24 Nov
2018
24 Nov
'18
1:47 p.m.
Going back to Dan's original post, I liked the passage
It might be interesting to determine which ratios of which pieces lead to
a fairly even game.
It would be delightful to come up with some meaningful asymptotic sense in which (for instance) knights and pawns are worth infinitesimally less than rooks and bishops, but a knight is worth exactly 5 pawns and a rook is worth exactly sqrt(2) bishops. (The numbers 5 and sqrt(2) are just random numbers that came to mind; I'm not suggesting that these are the actual values of the sorts of ratios Dan is asking about.) Jim Propp