Author: Robert Hyatt
Date: 18:25:05 05/21/01
Go up one level in this thread
On May 21, 2001 at 15:59:33, Uri Blass wrote:
>
>I read also the relevant position in a book of kasparov.
>I think that +0.74 is not convincing.
>Bxg7 does not lose material but white gets only Rook and a pawn and exposed king
>for bishop and knight.
>I am not convinced if white is winning and the position seems to be unclear to
>me.
>
>If Deep blue JR's move gives a clear advantage for white then it is clearly
>better
>
OK... now we have something to work with. I ran this position on a 4x550
machine for a good while. Here is what I get:
(2) 13 10:00 0.39 1. Nd2 Rcd8 2. Qg4 f6 3. Rc1 Bc5 4.
Rxd8 Qxd8 5. Ne4 Qe7 6. Nxc5 bxc5 7.
Bc3 Nb3 8. Re1
13 37:26 0.63 1. Rh5 Qc2 2. Qxe7 Qxb2 3. Qxa7 Nc4
4. a4 Qa2 5. Ng5 Rcd8 6. Nh7 Qe2 7.
Qc7 Qxh5 8. Nxf8
13-> 37:26 0.63 1. Rh5 Qc2 2. Qxe7 Qxb2 3. Qxa7 Nc4
4. a4 Qa2 5. Ng5 Rcd8 6. Nh7 Qe2 7.
Qc7 Qxh5 8. Nxf8
14 45:18 0.46 1. Rh5 Qc2 2. Qxe7 Qxb2 3. Qxa7 Nc4
4. a4 Qa2 5. Qd7 g6 6. Rd5 Ra8 7. Re1
Qxa4
(2) 14-> 82:37 0.46 1. Rh5 Qc2 2. Qxe7 Qxb2 3. Qxa7 Nc4
4. a4 Qa2 5. Qd7 g6 6. Rd5 Ra8 7. Re1
Qxa4
15 122:55 ++ 1. Rh5!!
15 146:03 0.87 1. Rh5 Qc2 2. Qxe7 Qxb2 3. Ne5 Rc1
4. g4 Rxf1+ 5. Kxf1 Qa1+ 6. Kg2 Qd4
7. Qh4 Qd5+ <HT>
It seems that I settle on DB Jr's move after 30+ minutes, although I am nowhere
near their score after 2+ hours of searching (they had +1.5, I have +.87... I
think Crafty might be "sniffing" what they are really seeing here. I will wait
to see if depth=16 finds something better.
But now we need some data from other programs as well to see if we all agree
that Rh5 is best, and by how much, and how long it takes to find it. I will
try to ask Murray how long DB Jr needs to find Rh5 and at what depth it finds
it, for reference...
This page took 0 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.