**How to reproduce:**
1. Add Cryptomatte Node
2. Enable 'Object' render passes
3. Connect Cryptomatte to output (composite or viewer node)
4. Optional: set matte ID
5. (Do not connect input image)
6. Render
7. Observe invalid output image (or assert in debug build)
(see also PR for an example blend file)
Only CPU is affected. The solution follows the behavior of the GPU backend, where the input color of the single value input image is multiplied with the matte result to produce a non constant output.
Pull Request: https://projects.blender.org/blender/blender/pulls/135786
The issue was that sometimes the group inputs of a node group were shuffled
around unexpectedly and thus inputs were passed to the wrong sockets.
The `or_socket_usages` function sorts the given span so that the key is more
likely to be reused, reducing the number of nodes inserted in the graph. The
issue was that `build_warning_node` passes `group_output_used_sockets_` into the
function the order of which is important. It thus should not be reordered.
The fix is to just never reorder the span passed to `or_socket_usages` but to
make a local copy instead which can be sorted without problems. Often this copy
is done already anyway when the span is inserted into
`graph_params.socket_usages_combination_cache` as `Vector`.
This fix also makes an assumption about `Map.lookup_or_add_cb` which was not
documented before. Namely it assumes that the key is moved into the map only
after the callback has been called. This behavior is now documented and there is
a unit test for it.
Pull Request: https://projects.blender.org/blender/blender/pulls/135528
The equivalent operation in edit mode reused existing vertices
rather than taking the new positions from the convex hull output.
This commit implements the same behavior for the geometry node.
Pull Request: https://projects.blender.org/blender/blender/pulls/135536
When calling the `lookup_or_add` functions on `MutableAttributeAccessor`
we need to make sure that the writer is actually created and the call
did not fail.
This fixes many of the instances where we would use an unchecked
attribute writer and potentially crash.
Pull Request: https://projects.blender.org/blender/blender/pulls/134413
Collapsing curves to a single point when just resampling is unexpected. This
patch changes it so that non-zero-length curves keep at least one segment.
The fix is fairly straight forward, but a bunch of additional code is added to
support the legacy option to avoid breaking backward compatibility.
Pull Request: https://projects.blender.org/blender/blender/pulls/133659
In setups where only very few elements are processed at once, the lazy-threading
hints that come from `parallel_for` are not triggered. However, often it can
still be worth to use multi-threading in such cases. Therefore, we need to find
additional places where the lazy-threading hints are sent. There are many
possibilities and we probably have to add them step by step as we find .blend
files that show missing multi-threading.
In this patch I choose to send the hint at the beginning of evaluating a repeat
zone if it has 10 or more iterations. The exact number here is just a guess and
could be fine tuned over time if necessary.
A slightly modified version of the file in #134230 speeds up from ~1100ms to
~100ms with this change. The exact speedup changes quite a bit each run, but is
always between 5 and 20x, so it's quite noticable.
Pull Request: https://projects.blender.org/blender/blender/pulls/134408
User may think of the `radius` vector value as RGB color, however it's
meant to be the depth of scattering for R/G/B channels. Now clarified in
the desctiption.
Pull Request: https://projects.blender.org/blender/blender/pulls/134088
It already doesn't work in a very fairly simple case when the Grease Pencil
geometry is transformed. This simple case used to work before we changed
the Transform Geometry node to transform layers instead of points.
Pull Request: https://projects.blender.org/blender/blender/pulls/134131
The issue was that the code did not handle input sockets that are fields
implicitly correctly. It just retrieved a single from them, instead of
treating them as having an unknown value.
The new Object and Collection constant geometry nodes do not register their dependency on their targets.
This causes the graph not to update properly when they are modified.
Pull Request: https://projects.blender.org/blender/blender/pulls/133784
The File Output node sometime ignores the transformations of their
inputs. That's due to the fact that transforms are now delayed and the
File Output node does not realize its inputs on its domain in case it
was not multi-layer.
To fix this, add another realization mode for transforms only. And use
that in the File Output node, as well as the Bokeh Blur, UV Map, and
Plane Track Deform, which also need this fix.
Pull Request: https://projects.blender.org/blender/blender/pulls/133850
There is a special case in the compositor code where viewer nodes are
treated as composite nodes. This patch renames relevant methods and
updates comments to clarify this use case.
Pull Request: https://projects.blender.org/blender/blender/pulls/133811
This patch changes how the Wrap option in the Translate node works. It
now repeats infinitely when realized, and not just in its own domain.
This is essentially how it used to work prior to version 4.2, so we are
reverting that change. We also rename Wrap to Repeat for clarity.
This is part of an ongoing to project to allow the user to control how
boundaries and empty areas are filled after transformations. So similar
options will be added to all transform nodes. But this patch is the
first step, since the functionality is already there, we just need to
extend it to work in all domains.
Pull Request: https://projects.blender.org/blender/blender/pulls/133766
Artists often want to quickly switch back and forth between two or more nodes while compositing.
This patch implements two operators `NODE_OT_viewer_shortcut_set` and `NODE_OT_viewer_shortcut_get` that allow users to map a viewer node to a shortcut. For example, pressing `cltr+1` while a node is selected, assigns that node to the shortcut `1`, creates a viewer node if it has none attached and sets that viewer node to active. Pressing `1` will set the active node with shortcut `1` to active.
Shortcuts are saved in DNA to preserve them after saving/loading blend files.
Limitations:
- Only compositor node tree is supported, because shading editor has no viewer node and geometry nodes viewer works differently.
Pull Request: https://projects.blender.org/blender/blender/pulls/123641
Previously, there was a `StringRef.copy` method which would copy the string into
the given buffer. However, it was not defined for the case when the buffer was
too small. It moved the responsibility of making sure the buffer is large enough
to the caller.
Unfortunately, in practice that easily hides bugs in builds without asserts
which don't come up in testing much. Now, the method is replaced with
`StringRef.copy_utf8_truncated` which has much more well defined semantics and
also makes sure that the string remains valid utf-8.
This also renames `unsafe_copy` to `copy_unsafe` to make the naming more similar
to `copy_utf8_truncated`.
Pull Request: https://projects.blender.org/blender/blender/pulls/133677
The Translate node is a transform operation that should not realize the
transformations of its input. So this patch marks its input accordingly.
Additionally, we also protect against operating on single values and
pass them without changes.
This patch introduces a new Derived Resources concept to the compositor.
Derived resources are resources that are computed from a particular
result and cached in it in case it is needed by another operation, which
can greatly improve performance in some cases at the cost of more memory
usage.
The first use case is to store denoised versions of the Denoising Albedo
and Denoising Normals passes if auxiliary pass denoising is enabled in
the denoise node. Consequently, multi-pass denoising setups where the
same auxiliary passes are used in multiple denoise nodes should be much
faster due to caching of the derived resources.
This implementation has the limitation that it can't preemptively
invalidate the cache when the derived resources are no longer needed to
free up memory. This requires a special resource tracking mechanism that
need to happen during node tree compilation, and will be submitted
later. The limitation is not significant in the particular derived
resources that is currently implemented. Since the auxiliary passes are
rarely used outside of denoising.
Fixes#131171.
Pull Request: https://projects.blender.org/blender/blender/pulls/125671
The issue is that the localized node tree has an invalid type if the linked file
is not available. So we essentially ended up with a geometry node group that
contains a group node that points to a non-geometry nodes group.
Pull Request: https://projects.blender.org/blender/blender/pulls/133650
This patch delays applying transformations until realization happens on
some other domain.
Currently, transformations are applied immediately at the point of
transform nodes, this is problematic for a few reasons:
- If that result was then realized on some other domain, interpolation
will have happened two times, at the transform nodes and at the node
that required realization, causing less than ideal precision issues.
- It is not possible to repeat or extend a rotated result because its
empty areas will be zero filled, leaving gaps in its extension. So
this patch is a prerequisite for #132371 if we want full support for
repetition.
- Doing inverse transformations will introduce interpolation artifacts
which might be undesirable. Inverse transformations might be used to
do pixelation for instance, so this change will be undesirable in this
case. But we decided that this is not a use case that we want to
support, and we added explicit pixel size control to the pixelate node
as an alternative.
So this has four implications, two that might be considered bad:
- Transformations will now be higher quality and more precise.
- Repetition and other boundary extension methods will now be possible.
- Downsampling then upsampling will no longer produce pixelated results.
- Realization might happen multiple times with identical results in some
cases.
The last point not a big issue, since domain realization is not a big
bottleneck in the compositor, and the plan is to move realization into
pixel operations, so it will even be more efficient than it is now.
Pull Request: https://projects.blender.org/blender/blender/pulls/133158
Fairly weird, debug builds were fine, but release ones with clang 19
and mold failed at linking step, complaining about misisng
`bNode::output_sockets()` symbol.
Guess debug config somehow pulls this symbol from somewhere else?
This reverts commit 19e9092cb6. The
minimum requirement is for builds that Blender officially provides. We
still need to maintain code for none SSE builds.
