Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Correction on the 1st diagram and analysis

Author: emerson tan

Date: 05:28:32 07/29/03

Go up one level in this thread


Deep Blue's 36.axb5 and 37.Be4 in game 2 of the match in 1997 along with Deep
Blue's 11...h5 in game 5 are the moves that made Kaparov to imply that there was
human intervention.

Kasparov doesnt believe that Deep Blue or any other chess program can play moves
like those mentioned.

I tried the positions on Deep Fritz 7 and Junior 5 and here are the results:
Comp Deep Blue - Kasparov,G


[D]r1r1q1k1/6p1/p2b1p1p/1p1PpP2/PPp5/2P4P/R1B2QP1/R5K1 w - - 0 1

Analysis by Deep Fritz 7:

36.Qb6!
  +-  (1.59)   Depth: 8/21   00:00:00  251kN
36.Qb6 Bc7 37.Qe6+ Qxe6 38.fxe6 Rcb8 39.axb5 Rxb5 40.Be4 Bd6 41.Rxa6 Rxa6
  +-  (1.50)   Depth: 9/22   00:00:00  497kN
36.Qb6 Bc7 37.Qe6+ Qxe6 38.fxe6 Rcb8 39.axb5 Rxb5 40.Be4 Bd6 41.Rxa6 Rxa6
42.Rxa6
  +-  (1.50)   Depth: 10/25   00:00:01  955kN
36.Qb6 Rd8 37.Be4 Be7 38.axb5 Rd6 39.Qa5 Bd8 40.Qa4 Bb6+ 41.Kh2 Qh5 42.Qc2 Qg5
43.bxa6
  +-  (1.56)   Depth: 11/29   00:00:03  2713kN
36.Qb6 Rd8 37.Be4 Be7 38.axb5 Rd6 39.Qa5 Bd8 40.Qa4 Ra7 41.Kh2 Bc7 42.bxa6 Qh5
  +-  (1.63)   Depth: 12/33   00:00:09  6451kN
36.Qb6 Rd8 37.Be4 Be7 38.axb5 Rd6 39.Qa5 Bd8 40.Qa4 Ra7 41.Kh2 Bc7 42.bxa6 Qh5
  +-  (1.63)   Depth: 13/34   00:00:19  14662kN
36.Qb6 Qe7 37.axb5 Rab8 38.Qxa6 e4 39.Bxe4 Qe5 40.Bf3 Qe3+ 41.Kf1 Qd3+ 42.Kf2
Be5 43.Qe6+ Kh8 44.d6 Qxc3
  +-  (1.41)   Depth: 14/35   00:00:41  31018kN
36.Qb6 Qe7 37.axb5 Rab8 38.Qxa6 e4 39.Re1 Qd8 40.Raa1 Ra8 41.Qb7 Rab8 42.Qa7 Bc7
43.Qa6
  +-  (1.41)   Depth: 15/38   00:01:30  69547kN
36.Qb6 Qe7 37.axb5 Rab8 38.Qxa6 e4 39.Re1 Qd8 40.Raa1 Ra8 41.Qb7 Rab8 42.Qa7 Bc7
43.Qa6
  +-  (1.41)   Depth: 16/40   00:03:10  146745kN
36.Qb6 Qe7 37.axb5 Rab8 38.Qxa6 e4 39.Re1 Qd8 40.Raa1 Ra8 41.Qb7 Rab8 42.Qa7 Be5
43.Qe3
  ±  (1.34)   Depth: 17/44   00:06:27  301496kN
36.Qb6--
  ±  (1.06)   Depth: 18/47   00:26:31  1204122kN
36.Qb6 Qe7 37.axb5 Rab8 38.Qxa6 e4 39.Bxe4 Qe5 40.Kf1 Rd8 41.Re2 Qxc3 42.Rae1
Bxb4
  ±  (1.03)   Depth: 18/72   00:37:04  1683194kN
36.Qb6 Qe7 37.axb5 Rab8 38.Qf2 axb5 39.Be4 Qd8 40.Ra6 Ra8
  ±  (1.03)   Depth: 19/54   01:01:56  2861739kN
36.axb5!
  ±  (1.06)   Depth: 19/54   01:21:20  3795485kN
36.axb5 axb5 37.Qb6 Rxa2 38.Rxa2 Ra8 39.Rxa8
  ±  (1.06)   Depth: 20/51   01:56:18  5485120kN
36.axb5 axb5 37.Qb6 Rxa2 38.Rxa2 Ra8 39.Rxa8
  ±  (1.06)   Depth: 21/60   03:57:20  11339720kN
36.axb5 axb5 37.Qb6 Rxa2 38.Rxa2 Ra8 39.Rxa8
  ±  (1.06)   Depth: 22/61   08:19:18  23943525kN

(, MyTown 26.07.2003)


Comp Deep Blue - Kasparov,G
[D]r1r1q1k1/6p1/3b1p1p/1p1PpP2/1Pp5/2P4P/R1B2QP1/R5K1 w - - 0 1

Analysis by Deep Fritz 7:

37.Qb6 Rxa2 38.Rxa2 Bc7 39.Qa6 Rd8 40.Qb7 Bb8 41.Be4 Bd6 42.Ra7
  ±  (0.75)   Depth: 9/20   00:00:00  245kN
37.Qb6 Bc7 38.Qe6+ Qxe6 39.dxe6 Rxa2 40.Rxa2 Kf8 41.Kf1 Bb6 42.Ra6 Be3 43.Be4
  ²  (0.69)   Depth: 10/28   00:00:00  428kN
37.Be4!
  ±  (0.72)   Depth: 10/28   00:00:00  488kN
37.Be4 Rxa2 38.Qxa2 Rd8 39.Qa6 Qd7 40.Qb6 Bc7 41.Qe6+ Qxe6 42.dxe6 Bb6+ 43.Kf1
Kf8 44.Ke2 Ke7
  ±  (0.75)   Depth: 11/31   00:00:02  1729kN
37.Be4 Rxa2 38.Qxa2 Rd8 39.Qa6 Qd7 40.Qb6 Bc7 41.Qc5 Bd6 42.Qe3 Qe8 43.Ra5 Rb8
44.Kh2 Qh5 45.Ra7
  ±  (0.72)   Depth: 12/31   00:00:04  3158kN
37.Be4 Rcb8 38.Ra6 Qd8 39.Kh2 Qc8 40.R6a5 Rxa5 41.Rxa5 Qd8 42.Qg3 Qd7 43.Qg6
  ±  (0.81)   Depth: 13/34   00:00:16  12565kN
37.Be4 Rcb8 38.Ra6 Qd8 39.Kh2 Rxa6 40.Rxa6 Kh8 41.Kh1 Kg8 42.Qa7
  ±  (0.84)   Depth: 14/39   00:00:34  26570kN
37.Be4 Rcb8 38.Ra6 Qd8 39.Kh2 Rxa6 40.Rxa6 Kh8 41.Kh1 Kg8 42.Qa7 Rb6 43.Rxb6
  ±  (0.84)   Depth: 15/37   00:00:50  40088kN
37.Be4 Rcb8 38.Ra6 Qd8 39.Kh2 Rxa6 40.Rxa6 Kh8 41.Kh1 Kg8 42.Qa7 Bf8 43.Kg1
  ±  (0.84)   Depth: 16/40   00:01:24  67869kN
37.Be4 Rcb8 38.Ra6 Qd8 39.Kh2 Rxa6 40.Rxa6 Kh8 41.Kh1 Kg8 42.Qa7 Bf8 43.Kg1 Bd6
44.Kh2
  ±  (0.84)   Depth: 17/47   00:02:34  126289kN
37.Be4 Rcb8 38.Ra6 Qd8 39.Kh2 Rxa6 40.Rxa6 Kh8 41.Kh1 Kg8 42.Qa7 Bf8 43.Kg1 Bd6
44.Kh2
  ±  (0.84)   Depth: 18/48   00:05:29  269014kN
37.Be4 Rcb8 38.Ra6 Qd8 39.Kh2 Rxa6 40.Rxa6 Rb7 41.Qg3 Rb8 42.Qe3 Kf8 43.Qa7
  ±  (0.84)   Depth: 19/48   00:10:24  516602kN
37.Be4 Rcb8 38.Ra6 Qd8 39.Kh2 Rxa6 40.Rxa6 Rb7 41.Qg3 Rb8 42.Qe3 Rb7 43.Rc6 Kh7
44.Kh1 Rb8
  ±  (0.84)   Depth: 20/52   00:21:52  1082064kN
37.Be4 Rcb8 38.Ra6 Qd8 39.Kh2 Rxa6 40.Rxa6 Rb7 41.Qg3 Rb8 42.Qe3 Rb7 43.Rc6 Kh7
44.Kh1 Kg8 45.Ra6
  ±  (0.84)   Depth: 21/55   00:45:23  2255740kN
37.Be4 Rcb8 38.Ra6 Qd8 39.Kh2 Rxa6 40.Rxa6 Rb7 41.Qg3 Rb8 42.Qe3 Rb7 43.Rc6 Kh7
44.Kh1 Kg8 45.Ra6
  ±  (0.84)   Depth: 22/59   01:39:26  4936283kN
37.Be4 Rcb8 38.Ra6 Qd8 39.Kh2 Rxa6 40.Rxa6 Rb7 41.Qg3 Rb8 42.Qe3 Rb7 43.Rc6 Kh7
44.Kg1 Kg8
  ±  (0.84)   Depth: 23/65   03:51:24  11313418kN
37.Be4 Rcb8 38.Ra6 Qd8 39.Kh2 Rxa6 40.Rxa6 Rb7 41.Qg3 Rb8 42.Qe3 Rb7 43.Rc6 Kh7
44.Kg1 Bb8
  ±  (0.84)   Depth: 24/76   09:47:23  28378832kN

(, MyTown 27.07.2003)

Kasparov,G - Comp Deep Blue
[D]r2qk2r/pp3ppp/2p1pn2/4n3/1b6/3P2PP/PPPN1PB1/R1BQK2R b KQkq - 0 1

Analysis by Junior 5.0:

11...Qa5 12.0-0
  ³  (-0.62)   Depth: 3   00:00:00  1kN
11...Qa5 12.d4 Ng6 13.c3 Bd6
  ³  (-0.39)   Depth: 6   00:00:00  8kN
11...Qd4 12.a3 Bxd2+ 13.Qxd2 Rd8
  ³  (-0.47)   Depth: 6   00:00:00  21kN
11...Qd4 12.a3 Bxd2+ 13.Qxd2 Rd8
  ³  (-0.47)   Depth: 6   00:00:00  23kN
11...Qd4 12.0-0 Bxd2 13.Qxd2 0-0-0 14.Qg5
  =  (-0.21)   Depth: 9   00:00:00  70kN
11...0-0 12.a3 Ba5 13.b4 Bb6 14.0-0 Qd4
  ³  (-0.31)   Depth: 9   00:00:00  111kN
11...0-0 12.a3 Ba5 13.b4 Bb6 14.0-0 Qd4
  ³  (-0.31)   Depth: 9   00:00:01  434kN
11...0-0 12.0-0 Ned7 13.c3 Be7 14.d4 e5
  =  (-0.21)   Depth: 12   00:00:01  838kN
11...0-0 12.0-0 Ned7 13.c3 Be7 14.d4 e5
  =  (-0.21)   Depth: 12   00:00:05  5059kN
11...0-0 12.0-0 Ba5 13.Ne4 Ned7 14.Bg5 e5 15.d4
  =  (-0.14)   Depth: 15   00:00:07  7238kN
11...h5 12.Qe2 Bd6 13.d4 Ng6 14.Be4 Nxe4 15.Nxe4 Bc7 16.Be3 Ba5+ 17.Bd2 e5
  =  (-0.24)   Depth: 15   00:00:20  19835kN
11...h5 12.Qe2 Bd6 13.d4 Ng6 14.Be4 Nxe4 15.Nxe4 Bc7 16.Be3 Ba5+ 17.Bd2 e5
  =  (-0.24)   Depth: 15   00:00:51  47571kN
11...h5 12.Qe2 Ng6 13.c3 Bd6 14.h4 0-0 15.0-0 Ng4 16.Bf3 N6e5
  =  (-0.15)   Depth: 17   00:01:22  77848kN
11...h5 12.Qe2 Ng6 13.c3 Bd6 14.h4 0-0 15.0-0 Ng4 16.Bf3 N6e5
  =  (-0.15)   Depth: 17   00:04:39  250610kN
11...h5 12.Qe2 Ng6 13.c3 Bd6 14.h4 0-0 15.0-0 Ng4 16.Bf3 N6e5
  =  (-0.15)   Depth: 18   00:05:31  298959kN
11...h5 12.Qe2 Ng6 13.c3 Bd6 14.h4 0-0 15.0-0 Ng4 16.Bf3 N6e5
  =  (-0.15)   Depth: 18   00:11:58  629071kN
11...h5 12.Qe2 Ng6 13.c3 Bd6 14.h4 0-0 15.Nc4 Be7 16.0-0 b5 17.Ne3 Qb6
  =  (-0.10)   Depth: 19   00:14:07  746855kN
11...h5 12.Qe2 Ng6 13.c3 Bd6 14.h4 0-0 15.Nc4 Be7 16.0-0 b5 17.Ne3 Qb6
  =  (-0.10)   Depth: 19   00:33:27  1757903kN
11...h5 12.Qe2 Bd6 13.Rf1 Qa5 14.d4 Ng6 15.c3 Qf5 16.Nc4 0-0-0 17.Ne3 Qb5
18.Qxb5 cxb5
  =  (-0.06)   Depth: 20   00:41:01  2184898kN
11...h5 12.Qe2 Bd6 13.Rf1 Qa5 14.d4 Ng6 15.c3 Qf5 16.Nc4 0-0-0 17.Ne3 Qb5
18.Qxb5 cxb5
  =  (-0.06)   Depth: 20   01:33:33  677195kN
11...h5 12.Qe2 Ng6 13.c3 Bd6 14.h4 Qd7 15.Ne4 0-0-0 16.Be3 Kb8 17.Nxf6 gxf6
18.a4
  =  (-0.05)   Depth: 21   01:46:57  1431959kN
11...h5 12.Qe2 Ng6 13.c3 Bd6 14.h4 Qd7 15.Ne4 0-0-0 16.Be3 Kb8 17.Nxf6 gxf6
18.a4
  =  (-0.05)   Depth: 21   03:59:12  4231602kN
11...h5 12.Qe2 Ng6 13.c3 Bd6 14.h4 Qd7 15.Ne4 0-0-0 16.Nxf6 gxf6 17.Be3 Kb8
18.Bf3 Bc7
  =  (-0.05)   Depth: 22   04:28:34  1610632kN
11...h5 12.Qe2 Ng6 13.c3 Bd6 14.h4 Qd7 15.Ne4 0-0-0 16.Nxf6 gxf6 17.Be3 Kb8
18.Bf3 Bc7
  =  (-0.05)   Depth: 22   09:01:24  3130210kN
11...h5 12.Qe2 Qc7 13.c3 Bd6 14.d4 Ng6 15.h4 0-0-0 16.0-0 e5 17.Ne4 Be7 18.Ng5
Rhf8 19.Qc4 Nd5 20.a4
  =  (-0.06)   Depth: 25   20:06:52  2794565kN

(, MyTown 28.07.2003)

It shows that given enough depth, some programs can replicate the moves.



 My computer was just running Deep Fritz 7 at an average of 600KNS, Deep Blue
was running around 125 million nodes per second


Actually, Mig Greengard already mentioned before that junior 5 was able to find
the 11...h5 move. He discovered that after he interviewed Kasparov regarding the
match with IBM. Mig mentioned that the Junior 5 program was even installed in
the laptop Gary was using.



It shows that Deep Blue did not cheat. The IBM team had several help during its
development from different Grandmasters like Joel Benjamin, Robert Byrne, Miguel
Illescas, Nick DeFirmain and John Fedorowincz. They may have gotten some
feedback also from Michael Rohde and Larry Christiansen during the training
matches. So the evaluation




This page took 0.01 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.