Philip Rebohle

> I think 2.5.3 may have fixed that? don't remember. The only thing 2.5.3 really did is give users an option to allow full-screen exclusive again, since forcing it off...

What's weird here is that current master changed it to always creating `UNORM` swapchains because using sRGB formats broke RTSS, and yet the issue remains the same. Apparently, what's causing...

Yeah, in either case, using a different *rendering* format should be entirely transparent, but evidently that is not the case. #4779 should work around this. Not that I'm a particularly...

Can't get that version to run, doesn't look like we get working a device / swapchain at all. This is very much still using the *non-recommended* way of using the...

> The V-Sync doesn't work (if I put 60Hz and enable the V-Sync in-game, the game fails to sync to the selected Hz, they sync with the screen Hz instead)...

AMD is known not to work with 2.6 on many systems, it's a pretty much undebuggable issue with their driver. Not sure if that's your problem since no long is...

Yeah this is the exact issue I was talking about, just swapchain creation failing for no apparent reason. I used to be able to trigger that bug on my Windows...

This has been a known issue for years at this point, the win32 apis we use to query display modes sometimes just don't seem to report available resolutions. There are...

That option interacts with the Vulkan swap chain, not with any of the APIs involved in querying display modes.

the d3d11 option implicitly applies to d3d10 already, not entirely sure what Crysis does here to make vegetation look weird but if there are things happening like resolve before tonemapping...