Problem with OliThink 5.1.3 and Xpdnt 080209

[opraus]

Graham,

A couple of questions:

When I feed these games into Scotti's xboard, the scores look really wierd. I first thought, 'wow, there is something wrong with eval!' but then noticed that even the winner seemed confused [his scores were inverted at end].

Should these work in Arena? [they did not when I tried]

In the comments, eval 1.10 ,12 would mean a score of 1.10 at depth 12?
and emt is time spent thinking?

Xpdnt seems to return many {eval: 1,1} and {emt: 0:0:0} - this is the first place I'll be looking, but I need clarification.

If my assumptions are correct above, it would seem to indicate Xpdnt is just 'spitting out' moves without thinking at certain points???

BTW, how did you filter out those games? I may have some such games also.

Thanks,

David

[Graham Banks]

Graham,

A couple of questions:

When I feed these games into Scotti's xboard, the scores look really wierd. I first thought, 'wow, there is something wrong with eval!' but then noticed that even the winner seemed confused [his scores were inverted at end].

Should these work in Arena? [they did not when I tried]

Not sure if Chessbase GUI pgns can be read using Arena. You might need to use a utility to clean up the pgn into a readable format.

In the comments, eval 1.10 ,12 would mean a score of 1.10 at depth 12?
and emt is time spent thinking?

Yes - emt is the time spent thinking.

Xpdnt seems to return many {eval: 1,1} and {emt: 0:0:0} - this is the first place I'll be looking, but I need clarification.

If my assumptions are correct above, it would seem to indicate Xpdnt is just 'spitting out' moves without thinking at certain points???

BTW, how did you filter out those games? I may have some such games also.

I used Chessbase 9. I quickly checked all the Xpdnt games, taking a note of the ones that I felt showed problems. I then selected those games and created a pgn file which I copied and pasted to here.

Thanks,

David

[OliverBr]

Thanks for the games. Let's start with game 1:

13. h3 is actually a big blunder, but not reproducable it all. If I test it here (in my lab;) then OliThink always finds the normal move a4.
Have such things happened with OliThink 5.1.2 and earlier, too?

Could you please post a position/game where OliThink 513 gave away a piece for no apparent reason?

Thank you.

I can email the pgn file of these games to you if you wish.

Regards, Graham.

[Event "JUNGLE NIGHTMARE-1"]
[Site "Home"]
[Date "2008.04.24"]
[Round "1.6"]
[White "Olithink 5.1.3"]
[Black "OBender 3.2.4"]
[Result "0-1"]
[ECO "C11"]
[Annotator "0.50;0.02"]
[PlyCount "72"]
[EventDate "2008.04.20"]
[EventType "tourn"]
[Source "Smith"]
[TimeControl "40/2700:40/2700:40/2700"]

{AMD Athlon(tm) 64 X2 Dual Core Processor 4200+ 2205 MHz W=13.5 ply; 1,192kN/
s; HS-8moves.ctg B=13.4 ply; 683kN/s; HS-8moves.ctg} 1. e4 {
[%eval 35534,0] [%emt 0:00:00]} e6 {[%eval 35534,0] [%emt 0:00:00]} 2. d4 {
[%eval 35534,0] [%emt 0:00:00]} d5 {[%eval 35534,0] [%emt 0:00:00]} 3. Nc3 {
[%eval 35534,0] [%emt 0:00:00]} Nf6 {[%eval 35534,0] [%emt 0:00:00]} 4. e5 {
[%eval 35534,0] [%emt 0:00:00]} Nfd7 {[%eval 35534,0] [%emt 0:00:00]} 5. f4 {
[%eval 35534,0] [%emt 0:00:00]} c5 {[%eval 35534,0] [%emt 0:00:00]} 6. Nf3 {
[%eval 35534,0] [%emt 0:00:00]} Nc6 {[%eval 35534,0] [%emt 0:00:00]} 7. Be3 {
[%eval 35534,0] [%emt 0:00:00]} cxd4 {[%eval 35534,0] [%emt 0:00:00]} 8. Nxd4 {
[%eval 35534,0] [%emt 0:00:00]} Bc5 {[%eval 35534,0] [%emt 0:00:00]} 9. g3 {
[%eval 50,12] [%emt 0:01:22]} Nxd4 {(Bxd4) [%eval 2,12] [%emt 0:00:22]} 10.
Bxd4 {[%eval 61,12] [%emt 0:00:54]} Qb6 {[%eval -4,13] [%emt 0:00:35]} 11. Bxc5
{[%eval 46,12] [%emt 0:01:23]} Nxc5 {(Qxc5) [%eval -1,13] [%emt 0:00:30]} 12.
Bb5+ {[%eval 43,12] [%emt 0:01:23]} Bd7 {(Nd7) [%eval -6,13] [%emt 0:00:20]}
13. h3 {(Qe2) [%eval -355,13] [%emt 0:08:00]} Bxb5 {
[%eval -320,14] [%emt 0:00:24]} 14. Nxb5 {[%eval -356,13] [%emt 0:00:47]} Qxb5
{[%eval -320,13] [%emt 0:00:22]} 15. Qd4 {[%eval -383,14] [%emt 0:01:09]} Ne4 {
[%eval -314,12] [%emt 0:00:23]} 16. Rg1 {[%eval -374,13] [%emt 0:00:43]} Qa5+ {
(Rc8) [%eval -319,13] [%emt 0:01:07]} 17. c3 {[%eval -370,13] [%emt 0:00:38]}
Rc8 {[%eval -317,12] [%emt 0:00:24]} 18. Rg2 {[%eval -386,12] [%emt 0:00:37]}
O-O {(Rc4) [%eval -321,12] [%emt 0:00:41]} 19. h4 {
(Qd3) [%eval -380,12] [%emt 0:01:38]} Rc4 {[%eval -322,12] [%emt 0:00:32]} 20.
Qd3 {[%eval -386,13] [%emt 0:00:58]} Rfc8 {(b5) [%eval -316,11] [%emt 0:00:25]}
21. Kf1 {[%eval -385,12] [%emt 0:00:50]} b5 {[%eval -318,11] [%emt 0:00:27]}
22. a3 {[%eval -390,12] [%emt 0:01:09]} b4 {
(Qb6) [%eval -322,12] [%emt 0:00:35]} 23. cxb4 {[%eval -400,14] [%emt 0:01:14]}
Rxb4 {[%eval -322,13] [%emt 0:00:30]} 24. Rd1 {[%eval -394,12] [%emt 0:01:14]}
Rb5 {(Rbc4) [%eval -319,12] [%emt 0:00:20]} 25. Qe3 {
[%eval -411,12] [%emt 0:02:14]} Qa4 {[%eval -422,13] [%emt 0:00:34]} 26. Rc1 {
(Rd3) [%eval -457,13] [%emt 0:01:10]} Re8 {
(Rxc1+) [%eval -420,13] [%emt 0:00:28]} 27. b4 {[%eval -459,15] [%emt 0:01:10]}
a5 {[%eval -417,14] [%emt 0:00:26]} 28. Kg1 {
(Rb2) [%eval -475,14] [%emt 0:01:10]} axb4 {[%eval -421,15] [%emt 0:00:39]} 29.
axb4 {[%eval -467,13] [%emt 0:01:10]} Rxb4 {[%eval -421,14] [%emt 0:00:37]} 30.
Kh1 {(Kh2) [%eval -476,13] [%emt 0:00:50]} Rb3 {
(Qb3) [%eval -426,13] [%emt 0:00:40]} 31. Qe1 {[%eval -591,14] [%emt 0:00:37]}
Ra8 {[%eval -523,12] [%emt 0:00:40]} 32. Kg1 {
(Rc7) [%eval -624,13] [%emt 0:01:58]} Qa3 {
(Qd4+) [%eval -528,12] [%emt 0:00:25]} 33. Rcc2 {
(Rb1) [%eval -638,12] [%emt 0:00:41]} Rxg3 {
(Nxg3) [%eval -621,14] [%emt 0:00:22]} 34. Rxg3 {[%eval -708,14] [%emt 0:00:47]
} Nxg3 {[%eval -621,14] [%emt 0:00:40]} 35. Kh2 {[%eval -712,13] [%emt 0:00:48]
} Nh5 {[%eval -628,14] [%emt 0:00:31]} 36. Qf2 {
(f5) [%eval -742,13] [%emt 0:01:24]} Qb4 {(Qd3) [%eval -724,13] [%emt 0:00:26]}
0-1

[Graham Banks]

Have such things happened with OliThink 5.1.2 and earlier, too?

Hi Oliver,

I only played a small number of games with OliThink 5.1.1 and 5.1.2. I'll see if I can locate them and take a look.
I can't recall seeing the same problem, but I'll get back to you.

[Olivier Deville]

Could you please post a position/game where OliThink 513 gave away a piece for no apparent reason?

Thank you.

I'll hunt one down for you. I normally discard such games, but I think I can find at least one I kept for this purpose.

Regards, Graham.

Hi Graham

Do you mean such games are replayed ?

Olivier

[Graham Banks]

Could you please post a position/game where OliThink 513 gave away a piece for no apparent reason?

Thank you.

I'll hunt one down for you. I normally discard such games, but I think I can find at least one I kept for this purpose.

Regards, Graham.

Hi Graham

Do you mean such games are replayed ?

Olivier

Usually, but not always.
Depends whether or not the result was clear cut at the point of the mishap.
Sometimes I set up the position and play it out - as Oliver stated, often the blunder occurs in a game situation, but doesn't reproduce from a position setup.

Regards, Graham.

[Olivier Deville]

Could you please post a position/game where OliThink 513 gave away a piece for no apparent reason?

Thank you.

I'll hunt one down for you. I normally discard such games, but I think I can find at least one I kept for this purpose.

Regards, Graham.

Hi Graham

Do you mean such games are replayed ?

Olivier

Usually, but not always.
Depends whether or not the result was clear cut at the point of the mishap.
Sometimes I set up the position and play it out - as Oliver stated, often the blunder occurs in a game situation, but doesn't reproduce from a position setup.

Regards, Graham.

Hi Graham

You are too kind to engine authors !

When such things happen here, I give the engine a loss and report the issue, and believe me, I usually get a bugfix in my mailbox very quickly

In real chess, if you let your Queen hanging, you are not allowed to replay the game.

Olivier

[Graham Banks]

Hi Graham

You are too kind to engine authors !

When such things happen here, I give the engine a loss and report the issue, and believe me, I usually get a bugfix in my mailbox very quickly

In real chess, if you let your Queen hanging, you are not allowed to replay the game.

Olivier

You're correct on all points there Olivier. Can't argue with you!
I'm a bit remiss also in that I don't usually keep logfiles.

Graham.