Duplicates the UI code for the "Stroke" panel and makes
sure it checks for the new GPv3 type in the poll functions.
This will allow us to slightly adapt the UI without
interfering with GPv2.
The sculpting template was created in Blender 2.80 and never adjusted
since then. The alpha socket is added as after linking step of the
versioning.
The GPU material clipping was done before clipping and would crash when
sockets don't exist.
Pull Request: https://projects.blender.org/blender/blender/pulls/122706
The system context is expected to be bound prior to the Blender
can be properly initialized. Otherwise GHOST will be doing OpenGL
calls without system context bound.
This follows code from DRW_render_context_enable().
Pull Request: https://projects.blender.org/blender/blender/pulls/122708
A regression since #118841.
It is possible that the selected preference device is not found, in which
case a default-initialized DeviceInfo would have added to the list. This
device is set to CPU, but with differnet other fields (such as description)
compared to the actual CPU device.
Pull Request: https://projects.blender.org/blender/blender/pulls/122701
The issue would happened in any situation where the light
moves (update, animation, jitter) or have a lot of LOD
tagged by moving casters. In these cases, the actual
effective LOD min is bigger than the one from the UI which
results in shadow acnee artifacts (because the computed bias
is too small).
This patch saves the effective min LOD per tilemaps and
amend the `light.lod_min` to replace it by the min of
all tilemaps in used by one light.
This adds the smooth post process option from GPv2 to the
GPv3 draw tool.
This now smoothes the positions, opacities and radii. In GPv2 the radii were not smoothed for some reason.
There were some private methods in `PaintOperation` that
were only called in `PaintOperation::on_stroke_done`. These
don't need to be private methods and can just be static functions.
There was also some code that changed the selection in the
function that was supposed to only trim the end points,
so that functionality is moved to it's own function instead.
The issue is visible when adding an assert in the delta accessors of
the TranslateOperation operation (get_delta_x and get_delta_y), and
rendering compositor-nodes-desintegrate-wipe-01.blend (either command
line or F12, doesn't matter).
Seems that under certain circumstances the system might skip determining
the area of interest. For such cases ensure delta from the beginning of
the threaded code.
Pull Request: https://projects.blender.org/blender/blender/pulls/122686
This fixes#69535 and #98930.
We use a equi-solid-angle sampling algorithm for rectangular area lights,
but it is not particularly robust for small area lights (either small
in general and/or small because it's being viewed from grazing angles).
The actual sampling part is fine since it just gets clamped into the
valid area anyways, and the difference isn't notable for small lights.
However, we also need to compute the solid angle to get the sampling PDF,
and that computation is quite sensitive to numerical issues for small
values.
Therefore, this commit adds a fallback path for small values, which instead
uses the classic equi-area sampling PDF term times the area-to-solid-angle
Jacobian term. This approximation assumes that all points on the light have
the same distance and angle to the sampling point, which is of course not
strictly the case, but it's close enough for small area lights and better
than failing altogether.
Pull Request: https://projects.blender.org/blender/blender/pulls/122323
Reformulates some terms in the equi-solid-angle rectangle sampling code to
handle small area lamps better, and allows for some rounding error in the
check whether the sampled position is inside the area light.
Pull Request: https://projects.blender.org/blender/blender/pulls/122323
In the original paper, the falloff inside `bcone.theta_e` is assumed to
be `pi/2`, which is too large for spot light and resulted in an
overestimation near the cone boundary.
To address this issue, attenuate the energy of a spot light using the
minimal possible angle formed by the light axis and the shading point
when traversing the light tree.
Ref: #122362
Pull Request: https://projects.blender.org/blender/blender/pulls/122667
This commit changes how users can interact with handles:
Unselected handles are not drawn anymore by default. Handles are bit
thinner. If handle can't be selected, because strip is too small,
it is not drawn.
When hovering over strip handle a cursor is changed to represent a
handle shape. It is possible to select 2 handles at once if strips are
adjoined.
When tweak event happens on unselected handle, handle selection is lost
ater tweaking.
This behavior can be disabled in preferences:
Editing > Video Sequencer > Tweak Handles.
Moving strips with G key works same way as before.
Pull Request: https://projects.blender.org/blender/blender/pulls/109522
Previously, GPU denoisers were ignoring settings about render
configuration and were using any available GPU. With these changes,
GPU denoisers will use the device selected in Blender Cycles
settings.
This allows any GPU denoiser to be used with CPU rendering.
Pull Request: https://projects.blender.org/blender/blender/pulls/118841
With #120283 the Win32 codepath for registration creates an endless
loop. Command-line arguments are detected for registration that causes
another blender to launch with the same arguments, that repeats
forever. But the only time these arguments is used is when Blender has
already been specifically launched with the permissions needed to
register. So this means it should just do so and then exit.
Pull Request: https://projects.blender.org/blender/blender/pulls/122682
The default width for sidebars was last updated for 2.80 when columns
didn't have headings and checkboxes were right-aligned.
Some strings no longer fit especially when using headings, and in
translations. Make it slightly wider while still being relatively out of
the way in a 1920 × 1080 setup.
Detailed images in the pull request description.
Pull Request: https://projects.blender.org/blender/blender/pulls/121795
Brush cursor is being drawn even if the selected tool is non-brush in
GPv3 paint mode. To prevent this, use `ED_gpencil_brush_cursor_poll` for
`pc` poll function. Above poll function returns false if active tool is not
brush, this will avoid drawing of paint cursor in
`pc->draw`/`grease_pencil_brush_cursor_draw`.
Resolves#122541
Pull Request: https://projects.blender.org/blender/blender/pulls/122657
Since ebb61ef30f reroute nodes couldn't be inserted on links anymore,
when their socket's type is currently incompatible.
Because the sockets of reroute nodes change types dynamically
according to the sockets they are linked to, it should always be
allowed to insert reroutes onto links.
Pull Request: https://projects.blender.org/blender/blender/pulls/122637
Previously, the old bake was only overwritten file by file. This caused issues like
the one fixed in 7d77691a83. Also, it could result in invalid
bake data if some bake is canceled and the data on disk is a mix of an old and
new bake.
The material blend mode Clip and Opaque are not supported
in EEVEE-Next and need to be emulated using math nodes.
This PR makes it possible to automatically convert the
simplest cases. More complex cases need user interactions
to fix.
Rel #122489
Pull Request: https://projects.blender.org/blender/blender/pulls/122608
Make the hack to get an Action selector slightly less hacky, in order to
prevent an infinite redraw loop.
No functional changes, except that `window_manager.selected_action` is only
set when it needs to change.
Blender crashes when toggling the compositor device from CPU to GPU.
This is because when submitting the compositor job, the device might be
set to CPU, so GPU related data are not initialized in the job, but
between the time the job is submitted and the time it gets executed, the
device might change to GPU, which then tries to accept the uninitialized
GPU data from the job.
To fix this, we use the evaluated scene from the depsgraph since it
captures the state of the scene used in the job initialization.
Pull Request: https://projects.blender.org/blender/blender/pulls/122661
When the sequencer overlays menu was restructured in #121591, the new
Sequencer panel types did not include polling, which means that they
also show up in the Preview-only view type.
Details and screenshots in the pull request.
Pull Request: https://projects.blender.org/blender/blender/pulls/122638
Reorganization of the mesh Sculpt mode menu for simplicity and consistency.
Changes include:
- Moved transform operators into its own menu, consistent with other modes.
- Group Box, Lasso, Line, and Polyline Show/Hide operators
- Box, Lasso, Line, and Polyline Trim/Add operators moved to a new "Trim/Add"
- Cleanup and simplification of labels
Detailed explanation and example images in the pull request.
Pull Request: https://projects.blender.org/blender/blender/pulls/122437
Fix for #122285
`Viewport Render Image` feature uses full parameters of the 3D view
to capture the scene.
Rendering `camera view` is a special case: viewport zoom/pan
parameters should be ignored in order to match render.
Pull Request: https://projects.blender.org/blender/blender/pulls/122467
This fixes an issue reported when launching Blender with
`--debug-gpu`. The newly added "thick wires" feature for
bone shapes caused that problem.
The issue was that the shader definitions used
the depth tex as a sampler and at the same time wrote to it.
The fix is to remove the code that uses the depth texture.
This is just the simplest solution for now, and it doesn't change
the visual appearance of the bone shapes.
------
Note: the code I copied from was `overlay_edit_mesh_frag.glsl`.
So I assume the depth sampler is needed, but the pipeline
for bone shapes isn't set up to handle that.
Pull Request: https://projects.blender.org/blender/blender/pulls/122558
