chess ai

[Guest]

Archived from groups: comp.ai.games,comp.programming (More info?)

What type of algorithm is generally implemented with chess ai engines?

 

[Guest]

Archived from groups: comp.ai.games,comp.programming (More info?)

j0mbolar wrote:
> What type of algorithm is generally implemented with chess ai engines?

Some derivative of alpha-beta: MTD(f), PVSearch, NegaScout, etc.

Doing a google search for "chess programming" yields wonderful results.
Then you will use the information you get from the basic tutorials
there to search the scientific articles in citeseer.

A search for "rookie thesis" might get to a master's thesis on chess.
Rookie 2.0 was the program.

The ng for this is rec.games.chess.computer

NR


-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----

 

[Malcolm]

Archived from groups: comp.ai.games,comp.programming (More info?)

"j0mbolar" <j0mbolar@engineer.com> wrote in message
>
> What type of algorithm is generally implemented with chess ai
> engines?
>
The heart of it is a position strength function. For a simple program this
would give 100% for checkmate (handled specially), and a crude points scheme
for material. Folr a good chess program the function gets very
sophisticated.
The second part is a tree analysis program. Because there are about fifty
possible moves each turn a brute force analysis is too slow. However using
the position strength function you can eliminate quickly those moves which
lead to bad positions, and concentrate on the rational lines of play.
Assuming that the opposition always take the strongest move, you play
whatever leads you to the best position at the end of analysis, as
determined by the position sterength function.