Page 1 of 1

Glaurung 1.02(BH) bug found?

PostPosted: 28 Feb 2006, 21:33
by Guenther Simon
Hi Tord,

It seems I have found a bug in Glaurung 1.02.
Glaurung crashed here after it reached depth 60 before
it could make the last move for a 3 time repetition.
(Max depth problem?)
I have played already 118 games with it at 40/15min
and this was the first time a problem occured.
(I have also a Polyglot log in case you need it)

[Event "GS15 Test 1"]
[Site "ESPRESSO"]
[Date "2006.02.28"]
[Round "2"]
[White "The Baron 1.7.1b8"]
[Black "Glaurung_102BH"]
[Result "*"]
[TimeControl "40/900"]
[Annotator "7. +0.27 7... -0.14"]

1. d4 Nf6 2. c4 e6 3. Nc3 Bb4 4. e3 O-O 5. Bd3 c5 6. Ne2 d5 7.
a3 {+0.27/10} Bxc3+ {-0.14/14} 8. Nxc3 {+0.39/12} b6 {-0.07/14} 9.
dxc5 {+0.42/11} bxc5 {-0.03/14} 10. O-O {+0.22/11} Nc6 {+0.18/14} 11.
cxd5 {+0.26/10} exd5 {-0.15/15} 12. e4 {+0.24/10} d4 {-0.39/14} 13.
Na4 {+0.38/10} Qd6 {-0.15/14} 14. f4 {+0.25/10} c4 {+0.04/14} 15.
e5 {+0.21/11} Qd5 {-0.12/15} 16. Be2 {+0.38/10} Nd7 {+0.15/13} 17.
Nc3 {+0.65/10} Qa5 {+0.20/12} 18. Ne4 {+0.70/10} d3 {+0.14/11} 19.
Bf3 {+0.55/9} Nd4 {+0.39/12} 20. Nd6 {+0.26/8} Rb8 {+0.25/12} 21.
Bd2 {+0.47/8} Qc5 {+0.40/12} 22. Bb4 {+0.41/9} Nxf3+ {+0.28/13} 23. Kh1
Qd5 {+0.28/13} 24. Qxf3 {+0.49/10} Qxf3 {+0.15/13} 25. Rxf3 {+0.63/11}
Ba6 {+0.14/14} 26. Rg3 {+1.06/10} g6 {+0.00/14} 27. Rf1 {+1.03/10}
Nb6 {+0.45/12} 28. Ne4 {+0.91/10} Rfd8 {+0.45/12} 29. Be7 {+0.99/9}
Rdc8 {+0.60/12} 30. Nf6+ {+0.75/9} Kg7 {+0.53/12} 31. f5 {+0.86/9}
Rb7 {+0.12/11} 32. fxg6 {+0.97/9} fxg6 {+0.00/11} 33. Bb4 {+0.49/9}
Na4 {+0.34/11} 34. Rh3 {+0.71/8} h5 {+0.04/12} 35. Rg3 {+0.75/8}
h4 {+0.00/11} 36. Rg4 {+0.40/9} Nxb2 {+0.00/11} 37. Rxh4 {+0.35/8}
g5 {+0.00/12} 38. Rg4 {+0.00/7} c3 {+0.00/12} 39. Rxg5+ {+0.00/7}
Kh6 {+0.00/12} 40. h4 {+0.00/7} d2 {+0.00/11} 41. Bxc3 {+0.00/9}
Rxc3 {+0.00/15} 42. Ng4+ {+0.00/11} Kh7 {+0.00/5} 43. Nf6+ {+0.00/12}
Kh6 {+0.00/16} 44. Ng4+ {+0.00/13} Kh7 {+0.00/5} 45. Nf6+
*

Code: Select all
4698095 <first : move h6h7
Interrupting second
4698095 >second: time 86317
4698095 >second: otim 86036
4698095 >second: h6h7
4721018 <second: 12. 0 22 5168459 Ng4-f6 Kh7-h6 Nf6-g4 Kh6-h7 <rep>
4721018 <second: move g4f6
4721018 >first : time 86036
4721018 >first : otim 84024
4721018 >first : usermove 4721018 >first : g4f6
4721059 <first : 2 +0 0 6 Kh6 Ng4+
4721059 <first : 3 +0 0 30 Kh6 Ng4+
4721059 <first : 4 +0 0 56 Kh6 Ng4+
4721059 <first : 5 +0 0 82 Kh6 Ng4+
4721059 <first : 6 +0 0 108 Kh6 Ng4+
4721059 <first : 7 +0 0 134 Kh6 Ng4+
4721059 <first : 8 +0 0 160 Kh6 Ng4+
4721059 <first : 9 +0 0 186 Kh6 Ng4+
4721059 <first : 10 +0 0 212 Kh6 Ng4+
4721059 <first : 11 +0 0 238 Kh6 Ng4+
4721059 <first : 12 +0 0 264 Kh6 Ng4+
4721059 <first : 13 +0 0 290 Kh6 Ng4+
4721159 <first : 14 +0 11 37070 Kh6 Ng4+
4726286 <first : 15 +0 524 1946628 Kh6 Ng4+
4742149 <first : 16 +0 2109 7792472 Kh6 Ng4+
4764070 <first : 16 +0 4209 15600000 Kh6 Ng4+
4764070 <first : move h7h6
Interrupting second
4764070 >second: time 84024
4764070 >second: otim 81731
4764070 >second: h7h6
4773334 <second: 12. 0 9 2212212 Nf6-g4 Kh6-h7 Ng4-f6 Kh7-h6 <rep>
4798470 <second: 13& 0 34 8353647 Nf6-g4 Kh6-h7 Ng4-f6 Kh7-h6 <rep>
4818068 <second: 13. 0 53 13056664 Nf6-g4 Kh6-h7 Ng4-f6 Kh7-h6 <rep>
4818068 <second: move f6g4
4818068 >first : time 81731
4818068 >first : otim 78624
4818068 >first : usermove 4818068 >first : f6g4
4818078 <first : 2 +0 0 2 Kh7
4818078 <first : 3 +0 0 3 Kh7
4818078 <first : 4 +0 0 4 Kh7
4818078 <first : 5 +0 0 5 Kh7
4818078 <first : 5 +0 0 5 Kh7
4818078 <first : move h6h7
Interrupting second
4818078 >second: time 78624
4818078 >second: otim 81730
4818078 >second: h6h7
4890082 <second: move g4f6
4890092 >first : time 81730
4890092 >first : otim 71423
4890092 >first : usermove 4890092 >first : g4f6
4890202 <first : 2 -9998 0 7 Kh8 Rg8#
4890222 <first : 2 +0 0 7 Kh6
4890222 <first : 3 +0 0 8 Kh6
4890222 <first : 4 +0 0 11 Kh6
4890222 <first : 5 +0 0 14 Kh6
4890222 <first : 6 +0 0 17 Kh6
4890222 <first : 7 +0 0 20 Kh6
4890222 <first : 8 +0 0 24 Kh6
4890222 <first : 9 +0 0 28 Kh6
4890222 <first : 10 +0 0 33 Kh6
4890222 <first : 11 +0 0 38 Kh6
4890222 <first : 12 +0 0 44 Kh6
4890222 <first : 13 +0 0 50 Kh6
4890222 <first : 14 +0 0 57 Kh6
4890222 <first : 15 +0 0 64 Kh6
4890222 <first : 16 +0 0 72 Kh6
4890222 <first : 17 +0 0 80 Kh6
4890222 <first : 18 +0 1 89 Kh6
4890222 <first : 19 +0 1 98 Kh6
4890222 <first : 20 +0 1 108 Kh6
4890222 <first : 21 +0 1 118 Kh6
4890222 <first : 22 +0 1 129 Kh6
4890222 <first : 23 +0 1 140 Kh6
4890222 <first : 24 +0 1 152 Kh6
4890222 <first : 25 +0 1 164 Kh6
4890222 <first : 26 +0 1 177 Kh6
4890222 <first : 27 +0 1 190 Kh6
4890222 <first : 28 +0 1 204 Kh6
4890222 <first : 29 +0 1 218 Kh6
4890222 <first : 30 +0 1 233 Kh6
4890222 <first : 31 +0 1 248 Kh6
4890222 <first : 32 +0 1 264 Kh6
4890222 <first : 33 +0 1 280 Kh6
4890222 <first : 34 +0 1 297 Kh6
4890222 <first : 35 +0 1 314 Kh6
4890222 <first : 36 +0 1 332 Kh6
4890222 <first : 37 +0 1 350 Kh6
4890222 <first : 38 +0 1 369 Kh6
4890222 <first : 39 +0 1 388 Kh6
4890222 <first : 40 +0 1 408 Kh6
4890222 <first : 41 +0 1 428 Kh6
4890222 <first : 42 +0 1 449 Kh6
4890222 <first : 43 +0 1 470 Kh6
4890222 <first : 44 +0 1 492 Kh6
4890222 <first : 45 +0 1 514 Kh6
4890222 <first : 46 +0 1 537 Kh6
4890222 <first : 47 +0 1 560 Kh6
4890222 <first : 48 +0 1 584 Kh6
4890222 <first : 49 +0 1 608 Kh6
4890222 <first : 50 +0 1 633 Kh6
4890222 <first : 51 +0 1 658 Kh6
4890222 <first : 52 +0 1 684 Kh6
4890222 <first : 53 +0 1 710 Kh6
4890222 <first : 54 +0 1 737 Kh6
4890222 <first : 55 +0 1 764 Kh6
4890222 <first : 56 +0 1 792 Kh6
4890222 <first : 57 +0 1 820 Kh6
4890222 <first : 58 +0 2 849 Kh6
4890222 <first : 59 +0 2 878 Kh6
4890222 <first : 60 +0 2 908 Kh6
Fatal Error: Error: first chess program (Glaurung_102BH.exe) exited unexpectedly
GameEnds(0, (null), 2)
4914266 >first : force
4914266 >first : ping 6
Interrupting second
4914266 >second: force
4914266 >second: ping 4
Exiting: Match Glaurung_102BH vs. The Baron 1.7.1b8: final score 0-1-1
4914366 >first : quit

Re: Glaurung 1.02(BH) bug found?

PostPosted: 01 Mar 2006, 00:13
by Tord Romstad
Guenther Simon wrote:Hi Tord,

It seems I have found a bug in Glaurung 1.02.
Glaurung crashed here after it reached depth 60 before
it could make the last move for a 3 time repetition.
(Max depth problem?)
I have played already 118 games with it at 40/15min
and this was the first time a problem occured.

This is strange. Glaurung 1.0.2 has been very thoroughly tested, and I haven't seen anyone report any crashes until now. I'll have a look at it and see what I can do.

(I have also a Polyglot log in case you need it)

Yes, that would be useful. Please send me the log.

Thanks for the bug report!

Tord

Re: Glaurung 1.02(BH) bug found?

PostPosted: 01 Mar 2006, 01:26
by Guenther Simon
This is strange. Glaurung 1.0.2 has been very thoroughly tested, and I haven't seen anyone report any crashes until now. I'll have a look at it and see what I can do.

Quote:
(I have also a Polyglot log in case you need it)

Yes, that would be useful. Please send me the log.

Thanks for the bug report!

Tord


Hi again,

I have looked into the Polyglot log and it seems Glaurung fell into a loop
for the two only possible moves Kh8 and Kh6.
After seldepth 60 I counted the incredible number of 29.962! lines
of such info: (I will send the log after I have cut it from the other games)

Guenther

Code: Select all
> XBOARD 60 +0 2 908 Kh6
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE info nodes 10200000 nps 10189810 time 1001
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
.
.
.
.
.
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE info currmove h7h6 currmovenumber 1
< ENGINE info currmove h7h8 currmovenumber 2
< ENGINE EOF


Re: Glaurung 1.02(BH) bug found?

PostPosted: 01 Mar 2006, 01:44
by Tord Romstad
Guenther Simon wrote:Hi again,

I have looked into the Polyglot log and it seems Glaurung fell into a loop
for the two only possible moves Kh8 and Kh6.
After seldepth 60 I counted the incredible number of 29.962! lines
of such info: (I will send the log after I have cut it from the other games)

Yes, this is a well known output bug (fixed in Glaurung CCT8). It creates a very annoying amount of debug output, but is otherwise harmless. I am fairly sure it was not this bug that caused the crash.

Tord