Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Attack detection with bitboards: alternative to "rotated bitboards"

Author: Robert Hyatt

Date: 08:00:19 12/05/99

Go up one level in this thread


On December 05, 1999 at 10:55:45, KarinsDad wrote:

>The problem is that some of us do not have access to 64 bit machines (at least I
>no longer have access to them).
>
>My method is similar to Alessandro's with 15 bytes to handle bishop and diagonal
>queen attacks and moves.
>
>The advantage of this for attacks is that you only have to check 1 of the 15
>bytes out of each bitboard (actually, you only need 13 bytes, but my structures
>are currently set up as 16 bytes just for boundary convenience) per direction
>(2). The same for moves (I would like to use a system in the future of only
>creating those diagonal bitmaps that can possibly be used by either side and
>then only filling in a new byte in the structure as I go down the tree and move
>a bishop or queen).
>
>The disadvantage is that you have to figure out which 2 bytes to use, but I
>think you have a similar issue with 64 bits (although I haven't studied the
>Crafty code to figure it out).
>
>What I haven't figured out from Alessandro's system yet is why he needs such
>large bitboard structures (64 * 256).
>
>KarinsDad :)


you need 256 entries because that is the number of possible states of any rank
or diagonal.

As far as 64 bits goes, within 5 years that is _all_ we will have.  Better to
plan for the future.  It is almost the present.  :)



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.