Search found 30 matches

by niel5946
Sun Feb 28, 2021 9:02 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: stockfish freezes chessbase
Replies: 1
Views: 129

Re: stockfish freezes chessbase

I don't know about any bugs concerning freezing in stockfish, and I myself have used stockfish in other GUIs without any problems with freezing. I would suspect the culprit to be in either your computer (perhaps not being able to run stockfish with too large hash table or too many cores) or the ches...
by niel5946
Sun Feb 28, 2021 7:44 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Transposition table causes illegal moves
Replies: 0
Views: 36

Transposition table causes illegal moves

Hi. I recently made a post concerning illegal moves as a result of faulty updating of the PV-stack (http://talkchess.com/forum3/viewtopic.php?f=7&t=76708 , the code referenced in the links are not the corrected version since I haven't pushed the changes yet), but after correcting this, I have isolat...
by niel5946
Thu Feb 25, 2021 4:29 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Illegal moves in PV stack
Replies: 6
Views: 202

Re: Illegal moves in PV stack

@Niels: 1) I suggest not to change the PV if score <= alpha or score >= beta. In that case you will restart the root search with a different aspiration window anyway (or you run out of time so you play the best move from previous iteration) so you do not need a PV (and it would also be misleading)....
by niel5946
Thu Feb 25, 2021 4:14 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Illegal moves in PV stack
Replies: 6
Views: 202

Re: Illegal moves in PV stack

Since you appear to take the move to play from the PV, how do you get that PV? Normally the root search would construct a new PV from a move that raised alpha, followed by the PV returned by the child node. And normally it would only search move for the player on move. So how could the first move o...
by niel5946
Thu Feb 25, 2021 2:58 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Illegal moves in PV stack
Replies: 6
Views: 202

Illegal moves in PV stack

Hi. I have noticed my engine making illegal moves over the past few days and I can't seem to figure out why. When I play a 200 game tournament - I have tested with time controls of 4s+0.1s, 12s+0.12s and 1min - between the current and the old version, at least two of those games are lost (by the cur...
by niel5946
Tue Feb 23, 2021 4:50 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: No TT hits with optimizations
Replies: 11
Views: 308

Re: No TT hits with optimizations

I see now that your problem is actually a non-determinism between debug and release build - might be some uninitialized variables? I've also noticed that you use masking when probing TT, but I don't see where you round numEntries to a power of two; while it still works for your default TT size, it ...
by niel5946
Tue Feb 23, 2021 12:05 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: No TT hits with optimizations
Replies: 11
Views: 308

Re: No TT hits with optimizations

Yes, I am using the C++ <random> library and the following preprocessor macro: #define RAND_64 ((uint64_t)std::rand() | \ (uint64_t)std::rand() << 15 | \ (uint64_t)std::rand() << 30 | \ (uint64_t)std::rand() << 45 | \ ((uint64_t)std::rand() & 0xf) << 60 ) What do you mean with unsafe optimizations?...
by niel5946
Tue Feb 23, 2021 11:29 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: No TT hits with optimizations
Replies: 11
Views: 308

Re: No TT hits with optimizations

Are you using the a library function for generating pseudo-random numbers? Different platforms use different PRNG, and if you use them for generating Zobrist keys this will alter the mapping from positions to TT entries, cause different replacements, and eventually a different search tree. This sho...
by niel5946
Tue Feb 23, 2021 11:23 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Endgame recognition
Replies: 6
Views: 413

Re: Endgame recognition

No, any sort of null move would make you fail to see zugzwangs. Because zugzwangs by definition are positions where passing your turn would be good, but doing any legal move is fatal. So the null move (turn pass) will return a score that is not representative for the position, but a gross overestim...
by niel5946
Tue Feb 23, 2021 9:36 am
Forum: Computer Chess Club: General Topics
Topic: New engine releases & news 2021
Replies: 105
Views: 19354

Re: New engine releases & news 2021

I have just released my new engine, Loki v1.0.2. It has support for 1 to 8 threads (Lazy SMP) and 1MB to 1000MB hash table and is probably rated around 1600-1800. It has only been tested on windows and has not shown any bugs on both 32-bit(x86) and 64-bit(x64). Download: https://github.com/BimmerBa...