Age | Commit message (Collapse) | Author |
|
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
|
|
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
|
|
This commit moves the normal field input to `BKE_geometry_set.hh`
from the node file so that normals can be used as an implicit input to
other nodes.
Differential Revision: https://developer.blender.org/D13779
|
|
MSVC used to warn about const mismatch for arguments passed by value.
Remove these as newer versions of MSVC no longer show this warning.
|
|
Compare the start of the range to zero to figure out whether the
indices for the instances to keep starts at zero. Also rename the
selection argument, since it made it seem like the selected indices
should be removed rather than kept.
|
|
Ever since the instance domain was added, this was exposed, it just
didn't do anything. This patch implements the instances domain in the
delete and separate geometry nodes, where it acts on the top-level
instances.
We act on a mutable instances input, with the idea that eventually
copy on write attribute layers will make this less expensive. It also
allows us to keep the instance references in place and to do less
work in some situations.
Ref T93554
Differential Revision: https://developer.blender.org/D13565
|
|
`GeometrySet::compute_boundbox_without_instances` may not initialize min
max in some cases such as meshes without vertices.
This can result in a Bounding Box with impossible dimensions
(min=FLT_MAX, max=-FLT_MAX).
So repeat the same solution seen in `BKE_object_boundbox_calc_from_mesh`
and set boundbox values to zero.
Reviewed By: HooglyBoogly
Differential Revision: https://developer.blender.org/D13664
|
|
Most of the comment block is similar to the text in the source
code documentation wiki. It's helpful to have some text in
a header file too, so it's closer for programmers already looking
at the code.
This also uses more consistent syntax and wording in the comments
about the attribute API in `GeometryComponent`.
Ref T93753
Differential Revision: https://developer.blender.org/D13661
|
|
Point cloud is two separate words, this just changes comments in
a few places where we were inconsistent. A small wording change
to another comment is also included.
|
|
|
|
This adds a bit more information to `GeometrySet` and each of the
geometry components. There is probably still more that can be written,
but this includes the most important information that I could think of.
I'd like to include some more general information about the
attribute API in a separate patch.
Differential Revision: https://developer.blender.org/D13501
|
|
- Added space below non doc-string comments to make it clear
these aren't comments for the symbols directly below them.
- Use doxy sections for some headers.
- Minor improvements to doc-strings.
Ref T92709
|
|
Most of our field inputs are currently specific to geometry. This patch introduces
a new `GeometryFieldInput` that reduces the overhead of adding new geometry
field input.
Differential Revision: https://developer.blender.org/D13489
|
|
`GeometrySet` contains at most one component of each type.
Previously, a map was used to make sure that each component
type only exists once. The overhead of a map (especially with
inline storage) is rather large though. Since all component types
are known at compile time and the number of types is low,
a simple `std::array` works as well.
Some benefits of using `std::array` here:
* Looking up the component of a specific type is a bit faster.
* The size of `GeometrySet` becomes much smaller from 192 to 40 bytes.
* Debugging a `GeometrySet` in many tools becomes simpler because
one can easily see which components exists and which don't
|
|
Now that we can store any dynamic attribute on the instances component,
we don't need the special case for `id`, it can just be handled by the
generic attribute storage. Mostly this just allows removing a bunch
of redundant code.
I had to add a null check for `update_custom_data_pointers` because
the instances component doesn't have any pointers to inside of
custom data.
Differential Revision: https://developer.blender.org/D13430
|
|
Adds an attribute provider for instance attributes.
A new domain `ATTR_DOMAIN_INSTANCE` is implemented.
Instance attributes are not yet realized correctly.
Differential Revision: D13149
|
|
Goals of this refactor:
* Simplify creating virtual arrays.
* Simplify passing virtual arrays around.
* Simplify converting between typed and generic virtual arrays.
* Reduce memory allocations.
As a quick reminder, a virtual arrays is a data structure that behaves like an
array (i.e. it can be accessed using an index). However, it may not actually
be stored as array internally. The two most important implementations
of virtual arrays are those that correspond to an actual plain array and those
that have the same value for every index. However, many more
implementations exist for various reasons (interfacing with legacy attributes,
unified iterator over all points in multiple splines, ...).
With this refactor the core types (`VArray`, `GVArray`, `VMutableArray` and
`GVMutableArray`) can be used like "normal values". They typically live
on the stack. Before, they were usually inside a `std::unique_ptr`. This makes
passing them around much easier. Creation of new virtual arrays is also
much simpler now due to some constructors. Memory allocations are
reduced by making use of small object optimization inside the core types.
Previously, `VArray` was a class with virtual methods that had to be overridden
to change the behavior of a the virtual array. Now,`VArray` has a fixed size
and has no virtual methods. Instead it contains a `VArrayImpl` that is
similar to the old `VArray`. `VArrayImpl` should rarely ever be used directly,
unless a new virtual array implementation is added.
To support the small object optimization for many `VArrayImpl` classes,
a new `blender::Any` type is added. It is similar to `std::any` with two
additional features. It has an adjustable inline buffer size and alignment.
The inline buffer size of `std::any` can't be relied on and is usually too
small for our use case here. Furthermore, `blender::Any` can store
additional user-defined type information without increasing the
stack size.
Differential Revision: https://developer.blender.org/D12986
|
|
|
|
|
|
Previously, every node had to create warnings for unsupported input
geometry manually. Now this is automated. Nodes just have to specify
the geometry types they support in the node declaration.
Differential Revision: https://developer.blender.org/D12899
|
|
Instance IDs serve no purpose for rendering when they aren't stable from
one frame to the next, and if the index is used in the end anyway, there
is no point in storing a vector of IDs and copying it around.
This commit exposes the `id` attribute on the instances component,
makes it optional-- only generated by default with the distribute points
on faces node.
Since the string to curves node only added the index as each instance's
ID, I removed it. This means that it would be necessary to add the ID
data manually if the initial index actually helps (when deleting only
certain characters, for example).
Differential Revision: https://developer.blender.org/D12980
|
|
This changes socket inspection for fields according to T91881.
Differential Revision: https://developer.blender.org/D13006
|
|
In order to address feedback that the "Stable ID" was not easy enough
to use, remove the "Stable ID" output from the distribution node and
the input from the instance on points node. Instead, the nodes write
or read a builtin named attribute called `id`. In the future we may
add more attributes like `edge_id` and `face_id`.
The downside is that more behavior is invisible, which is les
expected now that most attributes are passed around with node links.
This behavior will have to be explained in the manual.
The random value node's "ID" input that had an implicit index input
is converted to a special implicit input that uses the `id` attribute
if possible, but otherwise defaults to the index. There is no way to
tell in the UI which it uses, except by knowing that rule and checking
in the spreadsheet for the id attribute.
Because it isn't always possible to create stable randomness, this
attribute does not always exist, and it will be possible to remove it
when we have the attribute remove node back, to improve performance.
Differential Revision: https://developer.blender.org/D12903
|
|
Avoiding creating empty components can be a hassle for code that
interacts with a geometry set. One easy way to do that was calling
the functions that retrieved mutable access to geometry data directly,
like get_mesh_for_write. This commit makes it so that sort of direct
function does not create an empty component if there is no data.
Another way to create an empty component was calling the replace_*
methods with a null pointer. It's more convenient to have a nice API
that handles those cases without creating an empty component.
It's still convenient that the regular get_component_for_write adds
the component if it doesn't exist, because that's often a nice way to
add data to the geometry set.
Differential Revision: https://developer.blender.org/D12862
|
|
This reduces the compile time, because fewer symbols have to be generated
in translation units using geometry sets.
|
|
|
|
As described in T91672, often it can be much more efficient to run each
node only on the unique geometry of the instances, rather than realizing
all instances and potentially processing redundant data. Sometimes the
performance difference can be completely smooth vs. completely unusable.
Geometry nodes used to hide that choice from users by always realizing
instances, but recently we have decided to expose it. So this commit
makes nodes run once per unique reference in the entire tree of nested
instances in their input geometries, continuing the work started in
rB0559971ab377 and rBf94164d89629f0d2. For the old behavior, a realize
instances node can be added before the nodes, which is done in the
versioning code.
Differential Revision: https://developer.blender.org/D12656
|
|
|
|
|
|
This adds a new `GeometrySet::modify_geometry_sets` method that can be
used to update each sub-geometry-set separately without making any
instances real.
Differential Revision: https://developer.blender.org/D12650
|
|
This adds a new Instance on Points node that is a replacement
for the old Point Instance node. Contrary to the old node,
it does not have a mode to instance objects or collections
directly. Instead, the node has to be used with an Object/
Collection Info to achieve the same effect.
Rotation and scale of the instances can be adjusted in the node
directly or can be controlled with a field to get some variation
between instances.
The node supports placing different instances on different points.
The user has control over which instance is placed on which point
using an Instance Index input. If that functionality is used, the
Instance Geometry has to contain multiple instances that can are
instanced separately.
Differential Revision: https://developer.blender.org/D12478
|
|
This adds a replacement for the deprecated Point Distribute node.
Arguments for the name change can be found in T91155.
Descriptions of the sockets are available in D12536.
Thanks to Jarrett Johnson for the initial patch!
Differential Revision: https://developer.blender.org/D12536
|
|
This adds three new methods:
* `InstancesComponent::foreach_reference_as_geometry(...)`
* `GeometrySet::attribute_foreach(...)`
* `GeometrySet::gather_attributes_for_propagation(...)`
The goal is that these iteration primitives can be used in places
where we use more specialized iterators currently.
Differential Revision: https://developer.blender.org/D12613
|
|
With this commit, each referenced instance data will be converted to
a geometry instances and processed separately. This should result in
a large speedup when the instances component has many insances
referring to the same data.
This change can act as a blueprint for other nodes that need to
implement similar behavior. It adds some helper functions on the
instances component to make that easier.
Thanks to Erik Abrahamsson for a proof of concept patch.
Differential Revision: https://developer.blender.org/D12572
|
|
Previously, the node would always realize instances implicitly.
Now it can change the position of entire instances.
The Realize Instances node can be used before if the old
behavior is required.
Differential Revision: https://developer.blender.org/D12555
|
|
|
|
Since fields were committed to master, socket inspection did
not work correctly for all socket types anymore. Now the same
functionality as before is back. Furthermore, fields that depend
on some input will now show the inputs in the socket inspection.
I added support for evaluating constant fields more immediately.
This has the benefit that the same constant field is not evaluated
more than once. It also helps with making the field independent
of the multi-functions that it uses. We might still want to change
the ownership handling for the multi-functions of nodes a bit,
but that can be done separately.
Differential Revision: https://developer.blender.org/D12444
|
|
This implements the initial core framework for fields and anonymous
attributes (also see T91274).
The new functionality is hidden behind the "Geometry Nodes Fields"
feature flag. When enabled in the user preferences, the following
new nodes become available: `Position`, `Index`, `Normal`,
`Set Position` and `Attribute Capture`.
Socket inspection has not been updated to work with fields yet.
Besides these changes at the user level, this patch contains the
ground work for:
* building and evaluating fields at run-time (`FN_fields.hh`) and
* creating and accessing anonymous attributes on geometry
(`BKE_anonymous_attribute.h`).
For evaluating fields we use a new so called multi-function procedure
(`FN_multi_function_procedure.hh`). It allows composing multi-functions
in arbitrary ways and supports efficient evaluation as is required by
fields. See `FN_multi_function_procedure.hh` for more details on how
this evaluation mechanism can be used.
A new `AttributeIDRef` has been added which allows handling named
and anonymous attributes in the same way in many places.
Hans and I worked on this patch together.
Differential Revision: https://developer.blender.org/D12414
|
|
Previously, the Point Instance node in geometry nodes could only instance
existing objects or collections. The reason was that large parts of Blender
worked under the assumption that objects are the main unit of instancing.
Now we also want to instance geometry within an object, so a slightly larger
refactor was necessary.
This should not affect files that do not use the new kind of instances.
The main change is a redefinition of what "instanced data" is. Now, an
instances is a cow-object + object-data (the geometry). This can be nicely
seen in `struct DupliObject`. This allows the same object to generate
multiple geometries of different types which can be instanced individually.
A nice side effect of this refactor is that having multiple geometry components
is not a special case in the depsgraph object iterator anymore, because those
components are integrated with the `DupliObject` system.
Unfortunately, different systems that work with instances in Blender (e.g.
render engines and exporters) often work under the assumption that objects are
the main unit of instancing. So those have to be updated as well to be able to
handle the new instances. This patch updates Cycles, EEVEE and other viewport
engines. Exporters have not been updated yet. Some minimal (not master-ready)
changes to update the obj and alembic exporters can be found in P2336 and P2335.
Different file formats may want to handle these new instances in different ways.
For users, the only thing that changed is that the Point Instance node now
has a geometry mode.
This also fixes T88454.
Differential Revision: https://developer.blender.org/D11841
|
|
|
|
This commit moves the storage of `bDeformGroup` and the active index
to `Mesh`, `Lattice`, and `bGPdata` instead of `Object`. Utility
functions are added to allow easy access to the vertex groups given
an object or an ID.
As explained in T88951, the list of vertex group names is currently
stored separately per object, even though vertex group data is stored
on the geometry. This tends to complicate code and cause bugs,
especially as geometry is created procedurally and tied less closely
to an object.
The "Copy Vertex Groups to Linked" operator is removed, since they
are stored on the geometry anyway.
This patch leaves the object-level python API for vertex groups in
place. Creating a geometry-level RNA API can be a separate step;
the changes in this commit are invasive enough as it is.
Note that opening a file saved in 3.0 in an earlier version means
the vertex groups will not be available.
Differential Revision: https://developer.blender.org/D11689
|
|
* Reduce code duplication.
* Give methods more standardized names (e.g. `move_to_initialized` -> `move_assign`).
* Support wrapping arbitrary C++ types, even those that e.g. are not copyable.
|
|
|
|
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 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 code in the geometry set header was not directly related to
geometry sets, it makes more sense in the attribute access header.
This makes it clearer that code for geometry components uses attribute
code, rather than the other way around. It also allows adding more
functionality to `BKE_attribute_access.hh` that depends on these things
without including `BKE_geometry_set.hh` there.
|
|
This commit uses two changes to improve the performance of the point
instance node.
**Prevent Reallocations**
At 64 bytes, the transform matrix for every instance is rather large,
so reallocating the vector as it grows can become a performance bottle-
neck. This commit reserves memory for the instances that will be added
to prevent unecessary reallocations as the instance vector grows.
In a test with 4 million instances of 3 objects in a collection, the
node was about 40% faster, from 370ms to 270ms for the node.
**Parallelization**
Currently the instances are added by appending to a vector. By changing
this slightly to fill indices instead, we can parallelize the operation
so that multiple threads can fill data at the same time. Tested on a
Ryzen 3700x, this reduced the runtime from the above 270ms to 44ms
average, bringing the total speedup to ~8x.
Note that displaying the instances in the viewport is still much slower
than the calculations in node, this change doesn't affect that.
|