Sculpt Multires render: the Orco table should be made based on 'render'
level, not on '3d view' level. Now, how this could have worked even...
Nick, could you check?
Report showed a Mesh that crashed on entering editmode and use some tools.
Appeared to be a mesh index 4 equal to one of the other indices (which
makes a corrupted quad). Fix is a sanity check in addfacelist() for enter
editmode.
Save .blend file: when setting option "Relative Paths", blender tried to
save in the root directory.
Saving .blend files is always with a real path, do the relative option
should be skipped.
Also removed the totally annoying error popup when trying to save something
before you saved a .blend. That warning was almost never correct. I left
in this warning for loading files btw.
* moved from NMesh to Mesh
* added more error checking.
* tested with 1660 lightwave models
* faster dictionary usage.
BPyImage - disallowed recursive searching of '/' or a windows drive letter.
[ #5833 ] 2.43 RC2: Incorrect Bone rotation when 3D cursor set as Pivot
There's really ugly stuff going on with pose mode rotation in transform which I'll have to fix later, in the mean time, this commit fixes the problem (which happens when you rotate bones in post mode around an arbitrary point) and includes some juicy comments to further document the actual architectural problem.
end of the selection code. That gave bad errors, especially in solid
drawmode.
Instead, when an object is not selectable it is excluded from the opengl
selection code entirely now.
Redundant variables for postprocess gamma, multiply etc were removed.
Unfortunately that makes new saved .blend files initialize these vars
to 0.0, so renders come out black entirely.
Lets keep them in for at least 1 release cycle. Just 16 bytes per
scene :)
Mesh - removing UV or Color layers didnt check if the active object was in vpaint of uv mode, removing the last layer could crash blender - now switch to object mode if its the last layer like the UI.
Reverting Aligorith's fix for bug Bugfix #5833 (the fix was incorrect, see bug report for details).
That means the translation part when rotating "free" bones in pose mode is still broken.
When rotating bones around the 3d-cursor in posemode with a rotated
armature, rotation was around strange points other than the cursor.
This bug has been around for quite a few releases now. Somehow, the
maths used to convert the world/global space locations to local locations
only worked on things in editmode, but not bones.
Fixed bug #5917, shift-c doesn't work in the sculpt mode
* Added CKEY and SHIFT+CKEY to sculptmode's shortcuts (behavior is the same as for non-sculptmode.)
Bugfix: sound ipo calculation occasionally failed, leading to silenced
strips. Now we do it just like the video strips directly before usage...
Makes the code definitely more clean and even faster ;-)
Bugfix #5912
Fix for Angmap and Spheremap (sky render).
This is a rewind of bugfix #4999 (sept 18). That bug appeared to be a
user error... the option "real" for sky had to be set.
Bugfixes:
- Effect dependencies sometimes went wrong. seq3 was always calculated,
even if the effect already told the upper layer, that there is an early
out available...
- On render, only free _all_ buffers, if memory usage is above the memcache
limit. This made my render times drop from 17 hours to only 4 hours on
a 2 hour movie.
Ton: I don't want to start an edit war at this point, but if you really
need the old behaviour, just use a very low, non-zero memcache limit in the
preferences. (In most cases, the default of 32 Mb _is_ very low ;-)
Partial fix for bug #5766
* Moved the retopo call used during transforms to above the call to clip the transform (for the mirror modifier.) This stops retopo from moving a vertex across the axis.
