Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Detecting backward pawns

Author: James Robertson

Date: 20:01:10 06/18/99

Go up one level in this thread


On June 18, 1999 at 00:01:45, Robert Hyatt wrote:

>On June 18, 1999 at 00:00:43, Robert Hyatt wrote:
>
>>On June 17, 1999 at 00:10:11, James Robertson wrote:
>>
>>>I am trying to find the most efficient way to detect backward pawns, either with
>>>bitboards or the more conventional piece array. Any help is greatly
>>>appreciated!!
>>>
>>>James
>>
>>
>>I would suggest that you look at EvaluatePawns() in evaluate.c in crafty.  I
>>don't do 'backward pawns'.  Rather, I do a more general assessment of 'weak
>>pawns'  A pawn that can't advance without being attacked 1 more time than it
>>is defended is weak.  One that can't advance without being attacked two more
>>times than it is defended is _really_ weak.  These attacks only consider
>>pawns of course.
>>
>>The normal case many miss is (say) black pawns at c5/a5, white pawns at
>>a2/b2.  The b pawn is weak because it is undefended by a pawn at present,
>>and can't safely advance either...
>
>
>error.. make that black pawns at a4/c4 obviously... too late at night.
>
>

c5/a5 is still valid.... if White plays a4 then the b pawn can advance once and
not be attacked, yet it is still very weak....

James

>
>
>>
>>there are others, but what you are interested in finding is 'targets' that
>>will need attention throughout the game, as these are weaknesses that distract
>>pieces for defense..



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.