When cut a stroke of 1 point, the clean up done to avoid keep 1 point strokes removes the memory, but the pointer to the first stroke was not set to NULL. As this pointer is invalid now, any use of this produces a segment fault because the pointer is corrupted..
Sphinx has rather loose dependency requirements which can cause issues if we aren't careful.
As a solution they recommend that you pin sphinx dependency versions
Currently the emboss is only fading on left side of the widget,
resulting in the emboss extending vertically on the right side
and ending abruptly. This patch fixes this by also fading the
emboss on the right side and making it symmetric.
Differential Revision: https://developer.blender.org/D10810
This fixes T88346.
The code is also more readable by making a better distinction between
the texts used for Distances, "Proportional Size" and "AutoIK-Len".
And the text used to translate the "Proportional Size" is reused.
This patch fixes many minor spelling mistakes, all in comments or
console output. Mostly contractions like can't, won't, don't, its/it's,
etc.
Differential Revision: https://developer.blender.org/D11663
Reviewed by Harley Acheson
This problem has surprisingly been there for quite a few months.
For point clouds all attributes were handled the same, even "position",
which should be transformed when combining source points into the
destination.
Previously the code assumed that curve instances had no attributes.
This is true when the data came from curve objects, which don't support
attributes currently, but it isn't necessarily true when retrieving curves
from evaluated geometry sets.
Make the virtual functions protected and simpler, so that the logic is
better contained in the base class's implementation. Also introduce a
`copy_without_attributes` method to be used for realizing instances.
Add direct user feedback (as a warning report) to user when recursive
resync of overrides was needed.
And some timing (as CLOG logs) about main readfile process steps.
This is essentially adding a new BlendFileReadReport structure that wraps
BKE_reports list, and adds some extra info (some timing, some info about
overrides and (recursive) resync, etc.).
Allows to decompose a given amount of seconds into a random set of
days/hours/minutes/seconds/milliseconds values.
Also add matching test.
Differential Revision: https://developer.blender.org/D11581
Delay depsgraph visibility update tagging until it is known that
graph relations are up to date, and until it is known that the graph
is actually needed to be evaluated.
Differential Revision: https://developer.blender.org/D11660
Currently, the OptiX BVH build options are selected based on whether
we are in background mode (final renders) or not (viewport renders).
In background mode, the BVH is built for fast path tracing and low
memory footprint, while in viewport, it is built for fast updates.
However, on platforms without OpenGL support, the background flag is
always set to true and prevents using fast BVH builds in the viewport.
Now, the BVH options derive from the Scene BVH settings:
* if BVH is static, a fast to trace BVH is built
* if BVH is dynamic, a fast to update BVH is built
Reviewed By: #cycles, brecht
Differential Revision: https://developer.blender.org/D11154
Apply the same fix for T32214 (edge-select failing) to bones
which also failed when their end-points were outside of the view.
- Add V3D_PROJ_TEST_CLIP_CONTENT support for edit & pose bone iterator
and use for selection operators.
- Remove unnecessarily complicated checks with pose-mode lasso tagging.
- Correct error in pose-mode LassoSelectUserData.is_changed
(currently harmless as it's not read back).
`src` and `dst` are perfectly clear, and avoid repeating unecessary
characters when writing the variables many times, allowing more space
for everything else.
When CMake detects and incompatible Python version
it errors out with an error saying at-least python 3.9
is required, but doesn't mention the version it detected.
This makes troubleshooting the problem harder than it
needs to be.
This diff changes the error message to include the python
version CMake detected.
Differential Revision: https://developer.blender.org/D11666
Reviewed By: Ray Molenkamp
This adds preliminary VS 2022 support, since
there currently is no CMake version that
supports the VS2022 IDE only ninja support
was tested.
IDE support should work without any additional
changes as soon as an updated CMake becomes
available.
As VS2022 appears to keep binary compatibility
with earlier MSVC versions, the current SVN
libraries will work for this version.
This commit optimizes the node for the case where it works on many
splines by allowing it to generate mesh data from their combinations
in parallel. By itself, this made the node around twice as fast in my
test file with a result of 20 million vertices, around 600ms instead of
1.2s before.
That isn't actually a very good result; it reveals another bottleneck,
a single threaded loop over all face corners in the mesh normal
calculation code. As a simple change that might improve performance
in some situations, this commit moves normal calculation out of this
node, so at least the work isn't wasted if the mesh is changed later
on in the node tree anyway.
The depth cache (located in `RegionView3D::depths`) is used for quick
and simple occlusion testing in:
- particle selection,
- "Draw Curve" operator and
- "Interactive Light Track to Cursor" operator,
However, keeping a texture buffer in cache is not a recommended practice.
For displays with high resolution like 8k this represents something
around 132MB.
Also, currently, each call to `ED_view3d_depth_override` invalidates
the depth cache. So that depth is never reused in multiple calls from
an operator (this was not the case in blender 2.79).
This commit allows to create a depth cache and release it in the same
operator. Thus, the buffer is kept in cache for a short time, freeing
up space.
No functional changes.
This patch uses threading to flush selection from verts to edges and
from edges to faces. The solution is lockless and should scale well on
modern CPU architectures.
Master:{F10185359}
Patch:{F10185361}
End user performance went from 3.9 to 4.6 FPS (Stanford Dragon) but that
was measured on a Intel Core i7-6700 CPU and AMD RX480 Gpu. The more
cores the better improvements you get.
Reviewed By: mano-wii
Differential Revision: https://developer.blender.org/D11644
This resolves a long standing bug in edge selection
(picking, circle, box & lasso).
Now when one of the edges vertices fails to project into screen space,
the edge is clipped by the viewport to calculate an on-screen location
that can be used instead.
This isn't default as it may be important for the on the screen location
not to be clipped by the viewport.
Adds two new output modes to the CDT api which detect and remove
holes. A hole is a face from which a ray shot to the outside
intersects an even number of constraint edges, except we don't
count constraint edges in the same connected region of faces,
where a region is connected via non-constraint edges.
These modes are useful for filling in outlines meant to represent
text characters and the like.
Original patch was from Erik Abrahamsson, modified by me to work
with the "valid Bmesh" output type too. I also added tests
for the new modes.
Caused by improper testing on my part, assuming a helper function
existed in the industry compatible keymap file, and also assuming it
also used the N and T keys for the left and right side-regions.
A different data structure / implementation is used for curves in the
node tree currently. Converting from the DNA `Curve` structure to this
wasn't slow, but it's nice to decrease overhead. In a test of 14000
splines with 128000 points, this halves the runtime from about 5ms
to about 2.5ms.
The spreadsheet filter tried to apply the mesh selection filter on non-
mesh geometry components. Add a check for the component type,
and also refactor the function to be more easily readable.
The normals were broken because the normal calculation mode wasn't set.
This patch adds a default normal mode so all code creating a spline does
not necessarily have to set it manually. In the future there should be a
way to change this value in the node tree.
- Multi-thread BKE_mesh_recalc_looptri.
- Add BKE_mesh_recalc_looptri_with_normals,
this skips having to calculate normals for ngons.
Exact performance depends on number of faces, size of ngons and
available CPU cores.
For high poly meshes the isolated improvement to BKE_mesh_recalc_looptri
in my tests was between 6.7x .. 25.0x, with the largest gains seen in
meshes containing ngons with many sides.
The overall speedup for high poly meshes containing quads and triangles
is only ~20% although ngon heavy meshes can be much faster.
Remove `seq->tmp` usage from transform code. It was used to tag strips
that need to be "shuffled". Pass these strips in `SeqCollection`
instead.
Reviewed By: sergey, mano-wii
Differential Revision: https://developer.blender.org/D11631
This patch adds support for filtering rows based on rules and values.
Filters will work for any attribute data source, they are a property
of the spreadsheet rather than of the attribute system. The properties
displayed in the row filter can depend on data type of the currently
visible column with that name. If the name is no longer visible, the
row filter filter is grayed out, but it will remember the value until
a column with its name is visible again.
Note: The comments in `screen.c` combined with tagging the sidebar
for redraw after the main region point to a lack of understanding
or technical debt, that is a point to improve in the future.
**Future Improvements**
* T89272: A search menu for visible columns when adding a new filter.
* T89273: Possibly a "Range" operation.
Differential Revision: https://developer.blender.org/D10959
This code was disabled in 2.8 and all other associated code/comments
have been removed/cleared. These rna properties have been replaced with
`seq_prev_type`
Removal of unused local variable. Calculation of underline thickness
no longer needed with change to text output of underscore character.
Introduced in aee04d4960
Differential Revision: https://developer.blender.org/D11641
Currently B-Bone scaling can only be controlled via their
properties, thus requiring up to 8 drivers per joint between
B-Bones to transfer scaling factors from the handle bone.
A Scale Easing option is added to multiply the easing value
by the Y scale channels to synchronize them - this produces a
natural scaling effect where both the shape of the curve and
the scale is affected.
In addition, four toggles are added for each handle, which
multiply each of the X, Y, Z and Ease values by the matching
Local Scale channel of the handle bone, thus replacing trivial
drivers. The Scale Easing option has no effect on this process
since it's easy to just enable both Length and Ease buttons.
Differential Revision: https://developer.blender.org/D9870
Implement actual behavior for the B-Bone Y Scale channels added
to DNA and UI in the previous commit in addition to the existing
X and Z Scale inputs.
The two length scale inputs control the ratio between the lengths
of the start and end segments of the bone: although for convenience
two inputs are provided, the whole chain is still uniformly scaled
to fit the curve.
Differential Revision: https://developer.blender.org/D9870
In addition to the base bone transformation itself, B-Bones have
controls that affect transformation of its segments. For rotation
the features are quite complete, allowing to both reorient the
Bezier handles via properties, and to control them using custom
handle bones. However for scaling there are two deficiencies.
First, there are only X and Y scale factors (actually X and Z),
while lengthwise all segments have the same scaling. The ease
option merely affects the shape of the curve, and does not cause
actual scaling.
Second, scaling can only be controlled via properties, thus
requiring up to 6 drivers per joint between B-Bones to transfer
scaling factors from the handle bone. This is very inefficient.
Finally, the Z channels are confusingly called Y.
This commit adds a B-Bone Y Scale channel and extra B-Bone flag
fields to DNA with appropriate versioning (including for F-Curves
and drivers) in preparation to addressing these limitations.
Functionality is not changed, so the new fields are not used
until the following commits.
Differential Revision: https://developer.blender.org/D9870
When drawing mnemonic underlines for hotkeys, use text output of
underscore character instead of direct drawing a line. Otherwise these
are not visible in dialog buttons.
Introduced in 0fcc063fd9
Differential Revision: https://developer.blender.org/D11641
Reviewed by Campbell Barton
Using an ampersand here is more semantically correct. A slash indicates "or" while an ampersand indicates "and".
An ampersand here is best because the view type shows both the Sequencer and the Preview.
When having multiple materials in a mesh the triangles are sorted based
on material index. This sorting is done single threaded, but needs two
loops over the data. One to count the bucket size and the second one to
add the triangles to the right position in the buckets.
This patch will do the counting in a multithreaded approach that would
speed up the cache creation. It has been measured that this part is the
most blocking part of the cache creation.
Reviewed By: mano-wii
Differential Revision: https://developer.blender.org/D11615
During selections the total selection is refreshed at the end. This
process was done single threaded. This patch will do a parallel iter
approach.
Master: 0.043612s Threaded 0.017964s.
Master: {F10179586}
This patch: {F10179587}
Reviewed By: mano-wii
Differential Revision: https://developer.blender.org/D11622
`ED_fileselect_get_asset_params` would only return actual data pointer
when file browser is in ASSET mode.
Calling that whole section only makes sense if filebrowser is in asset
mode anyway.
Regression introduced in rBf6c5af3d4753 I think.
@Severin committing this fix now as this is a fairly critical bug for
the studio, feel free to revert and do proper fix if this one is not the
best solution.
This patch ensures that selection mode flushing updates total selection
counts internally. This reduces recounting when we are sure that the
input total selection counts were up to date.
For example for circle selection the total selection counts were
correct. But during flushing the selection could have been changed and
therefore the selection was always recounted.
This increased the performance on selected system from 6.90 FPS to 8.25
FPS during circle selection operations.
Before: {F10179981}
After: {F10179982}
Reviewed By: mano-wii
Differential Revision: https://developer.blender.org/D11647
When an object, whose mesh gets loaded from Alembic, gets animated in
Blender and the Alembic CacheFile datablock also gets animated, editing
keyframes causes both datablock to be re-copied for evaluation. This
caused a threading issue and a double-free of some memory. This is fixed
by expanding the scope of the spin lock in
`BKE_cachefile_reader_free()`.
Contributed by @erik85 in D11400. The idea from this patch was placed in
a more generic context: A new FOR macro has been added that loops
over the neighbors of a cell within a given radius.
This matches BMesh which also has tessellation in it's own file.
Using a separate file helps with organization when
extracting code into smaller functions.
The //Raycast// node intersects rays from one geometry onto another.
It computes hit points on the target mesh and returns normals, distances
and any surface attribute specified by the user.
A ray starts on each point of the input //Geometry//. Rays continue
in the //Ray Direction// until they either hit the //Target Geometry//
or reach the //Ray Length// limit. If the target is hit, the value of the
//Is Hit// attribute in the output mesh will be true. //Hit Position//,
//Hit Normal//, //Hit Distance// and //Hit Index// are the properties of the
target mesh at the intersection point. In addition, a //Target Attribute//
can be specified that is interpolated at the hit point and the result
stored in //Hit Attribute//.
Docs: D11620
Reviewed By: HooglyBoogly
Differential Revision: https://developer.blender.org/D11619
This node creates splines with more control points in between the
existing control points. The point is to give the splines more
definition for further tweaking like randomization with white noise,
instead of deforming a resampled poly spline with a noise texture.
For poly splines and NURBS, the node simply interpolates new values
between the existing control points. However, for Bezier splines,
the result follows the existing evaluated shape of the curve, changing
the handle positions and handle types to make that possible.
The number of "cuts" can be controlled by an integer input, or an
attribute can be used. Both spline and point domain attributes are
supported, so the number of cuts can vary using the value from the
point at the start of each segment.
Dynamic curve attributes are interpolated to the result with linear
interpolation.
Differential Revision: https://developer.blender.org/D11421
The minimum twist mode is important because it allows creating
normals without sudden changes in direction. The disadvantage
of minimum twist normals is that the normals depend on all control
points. So changing one control point can change the normals
everywhere. The computed normals do not match the existing
code exactly, although they do match quite well on non-cyclic and
on some cyclic curves. I also noticed that the existing implementation
has some fairly simple failure cases that I haven't found in the new
implementation so far.
Differential Revision: https://developer.blender.org/D11621
This makes the parts where a node is locked more explicit. Also, now the thread
is isolated when the node is locked. This prevents some kinds of deadlocks
(which haven't happened in practice yet).
- BKE_mesh_copy_parameters_for_eval to be used for evaluated meshes only
as it doesn't handle ID user-counts.
- BKE_mesh_copy_parameters is a general function for copying parameters
between meshes.
Refactor function `freeSeqData` so it is readable.
One strip can have multiple transform operations defined. To prevent
processing strip multiple times, build `SeqCollection` and use
sequencer iterator instead of iterating `TransData` directly.
No functional changes.
Differential Revision: https://developer.blender.org/D11618
Reduce complexity of sequencer transform code by removing recursivity.
This is possible by treating meta strips (mostly) as any other strip and
containing all transform code within SEQ_ functions.
Unfortunately internally meta strips still require special treatment,
but all complexity from code all over transform code seems to be
possible to contain within one function.
Functional change:
Previously adjusting handle of single image strip moved animation.
Now animation is not moved, which is behavior for all other strips.
Reviewed By: sergey, mano-wii
Differential Revision: https://developer.blender.org/D11493
Starts scrolling when dragging a node or node link and going outside the current window.
Largely copied from the VIEW2D_OT_edge_pan operator.
Edge panning operator customdata and supporting functions now in
UI_view2d.h, so they could be used by operators in other editor
libraries. The VIEW2D_OT_edge_pan operator also uses this customdata and
shared functions now. Operators properties can be used to configure
edge panning margins and speed for each use case, rather than using
hardcoded values.
The speed function for edge panning has been tweaked somewhat:
* "Speed per pixel" has been replaced with a "speed ramp" distance.
This is more intuitive and also creates an upper bound for the speed,
which can otherwise become extreme with large cursor distance.
* "Max speed" is reached at the end of the speed ramp.
* Padding the region inside and outside is applied as before, but both
values are operator properties now.
Node transform operator also supports edge panning. This requires
an offset for changes in the view2d rect, otherwise nodes are "stuck"
to the original view.
Transform operator had cursor wrapping categorically enabled, but this
gets quite confusing with the edge scrolling mechanism. A new TransInfo
option T_NO_CURSOR_WRAP has been introduced to disable this behavior.
The double negative is a bit annoying, but want to avoid affecting the
existing transform modes, so by default it should still set the
OP_IS_MODAL_GRAB_CURSOR flag (which then sets the WM_CURSOR_WRAP_XY
flag during modal execution).
Reviewed By: HooglyBoogly, JacquesLucke
Differential Revision: https://developer.blender.org/D11073
This allows line art to run only once for each modifier stacks,
with an option to toggle a specific line art modifier should
use cache or re-do their own calculations.
Reviewed By: Sebastian Parborg (zeddb), Hans Goudey (HooglyBoogly)
Differential Revision: https://developer.blender.org/D11291
This makes it easier to use task isolation in c++ code.
Previously, one either had to check `WITH_TBB` (possibly indirectly
through `WITH_OPENVDB`) or one had to use the C function which
is less convenient.
This namespace groups threading related functions/classes. This avoids
adding more threading related stuff to the blender namespace. Also it
makes naming a bit easier, e.g. the c++ version of BLI_task_isolate could
become blender::threading::isolate_task or something similar.
Differential Revision: https://developer.blender.org/D11624
Similar to what we do for constraints and modifiers, except that
currently adding or editing shaderfx in liboverride objects is
completely unsuported.
Fix T88974.
Used to detect if a shaderfx is purely local, or comes from linked data,
in case of a liboverride.
Not actually used yet since we do not currently support adding
shaderfx's to overrides, but will be in the future, and matches
constraints and modifiers code.
Mainly:
* Make `ED_operator_object_active_editable_ex` properly report poll
messages on failure.
* Add `ED_operator_object_active_local_editable_posemode_exclusive` for
bone constraints requiring pure local Object (non-override one).
* General cleanup and adding more poll messages on failures.
On selecting a multi-layer image with a combined pass, a "Combined"
socket is created and default combined pass socket "Image" is
disabled by setting `SOCK_UNAVAIL` flag. When converting into
operations, `ImageNode` converts alpha socket on finding any socket with
a combined pass without checking the flag.
Since commit rB93e2491ee724 an assertion fails when mapping sockets
twice because now map `add_new` is used.
Reviewed By: jbakker
Differential Revision: https://developer.blender.org/D11566
Adds full frame implementation to Value node operation.
No functional changes.
Reviewed By: jbakker
Differential Revision: https://developer.blender.org/D11594
This adds support for importing UV sets which are defined per vertex,
instead of per face corners. Such UV sets can be generated when the
mesh is split according to UV islands, or when there is only one UV
island, in which cases only a single UV value can be stored per
vertex since vertices will never be on a seam.
- Include all objects in pose mode.
- Show the number of objects in pose mode.
- Show the number of objects in edit mode for all types of objects
(not just meshes).
This adds support for importing UV sets which are defined per vertex,
instead of per face corners. Such UV sets can be generated when the
mesh is split according to UV islands, or when there is only one UV
island, in which cases only a single UV value can be stored per
vertex since vertices will never be on a seam.
Reviewed By: sybren
Differential Revision: https://developer.blender.org/D11584
Implementation of T86970. This node takes a geometry input with
multiple components and outputs them by component type. Meshes,
Curves, and Point Clouds support combining multiple input instances,
while volumes will only output the first volume component input until
suitable instance realization for multiple volumes is finished.
When direct geometry instancing is implemented it will be possible to
avoid realizing instances in this node.
Differential Revision: https://developer.blender.org/D11577
Creating a new full screen area had it's area initialized as empty,
updating the screen then set the area to a 3D view (as a fallback),
before the actual area type was set.
This made setting the intended space-type run the 3D views exit callback
on a 3D view without a View3D struct allocated, which the exit callback
needed to account for.
Resolve by calling ED_screen_change after the area type has been set.
This patch improves the 3DView statistics overlay to show LOCAL stats
while in local view. This means the stats can vary between 3DViews and
the statusbar when views are in local view, but this gives a much more
accurate count of the objects, and their components, that you are
directly working with rather than just scene values.
Differential Revision: https://developer.blender.org/D8883
Reviewed by Campbell Barton
Versioning code for converting strip offset property doesn't work, when
property was animated and disabled or when crop was used.
When offset property is animated and offset is enabled, animation is
converted to be used with new transform design. When offset is disabled,
animation is left untouched. New transform design doesn't have option
to disable offset, and therefore old unconverted animation is used
instead of converted static value.
Remove animation from propery if it was unused.
Another issue was that both X and Y offset animation was being corrected
by factor caluclated for X channel.
Reviewed By: sergey
Differential Revision: https://developer.blender.org/D11370
Use lookup string callback function for `sequences_all` RNA property
`rna_SequenceEditor_sequences_all_lookup_string` using a GHash for faster lookups.
When names are changed or strips are added/removed the lookup is tagged invalid.
The next time the lookup is used it will rebuild it.
Reviewed By: sergey, jbakker
Differential Revision: https://developer.blender.org/D11544
Just like the way we often have a choice between an attribute input and
a single float, this adds the ability to choose between attribute and
integer input sockets, useful for D11421.
There is no particular reason these two functions shouldn't be used
outside of the bezier spline implementation since they don't do anything
particularly controversial.
Create maps that specify which batches have vbo or ibo as a reference
and use these maps to discard batches along with buffers.
Differential Revision: https://developer.blender.org/D11588
Adds `source/blender/blendloader/intern/versioning_common.cc` and
`versioning_common.h` for version independent versioning functions.
I only placed `do_versions_add_region_if_not_found()` in there for now.
`blo_do_version_old_trackto_to_constraints()` could also be added, but
that's so old, I prefer keeping that in `versioning_legacy.c`.
Widget drawing code already supported drawing right-aligned, grayed out
shortcut strings. This patch generalizes things a bit so this can also
be used to draw other hints in the same way. There have been a few
instances in the past where this would've been useful, D11046 being the
latest one.
Note that besides some manual regression testing, I didn't check if this
works yet, as there is no code actually using it (other than the
shortcuts). Can be checked as part of further development for D11046.
A possible further improvement would be providing a way to define how
clipping should be done. E.g. sometimes the right-aligned text should be
clipped first (because it's just a hint), in other cases it should be
left untouched (like current code explicitly does it for shortcuts).
Removes the `UI_BUT_HAS_SHORTCUT` flag, which isn't needed anymore.
After looking into task isolation issues with Sergey, we couldn't find the
reason behind the deadlocks that we are getting in T87938 and a Sprite Fright
file involving motion blur renders.
There is no apparent place where we adding or waiting on tasks in a task group
from different isolation regions, which is what is known to cause problems. Yet
it still hangs. Either we do not understand some limitation of TBB isolation,
or there is a bug in TBB, but we could not figure it out.
Instead the idea is to use isolation only where we know we need it: when
holding a mutex lock and then doing some multithreaded operation within that
locked region. Three places where we do this now:
* Generated images
* Cached BVH tree building
* OpenVDB lazy grid loading
Compared to the more automatic approach previously used, there is the downside
that it is easy to miss places where we need isolation. Yet doing it more
automatically is also causing unexpected issue and bugs that we found no
solution for, so this seems better.
Patch implemented by Sergey and me.
Differential Revision: https://developer.blender.org/D11603
When using multiple materials in a single mesh the most time is spend in
counting the offsets of each material for the sorting.
This patch moves the counting of the offsets to render mesh data and
caches it as long as the geometry doesn't change.
This patch doesn't include multithreading of this code.
Reviewed By: mano-wii
Differential Revision: https://developer.blender.org/D11612
Clear the runtime data structs instead of individual members,
this simplifies adding new runtime members as there are at least
two places they would need to be cleared.
Resolves error in D8883.
This commit lets the join geometry node transfer dynamic attributes
to the result, the same way that point cloud and mesh attributes are
joined. The implementation is different though, because of an
optimization implemented for curves to avoid copying splines.
The result attribute is added with the highest priority domain (points
over splines), and the highest complexity data type. If one curve had
the attribute on the spline domain but not others, the point domain
values will be used.
Generally this is a bit lower level than I would have liked this code
to be, but should be efficient, and it's really not too complicated.
Differential Revision: https://developer.blender.org/D11491
Instead of building a set and then determining the final domain and
type for every attribute separately in the loop, construct a map with
the necessary data in the first place. This is simpler and should be
slightly more efficient.
Split from D11491
This is an implementation of T88722. It accepts a curve object and
for each spline, reverses the order of the points and all attributes.
This is more of a foundational node to support other nodes in the
future (like curve deform)
Selection takes spline domain attributes to determine which splines
are selected. If no selection is present all splines are reversed.
Differential Revision: https://developer.blender.org/D11538
This node implements the second option of T87429, creating points
along the input splines with the necessary evaluated information
for instancing: `tangent`, `normal`, and `rotation` attributes.
All generic curve point and spline attributes are copied to the
result points as well.
The "Count" and "Length" methods are just like the current options
in the resample node, but the output is points instead of a curve.
The "Evaluated" method uses the points you see on the curve directly,
and therefore should be the fastest.
The rotation data is retrieved from a transform matrix built with the
same method that the curve to mesh node uses. The radius attribute is
divided by 10 so the points don't look absurdly huge in the viewport.
In the future that could be an option.
For the implementation, one thing that could use an improvement
is the amount of temporary allocations while resampling to evaluated
points before the final points. I expect that reusing a buffer for
each thread would give a nice improvement.
Differential Revision: https://developer.blender.org/D11539
This patch allows Windows users to specify that their current blender
installation should be used to create thumbnails and be associated
with ".blend" files. This is done from Preferences / System. The only
way to do this currently is from the command-line and this is sometimes
inconvenient.
Differential Revision: https://developer.blender.org/D10887
Reviewed by Brecht Van Lommel
If the last decoded frame had the same timestamp as the GOP current
packet, then we would skip over this frame when fast forwarding and we
would seek until the end of the file.
This would could only be triggered reliably in single threaded mode.
Reviewed By: Richard Antalik
Differential Revision: http://developer.blender.org/D11601
The warning would appear when using the `ENUM_OPERATORS()` macro inside
of an `extern "C"` block.
Didn't cause a warning in master currently, but in the
`asset-browser-poselib` branch.
After macro expansion, there would be C++ code in code with C linkage
(`extern "C"`). So make sure the expanded C++ code always uses C++
linkage.
The syntax used is totally C++ compliant: the C++ standard requires that
in such nested linkage specifications, the innermost one determines the
linking language (e.g. see
https://timsong-cpp.github.io/cppwp/n4659/dcl.link#4).
On the Windows platform there will be some errors printed to the
console if the user's thumbnail cache folder doesn't already exist.
While creating those folders there is an attempt to do so multiple
times and so we get errors when trying to create when exists. This
is caused by paths that have hard-coded forward slashes, which causes
our path processing routines to not work correctly. This patch defines
those paths using platform-varying separator characters.
Differential Revision: https://developer.blender.org/D11505
Reviewed by Brecht Van Lommel
Support calculating face normals when tessellating. When this is done
before updating vertex normals it gives ~20% performance improvement.
Now vertex normal calculation only needs to perform a single pass on the
mesh vertices when called after tessellation.
Extended versions of normal & looptri update functions have been added:
- BM_mesh_calc_tessellation_ex
- BM_mesh_normals_update_ex
Most callers don't need to be aware of this detail by using:
- BKE_editmesh_looptri_and_normals_calc
- BKE_editmesh_looptri_and_normals_calc_with_partial
- EDBM_update also takes advantage of this,
where calling EDBM_update with calc_looptri & calc_normals
enabled uses the faster normal updating logic.
Bypass stored edge-vectors for ~16% performance gains.
While this increases unit-length edge-vector calculations by around ~4x
the overhead of a parallel loop over all edges makes it worthwhile.
Note that caching edge-vectors per-vertex performs better and may be
worth investigating further, although in my tests this increases code
complexity with barley measurable benefits over not using cache at all.
Details about performance and possible optimizations are noted in
bm_vert_calc_normals_impl.
Rename function EDBM_update_generic to EDBM_update, use a parameters
argument for better readability.
Also add calc_normals argument, which will have benefits when
calculating normals and tessellation together is optimized.
For some custom rendering engines it's advantageous not to write the image files to disk.
An example would be a network rendering engine which does it's own image writing.
This feature is only supported when bl_use_postprocess is also disabled, since render
engines can't influence the saving behavior of the sequencer or compositor.
Differential Revision: https://developer.blender.org/D11512
Better to avoid explicit vectors components direct manipulation when a
generic operation for whole vector exists, if nothing else it avoids
potential mistakes in indices.
Those were caused by various tools used on degenerate geometry, see
T79775.
Note that fixes are as low-level as possible, to ensure they cover as
much as possible of unreported issues too.
We still probably have many more of those hidden in BLI_math though.
This commit does two things:
* Disallows creating more than one link from one socket to a multi socket input.
* Properly count links if there happen to be more than one link between the same sockets.
The new link counting should also be more efficient asymptotically.
Differential Revision: https://developer.blender.org/D11570
- The argument with named with an `r_` prefix when it was in fact
also read from.
- The argument passed in had to be 'psys->clmd->hairdata',
if it was not - the function would not worked.
When moving a linked collection, we seem to only receive a depsgraph update
for an empty object so the Blender synchronization cannot discriminate it
and tag the object(s) (light or geometry) for an update through
id_map.set_recalc.
This missing transform update only affects lights since we do not check
manually if the transformations were modified like we do for objects.
To fix this, add a check to see if the transformation is different provided
that a light was already created.
Reviewed By: brecht
Maniphest Tasks: T88515
Differential Revision: https://developer.blender.org/D11574
This patch improves the positioning of child windows when on monitors
that are arranged vertically (any above any other). When calculating a
window position in Ghost coordinates from GL coordinates we were using
monitor height, which can give incorrect values when desktop is taller
than any single monitor. So use desktop height instead.
See D10637 for more details and examples.
Differential Revision: https://developer.blender.org/D10637
Reviewed by Brecht Van Lommel
This patch makes the "Render" window a top-level window, not a child of
the main window, which was the case in blender versions prior to 2.93.
This means it is no longer "on top", nor is the icon grouped on the
taskbar in the same way, but you can Alt-Tab between it and the main
window. This change only affects the Windows platform as the other
platforms behave this way.
See D11576 for links to negative feedback that prompts this change.
Differential Revision: https://developer.blender.org/D11576
Reviewed by Brecht Van Lommel
This moves the flash on mode transfer effect option from the overlays to
an operator property of the mode transfer operator.
- This effect is intended to show the target object when no overlays or
a minimal set of overlays is enabled. Making it part of the whole set of
overlays invalidates this use case.
- The effect is not intended to be configurable per viewport, it should
be a global option.
The effect is still implemented using the overlay engine (instead of a
draw modal callback) due to performance and drawing artifacts. Having it
implemented as an overlay with runtime timer data in the objects makes
also possible to run multiple animations at the same time without any
visual glitches.
Reviewed By: campbellbarton, JulienKaspar
Differential Revision: https://developer.blender.org/D11519
Note: Linking in this case as in link vs. append. Easily confused with linking
a data-block to multiple usages (e.g. single material used by multiple
objects).
Adds a drop-down to the Asset Browser header to choose between Link and Append.
This is probably gonna be a temporary place, T54642 shows where this could be
placed eventually.
Linking support is crucial for usage of the asset browser in production
environments. It just wasn't enabled yet because a) the asset project currently
focuses on single user, not production assets, and b) because there were many
unkowns still for the workflow that have big impact on production use as well.
With the recently held asset workshop I'm more confident with enabling linking,
as design ideas relevant to production use were confirmed.
Differential Revision: https://developer.blender.org/D11536
Reviewed by: Bastien Montagne
The operator run when dropping objects would duplicate the dropped object and
place that in the scene, even though that was just appended. Addressed by
making the duplication optional for the operator. If the duplication is not
requested, the object is just added to the scene (if needed), repositioned
based on the drop location and selected (deselecting other objects).
This makes the operator work as expected when using it to drop assets.
Reviewed as part of https://developer.blender.org/D11536.
Reviewed by: Bastien Montagne
Fix segmentation fault that can occur when reordering animation
channels.
Under some specific conditions, the list "act->curves" is empty in the
"join_groups_action_temp" function. In particular, this happens when a
scene contains an action that has not been pushed down, and with no
keyframe in it.
Reviewed By: sybren
Differential Revision: https://developer.blender.org/D11569
While this preprocessing does take some time upfront,
it avoids longer lookup later on, especially as nodes get
more sockets.
It's probably possible to make this more efficient in some cases
but this is good enough for now.
Scanline processor did its own heurestic what didn't scale well when
having a multiple cores. In stead of using our own code this patch will
leave it to TBB to determine how to split the scanlines over the
available threads.
Performance of the IMB_transform before this change was 0.002123s, with
this change 0.001601s. This change increases performance in other areas
as well including color management conversions.
Reviewed By: zeddb
Differential Revision: https://developer.blender.org/D11578
When using large sequences including audio the drawing of the audio on
top of the strip takes a lot of time. This effects the playback
performance heavily.
During the animation playback performance there was a solution for this
by only drawing the playhead overlay. This was reverted for the sequence
editor as it didn't update the color strips when they were animated.
This patch checks if there are animated color strips if so the full
screen is redrawn, otherwise only the playhead is redrawn.
Reviewed By: ISS
Differential Revision: https://developer.blender.org/D11580
One drawback to trying to predict the number of threads that will be
used in the `task_graph` is that we are only sure of the number when the
threads are running.
Using `BLI_task_parallel_range` allows the driver to
choose the best thread distribution through `parallel_reduce`.
The benefit is most evident on hardware with fewer cores.
This is the result on an 4-core laptop:
||before:|after:
|---|---|---|
|large_mesh_editing:|Average: 5.203638 FPS|Average: 5.398925 FPS
||rdata 15ms iter 43ms (frame 193ms)|rdata 14ms iter 36ms (frame 187ms)
Differential Revision: https://developer.blender.org/D11558
This is an adaptation of {D11488}.
A disadvantage of manually setting the iter ranges per thread is that
we don't know how many threads are running in the background and so we
don't know how to best distribute the ranges.
To solve this limitation we can use `parallel_reduce` and thus let the
driver choose the best distribution of ranges among the threads.
This proved to be especially beneficial for computers with few cores.
**Benchmarking:**
Here's the result on an 4-core laptop:
||master:|PATCH:
|---|---|---|
|large_mesh_editing:|Average: 5.203638 FPS|Average: 5.398925 FPS
||rdata 15ms iter 43ms (frame 193ms)|rdata 14ms iter 36ms (frame 187ms)
Here's the result on an 8-core PC:
||master:|PATCH:
|---|---|---|
|large_mesh_editing:|Average: 15.267482 FPS|Average: 15.906881 FPS
||rdata 9ms iter 28ms (frame 65ms)|rdata 9ms iter 25ms (frame 63ms)
|large_mesh_editing_ledge: |Average: 15.145966 FPS|Average: 15.520474 FPS
||rdata 9ms iter 29ms (frame 65ms)|rdata 9ms iter 25ms (frame 64ms)
|looptris_test:|Average: 4.001917 FPS|Average: 4.061105 FPS
||rdata 12ms iter 90ms (frame 236ms)|rdata 12ms iter 87ms (frame 230ms)
|subdiv_mesh_cage_and_final:|Average: 1.917769 FPS|Average: 1.971790 FPS
||rdata 7ms iter 37ms (frame 261ms)|rdata 7ms iter 31ms (frame 258ms)
||rdata 7ms iter 38ms (frame 252ms)|rdata 7ms iter 33ms (frame 249ms)
|subdiv_mesh_final_only:|Average: 6.387240 FPS|Average: 6.591251 FPS
||rdata 3ms iter 25ms (frame 151ms)|rdata 3ms iter 16ms (frame 145ms)
|subdiv_mesh_final_only_ledge:|Average: 6.247393 FPS|Average: 6.596024 FPS
||rdata 3ms iter 26ms (frame 158ms)|rdata 3ms iter 16ms (frame 148ms)
**Notes:**
- The improvement can only be noticed if all extracts are multithreaded.
- This patch touches different areas of the code, so it can be split into another patch if the idea is accepted.
These screenshots show how threads behave in a quadcore:
Master:
{F10164664}
Patch:
{F10164666}
Differential Revision: https://developer.blender.org/D11558
ffmpeg_generic_seek_workaround did work properly and our start pts
calculation was wrong.
Reviewed By: Richard Antalik
Differential Revision: http://developer.blender.org/D11562
Because of the added sanity checks in rB14508ef100c9 (D11492), seeking
in proxies would not work correctly any more. This is because it wasn't
working as intended before, but in most cases this wouldn't be
noticeable. However now when the sanity checks are tripped it is very
noticeable that something is wrong
The indexer tried to use dts values for time stamps when we used pts in
our decode functions to get the time positions. This would make it
start in the wrong GOP frames when searching. Now that we enforce no
crossing of GOP frames when decoding after seek, this would lead to
issues.
Now we correctly use pts (or dts if pts is not available) and thus we
don't have any seeking issues because of time stamp format missmatch.
Reviewed By: Richard Antalik
Differential Revision: http://developer.blender.org/D11561
When sampling ImBuf can be a char or a float buffer. Current sampling
functions added overhead by checking which kind of buffer was passed
every pixel that was sampled. When performing image processing this
check can be removed outside the inner loop adding 5% of performance
increase in the `IMB_transform` operator.
Remap {key Alt I} from `anim.delete_key_v3d` to `anim.delete_key`. This
makes it keyframe removal symmetrical with keyframe insertion ({key I}).
Both the default keymap {key Alt I} and the Industry Compatible keymap
{key Alt S} have been updated.
Reviewed By: sybren, #animation_rigging
Maniphest Tasks: T88068
Differential Revision: https://developer.blender.org/D11528
Inside the sequencer the cropping and transform of images/buffers were
implemented locally. This reduced the optimizations that a compiler
could do and added confusing code styles. This patch adds
`IMB_transform` to reduce the confusion and increases compiler
optimizations as more code can be inlined and we can keep track of
indices inside the inner loop.
This increases end-user performance by 30% when playing back aa video
in VSE.
Reviewed By: ISS, zeddb
Differential Revision: https://developer.blender.org/D11549
This patch changes occurrences of percentage to factor.
There are some usages of percentage left in there on purpose.
They are distinguished as follows:
- factor is 0-1 float
- percentage is 0-100 int
Ref D11361
Reviewed by: sybren, campbellbarton
Prepare node for conversion to Geometry Nodes.
There should be no functional changes.
Reviewed By: HooglyBoogly
Differential Revision: https://developer.blender.org/D11506
When using non-default system separator in filename path, code would end up
with an absolute path mixing regular and alternative separator,
confusing the rest of the path manipulations later on.
So this commit add proper replacements of alternative separators, and
path normalization.
Splitting out thread safe iteration logic means regular iteration
isn't checking for the thread-safe pointer each step.
This gives a small but measurable overall performance gain of 2-3%
when redrawing a high-poly mesh.
Ref D11564
Reviewed By: mont29
This change was prompted by D6408 which moves thumbnail extraction into
a shared function that happens use these endian defines but only links
blenlib.
There is no need for these defines to be associated with globals
so move into their own header.
Allows to define properties which will have proper units displayed
in the interface. The internal storage is expected to be seconds
(which matches how other times are stored in Blender).
Is not immediately used in Blender, but is required for the upcoming
feature in Cycles X (D11526)
The naming does not sound very exciting, but can't think of anything
better either.
For test it probably easiest to define FloatProperty with subdtype
of TIME_ABSOLUTE.
Differential Revision: https://developer.blender.org/D11532
Previously, a vertex from destination mesh would always be added to all
transferred vgroup (with a 0.0 weight), even if none of its matching
sources in source mesh belonged to the matching source vgroups.
Now a destination vertex is only added to a given destination vgroup if
at least one of its source vertices belong to the matching source
vgroup.
Issue found and initial investigation by @pls in D11524, thanks!
Doxygen by default leaves out any functions inside
#ifdef blocks that it thinks are disabled.
This change adds a DOXYGEN symbol, so you can
still get the functions included in the
documentation even if the #ifdef would
have normally excluded them.
before
#if defined(_WIN32)
after
#if defined(_WIN32) || defined(DOXYGEN)
Patch provided by Campbell Barton on chat.
TBBmalloc_proxy already takes care of any allocations
being done from MSVC compiled code, some of the dependencies
like GMP cannot be build with MSVC and we have to use
mingw to build them. mingw however links against the older
msvcrt.dll for its allocation needs, which TBBMallocProxy
does not hook.
GMP has an option to supply your own allocation functions
so we can still manually redirect them to TBBMalloc.
In a test-file with a boolean geometry node, this patch
uses 32s effective CPU time compared to 52s before.
Differential Revision: https://developer.blender.org/D11435
Reviewed by Campbell Barton, Ray Molenkamp
Remove an unnecessary call to pose_slide_mouse_update_percentage
That call was overriding the typed value
Reviewed By: #animation_rigging, Sybren A. Stüvel
Differential Revision: https://developer.blender.org/D11395
Ref D11395
To draw the overlay stats in columns the strings must be measured to
find the longest one. In some circumstances this measurement can be
incorrect. We draw the text with a specific size yet do not explicitly
set the size before calling BLF_size. This patch properly sets the size
so that the measurement will match what will be used for output.
See T88799 for examples of measurement failure.
Differential Revision: https://developer.blender.org/D11504
Reviewed by Hans Goudey
First, expand on the interpolation to evaluated points with a templated
helper function, and a function that takes a GSPan. Next, add a set of
functions to `Spline` for interpolating at arbitrary intervals between
the evaluated points. The code for doing that isn't that complicated
anyway, but it's nice to avoid repeating, and it might make it easier
to unroll the special cases for the first and last points if we require
the index factors to be sorted.
This commit adds a node to output the convex hull of any input geometry
as a mesh, which is an enclosing geometry around a set of points.
All geometry types are supported, besides volumes.
The code supports operating on instances to avoid copying all input
geometry before the operation. The implementation uses the same backend
as the operation in edit mode, but uses Mesh directly instead of BMesh.
Attribute transfer is not supported currently, but would be a point of
improvement for the future if it can work in a predictable way on
different geometry input types.
Differential Revision: https://developer.blender.org/D10925
Note that these changes are limited simple cases as these kinds of
changes could allow for errors when refactoring code when the known
state is not so obvious.
Use BM_iter_parallel for face tessellation, this gives around
6.5x speedup for BM_mesh_calc_tessellation on high poly meshes in my
tests, although exact speedup depends on available cores.
Support thread local storage for BLI_task_parallel_mempool,
as well as support for the reduce and free callbacks.
mempool_iter_threadsafe_* functions have been moved into a private
header thats only shared between task_iterator.c and BLI_mempool.c
so the TLS can be made part of the iterator array without having to
rely on passing in struct offsets.
Add test task.MempoolIterTLS that ensures reduce and free
are working as expected.
Reviewed By: mont29
Ref D11548
Also modified existing utility functions to take
care of the new surface normal interpolation and so on.
Reviewed By: Antonio Vazquez (antoniov)
Differential Revision: https://developer.blender.org/D11543
While this should not be allowed in general, there are some cases
(library overrides at least) where supporting this is mandatory.
Further more, comment stating that this could crash is from 2011, could
not reproduce any issue with current code. Commit comment was referring
to undo/redo, but in use cases here those should not affect things.
Note that in general, such relatively high-level checks should be
handled by high-level, close to user code (like in ED area e.g.), not in
low-level BKE code anyway.
This patch adds a specific extraction method when the mesh has only
one material. This method is multi-threaded.
There is a trade-off in this patch as the ibo isn't compressed (it adds
restart indexes for hidden faces). So it depends if threading is faster
than the additional GPU buffer upload.
# Subdivided cube
I used a cube subdivided 7 times, modifiers applied. that gives around 400000 faces.
The test is selecting some vertices and move them. During this test the next buffers are updated on each frame:
* vbo.pos_nor
* vbo.lnor
* vbo.edit_data
* ibo.tris
* ibo.points
System info:
|platform| Linux-5.11.0-7614-generic-x86_64-with-glibc2.33|
| renderer| AMD SIENNA_CICHLID (DRM 3.40.0, 5.11.0-7614-generic, LLVM 11.0.1)|
|vendor| AMD|
|version| 4.6 (Core Profile) Mesa 21.0.1|
|cpu| Intel(R) Core(TM) i7-6700 CPU @ 3.40GHz|
|compiler| gcc version 10.3.0|
Timing have been measured using DEBUG_TIME in `draw_cache_extract_mesh`.
master: `rdata 8ms iter 45ms (frame 153ms)`
this patch `rdata 6ms iter 36ms (frame 132ms)`
Reviewed By: mano-wii
Maniphest Tasks: T88352
Differential Revision: https://developer.blender.org/D11290
This commit adds a flag to disable displaying some socket labels which
just redundant eye sores. We still want to have a label, because that
information can potentially be accessed elsewhere in the UI.
The flag is used in a few geometry nodes.
Differential Revision: https://developer.blender.org/D11540
This is an optimization, but the difference is still not that
significant as some extractions are still done in single thread.
**Benchmarking**
||before:|after:
|---|---|---|
|large_mesh_editing:|Average: 14.246502 FPS|Average: 15.438118 FPS
||rdata 9ms iter 31ms (frame 69ms)|rdata 9ms iter 27ms (frame 65ms)
|large_mesh_editing_ledge: |Average: 14.913622 FPS|Average: 15.856538 FPS
||rdata 9ms iter 30ms (frame 67ms)|rdata 9ms iter 26ms (frame 63ms)
|looptris_test:|Average: 3.970774 FPS|Average: 4.095200 FPS
||rdata 11ms iter 90ms (frame 235ms)|rdata 12ms iter 87ms (frame 229ms)
Reviewed By: jbakker
Differential Revision: https://developer.blender.org/D11467
A single threaded task with thread data over 8192 bytes would leak.
While this didn't happen in practice, it could cause issues in the
future.
The free call for `task_parallel_iterator_do` wasn't running
if callbacks weren't set, also not an issue in practice but avoids
potential problems in the future too.
Changes introduced in commit rBe9f2f17e8518
can create different render results when there is
a Math or Mix operation after TextureOperation
on tiled execution model.
This is due to WriteBufferOperation forcing a single pixel
resolution when these operations use a preferred
resolution of 0 to check if their inputs have resolution.
Fixing this behaviour creates different renders too.
This patch keeps previous tiled implementation and
adds the new implementation only for full frame execution.
Reviewed By: Jeroen Bakker (jbakker)
Differential Revision: https://developer.blender.org/D11546
In order to reduce stack size this patch converts full frame
recursive methods into iterative.
- No functional changes.
- No performance changes.
- Memory peak may slightly vary depending on the tree because
now breadth-first traversal is used instead of depth-first.
Tests in D11113 have same results except for test1 memory peak:
360MBs instead of 329.50MBs.
Reviewed By: Jeroen Bakker (jbakker)
Differential Revision: https://developer.blender.org/D11515
CMake builder and install deps changes, precompiled libraries are still to be
committed.
Ref T88438, T88434
Differential Revision: https://developer.blender.org/D11486
During transforming an image, a matrix multiplication per pixel was done.
The matrix in itself is always linear so it could be replaced by two additions.
During testing in debug builds playing back a movie went from 20fps to
300 fps.
Reviewed By: zeddb
Differential Revision: https://developer.blender.org/D11533
Talked with Bastien and we ended up looking into this. Issue is that the
dupliation through drag & drop should also be considered a
"sub-process", like Shift+D duplicating does. Added a comment explaining
why this is needed.
Current index builder is designed to be used in a single thread.
This makes all index buffer extractions single threaded.
This patch adds a thread safe solution enabling multithreaded
building of index buffers.
To reduce locking the solution would provide a task/thread local
index buffer builder (called sub builder).
When a thread is finished this thread local index buffer builder
can be joined with the initial index buffer builder.
`GPU_indexbuf_subbuilder_init`: Initialized a sub builder. The
index list is shared between the parent and sub buffer, but the
counters are localized. Ensuring that updating counters would
not need any locking.
`GPU_indexbuf_subbuilder_finish`: merge the information of the
sub builder back to the parent builder. Needs to be invoked outside
the worker thread, or when sure that all worker threads have been
finished. Internal the function is not thread safe.
For testing purposes the extract_points extractor has been migrated to
the new API. Herefore changes to the mesh extractor were needed.
* When creating tasks, the task number of current task is stored in
ExtractTaskData including the total number of tasks.
* Adding two functions in `MeshExtract`.
** `task_init` will initialize the task specific userdata.
** `task_finish` should merge back the task specific userdata back.
* adding task_id parameter to the iteration functions so they can
access the correct task data without any need for locking.
There is no noticeable change in end user performance.
Reviewed By: mano-wii
Differential Revision: https://developer.blender.org/D11499
This node creates a boolean face attribute that is "true" for
every face that has the given material.
Differential Revision: https://developer.blender.org/D11324
Selection of an FCurve with box/circle select now selects the entire
curve and all its keys:
- Box selecting a curve selects all the keyframes of the curve.
- Ctrl + box selecting of the curve deselects all the keyframes of the
curve.
- Shift + box selecting of the curve extends the keyframe selection,
adding all the keyframes of the curves that were just selected to the
selection.
- In all cases, if the selection area contains a key, nothing is
performed on the curves themselves (the action only impacts the
selected keys).
Reviewed By: sybren, #animation_rigging
Differential Revision: https://developer.blender.org/D11181
Some of the primitive nodes can return null in an error condition.
This is confusing mixed with adding a maderial slot in calling
functions. This is the second crash caused by that confusion. It's
simpler to add the slot right when allocating the mesh, and it will
lend itself better to copy & paste coding in the future.
Differential Revision: https://developer.blender.org/D11530
Under some circumstances using task isolation can cause deadlocks.
Previously, our task pool implementation would run all tasks in an
isolated region. Now using task isolation is optional and can be
turned on/off for individual task pools.
Task pools that spawn new tasks recursively should never enable
task isolation. There is a new check that finds these cases at runtime.
Right now this check is disabled, so that this commit is a pure refactor.
It will be enabled in an upcoming commit.
This fixes T88598.
Differential Revision: https://developer.blender.org/D11415
Simplify vertex normal calculation by moving the main normal
accumulation function to operate on vertices instead of faces.
Using faces had the down side that it needed to zero, accumulate and
normalize the vertex normals in 3 separate passes, accumulating also
needed a spin-lock for thread since the face would write it's normal
to all of it's vertices which could be shared with other faces.
Now a single loop over vertices is performed without locking.
This gives 5-6% speedup calculating all normals.
This also simplifies partial updates, fixing a problem where
all connected faces were being read from when calculating normals.
While this could have been resolved separately,
it's simpler to operate on vertices directly.
Move BMesh conversion and all loading code into worker.
Reviewed By: Sebastian Parborg (zeddb)
Differential Revision: https://developer.blender.org/D11288
ChildWindowFromPoint retrieves the child of the provided window at a
point. In this case it always returns 0 because HWND_DESKTOP is flag
defined as 0, which is never a valid window handle and is not intended
for use in place of a window handle.
Forwarding of mousewheel events was added in adb08def61, and later
modified to the current unworking state in e9645806f5. Sending mouse
wheel events to the window under the cursor is a system preference and
therefore should not be overridden by Blender, therefore the noop code
has been removed.
Added a new api function to stich multires grids
on specific faces in a mesh,
subdiv_ccg_average_faces_boundaries_and_corners,
and changed multires normal calc to use it.
VTune profiling showed that this was a major
performance hit once you get above 10,000 or so
base mesh faces and/or have a high number of
subdivision levels.
Here's a video comparing the difference. Note the
bpy.app_debug switch is not in the final commit.
{F10145323}
And the .blend file:
{F10145346}
Reviewed By: Sergey Sharybin (sergey)
Differential Revision:
https://developer.blender.org/D11334
Added a new api function to stich multires grids
on specific faces in a mesh,
subdiv_ccg_average_faces_boundaries_and_corners,
and changed multires normal calc to use it.
VTune profiling showed that this was a major
performance hit once you get above 10,000 or so
base mesh faces and/or have a high number of
subdivision levels.
Here's a video comparing the difference. Note the
bpy.app_debug switch is not in the final commit.
{F10145323}
And the .blend file:
{F10145346}
Reviewed By: Sergey Sharybin (sergey)
Differential Revision:
https://developer.blender.org/D11334
Also use Curve as an argument instead of Object, since the object was
only used to retrieve the curve, and the calling code is already working
with curve data.
Some of the comments referenced code that was no longer there, or even
defines that were removed. Other comments were more confusing and
vague than helpful. Also adjust formatting in a few cases.
This patch improves the positioning of the little mnemonic underlines
shown under some hotkey letters in menus, especially when using custom
fonts.
see D11521 for details and examples.
Differential Revision: https://developer.blender.org/D11521
Reviewed by Campbell Barton
This revert went too far, only one line (the minimal version of FFMPEG
for `install_deps.sh` script`) actually needed to be reverted...
Sorry for the noise.
No functional changes as logic elsewhere already ensured this.
This just makes it obvious to anyone reading over the code that
these arguments are keyword only.
FOV was expanded to cover the shifting range,
rather than to precisely cut at the image border. Now fixed.
Reviewed By: Sebastian Parborg (zeddb)
Differential Revision: https://developer.blender.org/D11523
Moving the bounds code to the builder can be useful
for future optimizations like building multithreaded.
Reviewed By: fclem, jbakker
Differential Revision: https://developer.blender.org/D11455
Small bug that's causing edge count to be incorrect in
final culled list, just being offset exactly 1 entry.
Reviewed By: Sebastian Parborg (zeddb)
Differential Revision: https://developer.blender.org/D11513
This field was used for extend feature to get handle position of
metastrip children. Since D9972 extend feature works only on meta
strip itself, not it's children.
So `SEQ_transform_get_left_handle_frame()` second argument is always
false and can be removed.
Another instance of `seq->tmp usage` is hack to distinguish strips to be
shuffled, which is not covered by this patch.
Reviewed By: campbellbarton
Differential Revision: https://developer.blender.org/D10321
Use bounding box tests quickly tell that two components cannot
have a containment relation between each other. This change
cut about 0.6s off a test with 25 big icospheres.
Add a call to orthogonalize the matrix before processing for the
same reasons as D8915, and an early exit in case no limits are
enabled for a bit of extra efficiency.
Since the constraint goes through Euler decomposition, it would
in fact remove shear even before this change, but the resulting
rotation won't make much sense.
This change allows using the constraint without any enabled limits
purely for the purpose of efficiently removing shear.
Differential Revision: https://developer.blender.org/D9626
Since Limit Rotation is based on Euler decomposition, it should allow
specifying the order to use for the same reasons as Copy Rotation does,
namely, if the bone uses Quaternion rotation for its animation channels,
there is no way to choose the order for the constraint.
Ref D9626
Fixed the logic for seeking in ffmpeg video files.
The main fix is that we now apply a small offset in ffmpeg_get_seek_pos
to make sure we don't get the frame in front of the seek position when
seeking backward.
The rest of the changes is general cleanup and untangling code.
Reviewed By: Richard Antalik
Differential Revision: http://developer.blender.org/D11492
Due to the way we ship the CRT on windows TBB's
malloc proxy was unable to attach it self to
the memory management functions on windows 10.
This change moves ucrtbase.dll out of the blender.crt
folder and back into the main blender folder to side
step some undesirable behaviour on win10 making TBB
once more able to attach it self.
Having this work again, should give a speed
boost in memory allocation heavy workloads
such as mantaflow.
For details on how this only failed on Win10
see T88813
This patch exposes functionality for performing partial mesh updates
for normal calculation and face tessellation while transforming a mesh.
The partial update data only needs to be generated once,
afterwards the cached connectivity information can be reused
(with the exception of changing proportional editing radius).
Currently this is only used for transform, in the future it could be
used for other operators as well as the transform panel.
The best-case overall speedup while transforming geometry is about
1.45x since the time to update a small number of normals and faces is
negligible.
For an additional speedup partial face tessellation is multi-threaded,
this gives ~15x speedup on my system (timing tessellation alone).
Exact results depend on the number of CPU cores available.
Ref D11494
Reviewed By: mano-wii
Cycles, Eevee, OSL, Geo, Attribute
This operator provides consistency with the standard math node. Allows users to use a single node instead of two nodes for this common operation.
Reviewed By: HooglyBoogly, brecht
Differential Revision: https://developer.blender.org/D10808
rB847579b42250 updated the TBB build script
which had some unintended consequences for
windows as the directory layout slightly
changed.
This change adjusts the builder to the new
structure, there are no version/functional
changes.
When changing to another texture paint slot, the texture displayed in
the viewport should change accordingly (as well as the image displayed
in the Image Editor).
The procedure to find the texture to display in the viewport
(BKE_texpaint_slot_material_find_node) could fail
though because it assumed iterating nodes would always happen in the
same order (it was index based). This is not the case though, nodes can
get sorted differently based on selection (see ED_node_sort).
Now check the actual image being referenced in the paint slot for
comparison.
ref T88788 (probably enough to call this a fix, the other issue(s)
mentioned in the report are more likely a feature request)
Reviewed By: mano-wii
Maniphest Tasks: T88788
Differential Revision: https://developer.blender.org/D11496
The (tracking) camera presets have not been updated in the last 7 or
more years, so they are very outdated. I found it pointless to have a
few specific camera models in the list and instead add the most commonly
used sensor sizes/film sizes. This way the list is shorter, easier to
maintain/becomes later outdated, and is more user friendly for most people
who don't own any of the specific cameras. I added the Crop Factor to the
Beginning of the name, so it gets sortet in the correct order and presets
are easier to find based on the size.
Reviewed By: #render_cycles, #motion_tracking, brecht, sergey
Differential Revision: https://developer.blender.org/D10739
These were only showing in the Properties Editor, but there is no reason
to have the panels be different in the sidebar (they should not show in
the top bar though).
agreed upon by both @anoniov and @mendio
ref T88787
So far, linked IDs were not properly sorted at all, only the ones
explicitely linked from WM code would be, but any indirectly linked
data-blocks would end up in some random order in their lists.
While not ideal, this is not a huge issue in itself, but it had bad
side-effects, e.g. causing (recursive) resync of overrides to happen in
random order, leading to mismatches between name indices of newly-generated
override IDs and the one existings e.g.
And in general, it is much better to be consistent here.
Note that the file sub-version is bumped for this commit, since some
sorting (the directly linked IDs which we keep a reference to) should
never need to be re-done after relevant doversion process.
This commit adds a node that outputs the total length of all
evalauted curve splines in a geometry set as a float value.
Differential Revision: https://developer.blender.org/D11459
This commit implements support for deleting curve data in the geometry
delete node. Spline domain and point domain attributes are supported.
Differential Revision: https://developer.blender.org/D11464
This implements T87633
This overlay renders a flash animation on the target object when
transfering the mode to it using the mode transfer operator.
This provides visual feedback when switching between objects without
extra overlays that affect the general color and lighting in the scene.
Differences with the design task:
- This uses just a fade out animation instead of a fade in/out animation.
The code is ready for fade in/out, but as the rest of the overlays
(face sets, masks...) change instantly without animation, having a fade
in/out effect gives the impression that the object flashes twice (once
for the face sets, twice for the peak alpha of the flash animation).
- The rendering uses a flat color without fresnel for now, but this can
be improved in the future to make it look more like the shader in the
prototype.
- Not enabled by default (can be enabled in the overlays panel), maybe
the defaults can change for 3.0 to disable fade inactive and enable this
instead.
Reviewed By: jbakker, JulienKaspar
Differential Revision: https://developer.blender.org/D11055
This operator is needed in some cases to update image preview.
In workspaces with smaller timelines this is limiting, because users
need to first check that mouse cursor is in correct place, then press
CTRL+R shortcut.
The Wayland server will not update hidden surfaces. This will block the
main event loop and thus also block updates to visible windows in a multi-
window setup.
Keyboard click-drag events now use the "Drag Threshold".
This resolves a problem where keyboard click drag events
used a much smaller threshold when using a tablet.
In some cases e.g. only objects would actually need resync, so
collections on the override character would not be resynced, and if some
objects were sharing relationships with others those could be
lost/destroyed.
max unneccessarily
Since rB298d5eb66916 [which was needed to update buttons with custom
property range functions correctly], using tab would always clamp
(hardmin/hardmax) properties which were using FLT_MAX / INT_MAX as range
in their property definitions.
The clamping of rB298d5eb66916 was copied over from rB9b7f44ceb56c
[where it was used for the softmin/softmax], and while the re-evaluation
of hardmin/hardmax is needed for custom property range functions, the
clamping should actually not take place.
There are many properties using FLT_MAX / INT_MAX etc. and while it
probably would be good to update these with ranges that make more sense
-- not using FLT_MAX / INT_MAX would not have done the clamping here --
there should not be an arbitrary limit to these and they should stay as
they are.
Maniphest Tasks: T88762
Differential Revision: https://developer.blender.org/D11473
Ensure 'virtual' linked override IDs generated by the recursive resync
process are tagged as indirectly linked data.
This is needed to avoid the 'missing data' messages on those virtual
data-blocks after saving and reloading.
`BKE_main_collections_parent_relations_rebuild`,
`BKE_collection_parent_relations_rebuild` anf their internal
dependencies had two issues fixed by this commit:
* Main one was that a same collection could be processed several times,
sometimes even in an infinite loop (in some rare corner cases), by
`collection_parents_rebuild_recursive`.
* More exotic, code here would not ensure that the collections it was
processing were actually in Main (or a master one from a scene in
Main), which became an issue with some advanced ID management
processes involving partially out-of-main remapping, like liboverride
resync.
Use SEQ_time_strip_intersects_frame function to test if strip intersects with frame.
Note: There are cases where this function should not be used. For example splitting
strips require at least 1 frame "inside" strip. Another example is drawing, where
playhead technically doesn't intersect strip, but it is rendered, because current
frame has "duration" or "thickness" of 1 frame.
Reviewed By: sergey
Differential Revision: https://developer.blender.org/D11320
Remove unused flag `SEQ_DUPE_ANIM` and code used by this flag.
Remove flag `SEQ_DUPE_CONTEXT` and refactor code, to split operator
logic from duplication code.
Reduce indentation level in for loop.
Reviewed By: sergey
Differential Revision: https://developer.blender.org/D11318
Images with 4:2:2 and 4:4:4 chroma subsampling were blurred when
`SWS_FAST_BILINEAR` interpolation is set for `anim->img_convert_ctx`.
Use `SWS_BILINEAR` interpolation for all movies, as performance is
not impacted by this change.
Reviewed By: sergey
Differential Revision: https://developer.blender.org/D11457
Changes in rBce649c73446e, affected established proxy codec preset.
Presets were not working and all presets were similar to `veryfast`.
Tunes are now working too, so `fastdecode` tune can be used. I have
measured little improvement, but I tested this only on 2 machines and
I have been informed that `fastdecode` tune does influence decoding
performance for some users.
Change preset from `slow` to `veryfast` and add tune `fastdecode`
Reviewed By: sergey
Differential Revision: https://developer.blender.org/D11454
While encoder parameters for lossless encoding are set correctly,
output is not lossless due to pixel format being set to
`AV_PIX_FMT_YUV420P` which is inherently lossy due to chroma subsampling.
This was reported in T61569 and was merged to T57397, but there were
2 bugs - one for encoding and one for decoding.
Set pixel format to `AV_PIX_FMT_YUV444P` when rendering lossless H264
files. This format isn't available in `codec->pix_fmts[0]` and it looks,
that it has to be hard-coded.
Reviewed By: sergey
Differential Revision: D11458
This patch is from erik85, who says:
This patch makes populate_plane inside polymesh_from_trimesh_with_dissolve run in parallel.
On a test file with a boolean between two subdivided cubes (~6 million verts) this gives a 10% speed increase (49.5s to 45s) on my 6 core CPU.
Also there is an optimization of other_tri_if_manifold to skip the contains-call and get the pointer directly.
This reduces CPU time for find_patches from 5s to 2.2s on the same test file.
When there are many components (separate pieces of connected mesh),
a part of the algorithm to determine component containment was slow.
Using a float version of finding the nearest point on a triangle
as a prefilter sped this up enormously. A case of 25 icospheres
subdivided twice goes 11 seconds faster on my Macbook pro with this
change.
This just moves a couple of files in `space_node` to C++ and fixes
related errors.
The goal is to be able to use C++ data structures to simplify the code.
Differential Revision: https://developer.blender.org/D11451
The mesh to curve node generated an empty curve because no edges were
selected. This commit changes that node to not add a curve in that case.
This also changes the curve to mesh node to not add a material when
no mesh was created. Even though we don't expect null curves or meshes
in this case, the change is harmless.
EEVEE uses hashing to sync aov names and types with the gpu.
For the type a hashed value was overridden making `decalA`
and `decalB` choose the same hash. This patches fixes this
by removing the most significant bit.
EEVEE uses hashing to sync aov names and types with the gpu. For the type a hashed value was overridden making `decalA` and `decalB` choose the same hash. This patches fixes this by removing the most significant bit.
Sometimes is required to reset the thickness or the opacity of the strokes. Actually this was done using a modifier, but this operators solves this.
Reviewed By: mendio, filedescriptor
Maniphest Tasks: T87427
Differential Revision: https://developer.blender.org/D11453
Often you need to copy a spline to do an operation, but don't want
to manually copy over all of the settings. I've already forgotten to
do it once anyway. These functions copy a spline's "settings" into a
new spline, but not the data. I tried to avoid duplicating any copying
so this is easier to extend in the future.
Differential Revision: https://developer.blender.org/D11463
This is due to cam->obmat precision issue,
where it affects view vector precision.
Reviewed by Sebastian Parborg (zeddb)
Differential Revision: https://developer.blender.org/D11461
Often it would be beneficial to avoid the virtual array implementation
in `geometry_component_curve.cc` that flattens an attribute for every
spline and instead read an attribute separately for every input spline.
This commit implements functions to do that.
The downside is some code duplication-- we now have two places handling
this conversion. However, we can head in this general direction for the
attribute API anyway and support accessing attributes in smaller
contiguous chunks where necessary.
No functional changes in this commit.
Differential Revision: https://developer.blender.org/D11456
Use the D-key to access the view menu instead of the Tilda key,
which isn't accessible on some international layouts.
To resolve the conflicts the following changes have been made.
- `D` (motion) opens the view menu.
- `D` (mouse-button) uses grease-pencil (as it does currently).
- `Tilda` is used to for "Object Mode Transfer" instead of the View menu.
See T88092 for details.
Reviewed By: Severin
Ref D11189
Add a doc-string explaining the purpose of each call back and
how they should be used.
Also add a currently unused callback 'POST_FAIL' that is to be used in
cases the action fails - giving script authors, a guarantee that a
call to `pre` will always have a matching `post/post_fail` call.
- D11422: adds a callback that can use 'post_fail'.
- T88696: proposed these conventions.
This node is similar to the mask modifier, but it deletes the elements
of the geometry corresponding to the selection, which is retrieved as
a boolean attribute. The node currently supports both mesh and point
cloud data. For meshes, which elements are deleted depends on the
domain of the input selection attribute, just like how behavior depends
on the selection mode in mesh edit mode.
In the future this node will support curve data, and ideally volume
data in some way.
Differential Revision: https://developer.blender.org/D10748
Metadata panel was visible in each category. In other editors, this
panel is usually placed in category with other source media properties.
In sequencer, metadata is transfered over while compositing and relation
to particular strip is lost, therefore separate category for metadata
seems to be best option. Since Metadata panel is alone in this category,
it will be open by default.
This commit skips the eager recalculation of mesh normals in the
transform node. Often another deformation or topology-altering
operation will happen after the transform node, which means the
recalculation was redundant anyway.
In one of my test cases this made the node more than 14x faster.
Though depending on the situation the cost of updating the normals
may just be shifted elsewhere.
This was reported for the special case of mapping with "Strand /
Particle" coords, but was not working with other coordinates either.
Dont see a reason for not supporting Size influence textures for these
kinds of particles (and since these types of particles have an "age"
like all others as well, even the "Strand / Particle" coords are
supported here as well)
Maniphest Tasks: T88715
Differential Revision: https://developer.blender.org/D11449
This fixes T88455 by adding an empty material slot to newly
generated meshes. This allows the object to overwrite the
"default" material without any extra nodes. Technically,
all polygons reference the material index 0 already, so it
makes sense to add a material slot for this material index.
Differential Revision: https://developer.blender.org/D11439
Removing scripts that were placed in the source tree that would drive
the old buildbot. With the new buildbot in place these scripts aren't
being used anymore.
The buildbit is currently driven by
`build_files/config/pipeline_config.json`. In the near future make
update would also use this config.
Overview of the new buildbot: https://wiki.blender.org/wiki/Infrastructure/BuildBot
Work done by James Monteath.
`bvhtree_from_mesh_edges_create_tree` can actually leave the BVHTree
NULL (e.g. if no edges are present).
Now dont allocate `BVHTreeFromMesh` on the `SurfaceModifierData` at all
in case the tree would be NULL anyways.
Places like `get_effector_data` check for `SurfaceModifierData`-
>`BVHTreeFromMesh` and dont try to stuff like getting a closest point on
surface, which would crash as soon as BVHNodes would need to be accessed
(from the NULL BVHTree).
Maniphest Tasks: T88658
Differential Revision: https://developer.blender.org/D11443
This commit fixes two different issues:
* In some cases, when an object was added to a sub-collection and used
into a different subcollection, and the root common collection would
not need to be resynced, it would end up creating multiple overrides
of the new object. This was affecting both normal and recursive
resync.
* In recurisve resync case, the barrier code to define what is part or
not of a override group/hierarchy was wrong.
Note that the current solution for the first issue is sub-optimal (it
goes back to the root of the override hierarchy and resync the whole
thing), a better solution is TODO for now.
This function would considere that there was a name conflict even in
case existing ID would be a linked one.
This is only a (symbolic) perforance improvement and logical fix, since
`BKE_id_new_name_validate` would not do that mistake anyway.
This is mandatory for liboverride resync, since this feature may imply
we have to create linked overrides in libraries, and there may be
several copies of those.
This is also a first step to a more general support of IDmanagement-editing
library data.
Note that this commit should have absolutely no effect on current code,
as the only function allowed to check unique names for linked IDs
currently is `BKE_libblock_management_main_add`, which is unused.
This commit also adds some basic testing for `BKE_id_new_name_validate`.
This patch adds the base code needed to make the full-frame system work for both current tiled/per-pixel implementation of operations and full-frame.
Two execution models:
- Tiled: Current implementation. Renders execution groups in tiles from outputs to input. Not all operations are buffered. Runs the tiled/per-pixel implementation.
- FullFrame: All operations are buffered. Fully renders operations from inputs to outputs. Runs full-frame implementation of operations if available otherwise the current tiled/per-pixel. Creates output buffers on first read and free them as soon as all its readers have finished, reducing peak memory usage of complex/long trees. Operations are multi-threaded but do not run in parallel as Tiled (will be done in another patch).
This should allow us to convert operations to full-frame in small steps with the system already working and solve the problem of high memory usage.
FullFrame breaking changes respect Tiled system, mainly:
- Translate, Rotate, Scale, and Transform take effect immediately instead of next buffered operation.
- Any sampling is always done over inputs instead of last buffered operation.
Reviewed By: jbakker
Differential Revision: https://developer.blender.org/D11113
draw_cache_extract_mesh for task scheduling. Will be refactored to draw_cache_extract_mesh_scheduling later on after migrating to CPP.
draw_cache_extract_mesh_render_data extraction of mesh render data from edit mesh/mesh into a more generic structure.
draw_cache_extract_mesh_extractors containing all the extractors. This will be split up further into a single file per extractor.
This was kept since these blocks are easier to follow.
Remove as the overall result wasn't so readable
(especially with nested ifdef's).
Replace disabled code with comment on the indices used for quads/tris.
2 sided faces aren't supported and will cause problems in many areas
of Blender's code.
Removing (implied) support for faces with fewer than 3 sides
means the total number of triangles is known ahead of time.
This simplifies adding support for multi-threading and partial updates
to an existing tessellation - as the face and loop indices can be used
to access the range of triangles associated with a face.
Also correct outdated comments.
When projecting into screen space Z value isn't always needed.
Add 2D projection functions, renaming them to avoid accidents
happening again.
- Add GPU_matrix_project_2fv
- Add ED_view3d_project_v2
- Rename ED_view3d_project to ED_view3d_project_v3
- Use the 2D versions of these functions when the Z value isn't used.
This is from patch D11432 from Erik Abrahamsson. He found that
in some mpq3 functions called frequently from loops, passing in
buffers for termporary mpq3 values can save substantial time.
On my machine, his example in that patch went from 9.48s to 7.50s
for the boolean part of the calculation. On his machine, a running
time went from 17s to 10.3s.
The fseek() function on Windows only accepts a 32-bit long offset
argument. Because of this we have our own version, BLI_fseek(), which
will use 64-bit _fseeki64() on Windows. This patch just replaces some
fseek() calls with BLI_fseek().
Differential Revision: https://developer.blender.org/D11430
Reviewed by Brecht Van Lommel
Now FPS is displayed in the video source for videos to provide easy
access.
Reviewed By: Richard Antalik
Differential Revision: http://developer.blender.org/D11441
bf_bmesh historically always build with the /WX flag
on windows making all warnings errors, somewhere along
the way this has broken for msbuild, ninja still exhibits
the expected behaviour.
The flags are still passed to the target, and I've validated
they are there when the add_library call fires, but they
somehow never make it to the generated msbuild project files.
I suspect this is a cmake bug but I'm seemingly unable
to extract a repro case to file a bug upstream.
Setting the same options target_compile_options seems to work,
I'm not happy about the unexplained nature of the breakage
but this will have to do for now.
The same code existed in 2.82 and earlier so this should be safe. Removing the
custom implementation of shutil.copytree in f34d5d9 did not correctly add back
the option to copy symlinks.
This patch replaces / redoes the entire MeshExtractors system.
Although they were useful and facilitated the addition of new buffers, they made it difficult to control the threads and added a lot of threading overhead.
Part of the problem was in traversing the same loop type in different threads. The concurrent access of the BMesh Elements slowed the reading.
This patch simplifies the use of threads by merging all the old callbacks from the extracts into a single series of iteration functions.
The type of extraction can be chosen using flags.
This optimized the process by around 34%.
Initial idea and implementation By @mano-wii.
Fine-tuning, cleanup by @atmind.
MASTER:
large_mesh_editing:
- rdata 9ms iter 50ms (frame 155ms)
- Average: 6.462874 FPS
PATCH:
large_mesh_editing:
- rdata 9ms iter 34ms (frame 136ms)
- Average: 7.379491 FPS
Differential Revision: https://developer.blender.org/D11425
For 2.93 we bumped the minimum windows requirement
to windows 8.1, but did not do any clean-up of any
win 8/8.1 API usage we dynamically accessed though
LoadLibrary/GetProcAddress.
This patch bumps _WIN32_WINNT to 0x0603 (win 8.1)
and cleans up any API use that was accessed in a
more convoluted way than necessary
Differential Revision: https://developer.blender.org/D11331
Reviewed by: harley, nicholas_rishel
Use the appropriate notifier, listeners were already doing the rest
properly.
Maniphest Tasks: T88569
Differential Revision: https://developer.blender.org/D11436
Cause is that initializing the cryptomatte session would reset the
current frame of an image sequence. The solution is to always use the
scene current frame so it resets to the correct frame.
This was a todo that wasn't solved after it landed in master.
Needs to be backported to 2.93.
When compositor node tree has a texture node, TextureOperation vector inputs has always {0, 0} resolution instead of having same resolution as TextureOperation which is the expected behaviour for resolutions propagation.
Current TextureOperation determineResolution implementation doesn't determine inputs resolution, breaking propagation of preferred resolution and that's the reason why they are always 0. Setting scene resolution always would mean it is its own resolution and could make sense, but setting it only when preferred resolution is 0, breaks preferred resolution logic affecting other operations as explained in D10972. In any case scene resolution is already the default preferred resolution on viewer and compositor nodes.
Reviewed By: jbakker
Differential Revision: https://developer.blender.org/D11381
The issue was two fold. We didn't properly:
1. Initialize the codec default values which would lead to VLC
complaining because of garbage/wrong codec settings.
2.Calculate the time base for the video. FFmpeg would happily accept
this but VLC seems to assume the time base value is at least somewhat
correct and couldn't properly display the frames as the internal time
base was huge. We are talking about 90k ticks (tbn) for one second of
video!
This patch initializes all codecs to use their default values and fixes
the time base calculation so it follows the guidelines from ffmpeg.
Reviewed By: Sergey, Richard Antalik
Differential Revision: http://developer.blender.org/D11426
Code checking for potential collection loop dependencies can be called
in cases where we cannot guarantee that there is no NULL pointers, so we
need to check those. Was already done for objects.
NOTE: doubled-checked by @jbakker, thanks.
Code checking for potential collection loop dependencies can be called
in cases where we cannot guarantee that there is no NULL pointers, so we
need to check those. Was already done for objects.
Reuse loose geometry during selection (and other operations) from
previous calculation. Loose geometry stays the same, but was
recalculated to determine the size of GPU buffers. This patch would
reuse the previous loose geometry when geometry wasn't changed.
Although not the main bottleneck during selection it is measurable.
Master.
`rdata 46ms iter 55ms (frame 410ms)`
This patch.
`rdata 5ms iter 52ms (frame 342ms)`
Reviewed By: mano-wii
Differential Revision: https://developer.blender.org/D11339
This patch from Erik Abrahamsson uses a parallel_for to speed up
the case where the input is not manifold and the "hole_tolerant"
option is set.
In a test case on a 24 core (48 thread) machine, this sped up a
the boolean part on an object with 221k triangles from 12.06s to 0.46s.
When using ``Path`` alignment, if the stroke has one point the texture rotates randomly when move the viewport. This was because with one point is impossible to calculate a path.
Now, if the stroke has only one point, the texture for this stroke is aligned to Object.
Before the FFmpeg commit: github.com/FFmpeg/FFmpeg/commit/1c0885334dda9ee8652e60c586fa2e3674056586
FFmpeg would use deprecated variables to calculate the video fps.
We don't use these deprecated variables anymore, so ensure that the
duration is correct in ffmpeg versions without this fix.
Reviewed By: Sergey, Richard Antalik
Differential Revision: http://developer.blender.org/D11417
We didn't initialize the scaled proxy frame properly.
This would lead to issues in ffmpeg 4.4 as they are more strict that the API is properly used.
Now we initialize the size and format of the frame.
Based on the task T88006, there are a few simple changes
to make to improve the switch node:
- Change the label to "False" / "True" for clarity
- Change default to geometry, as it's the basic data container in
geometry nodes.
- Change node class to `NODE_CLASS_CONVERTOR`, which was an oversight
in the original patch.
I will add the new socket types (material and texture) in a separate commit.
Thanks to @EitanSomething for the original patch.
Differential Revision: https://developer.blender.org/D11165
This was caused by unsafe sqrt calls.
Fixes T86578 white artifacts in EEVEE
Reviewed By: brecht, dfelinto
Differential Revision: https://developer.blender.org/D11428
While we could make this node work for boolean inputs in the future,
currently it's really just designed to compare "float-like" inputs.
Many comparison modes don't even make sense for boolean inputs.
Therefore, the simplest fix for this bug is just to disable the
boolean attribute input modes for this node.
Differential Revision: https://developer.blender.org/D11427
When using ``Path`` alignment, if the stroke has one point the texture rotates randomly when move the viewport. This was because with one point is impossible to calculate a path.
Now, if the stroke has only one point, the texture for this stroke is aligned to Object.
This node creates poly curve splines from mesh edges. A selection
attribute input allows only using some of the edges from the mesh.
The node builds cyclic splines from branchless groups of edges where
possible, but when there is a three-way intersection, the spline stops.
The node also transfers all attributes from the mesh to the resulting
control points. In the future we could add a way to limit that to a
subset of the attributes to improve performance.
The algorithm is from Animation Nodes, written by @OmarSquircleArt.
I added the ability to use a selection, attribute transferring, and
used different variable names, etc, but other than that the algorithm
is the same.
Differential Revision: https://developer.blender.org/D11265
Not entirely sure why this was not an issue for 16.9
but TBB includes the Windows.h header which by default
will define min and max macro's
These collide with the stl versions in <algorithm>
This patch requests Windows.h not to define the
problematic macro's, resolving the conflict.
Caused by {rB66923031e6f2}.
Code would process unselected sequences and skip selected, needs to be
the other way around.
Maniphest Tasks: T88635
Differential Revision: https://developer.blender.org/D11424
When using the operator `ui.copy_data_path_button(full_path=True)` ({key
ctrl shift Alt C} on hover) the copied path does not consider the
library origin. That means that when there is a name clash the data path
is not accurate and refers to the local item instead.
This patch adds the library (if the ID is linked) of the returned string
from RNA_path_full_ID_py.
bpy.data.objects["Cube", "//library.blend"] instead of
bpy.data.objects["Cube"]
note: parsing this happens in
pyrna_prop_collection_subscript_str_lib_pair_ptr
Maniphest Tasks: T88499
Differential Revision: https://developer.blender.org/D11412
A utility that supports passing in actions as command line arguments for
writing reproducible interactions, benchmarking, profiling and testing.
Unlike regular scripts this is able to control model operators usefully.
Typical ways of controlling Blender using this utility are via
operator id's, menu search and explicit events.
Others methods can be added as needed.
See the doc-string for example usage.
This patch will use compute shaders to create the VBO for hair.
The previous implementation uses transform feedback.
Timings before: between 0.000069s and 0.000362s.
Timings after: between 0.000032s and 0.000092s.
Speedup isn't noticeable by end-users. The patch is used to test
the new compute shader pipeline and integrate it with the draw
manager. Allowing EEVEE, Workbench and other draw engines to
use compute shaders with the introduction of `DRW_shgroup_call_compute`
and `DRW_shgroup_vertex_buffer`.
Future improvements are possible by generating the index buffer
of hair directly on the GPU.
NOTE: that compute shaders aren't supported by Apple and still use
the transform feedback workaround.
Reviewed By: fclem
Differential Revision: https://developer.blender.org/D11057
Added missing topbar in VSE.
Also added the Stabilizer options to Topbar for all modes.
Reviewed By: mendio, pepeland
Maniphest Tasks: T86465
Differential Revision: https://developer.blender.org/D11347
We need to re-evaluate what needs to be resynced after each step of
processing overrides from a given 'indirect level' of libraries.
Otherwise, recusrive overrides (overrides of linked overrides) won't
work.
Note that this should not change too much in practice currently, since
there are other issues with recursive overrides yet.
Also, checks (CLOG errors) added show that some ID (node trees) seem to
be detected as needing resynced even after beig just resynced, this
needs further investigation still. Could be though that it is due to
limit currently set on nodetrees, those are always complicated
snowflakes to deal with...
Oversight in {rB470f17f21c06}.
Hiding was only done for the first mesh, then the operator finished (in
case of UV_SYNC_SELECTION).
Now just continue to the next.
Maniphest Tasks: T88625
Differential Revision: https://developer.blender.org/D11413
Both before and after can have artifacts with some normal maps, but this seems to give
worse artifacts on average which are not worth the minor performance increase.
This reverts commit 5c4d24e1fd.
Ref T88368, D10084
This allows choosing material and texture sockets for the group input
node in the modifier. Note that currently grease pencil materials are
displayed in the list, even though grease pencil data is not supported
yet by geometry nodes. That is more complicated to fix in this case,
since we use IDProperties to store the dynamic exposed inputs.
Differential Revision: https://developer.blender.org/D11393
blender-laucher.c was not an ideal name for this file
since it's not directly clear it is windows only.
This change renames it to blender_launcher_win32.c
to be more in line with other win32 specific files
we have.
This patch adds relatively small changes to the curve draw
cache implementation in order to draw the curve data in the
viewport. The dependency graph iterator is also modified
so that it iterates over the curve geometry component, which
is presented to users as `Curve` data with a pointer to the
`CurveEval`
The idea with the spline data type in geometry nodes is that
curve data itself is only the control points, and any evaluated
data with faces is a mesh. That is mostly expected elsewhere in
Blender anyway. This means it's only necessary to implement
wire edge drawing of `CurveEval` data.
Adding a `CurveEval` pointer to `Curve` is in line with changes
I'd like to make in the future like using `CurveEval` in more places
such as edit mode.
An alternate solution involves converting the curve wire data
to a mesh, however, that requires copying all of the data, and
since avoiding it is rather simple and is in-line with future plans
anyway, I think doing it this way is better.
Differential Revision: https://developer.blender.org/D11351
This patch removes unnecessary calls to `BKE_main_id_tag_all` where the
same job is done by `BKE_main_id_clear_newpoins` on the following line.
Reviewed By: campbellbarton, mont29
Ref D11379
Use array instead of ListBase for line art
bounding area linked triangles and edges.
Reviewed By: Sebastian Parborg (zeddb)
Differential Revision: https://developer.blender.org/D11302
This is not supposed to happen, but better be safe than sorry, and
assume it is beyond unlikely that someone would use chains of over 10k
linked libraries.
Allows to centralize storage and modification checks in a single place,
avoiding duplication in the synchronization code.
Ideally we would somehow be able to more granularly modify Cycles side
objects. Leaving this for a future decision, because it might be better
to implement it as a graph on the sync side.
Makes it more explicit they operate on shading/light.
Gives room to move more viewport related settings into this class and
cover with specific or generic modification checks.
Such pattern should only be used when it is really needed. Otherwise
just stick to a more regular design, without worrying who is the user
of the class. Otherwise it will be annoying to subclass or unit test.
No need to state that it is a viewport display pass, since the method
is within viewport parameters it is implied that parameters do belong
to the viewport.
Brings this code closer to the Cycles-X branch.
Very stupid mistake in libraries indirect-level building code, was not
skipping 'loop-back' ID pointers.
Note that we also need some level of checks for the case where there
would be an actual dependency loop between libraries, this is not
supposed to be possible, but better be safe than sorry. Will add in next
commit.
An arbitrary size offsets was used in float_array_to_string,
simplify the loop, use exact size limits.
Also rename variables so it's clear which array the length apply to.
Event though in practice this wasn't causing problems as the fixed size
buffers are generally large enough not to truncate text.
Using the result from `snprint` or `BLI_snprintf` to step over a fixed
size buffer allows for buffer overruns as the returned value is the size
needed to copy the entire string, not the number of bytes copied.
Building strings using this convention with multiple calls:
ofs += BLI_snprintf(str + ofs, str_len_max - ofs);
.. caused the size argument to become negative,
wrapping it to a large value when cast to the unsigned argument.
A deadlock could happen under certain circumstances when
geometry nodes is used on multiple objects.
Once T88598 is resolved, multi-threading can be enabled again.
Differential Revision: https://developer.blender.org/D11405
While the advantage isn't large,
it's simpler to skip the intermediate link.
Also remove unused next and previous struct members
from MeshUndoStep_Elem.
When editing more than 1 object at a time, complete copies of each mesh
were being stored. Now the most recent undo-data for each mesh is used
(when available).
This commit adds interpolation from the point domain to the spline
domain and the other way around. Before this, spline domain attributes
were basically useless, but now they are quite helpful as a way to use
a shared value in a contiguous group of points.
I implementented a special virtual array for the spline to points
conversion, so that conversion should be close to the ideal performance
level, but there are a few ways we could optimize the point to spline
conversion in the future:
- Use a function virtual array to mix the point values for each spline
on demand.
- Implement a special case for when the input virtual array is one of
the virtual arrays from the spline point attributes. In other words,
decrease curve attribute access overhead.
Differential Revision: https://developer.blender.org/D11376
This patch fixes a long-standing complaint from users:
the console window shortly flashing when they start
blender.
This is done by adding a new executable called
blender-launcher.exe which starts blender.exe while
hiding the console.
Any command line parameters given to blender-launcher
will be passed on to blender.exe so it'll be a drop
in replacement.
Starting blender.exe on its own will still function as
a proper console app so no changes required here for
users that use blender for batch processing.
Notable changes:
Registering blender (-R switch) will now register
blender-launcher as the preferred executable.
This patch updates the installer and updates the
shortcuts to start blender-launcher.exe rather
than blender.exe
Differential Revision: https://developer.blender.org/D11094
Reviewed by: brecht, harley
This reverts commit 8f9599d17e.
Mac seems to have an error with this change.
```
ERROR: /Users/blender/git/blender-vdev/blender.git/source/blender/draw/intern/draw_hair.c:115:44: error: use of undeclared identifier 'shader_src'
ERROR: /Users/blender/git/blender-vdev/blender.git/source/blender/draw/intern/draw_hair.c:123:13: error: use of undeclared identifier 'shader_src'
ERROR: make[2]: *** [source/blender/draw/CMakeFiles/bf_draw.dir/intern/draw_hair.c.o] Error 1
ERROR: make[1]: *** [source/blender/draw/CMakeFiles/bf_draw.dir/all] Error 2
ERROR: make: *** [all] Error 2
```
Both before and after can have artifacts with some normal maps, but this seems to give
worse artifacts on average which are not worth the minor performance increase.
This reverts commit 21bc1a99ba.
Ref T88368, D10084
The skin modifier was moving vertices without updating normals for the
connected faces, this happened when smoothing and welding vertices.
Reviewed By: mont29
Ref D11397
Recursive resync means also resyncing overrides that are linked from
other library files into current working file.
Note that this allows to get 'working' files even when their
dependencies are out of sync. However, since linked data is never
written/saved, this has to be re-done every time the working file is
loaded, until said dependencies are updated properly.
NOTE: This is still missing the 'report' side of things, which is part
of a larger task to enhance reports regarding both linking, and
liboverrides (see T88393).
----------
Technical notes:
Implementing this proved to be slightly more challenging than expected,
mainly because one of the key aspects of the feature was never done in
Blender before: manipulating, re-creating linked data.
This ended up moving the whole resync code to use temp IDs out of bmain,
which is better in the long run anyway (and more aligned with what we
generally want to do when manipulating temp ID data). It should also
give a marginal improvement in performances for regular resync.
This commit also had to carefully 'sort' libraries by level of indirect
usage, as we want to resync first the libraries that are the least directly
used, i.e. libraries that are most used by other libraries.
Even though shepkeys are not strictly speaking an embedded data, they
share quiet a few points with those, and from liboverride perspective
they are embedded, so...
While indeally we should only skip refcounting when relevant tag is set,
doing this in remapping code is too risky for now.
Related to previous commit and T88555.
While this is still very fuzzy in current code, this old behavior makes
it close to impossible to efficiently use out-of-main temp data, as it
implies that we'd need to update refcounts everytime we add something
back into BMain (an 'un-refcount' ID usages when removing from BMain).
Now that we have two separate flags/tags for those two different things,
let's not merge them anymore.
Note that this is somewhat on-going process, still needs more checks and
cleanup. Related to T88555.
This patch will use compute shaders to create the VBO for hair.
The previous implementation uses tranform feedback.
Timings master (transform feedback with GPU_USAGE_STATIC between 0.000069s and 0.000362s
Timings transform feedback with GPU_USAGE_DEVICE_ONLY. between 0.000057s and 0.000122s
Timings compute shader between 0.000032 and 0.000092s
Future improvements:
* Generate hair Index buffer using compute shaders: currently done single threaded on CPU, easy to add as compute shader.
Reviewed By: fclem
Differential Revision: https://developer.blender.org/D11057
With the compute pipeline calculation can be offloaded to the GPU.
This patch only adds the framework for compute. So no changes for users at
this moment.
NOTE: As this is an OpenGL4.3 feature it must always have a fallback.
Use `GPU_compute_shader_support` to check if compute pipeline can be used.
Check `gpu_shader_compute*` test cases for usage.
This patch also adds support for shader storage buffer objects and device only
vertex/index buffers.
An alternative that had been discussed was adding this to the `GPUBatch`, this
was eventually not chosen as it would lead to more code when used as part of a
shading group. The idea is that we add an `eDRWCommandType` in the near
future.
Reviewed By: fclem
Differential Revision: https://developer.blender.org/D10913
This patch adds an option in the Layers > Relations panel called "Disable Masks in Render".
When checked, no masks on this layer are included in the render.
Example:
| {F10087680} | {F10087681} |
See T88202 for why this is needed.
Reviewed By: antoniov
Maniphest Tasks: T88202
Differential Revision: https://developer.blender.org/D11234
Currently when rendering the view layer of a grease pencil layer that has
a mask layer attached, the mask layer would show in the rendered image.
This is inconsistent with the default behaviour with no mask on the
grease pencil layer, because it would only render what's on that
particular layer and not anything from any other layer.
This patch makes the masks invisible in the render.
Note: This might seem like not the best solution, but because masks are
just regular grease pencil layers, it's tricky to pass this edge-case to the
drawing code. The way it is handled right now is the best I could come
up with, without making changes that could affect something else.
Reviewed By: antoniov
Maniphest Tasks: T88202
Differential Revision: https://developer.blender.org/D11403
This issue was that `BKE_object_eval_uber_data` was not called for
the text object, because its geometry was not dependent upon
and its `is_directly_visible` tag was `false`. The crash happens in
rendering code, because the evaluated data is missing.
This not only affects text objects, but all object types that have a
geometry component that geometry nodes does not support yet.
The solution is to just add the missing dependencies.
Differential Revision: https://developer.blender.org/D11385
This operator is a common request of animators to convert the transformation (inluding modifiers) of one grease pencil object, into a new object, generating strokes.
Reviewed By: pepeland
Maniphest Tasks: T87424
Differential Revision: https://developer.blender.org/D11014
Calling BKE_nodetree_attribute_hint_add from multiple threads still
was not safe before..
One issue was that context_map embedded its values directly. So
when context_map grows, all NodeUIStorage would move as well.
I could patch around that by using std::unique_ptr in a few places,
but that just becomes too complex for now.
Instead I simplified the locking a bit by adding just locking a mutex
in NodeTreeUIStorage all the time while an attribute hint is added.
Differential Revision: https://developer.blender.org/D11399
When an `AttributeSet` is tagged as modified, which happens after the addition or
removal of an `Attribute` from the set, during the following GeometryManager device
update, we update and repack the kernel data for all attribute types. However, if we
only add or remove a `float` attribute, `float2` or `float3` attributes should not
be repacked for efficiency.
This patch adds some mechanisms to detect which attribute types are modified from
the AttributeSet.
Firstly, this adds an `AttrKernelDataType` to map the data type of the Attribute to
the one used in the kernel as there is no one to one match between the two since e.g.
`Transform` or `float4` data are stored as `float3s` in the kernel.
Then, this replaces the `AttributeSet.modified` boolean with a set of flags to detect
which types have been modified. There is no specific flag type (e.g.
`enum ModifiedType`), rather the flags used derive simply from the
`AttrKernelDataType` enumeration, to keep things synchronized.
The logic to remove an `Attribute` from the `AttributeSet` and tag the latter as modified
is centralized in a new `AttributeSet.remove` method taking an iterator as input.
Lastly, as some attributes like standard normals are not stored in the various
kernel attribute arrays (`DeviceScene::attribute_*`), the modified flags are only
set if the associated standard corresponds to an attribute which will be stored
in the kernel's attribute arrays. This makes it so adding or removing such attributes
does not trigger an unnecessary update of other type-related attributes.
Reviewed By: brecht
Differential Revision: https://developer.blender.org/D11373
Sharing data between duplicated IDs should be restricted to depsgraph
(CoW) cases, not all NO_MAIN ones...
While this was probably not an issue currently, we aim at using more and
more out-of-main IDs for temp data processing.
NOTE: Somewhat related to T88555, and similar issue as the one fixed in
rBdfb963c70df5.
Objects modified by geometry nodes modifiers were not caught as being
"dynamic".
Now add this modifier type to the list of modifiers making them "dynamic"
in the eyes of mantaflow.
(noticed by @sebbas in chat)
Maniphest Tasks: T88531
Differential Revision: https://developer.blender.org/D11389
(regression)
Code was actually checking for shapekeys, but these were not detected
properly (some effects like shape keys are added as virtual modifiers
before the user created modifiers)
Now go over virtual modifiers as well.
Maniphest Tasks: T88566
Differential Revision: https://developer.blender.org/D11388
Caused by {rB919558854d62}.
Same fix as in {rBdc8a43c8755a} -- let RNA enum item callbacks check
for NULL context.
The NULL context is used to extract items for document generation.
Maniphest Tasks: T88251
Differential Revision: https://developer.blender.org/D11391
Bump FFmpeg version to 4.4 to fix a problem where it would write the
wrong frame rate. Their old API was deprecated and Blender moved to the
new one in rB8d6264ea12bfac0912c7249f00af2ac8e3409ed1. The new one
produced files with the wrong frame rate, which was fixed in FFmpeg 4.4.
Manifest Task: T88568
Reviewed By: LazyDodo, zeddb
Differential Revision: https://developer.blender.org/D11392
This option is default off when creating line art objects
because line art seldom use lighting and the normal data
would be all over the place anyway.
Reviewed By: Antonio Vazquez (antoniov)
Differential Revision: https://developer.blender.org/D11372
This clarifies the data structures for storing edges
for different calculation stages.
Reviewed By: Sebastian Parborg (zeddb)
Differential Revision: https://developer.blender.org/D11386
Colors are often thought of as being 4 values that make up that can make any color.
But that is of course too limited. In C we didn’t spend time to annotate what we meant
when using colors.
Recently `BLI_color.hh` was made to facilitate color structures in CPP. CPP has possibilities to
enforce annotating structures during compilation and can adds conversions between them using
function overloading and explicit constructors.
The storage structs can hold 4 channels (r, g, b and a).
Usage:
Convert a theme byte color to a linearrgb premultiplied.
```
ColorTheme4b theme_color;
ColorSceneLinear4f<eAlpha::Premultiplied> linearrgb_color =
BLI_color_convert_to_scene_linear(theme_color).premultiply_alpha();
```
The API is structured to make most use of inlining. Most notable are space
conversions done via `BLI_color_convert_to*` functions.
- Conversions between spaces (theme <=> scene linear) should always be done by
invoking the `BLI_color_convert_to*` methods.
- Encoding colors (compressing to store colors inside a less precision storage)
should be done by invoking the `encode` and `decode` methods.
- Changing alpha association should be done by invoking `premultiply_alpha` or
`unpremultiply_alpha` methods.
# Encoding.
Color encoding is used to store colors with less precision as in using `uint8_t` in
stead of `float`. This encoding is supported for `eSpace::SceneLinear`.
To make this clear to the developer the `eSpace::SceneLinearByteEncoded`
space is added.
# Precision
Colors can be stored using `uint8_t` or `float` colors. The conversion
between the two precisions are available as methods. (`to_4b` and
`to_4f`).
# Alpha conversion
Alpha conversion is only supported in SceneLinear space.
Extending:
- This file can be extended with `ColorHex/Hsl/Hsv` for different representations
of rgb based colors. `ColorHsl4f<eSpace::SceneLinear, eAlpha::Premultiplied>`
- Add non RGB spaces/storages ColorXyz.
Reviewed By: JacquesLucke, brecht
Differential Revision: https://developer.blender.org/D10978
Colors are often thought of as being 4 values that make up that can make any color.
But that is of course too limited. In C we didn’t spend time to annotate what we meant
when using colors.
Recently `BLI_color.hh` was made to facilitate color structures in CPP. CPP has possibilities to
enforce annotating structures during compilation and can adds conversions between them using
function overloading and explicit constructors.
The storage structs can hold 4 channels (r, g, b and a).
Usage:
Convert a theme byte color to a linearrgb premultiplied.
```
ColorTheme4b theme_color;
ColorSceneLinear4f<eAlpha::Premultiplied> linearrgb_color =
BLI_color_convert_to_scene_linear(theme_color).premultiply_alpha();
```
The API is structured to make most use of inlining. Most notable are space
conversions done via `BLI_color_convert_to*` functions.
- Conversions between spaces (theme <=> scene linear) should always be done by
invoking the `BLI_color_convert_to*` methods.
- Encoding colors (compressing to store colors inside a less precision storage)
should be done by invoking the `encode` and `decode` methods.
- Changing alpha association should be done by invoking `premultiply_alpha` or
`unpremultiply_alpha` methods.
# Encoding.
Color encoding is used to store colors with less precision as in using `uint8_t` in
stead of `float`. This encoding is supported for `eSpace::SceneLinear`.
To make this clear to the developer the `eSpace::SceneLinearByteEncoded`
space is added.
# Precision
Colors can be stored using `uint8_t` or `float` colors. The conversion
between the two precisions are available as methods. (`to_4b` and
`to_4f`).
# Alpha conversion
Alpha conversion is only supported in SceneLinear space.
Extending:
- This file can be extended with `ColorHex/Hsl/Hsv` for different representations
of rgb based colors. `ColorHsl4f<eSpace::SceneLinear, eAlpha::Premultiplied>`
- Add non RGB spaces/storages ColorXyz.
Reviewed By: JacquesLucke, brecht
Differential Revision: https://developer.blender.org/D10978
Using displacement runs the shader eval kernel, but since OptiX modules are not loaded when
baking is active, those were not available and therefore failed to launch. This fixes that by falling
back to the CUDA kernels.
Issue is that due to the strange definition of render in grease pencil
(meaning should be rendered similar to rendering). This included normal
viewport rendering in OB_RENDER and OpenGL render in OB_RENDER.
For other rendering modes the overlay vertex opacity would be used. This
patch sets this value to 1 when rendering via a scene strip override.
NOTE: that this isn't a good solution as I expect that users want to use
the opacity of the Grease pencil object. Perhaps the GPencil team has a
better solution for it.
Tool settings for sequencer were not initialized, which caused crash
when adding strips.
There was fix for same issue in versioning rB0f81dafe6cec, but
subversion was not bumped, so files with uninitialized tool settings
may still exist.
Add `SEQ_tool_settings_get()` accessor function that will initialize
tool settings if they are missing. Change operator code to use
`SEQ_tool_settings_fit_method_get()` function instead of accessing
tool settings directly
Reviewed By: campbellbarton
Differential Revision: https://developer.blender.org/D11383
This will update active camera based on the maker for each frame.
Reviewed By: Sebastian Parborg (zeddb), Antonio Vazquez (antoniov)
Differential Revision: https://developer.blender.org/D11358
Do not allow a window to be created that has a top position that
can obscure all or part of title bar. Right and Left edges can
still be specified slightly outside of monitor bounds, but top
edge must be clamped to monitor top.
see D11371 for more details.
https://developer.blender.org/D11371
Reviewed by Ray Molenkamp
This fallback is an old hack.
It is difficult to have an orientation convention when several random
factors determine which one should be used.
In this case, to "fix" the problem, a new behavior had to be implemented.
Now the redo when moving the camera in `Camera View` has the default
orientation as `View`.
Deduplicates code by introducing a PlaneDirtortBaseOperation for common logic.
Reviewed By: #compositing, jbakker
Differential Revision: https://developer.blender.org/D11273
Prepare node for conversion to Geometry Nodes.
There should be no functional changes.
Reviewed By: JacquesLucke, LazyDodo
Differential Revision: https://developer.blender.org/D11226
When sound strip is above another strip such as movie strip, it prevents
from rendering movie strip.
This bug was introduced in 0b7744f4da. Function `must_render_strip()`
checks if there is any strip with `SEQ_BLEND_REPLACE` blending and
considers this strip as lowest strip in stack. Sound strips do have this
blend mode set, which caused the bug.
Remove all sound strips and muted strips from stack collection before
checking with `must_render_strip()` function
I'd like to use this file to draw curves from geometry nodes, which
would otherwise require implementing a C API. The changes in this
commit are minimal, mostly just casts and changing to nullptr.
Differential Revision: https://developer.blender.org/D11350
The CoW copy of the node group was not updated correctly after it
changed. Arguably, tagging the node tree could also be part of
`ntreeUpdateTree` (which needs to be called after changes to the
node tree anyway). However, there are many instances where the
depsgraph is tagged explicitly after `ntreeUpdateTree` is called,
so it should be fine here as well. This is similar to what is done
in `snode_dag_update`.
Differential Revision: https://developer.blender.org/D11342
The wrong matrix function was used and overwrote the custom bone shape
scale instead of reading from it.
Reviewed By: sybren
Differential Revision: https://developer.blender.org/D11330
When creating Win32 windows, the sizes and placements can be out by a
small amount, mostly noticeable near monitor edges. This is because
Windows 10 includes a thin invisible border (typically 7 pixels) when
determining position. Therefore the correct values can sometimes be
just outside the monitor bounds, but we clamp them at those bounds.
This patch fixes this by first clamping the requested values to monitor
bounds, adjusting for window chrome with AdjustWindowRectEx(), and then
using those adjusted values in CreateWindowExW().
see D11314 for more details.
Differential Revision: https://developer.blender.org/D11314
Reviewed by Ray Molenkamp
Accessing data through RNA should never implicitely create it. Objects'
and particles' forcefields and collision settings were doing this.
Note that UI code also had to be tweaked to properly handle `None`
(NULL) cases.
Differential Revision: https://developer.blender.org/D11341
This data (the force fields) are expected to always be there, but they
are currently created on the fly by RNA accessors (typically from UI
draw code), which is extremely wrong way to do it.
Differential Revision: https://developer.blender.org/D11341
This is an initial change to speed up the calculation of the Occlude eraser. In the future, we can add more optimizations, but at least this increase speed.
Intead to check always the 3 points, the check is skipped if it's not required.
Base in a solution by Philipp Oeser.
This is related to T88412
Root of the issue was actually hidden deep in depsgraph itself: it would
not properly update all of its COW IDs using a datablock when depsgraph
decides to evaluate or un-evaluate it.
This would lead to evaluated IDs pointing to either:
- orig IDs when there was an evaluated version of those (annoying bug,
but not a crashing one).
- old address of previously evaluated IDs that no longer exists in the
depsgraph (causing the crash from the report e.g.).
This commit adds an extra step at the end of nodes building, that goes
over all of already existing IDs in the depsgraph to check whether they
do one of the two things above, and tag them for COW update if so.
NOTE: This only affects depsgraph (re-)building, not its evaluation.
This remains consistent with the fact that operations that may change
the depsgraph content (like Collection exclusion etc.) need to trigger a
rebuild.
NOTE: Performances: Worst case scenarii, like (un-)excluding a whole
character collection in a production file, lead to 5% to 10% extra
processing time in depsgraph building. Most of it comming from extra COW
processing (in depsgraph's update in `build_step_finalize`), the detection
loop itself only accounts for 1% to 2% of the whole building time.
Maniphest Tasks: T85752
Differential Revision: https://developer.blender.org/D10907
message(FATAL_ERROR"32 bit compiler detected, blender no longer provides pre-build libraries for 32 bit windows, please set the LIBDIR cmake variable to your own library folder")
setBUILD_CMAKE_ARGS=%BUILD_CMAKE_ARGS% -G "Visual Studio %BUILD_VS_VER%%BUILD_VS_YEAR%%BUILD_GENERATOR_POST%"%BUILD_PLATFORM_SELECT%%TESTS_CMAKE_ARGS%%CLANG_CMAKE_ARGS%%ASAN_CMAKE_ARGS%%PYDEBUG_CMAKE_ARGS%
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
