 0} 7. Qd8 {(Qd8 Kc6 b4 cxb4 Qc8+ Kb6 c5+ Ka7 Qd7+ Kb8 c6 Ka8
Qb7+) +M7/22 2} Ka7 {(Ka7 Qc8 Kb6 Qb8+ Kc6 Qe8+ Kc7 Qxg6 Kc6 h5 Kd7 h6 Kc7
h7 Kb7 h8Q Ka6 Qhg7 Kb6 Qxe6+) -M11/9 0} 8. b4 {(b4 cxb4 Qc7+ Ka8 c5 g5 c6
g4+ Kxg4 b3 Qb7+) +M6/23 2} g5 {(g5 hxg5 cxb4 Qc7+ Ka8 g6 b3 g7 b2 g8Q+)
-M5/10 0 Black resigns} 1-0
Top
User avatar
Graham Banks
Posts: 30733
Joined: Sun Feb 26, 2006 9:52 am
Location: Auckland, NZ

Re: Chiron 1.1 bug?

Post by Graham Banks » Sat Dec 24, 2011 12:54 am

rvida wrote:Hi,

This is from a Critter beta tester and I am not able to verify because I do not own Chiron. He submitted some logs when he tested Critter's pawn blockage detection (and compared with Chiron).
Chiron 1.1a?
My email addresses:
gbanksnz at gmail.com
gbanksnz at yahoo.co.nz
Top
Robert Flesher
Posts: 1205
Joined: Tue Aug 18, 2009 1:06 am

Re: Chiron 1.1 bug?

Post by Robert Flesher » Sat Dec 24, 2011 3:35 am

rvida wrote:Hi,

This is from a Critter beta tester and I am not able to verify because I do not own Chiron. He submitted some logs when he tested Critter's pawn blockage detection (and compared with Chiron).

Starting from this position:
[d]8/6pp/1k2p3/p1p1P1P1/P1R2r1P/3P4/1P2P1K1/8 b - - 0 1

... both engines saw the blockage, and gave 0.00 evaluation. However, Chiron was not able to hold the draw...

[d]8/7p/1k2p1p1/p1p1P1P1/P1P4P/5K2/1P2P3/8 b - - 1 3
Here Chiron played 3. ...h5?? and lost the game. It's most probably caused by ignoring en passant in the pawn blockage detection code. Hope this helps...

pgn:

Code: Select all

[Event "Computer chess game"]
[Site "far_away_galaxy"]
[Date "2011.12.21"]
[Round "?"]
[White "Critter dev 64bit SSE"]
[Black "Chiron-1.1-64bit"]
[Result "*"]
[WhiteElo "2200"]
[TimeControl "60+1"]
[SetUp "1"]
[FEN "8/6pp/1k2p3/p1p1P1P1/P1R2r1P/3P4/1P2P1K1/8 b - - 0 1"]
[PlyCount "15"]

1. ... Rxc4 {(Rxc4 dxc4 g6 h5) 0.00/32 2} 2. dxc4 {(dxc4 g6 Kf3) 0.00/33 3}
g6 {(g6 h5 gxh5 e4) 0.00/34 2} 3. Kf3 {(Kf3) 0.00/64 0} h5 {(h5 gxh6)
0.00/42 2} 4. gxh6 {(gxh6 Kc6 h7 Kd7 h8Q Ke7 Qf6+ Kd7 Ke4 Kc7 Qxe6 Kd8 Qd6+
Kc8 e6 Kb7 e7 g5 e8Q gxh4 Qeb8+) +M11/22 2} Kc6 {(Kc6 h7 Kc7 h8Q Kb6 Qb8+
Kc6 Qe8+ Kc7 Qxg6 Kc6 h5 Kd7 h6 Kc7 h7 Kb7 h8Q Ka6 Qhg7 Kb6 Qxe6+) -M12/14
0} 5. h7 {(h7 Kd7 h8Q Ke7 Qf6+ Kd7 Ke4 Kc7 Qxe6 Kd8 Qd6+ Kc8 e6 Kb7 e7 g5
e8Q gxh4 Qeb8+) +M10/21 0} Kb6 {(Kb6 h8Q Kc7 Qg7+ Kc8 Qg8+ Kd7 Qxg6 Kc7 h5
Kd7 h6 Kc7 h7 Kb7 h8Q Ka6 Qhg7 Kb6 Qxe6+) -M11/9 0} 6. h8=Q {(h8Q Kb7 Qe8
Kb6 Qxe6+ Kc7 Qd6+ Kb7 e6 g5 Ke4 gxh4 e7 h3 e8Q Ka7 Qeb8+) +M9/21 2} Kb7
{(Kb7 Qg7+ Kc8 Qg8+ Kd7 Qxg6 Kc7 h5 Kd7 h6 Kc7 h7 Kb7 h8Q Ka6 Qhg7 Kb6
Qxe6+) -M10/6 0} 7. Qd8 {(Qd8 Kc6 b4 cxb4 Qc8+ Kb6 c5+ Ka7 Qd7+ Kb8 c6 Ka8
Qb7+) +M7/22 2} Ka7 {(Ka7 Qc8 Kb6 Qb8+ Kc6 Qe8+ Kc7 Qxg6 Kc6 h5 Kd7 h6 Kc7
h7 Kb7 h8Q Ka6 Qhg7 Kb6 Qxe6+) -M11/9 0} 8. b4 {(b4 cxb4 Qc7+ Ka8 c5 g5 c6
g4+ Kxg4 b3 Qb7+) +M6/23 2} g5 {(g5 hxg5 cxb4 Qc7+ Ka8 g6 b3 g7 b2 g8Q+)
-M5/10 0 Black resigns} 1-0

Must be a bug as Chiron 1.1a shows this


New game
8/7p/1k2p1p1/p1p1P1P1/P1P4P/5K2/1P2P3/8 b - - 0 1

Analysis by Chiron 1.1a 64bit:

1...Kb7 2.Ke4
= (0.00) Depth: 6/6 00:00:00 0kN
1...Kb7 2.Ke4
= (0.00) Depth: 7/7 00:00:00 0kN
1...Kb7 2.Ke4
= (0.00) Depth: 8/8 00:00:00 0kN
1...Kb7 2.Ke4
= (0.00) Depth: 9/9 00:00:00 0kN
1...Kb7 2.Ke4
= (0.00) Depth: 10/10 00:00:00 0kN
1...Kb7 2.Ke4
= (0.00) Depth: 11/11 00:00:00 0kN
1...Kb7 2.Ke4
= (0.00) Depth: 12/12 00:00:00 0kN
1...Kb7 2.Ke4
= (0.00) Depth: 13/13 00:00:00 0kN
1...Kb7 2.Ke4
= (0.00) Depth: 14/14 00:00:00 1kN
1...Kb7 2.Ke4
= (0.00) Depth: 15/15 00:00:00 1kN
1...Kb7 2.Ke4
= (0.00) Depth: 16/16 00:00:00 1kN
1...Kb7 2.Ke4
= (0.00) Depth: 17/17 00:00:00 1kN
1...Kb7 2.Ke4
= (0.00) Depth: 18/18 00:00:00 2kN
1...Kb7 2.Ke4
= (0.00) Depth: 19/19 00:00:00 3kN
1...Kb7 2.Ke4
= (0.00) Depth: 20/20 00:00:00 4kN
1...Kb7 2.Ke4
= (0.00) Depth: 21/21 00:00:00 6kN
1...Kb7 2.Ke4
= (0.00) Depth: 22/22 00:00:00 8kN
1...Kb7 2.Ke4
= (0.00) Depth: 23/23 00:00:00 11kN
1...Kb7 2.Ke4
= (0.00) Depth: 24/24 00:00:00 13kN
1...Kb7 2.Ke4
= (0.00) Depth: 25/25 00:00:00 16kN
1...Kb7 2.Ke4
= (0.00) Depth: 26/26 00:00:00 18kN
1...Kb7 2.Ke4
= (0.00) Depth: 27/27 00:00:00 22kN
1...Kb7 2.Ke4
= (0.00) Depth: 28/28 00:00:00 25kN
1...Kb7 2.Ke4
= (0.00) Depth: 29/29 00:00:00 29kN
1...Kb7 2.Ke4
= (0.00) Depth: 30/30 00:00:00 33kN
1...Kb7 2.Ke4
= (0.00) Depth: 31/31 00:00:00 37kN
1...Kb7 2.Ke4
= (0.00) Depth: 32/32 00:00:00 42kN
1...Kb7 2.Ke4
= (0.00) Depth: 33/33 00:00:00 48kN
1...Kb7 2.Ke4
= (0.00) Depth: 34/34 00:00:00 54kN
1...Kb7 2.Ke4
= (0.00) Depth: 35/35 00:00:00 56kN
1...Kb7 2.Ke4
= (0.00) Depth: 36/36 00:00:00 62kN
1...Kb7 2.Ke4
= (0.00) Depth: 37/37 00:00:00 71kN
1...Kb7 2.Ke4
= (0.00) Depth: 38/38 00:00:00 73kN
1...Kb7 2.Ke4
= (0.00) Depth: 39/39 00:00:00 81kN
1...Kb7 2.Ke4
= (0.00) Depth: 40/40 00:00:00 89kN
1...Kb7 2.Ke4
= (0.00) Depth: 41/41 00:00:00 97kN
1...Kb7 2.Ke4
= (0.00) Depth: 42/42 00:00:00 106kN
1...Kb7 2.Ke4
= (0.00) Depth: 43/43 00:00:00 116kN
1...Kb7 2.Ke4
= (0.00) Depth: 44/44 00:00:00 125kN
1...Kb7 2.Ke4
= (0.00) Depth: 45/45 00:00:00 133kN
1...Kb7 2.Ke4
= (0.00) Depth: 46/46 00:00:00 143kN
1...Kb7 2.Ke4
= (0.00) Depth: 47/47 00:00:00 152kN
1...Kb7 2.Ke4
= (0.00) Depth: 48/48 00:00:00 162kN
1...Kb7 2.Ke4
= (0.00) Depth: 49/49 00:00:00 170kN
1...Kb7 2.Ke4
= (0.00) Depth: 50/50 00:00:00 176kN
1...Kb7 2.Ke4
= (0.00) Depth: 51/51 00:00:00 187kN
1...Kb7 2.Ke4
= (0.00) Depth: 52/52 00:00:00 198kN
1...Kb7 2.Ke4
= (0.00) Depth: 53/53 00:00:00 208kN
1...Kb7 2.Ke4
= (0.00) Depth: 54/54 00:00:00 221kN
1...Kb7 2.Ke4
= (0.00) Depth: 55/55 00:00:00 237kN
1...Kb7 2.Ke4
= (0.00) Depth: 56/56 00:00:00 248kN
1...Kb7 2.Ke4
= (0.00) Depth: 57/57 00:00:00 262kN
1...Kb7 2.Ke4
= (0.00) Depth: 58/58 00:00:00 278kN
1...Kb7 2.Ke4
= (0.00) Depth: 59/59 00:00:00 301kN
1...Kb7 2.Ke4
= (0.00) Depth: 60/60 00:00:00 323kN
1...Kb7 2.Ke4
= (0.00) Depth: 61/61 00:00:00 339kN
1...Kb7 2.Ke4
= (0.00) Depth: 62/62 00:00:00 362kN
1...Kb7 2.Ke4
= (0.00) Depth: 63/63 00:00:00 395kN
1...Kb7 2.Ke4
= (0.00) Depth: 64/64 00:00:00 431kN
1...Kb7 2.Ke4
= (0.00) Depth: 65/65 00:00:00 459kN
1...Kb7 2.Ke4
= (0.00) Depth: 66/66 00:00:00 501kN
1...Kb7 2.Ke4
= (0.00) Depth: 67/67 00:00:00 542kN
1...Kb7 2.Ke4
= (0.00) Depth: 68/68 00:00:00 574kN
1...Kb7 2.Ke4
= (0.00) Depth: 69/69 00:00:00 607kN
1...Kb7 2.Ke4
= (0.00) Depth: 70/70 00:00:00 684kN
1...Kb7 2.Ke4
= (0.00) Depth: 71/71 00:00:00 797kN
1...Kb7 2.Ke4
= (0.00) Depth: 72/72 00:00:00 871kN
1...Kb7 2.Ke4
= (0.00) Depth: 73/73 00:00:00 911kN
1...Kb7 2.Ke4
= (0.00) Depth: 74/74 00:00:00 1025kN
1...Kb7 2.Ke4
= (0.00) Depth: 75/75 00:00:00 1138kN
1...Kb7 2.Ke4
= (0.00) Depth: 76/76 00:00:00 1340kN
1...Kb7 2.Ke4
= (0.00) Depth: 77/77 00:00:00 1516kN
1...Kb7 2.Ke4
= (0.00) Depth: 78/78 00:00:00 1804kN
1...Kb7 2.Ke4
= (0.00) Depth: 79/79 00:00:00 2009kN
1...Kb7 2.Ke4
= (0.00) Depth: 80/80 00:00:00 2166kN
1...Kb7 2.Ke4
= (0.00) Depth: 81/81 00:00:00 2377kN
1...Kb7 2.Ke4
= (0.00) Depth: 82/82 00:00:00 2608kN
1...Kb7 2.Ke4
= (0.00) Depth: 83/83 00:00:00 2928kN
1...Kb7 2.Ke4
= (0.00) Depth: 84/84 00:00:00 3311kN
1...Kb7 2.Ke4
= (0.00) Depth: 85/85 00:00:00 3634kN
1...Kb7 2.Ke4
= (0.00) Depth: 86/86 00:00:01 4397kN
1...Kb7 2.Ke4
= (0.00) Depth: 87/87 00:00:01 5202kN
1...Kb7 2.Ke4
= (0.00) Depth: 88/88 00:00:01 5761kN
1...Kb7 2.Ke4
= (0.00) Depth: 89/89 00:00:01 7052kN
1...Kb7 2.Ke4
= (0.00) Depth: 90/90 00:00:02 9559kN
1...Kb7 2.Ke4
= (0.00) Depth: 91/91 00:00:02 12486kN
1...Kb7 2.Ke4
= (0.00) Depth: 92/92 00:00:02 13546kN
1...Kb7 2.Ke4
= (0.00) Depth: 93/93 00:00:03 17247kN
1...Kb7 2.Ke4
= (0.00) Depth: 94/94 00:00:04 20480kN
1...Kb7 2.Ke4
= (0.00) Depth: 95/95 00:00:04 22583kN
1...Kb7 2.Ke4
= (0.00) Depth: 96/96 00:00:05 25341kN
1...Kb7 2.Ke4
= (0.00) Depth: 97/97 00:00:06 28571kN
1...Kb7 2.Ke4
= (0.00) Depth: 98/98 00:00:07 33973kN
1...Kb7 2.Ke4
= (0.00) Depth: 99/99 00:00:08 40351kN
1...Kb7 2.Ke4
= (0.00) Depth: 100/100 00:00:09 43015kN
1...Kb7 2.Ke4
= (0.00) Depth: 101/101 00:00:10 47995kN
1...Kb7 2.Ke4
= (0.00) Depth: 102/102 00:00:12 58071kN
1...Kb7 2.Ke4
= (0.00) Depth: 103/103 00:00:13 65273kN
1...Kb7 2.Ke4
= (0.00) Depth: 104/104 00:00:16 79557kN
1...Kb7 2.Ke4
= (0.00) Depth: 105/105 00:00:21 100mN
1...Kb7 2.Ke4
= (0.00) Depth: 106/106 00:00:31 149mN
1...Kb7 2.Ke4
= (0.00) Depth: 107/107 00:00:41 197mN
1...Kb7 2.Ke4
= (0.00) Depth: 108/108 00:00:48 234mN
1...Kb7 2.Ke4
= (0.00) Depth: 109/109 00:00:50 242mN
1...Kb7 2.Ke4
= (0.00) Depth: 110/110 00:00:55 269mN

(, Microsoft 23.12.2011)


, but it seems to be pondering h5? Strange!
Top
MM
Posts: 766
Joined: Sun Oct 16, 2011 9:25 am

Re: Chiron 1.1 bug?

Post by MM » Sat Dec 24, 2011 11:35 am

rvida wrote:Hi,

This is from a Critter beta tester and I am not able to verify because I do not own Chiron. He submitted some logs when he tested Critter's pawn blockage detection (and compared with Chiron).

Starting from this position:
[d]8/6pp/1k2p3/p1p1P1P1/P1R2r1P/3P4/1P2P1K1/8 b - - 0 1

... both engines saw the blockage, and gave 0.00 evaluation. However, Chiron was not able to hold the draw...

[d]8/7p/1k2p1p1/p1p1P1P1/P1P4P/5K2/1P2P3/8 b - - 1 3
Here Chiron played 3. ...h5?? and lost the game. It's most probably caused by ignoring en passant in the pawn blockage detection code. Hope this helps...

pgn:

Code: Select all

[Event "Computer chess game"]
[Site "far_away_galaxy"]
[Date "2011.12.21"]
[Round "?"]
[White "Critter dev 64bit SSE"]
[Black "Chiron-1.1-64bit"]
[Result "*"]
[WhiteElo "2200"]
[TimeControl "60+1"]
[SetUp "1"]
[FEN "8/6pp/1k2p3/p1p1P1P1/P1R2r1P/3P4/1P2P1K1/8 b - - 0 1"]
[PlyCount "15"]

1. ... Rxc4 {(Rxc4 dxc4 g6 h5) 0.00/32 2} 2. dxc4 {(dxc4 g6 Kf3) 0.00/33 3}
g6 {(g6 h5 gxh5 e4) 0.00/34 2} 3. Kf3 {(Kf3) 0.00/64 0} h5 {(h5 gxh6)
0.00/42 2} 4. gxh6 {(gxh6 Kc6 h7 Kd7 h8Q Ke7 Qf6+ Kd7 Ke4 Kc7 Qxe6 Kd8 Qd6+
Kc8 e6 Kb7 e7 g5 e8Q gxh4 Qeb8+) +M11/22 2} Kc6 {(Kc6 h7 Kc7 h8Q Kb6 Qb8+
Kc6 Qe8+ Kc7 Qxg6 Kc6 h5 Kd7 h6 Kc7 h7 Kb7 h8Q Ka6 Qhg7 Kb6 Qxe6+) -M12/14
0} 5. h7 {(h7 Kd7 h8Q Ke7 Qf6+ Kd7 Ke4 Kc7 Qxe6 Kd8 Qd6+ Kc8 e6 Kb7 e7 g5
e8Q gxh4 Qeb8+) +M10/21 0} Kb6 {(Kb6 h8Q Kc7 Qg7+ Kc8 Qg8+ Kd7 Qxg6 Kc7 h5
Kd7 h6 Kc7 h7 Kb7 h8Q Ka6 Qhg7 Kb6 Qxe6+) -M11/9 0} 6. h8=Q {(h8Q Kb7 Qe8
Kb6 Qxe6+ Kc7 Qd6+ Kb7 e6 g5 Ke4 gxh4 e7 h3 e8Q Ka7 Qeb8+) +M9/21 2} Kb7
{(Kb7 Qg7+ Kc8 Qg8+ Kd7 Qxg6 Kc7 h5 Kd7 h6 Kc7 h7 Kb7 h8Q Ka6 Qhg7 Kb6
Qxe6+) -M10/6 0} 7. Qd8 {(Qd8 Kc6 b4 cxb4 Qc8+ Kb6 c5+ Ka7 Qd7+ Kb8 c6 Ka8
Qb7+) +M7/22 2} Ka7 {(Ka7 Qc8 Kb6 Qb8+ Kc6 Qe8+ Kc7 Qxg6 Kc6 h5 Kd7 h6 Kc7
h7 Kb7 h8Q Ka6 Qhg7 Kb6 Qxe6+) -M11/9 0} 8. b4 {(b4 cxb4 Qc7+ Ka8 c5 g5 c6
g4+ Kxg4 b3 Qb7+) +M6/23 2} g5 {(g5 hxg5 cxb4 Qc7+ Ka8 g6 b3 g7 b2 g8Q+)
-M5/10 0 Black resigns} 1-0
It seems to be very strange.

With Fritz 13 GUI the engine clearly discards ...h5 because it sees that move loses with checkmate in 13 or something.

With Fritz 12 GUI the engine scores 0.00 with all lines, including ...h5...

Best Regards
MM
Top
rodolfoleoni
Posts: 263
Joined: Mon Nov 29, 2010 8:16 pm

Re: Chiron 1.1 bug?

Post by rodolfoleoni » Sat Dec 24, 2011 12:29 pm

rvida wrote:Hi,

This is from a Critter beta tester and I am not able to verify because I do not own Chiron. He submitted some logs when he tested Critter's pawn blockage detection (and compared with Chiron).

Starting from this position:
[d]8/6pp/1k2p3/p1p1P1P1/P1R2r1P/3P4/1P2P1K1/8 b - - 0 1

... both engines saw the blockage, and gave 0.00 evaluation. However, Chiron was not able to hold the draw...

[d]8/7p/1k2p1p1/p1p1P1P1/P1P4P/5K2/1P2P3/8 b - - 1 3
Here Chiron played 3. ...h5?? and lost the game. It's most probably caused by ignoring en passant in the pawn blockage detection code. Hope this helps...

pgn:

Code: Select all

[Event "Computer chess game"]
[Site "far_away_galaxy"]
[Date "2011.12.21"]
[Round "?"]
[White "Critter dev 64bit SSE"]
[Black "Chiron-1.1-64bit"]
[Result "*"]
[WhiteElo "2200"]
[TimeControl "60+1"]
[SetUp "1"]
[FEN "8/6pp/1k2p3/p1p1P1P1/P1R2r1P/3P4/1P2P1K1/8 b - - 0 1"]
[PlyCount "15"]

1. ... Rxc4 {(Rxc4 dxc4 g6 h5) 0.00/32 2} 2. dxc4 {(dxc4 g6 Kf3) 0.00/33 3}
g6 {(g6 h5 gxh5 e4) 0.00/34 2} 3. Kf3 {(Kf3) 0.00/64 0} [color=red]h5 {(h5 gxh6)
0.00/42 2}[/color] 4. gxh6 {(gxh6 Kc6 h7 Kd7 h8Q Ke7 Qf6+ Kd7 Ke4 Kc7 Qxe6 Kd8 Qd6+
Kc8 e6 Kb7 e7 g5 e8Q gxh4 Qeb8+) +M11/22 2} Kc6 {(Kc6 h7 Kc7 h8Q Kb6 Qb8+
Kc6 Qe8+ Kc7 Qxg6 Kc6 h5 Kd7 h6 Kc7 h7 Kb7 h8Q Ka6 Qhg7 Kb6 Qxe6+) -M12/14
0} 5. h7 {(h7 Kd7 h8Q Ke7 Qf6+ Kd7 Ke4 Kc7 Qxe6 Kd8 Qd6+ Kc8 e6 Kb7 e7 g5
e8Q gxh4 Qeb8+) +M10/21 0} Kb6 {(Kb6 h8Q Kc7 Qg7+ Kc8 Qg8+ Kd7 Qxg6 Kc7 h5
Kd7 h6 Kc7 h7 Kb7 h8Q Ka6 Qhg7 Kb6 Qxe6+) -M11/9 0} 6. h8=Q {(h8Q Kb7 Qe8
Kb6 Qxe6+ Kc7 Qd6+ Kb7 e6 g5 Ke4 gxh4 e7 h3 e8Q Ka7 Qeb8+) +M9/21 2} Kb7
{(Kb7 Qg7+ Kc8 Qg8+ Kd7 Qxg6 Kc7 h5 Kd7 h6 Kc7 h7 Kb7 h8Q Ka6 Qhg7 Kb6
Qxe6+) -M10/6 0} 7. Qd8 {(Qd8 Kc6 b4 cxb4 Qc8+ Kb6 c5+ Ka7 Qd7+ Kb8 c6 Ka8
Qb7+) +M7/22 2} Ka7 {(Ka7 Qc8 Kb6 Qb8+ Kc6 Qe8+ Kc7 Qxg6 Kc6 h5 Kd7 h6 Kc7
h7 Kb7 h8Q Ka6 Qhg7 Kb6 Qxe6+) -M11/9 0} 8. b4 {(b4 cxb4 Qc7+ Ka8 c5 g5 c6
g4+ Kxg4 b3 Qb7+) +M6/23 2} g5 {(g5 hxg5 cxb4 Qc7+ Ka8 g6 b3 g7 b2 g8Q+)
-M5/10 0 Black resigns} 1-0
In the PV it sees the ep capture, but it scores 0.0 at depth 42. Seemingly, something happens at search after that capture. Maurizio reports a different output and score between Fritz 12 and Fritz 13 GUI, while Robert reports it (apparently) fixed with 1.1a version.

It'd be useful to know what GUI is your beta tester using.
Rodolfo (The Baron Team)
Top
User avatar
rvida
Posts: 481
Joined: Thu Apr 16, 2009 10:00 am
Location: Slovakia, EU

Re: Chiron 1.1 bug?

Post by rvida » Sat Dec 24, 2011 2:32 pm

Graham Banks wrote: Chiron 1.1a?
Yes.
rodolfoleoni wrote: It'd be useful to know what GUI is your beta tester using.
The log files were produced by Arena, but I'm not sure which version.
Top
MM
Posts: 766
Joined: Sun Oct 16, 2011 9:25 am

Re: Chiron 1.1 bug?

Post by MM » Sat Dec 24, 2011 4:36 pm

rodolfoleoni wrote:
rvida wrote:Hi,

This is from a Critter beta tester and I am not able to verify because I do not own Chiron. He submitted some logs when he tested Critter's pawn blockage detection (and compared with Chiron).

Starting from this position:
[d]8/6pp/1k2p3/p1p1P1P1/P1R2r1P/3P4/1P2P1K1/8 b - - 0 1

... both engines saw the blockage, and gave 0.00 evaluation. However, Chiron was not able to hold the draw...

[d]8/7p/1k2p1p1/p1p1P1P1/P1P4P/5K2/1P2P3/8 b - - 1 3
Here Chiron played 3. ...h5?? and lost the game. It's most probably caused by ignoring en passant in the pawn blockage detection code. Hope this helps...

pgn:

Code: Select all

[Event "Computer chess game"]
[Site "far_away_galaxy"]
[Date "2011.12.21"]
[Round "?"]
[White "Critter dev 64bit SSE"]
[Black "Chiron-1.1-64bit"]
[Result "*"]
[WhiteElo "2200"]
[TimeControl "60+1"]
[SetUp "1"]
[FEN "8/6pp/1k2p3/p1p1P1P1/P1R2r1P/3P4/1P2P1K1/8 b - - 0 1"]
[PlyCount "15"]

1. ... Rxc4 {(Rxc4 dxc4 g6 h5) 0.00/32 2} 2. dxc4 {(dxc4 g6 Kf3) 0.00/33 3}
g6 {(g6 h5 gxh5 e4) 0.00/34 2} 3. Kf3 {(Kf3) 0.00/64 0} [color=red]h5 {(h5 gxh6)
0.00/42 2}[/color] 4. gxh6 {(gxh6 Kc6 h7 Kd7 h8Q Ke7 Qf6+ Kd7 Ke4 Kc7 Qxe6 Kd8 Qd6+
Kc8 e6 Kb7 e7 g5 e8Q gxh4 Qeb8+) +M11/22 2} Kc6 {(Kc6 h7 Kc7 h8Q Kb6 Qb8+
Kc6 Qe8+ Kc7 Qxg6 Kc6 h5 Kd7 h6 Kc7 h7 Kb7 h8Q Ka6 Qhg7 Kb6 Qxe6+) -M12/14
0} 5. h7 {(h7 Kd7 h8Q Ke7 Qf6+ Kd7 Ke4 Kc7 Qxe6 Kd8 Qd6+ Kc8 e6 Kb7 e7 g5
e8Q gxh4 Qeb8+) +M10/21 0} Kb6 {(Kb6 h8Q Kc7 Qg7+ Kc8 Qg8+ Kd7 Qxg6 Kc7 h5
Kd7 h6 Kc7 h7 Kb7 h8Q Ka6 Qhg7 Kb6 Qxe6+) -M11/9 0} 6. h8=Q {(h8Q Kb7 Qe8
Kb6 Qxe6+ Kc7 Qd6+ Kb7 e6 g5 Ke4 gxh4 e7 h3 e8Q Ka7 Qeb8+) +M9/21 2} Kb7
{(Kb7 Qg7+ Kc8 Qg8+ Kd7 Qxg6 Kc7 h5 Kd7 h6 Kc7 h7 Kb7 h8Q Ka6 Qhg7 Kb6
Qxe6+) -M10/6 0} 7. Qd8 {(Qd8 Kc6 b4 cxb4 Qc8+ Kb6 c5+ Ka7 Qd7+ Kb8 c6 Ka8
Qb7+) +M7/22 2} Ka7 {(Ka7 Qc8 Kb6 Qb8+ Kc6 Qe8+ Kc7 Qxg6 Kc6 h5 Kd7 h6 Kc7
h7 Kb7 h8Q Ka6 Qhg7 Kb6 Qxe6+) -M11/9 0} 8. b4 {(b4 cxb4 Qc7+ Ka8 c5 g5 c6
g4+ Kxg4 b3 Qb7+) +M6/23 2} g5 {(g5 hxg5 cxb4 Qc7+ Ka8 g6 b3 g7 b2 g8Q+)
-M5/10 0 Black resigns} 1-0
In the PV it sees the ep capture, but it scores 0.0 at depth 42. Seemingly, something happens at search after that capture. Maurizio reports a different output and score between Fritz 12 and Fritz 13 GUI, while Robert reports it (apparently) fixed with 1.1a version.

It'd be useful to know what GUI is your beta tester using.

Hi,
i looked better.
If i set a fixed time per move, ''2 seconds'', even in FRITZ 13 GUI Chiron 1.1a moves King b7 ... waiting for white's move King e4 and planning ....h5 (??) with the analysis gxh6...Kb8, h7 and score 0.00!
I am a bit confused.

Always using FRITZ 13 GUI now i try with 1.1 with 2 seconds per move:

same result, ...Kb7...waiting for Ke4...h5 (??) gxh6..Kb8, h7 and still score 0.00!

Always using FRITZ 13 GUI, if i give infinite time to think to the engines, black has 7 legal moves, h5 and h6 are instantly discarded cause checkmate in 12 or something.

Using FRITZ 12 GUI, Chiron 1.1, 2 seconds per move, plays: Kb7..(waiting for Ke4) and stops analysis so i try to play Ke4 and...it horribly plays ...h5 (??). I did it, of course, with ponder OFF.

Chiron 1.1a has exactly the same behaviour.

Again, if i run infinite analysis, Chiron 1.1a shows 7 legal moves and sees instantly that h5 and h6 drive to a checkmate in 14 or something and Chiron 1.1 the same.

Before probably i missed something, it seemed to me that there were different behaviours of the GUI, now it seems to me that it's a problem of the engine that in that position, after King b7 and King e4 likes to play h5...
Sorry for all these informations, i have no much experience with these kinds of test.

Anyway it seems to me that Mr Vida is right, it seems that Chiron doesn't consider en passant move, at least in some specific situations (perhaps it was in terrible time trouble?!)

Regards
MM
Top
rodolfoleoni
Posts: 263
Joined: Mon Nov 29, 2010 8:16 pm

Re: Chiron 1.1 bug?

Post by rodolfoleoni » Sat Dec 24, 2011 8:49 pm

MM wrote: Hi,
i looked better.
If i set a fixed time per move, ''2 seconds'', even in FRITZ 13 GUI Chiron 1.1a moves King b7 ... waiting for white's move King e4 and planning ....h5 (??) with the analysis gxh6...Kb8, h7 and score 0.00!
I am a bit confused.

Always using FRITZ 13 GUI now i try with 1.1 with 2 seconds per move:

same result, ...Kb7...waiting for Ke4...h5 (??) gxh6..Kb8, h7 and still score 0.00!

Always using FRITZ 13 GUI, if i give infinite time to think to the engines, black has 7 legal moves, h5 and h6 are instantly discarded cause checkmate in 12 or something.

Using FRITZ 12 GUI, Chiron 1.1, 2 seconds per move, plays: Kb7..(waiting for Ke4) and stops analysis so i try to play Ke4 and...it horribly plays ...h5 (??). I did it, of course, with ponder OFF.

Chiron 1.1a has exactly the same behaviour.

Again, if i run infinite analysis, Chiron 1.1a shows 7 legal moves and sees instantly that h5 and h6 drive to a checkmate in 14 or something and Chiron 1.1 the same.

Before probably i missed something, it seemed to me that there were different behaviours of the GUI, now it seems to me that it's a problem of the engine that in that position, after King b7 and King e4 likes to play h5...
Sorry for all these informations, i have no much experience with these kinds of test.

Anyway it seems to me that Mr Vida is right, it seems that Chiron doesn't consider en passant move, at least in some specific situations (perhaps it was in terrible time trouble?!)

Regards
Hi Maurizio,

I consider if a move is in the PV then engine has considered it. The problem here is, something wrong gets pruned right after that. Since you say the move h7 is in the PV too, it could be the promotion. But the strange thing is that the bug depends on the TCs. I think only Ubaldo Andrea Farina can try to figure out what really happens. As you own the engine, you could report this bug via mail.

You could run a different test too. You could delete some pawns from the position to see if it's really related to pawn blockage detection code. Or you could set some more pawn-blocked positions with a potential ep capture to have a confirm about that kind of bug.
Rodolfo (The Baron Team)
Top
Robert Flesher
Posts: 1205
Joined: Tue Aug 18, 2009 1:06 am

Re: Chiron 1.1 bug?

Post by Robert Flesher » Sat Dec 24, 2011 9:47 pm

MM wrote:
rodolfoleoni wrote:
rvida wrote:Hi,

This is from a Critter beta tester and I am not able to verify because I do not own Chiron. He submitted some logs when he tested Critter's pawn blockage detection (and compared with Chiron).

Starting from this position:
[d]8/6pp/1k2p3/p1p1P1P1/P1R2r1P/3P4/1P2P1K1/8 b - - 0 1

... both engines saw the blockage, and gave 0.00 evaluation. However, Chiron was not able to hold the draw...

[d]8/7p/1k2p1p1/p1p1P1P1/P1P4P/5K2/1P2P3/8 b - - 1 3
Here Chiron played 3. ...h5?? and lost the game. It's most probably caused by ignoring en passant in the pawn blockage detection code. Hope this helps...

pgn:

Code: Select all

[Event "Computer chess game"]
[Site "far_away_galaxy"]
[Date "2011.12.21"]
[Round "?"]
[White "Critter dev 64bit SSE"]
[Black "Chiron-1.1-64bit"]
[Result "*"]
[WhiteElo "2200"]
[TimeControl "60+1"]
[SetUp "1"]
[FEN "8/6pp/1k2p3/p1p1P1P1/P1R2r1P/3P4/1P2P1K1/8 b - - 0 1"]
[PlyCount "15"]

1. ... Rxc4 {(Rxc4 dxc4 g6 h5) 0.00/32 2} 2. dxc4 {(dxc4 g6 Kf3) 0.00/33 3}
g6 {(g6 h5 gxh5 e4) 0.00/34 2} 3. Kf3 {(Kf3) 0.00/64 0} [color=red]h5 {(h5 gxh6)
0.00/42 2}[/color] 4. gxh6 {(gxh6 Kc6 h7 Kd7 h8Q Ke7 Qf6+ Kd7 Ke4 Kc7 Qxe6 Kd8 Qd6+
Kc8 e6 Kb7 e7 g5 e8Q gxh4 Qeb8+) +M11/22 2} Kc6 {(Kc6 h7 Kc7 h8Q Kb6 Qb8+
Kc6 Qe8+ Kc7 Qxg6 Kc6 h5 Kd7 h6 Kc7 h7 Kb7 h8Q Ka6 Qhg7 Kb6 Qxe6+) -M12/14
0} 5. h7 {(h7 Kd7 h8Q Ke7 Qf6+ Kd7 Ke4 Kc7 Qxe6 Kd8 Qd6+ Kc8 e6 Kb7 e7 g5
e8Q gxh4 Qeb8+) +M10/21 0} Kb6 {(Kb6 h8Q Kc7 Qg7+ Kc8 Qg8+ Kd7 Qxg6 Kc7 h5
Kd7 h6 Kc7 h7 Kb7 h8Q Ka6 Qhg7 Kb6 Qxe6+) -M11/9 0} 6. h8=Q {(h8Q Kb7 Qe8
Kb6 Qxe6+ Kc7 Qd6+ Kb7 e6 g5 Ke4 gxh4 e7 h3 e8Q Ka7 Qeb8+) +M9/21 2} Kb7
{(Kb7 Qg7+ Kc8 Qg8+ Kd7 Qxg6 Kc7 h5 Kd7 h6 Kc7 h7 Kb7 h8Q Ka6 Qhg7 Kb6
Qxe6+) -M10/6 0} 7. Qd8 {(Qd8 Kc6 b4 cxb4 Qc8+ Kb6 c5+ Ka7 Qd7+ Kb8 c6 Ka8
Qb7+) +M7/22 2} Ka7 {(Ka7 Qc8 Kb6 Qb8+ Kc6 Qe8+ Kc7 Qxg6 Kc6 h5 Kd7 h6 Kc7
h7 Kb7 h8Q Ka6 Qhg7 Kb6 Qxe6+) -M11/9 0} 8. b4 {(b4 cxb4 Qc7+ Ka8 c5 g5 c6
g4+ Kxg4 b3 Qb7+) +M6/23 2} g5 {(g5 hxg5 cxb4 Qc7+ Ka8 g6 b3 g7 b2 g8Q+)
-M5/10 0 Black resigns} 1-0
In the PV it sees the ep capture, but it scores 0.0 at depth 42. Seemingly, something happens at search after that capture. Maurizio reports a different output and score between Fritz 12 and Fritz 13 GUI, while Robert reports it (apparently) fixed with 1.1a version.

It'd be useful to know what GUI is your beta tester using.

Hi,
i looked better.
If i set a fixed time per move, ''2 seconds'', even in FRITZ 13 GUI Chiron 1.1a moves King b7 ... waiting for white's move King e4 and planning ....h5 (??) with the analysis gxh6...Kb8, h7 and score 0.00!
I am a bit confused.

Always using FRITZ 13 GUI now i try with 1.1 with 2 seconds per move:

same result, ...Kb7...waiting for Ke4...h5 (??) gxh6..Kb8, h7 and still score 0.00!

Always using FRITZ 13 GUI, if i give infinite time to think to the engines, black has 7 legal moves, h5 and h6 are instantly discarded cause checkmate in 12 or something.

Using FRITZ 12 GUI, Chiron 1.1, 2 seconds per move, plays: Kb7..(waiting for Ke4) and stops analysis so i try to play Ke4 and...it horribly plays ...h5 (??). I did it, of course, with ponder OFF.

Chiron 1.1a has exactly the same behaviour.

Again, if i run infinite analysis, Chiron 1.1a shows 7 legal moves and sees instantly that h5 and h6 drive to a checkmate in 14 or something and Chiron 1.1 the same.

Before probably i missed something, it seemed to me that there were different behaviours of the GUI, now it seems to me that it's a problem of the engine that in that position, after King b7 and King e4 likes to play h5...
Sorry for all these informations, i have no much experience with these kinds of test.

Anyway it seems to me that Mr Vida is right, it seems that Chiron doesn't consider en passant move, at least in some specific situations (perhaps it was in terrible time trouble?!)

Regards

I just came to the same conclusions after testing again :shock: I hope the author makes note of this.
Top
MM
Posts: 766
Joined: Sun Oct 16, 2011 9:25 am

Re: Chiron 1.1 bug?

Post by MM » Sun Dec 25, 2011 3:40 am

rodolfoleoni wrote:
MM wrote: Hi,
i looked better.
If i set a fixed time per move, ''2 seconds'', even in FRITZ 13 GUI Chiron 1.1a moves King b7 ... waiting for white's move King e4 and planning ....h5 (??) with the analysis gxh6...Kb8, h7 and score 0.00!
I am a bit confused.

Always using FRITZ 13 GUI now i try with 1.1 with 2 seconds per move:

same result, ...Kb7...waiting for Ke4...h5 (??) gxh6..Kb8, h7 and still score 0.00!

Always using FRITZ 13 GUI, if i give infinite time to think to the engines, black has 7 legal moves, h5 and h6 are instantly discarded cause checkmate in 12 or something.

Using FRITZ 12 GUI, Chiron 1.1, 2 seconds per move, plays: Kb7..(waiting for Ke4) and stops analysis so i try to play Ke4 and...it horribly plays ...h5 (??). I did it, of course, with ponder OFF.

Chiron 1.1a has exactly the same behaviour.

Again, if i run infinite analysis, Chiron 1.1a shows 7 legal moves and sees instantly that h5 and h6 drive to a checkmate in 14 or something and Chiron 1.1 the same.

Before probably i missed something, it seemed to me that there were different behaviours of the GUI, now it seems to me that it's a problem of the engine that in that position, after King b7 and King e4 likes to play h5...
Sorry for all these informations, i have no much experience with these kinds of test.

Anyway it seems to me that Mr Vida is right, it seems that Chiron doesn't consider en passant move, at least in some specific situations (perhaps it was in terrible time trouble?!)

Regards
Hi Maurizio,

I consider if a move is in the PV then engine has considered it. The problem here is, something wrong gets pruned right after that. Since you say the move h7 is in the PV too, it could be the promotion. But the strange thing is that the bug depends on the TCs. I think only Ubaldo Andrea Farina can try to figure out what really happens. As you own the engine, you could report this bug via mail.

You could run a different test too. You could delete some pawns from the position to see if it's really related to pawn blockage detection code. Or you could set some more pawn-blocked positions with a potential ep capture to have a confirm about that kind of bug.
Hi Rodolfo,

thank you for the advice. I will make some tries and i will send this bug to Mr Farina.

Best Regards
MM
Top
Post Reply

Return to “Computer Chess Club: General Topics”