* BM_Collapse_Vert_Faces
* BM_Collapse_Vert_Edges
since these are both quite different operations and callers where checking for one case or another anyway.
* The code for snap to grid was completely wrong (only x pos of chan was affected… by z computed value!), and the same problem as with snap to cursor was present (using armature_loc_pose_to_bone is wrong here). Also, computing nearest grid point in world space! Else, it generally ends up completely out of grid, due to object transforms...
* Small cleanup of code (uneeded vars and copying).
In fact, the problem was present with any kind of constraint…
We need the cursor coordinates in the rest bone space (aka armature space), i.e with all "manual" and "procedural" (constraints, ...) pose transforms cleared (as said in armature_mat_pose_to_bone:
/* Remove the pchan's transforms from it's pose_mat.
* This should leave behind the effects of restpose +
* parenting + constraints
*/
, only "manual" transforms are "cleared" by armature_loc_pose_to_bone...).
Keir's comment:
Add support for detecting tracking failure in the ESM tracker component of
libmv. Since both KLT and Hybrid rely on ESM underneath, KLT and Hybrid now
have a minimum correlation setting to match. With this fix, track failures
should get detected quicker, with the issue that sometimes the tracker will
give up too easily. That is fixable by reducing the required correlation (in
the track properties).
Command used for merge: svn merge -r 42396:42397 -r 42399:42400 ^/branches/soc-2011-tomato
Moved tweak threshold value to user preferences
This threshold might be needed to be tweaked when working with tables, i.e.
to prevent tap+slight movement be treated as tweak event.
- Fixed bug with not putting disabled markers properly when doing backwards tracking
- Fixed margin size calculating from pattern size which used to be double-sized
and prevented to track things on the image edges.
In this case now just adjust the count value since theres no need to loop,
when the allocation is not big enough BLI_array_growone in a loop is still used though.
material node. This is a confusing system, but two features were missing from
2.4x that made this at least a bit more clear:
* The top right icon in the node now shows brighter again for the active node.
* Setting a material datablock in a node makes that node active.
* Added option "WITH_BF_CYCLES_CUDA_THREADED_COMPILE" for the people who have much RAM (8 or more) and can compile several kernels at the same time. If enabled, it uses the general BF_NUMJOBS flag.
* The option is off per default.
Warning! Tesselation uvs or vcol data got out of sync, "had to reset!
This would happen on every editmode edit with UV's and wasn't too reassuring that blender was handling uvs/vcols correctly.
From looking into the problem I found that creating the undo mesh would act as if it was tessellating the existing mesh each time and complain that the data was out of sync, when infact the mesh was just created and being filled in.
Also, allocating uv and vcol customdata arrats for tessfaces isn't needed for undo mesh, so save some memory and dont allocate these in the first place.
Fix#29475: remove node from properties editor crash on windows. This was a bug
in the UI code, which code access removed data.
Fix OpenCL still being used in a case where Experimental was disabled.
Fix msvc debug warning in md5 code.
