New version of Thinker

[Kerwin]

I have joined Lance in the development of the engine.

5.3B is out.

There is a new variety (Inert Thinker) which is another variation of the evaluation code. Compared to Passive Thinker, this one seems to perform better against other engines.

Passive Thinker should improve slightly from the fix in the check move generation. The improvement may not be that noticeable.

Active Thinker should have a noticeable improvement. The piece-square evaluation has been corrected.

Lastly, the SMP versions are significantly better. There was a nasty bug in the code that propagates the results from the helper board to the parent board. This has bee corrected.

Enjoy...

[Graham Banks]

Does the engine display the pv and evaluation while it's thinking yet?

[Roger Brown]

Does the engine display the pv and evaluation while it's thinking yet?

Hello Graham,

Please do not ride that horse over here.

A quick search over that other forum reveals that this subject has been flogged to death and I do not want an author to feel in any way obligated to do things any way but his.

Besides, would we really be having this mindset if this was a Rybka v 6 release?

Y'know, that mindset about the engine being worthless because it does not show this or that?

Please!

Later.

[Graham Banks]

Does the engine display the pv and evaluation while it's thinking yet?

Hello Graham,

Please do not ride that horse over here.

A quick search over that other forum reveals that this subject has been flogged to death and I do not want an author to feel in any way obligated to do things any way but his.

Besides, would we really be having this mindset if this was a Rybka v 6 release?

Y'know, that mindset about the engine being worthless because it does not show this or that?

Please!

Later.

I actually get on quite well with Lance, but I'll respect your order not to discuss the matter further.
I've also run several hundred games at CCRL 40/40 time control with various versions of Thinker, so I hardly think that Thinker is worthless.
I know that you dislike the way that I hounded the programmers who were claiming that Rybka 1.0 was probably a Fruit clone (and we're still waiting eagerly for the proof), but that hardly means that I'm a Rybka fanatic or that I don't appreciate all of the free and amateur engines.

Cheers, Graham.

[Kerwin]

Does the engine display the pv and evaluation while it's thinking yet?

I'm sorry Graham, but that feature is still not present. This is one feature that does not match the current design goals. I hope that does not prevent you from enjoying playing against Thinker.

Of the commonly requested features, this is what they look like:
1. running PV - distant future
2. EGTB/EGBB - very distant future
3. UCI - never

[Graham Banks]

Does the engine display the pv and evaluation while it's thinking yet?

I'm sorry Graham, but that feature is still not present. This is one feature that does not match the current design goals. I hope that does not prevent you from enjoying playing against Thinker.

Of the commonly requested features, this is what they look like:
1. running PV - distant future
2. EGTB/EGBB - very distant future
3. UCI - never

Okay. Thanks for this information.

Regards, Graham.

[Norm Pollock]

I tried the new version. In this game it unexpectedly agreed to a draw from a strong position, possibly a winning position. Thinker was not in time trouble. I use the Fritz9 gui with the Wb2Uci adapter for Thinker. The version used was "inert" for 32-bit single processor.

Code: Select all

[Event "C-55"]
[Site "?"]
[Date "2008.11.03"]
[Round "1.1"]
[White "Thinker 5.3B"]
[Black "GarboChess v2.20 x86"]
[Result "1/2-1/2"]
[ECO "A29"]
[Annotator "Pollock,Norman"]
[PlyCount "100"]
[EventDate "2008.11.03"]
[EventType "tourn"]
[Source "Pollock"]
[TimeControl "60+1"]

1. c4 {[%eval 0,0] [%emt 0:00:00]} e5 {[%eval 0,0] [%emt 0:00:00]} 2. Nc3 {
[%eval 0,0] [%emt 0:00:00]} Nf6 {[%eval 0,0] [%emt 0:00:00]} 3. Nf3 {
[%eval 0,0] [%emt 0:00:00]} Nc6 {[%eval 0,0] [%emt 0:00:00]} 4. g3 {
[%eval 0,0] [%emt 0:00:00]} Bb4 {[%eval 0,0] [%emt 0:00:00]} 5. Bg2 {
[%eval 0,0] [%emt 0:00:00]} O-O {[%eval 0,0] [%emt 0:00:00]} 6. O-O {
[%eval 0,0] [%emt 0:00:00]} e4 {[%eval 0,0] [%emt 0:00:00]} 7. Ng5 {
[%eval 0,0] [%emt 0:00:00]} Bxc3 {[%eval 0,0] [%emt 0:00:00]} 8. bxc3 {
[%eval 0,0] [%emt 0:00:00]} Re8 {[%eval 0,0] [%emt 0:00:00]} 9. f3 {
[%eval 0,0] [%emt 0:00:00]} e3 {[%eval 0,0] [%emt 0:00:00]} 10. d3 {
[%eval 0,0] [%emt 0:00:00]} d5 {[%eval 0,0] [%emt 0:00:00]} 11. cxd5 {
[%eval 0,0] [%emt 0:00:00]} Nxd5 {[%eval 0,0] [%emt 0:00:00]} 12. Ne4 {
[%eval 0,0] [%emt 0:00:00]} f5 {[%eval 0,0] [%emt 0:00:00]} 13. Qb3 {
[%eval 0,0] [%emt 0:00:00]} Be6 {[%eval 0,0] [%emt 0:00:00]} 14. Qxb7 {
[%eval 0,0] [%emt 0:00:00]} Na5 {[%eval 0,0] [%emt 0:00:00]} 15. Qb2 {
[%eval 0,0] [%emt 0:00:00]} fxe4 {[%eval 0,0] [%emt 0:00:00]} 16. fxe4 {
[%eval 1,1] [%emt 0:00:04]} Nf6 {[%eval 78,12] [%emt 0:00:05]} 17. Qa3 {
(Bxe3) [%eval 1,1] [%emt 0:00:03]} Nc6 {[%eval 79,11] [%emt 0:00:04]} 18. d4 {
(Bxe3) [%eval 1,1] [%emt 0:00:03]} Bg4 {[%eval 89,10] [%emt 0:00:04]} 19. Qa6 {
(Qa4) [%eval 1,1] [%emt 0:00:02]} Qd7 {[%eval 81,10] [%emt 0:00:04]} 20. Qc4+ {
[%eval 1,1] [%emt 0:00:03]} Qe6 {[%eval 69,11] [%emt 0:00:04]} 21. Qxe6+ {
[%eval 1,1] [%emt 0:00:03]} Bxe6 {[%eval 99,13] [%emt 0:00:04]} 22. d5 {
(e5) [%eval 1,1] [%emt 0:00:02]} Bd7 {[%eval 60,13] [%emt 0:00:04]} 23. dxc6 {
[%eval 1,1] [%emt 0:00:02]} Bxc6 {[%eval 54,14] [%emt 0:00:04]} 24. Rb1 {
(c4) [%eval 1,1] [%emt 0:00:02]} Bxe4 {[%eval 32,11] [%emt 0:00:03]} 25. Rb3 {
[%eval 1,1] [%emt 0:00:02]} Bxg2 {[%eval 49,10] [%emt 0:00:03]} 26. Kxg2 {
[%eval 1,1] [%emt 0:00:01]} Re4 {[%eval 47,10] [%emt 0:00:03]} 27. Rf4 {
[%eval 1,1] [%emt 0:00:04]} Rxf4 {[%eval 51,12] [%emt 0:00:03]} 28. gxf4 {
[%eval 1,1] [%emt 0:00:02]} Nd5 {[%eval 109,12] [%emt 0:00:03]} 29. Kf3 {
[%eval 1,1] [%emt 0:00:02]} Rf8 {[%eval 93,12] [%emt 0:00:03]} 30. Bxe3 {
[%eval 1,1] [%emt 0:00:03]} g5 {[%eval 105,12] [%emt 0:00:03]} 31. Ra3 {
(Rb5) [%eval 1,1] [%emt 0:00:03]} Nxf4 {[%eval 102,11] [%emt 0:00:03]} 32. Bxa7
{(Bxf4) [%eval 1,1] [%emt 0:00:03]} h6 {[%eval 78,8] [%emt 0:00:03]} 33. Bc5 {
(Rb3) [%eval 1,1] [%emt 0:00:02]} Rf5 {[%eval 98,10] [%emt 0:00:03]} 34. Ra8+ {
[%eval 1,1] [%emt 0:00:03]} Kh7 {[%eval 105,11] [%emt 0:00:02]} 35. Bd4 {
[%eval 1,1] [%emt 0:00:02]} Rf7 {[%eval 120,10] [%emt 0:00:02]} 36. Ra7 {
(a4) [%eval 1,1] [%emt 0:00:02]} Ne6+ {[%eval 132,10] [%emt 0:00:02]} 37. Ke4 {
[%eval 1,1] [%emt 0:00:02]} Kg6 {[%eval 154,11] [%emt 0:00:02]} 38. Ra6 {
[%eval 1,1] [%emt 0:00:02]} Re7 {[%eval 158,12] [%emt 0:00:02]} 39. Kd5 {
(a4) [%eval 1,1] [%emt 0:00:02]} Kf7 {[%eval 177,13] [%emt 0:00:02]} 40. e3 {
(a4) [%eval 1,1] [%emt 0:00:02]} Nxd4 {[%eval 165,11] [%emt 0:00:02]} 41. exd4
{[%eval 1,1] [%emt 0:00:02]} Re2 {[%eval 156,13] [%emt 0:00:02]} 42. Rxh6 {
[%eval 1,1] [%emt 0:00:02]} Rxa2 {[%eval 161,12] [%emt 0:00:02]} 43. Ke5 {
(h3) [%eval 1,1] [%emt 0:00:02]} Ke7 {[%eval 158,11] [%emt 0:00:02]} 44. Rh7+ {
[%eval 1,1] [%emt 0:00:02]} Kd8 {[%eval 158,12] [%emt 0:00:02]} 45. Kf5 {
(Kf6) [%eval 1,1] [%emt 0:00:03]} Rg2 {[%eval 149,12] [%emt 0:00:02]} 46. h3 {
[%eval 1,1] [%emt 0:00:01]} Rg3 {[%eval 153,12] [%emt 0:00:02]} 47. c4 {
[%eval 1,1] [%emt 0:00:01]} Rd3 {[%eval 165,13] [%emt 0:00:02]} 48. Ke5 {
(d5) [%eval 1,1] [%emt 0:00:01]} Rc3 {[%eval 160,11] [%emt 0:00:02]} 49. Kd5 {
(Rh5) [%eval 1,1] [%emt 0:00:02]} Re3 {[%eval 169,12] [%emt 0:00:02]} 50. c5 {
[%eval 1,1] [%emt 0:00:02]} Kc8 {[%eval 173,13] [%emt 0:00:02]} 1/2-1/2




[Kerwin]

I tried the new version. In this game it unexpectedly agreed to a draw from a strong position, possibly a winning position. Thinker was not in time trouble. I use the Fritz9 gui with the Wb2Uci adapter for Thinker. The version used was "inert" for 32-bit single processor.

Code: Select all

[Event "C-55"]
[Site "?"]
[Date "2008.11.03"]
[Round "1.1"]
[White "Thinker 5.3B"]
[Black "GarboChess v2.20 x86"]
[Result "1/2-1/2"]
[ECO "A29"]
[Annotator "Pollock,Norman"]
[PlyCount "100"]
[EventDate "2008.11.03"]
[EventType "tourn"]
[Source "Pollock"]
[TimeControl "60+1"]

1. c4 {[%eval 0,0] [%emt 0:00:00]} e5 {[%eval 0,0] [%emt 0:00:00]} 2. Nc3 {
[%eval 0,0] [%emt 0:00:00]} Nf6 {[%eval 0,0] [%emt 0:00:00]} 3. Nf3 {
[%eval 0,0] [%emt 0:00:00]} Nc6 {[%eval 0,0] [%emt 0:00:00]} 4. g3 {
[%eval 0,0] [%emt 0:00:00]} Bb4 {[%eval 0,0] [%emt 0:00:00]} 5. Bg2 {
[%eval 0,0] [%emt 0:00:00]} O-O {[%eval 0,0] [%emt 0:00:00]} 6. O-O {
[%eval 0,0] [%emt 0:00:00]} e4 {[%eval 0,0] [%emt 0:00:00]} 7. Ng5 {
[%eval 0,0] [%emt 0:00:00]} Bxc3 {[%eval 0,0] [%emt 0:00:00]} 8. bxc3 {
[%eval 0,0] [%emt 0:00:00]} Re8 {[%eval 0,0] [%emt 0:00:00]} 9. f3 {
[%eval 0,0] [%emt 0:00:00]} e3 {[%eval 0,0] [%emt 0:00:00]} 10. d3 {
[%eval 0,0] [%emt 0:00:00]} d5 {[%eval 0,0] [%emt 0:00:00]} 11. cxd5 {
[%eval 0,0] [%emt 0:00:00]} Nxd5 {[%eval 0,0] [%emt 0:00:00]} 12. Ne4 {
[%eval 0,0] [%emt 0:00:00]} f5 {[%eval 0,0] [%emt 0:00:00]} 13. Qb3 {
[%eval 0,0] [%emt 0:00:00]} Be6 {[%eval 0,0] [%emt 0:00:00]} 14. Qxb7 {
[%eval 0,0] [%emt 0:00:00]} Na5 {[%eval 0,0] [%emt 0:00:00]} 15. Qb2 {
[%eval 0,0] [%emt 0:00:00]} fxe4 {[%eval 0,0] [%emt 0:00:00]} 16. fxe4 {
[%eval 1,1] [%emt 0:00:04]} Nf6 {[%eval 78,12] [%emt 0:00:05]} 17. Qa3 {
(Bxe3) [%eval 1,1] [%emt 0:00:03]} Nc6 {[%eval 79,11] [%emt 0:00:04]} 18. d4 {
(Bxe3) [%eval 1,1] [%emt 0:00:03]} Bg4 {[%eval 89,10] [%emt 0:00:04]} 19. Qa6 {
(Qa4) [%eval 1,1] [%emt 0:00:02]} Qd7 {[%eval 81,10] [%emt 0:00:04]} 20. Qc4+ {
[%eval 1,1] [%emt 0:00:03]} Qe6 {[%eval 69,11] [%emt 0:00:04]} 21. Qxe6+ {
[%eval 1,1] [%emt 0:00:03]} Bxe6 {[%eval 99,13] [%emt 0:00:04]} 22. d5 {
(e5) [%eval 1,1] [%emt 0:00:02]} Bd7 {[%eval 60,13] [%emt 0:00:04]} 23. dxc6 {
[%eval 1,1] [%emt 0:00:02]} Bxc6 {[%eval 54,14] [%emt 0:00:04]} 24. Rb1 {
(c4) [%eval 1,1] [%emt 0:00:02]} Bxe4 {[%eval 32,11] [%emt 0:00:03]} 25. Rb3 {
[%eval 1,1] [%emt 0:00:02]} Bxg2 {[%eval 49,10] [%emt 0:00:03]} 26. Kxg2 {
[%eval 1,1] [%emt 0:00:01]} Re4 {[%eval 47,10] [%emt 0:00:03]} 27. Rf4 {
[%eval 1,1] [%emt 0:00:04]} Rxf4 {[%eval 51,12] [%emt 0:00:03]} 28. gxf4 {
[%eval 1,1] [%emt 0:00:02]} Nd5 {[%eval 109,12] [%emt 0:00:03]} 29. Kf3 {
[%eval 1,1] [%emt 0:00:02]} Rf8 {[%eval 93,12] [%emt 0:00:03]} 30. Bxe3 {
[%eval 1,1] [%emt 0:00:03]} g5 {[%eval 105,12] [%emt 0:00:03]} 31. Ra3 {
(Rb5) [%eval 1,1] [%emt 0:00:03]} Nxf4 {[%eval 102,11] [%emt 0:00:03]} 32. Bxa7
{(Bxf4) [%eval 1,1] [%emt 0:00:03]} h6 {[%eval 78,8] [%emt 0:00:03]} 33. Bc5 {
(Rb3) [%eval 1,1] [%emt 0:00:02]} Rf5 {[%eval 98,10] [%emt 0:00:03]} 34. Ra8+ {
[%eval 1,1] [%emt 0:00:03]} Kh7 {[%eval 105,11] [%emt 0:00:02]} 35. Bd4 {
[%eval 1,1] [%emt 0:00:02]} Rf7 {[%eval 120,10] [%emt 0:00:02]} 36. Ra7 {
(a4) [%eval 1,1] [%emt 0:00:02]} Ne6+ {[%eval 132,10] [%emt 0:00:02]} 37. Ke4 {
[%eval 1,1] [%emt 0:00:02]} Kg6 {[%eval 154,11] [%emt 0:00:02]} 38. Ra6 {
[%eval 1,1] [%emt 0:00:02]} Re7 {[%eval 158,12] [%emt 0:00:02]} 39. Kd5 {
(a4) [%eval 1,1] [%emt 0:00:02]} Kf7 {[%eval 177,13] [%emt 0:00:02]} 40. e3 {
(a4) [%eval 1,1] [%emt 0:00:02]} Nxd4 {[%eval 165,11] [%emt 0:00:02]} 41. exd4
{[%eval 1,1] [%emt 0:00:02]} Re2 {[%eval 156,13] [%emt 0:00:02]} 42. Rxh6 {
[%eval 1,1] [%emt 0:00:02]} Rxa2 {[%eval 161,12] [%emt 0:00:02]} 43. Ke5 {
(h3) [%eval 1,1] [%emt 0:00:02]} Ke7 {[%eval 158,11] [%emt 0:00:02]} 44. Rh7+ {
[%eval 1,1] [%emt 0:00:02]} Kd8 {[%eval 158,12] [%emt 0:00:02]} 45. Kf5 {
(Kf6) [%eval 1,1] [%emt 0:00:03]} Rg2 {[%eval 149,12] [%emt 0:00:02]} 46. h3 {
[%eval 1,1] [%emt 0:00:01]} Rg3 {[%eval 153,12] [%emt 0:00:02]} 47. c4 {
[%eval 1,1] [%emt 0:00:01]} Rd3 {[%eval 165,13] [%emt 0:00:02]} 48. Ke5 {
(d5) [%eval 1,1] [%emt 0:00:01]} Rc3 {[%eval 160,11] [%emt 0:00:02]} 49. Kd5 {
(Rh5) [%eval 1,1] [%emt 0:00:02]} Re3 {[%eval 169,12] [%emt 0:00:02]} 50. c5 {
[%eval 1,1] [%emt 0:00:02]} Kc8 {[%eval 173,13] [%emt 0:00:02]} 1/2-1/2


Thanks for the test Norm.

The Thinker engine actually does not respond to the draw command. It does not recognize that command. There is no code for it. Thus, it would not be possible for it to "accept" a draw request.

If you send me the log for the adapter, I will investigate it.

[Norm Pollock]

Kerwin,

I don't have any logs to send unfortunately but I can add a few observations. And I hope you still find a way to investigate this game.

GarboChess's evaluation had it losing by 1.73 pawns at a depth of 13 plies when the draw was agreed. Thinker does not give an evaluation.

At the end, Thinker had 31 seconds remaining, and GarboChess had 15 seconds, in a 1'+1" game. The draw came after black's (Garbo) 50th move.

-Norm

[Kerwin]

Kerwin,

I don't have any logs to send unfortunately but I can add a few observations. And I hope you still find a way to investigate this game.

GarboChess's evaluation had it losing by 1.73 pawns at a depth of 13 plies when the draw was agreed. Thinker does not give an evaluation.

At the end, Thinker had 31 seconds remaining, and GarboChess had 15 seconds, in a 1'+1" game. The draw came after black's (Garbo) 50th move.

-Norm

I loaded that game and had the engine play, and it completed the game. I'm guessing that the draw came from the adapter. Too bad we don't have the logs to troubleshoot this.

[George Bodkin]

Where can I get the new Thinker.

Thanks

George

[Kerwin]

Where can I get the new Thinker.

Thanks

George

geocities.com/thechessthinker