TalkChess.com

Generally speaking, it is harder to understand and debug code that might affect global variables in addition to whatever arguments are passed to the function (aka "side effects"). It shouldn't really affect speed though.

My code for this is complex, but the core of it sums attack "weights" for each piece attacking squares near the king. The weights depend on piece type and take into account "stacked" attackers like a Rook behind a Queen. The sum is then put through a sigmoid function. There is more to it than that -...

I am aware of the random move options. Actually as I have posted before, the way gensfen does this seems to me unsound, because it will score the whole game with the ultimate result, even though a random move inserted in the middle can change the eval significantly, and so really then the previous p...

Thanks for the responses. It just does not seem intuitive to me that the objective would depend on input size. That's not what I see with Texel tuning, for example. But maybe it is overfitted and/or the input should be more diverse. The gensfen program by default outputs all game positions past the ...

The EU Club Cup qualifers are happening, and the first round has some super-GMs paired against weaker players. This is what happens when you have to play Shirov. He offered a Queen sac on move 11, which was declined, but taking it loses, too. [Event "European Online Club-ch-A 2021"] [Site "Online"] ...

I am doing some experiments with https://github.com/bmdanielsson/nnue-trainer and have a couple of questions: 1. What's a typical number of positions for training and validation? 2. I am seeing that the objective values seem to go down with increasing training set size. For example, 5 million positi...

That book caused something of a stir when it first came out. It was basically advocating a kind of "Shannon Type B" strategy IIRC (https://www.chessprogramming.org/Type_B_Strategy). Botvinnik was involved in the development of a chess program called Pioneer that incorporated his ideas, but it was ne...

These are a few I have collected the past few months (besides the one I just posted with bm Bf7). I think Stockfish solves most if not all of these. But other engines may have trouble. r2qrbk1/5b1p/p1p2pp1/1pp5/P2nPNPP/2BPN3/1PP2R2/R5QK w - - bm axb5; c0 "SugarAI-Arasan, FICS 2021"; c1 "Arasan expec...

from an online game:



Bf7 wins, other moves only draw. Stockfish is fast on this. Other engines I have tried either don't select Bf7 or don't see it wins, at least with moderate depths.

Several commercial projects use Arasan as a chess engine. It is stronger than Crafty (didn't used to be, but is now). It is MIT licensed. I am pretty sure I don't even know how many software packages might use or include it, because they don't need to notify me.