Author: Robert Hyatt
Date: 22:44:26 04/11/03
Go up one level in this thread
On April 11, 2003 at 12:24:43, Uri Blass wrote: >On April 11, 2003 at 12:06:31, Robert Hyatt wrote: > >>Here is a position from a game Crafty played on ICC today. >> >>[d]8/8/1p4k1/3K1p1p/1R3P1P/6P1/3r4/8 w - - >> >>This was showing as a draw until white uncorked the Rd4 interposition to block >>the check. At this point, black's Rxd4 starts to climb as it should. The score >>goes to >>-10 quickly. >> >>This seems to be an example of an engine that misses the power of the "distant >>passed >>pawn". I won't name names to protect the guilty, but it shows that certain >>pieces of >>knowledge are key to prevent a draw from becoming a loss. > >There is some stage when the passed pawn is trapped and if the program >can see it and stop there to evaluate then it may evaluate it wrong. > >I suspect that the problem is in evaluating a pawn endgame >with no passed pawns. > >If you search deep but not deep enough you can see that b6 is gone >so to solve it by evaluation you need evaluation >of pawns relative to the kings to understand that black is losing inspite of the >fact that black is a pawn up. > >Uri That has to be part of the evaluation. IE you have to know that you can give the pawn up if your king is closer to the remaining pawns than the opposing king is... I do that obviously...
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.