Blender Git Loki

OpenSubdiv: Remove redundant section define from shader compilation

OpenSubdiv: Split shader source files

Was a bit annoying to do tweaks in a file which contained all
vertex, geometry and fragment shaders.

Merge branch 'master' into asset-engine

GPencil: Remove toggle option for pop-up specials menu

According UI rules, no toggle options in pop-ups menus, so remove it.

Fix T49349: Baking action doesnt bake multiple objects.

Tooltip was wrong here, it can only bake a single (active) object, 'only selected' option
is for pose bones only...

GPencil: Add a new special menu for editing and replace subdivision shortcut

Replace the W shortcut for subdivision by a new menu for edit specials
in order to keep consistency in UI.

Subdivision is not used all the time, so it's better assign this
shortcut to menu.

Fix T49372: Fresnel node: difference between 2.76 and 2.78 GLSL output

GPencil: New subdivide stroke operator

In some situations the artist needs to subdivide a stroke created with
few points before, specially for sculpting.

The subdivision is done for any pair of continuous selected points in
the same stroke.

The operator can be activated in edit mode with W key and has a
parameter for number of cuts.

OpenGL render: Move file writing to a separate thread

The idea is to have a dedicated thread which is responsive for all the
file writing to a separate thread, so slow disk will not slow down
OpenGL itself.

Gives really nice speedup around 1.5x when exporting barber shop layout
file to h264 video.

Fix T49369: Blender crashes/closes down application at alembic export of
any object

There were a couple of crashes caused by stupid typos in
rB631af9f930d2fd2c76751204ff22239aa95f761d and
rB78ea06fea4a74181c25254ed72d50d8a743b6954, as well as a shamefull lack
of 'testing before committing' which only affect exporting.

One crash was due to using RNA_boolean_get instead of RNA_enum_get, the
other one was a tricky case of order of deletion happening in the
destructors of AbcExporter and ArchiveWriter.

Should not affect RC or release.

Fixed a bug in the Disney BSDF that caused specular reflections to be too
bright and diffuse is now reacting to the roughness again

- A normalization for the fresnel was missing which caused the specular
reflections to become too bright for the single-scatter GGX
- The roughness value for the diffuse BSSRDF part has always been
overwritten and thus always 0
- Also the performance for refractive materials with roughness=0.0 has
been improved

removed access to manta grids from blender kernel code (bad sideeffects with new obstacle script code)

Gawain improvements

Fixed compile error in debug build (thanks mont29)

Renamed some functions for consistency.

New features:

Create a Batch with immediate mode! Just use immBeginBatch instead of immBegin. You can keep the result and draw it as many times as you like. This partially replaces the need for display lists.

Copy a VertexFormat, and create a VertexBuffer using an existing format.

Resize a VertexBuffer to a different number of vertices. (can only resize BEFORE using it to draw)

Merge branch 'master' into soc-2016-pbvh-painting

Extend Gawain to use Blender's built-in shaders

Was already done for immediate mode, but rearranged code to make a clean separation. Cleaned up #includes for code that uses this feature.

Added same for batched rendering.

Claude: add basic loading of selected images...

Note that you can link several images in a single run.

Also, images are using current 'link file as datablock' hack, awaiting for propper ID override to get really sane behavior here...

Gawain: batch rendering API

Follow-up to rBddb1d5648dbd

API is nearly complete but untested.

1) create batch with vertex buffer & optional index buffer
2) choose shader program
3) draw!

Cleanup: Redundant forward declaration

Fix mistake in previous commit

Was copying things in other way around and was not performing
proper color space conversion.

OpenGL render: Avoid byte-float-byte roundtrip

It was annoyingly slow to do roundtrip from byte OpenGL render to
float render result and back to byte image format (which is used
in 99% of cases for the OpenGL previews),

Now we use render result's rect32 to store render result which is
already supposed to be in the display space.

Gives about 30% speed improvement for OpenGL previews here.