The collection property `loop_triangles` was given the RNA type
`MeshLoopTriangle` (the type of the collection's items), instead
of `MeshLoopTriangles` (the actual expected RNA collection/array type).
The cutest, tiniest typo mistake in RNA code, leading to some complete
non-sense... It's fairly amazing that this did not cause more severe
issues actually.
The current code allocates and transfers a lot of memory to the GPU,
but only a small portion of this memory is actually used.
In addition, the code calls many costly gl operations during the
caching process.
This commit significantly reduce the amount of memory by allocating
and transferring a flat array without pads to the GPU.
It also calls as little as possible the gl operations during the cache.
This code also simulate a billinear filter `GL_LINEAR` using a 1D texture.
**Average drawing time:**
|before:|0.00003184 sec
|now:|0.00001943 sec
|fac:|1.6385156675048407
**5 worst times:**
|before:|[0.001075, 0.001433, 0.002143, 0.002915, 0.003242]
|now:|[0.00094, 0.000993, 0.001502, 0.002284, 0.002328]
Differential Revision: https://developer.blender.org/D6886
Since this control is set as you drag in the viewport anyway, this amount value is effectively useless.
It was only recently added by mistake with the Bevel GSOC.
- Move gizmo cursor check into ED_region_cursor_set so the result of
calling this function is the same as flagging for cursor update.
- Use tagging in ui_popup_block_remove which avoids adding a
mouse-move event in case the cursor needs to be changed again.
- Fixed uninitialized result used in DynamicMusic::seek().
The comment to this function says false is returned if the handle
is invalid, while in practice non-initialized value will be returned.
- Spelling typos in comment.
- Silence -Wdelete-non-abstract-non-virtual-dtor warning.
Differential Revision: https://developer.blender.org/D6896
File Browser using Windows file attributes for decorating and hiding items.
Differential Revision: https://developer.blender.org/D6816
Reviewed by Campbell Barton
This fixes a threading issue (T73593) between drivers that write to the
same memory address. Driver nodes in the depsgraph now get relations to
each other in order to ensure serialisation.
These relations are only added between drivers that target the same
struct in RNA, which is determined by removing everything after the last
period. For example, a driver with data path
`pose.bones["Arm_L"].rotation_euler[2]` will be grouped with all other
drivers on that datablock with a data path that starts with
`pose.bones["Arm_L"]` to form a 'driver group'.
To find a suitable relation within such a driver group, say the relation
(from → to), a depth-first search is performed (turned out to be
marginally faster than a breadth-first in my test case) to see whether
this will create a cycle, and to see whether there already is such a
connection (direct or transitive). This is done by recursively
inspecting the incoming connections of the 'to' node and thereby walking
from it towards the 'from' node. This is an order of magnitde faster
than inspecting the outgoing connections of the 'from' node.
This approach generalises the special case for array properties, so the
code to support that special case has been removed from
`DepsgraphRelationBuilder::build_animdata_drivers()`.
A test on the Spring rig [1] shows that this process adds approximately
8% to the build time of the dependency graph. In my test case, it takes
28 ms for this process on a total 329 ms construction time. However,
since it also made some code obsolete, it only adds 24 ms (=8%) to the
construction time. I have experimented with a simple cache to keep track
of known-connected (from, to) node pairs, but this did not significantly
improve the timing.
Note that animation data and drivers are already connected by a
relation, which means that animating a field and also changing it with a
driver will not cause conflicts.
[1] https://cloud.blender.org/p/spring/5d30a1076249366fa1939cf1
Differential Revision: https://developer.blender.org/D6905
Reviewed By: sergey, mont29
Updates include:
- A fix from Jacques that changed the loop order in the mesh creation function (the fix speeds up the function significantly due to fewer cache misses).
- Some of the grid copy helper functions are now multithreaded.
- A fix for Windows file IO. Now it possible to load files with non ASCII characters on Windows too.
bone constraints settings
'UI_context_copy_to_selected_list()' was working fine for constraints on
objects (would properly get the path and list of selected objects) and
also for bone settings (would properly get the path and list of bones).
There was no special case for constraints on bones though, so code would
work on selected objects (not selected bones).
So we now get the right bones and resolve the path to the constraint
setting _from the bones_.
Maniphest Tasks: T74006
Differential Revision: https://developer.blender.org/D6897
This partially fixes T73593.
The `add_relation(driver_key, property_entry_key, ...);` call can fail
in the following situation:
- A collection is linked, and instanced into the scene by an Empty.
- The collection contains an object with a driver on its `hide_render` or
`hide_viewport` property.
As the object doesn't exist as a real object in the scene, it's added with
`base_index=-1` to the depsgraph (see `DepsgraphNodeBuilder::build_collection()`).
As a result the node for syncing the restrictflags back to the base
isn't present in the depsgraph, and the `add_relation()` call failed.
This commit fixes the warning, simply by not attempting to add the
offending relation.
This patch adds new render passes to EEVEE. These passes include:
* Emission
* Diffuse Light
* Diffuse Color
* Glossy Light
* Glossy Color
* Environment
* Volume Scattering
* Volume Transmission
* Bloom
* Shadow
With these passes it will be possible to use EEVEE effectively for
compositing. During development we kept a close eye on how to get similar
results compared to cycles render passes there are some differences that
are related to how EEVEE works. For EEVEE we combined the passes to
`Diffuse` and `Specular`. There are no transmittance or sss passes anymore.
Cycles will be changed accordingly.
Cycles volume transmittance is added to multiple surface col passes. For
EEVEE we left the volume transmittance as a separate pass.
Known Limitations
* All materials that use alpha blending will not be rendered in the render
passes. Other transparency modes are supported.
* More GPU memory is required to store the render passes. When rendering
a HD image with all render passes enabled at max extra 570MB GPU memory is
required.
Implementation Details
An overview of render passes have been described in
https://wiki.blender.org/wiki/Source/Render/EEVEE/RenderPasses
Future Developments
* In this implementation the materials are re-rendered for Diffuse/Glossy
and Emission passes. We could use multi target rendering to improve the
render speed.
* Other passes can be added later
* Don't render material based passes when only requesting AO or Shadow.
* Add more passes to the system. These could include Cryptomatte, AOV's, Vector,
ObjectID, MaterialID, UV.
Reviewed By: Clément Foucault
Differential Revision: https://developer.blender.org/D6331
editmode
This was already supported in "Select" & "Deselect" but not in "Assign".
So similar to rB6b39dc7672eb, we now check if the material corresponding
to the currently selected material slot is found in other objects
materials and assign this (instead of always assigning their 'own'
actcol).
Maniphest Tasks: T73871
Differential Revision: https://developer.blender.org/D6869
PAINT_TEXTURE)
This fails because some tool contexts define their tools with functions
[see the following list for context that fail]:
- PARTICLE (_defs_particle.generate_from_brushes)
- SCULPT (_defs_sculpt.generate_from_brushes)
- PAINT_TEXTURE (_defs_texture_paint.generate_from_brushes)
- PAINT_VERTEX (_defs_vertex_paint.generate_from_brushes)
- PAINT_WEIGHT (_defs_weight_paint.generate_from_brushes)
- PAINT_GPENCIL (_defs_gpencil_paint.generate_from_brushes)
- SCULPT_GPENCIL (_defs_gpencil_sculpt.generate_from_brushes)
- WEIGHT_GPENCIL (_defs_gpencil_weight.generate_from_brushes)
ToolSelectPanelHelper._tools_flatten() is usually called with
cls.tools_from_context(context) [that already yields from the function].
But when registering a tool, _tools_flatten() will still give back this
function, not a ToolDef - and we cannot get a bl_idname from that.
Now check for this and yield None in that case.
Also share logic across all tool_flatten functions:
- _tools_flatten
- _tools_flatten_with_tool_index
- _tools_flatten_with_keymap
Maniphest Tasks: T63892
Differential Revision: https://developer.blender.org/D6763
The latest versions of Embree support Catmull-Rom splines
which use less memory than the previously used Hermite splines.
The representation is also much closer to Cycles own data structures
and can hopefully be unified in the future for more memory savings.
Memory savings using Victor benchmark scene:
Compared to previous Embree: ~400MB
Compared to Cycles' native BVH: ~1GB
Consequences of own refactor of usercount handling in readfile.c
(rB367ecff15d74).
Not super happy to have to call that function twice, but that should be
OK (not a real overhead here anyway).
The functionality of the mask modifier remains unchanged.
This patch updates the mask modifier so that it uses C++.
The manual memory management has been replaced with proper containers.
The large `applyModifier` function has been splitup into multiple smaller functions.
A large speedup is achieved by using simple arrays instead of hash tables in multiple places.
In my performance test file the playback speed increased from 1.1 to 5.1 fps on my laptop.
Reviewers: campbellbarton, brecht
Differential Revision: https://developer.blender.org/D6779
