Winboard Forum

In the documentary Game Over: Kasparov and the Machine Kasparov was accusing the IBM Deep Blue team of cheating. He was especially suspicious about game 2 and, as I understand, the move 37. Be4. I thought that you might be interested to know that in the following position of the second game of 1997 match Kasparov vs Deep Blue

[diag]r1r1q1k1/6p1/3b1p1p/1p1PpP2/1Pp5/2P4P/R1B2QP1/R5K1[/diag]
r1r1q1k1/6p1/3b1p1p/1p1PpP2/1Pp5/2P4P/R1B2QP1/R5K1 w - - 0 1

after a day of thinking Crafty 18.13.rj16 (a modified Crafty chess engine, my current Crafty version is 20.1.rj2) running on 1 CPU 2.6 GHz Windows XP machine has played 37. Be4, the same move Deep Blue played.

Code: Select all

depth=24 1/37 +0.37 37. Be4 Rcb8 38. Kh2 Rxa2 39. Qxa2 Qd8 40. Qa6 Kh7 41. g3 Rb6 42. Qa7 Kg8 43. Qa8 Qxa8 44. Rxa8+ Kf7 45. Ra7+ Kf8 46. Kg1 Bb8 47. Rd7 Bd6 48. Kf2 Ra6 49. Ke3 h5; Time: 24:17:06.52

I guess that Deep Blue could have come to the depth of 24 plies much faster than Crafty running on the mentioned hardware (possibly within a time limit of one move in tournament time controls) and it looks after all that computers can play this kind of move.

I would be interested to know what would other chess engines do in this position. If you intend to run the test and post back the results could you please specify: chess engine, machine speed and OS, time of thinking, ply reached and move made. Thanks.

Robert Jurjevic wrote:In the documentary Game Over: Kasparov and the Machine Kasparov was accusing the IBM Deep Blue team of cheating. He was especially suspicious about game 2 and, as I understand, the move 37. Be4. I thought that you might be interested to know that in the following position of the second game of 1997 match Kasparov vs Deep Blue

[diag]r1r1q1k1/6p1/3b1p1p/1p1PpP2/1Pp5/2P4P/R1B2QP1/R5K1[/diag]
r1r1q1k1/6p1/3b1p1p/1p1PpP2/1Pp5/2P4P/R1B2QP1/R5K1 w - - 0 1

after a day of thinking Crafty 18.13.rj16 (a modified Crafty chess engine, my current Crafty version is 20.1.rj2) running on 1 CPU 2.6 GHz Windows XP machine has played 37. Be4, the same move Deep Blue played.

Code: Select all

depth=24 1/37 +0.37 37. Be4 Rcb8 38. Kh2 Rxa2 39. Qxa2 Qd8 40. Qa6 Kh7 41. g3 Rb6 42. Qa7 Kg8 43. Qa8 Qxa8 44. Rxa8+ Kf7 45. Ra7+ Kf8 46. Kg1 Bb8 47. Rd7 Bd6 48. Kf2 Ra6 49. Ke3 h5; Time: 24:17:06.52

I guess that Deep Blue could have come to the depth of 24 plies much faster than Crafty running on the mentioned hardware (possibly within a time limit of one move in tournament time controls) and it looks after all that computers can play this kind of move.

I would be interested to know what would other chess engines do in this position. If you intend to run the test and post back the results could you please specify: chess engine, machine speed and OS, time of thinking, ply reached and move made. Thanks.

It is known that deep blue did not use null move pruning and used a lot of extensions that reduced it's nominal depth.

There are logfiles that can prove that it only could get depth 12 at tournament time control and it had a branching factor of more than 3 so there is no reason to believe that it could get depth 24 in less than
3*3^12 minutes and it means that it could not get depth 24 even in one year.

Note that I do not claim that deep blue depth is equivalent to Crafty's depth but I think that it is pointless to try to compare between a dead thing(deep blue) and another program that you use today.

Uri

Hi Robert,

not much of a problem for Shredder 9.1 on P4 with 2.8GHz

Code: Select all

 17/45    01:48    36.793.080    408.104  +0,41    1.Ta2a6 e5e4
 17/45+   01:57    40.007.401    403.605  +0,42    1.Lc2e4

 [...]

 22/56   16:52    417.751.126   420.325   +0,61   1.Lc2e4 Tc8b8 2.Ta2a6 De8d8 3.Kg1f1 Ta8xa6



Cheers,

Martin

On a P4 3.0Ghz (64bit)


Crafty-Cito 1.4.3:
7 00:00 132.888 1.661.100 +0.39 1. Ra5 Rxa5 2. Rxa5 Bc7 3. Ra7 Bb8 4. Ra6 e4
7 00:00 177.313 1.970.144 +0.52 1. Qb6 Rxa2 2. Rxa2 Rd8 3. Ra5 Bxb4 4. cxb4 Rxd5
7 00:00 190.483 1.731.663 +0.52 1. Qb6 Rxa2 2. Rxa2 Rd8 3. Ra5 Bxb4 4. cxb4 Rxd5
8 00:00 268.632 1.678.950 +0.79 1. Qb6 Rxa2 2. Rxa2 Qd7 3. Ra7 Rc7 4. Rxc7 Bxc7 5. Qe6+ Qf7 6. Be4
8 00:00 289.233 1.701.370 +0.79 1. Qb6 Rxa2 2. Rxa2 Qd7 3. Ra7 Rc7 4. Rxc7 Bxc7 5. Qe6+ Qf7 6. Be4
9 00:00 428.918 1.715.672 +0.82 1. Qb6 Rxa2 2. Rxa2 Qd7 3. Ra6 Rd8 4. Ra7 Qe8 5. Ra5 <HT>
9 00:00 455.061 1.685.411 +0.82 1. Qb6 Rxa2 2. Rxa2 Qd7 3. Ra6 Rd8 4. Ra7 Qe8 5. Ra5 <HT>
10 00:01 919.302 1.671.458 +1.22 1. Qb6!!
10 00:01 1.252.688 1.670.250 +1.24 1. Qb6 Rxa2 2. Rxa2 Bc7 3. Qe6+ Kf8 4. Ra6 Rd8 5. Rc6 Bb8 6. Be4 Ba7+ 7. Kh2
10 00:01 1.308.526 1.677.597 +1.24 1. Qb6 Rxa2 2. Rxa2 Bc7 3. Qe6+ Kf8 4. Ra6 Rd8 5. Rc6 Bb8 6. Be4 Ba7+ 7. Kh2
11 00:01 2.043.674 1.675.142 +1.13 1. Qb6 Rxa2 2. Rxa2 Bc7 3. Qe6+ Kf8 4. Ra6 Rd8 5. Be4 Qe7 6. Qc6 Qd7 7. Qe6
11 00:02 2.251.535 1.655.540 +1.13 1. Qb6 Rxa2 2. Rxa2 Bc7 3. Qe6+ Kf8 4. Ra6 Rd8 5. Be4 Qe7 6. Qc6 Qd7 7. Qe6
12 00:03 3.766.023 1.651.764 +1.17 1. Qb6 Rxa2 2. Rxa2 Bc7 3. Qe6+ Kf8 4. Ra6 Rd8 5. Rc6 Bb8 6. Be4 Ba7+ 7. Kh2 Bd4 8. cxd4 exd4
12 00:03 4.018.035 1.660.345 +1.17 1. Qb6 Rxa2 2. Rxa2 Bc7 3. Qe6+ Kf8 4. Ra6 Rd8 5. Rc6 Bb8 6. Be4 Ba7+ 7. Kh2 Bd4 8. cxd4 exd4
13 00:05 8.350.160 1.422.514 +1.12 1. Qb6 Rxa2 2. Rxa2 Bc7 3. Qe6+ Kf8 4. Ra6 Rd8 5. Rc6 Bb8 6. Be4 Ba7+ 7. Kh2 Be3 8. d6 Qd7
13 00:07 11.208.544 1.470.937 +1.12 1. Qb6 Rxa2 2. Rxa2 Bc7 3. Qe6+ Kf8 4. Ra6 Rd8 5. Rc6 Bb8 6. Be4 Ba7+ 7. Kh2 Be3 8. d6 Qd7
14 00:11 18.514.500 1.559.772 +1.15 1. Qb6 Rxa2 2. Rxa2 Ra8 3. Rxa8 Qxa8 4. Qxd6 Qa1+ 5. Kh2 Qc1 6. Bd1 Qxd1 7. Qe6+ Kh7 8. d6 e4 9. d7 e3 10. Kg3
14 00:15 23.523.611 1.568.240 +1.15 1. Qb6 Rxa2 2. Rxa2 Ra8 3. Rxa8 Qxa8 4. Qxd6 Qa1+ 5. Kh2 Qc1 6. Bd1 Qxd1 7. Qe6+ Kh7 8. d6 e4 9. d7 e3 10. Kg3
15 00:20 32.512.405 1.625.620 +0.87 1. Qb6 Rxa2 2. Rxa2 Ra8 3. Rxa8 Qxa8 4. Qxd6 Qa1+ 5. Kh2 Qc1 6. Bd1 Qxd1 7. Qb8+ Kh7 8. Qxb5 Qb3 9. d6 Qxc3 10. d7 Qd4
15 00:43 68.263.320 1.587.519 +0.87 1. Qb6 Rxa2 2. Rxa2 Ra8 3. Rxa8 Qxa8 4. Qxd6 Qa1+ 5. Kh2 Qc1 6. Bd1 Qxd1 7. Qb8+ Kh7 8. Qxb5 Qb3 9. d6 Qxc3 10. d7 Qd4
16 01:12 115.505.697 1.604.245 +1.13 1. Qb6 Rxa2 2. Rxa2 Ra8 3. Rxa8 Qxa8 4. Qxd6 Qa1+ 5. Kh2 Qc1 6. Bd1 Qxd1 7. Qb8+ Kh7 8. Qxb5 Qd3 9. d6 Qxd6 10. Qxc4 g6
16 01:16 122.699.863 1.614.471 +1.13 1. Qb6 Rxa2 2. Rxa2 Ra8 3. Rxa8 Qxa8 4. Qxd6 Qa1+ 5. Kh2 Qc1 6. Bd1 Qxd1 7. Qb8+ Kh7 8. Qxb5 Qd3 9. d6 Qxd6 10. Qxc4 g6
17 01:47 174.239.702 1.628.408 +0.60 1. Qb6 Rxa2 2. Rxa2 Ra8 3. Rxa8 Qxa8 4. Qxd6 Qa1+ 5. Kh2 Qc1 6. Bd1 Qxd1 7. Qb8+ Kh7 8. Qxb5 Qd3 9. d6 Qxd6 10. Qxc4 e4+ 11. g3 e3
17 08:16 791.885.234 1.596.542 +0.76 1. Be4 Rcb8 2. Kh2 Rxa2 3. Qxa2 Bc7 4. Qf2 Bd6 5. Ra6 Qd8 6. Qa7 <HT>
17 09:26 901.692.213 1.593.095 +0.76 1. Be4 Rcb8 2. Kh2 Rxa2 3. Qxa2 Bc7 4. Qf2 Bd6 5. Ra6 Qd8 6. Qa7 <HT>
18 11:35 1.111.143.577 1.598.767 +0.76 1. Be4 Rcb8 2. Kh2 Rxa2 3. Qxa2 Bc7 4. Qf2 Qd8 5. Qc5 Kh7 6. Kh1 Bd6 7. Qa7 Qc8 8. Qf7 Qf8 9. Qh5 Re8 10. Ra7