additions to the shannon algorithm?

Discussion of chess software programming and technical issues.

Moderators: hgm, Rebel, chrisw

User avatar
Greg Strong
Posts: 388
Joined: Sun Dec 21, 2008 6:57 pm
Location: Washington, DC

Re: additions to the shannon algorithm?

Post by Greg Strong »

Another idea that has been kicked around is adding an eval bonus for the side to move, although, personally, I don't think it's a good idea. I think it is reasonable, however, for the king safety algorithm to consider the side to move in determining how serious a threat is.
Gian-Carlo Pascutto
Posts: 1243
Joined: Sat Dec 13, 2008 7:00 pm

Re: additions to the shannon algorithm?

Post by Gian-Carlo Pascutto »

Another very important thing is knowing which endgames are drawn.
diep
Posts: 1822
Joined: Thu Mar 09, 2006 11:54 pm
Location: The Netherlands

Re: additions to the shannon algorithm?

Post by diep »

Note that 1,3,3,5,9 is not the shannon way of counting,
but rather the Max Euwe manner of counting.

http://en.wikipedia.org/wiki/Max_Euwe

Max Euwe was world champion 1935-1937 and a very influental person in the chess scene.

Thanks,
Vincent
Gerd Isenberg
Posts: 2250
Joined: Wed Mar 08, 2006 8:47 pm
Location: Hattingen, Germany

Re: additions to the shannon algorithm?

Post by Gerd Isenberg »

diep wrote:Note that 1,3,3,5,9 is not the shannon way of counting,
but rather the Max Euwe manner of counting.

http://en.wikipedia.org/wiki/Max_Euwe

Max Euwe was world champion 1935-1937 and a very influental person in the chess scene.

Thanks,
Vincent
http://en.wikipedia.org/wiki/Chess_piece_point_value
Another system is used by Max Euwe and Hans Kramer in Volume 1 of their The Middlegame, with values:

* pawn = 1
* knight = 3½
* bishop = 3½
* rook = 5½
* queen = 10.
http://www.pi.infn.it/~carosi/chess/shannon.txt
(1) The relative values of queen, rook, bishop, knight and pawn are about
9, 5, 3, 3, 1, respectively.
Gerd