This change makes it so baking displacement to the high-resolution mesh
(Use Low Resolution Mesh option is OFF, displacement is calculated
between top multi-resolution level and subdivided viewport level mesh)
uses texture UVs and tangent space from the high-res mesh.
This matches intended use-case when object baked with such configuration
have subdivision surface applied to them bringing overall resolution to
the same level as the highest multi-resolution level.
The issue was simple to see when baking high-res displacement for an
object which uses "Keep Corners, Junctions" UV smoothing.
The unfortunate aspect is increased memory usage due to calculation of
the grid index and grid UV mapping, but it is not too bad (12 bytes per
loop, so is like 48Mb per million face). Feels like there is a way to
optimize it by utilizing knowledge that high-res mesh faces are created
in a specific order, but also feels it is not that important at this
moment.
Pull Request: https://projects.blender.org/blender/blender/pulls/144935
The main idea is to switch Bake from Multires from legacy DerivedMesh
to Subdiv. On the development side of things this change removes a lot
of code, also making it easier easier to rework CustomData and related
topics, without being pulled down by the DerivedMesh.
On the user level switch to Subdiv means:
- Much more closer handling of the multi-resolution data: the derived
mesh code was close, but not exactly the same when it comes to the
final look of mesh.
Other than less obvious cases (like old DerivedMesh approach doing
recursive subdivision instead of pushing subdivided vertices on the
limit surface) there are more obvious ones like difference in edge
creases, and non-supported vertex creases by the DerivedMesh.
- UV interpolation is done correctly now when baking to non-base level
(baking to multi-resolution level >= 1).
Previously in this case the old derived mesh interpolation was used
to interpolate face-varying data, which gives different results from
the OpenSubdiv interpolation.
- Ngon faces are properly supported now.
A possible remaining issue is the fact that getting normal from CCG
always uses smooth interpolation. Based on the code it always has been
the case, so while it is something to look into it might be considered
a separate topic to dig into.
- RNA Patterns Unknown to the Manual -- only print rna_ids that do not return a URL
- Undocumented Sections -- consider the case of types that do not have props (a lot of nodes were returning as false positives)
With the aim of removing `seqbasep` to remove the complicated logic for
repairing pointers within the `Strip` struct when loading files and undo
steps, this commit just moves access of the variable behind a function.
In the future the function will retrieve the list from a Strip pointer,
for now it just returns the existing pointer.
Overall motivation is that blend file pointer manipulation is incompatible
with the changes required for #127706.
Pull Request: https://projects.blender.org/blender/blender/pulls/144624
It has ~1.2x speed-up on CPU and ~1.5x speed-up on GPU (tested on Metal
M2 Ultra).
Individual samples are noisier, but equal time renders are mostly
better.
Note that volume emission renders differently than before.
Pull Request: https://projects.blender.org/blender/blender/pulls/144451
It's important that frozen types are immutable, add a generic
check that mathutils types can be resized and check the frozen flag.
Also correct the exception types when Vector's cant be resized,
using a ValueError instead of a TypeError as the type is correct.
The float buffer should be tagged with the standard untonemapped colorspace,
so that when we convert to PQ/HLG the tonemapping is preserved.
With this change:
* Using AgX highlights properly go to white.
* Using the ACES configs, we can convert EXRs to a HDR video, exactly
matching colors with the HDR video on https://dpel.aswf.io/solemates/.
Pull Request: https://projects.blender.org/blender/blender/pulls/144493
Code checking for expected amount of local/linked/missing
linked/liboverrides IDs is now cleaner and more efficient.
Also add a test where the whole content of the source library is
removed, to validate that liboverrides using these linked IDs as
reference remain available as 'placeholders'.
Make registered py-defined property tests more flexible, by allowing to
pass the property type as an argument.
And add basic such tests for Bool and String types.
If an item name (and ID) is provided, never successfully match only
based on the item index.
This can lead to matching to a complete different ID than the intended
one, which can be catastrophic for the integrity of the next resync.
Also, do not require sucessful match on both source and destination
data, this will always fail in case e.g. an item is removed from a
collection, and can prevent detecting required resync from that
collection then.
This commit also enables new 'harder' test in unittests, added in
previous commit, which is now expected to pass.
Pull Request: https://projects.blender.org/blender/blender/pulls/144429
Add some initial data to test proper handling of resync when there are
specific changes in the reference collection hierarchy (in particular,
when the last child of a given collection is moved somewhere else).
Also will check for correct handling of recursive liboverrides resync in
that case.
The current failing part is commented out for now, until fix is
committed.
Changing a node tree item property (such as the default value) was using a very
broad and generic "tag" function which invalidates the runtime items cache.
This also invalidates any python iterators due to the API using the runtime
cache. Changing node tree items in a loop will then crash.
It's not necessary to invalidate the runtime items cache when the actual item
pointers have not changed. Most RNA updates only change superficial properties,
or at most require a recursive node tree update due to change of identifiers or
types.
This PR introduces a simpler "tag" function to only tag for tree updates by not
rebuild the entire runtime items cache. It also renames existing functions and
docstrings to better explain what each of them does and should be used for.
The `NodeTreeInterfaceChangedFlag` is removed completely because it is only ever
used as a simple boolean indicator of "item changes" that require a cache
rebuild. It is replaced with an atomic bool like flags used for runtime caches.
Pull Request: https://projects.blender.org/blender/blender/pulls/143932
Move the existing USDZ export test from C++ to Python for better
validation. The resulting file is now run through the `usdchecker`
system and we also check the contents of the resulting archive for the
expected texture file. This helped uncover a pathing issue in the
resulting archive where the 'textures' directory was misnamed on win32;
though it was still functional.
Pull Request: https://projects.blender.org/blender/blender/pulls/144176
This PR overwrites the FPS limit in the EEVEE performance tests to 1000 FPS.
Background: I have been using the performance tests with, e.g., the Temple scene from the Blender demo files. However, it has configured an animation playback frame rate of 25 FPS. I think for the performance tests it would be more meaningful to render at an unlimited frame rate. Otherwise, one can just see that Temple renders at 25 FPS on pretty much any GPU and it is not possible to measure performance improvements or regressions.
Pull Request: https://projects.blender.org/blender/blender/pulls/142984
Extend the existing benchmark framework to support the compositor.
Files are added separately.
Example output:
```
$ ./benchmark.py run comp
4.5 main
ghosts 1.2055s 1.0243s
ghosts 1.4936s 1.0454s
ghosts 0.4414s 0.4330s
file:///Users/habib/blender-git/benchmark/comp/results.html
```
Pull Request: https://projects.blender.org/blender/blender/pulls/136600
This test can't easily be updated because it relies on the old socket shape
design before #144119. Parts of it could be done from scratch again, but
most of this stuff is also tested by the new structure type inferencing test
in `bl_node_structure_type_inference.py`.
Pull Request: https://projects.blender.org/blender/blender/pulls/144185
Mixing dynamic data with other data always results in dynamic now. While this
wasn't true without lists, it is true now and also makes sense generally. Also
mixing lists and single values or fields results in a list now.
Pull Request: https://projects.blender.org/blender/blender/pulls/144183
The spec for .usdz permits only a small handful of file formats to be
contained in the archive, and HDR is not among those supported[1]. This
also causes validation errors with the `usdchecker` tool (will be
properly tested in a follow up change).
Ignoring the potential for a user to export a .usdz file with materials
referencing unsupported file formats, Blender itself should use only
the supported formats for its business.
[1] https://openusd.org/release/spec_usdz.html#usdz-specification
Pull Request: https://projects.blender.org/blender/blender/pulls/144101
This patch redesigns the File Output node to provide better UX and UI.
This is mainly achieved by allowing the user to create inputs by
dragging into an Extend socket and adjust existing inputs using the
familiar UI list design available in Blender. Additionally, various UI
changes were done:
- The Use Node Format option was renamed to Override Node Format for
clarity.
- Socket types are now fixed and do not change as new links are made,
allowing users to specify the exact output type and employ implicit
conversion if needed.
- The distinction between images and Multi-Layer EXR was made clearer.
- Final output paths are drawn in the UI to remove guess work.
- The Base Path was split into a Directory and a File Name.
- Panels were added to group options, include a panel for the node
format, items, and item formats.
Pull Request: https://projects.blender.org/blender/blender/pulls/141091
Temporary directory handling had a logical error, assuming the
"session" temporary directory was owned and created by Blender
and could be recursively removed on exit.
However, it's possible creating the session sub-directory fails,
in that case the temporary directory was used for the "session".
This meant setting `C:\` as the temporary directory in the preferences
would attempt to recursively remove `C:\` on exit.
Resolve with the following changes:
- Only perform a recursive removal on the temporary directory
if a session sub-directory was created.
- If the creating the user-preferences temporary "session" sub-directory
fails fall back to the systems temporary directory and try to
create the "session" directory there.
Previously this was only done if the preference path didn't exist.
The preferences path was still used if it existed but couldn't be
written to.
Include a test to ensure this is working as expected.
Ref !144042
Unit tests veryfying expectation for curves::nurbs::calculate_evaluated_num().
Expectation is computed from closed form expressions rather then hard coded
values. Purpose for this is to make the tests easier to adjust if, for example,
parameter sampling pattern is changed. It should also make them easier to read
and understand.
Additional purpose is to create a baseline and verify changes for #144000.
Implementation is essentially examples from:
https://link.springer.com/book/10.1007/978-3-642-59223-2
Pull Request: https://projects.blender.org/blender/blender/pulls/143920
'exr multilayer passes' had an invalid input, so its output was updated
The rest of the tests had outdated output sockets after some 'Color'
sockets were converted into 'Vector' sockets
Pull Request: https://projects.blender.org/blender/blender/pulls/144004
These tests check performances of ID python wrapper creation (these are
cached), regular RNA struct wrapper creation, acess to IDProperties,
(un)registering of py-defined PropertyGroup, and access to py-defined
properties (without and without custom get/set).
The goal of these tests is mainly to follow performances impact of
on-going refactors and improvements in the BPY/RNA area.
Many nodes operate on all the instances that are passed into them. For example,
the Subdivision Surface node subdivides the mesh at the root but also instanced
meshes. This works well for most nodes, but there are a few nodes were the old
`modify_geometry_sets` function was not very well defined and it was tricky to
use correctly.
The fundamental problem was that the behavior is not obvious when a node creates
or modifies instances and how those are integrated with the already existing
instances.
This patch solves this with the following changes:
* Remove the old `GeometrySet::modify_geometry_sets` and related
`*_during_modify` methods.
* Add a new `blender::geometry::foreach_real_geometry` function that is similar
to the old `modify_geometry_sets` but has a more well-defined interface:
* It never passes instances into the callback. So existing instances can't be
modified with it.
* The callback is allowed to create new instances. This will automatically be
merged back with potentially already existing instances. The callback does
not have to worry about accidentally invalidating existing instances like
before.
* A few existing usages used `modify_geometry_sets` to actually modify existing
instances (usually just removing attributes). Those can't use the new
`foreach_real_geometry`, so they just get a custom simple recursive
implementation instead of using a generic function.
Pull Request: https://projects.blender.org/blender/blender/pulls/143898
Add two sets of options to the IO Report class that allows tests to
control the verbosity of output.
- Add `Report.context_lines` controlling how many lines of context the
diff uses when there are failures (3 by default)
- Add `Report.side_to_print_single_line` (5 by default) and
`Report.side_to_print_multi_line` (3 by default) controlling how many
items are written out
The first option helps when a failure in the test might not produce
enough lines of output to know which object is affected. The second set
of options allows individual tests to ensure more values are taken into
consideration for test validation.
They are exposed as class variables due to all the inner methods using
them being static.
Pull Request: https://projects.blender.org/blender/blender/pulls/143922
This refactors the code for world to dome light to be shared between USD and
Hydra, and makes rotations work for Hydra the same way they do in USD.
One small behavior change is that missing image files now render black,
matching Cycles and EEVEE more closely.
Co-authored-by: Brecht Van Lommel <brecht@blender.org>
Pull Request: https://projects.blender.org/blender/blender/pulls/143035
This PR adds a new `fresnel_conductor_polarized` function, which calculates reflectance and phase shift (if requested) for both parallel and perpendicular polarized light. This is needed for applying thin film iridescence to conductors (see !141131).
For consistency, this PR also makes `fresnel_conductor` call `fresnel_conductor_polarized` instead of using a fast approximation of the Fresnel equations that is inaccurate at lower n and k values. This will change the output of some Metallic BSDF renders using Physical Conductor and prevent discrepancies when enabling thin film iridescence.
I didn't do any rigorous performance testing, but from timing the functions outside of Blender, `fresnel_conductor_polarized` is significantly slower than the approximation, between 1.5-3x depending on the compiler. This makes sense because it has three square roots and the approximation has none. In some informal tests with metallic_multiggx_physical.blend modified to have more spheres, the new renders took around 1-2% longer on both CPU and GPU.
There are some avoidable inefficiencies in this approach of just calling `fresnel_conductor_polarized`:
- one of the three square roots could be saved since `fresnel_conductor` never needs the phase shift and there are simplifications possible when only calculating the reflectance
- there are several unnecessary multiplications by 1.0 since `fresnel_conductor` uses relative IOR and `fresnel_conductor_polarized` doesn't, though those could get optimized out if inlined
Pull Request: https://projects.blender.org/blender/blender/pulls/143903
The EEVEE performance tests call scene.frame_set at the end of the warmup stage, which recursively
calls frame_change_handler. Eventually, this leads to a RecursionError. This PR adds a guard for
checking if frame_set is currently active and returns early for this recursive call of the function.
Pull Request: https://projects.blender.org/blender/blender/pulls/143203
The current glare streaks threshold for highlights is too high so the
input is just a black image, and no effects are output.
The threshold was reduced from 2.0 to 0.2 to produce a visible change.
Pull Request: https://projects.blender.org/blender/blender/pulls/142320
`node_pixelate` sets the output pixel size to 1, meaning no
transformation occurs and it is essentially the same as a single value
input. This will add a new test that pixelates the image, and renames
the original test to better reflect that edge case
Coverage:
- Function: 62.50% -> 100%
- Line: 37.36% -> 100%
- Region: 45.83% -> 100%
- Branch: 20.00% -> 90.00%
Pull Request: https://projects.blender.org/blender/blender/pulls/142438
The EEVEE performance tests wait for shader compilation to finish before
beginning to record performance. However, shader compilation can also
happen after the first frame. This PR adds a check to the warmup phase
to see if shader compilation is still happening. This happens, e.g., for
the Mr. Elephant demo scene.
Pull Request: https://projects.blender.org/blender/blender/pulls/143690
