Search found 147 matches

by sedicla
Thu Sep 22, 2011 2:04 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Internal Iterative Deepening questions
Replies: 12
Views: 3799

Re: Internal Iterative Deepening questions

The first test didn't show any improvments. I think I'll keep the way it is. I guess if you are on depth = 10 and call IID with depth=3 you have a hash_move of depth = 3. But if you call with depth - 3, then you eventually have a hash_move of depth=7 which probably is a better move than of depth=3. ...
by sedicla
Wed Sep 21, 2011 2:06 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Internal Iterative Deepening questions
Replies: 12
Views: 3799

Internal Iterative Deepening questions

Usually IID is implemented like this: if (hash_move == NONE && depth > 3 ...) search (alpha, beta, depth - 3, ... Has anyone tried searching with depth = 3 immediately, like: if (hash_move == NONE && depth > 3 ...) search (alpha, beta, 3 , ... I'm testing it now with 1000 games, 10sec+0.1sec. Doesn'...
by sedicla
Fri Jun 03, 2011 6:59 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Introduction and (hopefully) contribution - bitboard methods
Replies: 10
Views: 2664

Re: Introduction and (hopefully) contribution - bitboard met

I've been lurking here for awhile and first of all I'd like to say thank you to all for posting all this good stuff, I've learned a lot here and now i'd like to maybe contribute to the chess programming community. I am involved professionally with computer programming for 25 years. And some of my h...
by sedicla
Fri Jun 03, 2011 5:01 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Introduction and (hopefully) contribution - bitboard methods
Replies: 10
Views: 2664

Re: Introduction and (hopefully) contribution - bitboard met

Here is the code I used for tests: I'd suggest to test with random numbers, not with a linear loop that increments the target bitboard. Otherwise for the CPU becomes easy to predict the branches inside first_index() and friends and the results are better than what you would achieve in a real game. ...
by sedicla
Fri Jun 03, 2011 3:14 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Introduction and (hopefully) contribution - bitboard methods
Replies: 10
Views: 2664

Re: Introduction and (hopefully) contribution - bitboard met

Here is the code I used for tests: #define BBTSTSIZE 90000000 void bb_test_index_ok(void); void bb_test_fi_perf(void); void bb_test_li_perf(void); void bb_test_count_ok(void); void bb_test_bc_perft(void); void bb_test() { bb_test_index_ok(); bb_te...
by sedicla
Fri Jun 03, 2011 3:07 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Introduction and (hopefully) contribution - bitboard methods
Replies: 10
Views: 2664

Introduction and (hopefully) contribution - bitboard methods

I've been lurking here for awhile and first of all I'd like to say thank you to all for posting all this good stuff, I've learned a lot here and now i'd like to maybe contribute to the chess programming community. I am involved professionally with computer programming for 25 years. And some of my ho...