Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: a strange positional evaluation

Author: Simon Finn

Date: 06:49:15 09/30/01

Go up one level in this thread


On September 30, 2001 at 04:27:44, Uri Blass wrote:

>On September 29, 2001 at 21:05:58, allan johnson wrote:
>
>>On September 29, 2001 at 18:02:31, Uri Blass wrote:
>>
>>>PATZER 3.61 - Quark
>>>[D]8/6kB/8/6P1/P7/4p3/3bK3/8 w - - 0 1
>>>
>>>I wonder what is the reason that patzer evaluated
>>>this drawn position as +2.50 for white.
>>>
>>>Even programs with only material evaluation can see only +1
>>>for white.
>>>
>>>What is the reason for 1.5 pawns positional bonus?
>>>
>>>Uri
>> Uri: Perhaps the programmer did not include rules about end games when the
>>bishops are on opposite colours?
>>Al
>
>even without rules for opposite bishop the question still remain
>What is the reason for 1.5 pawns positional bonus.
>
>I do not think that a dustance passed pawn justifies 1.5 pawn positional bonus.

Move the other White pawn from g5 to g4:

[D]8/6kB/8/8/P5P1/4p3/3bK3/8 w - - 0 1

Now White appears to be winning after (for example)

1. Be4 Kf6 2. Kd3 Ke5 3. Bf3 Kd6 4. Kc4 Kc7 5. Kb5 Bc3 6. a5 etc.

I guess that it's very difficult to write an evaluation function that recognises
the key difference between the two positions (the White pawn is too vulnerable
on g5).

Simon






>
>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.