https://bugs.winehq.org/show_bug.cgi?id=31393
--- Comment #11 from Ruslan Kabatsayev b7.10110111@gmail.com --- (In reply to Henri Verbeet from comment #10)
How about this one? It tries harder to avoid invalid sample counts.
This works better. It now seems to even find the exact counts I've specified in the registry (at least it reports that number in the "Forcing sample count to", and the difference is visible between 4 and 5 samples choices) even in the cases which previously failed.