Atak 6.5 queen sac

[Graham Banks]

[d] 4rb1k/1b2rpp1/2p2n1p/q1PpNB2/p2P1P1Q/Pp4R1/1P1N1P1P/3R3K W
Atak 6.5 v Clueless 1.4 (White to move plays Qxf6)

Not bad for a 2200 elo engine.

[Graham Banks]

[Event "Round Robin"]
[Site "ChessGUI"]
[Date "2009.04.21"]
[Round "1"]
[White "Atak 6.5"]
[Black "Clueless 1.4"]
[Result "1-0"]
[PlyCount "75"]
[EventDate "2009.??.??"]
[TimeControl "40/2280:40/2280:40/2280"]

{Unknown Hardware} 1. d4 {[%eval 0,1] [%emt 0:00:00]} Nf6 {
[%eval 0,1] [%emt 0:00:00]} 2. c4 {[%eval 0,1] [%emt 0:00:00]} e6 {
[%eval 0,1] [%emt 0:00:00]} 3. g3 {[%eval 0,1] [%emt 0:00:00]} Bb4+ {
[%eval 0,1] [%emt 0:00:00]} 4. Bd2 {[%eval 0,1] [%emt 0:00:00]} Be7 {
[%eval 0,1] [%emt 0:00:00]} 5. Bg2 {[%eval 0,1] [%emt 0:00:00]} d5 {
[%eval 0,1] [%emt 0:00:00]} 6. Nf3 {[%eval 0,1] [%emt 0:00:00]} O-O {
[%eval 0,1] [%emt 0:00:00]} 7. O-O {[%eval 0,1] [%emt 0:00:00]} c6 {
[%eval 0,1] [%emt 0:00:00]} 8. Qb3 {[%eval 0,1] [%emt 0:00:00]} b6 {
[%eval 0,1] [%emt 0:00:00]} 9. Bf4 {[%eval 0,1] [%emt 0:00:00]} Ba6 {
[%eval 0,1] [%emt 0:00:00]} 10. Nbd2 {[%eval 0,1] [%emt 0:00:00]} Nbd7 {
[%eval 0,1] [%emt 0:00:00]} 11. Rfd1 {[%eval 0,1] [%emt 0:00:00]} Nh5 {
[%eval 0,1] [%emt 0:00:00]} 12. e4 {[%eval 0,1] [%emt 0:00:00]} Nxf4 {
[%eval 0,1] [%emt 0:00:00]} 13. gxf4 {[%eval -17,13] [%emt 0:00:56]} Nf6 {
[%eval 100,11] [%emt 0:01:21]} 14. Ne5 {[%eval 19,12] [%emt 0:01:29]} Bb7 {
[%eval 100,10] [%emt 0:01:21]} 15. exd5 {[%eval 16,12] [%emt 0:01:11]} exd5 {
[%eval 122,11] [%emt 0:01:21]} 16. Rac1 {[%eval 13,11] [%emt 0:01:22]} Rc8 {
[%eval 118,10] [%emt 0:01:21]} 17. Qa4 {[%eval 20,12] [%emt 0:00:55]} a6 {
[%eval 142,11] [%emt 0:01:21]} 18. Bh3 {[%eval 10,12] [%emt 0:00:57]} Rc7 {
[%eval 136,12] [%emt 0:01:21]} 19. c5 {[%eval 11,13] [%emt 0:00:48]} b5 {
[%eval 133,12] [%emt 0:01:21]} 20. Qa3 {[%eval 24,13] [%emt 0:00:42]} Re8 {
[%eval 131,11] [%emt 0:01:21]} 21. Qf3 {[%eval 35,12] [%emt 0:01:11]} Bf8 {
[%eval 146,12] [%emt 0:01:21]} 22. Kh1 {[%eval 17,11] [%emt 0:01:08]} Rce7 {
[%eval 174,12] [%emt 0:01:21]} 23. Qe3 {[%eval 32,11] [%emt 0:01:06]} b4 {
[%eval 176,10] [%emt 0:01:21]} 24. Qb3 {[%eval 41,11] [%emt 0:01:03]} a5 {
[%eval 159,12] [%emt 0:01:21]} 25. Qf3 {[%eval 8,11] [%emt 0:00:36]} Qb8 {
[%eval 177,11] [%emt 0:01:21]} 26. Re1 {[%eval 32,12] [%emt 0:00:44]} a4 {
[%eval 168,10] [%emt 0:01:21]} 27. Re3 {[%eval 28,11] [%emt 0:00:57]} Ba8 {
[%eval 140,10] [%emt 0:01:21]} 28. Qg3 {[%eval 56,11] [%emt 0:00:31]} b3 {
[%eval 132,10] [%emt 0:01:22]} 29. a3 {[%eval 50,11] [%emt 0:00:54]} Qd8 {
[%eval 148,10] [%emt 0:01:21]} 30. Rd1 {[%eval 48,11] [%emt 0:00:51]} Bb7 {
[%eval 144,11] [%emt 0:01:25]} 31. Qh4 {[%eval 36,12] [%emt 0:00:42]} Qa5 {
[%eval 120,10] [%emt 0:01:20]} 32. Bf5 {[%eval 94,12] [%emt 0:00:36]} h6 {
[%eval 103,11] [%emt 0:01:20]} 33. Rg3 {[%eval 103,12] [%emt 0:00:33]} Kh8 {
[%eval -28,11] [%emt 0:02:09]} 34. Qxf6 {[%eval 212,12] [%emt 0:00:45]} Rxe5 {
[%eval -691,11] [%emt 0:01:13]} 35. dxe5 {[%eval 963,14] [%emt 0:00:30]} Qd8 {
[%eval -691,11] [%emt 0:01:58]} 36. Qxf7 {[%eval 1055,16] [%emt 0:00:24]} Bc8 {
[%eval -768,11] [%emt 0:01:05]} 37. e6 {[%eval 1060,16] [%emt 0:00:26]} Bxe6 {
[%eval -1065,10] [%emt 0:01:05]} 38. Bxe6 {[%eval 1125,16] [%emt 0:00:21]} 1-0

[MattieShoes]

Hah, finally a problem my engine can solve too
Found Qxf6 on ply 11, 34.48 seconds.

[Graham Banks]

Hah, finally a problem my engine can solve too
Found Qxf6 on ply 11, 34.48 seconds.

Which is your engine?

[MattieShoes]

An unreleased one I've been working on. It plays on FICS as Moneypenny. In the 2000-2150 area

[Graham Banks]

An unreleased one I've been working on. It plays on FICS as Moneypenny. In the 2000-2150 area

Very good.
This one is a fairly straightforward queen sac. Try the following two:
http://www.talkchess.com/forum/viewtopi ... +queen+sac
http://www.talkchess.com/forum/viewtopi ... +queen+sac

[Roman Hartmann]

Seems it's not that hard to find as even my engine finds the move after a few seconds (at least the version I'm working on). In a 3+0 game it would miss the move though ...

Code: Select all

 D.   Time    Score  Best line
  1   0.00   -0.84   Nd2-f3 
  2   0.01   -0.44   f2-f3 qa5-c7 
  3   0.03   -0.80   Nd2-f1 nf6-e4 Bf5xe4 
  4   0.15   -0.41   Kh1-g2 qa5-c7 Kg2-f3 qc7-d8 
  5   0.63   -0.78   Nd2-f1 qa5-a6 Bf5-d3 qa6-a7 Nf1-e3 
  6   3.98   -0.42   Kh1-g2 qa5-c7 Kg2-f3 re8-d8 h2-h3 
  7   7.82    0.08   Qh4xf6 re7xe5 Qf6xf7 re5-e7 Qf7-g6 re7-e1+ Rg3-g1 
                     re1xg1+ Kh1xg1 
  8  21.77    1.05   Qh4xf6 re7xe5 Qf6xf7 re5-e1+ Kh1-g2 re1-e2 Qf7xb7 
                     re2xd2 Qb7xc6 
  9  56.51    1.91   Qh4xf6 re7xe5 Qf6xf7 re5xf5 Qf7xf5 qa5-d8 Qf5-g4 
                     qd8-e7 

best regards
Roman

[bob]

Seems it's not that hard to find as even my engine finds the move after a few seconds (at least the version I'm working on). In a 3+0 game it would miss the move though ...

Code: Select all

 D.   Time    Score  Best line
  1   0.00   -0.84   Nd2-f3 
  2   0.01   -0.44   f2-f3 qa5-c7 
  3   0.03   -0.80   Nd2-f1 nf6-e4 Bf5xe4 
  4   0.15   -0.41   Kh1-g2 qa5-c7 Kg2-f3 qc7-d8 
  5   0.63   -0.78   Nd2-f1 qa5-a6 Bf5-d3 qa6-a7 Nf1-e3 
  6   3.98   -0.42   Kh1-g2 qa5-c7 Kg2-f3 re8-d8 h2-h3 
  7   7.82    0.08   Qh4xf6 re7xe5 Qf6xf7 re5-e7 Qf7-g6 re7-e1+ Rg3-g1 
                     re1xg1+ Kh1xg1 
  8  21.77    1.05   Qh4xf6 re7xe5 Qf6xf7 re5-e1+ Kh1-g2 re1-e2 Qf7xb7 
                     re2xd2 Qb7xc6 

9 56.51 1.91 Qh4xf6 re7xe5 Qf6xf7 re5xf5 Qf7xf5 qa5-d8 Qf5-g4
qd8-e7
[/code]

best regards
Roman

I agree on this one. On my slow office box, one cpu:

Code: Select all

                6     0.14   1.45   1. Qxf6 Rxe5 2. Qxf7 R5e7 3. Qg6 Re1+
                                    4. Kg2
                6->   0.14   1.45   1. Qxf6 Rxe5 2. Qxf7 R5e7 3. Qg6 Re1+
                                    4. Kg2

If you have another half-second to spare (above the 0.14 it took here):

Code: Select all

               11     0.54     +1   1. Qxf6!                         
               11     0.62     +3   1. Qxf6!                         
               11     0.80     +M   1. Qxf6!                         
               11     1.23   9.41   1. Qxf6 Rxe5 2. fxe5 Qd8 3. Qxf7 Bc8
                                    4. Bd3 Re6 5. Re1 Qe8 6. Qf5 <HT>
               11->   1.27   9.41   1. Qxf6 Rxe5 2. fxe5 Qd8 3. Qxf7 Bc8
                                    4. Bd3 Re6 5. Re1 Qe8 6. Qf5 <HT>
               12     1.56     +1   1. Qxf6!                         
               12     2.11  10.51   1. Qxf6 Rxe5 2. dxe5 Qd8 3. Qxf7 Bc8
                                    4. e6 Bxe6 5. Bxe6 Rxe6 6. Qxe6 Bxc5
                                    7. Qxc6 Bxf2
               12->   2.18  10.51   1. Qxf6 Rxe5 2. dxe5 Qd8 3. Qxf7 Bc8
                                    4. e6 Bxe6 5. Bxe6 Rxe6 6. Qxe6 Bxc5
                                    7. Qxc6 Bxf2

[MattieShoes]

I found I had disabled check extensions, futility pruning, and extended futility pruning while testing something! Naturally I notice right AFTER running several hundred test games ( and unsurprisingly, finding that disabling those things makes the engine play worse )

Now it finds Qxf6 in 15 seconds on ply 8

For the next, it found Qxg7+ on ply 11, in about 45 seconds

The third... well, I'm sure it'd EVENTUALLY find it but the branching factor is still too high to find it in a reasonable amount of time I'm afraid If it takes Hiarcs 70 seconds it'd probably take mine 70 hours!

[Graham Banks]

Seems it's not that hard to find as even my engine finds the move after a few seconds (at least the version I'm working on). In a 3+0 game it would miss the move though ...

Code: Select all

 D.   Time    Score  Best line
  1   0.00   -0.84   Nd2-f3 
  2   0.01   -0.44   f2-f3 qa5-c7 
  3   0.03   -0.80   Nd2-f1 nf6-e4 Bf5xe4 
  4   0.15   -0.41   Kh1-g2 qa5-c7 Kg2-f3 qc7-d8 
  5   0.63   -0.78   Nd2-f1 qa5-a6 Bf5-d3 qa6-a7 Nf1-e3 
  6   3.98   -0.42   Kh1-g2 qa5-c7 Kg2-f3 re8-d8 h2-h3 
  7   7.82    0.08   Qh4xf6 re7xe5 Qf6xf7 re5-e7 Qf7-g6 re7-e1+ Rg3-g1 
                     re1xg1+ Kh1xg1 
  8  21.77    1.05   Qh4xf6 re7xe5 Qf6xf7 re5-e1+ Kh1-g2 re1-e2 Qf7xb7 
                     re2xd2 Qb7xc6 
  9  56.51    1.91   Qh4xf6 re7xe5 Qf6xf7 re5xf5 Qf7xf5 qa5-d8 Qf5-g4 
                     qd8-e7 

best regards
Roman

Hi Roman (and Bob),

did you try the other two slightly more difficult ones also?

Cheers, Graham.