Ancient issue; when you press keypad-0, it should assign the active camera
object to a scene, but this did not happen when that camera was already
assigned to the active 3d view. A rare case, causing 'error no camera' on
a render.
Composite: the Translate Node only worked in some cases (pixel processor
call), and not for:
- filter
- seperate rgba
- conversion of buffer types (value to rgba etc)
Blur still doesn't either, but that code is too optimized to add quick.
Will put on todo for checking on better unification of translations.
export would cause a problem since the temporary meshes created
would often reuse the same name for many objects. This fix mangles
the object name and uses this for the temporary mesh name to avoid
this problem. (The naming is important, since vrml has 'DEF' and 'USE'
statements that are used like C macros, so if two meshes have the
same name the exporter will try to 'recycle' data to keep file sizes
low).
Actually a bug since dark ages... the code that tries to find the view
correction (zoom) factor for grabbing stored result in a global. With
multiple 3d windows open, with different views, that could result in
wrong correction. Just made the factor a local property in View3D.
Bugfix: using Image.GetCurrent() and image.save() on the "Render Result"
image could result in a segfault since image->ibuf was NULL. This change
forces ibuf to be created if necessary.
NOTE: the Image API needs additional methods/attributes for image.save() to
really do anything useful. The image type, quality, etc., don't seem to be
gettable/settable so the resulting image file may not be in the format the
user would like.
N_T: the implementation of this feature isn't 100% nice, it currently does
a full recalc on each redraw, and it doesn't work properly unless it is
recalculated for each frame.
Repeat Image texture was broken... the repeat value for Y was accidentally
multiplied twice, typo... had to change a 1 into 2 :)
Error introduced in 2.41 btw, 2.40 and older were OK.
Render non-osa, with mblur, did not jitter the zbuffering, so it did not
result in basic AA for non moving parts.
Note to self: this now works on < 2.42 level again: it doesn't use
the correct subsampling, which can cause slight errors in rendering image
textures.
- readfile.c, version patch for new toolsettings used a malloc, using
unitialized variables
- despgraph, another malloc changed to calloc
- timer code for render ESC: unitialized variable in struct
- thread render stores per thread and per lamp the last intersected
shadow face
- the Isect struct, for intersections, could use a couple of extra
zeros on start.
Apparently not supported in Windows compiler, and although it has _isnan()
I've got no time to code wrappers for it, and #ifdefs here is plain bad
code.
Nice todo for Campbell one of these days after release!
- Updating Blender Lip Synchro to the newest version sent by the author,
Benoit Foucque (Dienben). Thanks!
I'm committing a version with a few minor (tested) changes to make it
easier and to fix a small problem, the author received the changes a few
days ago. He would try to send another update implementing a couple
suggestions, but it may not come in time, so better have this. Anyway,
interested users can check his blog for test files and more information:
http://blenderlipsynchro.blogspot.com/
Play option: if you set output to render to .avi, but in the same
directory also individual files exist, it always played the files, not
the movie.
Now it checks for the Output type, and plays back a movie or frames, as
indicated.
Negative scaled Objects get a OB_NEG_SCALE flag, to indicate to OpenGL
to invert lighting for solid drawing.
Works nice, but when you use Group-duplicators, the negative scale of
the duplicator should be taken into account as well.
- Added "Shift+P" note in 3d window menu for preview panel
- Added Preview, passepartout in spacebar toolbox
- Info pulldown "Render" -> "Render buttons" did cycle, should not.
