Created attachment 143171 [details]
There are two problems:
1. Trees are rendered incorrectly on Azure Coast
2. Reflections on some cars look strange. Like there are some random colored pixels.
Extended renderer info (GLX_MESA_query_renderer):
Vendor: X.Org (0x1002)
Device: AMD Radeon HD 7900 Series (TAHITI, DRM 3.27.0, 4.20.0-2-amd-staging-drm-next-git, LLVM 8.0.0) (0x6798)
Video memory: 3072MB
Unified memory: no
Preferred profile: core (0x1)
Max core profile version: 4.5
Max compat profile version: 4.5
Max GLES1 profile version: 1.1
Max GLES profile version: 3.2
OpenGL renderer string: AMD Radeon HD 7900 Series (TAHITI, DRM 3.27.0, 4.20.0-2-amd-staging-drm-next-git, LLVM 8.0.0)
OpenGL core profile version string: 4.5 (Core Profile) Mesa 19.0.0-devel (git-acdad24585)
Created attachment 143172 [details]
Created attachment 143358 [details] [review]
hack for the trees issue
Does the attached diff also fix the trees issue on your side?
For some reasons, this optimization doesn't work correctly with that game.
Created attachment 143361 [details]
Your patch works also for me.
BTW those bad reflections do only occur with MSAA. By also enabling FXAA makes it even more visible.
The trees issue should be fixed with https://cgit.freedesktop.org/mesa/mesa/commit/?id=334da034d8d91ca5a0a1bff8deaefd8ca762c42e
Does RADV_DEBUG=zerovram help?
Created attachment 143587 [details] [review]
hack reflection issue
Can you try this hack too?
Sorry for the late response. Without testing your patch, I think the reflection issue went away with a proton update. I suspect going from DXVK 1.01 to 1.03 fixed the issue.
Thx for your support!
Are you sure? Just tried with 1.0.3 and the issue still happens. Less frequently but still.
So you are right; this is still happening although much less often. Previously this bad effect was much more noticeable.
Its really hard to reproduce this now.
Also there isnt a change between DXVK versions. I didnt had time to test Mesa for some months and when I got back, there got a lot of things updated.
So I tried your patch and RADV_DEBUG=zerovram and each of those fixed the issue for me.
-- GitLab Migration Automatic Message --
This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.
You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/mesa/mesa/issues/855.