Sequencer preview was rendering all scenes with the context
resolution, which doesn't give proper result. Now made it so
sequencer scene preview uses scene's resolution.
BKE_sequencer_offset_animdata() was simply assuming bezt member of fcurve was always valid, while it might be NULL (e.g. when fcurve is using FPoints instead, like when generated from sound file).
Basically issue was caused by the fact that strip for proxy has been
post-processed but proxy files were considered pre-processed. This lead
to situation of postprocessing being applied twice.
The issue was caused by the fact that sequencer used to cross-over effect
result with strips used for this effect, which is really stupid.
Now made it so strips which are used for effect inputs are not in the
render stack to be sure they would only be used by effect itself and
wouldn't be blended in any other way.
Summary:
Made objects update happening from multiple threads. It is a task-based
scheduling system which uses current dependency graph for spawning new
tasks. This means threading happens on object level, but the system is
flexible enough for higher granularity.
Technical details:
- Uses task scheduler which was recently committed to trunk
(that one which Brecht ported from Cycles).
- Added two utility functions to dependency graph:
* DAG_threaded_update_begin, which is called to initialize threaded
objects update. It will also schedule root DAG node to the queue,
hence starting evaluation process.
Initialization will calculate how much parents are to be evaluation
before current DAG node can be scheduled. This value is used by task
threads for faster detecting which nodes might be scheduled.
* DAG_threaded_update_handle_node_updated which is called from task
thread function when node was fully handled.
This function decreases num_pending_parents of node children and
schedules children with zero valency.
As it might have become clear, task thread receives DAG nodes and
decides which callback to call for it.
Currently only BKE_object_handle_update is called for object nodes.
In the future it'll call node->callback() from Ali's new DAG.
- This required adding some workarounds to the render pipeline.
Mainly to stop using get_object_dm() from modifiers' apply callback.
Such a call was only a workaround for dependency graph glitch when
rendering scene with, say, boolean modifiers before displaying
this scene.
Such change moves workaround from one place to another, so overall
hackentropy remains the same.
- Added paradigm of EvaluaitonContext. Currently it's more like just a
more reliable replacement for G.is_rendering which fails in some
circumstances.
Future idea of this context is to also store all the local data needed
for objects evaluation such as local time, Copy-on-Write data and so.
There're two types of EvaluationContext:
* Context used for viewport updated and owned by Main. In the future
this context might be easily moved to Window or Screen to allo
per-window/per-screen local time.
* Context used by render engines to evaluate objects for render purposes.
Render engine is an owner of this context.
This context is passed to all object update routines.
Reviewers: brecht, campbellbarton
Reviewed By: brecht
CC: lukastoenne
Differential Revision: https://developer.blender.org/D94
For now just make sure conversion to sequencer space will ensure imbuf's
color space names is set properly.
Might be some further changes needed to make colorspace flow more clear
in sequencer, but that's for later.
Meta sound update (seq_update_sound_bounds_recursive_rec) was not taking into account hard trim (anim_startofs) when setting sound's start, while default sound strip update (sound_move_scene_sound_defaults) did... This could use some refactor, though, with a single func used in both cases, to avoid such issue.
Also added soft trim to sound panel, only hard one was available.
Byte images and movies will now fully follow input color space.
Before this non-sRGB input colorspace for byte images and movies
behave really doggy (results in preview and final render were
totally different).
To prevent data loss, if byte image is set not stored in sequencer's
space it'll be internally converted to float buffer.
In theory some setups might be rendering a bit different now, but
new behavior is totally expected and someone used non-sRGB input
space for byte images/movies had Convert Float enabled anyway.
Simply recalc sequence len for audio (and meta!) strips when modifying fps value. Note start, startofs and endofs are also updated, to try to keep final pos and length as consistent as possible.
existing code was very stupid.
- all ID pointers for clipboard strips are handled uniformly.
- clipboard stores a duplicate ID pointer which are restored on paste.
- restoring pointers...
-- use ID's that are still in the database (copy&paste within the same file).
-- fallback to name lookup.
-- fallback to loading them from the original filepath (movie-clip and sound only).
also fix bug pasting where initialing the sound wasn't done if there was no frame-offset.
Issue happened for scene. movie clip and mask strips, which contains
pointers to datablocks which are freeing on loading new file.
Also, scene strip would crash when pasted from clipboard after scene
was unlinked from file.
Before this change only old flag "Premultiply" was used to
detect alpha mode, which is not enough actually.
Now the logic here is:
- If "Premultiply" was enabled it is likely float image with
straight alpha, which shall be premultiplied before usage.
In this case image/sequence Alpha Mode is set to Straight.
- Otherwise use default alpha mode for image format based on
an extension. This could fail in some cases like TIFF, but
this wasn't handled fully correct in older blender anyway.
Initial discovered issue was that EXR images saved in older
Blender versions were set to Straight alpha mode, which is
obviously a straight way to lots of headache.
