Search found 686 matches

by smatovic
Wed Dec 21, 2011 4:18 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Cluster versions of chess programs.
Replies: 11
Views: 1845

Re: Minimax cluster

Now consider a 10,000 processor array running plain minimax. It will search as deeply as the A/B system in about the same time. If the mean branching factor in the minimax search is about 32, then increasing the minimax searcher processor count by a factor of 32 will make the system see an extra pl...
by smatovic
Mon Dec 19, 2011 11:53 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Introducing my chess engine: GM of Logic! (w/Gui)
Replies: 11
Views: 1818

Re: Introducing my chess engine: GM of Logic! (w/Gui)

Congratulations,
looks fast and i like the web-interface.

--
Srdja
by smatovic
Tue Dec 13, 2011 2:38 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Zeta OpenCL Chess - v 093x - developer release -
Replies: 0
Views: 323

Zeta OpenCL Chess - v 093x - developer release -

Heyho, i published the source of Zeta 0933 and 0934. https://github.com/smatovic/Zeta/tree/zeta_nvidia_0933 https://github.com/smatovic/Zeta/tree/zeta_nvidia_0934 I managed to get a a negamax without ab-pruning running on one SIMD Unit with 128 threads and based on a LIFO-Stack: Zeta 0932, search de...
by smatovic
Sun Dec 04, 2011 4:28 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: LIFO stack based parallel processing?
Replies: 12
Views: 2926

Re: LIFO stack based parallel processing?

But still don't have an idea how to handle AlphaBeta values, any suggestions? My quick n dirty solution is just to let the score-tree grow in memory.... Here is the LIFO scheme i use in pseudo code i had to realize that without a linked list i have to visit every generated position to do the same b...
by smatovic
Thu Dec 01, 2011 6:24 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: LIFO stack based parallel processing?
Replies: 12
Views: 2926

Re: LIFO stack based parallel processing?

...try and fail. It is nearly impossible to implement a parallel LIFO stack on a GPU considering that threads inside a SIMD Unit are again divided into Warps/Wavefronts. ..fail and error. fixed an bug and now i am able to distribute work by use of a LIFO Stack across threads within an SIMD Unit of ...
by smatovic
Tue Nov 15, 2011 10:23 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Want to know what savings to expect with pruning methods...
Replies: 23
Views: 3387

Re: Want to know what savings to expect with pruning methods

What do you guys recommend I should do next now? 1) check perft results http://chessprogramming.wikispaces.com/Perft+Results I had some problems with CastleMoves/Rights and En Passant Moves. 2) Setup some test positions http://chessprogramming.wikispaces.com/Test-Positions For a quick look how good...
by smatovic
Sun Oct 16, 2011 7:28 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Vintage Chess Programming
Replies: 24
Views: 5706

Re: Vintage Chess Programming

That is very slow indeed. I have never really measured it for Usurpator. but even on a 1MHz 6502 it must have been around 500 NPS (with virtually no evaluation). Most 6502 instructions executed in 2 clocks. The inner loop of the move generator (to step a slider over the board) did only contain a co...
by smatovic
Sat Oct 15, 2011 12:23 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Vintage Chess Programming
Replies: 24
Views: 5706

Re: Vintage Chess Programming

I did write a chess program for 6502 in those days (Usurpator II; it still competes in ChessWar through an emulator). How many nodes per second does Ursurpator compute? My current Version with an Mailbox Move Generator is quite slow, 10 NPS. Search depth 3 needs about 3 Minutes for the starting pos...
by smatovic
Fri Oct 14, 2011 1:48 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: chess software for dedicated chess computers
Replies: 7
Views: 998

Re: chess software for dedicated chess computers

The Citrine is a good example of a dedicated computer - but it's reliability and quality is a problem. It's just a Renesas H8/3687 chip which has 56k of program memory and 3k of RAM with CPU speed 20 MHz. Code for this type of chess computer must be available somewhere. I also suggest to use some x...
by smatovic
Sun Oct 09, 2011 2:01 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Vintage Chess Programming
Replies: 24
Views: 5706

Re: Vintage Chess Programming

The authors of the Northwestern Chess 4.x program noted that there was a benefit of having a transposition table with only 256 entries. However, that was with a program which had a speed of only about 400 nodes per second. Wonder what kind of TT they used. Is the computation of 32 bit Zobrist Hashe...