In https://lichess.org/d77I5RgC/, at the end of move 17, the local analysis and the "request computer analysis" analysis have very different evaluations - with the "request computer" showing an even game after a blunder, and the local analysis showing black holding the win.
What's up with this? Are they expected to differ so much, and if so, is one usually more correct?
Ah, OK, after a refresh the local analysis changed, the remote was correct....
This topic has been archived and can no longer be replied to.