Author: blass uri
Date: 08:42:10 06/05/00
Go up one level in this thread
On June 05, 2000 at 11:10:56, Albert Silver wrote: >On June 04, 2000 at 22:07:25, blass uri wrote: > >>On June 04, 2000 at 19:33:45, Albert Silver wrote: >><snipped> >>>What difference would it make if without special-case knowledge the >>>program judges this at +1.00 or with special-case knowledge it lowers this eval >>>to +0.4? In both cases it will believe it has the advantage and in both cases >>>will refuse to draw and play it out, only in the second case you'll be slowing >>>the program by giving it useless special knowledge. >> >>No >>It is not useless knoledge because you can avoid trading to the wrong endgame by >>this knowledge. >> >>Uri > >How? The position isn't a dead draw yet so why do you want the program to judge >it as such? +0.4 pawns advantage instead of +1 is not judging the position as a dead draw. True, in the given position, it is doubtlessly a draw with correct >play by both sides, but a slight change in the disposition of the pieces and >pawns, and nothing is clear yet. To say that just having an extra pawn in a rook >endgame with connected pawns all on one wing is a draw without looking at the >specific conditions would be a grave mistake. So not only is the knowledge >useless, but it's dangerous. You could add a large quantity of conditions to >make this knowledge more accurate and therefore useful, but I doubt very much >the cost would be worth it. The question is if all the programmers are interested only in doing a better program. It is possible to try to do a program with more knowledge without caring about the cost in speed and the level in practical games. Maybe this idea is productive for correspondence games because of diminishing returns from depth but I do not know. Uri
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.