Author: Jeremiah Penery
Date: 12:56:28 03/03/01
Go up one level in this thread
On March 02, 2001 at 09:15:23, Vincent Diepeveen wrote: >Hello, > >Deep Blue II - Kasparov > >As Seirawan and a few others indicated 44.Kh1 would have won the >game whereas the played 44.Kf1? by Deep blue is drawing the >game. > >On a dual 800 PIII diep needs 6:39 to get to Kh1. That's about >3 minutes at Bob's quad as in practical tournament play the quad is >exactly 2 times faster as a dual. > >So on todays hardware i would have found it with the IPCCC2001 version >(second half of the week, first half of the week had several bugs). This is the output of Diep when it changed from Kf1 to Kh1: 04:09 22359619 (2628719) 14 2.56 Kg1-f1 Rb7-b8 Ra8-a6 Qb6-e3 Qc6-d7 Kf7-g8 Qd7xd 6 Rb8-f8 Qd6-e6 Kg8-h7 Be4-f3 Qe3xc3 Qe6-e7 Qc3-d3 Kf1-g1 Qd3-e3 Kg1-h2 Qe3-f4 g 2-g3 Qf4-d2 Bf3-g2 ++ g1-h1 06:39 36563211 (3719499) 14 2.72 Kg1-h1 Rb7-b8 Ra8-a6 Qb6xc6 d5xc6 Rb8-c8 Ra6-a5 Kf7-e7 Ra5xb5 Rc8-b8 Rb5-a5 Rb8-c8 Be4-d5 Bd6-c7 Ra5-a6 Ke7-d6 Bd5xc4 You still show a score of +2.56 for Kf1. If Diep really saw the tactics it would have a much lower score. Kh1 is hardly better according to Diep - it could be a small evaluation thing where Diep likes the king to be hiding in the corner. If you let it run longer, how can you be sure it doesn't ever change its mind back to Kf1?
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.