Blender Git Commits

fix for outliner drawing when it should not; code was missing a break

yay better less hackish dupli fix

duplis in groups were not working

Render Branch: svn merge https://svn.blender.org/svnroot/bf-blender/trunk/blender -r28801:28850 ...note: dont merge my previous commit

Render Branch: implicit solver for hair now calculates forces and does
integration separate per hair. This increase the stability of the hair.

I found that usually the conjugate gradient takes about 4-8 steps, but
it sometimes takes 20 or even does not converge, and returns bogus
velocities. I'm not entirely sure why that is, but it seems logical
that a smaller system would be more numerically stable. What I suspect
is that since there are many hairs, and it tests for the average error
of those hairs, a few hairs may be able to hide their high error in this
average.

I've tried to not make the code too ugly but's there's a few stupid
hacks, maybe eventually it would be better to have a separate hair and
cloth function that calls common functions.

Also for threading, hairs can now be distributed over threads. There
still seems to be too much thread overhead though, should find a way
to keep all threads running for the entire timestep.

Render Branch: fix two point cache problems:
* cloth did not get reset correct in some situations, when startframe was
different than 1
* point cache could sometimes mark the cached frames as inexact, when
writing an info pointcache file, since that was indicated as cfra == 0.

Render Branch: some more code shuffling in the implicit solver:
* move force calc + implicit solve into separate function
* do the same for new goal code
* only run the second filtering solve in case it is needed
* only run rigid body damping in case it is enabled

Render Branch: implicit solver now only uses openmp on matrix multiplication
and dot product if num vertices > 25. This may help for cases where there are
few vertices and thread overhead actually slows things down, but mostly it's
to avoid useless threads in threads once I commit the code to solve each hair
individually.

Render Branch: minor code tweaks and warnings fixes in implicit solver, should
have no effect, committing these separately from the actual changes.

BLI_init_threads/BLI_end_threads with NULL listbase now raises thread level
and enables mutex protection on MEM_* functions, useful when you want to call
these functions from an OpenMP thread.

Render Branch: two utility functions:

BLI_linklist_find: find link at an index
BLI_linklist_insert_after: insert link after another one

particle system can start in negative frames now. also removed BIF_preview_changed, since it's not being used anymore.

Fix for the eyelid rig type. Vector(#, #, #) to Vector((#, #, #))

made trail settings a seperate panel. bug fixes will follow, after I work on some more urgent ones first.

Render Branch: svn merge https://svn.blender.org/svnroot/bf-blender/trunk/blender -r28759:28801 ...note: skipping changes from properties_particle.py, will need to be merged manually

Render Branch: disable threading for vector dot product in implicit solver
for now, due to non-commutative nature of floating point ops the sim would
give different results each time! Sometimes subtle, but sometimes also very
different.

version patch for 2.49 boids; be warned: this is not really at all correct, but hey at least no more crashing :)

svn merge https://svn.blender.org/svnroot/bf-blender/trunk/blender -r28726:28759

workaound for smoke pointcache not being read, this may not be a correct fix so committing to render-branch for now.

fix for mistake in commit 28570