Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Bit board representation

Author: Gian-Carlo Pascutto

Date: 07:15:13 05/28/01

Go up one level in this thread


On May 28, 2001 at 06:33:06, Cheok Yan Cheng wrote:

>By the way, has anyone make an experimental test on the speed of bit board vs
>array board?

How are you going to do this? The board structure has implications
in the whole program. You would have to write an identical program
using both bitboards and array boards, each fully optimized to use
all tricks allowed by one of the implementations.

You will run into trouble with the eval for example, where some
things are very fast with bitboards and others are very fast for
array boards. A real program will probably try to avoid using
the slower eval term unless it really has to, whereas it can be
a nearly free addition in the other. How are you going to compare
this?

You could look at the top programs and conclude that all except
crafty are array/attackboard based, but most of them have been
in development for a long time, from before bitboards were well-
known. Maybe if those top programmers had to restart now they
would use bitboards, then again maybe not.

That said, I think it's safe to say that bitboards wont be very
efficient if you aren't dealing with 32 or 64 square boards.

--
GCP



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.