Age | Commit message (Collapse) | Author |
|
Includes unwanted changes
This reverts commit 46e049d0ce2bce2f53ddc41a0dbbea2969d00a5d.
|
|
This patch implements the vector types (i.e:`float2`) by making heavy
usage of templating. All vector functions are now outside of the vector
classes (inside the `blender::math` namespace) and are not vector size
dependent for the most part.
In the ongoing effort to make shaders less GL centric, we are aiming
to share more code between GLSL and C++ to avoid code duplication.
####Motivations:
- We are aiming to share UBO and SSBO structures between GLSL and C++.
This means we will use many of the existing vector types and others
we currently don't have (uintX, intX). All these variations were
asking for many more code duplication.
- Deduplicate existing code which is duplicated for each vector size.
- We also want to share small functions. Which means that vector
functions should be static and not in the class namespace.
- Reduce friction to use these types in new projects due to their
incompleteness.
- The current state of the `BLI_(float|double|mpq)(2|3|4).hh` is a
bit of a let down. Most clases are incomplete, out of sync with each
others with different codestyles, and some functions that should be
static are not (i.e: `float3::reflect()`).
####Upsides:
- Still support `.x, .y, .z, .w` for readability.
- Compact, readable and easilly extendable.
- All of the vector functions are available for all the vectors types
and can be restricted to certain types. Also template specialization
let us define exception for special class (like mpq).
- With optimization ON, the compiler unroll the loops and performance
is the same.
####Downsides:
- Might impact debugability. Though I would arge that the bugs are
rarelly caused by the vector class itself (since the operations are
quite trivial) but by the type conversions.
- Might impact compile time. I did not saw a significant impact since
the usage is not really widespread.
- Functions needs to be rewritten to support arbitrary vector length.
For instance, one can't call `len_squared_v3v3` in
`math::length_squared()` and call it a day.
- Type cast does not work with the template version of the `math::`
vector functions. Meaning you need to manually cast `float *` and
`(float *)[3]` to `float3` for the function calls.
i.e: `math::distance_squared(float3(nearest.co), positions[i]);`
- Some parts might loose in readability:
`float3::dot(v1.normalized(), v2.normalized())`
becoming
`math::dot(math::normalize(v1), math::normalize(v2))`
But I propose, when appropriate, to use
`using namespace blender::math;` on function local or file scope to
increase readability.
`dot(normalize(v1), normalize(v2))`
####Consideration:
- Include back `.length()` method. It is quite handy and is more C++
oriented.
- I considered the GLM library as a candidate for replacement. It felt
like too much for what we need and would be difficult to extend / modify
to our needs.
- I used Macros to reduce code in operators declaration and potential
copy paste bugs. This could reduce debugability and could be reverted.
- This touches `delaunay_2d.cc` and the intersection code. I would like
to know @howardt opinion on the matter.
- The `noexcept` on the copy constructor of `mpq(2|3)` is being removed.
But according to @JacquesLucke it is not a real problem for now.
I would like to give a huge thanks to @JacquesLucke who helped during this
and pushed me to reduce the duplication further.
Reviewed By: brecht, sergey, JacquesLucke
Differential Revision: https://developer.blender.org/D13791
|
|
Reverted because the commit removes a lot of commits.
This reverts commit a2c1c368af48644fa8995ecbe7138cc0d7900c30.
|
|
This patch implements the vector types (i.e:float2) by making heavy
usage of templating. All vector functions are now outside of the vector
classes (inside the blender::math namespace) and are not vector size
dependent for the most part.
In the ongoing effort to make shaders less GL centric, we are aiming
to share more code between GLSL and C++ to avoid code duplication.
Motivations:
- We are aiming to share UBO and SSBO structures between GLSL and C++.
This means we will use many of the existing vector types and others we
currently don't have (uintX, intX). All these variations were asking
for many more code duplication.
- Deduplicate existing code which is duplicated for each vector size.
- We also want to share small functions. Which means that vector functions
should be static and not in the class namespace.
- Reduce friction to use these types in new projects due to their
incompleteness.
- The current state of the BLI_(float|double|mpq)(2|3|4).hh is a bit of a
let down. Most clases are incomplete, out of sync with each others with
different codestyles, and some functions that should be static are not
(i.e: float3::reflect()).
Upsides:
- Still support .x, .y, .z, .w for readability.
- Compact, readable and easilly extendable.
- All of the vector functions are available for all the vectors types and
can be restricted to certain types. Also template specialization let us
define exception for special class (like mpq).
- With optimization ON, the compiler unroll the loops and performance is
the same.
Downsides:
- Might impact debugability. Though I would arge that the bugs are rarelly
caused by the vector class itself (since the operations are quite trivial)
but by the type conversions.
- Might impact compile time. I did not saw a significant impact since the
usage is not really widespread.
- Functions needs to be rewritten to support arbitrary vector length. For
instance, one can't call len_squared_v3v3 in math::length_squared() and
call it a day.
- Type cast does not work with the template version of the math:: vector
functions. Meaning you need to manually cast float * and (float *)[3] to
float3 for the function calls.
i.e: math::distance_squared(float3(nearest.co), positions[i]);
- Some parts might loose in readability:
float3::dot(v1.normalized(), v2.normalized())
becoming
math::dot(math::normalize(v1), math::normalize(v2))
But I propose, when appropriate, to use
using namespace blender::math; on function local or file scope to
increase readability. dot(normalize(v1), normalize(v2))
Consideration:
- Include back .length() method. It is quite handy and is more C++
oriented.
- I considered the GLM library as a candidate for replacement.
It felt like too much for what we need and would be difficult to
extend / modify to our needs.
- I used Macros to reduce code in operators declaration and potential
copy paste bugs. This could reduce debugability and could be reverted.
- This touches delaunay_2d.cc and the intersection code. I would like to
know @Howard Trickey (howardt) opinion on the matter.
- The noexcept on the copy constructor of mpq(2|3) is being removed.
But according to @Jacques Lucke (JacquesLucke) it is not a real problem
for now.
I would like to give a huge thanks to @Jacques Lucke (JacquesLucke) who
helped during this and pushed me to reduce the duplication further.
Reviewed By: brecht, sergey, JacquesLucke
Differential Revision: http://developer.blender.org/D13791
|
|
The new triangulation mode for quads is the opposite of the current default
shortest diagonal mode. It is optimal for cloth simulations using quad meshes.
Differential Revision: http://developer.blender.org/D13777
|
|
This evaluator is used in order to evaluate subdivision at render time, allowing for
faster renders of meshes with a subdivision surface modifier placed at the last
position in the modifier list.
When evaluating the subsurf modifier, we detect whether we can delegate evaluation
to the draw code. If so, the subdivision is first evaluated on the GPU using our own
custom evaluator (only the coarse data needs to be initially sent to the GPU), then,
buffers for the final `MeshBufferCache` are filled on the GPU using a set of
compute shaders. However, some buffers are still filled on the CPU side, if doing so
on the GPU is impractical (e.g. the line adjacency buffer used for x-ray, whose
logic is hardly GPU compatible).
This is done at the mesh buffer extraction level so that the result can be readily used
in the various OpenGL engines, without having to write custom geometry or tesselation
shaders.
We use our own subdivision evaluation shaders, instead of OpenSubDiv's vanilla one, in
order to control the data layout, and interpolation. For example, we store vertex colors
as compressed 16-bit integers, while OpenSubDiv's default evaluator only work for float
types.
In order to still access the modified geometry on the CPU side, for use in modifiers
or transform operators, a dedicated wrapper type is added `MESH_WRAPPER_TYPE_SUBD`.
Subdivision will be lazily evaluated via `BKE_object_get_evaluated_mesh` which will
create such a wrapper if possible. If the final subdivision surface is not needed on
the CPU side, `BKE_object_get_evaluated_mesh_no_subsurf` should be used.
Enabling or disabling GPU subdivision can be done through the user preferences (under
Viewport -> Subdivision).
See patch description for benchmarks.
Reviewed By: campbellbarton, jbakker, fclem, brecht, #eevee_viewport
Differential Revision: https://developer.blender.org/D12406
|
|
Differentiate doc-strings from title/section text.
Also use explicit doxygen references to struct members
so it's not ambiguous which member is being referenced.
Note that these changes aren't complete (some files weren't touched).
|
|
The mirror modifiers merge option caused unnecessary re-ordering
to the vertex array with original vertices merging into their copies.
While this wasn't an error, it meant creating a 1:1 mapping from input
vertices to their final output wasn't reliable (when looping over
vertices first to last) as is done in
BKE_editmesh_vert_coords_when_deformed.
As merging in either direction is supported, keep the source meshes
vertices in-order since it allows the vertex coordinates to be extracted.
NOTE: Since this change introduce issues for some cases (e.g. bound
modifiers like SurfaceDeform), this change is only applied to newly
created modifiers, existing ones will still use the old incorrect merge
behavior.
Reviewed By: @brecht
Maniphest Tasks: T93321, T91444
Differential Revision: https://developer.blender.org/D13355
|
|
|
|
Previously fluid simulation and Alembic modifiers had a dedicated function
to query the velocity for motion blur. Now use a more generic system where
those modifiers output a velocity attribute.
Advantages:
* Geometry and particle nodes can output velocity through the same mechanism,
or read the attribute coming from earlier modifiers.
* The velocity can be preserved through modifiers like subdivision surface or
auto smooth.
* USD and Alembic previously only output velocity from fluid simulation, now
they work with velocity from other sources too.
* Simplifies the code for renderers like Cycles and exporters like
Alembic and USD.
This breaks compatibility:
* External renderers and exporters accessing these velocities through the
Python API now need to use the attribute instead.
* Existing modifier node setups that create an attribute named "velocity"
will render differently with motion blur.
Differential Revision: https://developer.blender.org/D12305
|
|
Allow blending the imported cache with the modifiers stack above the
MeshCache modifier.
This is particularly useful for instance when dealing with cloth
simulations performed in another software, where some parts of the cloth
are completely pinned (non-simulated, following the armature). Indeed,
this would allow modifying the animation in some areas without having to
rebake the other parts or the cloth, resulting in a much more flexible
workflow.
Reviewed By: #modeling, campbellbarton, mont29
Ref D9898
|
|
Add an option to the mask modifier to use the vertex weights to generate
smooth in between geometry, instead of just deleting non complete faces.
This can be used to make all sorts of smooth dissolve animations
directly with geometry, which are usually hacked together with shaders.
It also allows for implicit function plotting using geometry nodes and
boolean like operations on non manifold geometry with the proximity
modifier.
Reviewed By: campbellbarton
Ref D10979
|
|
Some of the dna structs were not properly
aligned for 32 bit builds causing issues
for some of the 32 platforms Debian builds
for.
Reviewed By: sergey, brecht
Differential Revision: https://developer.blender.org/D9389
|
|
|
|
This improve the cloth modeling workflow by allowing you to weld only the
edges that are used for the sewing forces.
Reviewed By: mano-wii, weasel
Differential Revision: https://developer.blender.org/D10710
|
|
When a vertex group is used to limit the influence of the modifier
to a subset of vertices, binding data for vertices with zero weight
is not needed. This wastes memory, disk space and CPU cycles.
If the vertex group contents is known to be final and constant,
it is reasonable to optimize by only storing data group vertices.
This has to be an option in case the group can change.
Supporting this requires adding a vertex index field and spliting
the vertex count into mesh and bind variants, but both happen to
fit in available padding. The old numverts field is renamed to the
new bound vertex count field to maintain the array length invariant.
Versioning is used to initialize the other new fields.
If a file with sparse binding is opened in an old blender version,
it is corrupted into a non-sparse bind with vertex count mismatch,
preventing the modifier from working until rebind.
Differential Revision: https://developer.blender.org/D11924
|
|
Many ui features for geometry nodes need access to information generated
during evaluation:
* Node warnings.
* Attribute search.
* Viewer node.
* Socket inspection (not in master yet).
The way we logged the required information before had some disadvantages:
* Viewer node used a completely separate system from node warnings and
attribute search.
* Most of the context of logged information is lost when e.g. the same node
group is used multiple times.
* A global lock was needed every time something is logged.
This new implementation solves these problems:
* All four mentioned ui features use the same underlying logging system.
* All context information for logged values is kept intact.
* Every thread has its own local logger. The logged informatiton is combined
in the end.
Differential Revision: https://developer.blender.org/D11785
|
|
Also use doxy style function reference `#` prefix chars when
referencing identifiers.
|
|
|
|
|
|
The `bisect_distance` in the mirror modifier was hard-coded to `0.001`.
This would result in some unexpected behavior like vertices close
to the mirror plane being deleted or merged.
The fix now adds a parameter to the mirror modifier to expose the
bisect distance to the user. The default is set to the previous
hard-coded value to not "change" previous files.
Ref D10201
|
|
|
|
The commit rB6f63417b500d that made exact boolean work on meshes
with holes (like Suzanne) unfortunately dramatically slowed things
down on other non-manifold meshes that don't have holes and didn't
need the per-triangle insideness test.
This adds a hole_tolerant parameter, false by default, that the user
can enable to get good results on non-manifold meshes with holes.
Using false for this parameter speeds up the time from 90 seconds
to 10 seconds on an example with 1.2M triangles.
|
|
|
|
|
|
|
|
Things like pointers to particle systems, or softbody data being stored
outside of its modifier, make it impossible for internal modifier copy
data code to be self-contained currently. It requires extra processing.
In existing code this was handled in several different places, in
several ways, and alltogether fairly inconsistently. Some cases were
even not properly handled, causing e.g. crashes as in T82945.
This commit addresses those issues by:
* Adding comments about the hackish/unsafe parts `psys` implies when
copying some modifier data (since we need to ensure particle system
copying and remapping of those pointers separately).
* Adding as-best-as-possible handling of those cases to
`BKE_object_copy_modifier` (note that it remains fragile, but is
expected to behave 'good enough' in any practical usecase).
* Remove special handling for specific editor code
(`copy_or_reuse_particle_system`). This should never have been
accepted in ED code area, and is now handled by
`BKE_object_copy_modifier`.
* Factorize copying of the whole modifier stack into new
`BKE_object_modifier_stack_copy`, now used by both `object_copy_data`
and `BKE_object_link_modifiers`.
Note that this implies that `BKE_object_copy_modifier` and
`BKE_object_copy_gpencil_modifier` are now to be used exclusively to
copy single modifiers. Full modifier stack copy should always use
`BKE_object_modifier_stack_copy` instead.
Fix T82945: Crash when dragging modifiers in Outliner.
Maniphest Tasks: T82945
Differential Revision: https://developer.blender.org/D10148
|
|
|
|
Implement improvement from T73139 for merging along edges.
It is now called "Connected" mode, while the default is called "All".
With the recent performance improvement, the Connected Mode is in some
cases only double the speed than the usual merge all strategy but in
other cases it may be even faster. The bottleneck is somewhere further
down the line of merging geometry.
The motivation for this patch came from T80897, because the merging in
complex solidify is making it very slow.
Now merging can be removed from solidify without greater consequences,
as this is just a quicker and more advanced algorithm to do the same
thing that solidify currently does slowly.
Reviewed by: mano-wii, campbellbarton
Ref D8966
|
|
Rerun `make format`.
No functional changes.
|
|
This commit adds functions to set and get the object's active
modifier, which is stored as a flag in the ModifierData struct,
similar to constraints. This will be used to set the context in
the node editor. There are no visible changes in this commit.
Similar to how the node editor context works for materials, this commit
makes the node group displayed in the node editor depend on the active
object and its active modifier. To keep the node group from changing,
just pin the node group in the header.
* Shortcuts performed while there is an active modifier will affect
only that modifier (the exception is the A to expand the modifiers).
* Clicking anywhere on the empty space in a modifier's panel will make it active.
These changes require some refactoring of object modifier code. First
is splitting up the modifier property invoke callback, which now needs
to be able to get the active modifier separately from the hovered
modifier for the different operators.
Second is a change to removing modifiers, where there is now a separate
function to remove a modifier from an object's list, in order to handle
changing the active.
Finally, the panel handler needs a small tweak so that this "click in panel"
event can be handled afterwards.
|
|
This is the initial merge from the geometry-nodes branch.
Nodes:
* Attribute Math
* Boolean
* Edge Split
* Float Compare
* Object Info
* Point Distribute
* Point Instance
* Random Attribute
* Random Float
* Subdivision Surface
* Transform
* Triangulate
It includes the initial evaluation of geometry node groups in the Geometry Nodes modifier.
Notes on the Generic attribute access API
The API adds an indirection for attribute access. That has the following benefits:
* Most code does not have to care about how an attribute is stored internally.
This is mainly necessary, because we have to deal with "legacy" attributes
such as vertex weights and attributes that are embedded into other structs
such as vertex positions.
* When reading from an attribute, we generally don't care what domain the
attribute is stored on. So we want to abstract away the interpolation that
that adapts attributes from one domain to another domain (this is not
actually implemented yet).
Other possible improvements for later iterations include:
* Actually implement interpolation between domains.
* Don't use inheritance for the different attribute types. A single class for read
access and one for write access might be enough, because we know all the ways
in which attributes are stored internally. We don't want more different internal
structures in the future. On the contrary, ideally we can consolidate the different
storage formats in the future to reduce the need for this indirection.
* Remove the need for heap allocations when creating attribute accessors.
It includes commits from:
* Dalai Felinto
* Hans Goudey
* Jacques Lucke
* Léo Depoix
|
|
The Vertex Weight Edit Modifier already got the Custom Curve, there was no
real reason for the proximity not to have it as well.
With some fixes by Bastien Montagne (@mont29).
Reviewed By: mont29
Differential Revision: https://developer.blender.org/D9594
|
|
This commit uses an enum to access expansion for specific panels for
each modifier, constraint, etc. Even though these values are quite simple,
this can help make the code more explicit when the ui_expand_flag is
accessed directly. Also update comments about this bitfield to make
them consistent.
|
|
The simple subdivision as a type only causes issues like no-continuous
normals across edges, inability to reliably switch the type and things
like this.
The new subdivision operators supports wider variety of how to add
details to the model, which are more powerful than a single one-time
decision on the subdivision type.
The versioning code is adjusting topology converter to specify all
edges as infinitely sharp. The reason for this (instead of using
settings.is_simple) is because in a longer term the simple subdivision
will be removed from Subsurf modifier as well, and will be replaced
with more efficient bmesh-based modifier.
This is finished up version of D8436.
Differential Revision: https://developer.blender.org/D9350
|
|
These flags shouldn't be used except in versioning code.
|
|
The design for how we approach the "Everything Nodes" project
has changed. We will focus on a different part of the project initially.
While future me will likely refer back to some of the code I remove here,
there is no point in keeping this code around in master currently.
It would just confuse other developers working on the project.
This does not remove the simulation modifier and data block. Those are
just cleaned up, so that the boilerplate code can be reused in the future.
|
|
This modifier is the opposite of the recently added Mesh to Volume modifier.
It converts the "surface" of a volume into a mesh. The "surface" is defined
by a threshold value. All voxels with a density higher than the threshold
are considered to be inside the volume, while all others will be outside.
By default, the resolution of the generated mesh depends on the voxel
size of the volume grid. The resolution can be customized. It should be
noted that a lower resolution might not make this modifier faster. This
is because we have to downsample the openvdb grid, which isn't a cheap
operation.
Converting a mesh to a volume and then back to a mesh is possible,
but it does require two separate mesh objects for now.
Reviewers: brecht
Differential Revision: https://developer.blender.org/D9141
|
|
|
|
This modifier uses a 3D texture to displace a volume.
For now, this can only use the previously existing texture system,
because we do not have a better alternative yet. Still, the results
can be quite good and interesting. See D9075 for some examples.
Reviewers: brecht, simonthommes
Differential Revision: https://developer.blender.org/D9075
|
|
This adds the option to either smooth the entire boundary, or to keep
corners sharp, for the Subdivision Surface and Multiresolution modifiers.
This mainly helps with compatibility with other software. The default
behavior remains to smooth the entire boundary.
Differential Revision: https://developer.blender.org/D8485
|
|
This makes subdivision surfaces compatible with the old subdivision
surface modifier and other applications that do not use the limit surface.
This option is available on the Subdivision Surface modifier.
Differential Revision: https://developer.blender.org/D8413
|
|
Add an option to disable Alembic vertex interpolation.
Bump subversion from 5 to 6.
Alembic stores mesh samples at specific time keys; when a frame in
Blender maps to a timecode between two samples, Blender will interpolate
the mesh vertex positions. This interpolation only happens when the mesh
has a constant topology, but sometimes this was not detected properly
when the vertices change order, but the number of mesh elements remains
the same. This would result in a mesh with jumbled up vertices (T71981).
With this patch, users have the ability to disable vertex interpolation.
An alternative would be to have better detection of topology changes,
but that that'll cause a considerable slowdown.
Maniphest Tasks: T71981
Differential Revision: https://developer.blender.org/D9041
|
|
This modifier can only be added to Volume objects. It takes a mesh
as input and generates a "density" grid near the surface or in
the enclosed volume.
Ref T73201.
Reviewers: brecht
Differential Revision: https://developer.blender.org/D9032
|
|
This commit replaces the BVH Tree currently used by the Weld Modifier
with the KD Tree used by `Merge > By Distance`.
This changes the result of the Weld Modifier to more closely match
`Merge > By Distance`.
There is also a big performance advantage.
Here is an overview (models in D8995):
| 2.90 (Duplicate Limit = 0) | 2.90 (Duplicate Limit = 1) | master (BVH) (Duplicate Limit = 1) | patch (KD) |
| 1.69s| 0.17s | 0.12s | 0.029s |
Differential Revision: https://developer.blender.org/D8995
|
|
Also added code so that exact solver does the whole collection at once.
This patch allows users to use a collection (as an alternative to Object)
for the boolean modifier operand, and therefore get rid of a long modifier stack.
|
|
With this option, self-intersections in either or both operands
will be handled properly (if both sides are piecewise winding
number constant, and maybe some other cases too).
In the Boolean tool, this flag was there already but the code
forced a unary operation in that case; this commit corrects it
to make a binary operation. This flag makes the code slower, which
is why it is an option and not an always-on thing.
|
|
This is for design task T67744, Boolean Redesign.
It adds a choice of solver to the Boolean modifier and the
Intersect (Boolean) and Intersect (Knife) tools.
The 'Fast' choice is the current Bmesh boolean.
The new 'Exact' choice is a more advanced algorithm that supports
overlapping geometry and uses more robust calculations, but is
slower than the Fast choice.
The default with this commit is set to 'Exact'. We can decide before
the 2.91 release whether or not this is the right choice, but this
choice now will get us more testing and feedback on the new code.
|
|
This adds an option to the Multires modifier to sculpt directly on the
base mesh while previewing the displacement of a higher subdivisions
level. What this does it considering Multires as a regular modifier
without exposing the grid displacement to sculpt mode.
This allows to see the propagation happening in real time, which enables
to use complex tools like Cloth or Pose in much higher resolutions and
without surface noise and artifacts.
Reviewed By: sergey, Severin
Differential Revision: https://developer.blender.org/D8555
|
|
Allows to keep track of modifiers, which is required, for example,
for runtime data preservation in depsgraph.
|