This PR adds world probe baking to Eevee-next. The world is baked to a
cubemap and is used for reflective light in the deferred render pass.
The world probe is baked to a resolution of 2048x2048. In the future this
would become a user facing setting, but wasn't considered essential for
the first implementation.
When updating the world cubemap the world surface shader is reused.
Currently the world surface shader clears many render passes. It was
decided to replace the render passes with dummy textures as the effort
and potential slowdown didn't weigh against the benefit of doing this nicely.
Updating the world reflection probe isn't expected to happen often.
A big difference with Eevee(-legacy) is that the roughness GGX reflection
parameter isn't baked into the texture (as mipmap levels), but is calculated
during shading. This improves accuracy as we don't assume that every
object is an infinitive small sphere. The result has more noise and that
will be tackled after SSR will land.
Pull Request: https://projects.blender.org/blender/blender/pulls/108149
`CD_HAIRLENGTH` is not really a custom data type, it was just used to
specify the hair particle "length" attribute to pass it to EEVEE and
material preview. Because of the "typemap" array in `CustomData`,
it's better not to have this unnecessary type. Instead, use the same
mechanism used to request the active color attribute.
Pull Request: https://projects.blender.org/blender/blender/pulls/109449
The Voronoi Smooth F1 mode breaks when the Smoothness is 0 for OSL. This is
due to a zero division in the shader.
To fix this, standard F1 is used when Smoothness is 0.
Pull Request: https://projects.blender.org/blender/blender/pulls/109255
Replaces vector of allocations with dynamic linked list.
Bug caused by previously freed buffers still having been in the
list. Linked list enables fast removal of already-released buffers.
Also ensured that the memory manager classes are included in
memory tracking.
Authored by Apple: Michael Parkin-White
Pull Request: https://projects.blender.org/blender/blender/pulls/108940
The current Vulkan resource management has some issues as context that
are not active can still use resources that are freed via another
context.
When this happens incorrect data can be read on the GPU and even crash
Blender. When trying to bind something that now contains other memory
pointers.
This change introduces that contexts are tracked via the device.
Context will be registered/unregistered with the device instance.
Unbinding of resources must pass the device and the device will check
all registered contexts. Binding of resources will happen via the active
context only.
On user perspective this now allowes:
- Opening/switching files
- Switching workspaces
- Switching render engines
Pull Request: https://projects.blender.org/blender/blender/pulls/108968
Fractal noise is the idea of evaluating the same noise function multiple times with
different input parameters on each layer and then mixing the results. The individual
layers are usually called octaves.
The number of layers is controlled with a "Detail" slider.
The "Lacunarity" input controls a factor by which each successive layer gets scaled.
The existing Noise node already supports fractal noise. Now the Voronoi Noise node
supports it as well. The node also has a new "Normalize" property that ensures that
the output values stay in a [0.0, 1.0] range. That is except for the F2 feature where
in rare cases the output may be outside that range even with "Normalize" turned on.
How the individual octaves are mixed depends on the feature and output socket:
- F1/Smooth F1/F2:
- Distance/Color output:
The individual Distance/Color octaves are first multiplied by a factor of
`Roughness ^ (#layers - 1.0)` then added together to create the final output.
- Position output:
Each Position octave gets linearly interpolated with the combined output of the
previous octaves. The Roughness input serves as an interpolation factor with
0.0 resutling in only using the combined output of the previous octaves and
1.0 resulting in only using the current highest octave.
- Distance to Edge:
- Distance output:
The Distance octaves are mixed exactly like the Position octaves for F1/Smooth F1/F2.
It should be noted that Voronoi Noise is a relatively slow noise function, especially
at higher dimensions. Increasing the "Detail" makes it even slower. Therefore, when
optimizing a scene one should consider trying to use simpler noise functions instead
of Voronoi if the final result is close enough.
Pull Request: https://projects.blender.org/blender/blender/pulls/106827
This PR adds support for indirect compute.
Indirect compute is almost the same as regular compute. The
only difference is that the parameters for the compute dispatch
isn't passed as a parameter, but that these parameters are part
of a buffer.
Pull Request: https://projects.blender.org/blender/blender/pulls/108879
Resolves failing unit tests where missing texture read support for
1D/3D textures was required. Also resolves a number of missing
format support cases when determining texture component
count and byte size for reading/writing.
Authored by Apple: Michael Parkin-White
Pull Request: https://projects.blender.org/blender/blender/pulls/108751
Math tests failing in Metal due to non-zero-initialized values
and shader compilation error caused by mutli-component
boolean being implicitly evaluated.
Authored by Apple: Michael Parkin-White
Pull Request: https://projects.blender.org/blender/blender/pulls/108758
Bug in Metal when gl_FragDepth is used but not written to by a shader
causing depth-based navigation to fail. Patch ensures gl_FragDepth
is assigned a default value at the current depth from the vertex
shader to catch all edge-cases of this.
Authored by Apple: Michael Parkin-White
Co-authored-by: Michael Parkin-White <mparkinwhite@apple.com>
Pull Request: https://projects.blender.org/blender/blender/pulls/108453
This PR adds initial cube (array) support. Depending on how the texture
is used a different image view is created. When used as a framebuffer
attachment only a single side of the cubemap is attached. The image
view is attached as a 2d texture array. When used as a shader resource
the image view is a cubemap.
Also adds test cases to test both scenarios.
Pull Request: https://projects.blender.org/blender/blender/pulls/108794
Workbench normals where inverted resulting to drawing artifacts like:
- Black pixels during matcaps (normals contained NAN values)
- Studio light inverted
The root cause for this was that Vulkan default to determine if a
primitive is facing the camera should be switched. The default was still
incorrect.
Although this fixes it, a better solution is to determine the front
facing based on if the viewport should be switched. As all rendering
occurs when the viewport is inverted it doesn't matter at this moment.
Pull Request: https://projects.blender.org/blender/blender/pulls/108797
Ensure buffer read operations wait for pending GPU work to
complete. This is only required for unit tests, but required to
ensure correct results in all cases. Also adds support for staging
buffer if GPU memory type is private.
Buffer clearing routine added to support values above 1 byte.
The previous buffer clearing functionality using fillBuffer writes
values to each byte, rather than the full 32 bit uint used by the
GPU API.
Authored by Apple: Michael Parkin-White
Pull Request: https://projects.blender.org/blender/blender/pulls/108757
Addressing a number of small issues and OOB reads/writes occuring in
EEVEE Next shadows + lighting passes. Improving correctness for unit
tests. Shadows are not yet working overall, but this unblocks progress
towards unit tests.
Authored by Apple: Michael Parkin-White
Pull Request: https://projects.blender.org/blender/blender/pulls/108768
The ownership of image views depends on how they are used. When used as
an framebuffer attachment the image view is owned by the framebuffer.
When used as a image/texture in a shader the ownership of the image view
is the texture itself.
Pull Request: https://projects.blender.org/blender/blender/pulls/108765
* opengl_context -> system_gpu_context. This is the operating system OpenGL,
Metal or Vulkan context provided by GHOST.
* gpu_context -> blender_gpu_context. This is the GPUContext provided by
the Blender GPU module, which wraps the GHOST context and adds some state.
* Various functions create/destroy/enable/disable both contexts, these have
just gpu_context in the name now.
Pull Request: https://projects.blender.org/blender/blender/pulls/108723
This makes tracking shaders easier across the pipeline. Durring logging
and debugging the vulkan driver now remembers the shader associate to
the descriptor and provides this information in its messages.
Pull Request: https://projects.blender.org/blender/blender/pulls/108743
Adds support to create SRGBA8 textures using UBYTE as input data.
It is assumed that the UBYTE are already converted to SRGB and no
conversion actually needs to happen.
Pull Request: https://projects.blender.org/blender/blender/pulls/108741
Some vulkan platforms don't support 24bit depth components. In this
case we will allocate float depth component. During readback the data
should be converted back to what is expected.
Pull Request: https://projects.blender.org/blender/blender/pulls/108698
This PR adds conversion template to convert between Low Precision float
formats. These include Binary32 floats and lower. It also adds support
to convert between unsigned and signed float formats and float formats
with different mantissa and exponents.
Additionally overflows (values that don't fit in the target float
format) will be clamped to the maximum value.
**Reasoning**:
Up to now the Vulkan backend only supported float and half float
formats, but to support workbench, 11 and 10 unsigned floats have to be
supported as well. The available libraries that support those float
formats targets scientific applications. Where the final code couldn't
be optimized that well by the compiler.
Data conversion for color pixels have different requirements about
clamping and sign, what could eliminate some clamping code in other
areas in Blender as well. Also could fix some undesired overflow when
using pixels with high intensity that didn't fit in the texture format
leading to known artifects in Eevee and slow-down in the image editor.
**Future**
In the future we might want to move this to the public part of the GPU
module so we can use this as well in other areas (Metal backend), Imbuf clamping
See 3c658d2c2e69e9cf97dfaa7a3c164262aefb9e76 for a commit that uses
this and improves image editor massively as it doesn't need to reiterate over
the image buffer to clamp the values into a known range.
Pull Request: https://projects.blender.org/blender/blender/pulls/108168
