Search found 64 matches

by ymatioun
Thu Aug 20, 2015 12:02 am
Forum: Computer Chess Club: General Topics
Topic: Fizbo 1.5 released
Replies: 18
Views: 4415

Re: Fizbo 1.5 released

Clarification: Fizbo uses EGTBs two ways: W/L/D tables are used inside search. So they get accessed as soon as number on pieces on board drops to 5. This part requires tables to be on SSD, because they are accessed a lot. Second way is using depth-to-zero tables when there are 5 or fewer piece on th...
by ymatioun
Wed Aug 19, 2015 11:48 pm
Forum: Computer Chess Club: General Topics
Topic: Fizbo 1.5 released
Replies: 18
Views: 4415

Re: Fizbo 1.5 released

I just ran a few games with the latest file (to make sure i did not mess up compile) and i don't see any memory increase. Normally i run 1,000 games in testing, and that does not lead no any memory increase. Slow search with EGTB turned on: i don't see it either. I just ran search on a simple positi...
by ymatioun
Wed Aug 19, 2015 10:12 pm
Forum: Computer Chess Club: General Topics
Topic: Fizbo 1.5 released
Replies: 18
Views: 4415

Re: Fizbo 1.5 released

OK, i will remove "Nalimov" option right away - updated files will be on my site within an hour.

Youri.
by ymatioun
Wed Jun 24, 2015 11:46 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: There are compilers and there are compilers
Replies: 31
Views: 5211

Re: There are compilers and there are compilers

New Intel Xeon CPUs include something called "Transaction Synchronization Extensions". I don't know exactly how they work, but those instructions were specifically designed to reduce synch overhead. Perhaps ICC emits those instructions, while GCC most certainly does not? That would explain why impro...
by ymatioun
Mon Jun 15, 2015 10:43 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: thoughts on null-move reduction
Replies: 6
Views: 2507

Re: thoughts on null-move reduction

Thank you for all your comments. What I am trying to get out of this is to make Null-Move reduction more effective. I think this can be done, if I find a way to vary its intensity. Example: LMR. We can vary intensity of it by reducing more or by reducing less. Optimal result is achieved when margina...
by ymatioun
Sun Jun 14, 2015 12:41 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: thoughts on null-move reduction
Replies: 6
Views: 2507

thoughts on null-move reduction

I am trying to understand why null-move reduction works so well. And possibly find a way to make it even better. Here are some thoughts and questions i have on this topic. Null-move reduction works by searching a node with reduced depth and a margin. In this case margin is loss of turn. And reductio...
by ymatioun
Fri May 29, 2015 12:39 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Ostrich tactics and pointless checks
Replies: 8
Views: 2591

Re: Ostrich tactics and pointless checks

I also think this is driven by eval: terms like king safety and passed pawns can be used to predict checkmate and promotions way before they can be seen by search. If your eval has those terms and your opponent's eval has not, he will let you get there, and you will see your score steadily increasin...
by ymatioun
Sat Mar 21, 2015 5:57 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: On-line engine blitz tourney March
Replies: 38
Views: 7340

Re: On-line engine blitz tourney March

Looks like now US is on summer time but Amsterdam is still on winter time. S o the tournament is 9pm Amsterdam 4pm (not 3pm) Boston. That is, 2 hours form now. Correct?
by ymatioun
Tue Dec 09, 2014 11:49 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Scorpio EGBB bug?
Replies: 8
Views: 1342

Re: Scorpio EGBB bug?

Is this because black pawn is on rank 7 and could be moved 2 squares, leading to possible EP move by a white pawn?

Youri.
by ymatioun
Tue Dec 09, 2014 12:40 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Scorpio EGBB bug?
Replies: 8
Views: 1342

Scorpio EGBB bug?

I have been using 5-piece Scorpio EGBB, and it works fine. But i just came across the following position:FEN=8/4kp2/8/5KP1/4P3/8/8/8 b - - 0 60. Scorpio (called through "probe_egbb_fen") calls it a draw. But this is an obvious loss for black - as confirmed by deep search by Stockfish and online vers...