* Use the PBVHType consistently in pbvh_update_draw_buffers().
* Split the raycast function up, mesh and grid raycast get their own
functions now.
* Replace duplicated code in BLI_pbvh_node_add_proxy() with call to
BLI_pbvh_node_num_verts().
* Changes to DerivedMesh interface: DMGridData has been removed,
getGridData() now returns an array of CCGElem pointers. Also added
getGridKey() to initialize a CCGKey (implemented only by
CCGDerivedMesh.)
* PBVH: added BLI_pbvh_get_grid_key().
* A lot of code is affected, but mainly is just replacing
DMGridData.co, DMGridData.no, and sizeof(DMGridData) with the
CCG_*_elem functions, removing the reliance on grid elements of
exactly six floats.
No functional changes, this is just a bit easier to read I think (and
nicer if we add more PBVH types in the future, e.g. for dynamic
topology sculpting.)
Previously, the shading and material was set once per PBVHNode when
drawing. This is still the case, but PBVHNodes are now built to
contain only one material and shading mode.
This is done with an extra partitioning step; once the number of
primitives in the node falls below the PBVH leaf limit, it's
primitives are checked for matching materials. If more than one
material or shading mode is present in the node, it is split and
partitioned (partitioned by material rather than 3D location.)
Given a sufficiently 'annoying' input, like a dense mesh with
thousands of materials randomly scattered across it, this could
greatly increase PBVH build time (since nodes might end up containing
a single primitive), but in general this shouldn't come up.
In order to support materials for grids, the CCGDM is building another
grid array (of DMFaceMat structs). This could be used to replace
CCGDM.faceFlag for some small memory savings (TODO).
Used a crazyspace approach (like in edit mode), but only modifiers with
deformMatricies are allowed atm (currently shapekeys and armature modifiers only).
All the rest modifiers had an warning message that they aren't applied because
of sculpt mode. Deformation of multires is also unsupported.
With all this restictions users will always see the actual "layer" (or maybe
mesh state would be more correct word) they are sculpting on.
Internal changes:
- All modifiers could have deformMatricies callback (the same as deformMatriciesEM but
for non-edit mode usage)
- Added function to build crazyspace for sculpting (sculpt_get_deform_matrices), but it
could be generalized for usage in other painting modes (particle edit mode, i.e)
Todo:
- Implement crazyspace correction to support all kinds of deformation modifiers
- Maybe deformation of multires isn't so difficult?
- And maybe we could avoid extra bad-level-stub for ED_sculpt_modifiers_changed
without code duplicating?
PBVH used the same verts array as mesh data and shape key/reference key coords
were applying on the mesh data, so on some refreshing undeformed mesh was
displayed.
Added utility functions to get vert coords from key block, apply new vert coords
on keyblock and function to apply coords on bpvh, so now pbvh uses it's ovn
vertex array and no changes are making to the mesh data.
Additional change:
Store key block name in SculptUndoNode, so now shape wouldn't be copied to
wrong keyblock on undo
Fixes various crashes and redraw problems, most noticeable new feature
is that you can now sculpt on a multires mesh with deforming modifiers
preceding it.
I've left out support for sculpting on multires with enabled modifiers
following it, in this case only the base mesh can be sculpted now. The
code changes needed to do this are just too ugly in my opinion, would
need a more torough redesign which I don't think we should try now. In
my opinion this is also not really an important case, since it's going
to be incredibly slow anyway to run a modifier on a high res mesh while
sculpting.
So, to summarize current state:
* Fastest sculpting: base mesh with no modifiers or multires with only
modifiers preceding it.
* Slower sculpting: base mesh with modifiers, depends on the speed of
the modifiers.
* Not supported: multires mesh with modifiers following it.
* Temporary workaround for sculpt not working well with small polygons,
still seems to be some issues, but can at least paint now.
* Small optimization avoiding local function variable aliasing.
* Added detection if VBO extension is supported.
* Redraw other 3d views after sculpting.
* Fix brush sometimes punching through mesh with very small polygons,
added an extra epsilon to the ray-triangle intersection.
* PBVH can now be created contain both from face grids or standard
meshes. The former is much quicker to build for high res meshes.
* Moved some drawing code into pbvh (mostly for the frustum test).
* Moved ray intersection code into pbvh.
* GPU buffers also can be built from either mesh or grids now.
* Updated sculpt code to work with this. The ugly part is that there
is now a macro for iterating over vertices, to handle both cases,
and some duplicated code for e.g. undo.
* Smooth brush does not work yet with grids.
* Draw/Inflate/Layer now keep working on the original mesh coordinates and
normals from when the stroke started. This helps avoid the mesh blowing
up, but can still be better. The old behavior is still available as
"Accumulate" in the UI.
* This requires some more memory usage for the BVH, would like to find a
way to avoid that.
* Smooth falloff is now the default.
* Spacing is now enabled by default, with a value of 7.5.
* Anchored now stores normals per node to save some memory.
The weak point now is the thread-safe atomic access to normals from multiple
threads, did not seem to be a bottleneck in my tests but I don't really trust
it to be fast.
* Sculpting, normal update and bounding box code is now multithreaded
using OpenMP.
* Fix a number of update issues: normals on node boundaries, outdated
bounding boxes, partial redraw, .. . There's probably still a few
left, but should be better now.
* Clicking once now does a single paint instead of two (was also
painting on mouse up event).
* Smooth shading now is enabled for the full mesh when the first face
uses it (so it can be tested at least).
Implementation Notes:
* PBVH search can now be done either using a callback or bt gathering the
nodes in an array. The latter makes multithreading with OpenMP easier.
* Normals update code is now inside PBVH, was doing it per node before but
should do all faces first and only then vertices.
* Instead of using search modes + 1 modified flag, now nodes get 4 flags
to indicate what needs to be updated for them, found that this makes it
easier for me to understand the code and fix update bugs.
* PBVHNode is now exposed as an abstract type, I think this makes it more
clear what is happening than having it's data passed as part of callback
functions.
* Active_verts list was replaced by looping over nodes and the vertices
inside them. However the grab brush still uses the active_verts system,
will fix that later.
* Some micro-optimizations, like avoiding a few multiplications/divisions,
using local variables instead of pointers, or looping over fewer vertices
to update the bounding boxes.
