Author: Ralf Elvsén
Date: 07:48:48 01/16/02
Go up one level in this thread
On January 16, 2002 at 07:41:28, Graham Laight wrote: >It has occurred to me that it is wrong to evaluate a position in terms of >relative pawns (the "de facto" standard - whereby an evaluation of 2 means that >you're approximately the equivalent of 2 pawns ahead). > >This means that many aspects of evaluation have to be squeezed into a dimension >which is not appropriate at all. > >A better way would be to evaluate "winning probability". If a position was a >draw, the value would be 0.50 (or 50%). If the player should win 3 out of 4 >times, the eval should be 75%. If the player must win from here, then the >evaluation should be 100%. > >It seems strange when you think about it that all programmers have chosen to >adopt the traditional "pawn equivalence" standard. > >-g I have had the same idea, but on second thought it wasn't so smart. First of all, if we call the probability P, then any one-to-one mapping f(P) will do as good as a measure, since all you want to do is to maximize the score. Programmers make their evaluation (even without thinking about it) in this way. Secondly, you will probably get problems with nonlinearities in the search (i.e. the code gets uglier), so you would have to convert to the ordinary scale anyway. But maybe as an option for the GUI to display probability instead of pawns? :) Ralf
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.