These should not have any effect on render results, except in some cases with
you have overlapping faces, where the noise seems to be slightly reduced.
There are some performance improvements, for simple scenes I wouldn't expect
more than 5-10% to be cut off the render time, for sintel scenes we got about
50% on average, that's with millions of polygons on intel quad cores. This
because memory access / cache misses were the main bottleneck for those scenes,
and the optimizations improve that.
Interal changes:
* Remove RE_raytrace.h, raytracer is now only used by render engine again.
* Split non-public parts rayobject.h into rayobject_internal.h, hopefully
makes it clearer how the API is used.
* Added rayintersection.h to contain some of the stuff from RE_raytrace.h
* Change Isect.vec/labda to Isect.dir/dist, previously vec was sometimes
normalized and sometimes not, confusing... now dir is always normalized
and dist contains the distance.
* Change VECCOPY and similar to BLI_math functions.
* Force inlining of auxiliary functions for ray-triangle/quad intersection,
helps a few percentages.
* Reorganize svbvh code so all the traversal functions are in one file
* Don't do test for root so that push_childs can be inlined
* Make shadow a template parameter so it doesn't need to be runtime checked
* Optimization in raytree building, was computing bounding boxes more often
than necessary.
* Leave out logf() factor in SAH, makes tree build quicker with no
noticeable influence on raytracing on performance?
* Set max childs to 4, simplifies traversal code a bit, but also seems
to help slightly in general.
* Store child pointers and child bb just as fixed arrays of size 4 in nodes,
nearly all nodes have this many children, so overall it actually reduces
memory usage a bit and avoids a pointer indirection.
Patch #25901 by Tobias Oelgarte.
Bone transformations would be converted back and forth between different
representations when changing modes, which due to numerical errors could
lead to bone transformations slowly changing as you edit the armature.
Now the editmode head, tail and roll values are stored in bones and used
directly when entering edit mode. Head and tail were already there but
now we ensure they are the exact same value, roll was not yet there, so
we have a version patch for it.
The sub version was incremented to 1 for the version patch.
Manually disable jitter usage for anchored and drag dot brush stroke metdhods.
Jitter slider is hidden in UI for this strokes so users can't set it to 0 by hand
and even if this slider would be visible in UI jitter gives wierd result for
this stroke methods.
* Particles that aren't shown are now actually deleted (huge memory savings for flat objects).
* Grid distribution for flat objects is now done on the surface object surface without offset.
* Invert grid option wasn't in ui and it didn't work for non-volume grids.
* New parameter to randomize the grid point locations.
* Resolution soft/hard limits changed to even 50/250.
- use BLI_math functions for removing rotations from objects and pose channels.
- add unit_axis_angle() to avoid setting the Y axis inline anywhere rotation needs removing.
calling
bpy.ops.wm.read_factory_settings()
... would clear a scripts namespace if running directly, not in a module.
Fix by backing up and restoring the __main__ module.
Also found BKE_reportf wasnt printing all reports in background mode as BKE_report() was doing.
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?
oldbump -> original
newbump -> compatible
*new* -> default (3tap)
*new* -> best quality (5tap)
the latter two have an option to apply bumpmapping in
viewspace - much like displacement mapping
objectspace - default (scales with the object)
texturespace - much like normal mapping (scales)
* Caching of the start and end stills were just referencing the original imbuf (which got premultiplied after the caching), so as a result most of the time the premul was applied twice.
* Now the start and end stills are stored in the cache as duplicates of the original (non modified) imbuf.
they have some rotation-affecting constraint (i.e. Track To and Copy
Rotation) targetting, transforming the objects (directly, using GKEY
-> grab) becomes unreliable.
This was caused by a typo in some code checking for some
OB_NO_CONSTRAINTS under "flag" instead of "transflag"
For Constraints, there's now a working "Local" Space for Objects
without parents. This is defined as relative to the object's rotated
set of axes which results from rotation that gets set via "rotation"
transform properties.
I'm not sure whether this different behaviour between parented and
unparented objects will be too confusing (and thus require separate
settings + a round of version patching), so I'll wait until we get
proper testing from experienced riggers first.
- values were added to both the classes __dict__ as well as the internal StructRNA.
- made properties available from the type since this is where the python api assigns them:
>>> bpy.types.Scene.frame_start
<bpy_struct, IntProperty("frame_start")>
- rename RNA_struct_type_properties() -> RNA_struct_type_properties(), added RNA_struct_type_find_property()
Support for physics is done by skiping the modifiers that
don't support mapping to original mesh. This mapping is
required to report the hit polygon to the application
by the rayCast() function.
Support for graphics is done by using the same render
function that blender uses for the 3D view. This guantees
equal result.
Limitation: there is still a known bug if all these conditions are met:
- Display list enabled
- Old tex face with a several textures mapped to the same material
- no armature or shape keys
- active modifiers
In this case, only a part of the mesh will be rendered
with the wrong texture. To avoid this bug, use the GLSL
materials or make sure to have 1 material=1 texture in
your old tex face objects.
* The whole case of lib linking and pointcaches is not very well defined currently, but this fix sets the behavior of sb to the same as other physics currently.
* A proper fix will be easy to implement after a good physics baking ui is added.
1) Added a new operator to swap the order of strips within a track
(Alt-F).
This makes it possible to select two strips (or more precisely, two
islands of consecutive + selected strips) in a single track and change
the order in which the appear without needing a extra tracks to
perform the move through.
As usual, the non-overlapping rules apply, so there may be some cases
where swapping in this way is not possible without adjusting the
intermediate strips first manually. Otherwise, everything just gets
too tricky to manage deciding what adjustments should be done to the
obstructing strips to make a fit.
2) Freeing meta-strips didn't free their local data properly (i.e.
modifiers they may have had).
3) Adding strips to tracks, where the endframes for the strips
overlapped would cause problems with incorrect ordering of strips. I
still need to double-check whether evaluation works ok in this case...
* When a simulation is baked no dynamics calculations should happen anymore (even outside of baked frame range) since these can be very time consuming and baked simulations are supposed to be fast!
- metaball tessellation functuion was calculating density when it didn't need to.
- image drawing was using a float as a loop counter, in extreme cases this could cause an infinite loop.
- remove/comment unused vars.
Added new option for applyong modifiers on splines' points. This moves
tesselation point and path would be affected by modifiers which are applied on
splines' points.
* When euler-rotation F-Curves had a single keyframe only, the view
would be artifically extended to fill up to 57 (this comes from the
radians to degrees calculations) due to a combination of the bounds-
finding function enforcing a minimum separation of 1 unit between
min/max. This has now been moved to the operator-level where it gets
applied AFTER these conversions have taken effect
* F-Curves with samples only (i.e. baked F-Curves) would be ignored by
these operators. Was caused by using a poll calback that only
considered whether there were keyframes. Hopefully this is sufficient;
otherwise a hybrid poll method will be needed.
The tool-redo depends on a working undo system, so it can rewind
a step and then redo operator with new settings. When a user
disables undo, this won't work.
Now the properties for redo operator (toolbar, F6) will grey out
when a redo isn't possible.
