| Age | Commit message (Collapse) | Author |
|---|
|
|
|
for getting object motion blur ready.
|
|
|
|
|
|
|
|
That was an old check whether tiled EXRs are used during rendering
since version 2.42 where there indeed was a special check for tile
size in EXR tile code.
Now it seems EXR could handle tiles with non-equal size and no
extra tile size check happens for EXR. Anyway EXR tile initialization
happens after initparts, so clamping size in initparts should be
safe for EXR tiles as well.
|
|
No need to linearize byte buffer when converting to display space which is data space.
|
|
|
|
|
|
conversion
|
|
|
|
|
|
|
|
|
|
This operator used to be called "Jump to Frame". It basically takes the midpoint
(frame number and/or value) of selected keyframes, and positions the current
frame (or2d-cursor in Graph Editor) at this point.
The hotkey for this is now Ctrl-G (i.e. as it's similar to a "Goto Frame"
feature). It is also now in the Key menu instead of in the relatively obscure
View menu, even though it doesn't actually result in any keyframe edits taking
place.
(Also, fixed a typo/grammer issue with one of Remove Bone Group operator)
|
|
across the edges of the empty space they spanned) only noticeable when the customdata on either size of the rip didnt match.
|
|
|
|
cleanup and remove unused define.
|
|
check if the object is already parented to said deformer before trying to add a
new modifier
This should help reduce the number of cases where users inadvertantly end up
creating multiple deform modifiers pointing to the same object, which has been
known to be a cause of "double-transform" artifacts.
Note that this is only able to detect these cases by checking if the parent
object is selected, so this will only really work for the Ctrl-P shortcut where
you have to select both objects first. However, it shouldn't be a problem either
in the Outliner (drag and drop), as the object probably won't be a child of its
parent already if you're doing this.
|
|
modifiers now
This makes it harder for users to unwittingly create multiple deform modifiers
by parenting and unparenting a number of times, with the net result being that
"Clear Parent" is now the true inverse operation of "Make Parent".
|
|
|
|
|
|
single wrap when input line width equals console width
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
few of selected objects failed to bake.
|
|
Alt+V will fill the area inbetween the ripped faces - a bit like extrude.
faces are flipped to match existing geometry and customdata (uv, vcols etc) is copied from surrounding geometry too.
|
|
use uninitialized memory. I had fixed this previously in Swiss, but it looks like I missed grabbing the fix when bringing the GetViewPort() changes into trunk.
|
|
BLI_endian_switch_int64()
|
|
to use MAX_ID_NAME (now greater then 32).
|
|
template...
Note: I should really try to remove this stupid limitation!
|
|
Was discussed in De Balie with lots of artists and we agreed it makes more
sense to behave this way
|
|
Also forgot to translate reports' titles, and change some usages of BKE_reportf to simple BKE_report, when the former is not needed!
|
|
BKE_report(f)/BKE_reports_append(f) funcs for now). Already adds quite a bunch of new msgids!
|
|
|
|
Just makes progressive refine :)
This means the whole image would be refined gradually using as much
threads as it's set in performance settings. Having enough tiles is
required to have this option working as it's expected.
Technically it's implemented by repeatedly computing next sample for
all the tiles before switching to next sample.
This works around 7-12% slower than regular tile-based rendering, so
use this option only if you really need it.
This commit also fixes progressive update of image when Save Buffers
option is enabled.
And one more thing this commit fixes is handling display buffer with
Save Buffers option enabled. If this option is enabled image buffer
wouldn't have neither byte nor float buffer until image is fully
rendered which could backfire in missing image while rendering in
cases color management cache became full.
This issue solved by allocating byte buffer for image buffer from
tile update callback.
Patch was reviewed by Brecht. He also made some minor edits to
original version to patch. Thanks, man!
|
|
This operator (Ctrl-F) allows you to flip the lattice coordinates without
inverting the normals of meshes deformed by the lattice (or the lattice's
deformation space for that matter). Unlike the traditional mirror tool, this
tool is aware of the fact that the vertex order for lattice control points
matters, and that simply mirroring the coordinates will only cause the lattice
to have an inverted deform along a particular axis (i.e. it will be scaling by a
negative scaling factor along that axis).
The problems (as I discovered the other day) with having such an inverted
deformation space are that:
- the normals of meshes/objects inside that will be incorrect/flipped (and will
disappear in GLSL shading mode for instance)
- transforming objects deformed by the lattices will become really tricky and
counter-intuitive (e.g. rotate in opposite direction by asymmetric amounts to
get desired result)
- it is not always immediately obvious that problems have occurred
Specific use cases this operator is meant to solve:
1) You've created a lattice-based deformer for one cartoonish eye. Now you want
to make the second eye, but want to save some time crafting that basic shape
again but mirrored.
2) You've got an even more finely crafted setup for stretchy-rigs, and now need
to apply it to other parts of the rig.
Notes:
* I've implemented a separate operator for this vs extending/patching mirror
transform tool as it's easier to implement this way, but also because there are
still some cases where the old mirroring seems valid (i.e. you explicitly want
these sort of distortion effects).
* Currently this doesn't take selections into account, as it doesn't seem useful
to do so.
|
|
|
|
common case ripping an edge of the default cube, it didn't run the edge size check.
|
|
|
|
vertex has 3 surrounding verts & faces.
|
|
render
When adjusting settings for world textures (with Both/World preview modes),
every tweak would result in the usercount of the texture increasing. As a
result, before long the texture would claim to have over 100 users. Fortunately,
this only appeared to be just a cosmetic issue (i.e. no real memory leak here),
though it was a bit unsettling.
NOTE: this is still a bit glitchy, as now we have flickering when updating
texture settings - the texture still temporarily has a second user during
preview rendering.
|
|
|
|
|
|
PySequence_Fast.
|
