Sharper 0.12 released

[Albert Bertilsson]

Geschrieben von: / Posted by: Albert Bertilsson at 31 July 2003 15:34:53:

Sharper 0.12 is released
A few bug fixes, really recomended that you download this.
/Regards Albert

[Olivier Deville]

Geschrieben von: / Posted by: Olivier Deville at 31 July 2003 16:56:45:
Als Antwort auf: / In reply to: Sharper 0.12 released geschrieben von: / posted by: Albert Bertilsson at 31 July 2003 15:34:53:

Sharper 0.12 is released
A few bug fixes, really recomended that you download this.
/Regards Albert

Hi Albert ! Sharper is still losing on time here after the 40th move.
Olivier


Chesswar

[Albert Bertilsson]

Geschrieben von: / Posted by: Albert Bertilsson at 31 July 2003 20:59:07:
Als Antwort auf: / In reply to: Re: Sharper 0.12 released geschrieben von: / posted by: Olivier Deville at 31 July 2003 16:56:45:

Please send me logs without that there isn't much I can do.
Preferably I'd like to have the logs from sharper (log.txt) logging can be turned on with "logon" in the .ini file.
/Regards Albert

[Olivier Deville]

Geschrieben von: / Posted by: Olivier Deville at 31 July 2003 21:13:00:
Als Antwort auf: / In reply to: Please give me logs, I'll fix this one even if it is the last thing I do! geschrieben von: / posted by: Albert Bertilsson at 31 July 2003 20:59:07:

[Albert Bertilsson]

Geschrieben von: / Posted by: Albert Bertilsson at 31 July 2003 21:30:34:
Als Antwort auf: / In reply to: Sent (n/t) geschrieben von: / posted by: Olivier Deville at 31 July 2003 21:13:00:

level 40 4 0
force
e2e4
e7e5
b1c3
g8f6
f1c4
f8c5
g1f3
time 24000
otim 23700
go
force
d2d3
e8g8
e1g1
time 23700
otim 23500
go

Time: 240000 ms, Type: Tournament, Inc: 0 s, Moves: 40, Moves left: 40, Use: 5950 ms
rnbqk2r/pppp1ppp/5n2/2b1p3/2B1P3/2N2N2/PPPP1PPP/R1BQK2R b KQkq - 5 4
--------------------------------------------------------------
| move | freq | played | won | drawn | lost | prob |
--------------------------------------------------------------
| b8c6 | 1 | 0 | 0 | 0 | 0 | 1.0000 |
--------------------------------------------------------------
Total time: 1810 ms
move b8c6
Time used: 1810 ms
Time: 237000 ms, Type: Tournament, Inc: 0 s, Moves: 40, Moves left: 39, Use: 6025 ms
r1bq1rk1/pppp1ppp/2n2n2/2b1p3/2B1P3/2NP1N2/PPP2PPP/R1BQ1RK1 b - - 2 6
1 10 0 74 c5d6
2 -38 0 524 c5d6 c1d2
3 0 6 2335 d7d6 c1d2 c8d7
4 -11 11 9216 d7d6 c1d2 c8d7 f1e1
5 -11 55 33682 d7d6 c1d2 c8e6 f1e1 e6d7
6 -11 319 182847 d7d6 c3a4 c6a5 a4c5 a5c4 f1e1
move d7d6
I don't understand what happens here, there are moves made in force mode?
Sharper runs out of time because the moves in force mode increases the GUI's counter of when to give the new increased time, but Sharper doesn't count them!
How do you use the GUI to get this effect? Is it common practice to do so?
/Regards Albert

[Olivier Deville]

Geschrieben von: / Posted by: Olivier Deville at 31 July 2003 21:50:34:
Als Antwort auf: / In reply to: I don't understand what did you do? geschrieben von: / posted by: Albert Bertilsson at 31 July 2003 21:30:34:

level 40 4 0
force
e2e4
e7e5
b1c3
g8f6
f1c4
f8c5
g1f3
time 24000
otim 23700
go
force
d2d3
e8g8
e1g1
time 23700
otim 23500
go

Time: 240000 ms, Type: Tournament, Inc: 0 s, Moves: 40, Moves left: 40, Use: 5950 ms
rnbqk2r/pppp1ppp/5n2/2b1p3/2B1P3/2N2N2/PPPP1PPP/R1BQK2R b KQkq - 5 4
--------------------------------------------------------------
| move | freq | played | won | drawn | lost | prob |
--------------------------------------------------------------
| b8c6 | 1 | 0 | 0 | 0 | 0 | 1.0000 |
--------------------------------------------------------------
Total time: 1810 ms
move b8c6
Time used: 1810 ms
Time: 237000 ms, Type: Tournament, Inc: 0 s, Moves: 40, Moves left: 39, Use: 6025 ms
r1bq1rk1/pppp1ppp/2n2n2/2b1p3/2B1P3/2NP1N2/PPP2PPP/R1BQ1RK1 b - - 2 6
1 10 0 74 c5d6
2 -38 0 524 c5d6 c1d2
3 0 6 2335 d7d6 c1d2 c8d7
4 -11 11 9216 d7d6 c1d2 c8d7 f1e1
5 -11 55 33682 d7d6 c1d2 c8e6 f1e1 e6d7
6 -11 319 182847 d7d6 c3a4 c6a5 a4c5 a5c4 f1e1
move d7d6
I don't understand what happens here, there are moves made in force mode?
Sharper runs out of time because the moves in force mode increases the GUI's counter of when to give the new increased time, but Sharper doesn't count them!
How do you use the GUI to get this effect? Is it common practice to do so?
/Regards Albert

Yes it is very common : the Fritz GUI plays the opening. You should read the messages below, it seems that the author of Kanguruh got the same problem - and Uri Blass had the solution.
I am a patzer at programming, but I feel this should not be too difficult to fix. Otherwise for the next games (but there are no more right now for Sharper, the tourney is finished) I can let Sharper play with its ownbook. Strangely enough, SharpChess seems to have no trouble with time control in my tourney (only 1 game lost on time out of 9).
Best
Olivier



Chesswar

[Albert Bertilsson]

Geschrieben von: / Posted by: Albert Bertilsson at 31 July 2003 21:57:50:
Als Antwort auf: / In reply to: Re: I don't understand what did you do? geschrieben von: / posted by: Olivier Deville at 31 July 2003 21:50:34:

Yes it is very common : the Fritz GUI plays the opening. You should read the messages below, it seems that the author of Kanguruh got the same problem - and Uri Blass had the solution.
I am a patzer at programming, but I feel this should not be too difficult to fix. Otherwise for the next games (but there are no more right now for Sharper, the tourney is finished) I can let Sharper play with its ownbook. Strangely enough, SharpChess seems to have no trouble with time control in my tourney (only 1 game lost on time out of 9).

Thanx a lot, you probably didn't have any problems before because the problems was related to how certain engines use the level command. I tested with Arena 0.92, 0.93 and WB 4.2.6. Now you have a third example of how engines use the protocol.
I'll study this log for a while and think about a solution that works for all these GUI's (hope there is no fourth special way to do it). I'll get back to you when I have anything worth trying.
/Regards Albert

[Günther Simon]

Geschrieben von: / Posted by: Günther Simon at 31 July 2003 22:15:01:
Als Antwort auf: / In reply to: Re: I don't understand what did you do? geschrieben von: / posted by: Olivier Deville at 31 July 2003 21:50:34:

level 40 4 0
force
e2e4
e7e5
b1c3
g8f6
f1c4
f8c5
g1f3
time 24000
otim 23700
go
force
d2d3
e8g8
e1g1
time 23700
otim 23500
go

Time: 240000 ms, Type: Tournament, Inc: 0 s, Moves: 40, Moves left: 40, Use: 5950 ms
rnbqk2r/pppp1ppp/5n2/2b1p3/2B1P3/2N2N2/PPPP1PPP/R1BQK2R b KQkq - 5 4
--------------------------------------------------------------
| move | freq | played | won | drawn | lost | prob |
--------------------------------------------------------------
| b8c6 | 1 | 0 | 0 | 0 | 0 | 1.0000 |
--------------------------------------------------------------
Total time: 1810 ms
move b8c6
Time used: 1810 ms
Time: 237000 ms, Type: Tournament, Inc: 0 s, Moves: 40, Moves left: 39, Use: 6025 ms
r1bq1rk1/pppp1ppp/2n2n2/2b1p3/2B1P3/2NP1N2/PPP2PPP/R1BQ1RK1 b - - 2 6
1 10 0 74 c5d6
2 -38 0 524 c5d6 c1d2
3 0 6 2335 d7d6 c1d2 c8d7
4 -11 11 9216 d7d6 c1d2 c8d7 f1e1
5 -11 55 33682 d7d6 c1d2 c8e6 f1e1 e6d7
6 -11 319 182847 d7d6 c3a4 c6a5 a4c5 a5c4 f1e1
move d7d6
I don't understand what happens here, there are moves made in force mode?
Sharper runs out of time because the moves in force mode increases the GUI's counter of when to give the new increased time, but Sharper doesn't count them!
How do you use the GUI to get this effect? Is it common practice to do so?
/Regards Albert

Yes it is very common : the Fritz GUI plays the opening. You should read the messages below, it seems that the author of Kanguruh got the same problem - and Uri Blass had the solution.
I am a patzer at programming, but I feel this should not be too difficult to fix. Otherwise for the next games (but there are no more right now for Sharper, the tourney is finished) I can let Sharper play with its ownbook. Strangely enough, SharpChess seems to have no trouble with time control in my tourney (only 1 game lost on time out of 9).
Best
Olivier

Its possible that a lot more programs in your group D
had the same problem, because a lot of weaker or newer
programs might not support edit or setboard mode.
Have you checked the pgn for other time losses?
Those programs should better all play with own/no book
instead...well another way would be to use WB or Arena
for group D.
Regards,
Günther

[Olivier Deville]

Geschrieben von: / Posted by: Olivier Deville at 01 August 2003 05:06:37:
Als Antwort auf: / In reply to: Re: I don't understand what did you do? geschrieben von: / posted by: Günther Simon at 31 July 2003 22:15:01:

Olivier

Its possible that a lot more programs in your group D
had the same problem, because a lot of weaker or newer
programs might not support edit or setboard mode.
Have you checked the pgn for other time losses?
Those programs should better all play with own/no book
instead...well another way would be to use WB or Arena
for group D.
Regards,
Günther

Yes, there are other losses on time : I though that these engines were simply buggy, but next time I'll let them play with their ownbook.
Olivier



Chesswar

[Olivier Deville]

Geschrieben von: / Posted by: Olivier Deville at 01 August 2003 05:10:34:
Als Antwort auf: / In reply to: Thanx alot, I'll be contacting you again... geschrieben von: / posted by: Albert Bertilsson at 31 July 2003 21:57:50:

Yes it is very common : the Fritz GUI plays the opening. You should read the messages below, it seems that the author of Kanguruh got the same problem - and Uri Blass had the solution.
I am a patzer at programming, but I feel this should not be too difficult to fix. Otherwise for the next games (but there are no more right now for Sharper, the tourney is finished) I can let Sharper play with its ownbook. Strangely enough, SharpChess seems to have no trouble with time control in my tourney (only 1 game lost on time out of 9).

Thanx a lot, you probably didn't have any problems before because the problems was related to how certain engines use the level command. I tested with Arena 0.92, 0.93 and WB 4.2.6. Now you have a third example of how engines use the protocol.
I'll study this log for a while and think about a solution that works for all these GUI's (hope there is no fourth special way to do it). I'll get back to you when I have anything worth trying.
/Regards Albert

Thanks Albert The Fritz GUI is a very popular one indeed
Olivier


Chesswar