TalkChess.com

Hi, Fulvio. Thanks for your work on Scid. One thing that first it was not possible to do was to save/retrieve hash memory to/from disk. I am pretty sure that many users would really appreciate such a feature.
Thanks again.

Stockfish contains the nice 'eval' command that prints out the detailed score of its static evaluation function. It regards Mobility, Material, Bishop, Rooks, etc. I was wondering if there was any possibility to get even more detailed scores, like, for example the one provided by the Bishop xray att...

@dkappe Impressive, many thanks. One thing I was curious about is the following. Is there any resource available that I could use If I would like to replicate the generation of the Bad Gyal neural net? I mean a resource that would tell me: these are the games and these are the scripts you need to us...

It was my understanding of Oxyron thinking that playing a, say, 10 games match /10 plies long, would allow to deepen the search and back propagate the final score to the root, ultimately leading to beta cutoffs in the hash table. Then, iterate the process with the improved HT, until the original sco...

Well, I was hoping that following the Oxyron idea to play a short match among the engines after after an infinite analysis would have worked by pushing the search beyond a certain horizon. That was a way to automate things. Another attempt might be to combine moves from a multi-PV and see if you get...

Usually, hash tables contain information that is continuously written upon over and over during a game. However, when analyzing a single position it is possible to prime a meaningful storage of the information by moving pieces back and the forth from the starting position and confuting 'by hand' all...

I am letting stockfish to play a self match starting from a given position. Is there a way to avoid that Stockfish clears the hash every time it starts a new game? Alternatively, Sugar program used to do that, but I have a rather old version and the new one is not for linux yet. https://github.com/Z...

Sorry if this question sounds too naive, but I was wondering if there was a tutorial on how starting to experiment with network building for lc0. Ideally, it should be a tutorial starting with simple networks that also people without GPU can efficiently use: http://talkchess.com/forum3/viewtopic.php...

Could someone please post why in this position Nc3 is stronger than h5 and which is the plan that white needs to follow? Dann said that his computer solved at depth 46, but lastest stockfish couldn't on my computer well beyond depth 60. 2bqrr1k/p5b1/1p1p2pp/nPpBp3/P1P1N2P/3PN1P1/R4P1K/3Q1R2 w - - b...

Could someone please post why in this position Nc3 is stronger than h5 and which is the plan that white needs to follow? Dann said that his computer solved at depth 46, but lastest stockfish couldn't on my computer well beyond depth 60. 2bqrr1k/p5b1/1p1p2pp/nPpBp3/P1P1N2P/3PN1P1/R4P1K/3Q1R2 w - - b...