I think its something like:
old was: 4*nlogn + 3*(n*6)
new is: (2*nlogn + 3*(n*6)) * f, with f<1
Still missing changing the sorting function to an introsort instead of qsort
Other options like bucketing sort may be worth trying (for very large trees)
only for 2childs splits
with "worse case heuristic" means each child is considered to have a cost linear on the number of leafs
no termination criteria
number of BB test/hits expected to "reduced" by some factor
tree building is also expected to be slower as previous split was "object mean", which is quite fast to evaluate
*"Added" SCE_PASS_RAYHITS to visually see each pixel primitive and BB tests (not-completed for UI)
*Added runtime exchange of tree structure
*Removed FLOAT_EPSILON from BLI_bvhkdop BB's
* Baked normal particles can now use the "Path" visualization.
* Path "max length" & "abs length" are now history:
- New option to set path start & end times + random variation to length.
- Much more flexible (and calculated better) than previous options.
- This works with parents, children, hair & normal particles unlike old length option.
- Only known issue for now is that children from faces don't get calculated correctly when using path start time.
* New option "trails" for "halo", "line" and "billboard" visualizations:
- Draws user controllable number of particle instances along particles path backwards from current position.
- Works with children too for cool/weird visualizations that weren't possible before.
* Normal particle children's velocities are now approximated better when needed so that "line" visualization trails will look nice.
* New particle instance modifier options:
- "path"-option works better and has controllable (max)position along path (with random variation possible).
- "keep shape"-option for hair, keyed, or baked particles allows to place the instances to a single point (with random variation possible) along particle path.
- "axis" option to make rotation handling better (still not perfect, but will have to do for now).
Some fixes & cleanup done along the way:
* Random path length didn't work for non-child particles.
* Cached & unborn particles weren't reset to emit locations.
* Particle numbers weren't drawn in the correct place.
* Setting proper render & draw visualizations was lost somewhere when initializing new particle settings.
* Changing child mode wasn't working correctly.
* Some cleanup & modularization of particle child effector code and particle drawing & rendering code.
* Object & group visualizations didn't work.
* Child simplification didn't work.
* Fix crash in python with enum properties, and don't throw
error if no matching identifier is found. This shouldn't
happen, but it should break a python script either, which
is not at fault.
* Fix a wrong variable initialization in fluidsim.
* Lattices: properties editable, editmode operators, menus working
again. As a bonus you can now edit u/v/w in editmode.
* Shape Keys: some code cleanup, and added more buttons. The
value/min/max buttons don't work correct yet though.
* Fix issue with uv textures, vertex colors not being visible outside
editmode, and a few other issue. Mesh.edit_mesh is now NULL when
not in editmode.
More scroller work:
- Added subtle arrow widgets to denote a scroller can zoom the view.
- Made zoom symmetric (old convention to only zoom "one side" just
worked badly)
All kinds of changes to get it ready for UI layouts. This means RNA
and operators should be working correct, but most buttons are still
not actually there yet.
* Added near empty soft body, fluid, field and collision panels,
tweaks to cloth panels.
* Fluid bake works, but without escape or showing any progress.
* Fluid/Softbody/Cloth/Collision can now be both added as modifiers
or in the physics panels.
* Missing: fields & soft body for particles.
* Missing: proper updating softbodies, guess this code still needs
updates after pointcache refactor?
Cleanup of scroller drawing in 2D windows.
Before:
http://download.blender.org/institute/rt11.jpg
After:
http://download.blender.org/institute/rt12.jpg
Will add 'zoom' widget circles later, as mockupped here:
http://www.reynish.com/files/blender25/fcurve_scrollbar.png
Also note the scale values are inside scroller; drawing it
on top conflicts with current frame item and markers.
Currently scroller disappear entirely when view is total.
For Joshua:
To make sliders behave nicely, the boundbox (v2d->tot) has to
be refreshed on each change. I've added it in graph drawing
now, but it could be notifier based I guess... not sure what
the correct anim api call would be. Can discuss tomorrow!
On todo:
Layout config hints so people can make scroller positions swap.
More toolbar functionality for workflow review.
- Split the region in two parts, bottom has the Tool Properties, the
top part shows 2 panels, one for python defined tools, other for a
"tool shelf" which (later) will get saved in files.
- Added a full context driven framework for this toolbar, showing
the tools depending on 3D window 'mode'. Both python defined tools
as the shelf respect this. So - for example - you will see different
tools in editmode mesh, as in vertex paint mode, etc.
- First template for the python tools will be committed after this commit;
it has placeholder tools to just show/test functioning.
NOTE: if you had saved a layout that shows tools region, open/close it
once to get the new region created for properties.
TODO:
- Moving paint properties to tool settings
- Test a layout with horizontal toolbar (without properties)
- Bring back floating panels, and put tool-properties here. (as option)
RNA
* Added the relevant active_*_index properties, with proper
get/set/range, updates and notifiers.
* Context.tool_settings.
* ToolSettings.vertex_group_weight.
Operators
* MESH_OT_uv_texture_add/remove
* MESH_OT_vertex_color_add/remove
* MESH_OT_sticky_add/remove
* OBJECT_OT_vertex_group_add/remove/assign/remove_from/
select/deselect/copy/copy_to_linked
* OBJECT_OT_shape_key_add/remove
UI
* Some updates and cleanups in list template code.
Known issue: when going in & out of editmode, uv textures and vertex
colors dissappear. I thought me->edit_mesh would be NULL when not in
edit mode but it is not?
RNA_function_call_direct_lookup(C, reports, ptr, "do_something", "sC", "some string value", &RNA_SomeItemType, listbase);
Note that from python you could in theory pass collections of items each of a different type while using this API you can't. I don't think this should be a problem as RNA supports collections this way anyway (i.e. where items are all of the same type or inherit from the same base type).
Also a small fix for pointer parameters.
Added slice to PyRNA collections and arrays (py3 only).
eg.
some_verts = mesh.verts[0:10]
some_rna_array[4:-1] = [0,1,2,3]
Collections dont support assignment, when assigning slices, resizing the array isnt support like with python lists.
* Mesh.add_geometry, Mesh.update and make indices editable. This
is without checking if they are valid still, no time now to
implement this.
* Also fix warnings in rna_ui.c, and a bug in CDDM_calc_edges.
Example code:
co = [0.0, 0.0, 0.0] + [1.0, 0.0, 0.0] + [0.0, 1.0, 0.0] + [1.0, 1.0, 0.0]
faces = [0, 1, 2, 0] + [1, 3, 2, 0]
mesh.add_geometry(4, 0, 2)
mesh.verts.foreach_set("co", co)
mesh.faces.foreach_set("verts", faces)
mesh.update()
== Re-added smooth stroke ==
UI: toggle is just in the sculpt menu for now. Also changes the sculpt paint cursor slightly, draws a line between previous and current locations.
It's a different implementation than in 2.4, works like this:
The stroke interpolates between the last mouse location and the current location, weighted towards the previous location. If the stroke gets within a certain radius of the current mouse location, the stroke stops. This radius allows for sharp turns in the stroke.
Todo: there are two hard-coded parameters that should become user settable, that's the weighting between previous and current locations, and most important, the no-update radius.
Note also that this option was added as a per-brush flag, worth discussing whether that's the correct place, or whether it should be a sculpt setting like symmetry?
== Improved stroke spacing ==
The previous implementation of stroke spacing simply guaranteed that stroke dots would not occur any closer than the space setting. It now forces stroke dots to always be the specified distance apart.
Todo: Performance gets pretty awful with a small spacing setting, this needs optimization.
* show only name of the last directory for the bookmark
* small fix of projectfile: header BLI_fileops.h was moved
Note: full path should appear in tool tip later, also for renaming bookmarks later on.
