| Age | Commit message (Collapse) | Author |
|---|
|
This was crashing, when looking into a fix I noticed that it gave
hap-hazard results dissolving past forks in the parent/child hierarchy
arbitrarily following one chain.
This functionality is almost identical to "dissolve" which delimits
forks in the chain predictably.
So remove this in favor of dissolve (available from the delete menu).
|
|
Resolve issue noted in D5482, texture slots '+' icon for example
was noticeably off-center.
|
|
|
|
Match names between the redo popup and the menu item.
|
|
Without this, adjusting properties always re-initialized
from the view-vector.
|
|
|
|
This works for all selection modes,
include last below a separator since it's a specialized function.
Note that the previous commit was raising an exception as operator
properties don't support 'or'.
|
|
|
|
Edges were hard to see in some cases in edit-mesh vertex/face modes.
Since 804e90b42d72 alpha is handled differently,
update edge alpha to visually match 2.82.
|
|
|
|
Improved loading times for particles files by reading bigger chunks of data from the disk at once.
|
|
Required for collisions with moving rigid bodies. Otherwise the static optimization mode will be kept and the obstacles would be calculated only once at the beginning.
|
|
Edges with sharpness, seam and bevel are thicker and thus needs more offset
to not appear aliased.
Based on D5448 by @oficsu
|
|
The previous offset was done in view space. It is now done in NDC space
to avoid differences when adjusting the near/far plane distances.
|
|
Differential Revision: https://developer.blender.org/D7263
|
|
The button seems to behave more as I'd expect without these
additional checks. Previously, the button was often grayed out when
it was actually working.
Reviewers: ISS
Differential Revision: https://developer.blender.org/D7252
|
|
|
|
|
|
|
|
|
|
The Screw Modifier had a lower limit for the steps value, which not only
was inconsistent between render and viewport steps, but also was capped
to 2 in UI and also in the code internally.
|
|
Error in recent patch D6820
|
|
|
|
Was added when utf8 was originally introduced - for testing,
but is no longer needed.
|
|
|
|
|
|
Add NULL checks to other deform modifiers too.
|
|
|
|
Also fixed T75162
|
|
It's possible create a stroke with 0 points using python
|
|
|
|
Although indicated, the `rna_Object_active_vertex_group_set`
function was missing.
|
|
|
|
Removing the GHOST_kKeyUnknown check from processKeyEvent() produces
epeated unknown key events whenever a modifier key is held down, due
to the way ghost uses GHOST_kKeyUnknown as a filter value for modifier
key events.
Differential Revision: https://developer.blender.org/D7257
|
|
Result of poor shader pre-processing on Intel HD 4000 drivers
|
|
|
|
|
|
|
|
Differential Revision: https://developer.blender.org/D7250
|
|
This commit introduces a new mode for calculating the positions and
weights of the IK segments in the Pose Brush based on the Face Sets.
The first segment of the chain will always include all face sets inside
the brush radius and it will propagate until the boundary of the last
face sets added in the flood fill. Then consecutive connected face sets
are added to the chain until the chain length limit is reached or all
face sets of the mesh are already part of the chain.
This feature enables complete control over the pose brush origins in
case that is needed. Also, with this mode, the user can have a library
of base meshes with face sets already configured to get to the initial
pose as fast as possible.
Reviewed By: jbakker
Differential Revision: https://developer.blender.org/D7235
|
|
This makes more clear what this automasking operation does and helps to differenciate it from the future face sets boundary automasking.
Reviewed By: jbakker
Differential Revision: https://developer.blender.org/D7185
|
|
Previously, all vertices inside the brush radius were taken into account
equally when calculating the sculpt normal and area. This was causing
artifacts and unpredictable results with large brushes or meshes with
curvatures, as the strongest deformation point of all brushes is usually
in the center. By weighting the vertex normal and position towards the
center when sampling, all brushes should now behave in a more
predictable way in non-uniform surfaces.
Reviewed By: jbakker
Differential Revision: https://developer.blender.org/D6989
|
|
Last time I checked Face Sets were preserved in a more or less
predictable way when modifying the mesh with dyntopo. As it looks that
in some problems this may cause bugs and you can't see or use face sets
when modifying the topology of the mesh whith dyntopo active, it is
probably better to reset them when going from dyntopo to mesh. This way
you know that you are always going to get a predictable face sets state.
Reviewed By: jbakker
Maniphest Tasks: T74637
Differential Revision: https://developer.blender.org/D7099
|
|
|
|
This is annoying when one want to use system python and hence disables
the `WITH_PYTHON_INSTALL` options...
|
|
Update for D3078, I think it should be fixed
Benefits:
- after installing python 3.7 with `./build_files/build_environment/install_deps.sh`, user will be able to run `make bpy` without linking error:
- https://blender.stackexchange.com/questions/102933/a-working-guidance-for-building-blender-as-bpy-python-module
- https://stackoverflow.com/questions/36779834/compiling-blender-bpy-recompile-with-fpic
To prevent errors like `/opt/lib/python-3.7.4/bin/python3.7: error while loading shared libraries: libpython3.7m.so.1.0: cannot open shared object file: No such file or directory`, add python .so lib to ldconfig
Reviewed By: mont29
Differential Revision: https://developer.blender.org/D7177
|
|
Differential Revision: https://developer.blender.org/D7253
|
|
Reviewers: sebbas
Differential Revision: https://developer.blender.org/D7227
|
|
|
|
|
