Search found 99 matches

by Tony P.
Tue Jan 28, 2020 7:42 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: hash collisions
Replies: 370
Views: 28478

Re: hash collisions

I think lockless hashing was pretty safe at modest number of threads most people have removed the need for legality checking of the hash move. Now it may be necessary to bring that back in. So it looks like the SF devs were right when they decided not to store the full hash key in the TT (currently...
by Tony P.
Tue Jan 28, 2020 5:39 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: multi-dimensional piece/square tables
Replies: 11
Views: 2773

Re: multi-dimensional piece/square tables

The Chinese chess paper (section V.B) reports that, in a match between trained EVAL0 versions, the one with tapered eval (which, in that case, interpolated between the opening and the endgame) beat the one without convincingly, with a 62.75% winrate, but of course, the tapering coefficient doesn't h...
by Tony P.
Tue Jan 28, 2020 6:30 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: hash collisions
Replies: 370
Views: 28478

Re: hash collisions

CPW suggests some remedies for data races... I have no idea how much overhead they add in practice, though.

As for collisions, maybe store >1 move per hash entry?
by Tony P.
Tue Jan 28, 2020 4:57 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: multi-dimensional piece/square tables
Replies: 11
Views: 2773

Re: multi-dimensional piece/square tables

Also, while we're at it, let me necro HGM's rotated nibbleboard idea from the WB forum as well :wink: I'm still not sure if it's best to store 2, 3 or 4 bits of info per square. The problem with 4 bits/sq (hence 32 bits per ray) is that the LUT would be too large to fit into RAM, but unlike in 2006,...
by Tony P.
Tue Jan 28, 2020 4:32 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: multi-dimensional piece/square tables
Replies: 11
Views: 2773

Re: multi-dimensional piece/square tables

Sorry for necromancy - I want to credit Pawel for considering PSTs involving more than 1 piece location. I've stumbled upon a 2018 preprint 'Comparison Training for Computer Chinese Chess' (the final version of the article was published in IEEE Transactions on Games in January 2019) by Wen-Jie Tseng...
by Tony P.
Sun Jan 26, 2020 5:38 pm
Forum: Computer Chess Club: General Topics
Topic: Another tactical test position
Replies: 7
Views: 982

Re: Another tactical test position

It's worth noting that Alpha sub0 had 4. f4 in the PV, a rather counterintuitive move giving up a piece desperately, like SF11 did at depths 22-29 :wink:
by Tony P.
Fri Jan 24, 2020 9:18 pm
Forum: Computer Chess Club: General Topics
Topic: Another tactical test position
Replies: 7
Views: 982

Re: Another tactical test position

Oh, that's awesome, thanks! I just didn't have the resources and patience to run a longer search (in terms of nodes) with a preciser net.
by Tony P.
Fri Jan 24, 2020 9:04 pm
Forum: Computer Chess Club: General Topics
Topic: Another tactical test position
Replies: 7
Views: 982

Re: Another tactical test position

Oops, my bad, the correct FEN for the last one is:

7r/p4q1k/1r1P1b1n/2NQp1pn/2P1Pp2/R1B2P1b/PP2B2P/3R3K b - - 1 1

So it's Black to move, not White.
by Tony P.
Fri Jan 24, 2020 8:25 pm
Forum: Computer Chess Club: General Topics
Topic: Another tactical test position
Replies: 7
Views: 982

Re: Another tactical test position

I've modified the position after Rd1 by adding a white pawn on b2 to ensure that Black's inferior moves don't win. 7r/p4q1k/1r1P1b1n/2NQp1pn/2P1Pp2/R1B2P1b/PP2B2P/3R3K w - - 1 1 Now Ng3+ is the only winning move; Kg6 and a6 likely draw, the other moves probably lose. Black needs to find a few only m...
by Tony P.
Fri Jan 24, 2020 6:57 am
Forum: Computer Chess Club: General Topics
Topic: Another tactical test position
Replies: 7
Views: 982

Re: Another tactical test position

Actually, the whole game was a KID mess :lol: [Event "Computer chess game"] [White "Stockfish_9_x64"] [Black "Stockfish_20011801_x64"] [Result "0-1"] [ECO "E99"] [Opening "King's Indian"] [Variation "Mar del Plata, Benko Attack, 11...Nf6 12.Nd3"] [TimeControl "180+2"] [Termination "normal"] [PlyCoun...