Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Are there new ideas to make movegeneration faster ?

Author: Bruce Moreland

Date: 09:22:45 09/01/99

Go up one level in this thread


On September 01, 1999 at 09:02:34, Bas Hamstra wrote:

>On September 01, 1999 at 08:26:48, stefan wrote:
>
>>Where can I read about it?
>>
>>Thank you.
>
>If you want to go fast, you shouldn't only look at the movegenerator.
>What is critital is:
>
>- fast capture generation (non captures are generated far less often)
>- fast attack generation
>- fast make/unmake
>
>If you are going to make a fast searcher above things are important. If you make
>a smart (more evaluation, but slower) program, they become less important,
>because most of the time will be spent in the eval anyway.
>
>So I wouldn't focus too much on move generation alone. For instance captures are
>generated 10x as often as non-captures.
>
>Ways of generating captures and attacks fast (in my opinion):
>
>- 0x88
>- Bitboards
>- Incremental attackmap (IE attackmap is updated at make/unmake time)
>
>All these methods can see if a piece is attacked by another piece very fast,
>without scanning the board.

Mine does quite well enough scanning the board.

bruce



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.