Computer Chess Club Archives


Search

Terms

Messages

Subject: 10^46 positions in chess - by Shirish S. Chinchalkar

Author: S. Loinjak

Date: 05:38:52 01/08/03


Dear chess friends,

I'm interested in upper bounds on the number of legal positions in chess.

------------------------------------------------------------

The most interesting article about this seems to be the one by Shirish S.
Chinchalkar from the ICCA Journal, Volume 19:  Number 3  (September 1996):

'An Upper Bound for the Number of Reachable Positions'

Unfortunately the abstract is missing on the ICGA page:
http://www.cs.unimaas.nl/ICGA/journal/contents/node19.html

I read the limit he has calculated was about 10^46.
As I don't have access to the ICCA Journal I want to ask in this forum whether
someone has read the article and could tell me the exact number?

In my opinion all positions should be treated different which would generate
different sets of legal moves in a chess program (of course a bug free one - at
least in the move generator).

Therefor I'd also like to know whether the positions have been distinguished by
- color to move
- castle rights
- e.p. rights

I think to solve chess completely it might not be neccesary to take care about
the 3-fold-repetition rule and the 50-moves-draw rule as they don't affect the
set of legal moves in a position.

------------------------------------------------------------

Maybe also someone knows how Richard Schroeppel came to his estimation(?!) of
10^40 legal positions. I'm quite sure (after having done some calculations) that
this is too low.

--> http://keithlynch.net/cryonet/f/SchroeppelR.html (msg. 223 - on top)
--> http://www.semaphorecorp.com/chess/

------------------------------------------------------------


Best Regards and a Happy New Year,
Sini



This page took 0.01 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.