If this is a Mesa regression, why not report this to Mesa instead?
If this is a Mesa regression, why not report this to Mesa instead?
Well, I listed those reasons in the first paragraph of the description. They might not be valid, I don't know.
Anyway, from your response I'm assuming this doesn't ring any bells and that you think it's a Mesa issue?
I've created a bug for Mesa https://gitlab.freedesktop.org/mesa/mesa/issues/2458 so feel free to close if this is not a DXVK bug.
Well the problem is that with a screenshot alone I can't do all that much, I'd need something to reproduce the issue with and then bisect Mesa in order to figure out what's broken. It's fine to report things like this here, but then I need something to work with (i.e. apitrace, or at the very least instructions on how to reproduce the problem easily, including in-game settings).
Nothing extracted yet.
This might be a Mesa issue since it's manifested itself after updating to 19.3 and when downgrading to 19.2 I don't have any issue but I figured I should check here first if there's an obvious reason this is happening + not sure how much support one gets for bugreports on Mesa when using applications through Wine/DXVK.
I'm seeing some sort of corruption, it looks a bit like it's at the edges of the tiles.
Screenshot below
GPU: RX480
Driver: RADV, Mesa 19.3.3 ACO not enabled, although I get the same corruption with ACO enabled.
Wine: Staging 5.0
DXVK: 1.5.3
dxgi and d3d11 logs below
HeroesOfTheStorm_x64_dxgi.log
HeroesOfTheStorm_x64_d3d11.log
If this is something that is caused by/should be fixed by DXVK I can add an Apitrace of course.