EWA sampling is designed for downsampling images, i.e. scaling down the size of
input image pixels, which happens regularly in compositing. While the standard
sampling methods (linear, cubic) work reasonably well for linear
transformations, they don't yield good results in non-linear cases like
perspective projection or arbitrary displacement. EWA sampling is comparable to
mipmapping, but avoids problems with discontinuities.
To work correctly the EWA algorithm needs partial derivatives of the mapping
functions which convert output pixel coordinates back into the input image
space (2x2 Jacobian matrix). With these derivatives the EWA algorithm
projects ellipses into the input space and accumulates colors over their
area. This calculation was not done correctly in the compositor, only the
derivatives du/dx and dv/dy were calculation, basically this means it only
worked for non-rotated input images.
The patch introduces full derivative calculations du/dx, du/dy, dv/dx, dv/dy for
the 3 nodes which use EWA sampling currently: PlaneTrackWarp, MapUV and
Displace. In addition the calculation of ellipsis area and axis-aligned
bounding boxes has been fixed.
For the MapUV and Displace nodes the derivatives have to be estimated by
evaluating the UV/displacement inputs with 1-pixel offsets, which can still have
problems on discontinuities and sub-pixel variations. These potential problems
can only be alleviated by more radical design changes in the compositor
functions, which are out of scope for now. Basically the values passed to the
UV/Displacement inputs would need to be associated with their 1st order
derivatives, which requires a general approach to derivatives in all nodes.
inputs.
http://wiki.blender.org/uploads/4/4c/Compo_image_interpolation_borders.png
Problem is that all image buffer reader nodes (RenderLayer, Image,
MovieClip) were clipping pixel coordinates to 0..N range (N being width
or height respectively). Bilinear interpolation works ok then on the
upper-right borders (x, N) and (N, y), since the last (N-1) pixel fades
out to N (background). But the lower-left (x, 0) and (0, y) borders are
not correctly interpolated because the nodes cut off the negative pixels
before the interpolation function can calculate their value.
To fix this, the interpolation functions are now entirely responsible
for handling "out of range" cases, i.e. setting (0,0,0,0) results for
invalid pixels, while also handling interpolation for borders.
Callers should not do pixel range checks themselves, which also makes
the code simpler. Should not have any real performance penalty,
the interpolation functions do this check anyway, so is probably even
slightly faster.
Also tweaked a bit how nodes with more than one socket available are drawn:
* The node entry itself (a label) has no more any icon, so that it is aligned with the other single-socket nodes.
* The sub-entries (i.e. node's sockets) are more clearly indented (using a blank icon).
Details:
Another case of bad internal UI_HAS_ICON flag erasing by brutally assigning an uneeded value to but->flag.
Revealed by recent split of some button flags into but->drawflag.
Thanks to Thomas Dinge for noting the issue.
This patch changes most of the reamining degrees usage in internal code into radians.
I let a few which I know off asside, for reasons explained below - and I'm not sure to have found out all of them.
WARNING: this introduces forward incompatibility, which means files saved from this version won't open 100% correctly
in previous versions (a few angle properties would use radians values as degrees...).
Details:
- Data:
-- Lamp.spotsize: Game engine exposed this setting in degrees, to not break the API here I kept it as such
(using getter/setter functions), still using radians internally.
-- Mesh.smoothresh: Didn't touch to this one, as we will hopefully replace it completely by loop normals currently in dev.
- Modifiers:
-- EdgeSplitModifierData.split_angle, BevelModifierData.bevel_angle: Done.
- Postprocessing:
-- WipeVars.angle (sequencer's effect), NodeBokehImage.angle, NodeBoxMask.rotation, NodeEllipseMask.rotation: Done.
- BGE:
-- bConstraintActuator: Orientation type done (the minloc[0] & maxloc[0] cases). Did not touch to 'limit location' type,
it can also limit rotation, but it exposes through RNA the same limit_min/limit_max, which hence
can be either distance or angle values, depending on the mode. Will leave this to BGE team.
-- bSoundActuator.cone_outer_angle_3d, bSoundActuator.cone_inner_angle_3d: Done (note I kept degrees in BGE itself,
as it seems this is the expected value here...).
-- bRadarSensor.angle: Done.
Reviewers: brecht, campbellbarton, sergey, gaiaclary, dfelinto, moguri, jbakker, lukastoenne, howardt
Reviewed By: brecht, campbellbarton, sergey, gaiaclary, moguri, jbakker, lukastoenne, howardt
Thanks to all!
Differential Revision: http://developer.blender.org/D59
Summary:
Issue was caused by GLSL samplers not valid anymore after
dupliobject was drawn into OpenGL displist.
Solution for now is to not use displist for active object
when matcap is used.
The same issue exists for regular GLSL viewport draw, and
this was solved by disabling diplist if draw_glsl_material
returns truth for current dupliobject.
Reviewers: brecht
Reviewed By: brecht
Maniphest Tasks: T37663
Differential Revision: http://developer.blender.org/D70
Summary:
Seems to be known TODO in the code, but no idea why it
was never solved, especially since tweak is so much easy.
It might be arguable that we need to support painting
color space, but it's still much better to convert to
sRGB space. It's gonna to cover 90% of cases anyway.
Reviewers: campbellbarton, brecht
Reviewed By: brecht
Maniphest Tasks: T37670
Differential Revision: http://developer.blender.org/D65
This is a addtion to the dynamic fly mode.
It behaves as the first person navigation system available in most 3d world games nowadays.
You can alternate between the old mode (Fly) and the new mode (Walk) in User Preferences > Inputs
Manual:
-------
http://wiki.blender.org/index.php/Doc:2.6/Manual/3D_interaction/Navigating/3D_View#View_Navigationhttp://wiki.blender.org/index.php/Doc:2.6/Manual/3D_interaction/Navigating/3D_View/Navigation_Modes
Shortcuts:
----------
WASD (hold) - Move forward/backward and straft left/right
QE (hold) - Move up and down
Tab - Alternate between Walk and Fly modes
Shift (hold) - Speed up movement
Alt (hold) - Slow down movement
Space or MMB - Teleport
V - Jump
+/- or mouse wheel - speed increase/decrease speed for this Blender session
User Preferences Options:
-------------------------
Navigation Mode - fly/walk navigation systems (fly is the old, walk is the new, next options are for walk mode only)
Gravity - alternate between free navigation and walk with gravity modes
Mouse Sensitivity - sensitivity factor to mouse influence to look around
Teleport Duration - how long the teleport lasts
Camera Height - camera height to use in gravity mode
Jump Height - maximum jump speed in m/s
Move Speed - base move speed in m/s
Boost Factor - multiplication factor when running or going slow (1/boost)
Development Notes:
------------------
* The initial code was based on view3d_fly.c.
* The NDoF code was not touched, so it most likely is not working.
Pending Issues:
---------------
* Draw in the UI the shortcut options, and current values (e.g., Mode: Fly/Walk)
(we need a proper API for that)
* OSX seems to present issues if we re-center the mouse every time. We implemented a workaround for that, but a real fix would be welcome.
Code reviewed and with collaborations from Campbell Barton - @campbellbarton
Differential Revision: http://developer.blender.org/D30
When the desired widths (offsets) of beveled edges cannot be
satisfied, often because we want them to meet on an intermediate
non-beveled edge, we need to compromise on the widths somehow.
This code changes the compromise to minimize the sum of squares
of errors in the offsets. It also adds a global width consistency
pass: starting from a vertex that needed width adjustment, it
uses a breadth-first search to try to propagate the adjustments
and keep the bevel widths from having to taper along the edges.
Also fixed a case where a reflex angle would cause bad results.
Also fixed the way the 'percentage' width method was calculated.
Issue was caused by linear float buffer creating for every working
thread. This buffer actually duplicated original buffer which doubles
amount of required memory.
We can not avoid such a duplication, because OCIO needs to work on a
float buffer and it modifies this buffer.
Alternative for now is to not allocate linear buffer for the whole chunk
which needs to be handled by the thread and use further chunk cutting in
thread itself.
So now every thread will handle the chunk in blocks of 64 scanlines.
This reduces memory overhead significantly without speed loss in own
tests.
Ideally, IMB_processor_apply_threaded need to be switched to generic
task scheduler and made it so this function generates tasks with
reasonable number of scanlines. This requires much more testing to
be sure there're no conflicts with object update and so.
Such a change to IMB_processor_apply_threaded would not be noticed by
users, so do not consider this is a crucial to do right now.
- could crash if triangulate attempted to create an existing face.
- tagging edges to rotate was unreliable, don't do this anymore.
now check if edge is in the array passed to the beauty function.
Simple/predictable polygon filling functions (no hole support)
originally from libgdx which have some advantages over scanfill.
- always creates the same number of triangles (never any missing faces).
- gives same results for any affine transformation.
- doesn't give so many skinny faces by default.
made some changes for Blender.
- remove last ears first (less to memmove)
- step over the ears while clipping to avoid some verts becoming fans to most of the other.
The property sensor was using CValue::FindIdentifier(), which does an AddRef(). However,
the property sensor was not calling Release() when it was done with the value. This could
cause more leaks when used in conjunction with the copy property actuator since it would
really throw off ref counts.
This attribute allows to open a blend file as an alternative
start-up file. The attribute is only available from python.
This is an example call:
bpy.ops.wm.read_homefile(filepath='path/to/a/file.blend')
This patch also changes readfile.c to ensure that unintentionally
stored cursor states are not imported from the loaded file.
Changes for VC2013
Now, I can build Blender with VC2013 with Cycles, Collada, OpenExr,OpenImageIO disabled. Also, you need VC2008 sp1 installed to make old libs compatible.
Summary:
Now it's possible to assign an image to plane tracks
in clip editor. This image is only used for display
in clip editor and this image is being warped into
the plane track rectangle.
Main purpose of this is to get early feedback about
how good image warping matches the footage, before
clip goes to the compositor.
Pretty much straightforward change: just compute
homography from undeformed normalized frame corner
coordinates (unity square) to plane marker corners
and apply this matrix to opengl stack.
Still could improve behavior when perspective
plane transform is degenerate, but that's not so
much critical for now i'd say.
Reviewers: brecht, campbellbarton
Reviewed By: brecht
CC: sebastian_k
Differential Revision: http://developer.blender.org/D57
Summary:
Makes readfile.c more clear and easier to navigate in.
We would need to do such a move when we'll do 2.70 bump
anyway. So better be prepared.
In fact, as soon 2.70 bumo is done i would think we'd
better create versioning_270 file and do version code
there. Otherwise readfile.c will grow up dramatically
again.
Reviewers: brecht, campbellbarton
Reviewed By: brecht
Differential Revision: http://developer.blender.org/D60
There is a bunch of internal refactoring going on too:
* No longer use operators to handle these directory and file fields, only makes
things more complicated than they should be.
* Handle autocomplete partial/full match deeper in the UI code
* Directory field still does not keep focus, that's for another time to fix,
you can already do pretty quick keyboard only navigation with the file field.
Reviewed By: elubie
Differential Revision: http://developer.blender.org/D29
