dubslow

> I think if the server sets the info field then the worker will do the right thing (that's the intention anyway). It does not do anything except print a...

Duplicating my discord comment: > If the message is formed by the worker, it can figure exactly the reason, can display language-specific messages, and has various options on how to...

I've pushed a new version which moves all message interpretation from the worker to the server. Now the worker looks almost like master, only now it prints whatever string the...

> This https://github.com/official-stockfish/Stockfish/commit/9eccba776198c210563666787a9b5fe44b386fdc commit and the mentioned bugs may be of interest. Fascinating! If indeed this PR turns out to not be mergeable, that means then that the code is...

I agree on the following two points: 1) making SF a library will have enormous benefit to the project, for foreseeable applications (android or other mobile apps), and for unforeseeable...

yea 10s 1 thread much better choice

> As I said: Unit tests are also for testing regressions in case of refactorings. Fishtest tests for both nonregression of elo and gain of elo. In fact I'm tempted...

> Also, it's pretty easy to notice that this must be a bug/laziness/some oversight turned into a feature.. This is clearly false, as niklasf repeatedly stated this is specifically and...

What pb is saying, jmu, is that removing the abs(beta) < blah condition is fine, but removing the nmpPly stuff is less fine. It is interesting that removing verification entirely...

> It might be that nowadays doing enhanced zugzwang detection comes with a certain cost (~0.5 - 1 Elo) It does not, as I demonstrated the other week. If it...