Quoting: ShmerlJózef also made a [patch](https://bugs.winehq.org/attachment.cgi?id=59744&action=diff&context=patch&collapsed=&headers=1&format=raw) that fixes invisible surfaces ([bug #43131](https://bugs.winehq.org/show_bug.cgi?id=43131)).

This is good news! That was the thing that annoyed me the most in the past versions of wine.

Any idea in which version of wine-staging it will get added?

Quoting: devlandAny idea in which version of wine-staging it will get added?

No idea, it's just a manual patch for now. Whether it will go in staging, or in Wine master itself is something developers will decide. It may remain such manual patch for the time being.

Quoting: Shmerl20-30 fps, flickering
OpenGL renderer string: AMD Radeon (TM) RX 480 Graphics (POLARIS10 / DRM 3.18.0 / 4.13.0-1-amd64, LLVM 5.0.0)
OpenGL core profile version string: 4.5 (Core Profile) Mesa 17.4.0-devel (git-386f6cd041)

So Mesa is almost on par with NVIDIA again? :woot: :woot: ;)

Quoting: bekoSo Mesa is almost on par with NVIDIA again? :woot: :woot: ;)

It is not impossible that Mesa implemented recently that nvidia drivers already had. Maybe Shmerl will figure out exactly what patch caused it and make a Mesa-tw3 build. :whistle:

Quoting: ShmerlIt may remain such manual patch for the time being.

At least it works as advertised. Applied it against wine-2.21 before the few remaining staging patches and the invisible areas of ground are gone. Would be interesting to know if it also fixes the invisible enemies, or if that's an entirely different issue.

At any rate, I'm now well into Blood and Wine, so the fun will be over all too soon :-(.

Shmerl quick to aid! Movement on #43872 (invisible monsters) by J. Kucia :D

Quoting: kaimanWould be interesting to know if it also fixes the invisible enemies, or if that's an entirely different issue.

It's a different issue and that patch doesn't fix it.

Quoting: bekoMovement on #43872 (invisible monsters) by J. Kucia :D

Yeah, that's one of the major remaining issues (besides performance of course). As Józef wrote:

The bug is caused by two issues:
* wined3d doesn't support draw calls with no attachments,
* wined3d tries to map transform feedback buffers while transform feedback is paused.
I should have a fix soon.

Quoting: EhvisMaybe Shmerl will figure out exactly what patch caused it and make a Mesa-tw3 build. :whistle:

I'll try to isolate what causes that drop and flickering and will file a Mesa bug.

mesa-git 24e90047088599e686b636fde9bda3a96f34a35c: performance is better (but still not fully better), but flickering still exists. So that's an unrelated issue and should be narrowed down separately (the above is just around half way between current master and last known good revision, it's not the hash of the change that broke something).

It's basically a binary tree search for a good revision :)

b9446172249f05f577072f09ea17a391f3ee5d90 - no flickering, performance is good.

e80bbd6f52341cbf9363f3c0c8b7ad3be851b1e6 - no flickering, performance is good.

de126b0402c53ed9011b8ea5237daea554537aa1 - no flickering, performance is good.

I have a suspiction it's realted to this:
adab7f16ffd3ea4c52e8d07f40ca6ae4868c3706 2017-11-06 radeonsi: don't map big VRAM buffers for the first upload directly | Marek Olšák

But I'll continue narrowing it down gradually.

60a9705e0040a0962b82ec5363c14503d8a96d42 - no flickering, performance is good. Getting close to it.

OK, one of these introduces flickering:

adab7f16ffd3ea4c52e8d07f40ca6ae4868c3706 2017-11-06 radeonsi: don't map big VRAM buffers for the first upload directly | Marek Olšák
4b0dc098b2561c07c59f7dab2813640a25789bf1 2017-11-06 gallium/u_threaded: don't map big VRAM buffers for the first upload directly | Marek Olšák
a5d3999c31e2460f690b561b41170bb7bc24fc65 2017-11-06 gallium/u_threaded: clean up tc_improve_map_buffer_flags and prevent reentry | Marek Olšák

https://bugs.freedesktop.org/show_bug.cgi?id=103874

Jozef kucia send new patch for solve bug 43131, marked as wined3d: Fix SM4 discard_z instruction

https://source.winehq.org/patches/data/138856

Maybe in next hours can be approved

^_^

Kudos for all the testing <3

So I tried to find monsters with bug [#43828](https://bugs.winehq.org/show_bug.cgi?id=43828) happening on them, but for the life of me I couldn't find any that exhibited the bug. A while back it seemed to happen almost always on Alghouls or at least on one Harpy/Ekhidna/Siren or Endrega of a pack.

Is this still happening for other people?

Quoting: FaattoriSo I tried to find monsters with bug [#43828](https://bugs.winehq.org/show_bug.cgi?id=43828) happening on them, but for the life of me I couldn't find any that exhibited the bug. A while back it seemed to happen almost always on Alghouls or at least on one Harpy/Ekhidna/Siren or Endrega of a pack.

Is this still happening for other people?

I noticed that instead of those polygons sticking out, rotfiends became invisible for me. It's possible it's caused by the same bug as invisible monsters, so I'll wait for that to be fixed to confirm before closing this one.

Quoting: bekoKudos for all the testing <3

Quoting: chancho_zombieagreed thanks for the testing I hate git bisecting, it's really time consuming.

Thanks! Yeah, it's kind of tedious, especially if you have a lot of revisions in between the last known good one and current master. I saved some time focusing on changes to radeonsi here (in addition to regular binary search bisection).