Optimize generate_piece_blocking_evasions()
authorMarco Costalba <mcostalba@gmail.com>
Wed, 11 Feb 2009 11:58:45 +0000 (12:58 +0100)
committerMarco Costalba <mcostalba@gmail.com>
Wed, 11 Feb 2009 18:01:06 +0000 (19:01 +0100)
commit1dc1cecf01352a39bc1b80c83f6e8e4e97d859e1
tree41cb8c8023cc04880b8a23ac4cc9cb3b3d710a63
parent67535711e8d86a975ff9aac9c2d2fdad24ec5915
Optimize generate_piece_blocking_evasions()

Rewrite as in generate_piece_moves() using a for
loop instead of the slower serializing of the
bitboard with pop_1st_bit()

This will allow also to merge with generate_piece_moves()
when we will drop legality constrain on generate_evasions()

Generated moves are not changed, but are generated in a
different order, this changes the number of nodes at fixed
depth test.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
src/movegen.cpp