Page 1 of 1

Something seriously wrong with Chispa

PostPosted: 15 Dec 2004, 10:16
by Gábor Szots
This game has just been finished. Look at the scores Chispa assigns to its moves. Its evaluation is still around 0 only a few moves before being mated!
The game was played under Arena, Chispa installed as an UCI engine.

[Event "Kar?csony"]
[Site "SZOTS"]
[Date "2004.12.15"]
[Round "1"]
[White "KnightX 1.90a"]
[Black "Chispa 4.0"]
[Result "1-0"]
[ECO "C28"]
[Opening "Vienna Game"]
[Time "10:01:02"]
[Variation "4.d3"]
[TimeControl "180+1"]
[Comment " white engine resigns for black (illegal)"]
[Termination "normal"]
[PlyCount "75"]
[WhiteType "program"]
[BlackType "program"]

1.e4 {0.00/2} e5 2.Nc3 {0.00/2} Nf6 3.Bc4 {0.00/2} Nc6 4.d3 {0.00/2} Bc5 5.Bg5
{0.00/2} h6 6.Bxf6 {0.00/2} Qxf6 {+0.36/7 7} 7.Nf3 {0.00/2} d6 8.Nd5 {0.00/2} Qd8 9.c3
{0.00/2} Ne7 10.Qa4+ {+0.39/9 8} c6 {+0.12/5 7} 11.Nxe7 {+0.04/8 5} Qxe7 {+0.29/6 6}
12.b4 {-0.33/9 7} Bb6 {+0.26/6 6} 13.b5 {-0.43/9 4} d5 {+0.67/6 9} 14.bxc6 {+1.76/10
6} O-O {+0.68/5 11} 15.Bxd5 {+1.95/10 5} bxc6 {+0.12/6 10} 16.Bxc6 {+1.81/10 4}
Rd8 {+0.13/7 10} 17.Bxa8 {+4.82/9 6} Rxd3 {+0.17/7 9} 18.Qc4 {+6.24/9 5} Qd8
{+0.15/6 9} 19.Bd5 {+8.00/9 6} Rxd5 {+0.09/7 8} 20.Qxd5 {+8.99/9 6} Qe7 {+0.08/7 7}
21.Qxe5 {+9.18/10 5} Be6 {+0.08/7 5} 22.O-O Bc7 {+0.13/8 7} 23.Qb5 {+9.29/8 5} Bd7
{+0.24/7 5} 24.Qb7 {+9.43/9 4} Be8 {+0.07/7 5} 25.e5 {+9.75/8 4} Qc5 {-0.01/7 4} 26.c4
{+9.73/9 3} Qc6 {-0.22/7 6} 27.Qxa7 {+10.64/10 4} Bb6 {+0.10/8 4} 28.Qb8 {+10.66/10 4}
Kh7 {-0.22/8 10} 29.Rab1 {+11.09/9 3} Bc7 {-0.31/7 2} 30.Qa7 {+12.16/10 3} Kg6
{-0.19/6 1} 31.Rb7 {+13.88/9 3} Qxf3 {-0.43/7 1} 32.gxf3 {+20.79/8} Bxe5 {-0.35/8 2}
33.Re1 {+22.06/9 3} Bc6 {+0.09/6 2} 34.Rxe5 {+30.39/10 4} h5 {-1.56/7 2} 35.Rxf7
{+m414/7 2} Bd7 {-m3/6 2} 36.Qxd7 {+m413/5} Kh6 {-m2/6 2} 37.Qf5 {+m413/4} h4 {-m1/6 1}
38.Qh5# {+m413/3} {matt} 1-0

Re: Something seriously wrong with Chispa

PostPosted: 15 Dec 2004, 11:06
by Gábor Szots
G?bor Szots wrote:This game has just been finished. Look at the scores Chispa assigns to its moves. Its evaluation is still around 0 only a few moves before being mated!
The game was played under Arena, Chispa installed as an UCI engine.

[Event "Kar?csony"]
[Site "SZOTS"]
[Date "2004.12.15"]
[Round "1"]
[White "KnightX 1.90a"]
[Black "Chispa 4.0"]
[Result "1-0"]
[ECO "C28"]
[Opening "Vienna Game"]
[Time "10:01:02"]
[Variation "4.d3"]
[TimeControl "180+1"]
[Comment " white engine resigns for black (illegal)"]
[Termination "normal"]
[PlyCount "75"]
[WhiteType "program"]
[BlackType "program"]

1.e4 {0.00/2} e5 2.Nc3 {0.00/2} Nf6 3.Bc4 {0.00/2} Nc6 4.d3 {0.00/2} Bc5 5.Bg5
{0.00/2} h6 6.Bxf6 {0.00/2} Qxf6 {+0.36/7 7} 7.Nf3 {0.00/2} d6 8.Nd5 {0.00/2} Qd8 9.c3
{0.00/2} Ne7 10.Qa4+ {+0.39/9 8} c6 {+0.12/5 7} 11.Nxe7 {+0.04/8 5} Qxe7 {+0.29/6 6}
12.b4 {-0.33/9 7} Bb6 {+0.26/6 6} 13.b5 {-0.43/9 4} d5 {+0.67/6 9} 14.bxc6 {+1.76/10
6} O-O {+0.68/5 11} 15.Bxd5 {+1.95/10 5} bxc6 {+0.12/6 10} 16.Bxc6 {+1.81/10 4}
Rd8 {+0.13/7 10} 17.Bxa8 {+4.82/9 6} Rxd3 {+0.17/7 9} 18.Qc4 {+6.24/9 5} Qd8
{+0.15/6 9} 19.Bd5 {+8.00/9 6} Rxd5 {+0.09/7 8} 20.Qxd5 {+8.99/9 6} Qe7 {+0.08/7 7}
21.Qxe5 {+9.18/10 5} Be6 {+0.08/7 5} 22.O-O Bc7 {+0.13/8 7} 23.Qb5 {+9.29/8 5} Bd7
{+0.24/7 5} 24.Qb7 {+9.43/9 4} Be8 {+0.07/7 5} 25.e5 {+9.75/8 4} Qc5 {-0.01/7 4} 26.c4
{+9.73/9 3} Qc6 {-0.22/7 6} 27.Qxa7 {+10.64/10 4} Bb6 {+0.10/8 4} 28.Qb8 {+10.66/10 4}
Kh7 {-0.22/8 10} 29.Rab1 {+11.09/9 3} Bc7 {-0.31/7 2} 30.Qa7 {+12.16/10 3} Kg6
{-0.19/6 1} 31.Rb7 {+13.88/9 3} Qxf3 {-0.43/7 1} 32.gxf3 {+20.79/8} Bxe5 {-0.35/8 2}
33.Re1 {+22.06/9 3} Bc6 {+0.09/6 2} 34.Rxe5 {+30.39/10 4} h5 {-1.56/7 2} 35.Rxf7
{+m414/7 2} Bd7 {-m3/6 2} 36.Qxd7 {+m413/5} Kh6 {-m2/6 2} 37.Qf5 {+m413/4} h4 {-m1/6 1}
38.Qh5# {+m413/3} {matt} 1-0


Here is another one:

[Event "Kar?csony"]
[Site "SZOTS"]
[Date "2004.12.15"]
[Round "1"]
[White "Chispa 4.0"]
[Black "Glaurung 0.1.5"]
[Result "0-1"]
[ECO "D45"]
[Opening "QGD"]
[Time "10:58:24"]
[Variation "Semi-Slav (5.e3)"]
[TimeControl "180+1"]
[Termination "normal"]
[PlyCount "67"]
[WhiteType "program"]
[BlackType "program"]

1.d4 d5 2.c4 c6 3.Nc3 Nf6 4.Nf3 e6 5.e3 Be7 6.Bd3 {+0.17/7 12} dxc4 7.Bc2
{+0.09/8 12} Nbd7 {+0.82/10 6} 8.Qe2 {+0.10/8 11} b5 {+0.86/11 9} 9.O-O {+0.10/8 11} Bb7
{+0.93/10 5} 10.e4 {+0.14/7 10} Qb6 {+0.93/10 9} 11.e5 {+0.44/6 9} Nd5 {+1.00/10 5} 12.a4
{+0.14/6 9} b4 {+0.96/10 7} 13.a5 {+0.29/6 8} Qa6 {+0.99/10 4} 14.Nxd5 {+0.34/6 16} cxd5
{+1.03/11 4} 15.Bg5 {+0.19/6 4} Bd8 {+1.11/10 5} 16.Ba4 {+0.44/5 10} Bxa5 {+1.18/10 9}
17.Qc2 {+0.51/6 6} b3 {+1.80/10 4} 18.Qd2 {+0.41/6 5} Bxd2 {+4.46/14 5} 19.Bxd2
{-0.06/7 3} Bc6 {+8.18/13 3} 20.Bxc6 {-0.11/7 4} Qxc6 {+8.30/12 5} 21.Rfe1 {-0.18/7 4}
O-O {+8.51/12 5} 22.Ra3 {-0.24/6 2} f6 {+8.53/11 3} 23.Raa1 {-0.47/6 2} fxe5
{+8.65/12 5} 24.Nxe5 {-0.64/8 2} Nxe5 {+8.70/12 2} 25.Rxe5 {-0.62/6 2} Qb6 {+8.71/12 4}
26.Be3 {-0.77/6 1} c3 {+10.04/10 5} 27.bxc3 {-0.70/6 2} b2 {+10.09/9 2} 28.Rb1
{-0.43/6 2} Qb3 {+10.96/12 4} 29.f3 {-0.55/6 2} Qa2 {+12.55/12 3} 30.Rxb2 {-0.27/7 1}
Qa1+ {+13.24/11 4} 31.Rb1 {-0.55/7 2} Qxb1+ {+13.88/11 2} 32.Kf2 {-0.46/7 2} Qc2+
{+13.85/11 3} 33.Kg3 {-0.38/6 2} Rab8 {+16.96/11 6} {0-1 Felhaszn?l?i d?nt?s} 0-1

same problem here

PostPosted: 15 Dec 2004, 18:46
by Dr. Axel Schumacher
I tried to run Chispa on my Athlon and it lost every game (out of 40), even against engines of Elo 1500.
Something is broken it seems.

Axel

Re: Something seriously wrong with Chispa

PostPosted: 15 Dec 2004, 21:22
by Uri Blass
If you add the letters PR9 to chispa it seems to play better and it even can beat movei(inspite of losing the match 3-1)

[Event "Test Tourney 165"]
[Site "Mariano Comense"]
[Date "2004.12.08"]
[Round "12.1"]
[White "Chispa 4.0 PR9"]
[Black "Movei 00.8.295"]
[Result "1-0"]
[ECO "D19"]
[WhiteElo "2220"]
[BlackElo "2419"]
[PlyCount "121"]
[EventDate "2004.12.05"]
[TimeControl "300"]

1. d4 d5 2. c4 c6 3. Nf3 Nf6 4. Nc3 dxc4 5. a4 Bf5 6. e3 e6 7. Bxc4 Bb4 8. O-O
O-O 9. Qe2 Bg6 10. Ne5 Nbd7 11. Nxg6 hxg6 12. Rd1 Qa5 13. Qc2 Qh5 14. e4 e5 15.
d5 Rac8 16. Be2 Qh4 17. g3 Qh3 18. Bf1 Qh5 19. h4 Qg4 20. Be2 Qh3 21. dxc6 Bc5
22. Qd3 Rxc6 23. Qf3 Bd4 24. Rb1 Nc5 25. Bf1 Qg4 26. Be2 Qxf3 27. Bxf3 Nb3 28.
Be3 Bxe3 29. fxe3 Rc4 30. Kg2 Rfc8 31. g4 Rb4 32. g5 Nh5 33. Rd5 Nc5 34. Bg4
Rc6 35. Rxe5 Nxa4 36. Nxa4 Rxa4 37. Re8+ Kh7 38. Re7 f6 39. Be6 Rc2+ 40. Kf3
Rb4 41. b3 Rh2 42. Rf1 Ra2 43. Rc1 b6 44. Rcc7 fxg5 45. hxg5 Rb5 46. e5 Rc5 47.
Rxc5 bxc5 48. b4 Ra1 49. bxc5 Rf1+ 50. Kg2 Rc1 51. Re8 Rc2+ 52. Kf3 Rc1 53.
Bg8+ Kh8 54. c6 a5 55. Bf7+ Kh7 56. Bd5 a4 57. e6 Rf1+ 58. Kg2 Rf5 59. c7 Rxg5+
60. Kf3 Nf6 61. c8=Q {Movei->black resign} 1-0

Uri

Re: Something seriously wrong with Chispa

PostPosted: 16 Dec 2004, 04:06
by Federico Corigliano
Seems that the problem is only in UCI mode because don't uses the configuration file chispa.ini. The problem is caused by the default piece values. I fixed the problem and I released Chispa 4.0.1. Thanks for the bug report.