Test Your Best

Discussion of anything and everything relating to chess playing software and machines.

Moderator: Ras

Terry McCracken
Posts: 16465
Joined: Wed Aug 01, 2007 4:16 am
Location: Canada

Test Your Best

Post by Terry McCracken »

[d]r4rk1/pp3ppp/4p3/q1bbP1B1/5Q1P/2N5/PPP3P1/2KR3R w - - 0 17

17. Bf6!!

[Event "Sochi"]
[Site "Sochi"]
[Date "1973.??.??"]
[Round "?"]
[White "Boris Spassky"]
[Black "Valeri S Korensky"]
[Result "1-0"]
[ECO "B65"]
[PlyCount "55"]
[EventDate "1973.??.??"]

1. e4 c5 2. Nf3 Nc6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 d6 6. Bg5 e6 7. Qd2 Be7 8. O-O-O Nxd4 9. Qxd4 O-O 10. f4 Qa5 11. Bc4 Bd7 12. e5 dxe5 13. fxe5 Bc6 14. h4 Bc5 15. Qf4 Nd5 16. Bxd5 Bxd5 17. Bf6 Rfc8 18. Bxg7 Kxg7 19. Qf6+ Kf8 20. Rhf1 Rc7 21. Nxd5 exd5 22. e6 Qxa2 23. e7+ Kg8 24. Qxf7+ Kh8 25. e8=Q+ Rxe8 26. Qxe8+ Kg7 27. Qe5+ Kg8 28. Qg5+ 1-0
User avatar
AdminX
Posts: 6396
Joined: Mon Mar 13, 2006 2:34 pm
Location: Acworth, GA

Re: Test Your Best

Post by AdminX »

Hi Terry,

I remember going over this game back in the day. :lol: Anyway ....

Hardware: Q6700 /1024 Hash
Analysis by 64-bit Deep Rybka 3:

1.Nxd5 exd5[] 2.Kb1[] Qb6 3.Rxd5 h6 4.Qe4 Rfe8 5.Bf4 Rad8 6.Rhd1 Rxd5 7.Rxd5 Qc6 8.g4 b6 9.a3 a5
+/= (0.30) Depth: 6 00:00:00 5kN
+/- (0.78) Depth: 15 00:00:41 7513kN

1.h5 Bxg2 2.Rh4 Rad8 3.Rxd8 Rxd8 4.Qg3 Bd5 5.Rg4 g6 6.hxg6 fxg6 7.Bxd8 Qxd8 8.Qd3 Bb6 9.Ne4 Bxe4 10.Qxe4 Qd5 11.Qxd5 exd5
+/- (1.09 !) Depth: 15 00:01:49 17747kN
+/- (1.04) Depth: 16 00:03:46 36393kN

1.Bf6 gxf6 2.Rh3[] Bf2 3.Qxf2 f5 4.Nxd5 exd5 5.Qxf5 Qb6 6.Rg3+ Kh8 7.Rxd5 Qh6+ 8.Rg5 f6 9.exf6 Qxf6 10.Qxf6+ Rxf6 11.Rd7 h6 12.Rgg7 b6 13.b3 a5 14.Rh7+ Kg8 15.Rdg7+ Kf8 16.Rc7
+/- (1.25 !) Depth: 16 00:04:53 46213kN
+- (2.61) Depth: 16 00:11:38 111mN


Hardware: Q6700 /1024 Hash
Analysis by Deep Sjeng WC2008 x64:

1.Qg3 Rfc8 2.Kb1 Bf8 3.Nxd5 exd5 4.Rhf1 Rc7 5.c3 Rac8 6.Qf3 Bc5 7.Rxd5 a6 8.Be3
+/= (0.30) Depth: 6/14 00:00:00 39kN
+/= (0.63) Depth: 12/31 00:00:01 3621kN

1.Rh3 Rfc8 2.Kb1 Bf8 3.Nxd5 exd5 4.Qf5 g6 5.Qd7 Rc7 6.Qxd5 Qxd5 7.Rxd5 Rac8 8.c3 Be7 9.Rhd3 Bxg5 10.hxg5 Rc5 11.R3d4 Rxd5 12.Rxd5 Kg7
+/- (0.76) Depth: 12/31 00:00:03 8439kN
+/- (0.95) Depth: 17/60 00:02:09 424mN

1.Bf6 h6 2.Nxd5 exd5 3.Qg3 g6 4.Qf4 Kh7 5.h5 g5 6.Qf5+ Kg8 7.e6 Qxa2 8.exf7+ Rxf7 9.Qg6+ Kf8 10.Qxh6+ Ke8 11.Rhe1+ Kd7 12.Qxg5 Bd6 13.Qf5+ Kc7 14.c3 Rg8 15.g4 Rgf8 16.g5
+/- (1.34 !) Depth: 17/60 00:04:12 833mN
+- (1.92) Depth: 17/61 00:06:07 1210mN
"Good decisions come from experience, and experience comes from bad decisions."
__________________________________________________________________
Ted Summers
Terry McCracken
Posts: 16465
Joined: Wed Aug 01, 2007 4:16 am
Location: Canada

Re: Test Your Best

Post by Terry McCracken »

I was hoping it would be a stumper...Thanks Ted.

Terry

P.S. We'll see if others post, what programs and hardware.

P.P.S Did you see my post on the new Cray XT Jaguar that will sell next year? A whopping 1.64 Petaflops! It just ate the Roadrunner! :lol:

http://www.nccs.gov/jaguar/

Wylie Coyote
User avatar
Zach Wegner
Posts: 1922
Joined: Thu Mar 09, 2006 12:51 am
Location: Earth

Re: Test Your Best

Post by Zach Wegner »

Here's ZCT, Q6600:

Code: Select all

 15/34         3:00     +1.07   17. Rh3 Rfc8 18. Rg3 Qb4 19. Nxd5 exd5 20.
                                  Qxb4 Bxb4 21. Rxd5 Rc4 22. Bf6 g6 23. Rd7
                                  Be1 24. b3 Bxg3 25. bxc4 (327370745)
[15/36]        3:44     +1.07   17. Rh3 Rfc8 18. Rg3 Qb4 19. Nxd5 exd5 20.
                                  Qxb4 Bxb4 21. Rxd5 Rc4 22. Bf6 g6 23. Rd7
                                  Be1 24. b3 Bxg3 25. bxc4 (405199418)
 16/36         5:07     +1.11   17. Rh3 Rfc8 18. Rg3 Qb4 19. Nxd5 exd5 20.
                                  Qxb4 Bxb4 21. Rxd5 Rab8 22. Bh6 g6 23. Rd7
                                  Be1 24. Rg4 Ba5 (550558707)
[16/36]        6:02     +1.11   17. Rh3 Rfc8 18. Rg3 Qb4 19. Nxd5 exd5 20.
                                  Qxb4 Bxb4 21. Rxd5 Rab8 22. Bh6 g6 23. Rd7
                                  Be1 24. Rg4 Ba5 (647583301)
 17/36         7:46     +1.07   17. Rh3 Rfc8 18. Rg3 Qb4 19. Nxd5 exd5 20.
                                  Qxb4 Bxb4 21. Rxd5 Rab8 22. Bh6 g6 23. Rd7
                                  Rc4 24. h5 Rh4 25. Be3 Rh1+ 26. Rd1
                                  (831348642)
 17/42        11:26     +1.60   17. Bf6 gxf6 18. Rh3 Rfd8 19. Rg3+ Kf8 20.
                                  Qxf6 Bxg2 21. Rg7 Rxd1+ 22. Nxd1 Qc7 23.
                                  Rxg2 Rc8 24. Rd2 Bb4 25. Rf2 Qd7 26. c3
                                  (1234861722)
[17/42]       11:46     +1.60   17. Bf6 gxf6 18. Rh3 Rfd8 19. Rg3+ Kf8 20.
                                  Qxf6 Bxg2 21. Rg7 Rxd1+ 22. Nxd1 Qc7 23.
                                  Rxg2 Rc8 24. Rd2 Bb4 25. Rf2 Qd7 26. c3
                                  (1270888110)
 18/42        14:47     +1.54   17. Bf6 gxf6 18. Rh3 Rfd8 19. Rg3+ Kf8 20.
                                  Qxf6 Bxg2 21. Rg7 Rxd1+ 22. Nxd1 Qc7 23.
                                  Rxg2 Rc8 24. Rd2 Bb4 25. c3 Bxc3 26. Nxc3
                                  Qc5 (1620025998)
[18/42]       16:58     +1.54   17. Bf6 gxf6 18. Rh3 Rfd8 19. Rg3+ Kf8 20.
                                  Qxf6 Bxg2 21. Rg7 Rxd1+ 22. Nxd1 Qc7 23.
                                  Rxg2 Rc8 24. Rd2 Bb4 25. c3 Bxc3 26. Nxc3
                                  Qc5 (1860070740)
User avatar
Ovyron
Posts: 4562
Joined: Tue Jul 03, 2007 4:30 am

Re: Test Your Best

Post by Ovyron »

Engine: bright-0.2c (64 MB/Athlon 1.8Ghz)
by AJ Siemelink
[snip]
9/27 0:00 +0.29 1.Qg3 Rfe8 2.Nxd5 exd5 3.Kb1 Bf8 4.Bd2 Qb5 5.Rhf1 Rac8 (681.416) 0
10/31 0:02 +0.23 1.Qg3 Rfc8 2.Nxd5 exd5 3.Qb3 h6 4.Bf4 d4 5.e6 Rf8 6.exf7+ Rxf7 (1.734.053) 0
11/36 0:04 +0.23 1.Qg3 Rfc8 2.Nxd5 exd5 3.Qb3 h6 4.Bf4 Rd8 5.Kb1 Bb6 6.Be3 Bxe3 7.Qxe3 (3.383.237) 0
12/45 0:17 +0.30 1.Qg3 Rfe8 2.Nxd5 exd5 3.Kb1 Bf8 4.Qb3 Rxe5 5.Qxb7 Rae8 6.Bf4 R8e7 7.Qc8 (12.782.060) 0
13/45 0:39 +0.26 1.Qg3 Rfe8 2.Nxd5 exd5 3.Kb1 Bf8 4.Qb3 Rxe5 5.Qxb7 Rae8 6.Bf4 R8e7 7.Qb3 Re1 8.Rhxe1 Rxe1 (28.989.847) 0
13/47 1:13 +0.35 1.Rh3 Rfc8 2.Nxd5 exd5 3.Kb1 Rc7 4.Rxd5 Qe1+ 5.Qc1 Qe4 6.Qd2 Re8 7.Bf4 Bb4 8.Rd8 (52.104.068) 0
13/47 1:29 +0.42 1.Nxd5 exd5 2.Kb1 d4 3.Rh3 Qb6 4.Rb3 Qc6 5.Rg3 Kh8 6.Qf5 Rfe8 7.Qxf7 Rxe5 (61.377.775) 0
13/55 2:04 +0.44 1.Bf6 Rfc8 2.Nxd5 exd5 3.Qg4 Bf8 4.e6 fxe6 5.Qxe6+ Kh8 6.Rxd5 Qa4 7.Bc3 Re8 8.Qd7 Qxa2 9.Ra5 (84.799.188) 0

(Considered reasonable time because this iteration started at second 39.)
User avatar
michiguel
Posts: 6401
Joined: Thu Mar 09, 2006 8:30 pm
Location: Chicago, Illinois, USA

Re: Test Your Best

Post by michiguel »

Terry McCracken wrote:[d]r4rk1/pp3ppp/4p3/q1bbP1B1/5Q1P/2N5/PPP3P1/2KR3R w - - 0 17

17. Bf6!!

[Event "Sochi"]
[Site "Sochi"]
[Date "1973.??.??"]
[Round "?"]
[White "Boris Spassky"]
[Black "Valeri S Korensky"]
[Result "1-0"]
[ECO "B65"]
[PlyCount "55"]
[EventDate "1973.??.??"]

1. e4 c5 2. Nf3 Nc6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 d6 6. Bg5 e6 7. Qd2 Be7 8. O-O-O Nxd4 9. Qxd4 O-O 10. f4 Qa5 11. Bc4 Bd7 12. e5 dxe5 13. fxe5 Bc6 14. h4 Bc5 15. Qf4 Nd5 16. Bxd5 Bxd5 17. Bf6 Rfc8 18. Bxg7 Kxg7 19. Qf6+ Kf8 20. Rhf1 Rc7 21. Nxd5 exd5 22. e6 Qxa2 23. e7+ Kg8 24. Qxf7+ Kh8 25. e8=Q+ Rxe8 26. Qxe8+ Kg7 27. Qe5+ Kg8 28. Qg5+ 1-0
For some reason, Gaviota 0.44 AMD 2.4 Ghz finds it relatively quick (48 seconds, ply 9):

Miguel

Code: Select all

        61   1:      0.0    +0.32  Nc3xd5   e6xd5    Rd1xd5   
       438   2:      0.0    +0.32  Nc3xd5   e6xd5    Rd1xd5   
                          ~ -0.07  will fail low? (  -19)
      2764   3       0.0      :-(  
                          ~ -0.95  will fail low? ( -244)
                          ~ -0.05  ok             (  -12)
      6083   3:      0.0    -0.21  Nc3xd5   e6xd5    Kc1-b1   h7-h6    
     27570   4:      0.1    -0.07  Qf4-g3   Rf8-c8   Bg5-f6   g7-g6    
                                   Nc3xd5   e6xd5    
     69227   5:      0.2    -0.07  Qf4-g3   Rf8-c8   Bg5-f6   g7-g6    
                                   Nc3xd5   e6xd5    
    207229   6:      0.5    +0.04  Qf4-g3   Kg8-h8   Nc3xd5   e6xd5    
                                   Kc1-b1   h7-h6    Bg5-e3   Bc5xe3   
                                   Qg3xe3   
    746458   7:      1.8    +0.07  Qf4-g3   Kg8-h8   Nc3xd5   e6xd5    
                                   Qg3-b3   d5-d4    Kc1-b1   Bc5-b6   
   1247490   8       2.9    +0.12  Qf4-g3   Kg8-h8   Nc3xd5   e6xd5    
                                   Qg3-b3   d5-d4    Qb3xb7   Qa5xa2   
                                   b2-b3    
   2384785   8:      5.5    +0.12  Qf4-g3   Kg8-h8   Nc3xd5   e6xd5    
                                   Qg3-b3   d5-d4    Qb3xb7   Qa5xa2   
                                   b2-b3    
   3545713   9       8.1    +0.11  Qf4-g3   Kg8-h8   Nc3xd5   e6xd5    
                                   Qg3-b3   d5-d4    Rd1-f1   Kh8-g8   
                                   Kc1-b1   Bc5-b4   
  21512687   9      48.6    +0.11  Bg5-f6   Rf8-c8   Nc3xd5   e6xd5    
                                   Qf4-g3   Bc5-f8   Kc1-b1   Kg8-h8   
                                   Bf6-g5   Bf8-b4   Bg5-e3  
  22247721   9:     50.2    +0.11  Bg5-f6   Rf8-c8   Nc3xd5   e6xd5    
                                   Qf4-g3   Bc5-f8   Kc1-b1   Kg8-h8   
                                   Bf6-g5   Bf8-b4   Bg5-e3   
  30856447  10      69.3      :-)  Bg5-f6   
  51036375  10     113.1    +0.67  Bg5-f6   Rf8-e8   Bf6xg7   Kg8xg7   
                                   Qf4-f6   Kg7-g8   Rh1-h3   Bc5-f2   
                                   Qf6xf2   Bd5xa2   Rh3-g3   Kg8-f8   
                                   Nc3xa2   Qa5xa2   
  55680916  10:    122.9    +0.67  Bg5-f6   Rf8-e8   Bf6xg7   Kg8xg7   
                                   Qf4-f6   Kg7-g8   Rh1-h3   Bc5-f2   
                                   Qf6xf2   Bd5xa2   Rh3-g3   Kg8-f8   
                                   Nc3xa2   Qa5xa2   
 118071554  11     259.4    +0.80  Bg5-f6   Rf8-c8   Bf6xg7   Kg8xg7   
                                   Qf4-f6   Kg7-g8   Rh1-h3   h7-h5    
                                   <-transp 
 130163568  11:    285.4    +0.80  Bg5-f6   Rf8-c8   Bf6xg7   Kg8xg7   
                                   Qf4-f6   Kg7-g8   Rh1-h3   h7-h5    
                                   <-transp 
 206304031  12     453.6    +0.67  Bg5-f6   Rf8-c8   Bf6xg7   Kg8xg7   
                                   Qf4-f6   Kg7-g8   Rh1-h3   Qa5-d8   
                                   Nc3xd5   e6xd5    Rh3-g3   Kg8-f8   
                                   Qf6-h6   Kf8-e7   Rd1-f1   Qd8-c7   
                                   Rf1-f5   
 336262692  12:    731.5    +0.67  Bg5-f6   Rf8-c8   Bf6xg7   Kg8xg7   
                                   Qf4-f6   Kg7-g8   Rh1-h3   Qa5-d8   
                                   Nc3xd5   e6xd5    Rh3-g3   Kg8-f8   
                                   Qf6-h6   Kf8-e7   Rd1-f1   Qd8-c7   
                                   Rf1-f5   
 608547266  13    1320.5    +0.42  Bg5-f6   Rf8-c8   Bf6xg7   Kg8xg7   
                                   Qf4-f6   Kg7-g8   Rh1-h3   Qa5-d8   
                                   Rh3-g3   Kg8-f8   Qf6-h6   Kf8-e7   
                                   Rg3-g7   Qd8-f8   Nc3xd5   e6xd5    
                                   Qh6xh7   Bc5-e3   Kc1-b1   Rc8-c5   
2781337143  13:   1870.1    +0.42  Bg5-f6   Rf8-c8   Bf6xg7   Kg8xg7   
                                   Qf4-f6   Kg7-g8   Rh1-h3   Qa5-d8   
                                   Rh3-g3   Kg8-f8   Qf6-h6   Kf8-e7   
                                   Rg3-g7   Qd8-f8   Nc3xd5   e6xd5    
                                   Qh6xh7   Bc5-e3   Kc1-b1   Rc8-c5   
 422287971  14    1780.6      :-)  Bg5-f6
Terry McCracken
Posts: 16465
Joined: Wed Aug 01, 2007 4:16 am
Location: Canada

Re: Test Your Best

Post by Terry McCracken »

48.6 sec., that's a hard time to beat!

As a human player, 17. Bf6 was the first move I considered as I'm familar with the pattern and have played it in a few of my games.

I felt this position would be easier for humans than machines and I think I'm right about that, still that's impressive!

Can you make an educated guess why your program picked up on the quiet move well under a min.?

Terry
User avatar
michiguel
Posts: 6401
Joined: Thu Mar 09, 2006 8:30 pm
Location: Chicago, Illinois, USA

Re: Test Your Best

Post by michiguel »

Terry McCracken wrote:48.6 sec., that's a hard time to beat!

As a human player, 17. Bf6 was the first move I considered as I'm familar with the pattern and have played it in a few of my games.

I felt this position would be easier for humans than machines and I think I'm right about that, still that's impressive!

Can you make an educated guess why your program picked up on the quiet move well under a min.?

Terry
I am not sure without spending a bit of time digging it, but I think that
1) Gaviota is not optimistic with the position without Bf6 (that allows to find it with a not so high score)
2) a key is to trust in the Bxg7 sacrifice, and Gaviota may like the black king in the center in some variations
3) I have the feeling that null move may have something to do with this. The reason I say this is because Gaviota sees it several plies before other programs. Gaviota does not null-move "prune" variations, it "reduces" them. It may have a similar effect as "null move verification". In certain positions in which there is an inevitable mate, null move prune those variations. Generally, Gaviota does well in those situations. It certainly does NOT make Gaviota stronger, but it is not blind to certain situations.
I may be wrong with this. If my instinct is right, I wonder how other programs will do in this position with null-move off.

Also, I agree about this move is easy for humans. I think I would play it even in blitz.

Miguel
Tony Thomas

Re: Test Your Best

Post by Tony Thomas »

I am happy to hear that Gaviota is still being developed. It used to be one of my base engines for testing new engines (mainly due to stability).. now almost everything seems to have surpassed it. I am sure Leo would be happy if you send it to him.
User avatar
Ovyron
Posts: 4562
Joined: Tue Jul 03, 2007 4:30 am

Re: Test Your Best

Post by Ovyron »

michiguel wrote:If my instinct is right, I wonder how other programs will do in this position with null-move off.
Zappa Mexico II with Null-move off:

Engine: Zappa Mexico II (64 MB-Athlon 1.87Ghz)
by Anthony Cozzie
[snip]
8/25 0:02 +0.24 1.Qg4 Kh8 2.Nxd5 exd5 3.Kb1 h6 4.Bf4 Rad8 (368.076) 157
8/26 0:04 +0.70 1.Qg3 Rfe8 2.Rhf1 Bc4 3.Rf4 b5 4.Rd7 Rf8 (669.539) 156
8/26 0:05 +0.70 1.Qg3 Rfe8 2.Rhf1 Bc4 3.Rf4 b5 4.Rd7 (895.165) 159
9/27 0:08 +0.59 1.Qg3 Rfc8 2.Nxd5 exd5 3.Qb3 d4 4.Rhg1 h6 (1.408.156) 158
9/30 0:13 +0.59 1.Qg3 Rfc8 2.Nxd5 exd5 3.Qb3 d4 4.Rhg1 (2.143.849) 160
10/30 0:25 +0.61 1.Qg3 Rfe8 2.Rhf1 Bc6 3.h5 Bb4 4.Bf6 Bf8 (4.016.319) 156
10/39 0:41 +0.61 1.Qg3 Rfe8 2.Rhf1 Bc6 3.h5 Bb4 4.Bf6 (6.479.338) 157
11/39 1:13 +0.46 1.Qg3 Kh8 2.Nxd5 exd5 3.Kb1 d4 4.Rhf1 Rae8 5.h5 (11.137.243) 150
11/39 2:51 +0.52 1.Rh3 Rfc8 2.Rg3 Bf8 3.Nxd5 exd5 4.Rf1 Rc7 5.Kb1 (24.663.346) 143
11/39 3:34 +0.57 1.Qg4 h5 2.Qg3 Rfc8 3.Nxd5 exd5 4.Kb1 Qb6 5.Rxd5 (29.982.744) 140
(Not found in reasonable time.)


With Null-move ON:

Engine: Zappa Mexico II (64 MB-Athlon 1.87Ghz)
by Anthony Cozzie
[snip]
8/28 0:01 +0.29 1.Qg4 h5 2.Qxh5 Bxg2 3.Rh2 Bc6 4.Rhd2 Bb4 (305.644) 156
8/28 0:02 +0.48 1.Qg3 Rfe8 2.Nxd5 exd5 3.Kb1 Bf8 4.Bd2 Qa4 (450.702) 156
8/28 0:03 +0.48 1.Qg3 Rfe8 2.Nxd5 exd5 3.Kb1 Bf8 4.Bd2 (500.404) 157
9/28 0:04 +0.50 1.Qg3 Rfe8 2.Kb1 Bb4 3.Nxd5 exd5 4.c3 Bf8 (692.263) 161
9/33 0:05 +0.50 1.Qg3 Rfe8 2.Kb1 Bb4 3.Nxd5 exd5 4.c3 (813.645) 161
10/33 0:07 +0.51 1.Qg3 Rfe8 2.Kb1 Bc6 3.h5 Rac8 4.h6 g6 (1.189.694) 163
10/33 0:12 +0.51 1.Qg3 Rfe8 2.Kb1 Bc6 3.h5 Rac8 4.h6 (1.987.151) 157
11/33 0:20 +0.49 1.Qg3 Kh8 2.Nxd5 exd5 3.Kb1 d4 4.Rhf1 Rae8 5.h5 (3.313.410) 158
11/39 1:04 +0.52 1.Rh3 Rfc8 2.Rg3 Bf8 3.Nxd5 exd5 4.Rf1 Rc7 5.Kb1 (9.635.452) 149
11/39 1:05 +0.52 1.Rh3 Rfc8 2.Rg3 Bf8 3.Nxd5 exd5 4.Rf1 Rc7 5.Kb1 (9.837.673) 149
12/39 1:27 +0.63 1.Rh3 Rfc8 2.Nxd5 exd5 3.Kb1 Qb6 4.Rf1 Qe6 5.Rg3 a5 (12.811.668) 146

12/40 2:08 +0.90 1.h5 h6 2.Bxh6 gxh6 3.Qxh6 Qd8 4.Nxd5 exd5 5.Rh3 Bf2 (18.519.920) 144
(Not found in reasonable time.)