Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Game 2, IBM Kasparov vs. Deep Blue correction on Qe3 draw position

Author: Uri Blass

Date: 10:34:47 03/04/06

Go up one level in this thread


On March 04, 2006 at 12:56:17, Lar Mader wrote:

>I gave Fritz 9 and Deep Shredder 9 about 20 minutes each.  They both settled on
>Qe3 as the best move, but they did not understand that there was a forced draw
>by repetition.  They still evaluated black's position as worse by roughly a
>pawn, plus or minus.  Sorry for not clipping the analysis, I just figured the
>point was to see if they could find the draw...

There is no forced draw and I remember that even yace after long analysis saw a
small positive score for white.

I think that the main question is how much time they need to find a line like
1...Qe3 2.Qxd6 Re8 3.Qd7+(or Qc7+) Re7 4.Qc6 with no perpetual check(probably a
draw but I am not sure).

3.h4 h5 4.Bf3 Qc1+ is leading to a forced draw and this is the problem that they
cannot see in tournament time control from the root.


I know that programs of today can find both Qe3 and avoid Kf1

I did not test rybka in the Kf1 problem but finding Qe3 is easy for it.
Here is rybka's analysis on Athlon64 3000(512 mbytes hash 2 ghz)

New game
1r6/5kp1/RqQb1p1p/1p1PpP2/1Pp1B3/2P4P/6P1/5K2 b - - 0 1

Analysis by Rybka 1.01 Beta 13d 32-bit:

1...Qxc6
  ±  (1.37)   Depth: 3   00:00:00
1...Qxc6
  +-  (1.78)   Depth: 4   00:00:00
1...Qxc6 2.dxc6
  +-  (1.79)   Depth: 5   00:00:00
1...Qxc6 2.dxc6 Be7
  +-  (1.70)   Depth: 6   00:00:00  2kN
1...Qxc6 2.dxc6 Bc7 3.Ra7
  +-  (2.12)   Depth: 7   00:00:00  8kN
1...Qxc6 2.dxc6 Be7 3.Ke2 Ke8
  +-  (2.38)   Depth: 8   00:00:00  17kN
1...Qxc6 2.dxc6 Rc8 3.Ra7+ Kf8 4.Rb7 Rc7
  +-  (2.58)   Depth: 9   00:00:00  46kN
1...Qxc6 2.dxc6 Rc8 3.Ra7+ Rc7 4.Ra5 Rc8 5.Ke2
  +-  (2.48)   Depth: 10   00:00:00  67kN
1...Qxc6 2.dxc6 Rc8 3.Ra7+ Kf8 4.Rb7 Rc7 5.Rxb5 Ra7
  +-  (2.54)   Depth: 11   00:00:01  108kN
1...Qxc6 2.dxc6 Rc8 3.Ra7+ Kf8 4.Ke2 Rc7 5.Ra5 Ke8 6.Ke3
  +-  (2.76)   Depth: 12   00:00:03  283kN
1...Qxc6 2.dxc6 Rc8 3.Ra7+ Rc7 4.Ra8 h5 5.Bd5+ Ke7 6.Rg8 e4 7.Rxg7+ Kf8 8.Rg8+
  +-  (2.81)   Depth: 13   00:00:04  473kN
1...Qe3 2.Qxd6 Re8 3.h4 Re7 4.Bf3 Qc1+ 5.Kf2 Qd2+ 6.Kg3 Qe1+ 7.Kh3 Qh1+ 8.Kg4
  +-  (2.05)   Depth: 13   00:00:26  1741kN
1...Qe3 2.Qxd6 Re8 3.h4 h5 4.Bf3 Qc1+ 5.Kf2 Qd2+ 6.Be2 Qf4+ 7.Kg1 Qc1+ 8.Kh2
  +-  (2.58)   Depth: 14   00:00:44  2883kN
1...Qe3 2.Qxd6 Re8 3.h4 h5 4.Bf3 Qc1+ 5.Kf2 Qd2+ 6.Be2 Qf4+ 7.Kg1 Qc1+ 8.Kh2
  +-  (2.58)   Depth: 15   00:01:08  4480kN
1...Qe3 2.Qxd6 Re8 3.h4 h5 4.Bf3 Qc1+ 5.Kf2 Qd2+ 6.Be2 Qf4+ 7.Kg1 Qc1+ 8.Kh2
  +-  (2.38)   Depth: 16   00:02:04  7985kN
1...Qe3 2.Qxd6 Re8 3.h4 h5 4.Bf3 Qc1+ 5.Kf2 Qd2+ 6.Be2 Qf4+ 7.Kg1 Qc1+ 8.Kh2
  +-  (1.51)   Depth: 17   00:04:31  17205kN
1...Qe3 2.Qxd6 Re8 3.h4 h5 4.Bf3 Qc1+ 5.Kf2 Qd2+ 6.Be2 Qf4+ 7.Kg1 Qc1+ 8.Kh2
  +-  (1.51)   Depth: 18   00:07:45  29910kN

(,  04.03.2006)


Uri



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.