Search found 88 matches

by Tony P.
Wed Sep 20, 2017 1:34 am
Forum: Computer Chess Club: General Topics
Topic: Fizbo 1.9 released
Replies: 57
Views: 15400

Re: Fizbo 1.9 released

Judging by the fact that the Qxe4 'PVs' like 21/45- are printed almost instantly after the previous search iteration ends, I'm afraid you're right. Another clue is that Fizbo's TT has ' dense 8 byte entries' - it looks like the TT has no buckets (that Stockfish has) and, indeed, there's no insurance...
by Tony P.
Wed Sep 20, 2017 12:29 am
Forum: Computer Chess Club: General Topics
Topic: Fizbo 1.9 released
Replies: 57
Views: 15400

Re: Fizbo 1.9 released

Perhaps Fizbo's error occurs because the key length for one of the auxiliary hashes (most likely the material hash) is too small even when the TT is large.
by Tony P.
Wed Sep 20, 2017 12:08 am
Forum: Computer Chess Club: General Topics
Topic: Fizbo 1.9 released
Replies: 57
Views: 15400

Re: Fizbo 1.9 released

If these are indeed hash errors (with the hash size set to 2 GB), then it's surprising that Fizbo is the 6th strongest engine in the world - I've seen such erratic PVs (always involving obviously incorrect weird sacrifices) many times in fewer than 100 Fizbo games played on my laptop, primarily in e...
by Tony P.
Tue Sep 19, 2017 11:48 pm
Forum: Computer Chess Club: General Topics
Topic: Fizbo 1.9 released
Replies: 57
Views: 15400

Re: Fizbo 1.9 released

Here are a couple of quirks in the way Fizbo 1.9 writes PVs that are disturbing me. 1. In its comments to a game vs Andscacs 0.92 , specifically, in the PVs for its moves 45 and 46, Fizbo wrote illegal moves (45. d7 c2 46. d8Q cxd1Q+ h5+ Qf4+ ), and as far as I remember, it was writing those moves i...
by Tony P.
Tue Sep 19, 2017 7:14 pm
Forum: Computer Chess Club: General Topics
Topic: Where's the joy in writing a chess engine?
Replies: 44
Views: 10704

Re: Where's the joy in writing a chess engine?

4) While implementing it correctly is indeed very important, a common misconception is that it has to be 'fast' to be any good. Certainly at some point speed optimisations will become the low hanging fruit but there's a lot of more important work before you get to that point. Yes, exactly! I don't ...
by Tony P.
Tue Sep 19, 2017 7:06 pm
Forum: Computer Chess Club: General Topics
Topic: Where's the joy in writing a chess engine?
Replies: 44
Views: 10704

Re: Where's the joy in writing a chess engine?

Thanks for the explanations, it's good to have you both as counterexamples ITT! I wonder why those who enjoy their job, and would like to do it as a hobby too, don't simply take extra working hours. I can think of a couple of reasons for this. Firstly, there are FOSS proponents who feel that their c...
by Tony P.
Tue Sep 19, 2017 6:29 pm
Forum: Computer Chess Club: General Topics
Topic: Where's the joy in writing a chess engine?
Replies: 44
Views: 10704

Where's the joy in writing a chess engine?

I have a conjecture that I'd like to check with the forum residents. Namely, writing an engine can only be enjoyed by someone who already does software development as a job. The thing is that the actual development is too stressful for an uninitiated person. Algorithm design in terms of pseudocode i...
by Tony P.
Tue Sep 19, 2017 5:20 pm
Forum: Computer Chess Club: General Topics
Topic: Opening analyses: SF / Komodo to 88% similar, not nice!!
Replies: 24
Views: 5035

Re: Opening analyses: SF / Komodo to 88% similar, not nice!!

right, reason is both top engines are fully clueless about the opening. for example, taking the Dutch. after 1.d4 f5, there is a single move that would give white advantage and winning chances, 2.d5! [d]rnbqkbnr/ppppp1pp/8/3P1p2/8/8/PPP1PPPP/RNBQKBNR b KQkq - 0 2 unfortunately, SF and Komodo are fu...
by Tony P.
Tue Sep 19, 2017 4:27 pm
Forum: Computer Chess Club: General Topics
Topic: Stockfish Chess 8 - en passant rule
Replies: 3
Views: 771

Re: Stockfish Chess 8 - en passant rule

There is code for e.p. move making , though (see the below code snippet from bitboards.cpp, lines 311-331), and the e.p. square is saved in the game state variable 'st', but for some reason, the code isn't working correctly yet. The zobrist.zEnPassant function is definitely not working yet. if (...
by Tony P.
Tue Sep 19, 2017 4:00 pm
Forum: Computer Chess Club: General Topics
Topic: Stockfish Chess 8 - en passant rule
Replies: 3
Views: 771

Re: Stockfish Chess 8 - en passant rule

I'm not a specialist in Stuckfish's source code , but a quick look at it reveals that its zobristh.cpp (the Zobrist hash implementation that's apparently different from Stockfish's) contains the following comment (line 146): //EnPassant and castling stuff add later Besides, the author (MaxCarlson) s...