The problem here was that the sync code overwrites the point data, but didn't
reset the transform_applied flag, so the rest of the code assumed that the
points are already in world space and didn't apply the transform again.
This was useful in the beginning to ensure we supported most surface
configurations. However it seems that there are not that many surface
configurations to support.
Pull Request: https://projects.blender.org/blender/blender/pulls/133186
Have `eyedropper_color_sample_fl` return success, in case platform
sampling starts but is canceled (Async return on MacOS). Removes any
chance of returning black color on failures.
Pull Request: https://projects.blender.org/blender/blender/pulls/133124
Cycles has a sample offset feature allowing users to render X samples
in a single frame on one device, then the remaining Y samples later or
on a different device and combine them back together at the end.
However in most situations the result from using this method was
different, and usually lower quality than rendering all the samples in
one go.
This was because Cycles tunes it's random number sequence for the
number of samples being rendered. And the random number sequence was
being tuned for the wrong number of samples in the case that a user
was using the sample offset.
This commit fixes this issue by adding a "sample subset" feature.
The user specifies the total sample count being rendered across all
devices in the existing `Max Samples` parameter, then specifies per
device which subset of samples will be rendered (E.g. Render samples
0-1024 out of a 0-2048 range).
This commit also contains some additional clean up work
inside Cycles related to the area being changed.
Co-authored-by: Brecht Van Lommel <brecht@blender.org>
Pull Request: https://projects.blender.org/blender/blender/pulls/132961
#115187 introduced the eyedropper tool for use outside blender on macOS,
but the following issues remain:
1. Users must grant screen capture permission to blender.
2. No visual cue. (cursor icon cannot display as an eyedropper outside
blender)
3. Users must press `Enter` to activate the picker.
4. `CGWindowListCreateImage` function has been deprecated as of macOS
14.0
The proposed fix is to use `NSColorSampler` class to select colours
directly from onscreen content. It offers:
1. Permission free (no screen capture permission required).
2. A clear visual cue when selecting colours.
Known issue:
1. Still need press `Enter` to activate the picker.
Pull Request: https://projects.blender.org/blender/blender/pulls/131593
This will add support for `VK_KHR_dynamic_rendering_local_read` when supported.
The extension allows reading from an attachment that has been written to by a
previous command.
Per platform optimizations still need to happen in future changes. Change will
be limited to Qualcomm devices (in a future commit).
On Qualcomm devices this provides an uplift of 16% when using shader_balls.blend
Pull Request: https://projects.blender.org/blender/blender/pulls/131053
The `ZDepth` output of the camera data node was different between SVM
and OSL.
SVM would output the `ZDepth`, with negative distances for points
behind the camera. While OSL would output the absolute of the distance,
which resulted in points behind the camera becoming positive.
Align OSL to SVM and allow outputting the negative distance as it
allows users to differentiate between what's in front or behind the
camera.
Pull Request: https://projects.blender.org/blender/blender/pulls/132837
In a recent refactor (1), the subsurface weight was set to be
constant, when it can be modified in the case that `__SUBSURFACE__`
is false, such as when using the adaptive compilation feature.
This commit fixes this issue by rearranging the code so the subsurface
weight is never overwritten.
(1) blender/blender@dd51c8660b
Pull Request: https://projects.blender.org/blender/blender/pulls/132620
When using clangd or running clang-tidy on headers there are
currently many errors. These are noisy in IDEs, make auto fixes
impossible, and break features like code completion, refactoring
and navigation.
This makes source/blender headers work by themselves, which is
generally the goal anyway. But #includes and forward declarations
were often incomplete.
* Add #includes and forward declarations
* Add IWYU pragma: export in a few places
* Remove some unused #includes (but there are many more)
* Tweak ShaderCreateInfo macros to work better with clangd
Some types of headers still have errors, these could be fixed or
worked around with more investigation. Mostly preprocessor
template headers like NOD_static_types.h.
Note that that disabling WITH_UNITY_BUILD is required for clangd to
work properly, otherwise compile_commands.json does not contain
the information for the relevant source files.
For more details see the developer docs:
https://developer.blender.org/docs/handbook/tooling/clangd/
Pull Request: https://projects.blender.org/blender/blender/pulls/132608
Windows Only. On some keyboard layouts, the right Alt key behaves as an
"AltGR" key. This key emits both right alt and left ctrl key events,
which means that it won't match keyboard shortcuts set to just Alt.
This PR, for just layouts that include AltGr, if the scancode is for
right Alt yet the virtual key is actually the control key then ignore
these key events. This treats AltGr exactly as Alt.
Pull Request: https://projects.blender.org/blender/blender/pulls/132536
This PR will add timeline semaphores to be required. It doesn't use
the timeline semaphores yet, but as multiple developments will
rely on it it is better to add the requirement.
Pull Request: https://projects.blender.org/blender/blender/pulls/132683
`ftruncate()` is annotated such that the compiler (at least GCC 13)
warns when the returned status is not used.
This commit adds a `GHOST_ASSERT()` to check that result in debug
builds. In non-debug builds it's discarded without checking, but
explicitly as to not trigger this warning.
No functional changes.
Pull Request: https://projects.blender.org/blender/blender/pulls/132108
Update the OptiX denoiser model type to the newer and more frequently
updated `OPTIX_DENOISER_MODEL_KIND_AOV` denoiser.
The output of the new denoiser type is typically higher quality.
And the old denoiser is no longer being updated.
Original change was proposed by Patrick Mours from NVIDIA.
Pull Request: https://projects.blender.org/blender/blender/pulls/132542
Check was misc-const-correctness, combined with readability-isolate-declaration
as suggested by the docs.
Temporarily clang-format "QualifierAlignment: Left" was used to get consistency
with the prevailing order of keywords.
Pull Request: https://projects.blender.org/blender/blender/pulls/132361
