Hacker News new | ask | show | jobs
by tromp 813 days ago
> For regular chess, I've calculated that there are most legal positions for 29 pieces

Improved counting shows the maximum to occur at 28 pieces [1] (also see the issue I filed on your github repo).

[1] https://www.chess.com/forum/view/general/on-the-number-of-ch...
1 comments
zone411 812 days ago
I don't have time to go through the thread at this time but I see that in the first post of this thread you just confirmed the number I calculated (8.7E+45) instead of your earlier estimate of 4.5E+46?
link
tromp 811 days ago
Your page only mentions

> Upper bound estimate of possible legal chess position (counts en passant, castling, sides to move, allows underpromotions): 8.7E+45.

But how exactly is this number determined? Is this a sampling based estimate?

My 8726713169886222032347729969256422370854716254 is an exact upperbound on the number of so-called urpositions, and is not sampling based.

In any case it will be easier to continue this discussion at https://github.com/lechmazur/ChessCounter/issues/1

link