Francesca M.A.D v0.15 released

Discussion of anything and everything relating to chess playing software and machines.

Moderator: Ras

User avatar
silentshark
Posts: 327
Joined: Sat Mar 27, 2010 7:15 pm

Re: Francesca M.A.D v0.15 released

Post by silentshark »

Actually, I'm probably wrong and bitboards were invented back then.. maybe way earlier.

(But it was still fun to try to build a program back then, when no other source code was readily available. It was a couple of years after I had a working program that I first saw any other chess program source code.. an early version of GNU if I remember).
Dann Corbit
Posts: 12777
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Francesca M.A.D v0.15 released

Post by Dann Corbit »

With an array representation of any kind, I would not expect any speedup going to 64 bits.

The only real advantage of such a compile is to allow for very large hash tables.
User avatar
silentshark
Posts: 327
Joined: Sat Mar 27, 2010 7:15 pm

Re: Francesca M.A.D v0.15 released

Post by silentshark »

Dann Corbit wrote:With an array representation of any kind, I would not expect any speedup going to 64 bits.

The only real advantage of such a compile is to allow for very large hash tables.
ok, thanks. Are there certain data types I should look to consolidate toward?
Dann Corbit
Posts: 12777
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Francesca M.A.D v0.15 released

Post by Dann Corbit »

silentshark wrote:
Dann Corbit wrote:With an array representation of any kind, I would not expect any speedup going to 64 bits.

The only real advantage of such a compile is to allow for very large hash tables.
ok, thanks. Are there certain data types I should look to consolidate toward?
A switch to bitboard would be pretty major surgery.
For bitboards, the fundamental underlying type is unsigned long long.