5 years agoCycles: Cleanup, trailing whitespace
Sergey Sharybin [Tue, 2 Aug 2016 09:28:20 +0000 (11:28 +0200)]
Cycles: Cleanup, trailing whitespace

5 years agoCycles: Some cleanup in CUDA device file
Sergey Sharybin [Tue, 2 Aug 2016 09:27:59 +0000 (11:27 +0200)]
Cycles: Some cleanup in CUDA device file

5 years agoCycles: Include NVCC compiler flags into md5 hash
Sergey Sharybin [Tue, 2 Aug 2016 09:24:42 +0000 (11:24 +0200)]
Cycles: Include NVCC compiler flags into md5 hash

This way we can easily switch between toolkits without worrying
whether some kernel was compiled with old or new CUDA toolkit.

It's also now possible to switch machine architecture and have
proper cached kernel detected. Not as if it happens every day,
but i did such a bitness switch back in the days :)

5 years agoFix T48998: 'header input' of rotation transform was clamped in [-PI, PI[ range.
Bastien Montagne [Tue, 2 Aug 2016 10:41:43 +0000 (12:41 +0200)]
Fix T48998: 'header input' of rotation transform was clamped in [-PI, PI[ range.

Can't see any reason for this behavior (inherited from 2.4x code), so for now just removing it.

5 years agoFix T47591: Smoke keeps both .bphys and .vdb cache files
Kévin Dietrich [Tue, 2 Aug 2016 10:20:54 +0000 (12:20 +0200)]
Fix T47591: Smoke keeps both .bphys and .vdb cache files

Delete old caches when changing cache type.

5 years agoFix T47520: Compositor Node "Filter" Outputs clear image
Kévin Dietrich [Tue, 2 Aug 2016 09:38:03 +0000 (11:38 +0200)]
Fix T47520: Compositor Node "Filter" Outputs clear image

Make 'Laplace' filter an edge filter operation, since this what it is
typically used for, and such operation does not affect the input's alpha

Reviewers: sergey, campbellbarton

Reviewed By: sergey

Differential Revision:

5 years agoCycles: tweak some folding tests to avoid unnecessary type conversion.
Alexander Gavrilov [Tue, 2 Aug 2016 09:08:04 +0000 (12:08 +0300)]
Cycles: tweak some folding tests to avoid unnecessary type conversion.

It's easy to connect a simple value to output without forcing
conversion to color by using the Strength socket of Emission.

5 years agoFix tmp screen becoming permanent after going fullscreen
Julian Eisel [Mon, 1 Aug 2016 20:57:12 +0000 (22:57 +0200)]
Fix tmp screen becoming permanent after going fullscreen

Doubt this is intentional.

5 years agoFix T48991: some particles emitted at a distance from object.
Alexander Gavrilov [Mon, 1 Aug 2016 18:45:42 +0000 (21:45 +0300)]
Fix T48991: some particles emitted at a distance from object.

The cause seems to be that despite dt_frac being computed as
1/(subframes+1) with integer subframes value, it doesn't always
add up to exactly 1.0 due to precision limitations. If the sum
is similar to 1.00000???, the last subframe is skipped, and all
particles that were supposed to be emitted in that interval are
emitted next frame, with the code working incorrectly due to
skewed time range.

To fix, separate the code from the dynamic timestep feature that
adjusts the last subframe length into a separate function, and
use it even when dynamic timestep is disabled.

5 years agoAudapsace: Make it work with Jack and no libjack-dev package installed
Sergey Sharybin [Mon, 1 Aug 2016 18:39:54 +0000 (20:39 +0200)]
Audapsace: Make it work with Jack and no libjack-dev package installed

@neXyon, mind backporting it to your upstream? :)

5 years agoBuildbot: Enable proper Pascal architecture
Sergey Sharybin [Mon, 1 Aug 2016 16:26:53 +0000 (18:26 +0200)]
Buildbot: Enable proper Pascal architecture

5 years agoFix strict compiler flags with older GCC
Sergey Sharybin [Mon, 1 Aug 2016 15:58:45 +0000 (17:58 +0200)]
Fix strict compiler flags with older GCC

5 years agoCycles: add unit tests for supported constant folding rules.
Alexander Gavrilov [Mon, 1 Aug 2016 15:53:20 +0000 (18:53 +0300)]
Cycles: add unit tests for supported constant folding rules.

Code coverage of different combinations of secondary conditions
is obviously not complete because there are so many of them, but
all main rules should be there.

The reason for CORRECT vs INVALID is that both words have the same
number of characters so calls line up, but look quite different.

Reviewers: #cycles, sergey

Reviewed By: #cycles, sergey

Subscribers: dingto, sergey, brecht

Differential Revision:

5 years agoBuildbot: Seems quotes are not really needed when passing list
Sergey Sharybin [Mon, 1 Aug 2016 14:25:03 +0000 (16:25 +0200)]
Buildbot: Seems quotes are not really needed when passing list

5 years agoBuildbot: Attempt to enable sm_60 CUDA architecture for build slaves
Sergey Sharybin [Mon, 1 Aug 2016 14:05:33 +0000 (16:05 +0200)]
Buildbot: Attempt to enable sm_60 CUDA architecture for build slaves

5 years agoCycles: Tweaks to support CUDA 8 toolkit
Sergey Sharybin [Mon, 1 Aug 2016 13:40:46 +0000 (15:40 +0200)]
Cycles: Tweaks to support CUDA 8 toolkit

All the changes are mainly giving explicit tips on inlining functions,
so they match how inlining worked with previous toolkit.

This make kernel compiled by CUDA 8 render in average with same speed
as previous kernels. Some scenes are somewhat faster, some of them are
somewhat slower. But slowdown is within 1% so far.

On a positive side it allows us to enable newer generation cards on
buildbots (so GTX 10x0 will be officially supported soon).

5 years agoFix T48901: Blender ignores xinput cursor matrix
Campbell Barton [Mon, 1 Aug 2016 03:40:59 +0000 (13:40 +1000)]
Fix T48901: Blender ignores xinput cursor matrix

5 years agoCleanup: unused vars, imports, pep8
Campbell Barton [Mon, 1 Aug 2016 01:54:02 +0000 (11:54 +1000)]
Cleanup: unused vars, imports, pep8

5 years agoReplace pep8 with flake8 checker
Campbell Barton [Mon, 1 Aug 2016 01:02:01 +0000 (11:02 +1000)]
Replace pep8 with flake8 checker

5 years agoSupport Auto-Clamped Handle for Curve-Mapping
Campbell Barton [Sun, 31 Jul 2016 23:01:43 +0000 (09:01 +1000)]
Support Auto-Clamped Handle for Curve-Mapping

This patch supports auto-clamped handles for curves,
useful since without this it can be difficult to have 'flat' sections of a curve.

5 years agoFix T48980: crash when loading a file that contains a custom node tree.
Bastien Montagne [Sun, 31 Jul 2016 18:07:33 +0000 (20:07 +0200)]
Fix T48980: crash when loading a file that contains a custom node tree.

Syupid logical mistake in own recent rB018d336cbd51...

5 years agoCorrection to 0f2ff52c0866
Julian Eisel [Sun, 31 Jul 2016 16:35:38 +0000 (18:35 +0200)]
Correction to 0f2ff52c0866

This info is shown while dragging, so should stay where it was.

5 years agoEnable Undo for "Add background Image"
Julian Eisel [Sun, 31 Jul 2016 15:33:57 +0000 (17:33 +0200)]
Enable Undo for "Add background Image"

Enabled the Undo option for the operation "Add background Image". This is to treat adding a background image like opening an image. This behavior is useful when a background image is add using 'drag and drop'.

Reviewers: Severin

Reviewed By: Severin

Subscribers: brecht

Tags: #bf_blender

Maniphest Tasks: T47050

Differential Revision:

5 years agoFix mistake in Cycles mix node multiply constant folding.
Brecht Van Lommel [Sun, 31 Jul 2016 12:56:49 +0000 (14:56 +0200)]
Fix mistake in Cycles mix node multiply constant folding.

5 years agoCleanup: reduce Vector_getseters size using macros
Campbell Barton [Sun, 31 Jul 2016 10:41:24 +0000 (20:41 +1000)]
Cleanup: reduce Vector_getseters size using macros

5 years agoCorrect own error in last commit
Campbell Barton [Sun, 31 Jul 2016 09:12:40 +0000 (19:12 +1000)]
Correct own error in last commit

5 years agoSupport limiting collisions by group for softbody and particles
Alexander Gavrilov [Sun, 31 Jul 2016 08:56:44 +0000 (18:56 +1000)]
Support limiting collisions by group for softbody and particles

This feature is extremely useful for layering multiple cloth objects,
and there is no reason there shouldn't be the same kind of feature for softbody.

5 years agoCleanup: spelling, style
Campbell Barton [Sun, 31 Jul 2016 07:41:05 +0000 (17:41 +1000)]
Cleanup: spelling, style

5 years agoPyAPI: minor optimization for dictionary creation
Campbell Barton [Sun, 31 Jul 2016 07:22:04 +0000 (17:22 +1000)]
PyAPI: minor optimization for dictionary creation

Pass size when its known.

5 years agoPyAPI: fix leak in unlikely case converting idprops fail
Campbell Barton [Sun, 31 Jul 2016 06:59:08 +0000 (16:59 +1000)]
PyAPI: fix leak in unlikely case converting idprops fail

5 years agoPyAPI: Add PyC_UnicodeAsByteAndSize
Campbell Barton [Sun, 31 Jul 2016 06:52:44 +0000 (16:52 +1000)]
PyAPI: Add PyC_UnicodeAsByteAndSize

Read the string length from Python directly when assigning id-properties

5 years agoDocs: add initial idprop.types API docs
Campbell Barton [Sun, 31 Jul 2016 05:39:30 +0000 (15:39 +1000)]
Docs: add initial idprop.types API docs

5 years agoDocs: simplify EXCLUDE_MODULES handling
Campbell Barton [Sun, 31 Jul 2016 04:37:25 +0000 (14:37 +1000)]
Docs: simplify EXCLUDE_MODULES handling

5 years agoCleanup: capitalize Blender & Python
Campbell Barton [Sun, 31 Jul 2016 04:31:39 +0000 (14:31 +1000)]
Cleanup: capitalize Blender & Python

5 years agosimplify redundant conditionals
Mike Erwin [Sun, 31 Jul 2016 05:22:36 +0000 (01:22 -0400)]
simplify redundant conditionals

The redundant terms were harmless but check an expression we already
know to be true (from earlier in the same conditional).

Found by PVS-Studio T48917

5 years agoImprove description for use_self py-driver option
Campbell Barton [Sun, 31 Jul 2016 04:21:16 +0000 (14:21 +1000)]
Improve description for use_self py-driver option

5 years agoCurve Fitting: circular fit could give NAN handles
Campbell Barton [Sun, 31 Jul 2016 03:55:50 +0000 (13:55 +1000)]
Curve Fitting: circular fit could give NAN handles

Fitting lines that exactly double back on themselves could give NAN length handles.

5 years agoCleanup: rename btree (incorrect & confusing)
Campbell Barton [Sun, 31 Jul 2016 02:02:33 +0000 (12:02 +1000)]
Cleanup: rename btree (incorrect & confusing)

5 years agoFix Cycles OpenCL compile error on Windows.
Brecht Van Lommel [Sat, 30 Jul 2016 23:54:05 +0000 (01:54 +0200)]
Fix Cycles OpenCL compile error on Windows.

5 years agoPy-Driver: re-use self PyObject when its unchanged.
Campbell Barton [Sun, 31 Jul 2016 01:43:24 +0000 (11:43 +1000)]
Py-Driver: re-use self PyObject when its unchanged.

5 years agoPy-Driver: use intern strings
Campbell Barton [Sun, 31 Jul 2016 01:22:02 +0000 (11:22 +1000)]
Py-Driver: use intern strings

Avoids managing strings inline

5 years agoFix for previous commit changing the addons submodule.
Brecht Van Lommel [Sun, 31 Jul 2016 01:23:51 +0000 (03:23 +0200)]
Fix for previous commit changing the addons submodule.

5 years agoCycles: refactor kernel closure storage to use structs per closure type.
Brecht Van Lommel [Mon, 25 Jul 2016 01:03:23 +0000 (03:03 +0200)]
Cycles: refactor kernel closure storage to use structs per closure type.

Reviewed By: dingto, sergey

Differential Revision:

5 years agoCycles: constant fold add/mul type nodes with known 0 and 1 arguments.
Alexander Gavrilov [Sat, 30 Jul 2016 18:03:49 +0000 (20:03 +0200)]
Cycles: constant fold add/mul type nodes with known 0 and 1 arguments.

These values often either turn the node into a no-op, or even make it
evaluate to 0 no matter what the other input value is, thus allowing
deletion of a branch of the node graph that otherwise is not constant.

Reviewed By: brecht

Differential Revision:

5 years agoCycles: constant folding for RGB/Vector Curves and Color Ramp.
Alexander Gavrilov [Sat, 30 Jul 2016 21:30:36 +0000 (23:30 +0200)]
Cycles: constant folding for RGB/Vector Curves and Color Ramp.

These are complex nodes, and it's conceivable they may end up constant
in some circumstances within node groups, so folding support is useful.

Reviewed By: brecht

Differential Revision:

5 years agoFix Cycles OpenCL not always allocating enough space for closures.
Brecht Van Lommel [Sat, 30 Jul 2016 17:16:29 +0000 (19:16 +0200)]
Fix Cycles OpenCL not always allocating enough space for closures.

5 years agoFix CUDA warning, due to extra ; at the line ending.
Thomas Dinges [Sat, 30 Jul 2016 19:37:20 +0000 (21:37 +0200)]
Fix CUDA warning, due to extra ; at the line ending.

5 years agoCycles OpenCL: use #line directives for better error messages.
Brecht Van Lommel [Sat, 30 Jul 2016 14:54:48 +0000 (16:54 +0200)]
Cycles OpenCL: use #line directives for better error messages.

5 years agoCycles OpenCL: detect incorrect usage of SOA members in the split kernel.
Brecht Van Lommel [Sat, 30 Jul 2016 13:18:21 +0000 (15:18 +0200)]
Cycles OpenCL: detect incorrect usage of SOA members in the split kernel.

5 years agoFix Cycles OpenCL address space compile error with amdgpu-pro drivers on Linux.
Brecht Van Lommel [Sat, 30 Jul 2016 13:17:30 +0000 (15:17 +0200)]
Fix Cycles OpenCL address space compile error with amdgpu-pro drivers on Linux.

5 years agoCleanup: pep8
Campbell Barton [Sat, 30 Jul 2016 08:39:22 +0000 (18:39 +1000)]
Cleanup: pep8

5 years agoFix bplayer (c)
Thomas Beck [Sat, 30 Jul 2016 08:35:30 +0000 (10:35 +0200)]
Fix bplayer (c)

5 years agoCleanup: pep8
Campbell Barton [Sat, 30 Jul 2016 08:15:16 +0000 (18:15 +1000)]
Cleanup: pep8

5 years agoCleanup: minor changes from last commit
Campbell Barton [Sat, 30 Jul 2016 06:47:18 +0000 (16:47 +1000)]
Cleanup: minor changes from last commit

Un-indent blocks, rename vars.

5 years agoPy-Driver: add 'self' option
Campbell Barton [Sat, 30 Jul 2016 06:34:01 +0000 (16:34 +1000)]
Py-Driver: add 'self' option

Drivers can use this to refer to the data which the driver is applied to,
useful for objects, bones, to avoid having to create a variable pointing to its self.

5 years agoCloth Simulation: add time scale property
Campbell Barton [Sat, 30 Jul 2016 04:46:19 +0000 (14:46 +1000)]
Cloth Simulation: add time scale property

This setting can also be animated, to create a "time warp" effect.

D2122 by @LucaRood

5 years agoCloth: option to use dynamic base mesh
Campbell Barton [Sat, 30 Jul 2016 04:47:31 +0000 (14:47 +1000)]
Cloth: option to use dynamic base mesh

This adds the ability for cloth simulations to respect changes in the underlying mesh.
So you can for instance, animate shape keys, armatures, or add any deformation modifiers (above the cloth modifier).

This is mainly useful for (but not limited to) cartoon animations,
where your character might stretch or change shape, and you want the clothes to follow accordingly.

D1903 by @LucaRood

5 years agoCloth: UI cleanup & quality hard limit removal
Campbell Barton [Sat, 30 Jul 2016 04:44:28 +0000 (14:44 +1000)]
Cloth: UI cleanup & quality hard limit removal

D2121 by @LucaRood

5 years agoCleanup: pep8
Campbell Barton [Fri, 29 Jul 2016 11:22:27 +0000 (21:22 +1000)]
Cleanup: pep8

5 years agoAdd copy/make_local to movieclip datablock.
Bastien Montagne [Fri, 29 Jul 2016 21:13:31 +0000 (23:13 +0200)]
Add copy/make_local to movieclip datablock.

As title says... Copying tracking data from movieclip was not the simplest thing...

Reviewers: sergey

Differential Revision:

5 years agoGtests: Re-apply strict compiler fixes
Sergey Sharybin [Fri, 29 Jul 2016 16:59:54 +0000 (18:59 +0200)]
Gtests: Re-apply strict compiler fixes

5 years agoCycles: Add fundamentals to test constant folding
Sergey Sharybin [Fri, 29 Jul 2016 16:44:29 +0000 (18:44 +0200)]
Cycles: Add fundamentals to test constant folding

This commit adds some easy to use way to check whether the graph
was properly optimized. The idea is based on using mock glog sync
which keeps track on all messages and expects specific log message
to appear in the logging.

This means each optimization step should report that it optimized
something (specifying what exactly was optimized) and what was used
for optimization.

There's also some ease-ish way to define shader networks which we
might want to make a bit more global and available for all tests
in the future.

For now only RGBToBW node is covered as an example, really hope
to get help from active guys in the community to finish covering
all existing optimization cases. Ready to fix any possible issues
with builder when needed tho :)

5 years agoAdd GMock library which is responsive for dealing with mock objects
Sergey Sharybin [Fri, 29 Jul 2016 16:34:33 +0000 (18:34 +0200)]
Add GMock library which is responsive for dealing with mock objects

5 years agoGTests: Update to a newer Git version
Sergey Sharybin [Fri, 29 Jul 2016 16:33:27 +0000 (18:33 +0200)]
GTests: Update to a newer Git version

Required to get GMock working with GTest.

5 years agoFix missing redraw of timeline when changing GPencil layer settings
Julian Eisel [Fri, 29 Jul 2016 18:42:57 +0000 (20:42 +0200)]
Fix missing redraw of timeline when changing GPencil layer settings

5 years agoCleanup space/tabs, Grrr.
Bastien Montagne [Fri, 29 Jul 2016 16:49:50 +0000 (18:49 +0200)]
Cleanup space/tabs, Grrr.

5 years agoFix a bunch of missing expand calls in object/particle physics area, was breaking...
Bastien Montagne [Fri, 29 Jul 2016 15:35:31 +0000 (17:35 +0200)]
Fix a bunch of missing expand calls in object/particle physics area, was breaking linking of those.

Initial report/patch by Alexander Gavrilov (@angavrilov), found more on the road.

Nice demo of why we should use libquery ID looper in read code too - but that's for another day
(also because read code needs to take care of some deprecated pointers sometimes)...

5 years agoCycles: Fix wrong projection coordinates for points behind the camera
Sergey Sharybin [Fri, 29 Jul 2016 15:28:36 +0000 (17:28 +0200)]
Cycles: Fix wrong projection coordinates for points behind the camera

Points behind camera will be flipped on projection, but that did not

This solves second issue reported at T48848.

5 years agoCycles: Fix wrong check for object is behind camera after recent fixes
Sergey Sharybin [Fri, 29 Jul 2016 14:37:02 +0000 (16:37 +0200)]
Cycles: Fix wrong check for object is behind camera after recent fixes

5 years agoFix T48971: Append creates linked image textures if object has shape keys.
Bastien Montagne [Fri, 29 Jul 2016 15:00:29 +0000 (17:00 +0200)]
Fix T48971: Append creates linked image textures if object has shape keys.

Hating all those not-so-real ID types... Here there were two causes for the issue:
1) Linked shapekey ID was made local twice (once from mesh's make local, once by itself).
   Solved by not explicitely making shapekeys (nor any other non-linkable datatype) local.
2) Key->from 'back pointer' to its owner was messing 'still in used' detection of linked data
   after localization. Fixed with a hack for now, thinking correct solution might actually
   be to not consider this pointer at all in libquery ID looper, since it's nothing like
   and actual usage of mesh/lattice/curve.

Again, shapekeys as ID is a joke, those should be mere struct, they have absolutely nothing to do in Main, period. :(

Point 2) still demonstrates the need for better handling of IDs dependencies though,
so far we only hit corner cases, but think there could also be valid cases generating those
'dependency cycles' between IDs (ID a using ID b which uses ID a), this will have to be addressed some day...

5 years agoRemove "debug build only" event printing.
Bastien Montagne [Fri, 29 Jul 2016 12:45:15 +0000 (14:45 +0200)]
Remove "debug build only" event printing.

This makes no sense since user usually only has release build... And those noisy printings are
behind debug flag anyway, so really no reason to only print them out in debug build.

5 years agoCycles: Fix strict compilation warning
Sergey Sharybin [Fri, 29 Jul 2016 13:06:13 +0000 (15:06 +0200)]
Cycles: Fix strict compilation warning

5 years agoFix T48679: Bone transform only alters between translation and rotation
Sergey Sharybin [Fri, 29 Jul 2016 12:45:27 +0000 (14:45 +0200)]
Fix T48679: Bone transform only alters between translation and rotation

There was some smart code in the transform which would alter between translation
and rotation based on whether bone is connected or not and whether translation is
locked or not.

It could be handy to also fallback to scale if both rotation and translation are

5 years agoCycles: remove narrowing conversion from recent commit
Mai Lavelle [Fri, 29 Jul 2016 08:00:37 +0000 (04:00 -0400)]
Cycles: remove narrowing conversion from recent commit

5 years agoCycles microdisplacement: move subdivision options to subsurf modifier
Mai Lavelle [Sat, 16 Jul 2016 23:56:45 +0000 (19:56 -0400)]
Cycles microdisplacement: move subdivision options to subsurf modifier

Subdivision options can now be found in the subsurf modifier. The modifier must
be the last in the stack or the options will be unavailable. Catmull-Clark
subdivision is still unavailable and will fallback to linear subdivision instead

Reviewed By: brecht

Differential Revision:

5 years agoCycles microdisplacement: ngons and attributes for subdivision meshes
Mai Lavelle [Sat, 16 Jul 2016 23:42:28 +0000 (19:42 -0400)]
Cycles microdisplacement: ngons and attributes for subdivision meshes

This adds support for ngons and attributes on subdivision meshes. Ngons are
needed for proper attribute interpolation as well as correct Catmull-Clark
subdivision. Several changes are made to achieve this:

- new primitive `SubdFace` added to `Mesh`
- 3 more textures are used to store info on patches from subd meshes
- Blender export uses loop interface instead of tessface for subd meshes
- `Attribute` class is updated with a simplified way to pass primitive counts
  around and to support ngons.
- extra points for ngons are generated for O(1) attribute interpolation
- curves are temporally disabled on subd meshes to avoid various bugs with
- old unneeded code is removed from `subd/`
- various fixes and improvements

Reviewed By: brecht

Differential Revision:

5 years agoMissing free for single-sided edge-slide failure
Campbell Barton [Thu, 28 Jul 2016 23:50:55 +0000 (09:50 +1000)]
Missing free for single-sided edge-slide failure

5 years agoFix Cycles OSL hair BSDF inconsistencies with SVM.
Brecht Van Lommel [Fri, 29 Jul 2016 00:22:01 +0000 (02:22 +0200)]
Fix Cycles OSL hair BSDF inconsistencies with SVM.

5 years agoFix Cycles ctests not always deleting fail.png files on success.
Brecht Van Lommel [Fri, 29 Jul 2016 00:04:45 +0000 (02:04 +0200)]
Fix Cycles ctests not always deleting fail.png files on success.

5 years agoFix use of uninitialized variable in Cycles, mostly harmless.
Brecht Van Lommel [Thu, 28 Jul 2016 23:04:01 +0000 (01:04 +0200)]
Fix use of uninitialized variable in Cycles, mostly harmless.

5 years agoMake ID types typed enum, to easily spot missing ones in core switches of library.c
Bastien Montagne [Thu, 28 Jul 2016 13:21:53 +0000 (15:21 +0200)]
Make ID types typed enum, to easily spot missing ones in core switches of library.c

Note that all deprecated/non-real ID types are kept as defines.

5 years agoFix T48965: Cannot Append Palette As Local Datablock.
Bastien Montagne [Thu, 28 Jul 2016 12:40:26 +0000 (14:40 +0200)]
Fix T48965: Cannot Append Palette As Local Datablock.

Palette and PaintCurve were totally missing from id_copy/id_make_local switch... :/

5 years agoCycles: Fix difference in Ashikhmin Shirley shader between CPU and GPU
Sergey Sharybin [Thu, 28 Jul 2016 16:46:29 +0000 (18:46 +0200)]
Cycles: Fix difference in Ashikhmin Shirley shader between CPU and GPU

The issue was caused by some NaN appearing in calculations.

Visible with scifi_armor_concept.blend from the cloud.

5 years agoCleanup: Indentation
Sergey Sharybin [Thu, 28 Jul 2016 12:17:35 +0000 (14:17 +0200)]
Cleanup: Indentation

5 years agoFix T48838: Tiling sculpt problem with custom brush
Sergey Sharybin [Thu, 28 Jul 2016 12:00:44 +0000 (14:00 +0200)]
Fix T48838: Tiling sculpt problem with custom brush

5 years agoFix T48962: Edge-slide, random edge switch order
Campbell Barton [Thu, 28 Jul 2016 11:48:07 +0000 (21:48 +1000)]
Fix T48962: Edge-slide, random edge switch order

5 years agoCycles: Fix missing mesh flags update when tweaking shaders
Sergey Sharybin [Thu, 28 Jul 2016 10:37:44 +0000 (12:37 +0200)]
Cycles: Fix missing mesh flags update when tweaking shaders

This is a bit weak, but better than tagging whole mesh manager for update.

Maybe we'll solve such dual-look up in the future.

This commit finally solves T48963: Noise when changing Diffuse node to Emission node

5 years agoCycles: Fix wrong order of light manager update tag
Sergey Sharybin [Thu, 28 Jul 2016 10:31:01 +0000 (12:31 +0200)]
Cycles: Fix wrong order of light manager update tag

Shader's has_surface_emission will only be known for sure after compilation.

5 years agoCycles: De-duplicate more checks around light emisive meshes
Sergey Sharybin [Thu, 28 Jul 2016 10:27:24 +0000 (12:27 +0200)]
Cycles: De-duplicate more checks around light emisive meshes

Once again, should be no functional changes.

5 years agoCycles: De-duplicate check for MIS shaders in meshes
Sergey Sharybin [Thu, 28 Jul 2016 10:20:47 +0000 (12:20 +0200)]
Cycles: De-duplicate check for MIS shaders in meshes

Should be no functional changes.

5 years agoCycles: Add some extra logging for light debugging
Sergey Sharybin [Thu, 28 Jul 2016 10:13:11 +0000 (12:13 +0200)]
Cycles: Add some extra logging for light debugging

5 years agoDupliObject: Remove unused and confusing index argument from the context
Sergey Sharybin [Thu, 28 Jul 2016 08:42:52 +0000 (10:42 +0200)]
DupliObject: Remove unused and confusing index argument from the context

5 years agoCurve Fitting: Use normalized projection function
Campbell Barton [Thu, 28 Jul 2016 08:33:14 +0000 (18:33 +1000)]
Curve Fitting: Use normalized projection function

5 years agoFix T48404: Translate fails w/ snap orientation option
Campbell Barton [Thu, 28 Jul 2016 08:11:43 +0000 (18:11 +1000)]
Fix T48404: Translate fails w/ snap orientation option

Rotate the object around the snap-target so it meets the snap-point.

5 years agoCleanup: unused transform member
Campbell Barton [Thu, 28 Jul 2016 06:55:57 +0000 (16:55 +1000)]
Cleanup: unused transform member

5 years agoFix T48961: Material slots become easily desynced over linked duplicates.
Bastien Montagne [Thu, 28 Jul 2016 07:23:46 +0000 (09:23 +0200)]
Fix T48961: Material slots become easily desynced over linked duplicates.

Own mistake in recent changes (split to test_object_materials()/test_all_objects_materials()).

5 years agoFix crash fitting single point curve
Campbell Barton [Thu, 28 Jul 2016 04:12:12 +0000 (14:12 +1000)]
Fix crash fitting single point curve

5 years agoUse COMPAT_ENGINES instead of RenderEngine.use_game_engine for panel poll methods
Mitchell Stokes [Tue, 5 Jul 2016 05:05:32 +0000 (22:05 -0700)]
Use COMPAT_ENGINES instead of RenderEngine.use_game_engine for panel poll methods

This mostly affects physics panels. Any engines relying on
RenderEngine.use_game_engine flag to show/hide panels will need to be
updated. The COMPAT_ENGINES technique is how we usually deal with this.
One issue with use_game_engine is that I cannot find a way to set it; it
appears only the BGE can set it. This means (without this commit)
external RenderEngines cannot get rid of the default physics panels.

The RE_GAME flag (the C flag behind use_game_engine) is pretty hacky
and we should look into removing its usage where possible.

5 years agoFix T48951: Crash refreshing some popups
Campbell Barton [Thu, 28 Jul 2016 01:08:01 +0000 (11:08 +1000)]
Fix T48951: Crash refreshing some popups

Fix for T48874 caused popups to be tagged to be refreshed, that didnt support it.

Tag popups that can't be refreshed.

5 years agoAdd comments on behavior of spacedata and regionbase lists
Julian Eisel [Thu, 28 Jul 2016 00:36:12 +0000 (02:36 +0200)]
Add comments on behavior of spacedata and regionbase lists