boarquantile

Load testing would have to be quite advanced to find this, doing more than just simulating users for the most common features. It took more than 24 hours of real-world usage to visit sufficiently many niche code paths to fill the code cache.

And from the same section, importantly:

> [...] reinstated (a) provisionally, unless and until the copyright holder explicitly and finally terminates your license

which is what the Stockfish developers have now done. I.e. if Chessbase starts to comply now, it would be too late.

ITT: People missing the point by making it about software licenses, when the real issue is lying about the product.

ChessBase could easily comply with any open source license by linking the full source code, including the method they used to generate NN weights (if they think the license applies here as well), in a footnote.

It would still be a scam, because their claims of superiority, originality ("almost from scratch"), and innovation (see the funny part about technology from Japan) are wrong, and purposely so.

The most boring chess game according to Lichess move compression: https://lichess.org/study/5jQeJXXb

Lichess tries to use (in this order): PNaCl, WASM, asm.js.

PNaCl required almost no modifications to Stockfish (https://github.com/niklasf/stockfish.pexe), supports multithreading and is still the fastest.

WASM and asm.js require the same patches (https://github.com/niklasf/stockfish.js). Most notably WASM does not yet have multi-threading. In the original Stockfish one dedicated thread is listening for new commands. Instead now the main search thread (since it's the only thread) has to check for new commands from time to time, which costs a bit of performance.