Commit Graph

3135 Commits

Author SHA1 Message Date
Thomas Dinges
1cd3676d4d Cleanup, remove unneeded variable. 2016-06-12 00:01:57 +02:00
Thomas Dinges
324bed3843 Fix compiler warning for unused variables. 2016-06-11 23:58:12 +02:00
ebdd2e0b6d Cycles: make shader node enums consistently lower case, update OSL shaders accordingly. 2016-06-11 23:50:11 +02:00
Thomas Dinges
5b061ddf1e Fix Gradient Texture and OSL after refactor. 2016-06-11 21:40:00 +02:00
42aec3b355 Cycles: nodify shader nodes
Differential Revision: https://developer.blender.org/D2038
2016-06-11 20:32:24 +02:00
Sergey Sharybin
6307e823ca Cycles: Fix crash after recent zero scale instance optimization 2016-06-08 12:25:35 +02:00
Lukas Stockner
654019fa01 Cycles: Fix two numerical issues in the volume code
This hopefully fixes T48383 by avoiding two numerical problems that I found in the volume code.

Reviewers: sergey, dingto, brecht

Reviewed By: sergey, dingto, brecht

Maniphest Tasks: T48383

Differential Revision: https://developer.blender.org/D2051
2016-06-08 03:17:19 +02:00
Sergey Sharybin
b595a692c8 Cycles: Limit degenerated triangle check got CUDA only
OpenCL seems to work fine here, and for some reason that comparison was
giving compilation error on OpenCL here.

Better to compile OpenCL kernel than to be fully robust to weird corner
cases.
2016-06-07 15:48:56 +02:00
Mai Lavelle
b6954c8da1 Cycles: Fix regression introduced in c96a4c8
A few places still needed to be updated to use the new Mesh::num_triangles()
method; wrong number from triangles.size() was causing crashes.
2016-06-07 07:38:09 -04:00
Lukas Stockner
7a5a02509b Cycles: Use faster ray-quad-intersection test
The original quad intersection test works by just testing against the two triangles that define the quad.
However, in this case it's actually faster to use the same test that's also used for portals: Determining
the distance to the plane in which the quad lies, calculating the hitpoint and checking whether it's in the
quad by projecting onto the sides.

Reviewers: brecht, sergey, dingto

Reviewed By: dingto

Differential Revision: https://developer.blender.org/D2045
2016-06-06 23:38:50 +02:00
Sergey Sharybin
14f9a5aa1d Fix T48571: Cycles/GPU - A lot of fireflies on SSS+Volume
Was some accumulated precision error happening.
2016-06-06 15:56:22 +02:00
Sergey Sharybin
c276480b0f Fix compilation error on 32 bit Windows 2016-06-06 14:01:49 +02:00
Sergey Sharybin
b277ba5c0d Cycles: Fix compilation error on OSX 2016-06-06 13:52:57 +02:00
Sergey Sharybin
6046c03f5c Cycles: Ignore zero size instances in BVH
In certain types of animation it's possible to have some objects
scaling to zero. In this case we can save render times by avoid
traversing such instances.

Better to do ti ahead of a time, so traversal stays simple.

Reviewers: lukasstockner97, dingto, brecht

Reviewed By: brecht

Differential Revision: https://developer.blender.org/D2048
2016-06-06 09:23:53 +02:00
Sergey Sharybin
b62faa54de Cycles: Add support of processor groups
Currently for windows only, this is an initial commit towards native
support of NUMA.

Current commit makes it so Cycles will use all logical processors on
Windows running on system with more than 64 threads.

Reviewers: juicyfruit, dingto, lukasstockner97, maiself, brecht

Subscribers: LazyDodo

Differential Revision: https://developer.blender.org/D2049
2016-06-06 09:14:37 +02:00
04bcaf07dc Fix T48580: path / branched path UI grayed out with OpenCL device that is not used. 2016-06-04 23:21:44 +02:00
Thomas Dinges
234801c8dc Fix T48587, constant fold should only be done for Mix Type. 2016-06-04 21:39:07 +02:00
Martijn Berger
50f432b1e0 CMake, minor changes to make Visual studio 2015 use a compatible numpy and
the standard cmake CUDA/NVCC arguments flag allowing 2015 build to use
msvc 2013 for cuda
2016-06-04 11:42:48 +02:00
Sergey Sharybin
b406b7be00 Cycles: Mark which CUDA device is used for display
It is really handy to know which one is display when having two cards of
same type in the machine.
2016-06-03 11:52:08 +02:00
Sergey Sharybin
f54a98a1c5 Cycles: Simplify check for degenerated faces on GPU
Still not sure how to properly solve the issue, needs some trickery to get
actual optimized values from intersection function (using printf() avoids
some optimization and makes stuff render correct).

For the time being let's just simplify check.
2016-06-03 10:36:04 +02:00
Bastien Montagne
24712b1c0b Usual UI/i18n message cleanup (get rid of last remaining 'addon' too). 2016-06-01 20:38:30 +02:00
Sergey Sharybin
d2bb0e660b Fix T46207: Slow OpenCL GPU bake and blown out baking Cycles render 2016-05-31 17:48:42 +02:00
Sergey Sharybin
d5220d23f9 Cycles: Fixes for recent refactor
- add_vertex() can be called from split_vertex() which does not guarantee
  to have properly pre-allocate arrays.

- Need to check whether Cycles is compiled with OSL in XML reader.
2016-05-31 15:32:31 +02:00
Mai Lavelle
4388b29e98 Cycles: Add human readable sizes to debug output
Some of these values can get quite large and are hard to read, adding this
makes it easy to read them at a glance.

Reviewed By: sergey

Differential Revision: https://developer.blender.org/D2039
2016-05-31 06:13:54 -04:00
Sergey Sharybin
a2aa44370b Fix T48556: Missing transparent shadows on AMD OpenCL
We had transparent shadows disabled for some time because they were causing
drivers to crash. Can't reproduce that issue anymore with current drivers,
so will enable them and see how it goes.
2016-05-31 11:48:07 +02:00
0d4a7d50c6 Fix broken Cycles curve motion radius after recent refactor, and fix warnings. 2016-05-31 00:21:33 +02:00
Sergey Sharybin
673fabbb64 Depsgraph: Fix wrong layers flush form children to parent
It was possible to have issues in cases when several child dependencies
goes to IDs with different layers. In this case order of flushing was not
really well defined, which could lead to cases when indirect dependency
via invisible object wouldn't work.

Need some sort of barrier to prevent scheduling of parent nodes for until
all children are done, but that's becoming quite nasty thing to implement.

Added a temp field to component for now. maybe it's not so crazy actually
and we might use it for evaluation as well, so we wouldn't flush updates
to components which does not affect visible stuff.
2016-05-30 12:35:03 +02:00
11b0874db0 Code refactor: store ShaderNode enums as enum rather than ustring. 2016-05-29 20:30:16 +02:00
a70a435f28 Code refactor: centralize OSL node creation in shader manager. 2016-05-29 20:30:16 +02:00
af073e149b Code refactor: pass ShaderInput to constant_fold, so it supports arbitrary types. 2016-05-29 20:30:16 +02:00
9bd2820aaf Code refactor: add separate RGB to BW node and rename some sockets. 2016-05-29 20:30:16 +02:00
eac7ed8d04 Code refactor: minor node and node type utility functions and changes. 2016-05-29 20:30:16 +02:00
7cd18dda7d Fix debug mode assert in subd code after recent refactoring. 2016-05-29 20:30:16 +02:00
001ba5bdf5 Code refactor: nodify object and mesh, but not used for XML yet.
Differential Revision: https://developer.blender.org/D2016
2016-05-28 18:40:09 +02:00
c96a4c8a2a Code refactor: modify mesh storage to use arrays rather than vectors, separate some arrays.
Differential Revision: https://developer.blender.org/D2016
2016-05-28 18:31:00 +02:00
b94bfe4cd8 Code refactor: make ShaderNode match Node a bit more, reusing types and enums.
Differential Revision: https://developer.blender.org/D2016
2016-05-28 15:49:15 +02:00
98ad473324 Code refactor: nodify Cycles camera and fix some mistakes in XML node read.
Differential Revision: https://developer.blender.org/D2016
2016-05-28 14:07:19 +02:00
62f85add48 Fix T48526: Cycles viewport render incorrectly influenced by pixel aspect ratio.
This reverts 72592cfb, needs more refactoring to do that cleanup.
2016-05-28 11:57:27 +02:00
Thomas Dinges
dc07a5561f Cleanup: Further tweaks for consistency and simplifications.
Now I can start adding half float without adding even bigger mess to all these functions. ;)
2016-05-27 23:35:29 +02:00
Thomas Dinges
2f457b7649 Cleanup: Simplify and de-duplicate assignements in the ImageManager constructor. 2016-05-27 23:12:13 +02:00
Thomas Dinges
2ee063868d Cleanup: Shorten texture variables, tex and image was kinda redundant.
Also make prefix consistent, so it starts with either TEX_NUM or TEX_START, followed by texture type and architecture.
2016-05-27 22:58:33 +02:00
Sergey Sharybin
7dae62cde0 Cycles: Simplify code around debug stats in BVH traversing 2016-05-27 10:55:48 +02:00
f2ba13964d Fix T48514: Cycles toon glossy BSDF not respecting reflective caustics option. 2016-05-25 21:13:24 +02:00
b49185df99 Cycles CUDA: reduce branched path stack memory by sharing indirect ShaderData.
Saves about 15% for the branched path kernel.
2016-05-25 21:13:24 +02:00
Sergey Sharybin
84c7f427f2 Cycles: Yet another fix for textures limit 2016-05-25 12:49:02 +02:00
Sergey Sharybin
18b418e8e0 Cycles: Fix wrong cuda textures limits 2016-05-24 17:02:19 +02:00
Sergey Sharybin
dc5a6a0aed Cycles: Fix wrong render results when using multiple OpenCL devices 2016-05-24 16:57:57 +02:00
Sergey Sharybin
42b26206c6 Fix T48508: Cycles Regression / Crash 2016-05-24 14:53:34 +02:00
Sergey Sharybin
e1dd83b399 Fix T48486: Not enough precision in Volume Sampling Step Size input
Volumes might be small, makes sense to use more precision here.
2016-05-24 12:17:06 +02:00
999d5a6785 Cycles CUDA: reduce stack memory by reusing ShaderData.
57% less for path and 48% less for branched path.
2016-05-23 22:29:24 +02:00