Speedup by bitboards

Programming Topics (Computer Chess) and technical aspects as test techniques, book building, program tuning etc

Moderator: Andres Valverde

Re: Speedup by bitboards

Postby Onno Garms » 31 Jul 2007, 18:42

Uri Blass wrote:If somebody can write a program that gives the same analysis as fruit except doing it twice faster then this is going to be convincing.


A bitboard rewrite won't produce absolutely identical results. This is because the rewrite will create moves in a different order. Move ordering depends on the order of generation in case of ties.
User avatar
Onno Garms
 
Posts: 128
Joined: 17 Feb 2007, 11:17
Location: Bonn, Germany

Re: Speedup by bitboards

Postby Dann Corbit » 07 Aug 2007, 02:47

Onno Garms wrote:
Dann Corbit wrote:I guess that the switch to bitboard, together with being table driven gives about a 2x speedup over array representation.


What do you mean by being table driven? Have tables like the bitboards of attacked squares for every square? Doesn't every bitboard engine have that?

Or do you think of some other kind of tables?


Some other kinds of tables. Perhaps the author will send you the source code if you ask for it.

I do not think I should divulge too much, because I was given the source code for a specific purpose only.
Dann Corbit
 

Previous

Return to Programming and Technical Discussions

Who is online

Users browsing this forum: No registered users and 4 guests