Search found 748 matches

by smatovic
Thu Jun 14, 2012 5:40 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Another Sliding Piece Move Generator -> BlockerTable Look
Replies: 2
Views: 784

Another Sliding Piece Move Generator -> BlockerTable Look

i am sure someone else tested this before....and my first implementation is slower than magic bitboards...anyway.. The Idea is to compute for every piece and all 8 directions the possible sliding-blockers and check this table against an precomputed AttackTable. Here some pseudocode: BlockerTable[9*6...
by smatovic
Tue Jun 12, 2012 2:35 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Is there such a thing as branchless move generation?
Replies: 55
Views: 17124

Re: Zeta CL uses Magic Bitboards

Maybe he could have taken the legality test out side the inner loop since it is better to do pseudo-legal move generation first.
Thanks,
i will test an pseudo-legal move generator in future..
by smatovic
Tue Jun 12, 2012 1:16 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Is there such a thing as branchless move generation?
Replies: 55
Views: 17124

Re: Zeta CL uses Magic Bitboards

Hey Daniel, I agree with reducing the memory usage, BUT I do not think your code is ugly at all. Thanks for this one. I guess the code of an GPU-Chess-Programm must be a bit arbitrary :) AFAIK you and I have been trying to use each core for doing real search. So unless you think it is impossible tor...
by smatovic
Tue Jun 12, 2012 1:05 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Is there such a thing as branchless move generation?
Replies: 55
Views: 17124

Re: Zeta CL uses Magic Bitboards vs Kogge-Stone

just for the record: i retested my old kogge-stone implementation, now with only one thread, looks like it is a bit faster: // every direction for(i=0;i<8;i++) { pro = ~bbBlockers; gen = SetMaskBB[pos]; r = shift[(piece>>1)*8+i]; pro &= avoidWrap[(piece>>1)*8+i]; // do kogge...
by smatovic
Mon Jun 11, 2012 9:35 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Is there such a thing as branchless move generation?
Replies: 55
Views: 17124

Re: Zeta CL uses Magic Bitboards

Hey Gerd, Your qbb legal move generation in zeta.cl with domove/undomove, king in check test, x-times getting piece codes from qbb in domove/undomove and elsewhere, and lots of branches in inner loops, looks horrorful inefficient and ugly to me Wink Yep, i admit it is ugly, copy and paste from diffe...
by smatovic
Mon Jun 11, 2012 2:53 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Is there such a thing as branchless move generation?
Replies: 55
Views: 17124

Re: Zeta CL uses Magic Bitboards

Hi John, maybe you want to take a look at "Zeta CL" http://zeta-chess.blogspot.de/ Code is published under GPL: https://github.com/smatovic/Zeta/ Zeta uses QuadBitboards (thx to Gerd) for Board Presentation and a Magic-Bitboard Move Generator (parts ported from Stockfish). I tried an 0x88 move gener...
by smatovic
Mon Jun 04, 2012 5:08 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: YBWC: Active Reparenting
Replies: 63
Views: 16765

Re: GPUs

Did i hear the word GPU?

IMO there is no way to implement YBWC or DTS on current GPUs in an efficient matter.

--
Srdja

PS: ...i have read something about an upcoming recursion feature....maybe...
by smatovic
Wed Apr 04, 2012 6:54 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: New engine: Rhetoric
Replies: 13
Views: 3537

Re: New engine: Rhetoric

Congratulations and Welcome to the Jungle.

--
Srdja