- Do not report MIP_AUTOGEN if the image format cannot
be used as a color attachment
- Do not report SAMPLE_COMPARISON and GATHER_COMPARISON
if the DXGI format has no corresponding depth format
- Only report image-related features if the image format
can actually be used as a sampled image
Some games do not compute the number of mip levels of
a texture or texture view correctly, so we should work
around this by capping it to the highest possible value.
If this method is used to clear a view with a floating point format,
we need to create a compatible view with an integer format in order
to clear the resource with the correct value. Fixes some calls to
this function in Rise of the Tomb Raider and other games.
No `std::mbstowcs` string conversion needed than.
W/o this change I have different output when running `dxgi-factory.exe` compiled with MinGW:
* `Output 0:`
```
\\.\DISPLAY1 (default)
\\.\DISPLAY1 (DXVK MinGW)
\ (DXVK winebuild)
```
With this patch all three variants are identical (`\\.\DISPLAY1`)
p.s. same problem in dxgi_adapter.cpp, but `deviceProp.deviceName` is vulkan structure parameter (char deviceName[VK_MAX_PHYSICAL_DEVICE_NAME_SIZE]).
This is part of a major refactoring process regarding DXGI->Vulkan
format conversions. Since we don't patch format lookup tables any
longer, we can create a global lookup table.
The D3D11 ClearUnorderedAccessView* and ClearView functions
will have to be emulated using compute shaders rather than
clear operations, since Vulkan clear operations do not take
image views and their format into account.
* Added instructions for installing from the easy build script.
* Skipped the packaging part by adding --no-package
* Update README.md
Rephrased some things a bit.
We don't need to iterate over the full shader code when creating
a new shader module. This optimization may slightly reduce the
initial pipeline creation time.
Drivers from both major vendors implement their own shader cache
already, and storing a cache per game causes more issues than it
solves. Should fix#261.
