dav1312

> my stockfish version is `stockfish_15_x64_avx2` , download this version & do test to see your problem solved or not ... Again... stop using an older/different version and expect the...

> maybe exist a bug in Nnue Evaluation No, this happens in [Stockfish Classical](https://github.com/official-stockfish/Stockfish/releases/tag/SF_classical) too ``` Stockfish 291021 64 POPCNT by T. Romstad, M. Costalba, J. Kiiski, G. Linscott position...

> I think that this is the commit that causes the issue. [95d7369](https://github.com/official-stockfish/Stockfish/commit/95d7369e54f20715345cf5408040f3c7d1ec8415) This doesn't seem to be a new issue created by a recent commit as it can be...

> there are no terms that obviously favor repetition. By "on purpose" I don't really mean "favor" as in "preferring" I mean more like "knowing and not caring". If Stockfish...

> Lets assume that w1 and b1 are played and now SF searches on this positions, gets a deeper search and finds a new move w2' that is better than...

> Evaluating 2-fold repetition as a draw even in root (if expected score is > 0) should do what you want. No idea if this was already tested at some...

Hmm yeah I didn't consider MultiPV... but I guess this behaviour could've been disabled only when MultiPV was more than 1? Why not? That way you "hide" the trick when...

> What Elo does, for example, _Skill Level 0_ represent? If not 1000, what then? ``` 0 1347 1 1490 2 1597 3 1694 4 1785 5 1871 6 1954...

The evaluations have to change your winning chances by at least 10% to get an inaccuracy. This does mean that theoretically, a player can lose a game while not getting...

This seems similar to the discussion in #9119