Author: Robert Hyatt
Date: 20:52:45 01/30/00
Go up one level in this thread
On January 30, 2000 at 22:17:56, David Eppstein wrote: >I noticed that three out of four of crafty's wins were ground-out endgames. I >guess the new potential-passed-pawn evaluation term helped a lot? I think it has helped. IE I didn't feel like I was badly outsearching _anybody_ so far... I simply saw lots of significant positional disagreement. Two games come to mind. round 3 vs Nimzo... crafty played an 'iffy' variant in the bishops opening (as black) but Nimzo allowed a queen trade. That turned it into at least a draw, but the resulting king and pawn ending was nicely played. I didn't see all of this game so I am not sure exactly what happened. In round 4, vs LG, LG allowed crafty to create a passed a-pawn early in the game, then refused to blockade it. It eventually reached the 7th rank and caused tactical problems... I don't mind endgames nowadays, compared to a few years ago. Some programs have problems knowing where to put their kings, some don't understand passed pawns, outside passed pawns, and candidates/majorities. I'm a _long_ way from perfect. But also a long way from "stupid". :)
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.