Search found 64 matches

by ymatioun
Sun Sep 14, 2014 12:50 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: More details from test suites
Replies: 8
Views: 1440

Re: More details from test suites

I use "strategic test suite". It is big (1,000 positions), and moves are scored according to their value. I usually run it at 0.5 second/position, for 500 seconds total run time. My results on that test match reasonably close to playing strength (the match is not perfect, so it does not work for sma...
by ymatioun
Wed Sep 03, 2014 12:49 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Implementing parallel search
Replies: 22
Views: 4382

Re: Implementing parallel search

On my 4-core computer this algorithm gets average CPU utilization of around 75%, so on average 3 out of 4 cores are working.
This is not perfect, but good enough for 4 cores. But this would not scale well beyond 4 cores; if i get a bigger computer, i'll have to do something more complicated.

Youri.
by ymatioun
Tue Sep 02, 2014 9:52 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Implementing parallel search
Replies: 22
Views: 4382

Re: Implementing parallel search

I only allow splits by master thread. So when a slave thread has no more work, but some slaves are still working, CPU utilization drops, from 4 running threads down to 3, 2, then 1. This is certainly suboptimal, but very easy to implement.

Youri.
by ymatioun
Tue Sep 02, 2014 11:52 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Implementing parallel search
Replies: 22
Views: 4382

Re: Implementing parallel search

I also found this confusing when i was implementing parallel search. So the way i did it was several slave threads (4 for 4-core CPU), waiting on an event. When master thread reaches split point, it puts position data into global variable, triggers slave event and goes to sleep. Slaves pick-up the s...
by ymatioun
Mon May 05, 2014 10:48 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: investigating why stockfish is strong idea
Replies: 31
Views: 9051

Re: investigating why stockfish is strong idea

I did some testing on SF4, in a way similar to what you described: i started with full SF4, and then disabled one search feature at a time (cumulatively), running it against my program and recording incremental impact of each feature. Here are the results: LMR=100 Sing. Ext=38 prob cut=28 step 7- fu...
by ymatioun
Tue Apr 08, 2014 11:50 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: problem with syzygy tablebases
Replies: 8
Views: 2631

Re: problem with syzygy tablebases

OK, thanks. I'll keep digging.
by ymatioun
Tue Apr 08, 2014 2:21 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: problem with syzygy tablebases
Replies: 8
Views: 2631

Re: problem with syzygy tablebases

So, with Roland's help i set-up 5-piece tables, and they all work. I just tried using 6-piece tables, and i am having problems again. When i start with position 4k3/8/4b3/5p2/8/3P4/1B1K4/8 w - - 0 0, it crashed on call to probe_wdl() function. Call history is probe_wdl() calls probe_ab(-2,2,success)...
by ymatioun
Sat Mar 08, 2014 2:01 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: SEE logic
Replies: 35
Views: 7267

SEE logic

From all the posts here i got the impression that everybody uses just captured piece values in SEE. But i set-up my SEE to also include piece-square values of all the moves made. It is definitely slower, but i think it adds a lot of value. But i never tested impact of this on engine playing strength...
by ymatioun
Sun Feb 02, 2014 3:52 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: problem with syzygy tablebases
Replies: 8
Views: 2631

Re: problem with syzygy tablebases

I got it working, and everything looks fine. Thank you for your help.

Youri.