My first two Cheng3 games

[mar]

Is it too early to suspect a pattern?

[Event "40 Moves in 3 min"]
[Site "Cheng3 1.02 64-bit - Apr 25, LUCIFER"]
[Date "2011.04.25"]
[Round "1.1"]
[White "Cheng3 1.02 64-bit"]
[Black "Alex 2.07a"]
[Result "0-1"]

1. e4 {book 0s} c5 {book 0s} 2. Nf3 {book 0s} d6 {book 0s}
3. d4 {book 0s} cxd4 {book 0s} 4. Nxd4 {book 0s} Nf6 {book
0s} 5. Nc3 {book 0s} a6 {book 0s} 6. Be3 {book 0s} e5 {book
0s} 7. Nb3 {book 0s} Be6 {book 0s} 8. f3 {book 0s} Be7
{book 0s} 9. Nd5 {-0.04/12 3s} Nxd5 {-0.10/10 6s} 10. exd5
{-0.09/13 2s} Bd7 {0.00/11 5s (Bf5)} 11. Bd3 {+0.11/12 5s}
O-O {+0.05/10 5s} 12. O-O {+0.16/11 2s (c3)} a5 {+0.08/10
6s (f5)} 13. f4 {+0.44/12 5s (c3)} a4 {-0.07/10 5s}
14. fxe5 {+0.44/12 5s (Nd2)} axb3 {-2.02/9 1s (Bg5)}
15. Qh5 {+2.99/10 4s (axb3)} g6 {-2.05/11 5s (h6)}
16. Qxg6+ {+M2/60 0s (Qf3)} fxg6 {-10.86/11 5s (Kh8) time}
0-1

[Event "40 Moves in 3 min"]
[Site "Cheng3 1.02 64-bit - Apr 25, LUCIFER"]
[Date "2011.04.25"]
[Round "1.2"]
[White "Anechka 0.08"]
[Black "Cheng3 1.02 64-bit"]
[Result "1-0"]

1. e4 {book 0s} c5 {book 0s} 2. Nf3 {book 0s} d6 {book 0s}
3. d4 {book 0s} cxd4 {book 0s} 4. Nxd4 {book 0s} Nf6 {book
0s} 5. Nc3 {book 0s} a6 {book 0s} 6. Be3 {book 0s} e5 {book
0s} 7. Nb3 {book 0s} Be6 {book 0s} 8. f3 {book 0s} Be7
{book 0s} 9. Be2 {+0.11/9 4s} Nbd7 {-0.05/12 5s (O-O)}
10. Nd5 {+0.13/9 3s (f4)} Nxd5 {-0.18/13 4s} 11. exd5
{+0.10/12 5s} Bf5 {-0.12/13 3s} 12. O-O {+0.09/11 5s} O-O
{-0.13/12 3s} 13. f4 {+0.11/10 6s} Be4 {-0.14/11 2s (Rc8)}
14. fxe5 {+0.18/9 4s (Bd3)} Nxe5 {-0.26/11 3s} 15. c4
{+0.17/10 6s} Qc8 {-0.19/10 2s (Rc8)} 16. Rc1 {+0.21/9 4s
(Bf4)} Bxg2 {-3.06/11 3s (Qd7)} 17. Kxg2 {+1.95/10 5s
(Rf2)} Qg4+ {-3.74/14 5s (Qd7)} 18. Bxg4 {+11.25/10 5s
(Kh1) time} 1-0

Hi Gabor,
do you suspect a bug in time control? did cheng3 flag? I recently played
a 100-game tournament (40 moves in 4 min repeating) and it didn't flag in a single case (used Arena 3.0). I've only tested the 32-bit version though but there should be no difference...

Regards
Martin

[mar]

On the second thought it could have been a crash as well?

[mar]

huh such blunders...in the first game my version doesn't even consider 14. fxe5 (plays Nd2)... even after the blunder it then considers axb3 instead of Qh5?? and at move 16 it would play Qh6 instead of Qxg6+??
seems that there's a strange bug that only occurs in 64-bit versions?... So I recommend using the 32-bit version instead. Unfortunately I can't test the 64-bit versions...
or perhaps try both 64-bit versions. if the problem persists I'll probably remove the 64-bit versions until the problem is resolved. please let me know
whether the 32-bit version has same problems.

thanks
Martin

[mar]

Ok I just found the bug. It only affects 64-bit versions. The fixed source is here:
http://www.mediafire.com/?8raah1c83d4bsxw. So Dann please if you read this could you please recompile the sources and upload the fixed executables somewhere?

Thanks and sorry for that
Martin