Blender has the option to disable SSBO support. This was accessible
as a command line option `--debug-gpu-disable-ssbo`.
Blender 4.0 has a hard requirement for OpenGL 4.3 which includes
SSBO support by default.
This PR removes the command line option as it makes no sense to
have it anymore.
Related to #112224
Pull Request: https://projects.blender.org/blender/blender/pulls/112571
In Vulkan the grid overlay wasn't rendering correctly. The reason was
a mis alignment in the std430 layout. This PR fixes the std430 layout
so the grid overlay (and perhaps also other sections) would work.
Root cause is that the reserve size if an nVec3 is different when used
as an array versus a single attribute. This was not taken into account
where the overlay flag was read from uninitialized memory.
Pull Request: https://projects.blender.org/blender/blender/pulls/112564
Array textures stores the array length inside the size of the image. In
vulkan the size mustn't contain the array length, but it should be set
as a separate parameter. This logic wasn't implemented when updating
the mipmap chain.
Patch also includes some smaller fixes where textures would not yet be
allocated on the device, but was expected to be.
Pull Request: https://projects.blender.org/blender/blender/pulls/112518
Uniform buffers can have data provided during construction, or as
attached data. Attached data should be uploaded during bind.
This PR implements `GPU_uniformbuf_create_from_list`.
Pull Request: https://projects.blender.org/blender/blender/pulls/112517
Resources can still be in use inside the vulkan command queue, when
they are destroyed. Especially as Vulkan doesn't allow resetting
bindings to nothing.
This PR will collect the resources that needs to be destroyed inside
the VKDevice and actually destroy the resources when on frame end.
The resources currently include:
* VkBuffer
* VkImage
* VkImageView
* VkFramebuffer
* VkRenderPass
Pull Request: https://projects.blender.org/blender/blender/pulls/112514
Previous implemetation had some limitations that were not working
for complex shaders. This limitations included:
* All bindings should be bound to a location < 16.
* Collision could happen when image locations where read that
overlapped with the texture locations.
This PR replaces the fixed binding locations with a dynamic vector.
When allocating locations for images the image offset is determined
that it is always clear if the requested locations is already
offsetted or not. This is done by ensuring the offset to larger than
the max slot of samplers, but also of image slots.
When using Eevee-next materials the slot numbers can be very high
as the location of the first material texture is set to
`GPU_max_textures_frag()` On current NVIDIA drivers this number is
set to 1K. Images are stored after the textures slots upwards.
In the future driver could set this to MAX_INT which will not work
with the new implementation. This isn't the case at this moment
and perhaps also not realistic. We don't check for this situation yet.
Pull Request: https://projects.blender.org/blender/blender/pulls/112511
The NDEBUG is a toggle define, and in debug builds it is not defined.
This change solves the warning
mtl_context.mm:2282:5: warning: 'NDEBUG' is not defined, evaluates to 0 [-Wundef]
Pull Request: https://projects.blender.org/blender/blender/pulls/112504
because it contains reflectance and transmittance, so BSDF would be a
morep proper name.
Also rename BSDF to BRDF at places where only reflectance is returned.
Previously, the Principled BSDF used the Subsurface input to scale the radius.
When it was zero, it used a diffuse closure, otherwise a subsurface closure.
This sort of scaling input makes sense, but it should be specified in distance
units, rather than a 0..1 factor, so this commit changes the unit and renames
the input to Subsurface Scale.
Additionally, it adds support for mixing diffuse and subsurface components.
This is part of e.g. the OpenPBR spec, and the logic behind it is to support
modeling e.g. dirt or paint on top of skin. Before, materials would be either
fully diffuse (radius=0) or fully subsurface.
For typical materials, this mixing factor will be either zero or one
(just like metallic or transmission), but supporting fractional inputs makes
sense for e.g. smooth transitions at boundaries.
Another change is that there is no separate Subsurface Color anymore - before,
this was mixed with the Base Color using the Subsurface input as the factor,
but this was not really useful since that input was generally very small.
And finally, the handling of how the path enters the material for random walk
subsurface scattering is changed. Before, this always used lambertian (diffuse)
transmission, but this caused some problems, like overly white edges.
Instead, two different methods are now used, depending on the selected mode.
In Fixed Radius mode, the code assumes a simple medium boundary, and performs
refraction into the material using the main Roughness and IOR inputs.
Meanwhile, when not using Fixed Radius, the code assumes a more complex
boundary (as typically found on organic materials, e.g. skin), so the entry
bounce has a 50/50 chance of being either diffuse transmission or refraction
using the separate Subsurface IOR input and a fixed roughness of 1.
Credit for this method goes to Christophe Hery.
Pull Request: https://projects.blender.org/blender/blender/pulls/110989
- Adds tint control, which simulates volumetric absorption inside the coating.
This results in angle-dependent saturation and affects all underlying layers
(diffuse, subsurface, metallic, transmission). It provides a physically-based
alternative to ad-hoc effects such as tinted specular highlights.
- Renames the component from "Clearcoat" to "Coat", since it's no longer
necessarily clear now. This matches naming in e.g. other renderers or OpenPBR.
- Adds an explicit Coat IOR input, in preparation for future smarter IOR logic
around the interaction between Coat and main IOR. This used to be hardcoded
to 1.5.
- Removes hardcoded 0.25 weight multiplier, and adds versioning code to update
existing files accordingly. OBJ import/export still applies the factor.
- Replaces the GTR1 microfacet component with regular GGX. This removes a corner
case in the Microfacet code, solves #53038, and makes us more consistent with
other standard surface shaders. The original Disney BSDF used GTR1, but it
doesn't appear that it caught on in the industry.
Co-authored-by: Weizhen Huang <weizhen@blender.org>
Pull Request: https://projects.blender.org/blender/blender/pulls/110993
5cf7089e43 added the `BuiltinBits::LAYER` to shaders with a geometry
stage. This causes compilation errors when
`GLContext::layered_rendering_support` is false (otherwise the flag
does nothing).
This PR moves the `LAYER` flags to the `no_geom` shader versions and
adds a check to `ShaderCreateInfo::finalize()` to ensure the `LAYER` flag
is not used in shaders with a geometry stage.
Pull Request: https://projects.blender.org/blender/blender/pulls/112245
Component swizzling is part of an image view and localized in the code.
With component swizzling a different channel (or 0 or 1) can be read
when the component is accessed.
Pull Request: https://projects.blender.org/blender/blender/pulls/112273
Adding support for Texture View in Vulkan backend.
When a VKTexture is a texture view many options become
irrelevant, but isn't actually clear via its API. Added
asserts to make sure that the usage is supported.
This patch allows rendering simple workbench-next scenes
without any validation errors.
Pull Request: https://projects.blender.org/blender/blender/pulls/110887
When `COMP_U*` is used together with a `GPU_FETCH_INT` the backend
tried to convert them to floats. What is of course not what should
happen.
Issue was that these cases were never correctly tested. With this
change the `data` attribute of the `overlay_edit_mesh` shaders are
transferred correctly and can draw some pixels as intended.
Pull Request: https://projects.blender.org/blender/blender/pulls/112270
Shader draw parameter support is needed for Workbench-next. (Almost) All
devices that we support also supports shader draw parameters. This PR
doesn't include a workaround for devices that don't have support.
For end-users this PR will allow workbench-next to render the object on
the correct location. Before this patch they were always drawn on the origin.
The catch is that OpenGL uses gl_InstanceID which always starts counting at 0.
Vulkan has gl_InstanceIndex which holds the correct instance number. By
subtracting the gl_InstanceIndex from gl_BaseInstance will get the correct
gl_InstanceID.
This patch also activates some device extensions to perform in shader printf
statements inside a shader. Documentation for this will be added to the dev
docs how to use them.
Pull Request: https://projects.blender.org/blender/blender/pulls/111461
Schlick's approaximation used by EEVEE is not accurate near `IOR == 1`,
especially when IOR is exactly one, there is no specular reflection and
the material should appear diffuse.
Cycles bypass the issue by lerping between the f0 and f90 color using
the factor derived from real Fresnel curve. In EEVEE we can use the same
trick as in Glass BSDF to smooth the transition at `IOR == 1`.
Note that at `IOR < 1` there is still mismatch, because f0 is prebaked
in the BTDF look up table. In the future if we color f0 using
`specular_tint`, we can split the table and use the BTDF LUT for the
specular component too.
Enables performance optimizations and new rendering
features through allowing colour attachments to be
tagged with a rasterization order group. This means that
all accesses to the same pixel location are guaranteed
to happen in submission order.
This lays the ground work for tile-based architecture
optimizations, enabling additional features for
deferred rendering which allow subsequent passes
which operate on pixel data to occur in order, while
memory remains on-tile.
This patch allows fragment outputs to be tagged
with a raster order group and also adds a new
FragmentTileIn parameter to a shader, which allows
speciication of incoming parameters from a previous
draw.
Authored by Apple: Michael Parkin-White
Pull Request: https://projects.blender.org/blender/blender/pulls/111748
Enhance custom framebuffer binding state to allow
specification of clear color as part of the loadstore
state at framebuffer bind time.
This ensures all parameters controlling attachment
loading and storage behaviour can be explicitly
specified when binding a framebuffer.
This change enables optimizations which leverage
explicit framebuffer load store state to also specify
a clear color without prematurely triggering a
clear which may occur independently to
render work when using GPU_framebuffer_clear(..).
Authored by Apple: Michael Parkin-White.
Pull Request: https://projects.blender.org/blender/blender/pulls/111810
Removes old unused bind state parameters from
prior to the addition of explicit resource location support.
Patch also ensures compute state is reset between each
encoder and adds a debug option to dispatch each
compute command within its own encoder for debugging
synchronization issues.
Authored by Apple: Michael Parkin-White
Pull Request: https://projects.blender.org/blender/blender/pulls/111753
Memoryless textures are only used as intermediate attachments
during rasterization, but do not have any backing storage. This is
particularly useful if a virutal framebuffer is needed, or, there is
a situation where a depth buffer is only needed within the pass
itself and the results are discarded once the pass completes.
Authored by Apple: Michael Parkin-White
Pull Request: https://projects.blender.org/blender/blender/pulls/111749
imageStoreFast provides a variant of imageStore which does
not perform any bounds checking, reducing shader divergence,
register pressure and increasing performance through fewer
instructions.
However, this should only be used for cases where the writing
coordinate is guaranteed to fall within the texture.
Authored by Apple: Michael Parkin-White
Pull Request: https://projects.blender.org/blender/blender/pulls/111750
Some shaders using gl_Layer/gpu_Layer were missing
the correct usage bit in the shader create info. Shaders
also need to inherit feature usage bits from included
additional_info.
Authored by Apple: Michael Parkin-White
Pull Request: https://projects.blender.org/blender/blender/pulls/111751
This ports the LUT using compute shader.
All LUT are computed by the same compute shader
to avoid boiler plate code to add new LUTs.
As for the generation code itself it is mostly the
same except for the use of `hammersley_2d` instead of
regular grid sampling. Regular grid did not improve
anything and was a bit more cumbersome.
This also bumps the number of samples very high
for more precision.
The new utility class for computing the LUT has
methods to write the content to a PFM image file
or as C++ array header.
The hash tables and vector blenlib headers were pulling many more
headers than they actually need, including the C base math header,
our C string API header, and the StringRef header. All of this
potentially slows down compilation and polutes autocomplete
with unrelated information.
Also remove the `ListBase` constructor for `Vector`. It wasn't used
much, and making it easy to use `ListBase` isn't worth it for the
same reasons mentioned above.
It turns out a lot of files depended on indirect includes of
`BLI_string.h` and `BLI_listbase.h`, so those are fixed here.
Pull Request: https://projects.blender.org/blender/blender/pulls/111801
changes include:
* Use microfacet normal instead of macronormal. Previously Cycles used
macronormal for Glass BSDF and Transmission component in Principeld
BSDF, leading to artefacts at grazing angles. This has been corrected
in 5f9b518a8b and 89218b66c2. Now change EEVEE to match this behaviour.
* GGX distribution is now darker due to the shadowing-masking term,
while Multiscatter GGX preserves energy. This now matches Cycles too.
Pull Request: https://projects.blender.org/blender/blender/pulls/111687
