This commit contains the first part of the new Cycles denoising option,
which filters the resulting image using information gathered during rendering
to get rid of noise while preserving visual features as well as possible.
To use the option, enable it in the render layer options. The default settings
fit a wide range of scenes, but the user can tweak individual settings to
control the tradeoff between a noise-free image, image details, and calculation
time.
Note that the denoiser may still change in the future and that some features
are not implemented yet. The most important missing feature is animation
denoising, which uses information from multiple frames at once to produce a
flicker-free and smoother result. These features will be added in the future.
Finally, thanks to all the people who supported this project:
- Google (through the GSoC) and Theory Studios for sponsoring the development
- The authors of the papers I used for implementing the denoiser (more details
on them will be included in the technical docs)
- The other Cycles devs for feedback on the code, especially Sergey for
mentoring the GSoC project and Brecht for the code review!
- And of course the users who helped with testing, reported bugs and things
that could and/or should work better!
The idea is to make include statements more explicit and obvious where the
file is coming from, additionally reducing chance of wrong header being
picked up.
For example, it was not obvious whether bvh.h was refferring to builder
or traversal, whenter node.h is a generic graph node or a shader node
and cases like that.
Surely this might look obvious for the active developers, but after some
time of not touching the code it becomes less obvious where file is coming
from.
This was briefly mentioned in T50824 and seems @brecht is fine with such
explicitness, but need to agree with all active developers before committing
this.
Please note that this patch is lacking changes related on GPU/OpenCL
support. This will be solved if/when we all agree this is a good idea to move
forward.
Reviewers: brecht, lukasstockner97, maiself, nirved, dingto, juicyfruit, swerner
Reviewed By: lukasstockner97, maiself, nirved, dingto
Subscribers: brecht
Differential Revision: https://developer.blender.org/D2586
The intention of this commit it to address issues mentioned in the
reports T43865,T50164 and T50452.
The code is based on Embree code with some extra vectorization
to speed up single ray to single triangle intersection.
Unfortunately, such a fix is not coming for free. There is some
slowdown for AVX2 processors, mainly due to different vectorization
code, which caused different number of instructions to be executed
and different instructions-per-cycle counters. But on another hand
this commit makes pre-AVX2 platforms such as AVX and SSE4.1 a bit
faster. The prerformance goes as following:
2.78c AVX2 2.78c AVX Patch AVX2 Patch AVX
BMW 05:21.09 06:05.34 05:32.97 (+3.5%) 05:34.97 (-8.5%)
Classroom 16:55.36 18:24.51 17:10.41 (+1.4%) 17:15.87 (-6.3%)
Fishy Cat 08:08.49 08:36.26 08:09.19 (+0.2%) 08:12.25 (-4.7%
Koro 11:22.54 11:45.24 11:13.25 (-1.5%) 11:43.81 (-0.3%)
Barcelone 14:18.32 16:09.46 14:15.20 (-0.4%) 14:25.15 (-10.8%)
On GPU the performance is about 1.5-2% slower in my tests on GTX1080
but afraid we can't do much as a part of this chaneg here and
consider it a price to pay for more proper intersection check.
Made in collaboration with Maxym Dmytrychenko, big thanks to him!
Reviewers: brecht, juicyfruit, lukasstockner97, dingto
Differential Revision: https://developer.blender.org/D1574
The title says it all actually. Gives up to 10% speedup on test scenes here
on i7-6800K.
Render times on GPU are unreliable here, but there might be some slowdown
caused by watertight nature of intersections.
Avoid construction of temporary array and make utility function force-inlined.
Additionally avoid calling float4_to_float3 twice.
This brings render times to the same values as before current patch series.
This is a preparation work for the followup commit which wil l move
remaining parts of Woop intersection logic to an utility file.
Doing it as a separate commit to keep changes more atomic and easier
to bisect when/if needed.
There seems to be a compiler bug of MSVC2013. The issue does not happen on Linux and
does not happen on Windows when building with MSVC2015.
Since it's reallly a pain to debug release builds with MSVC2013 the AVX2 optimization
is disabled for curve sergemnts for this compiler.
Decoupled ray marching is not supported yet.
Transparent shadows are always enabled for volume rendering.
Changes in kernel/bvh and kernel/geom are from Sergey.
This simiplifies code significantly, and prepares it for
record-all transparent shadow function in split kernel.
The issue here was mainly coming from minimal pixel width feature
which is quite commonly enabled in production shots.
This feature will use some probabilistic heuristic in the curve
intersection function to check whether we need to return intersection
or not. This probability is calculated for every intersection check.
Now, when we use multiple BVH nodes for curve primitives we increase
probability of that primitive to be considered a good intersection
for us. This is similar to increasing minimal width of curve.
What is worst here is that change in the intersection probability
fully depends on exact layout of BVH, meaning probability might
change differently depending on a view angle, the way how builder
binned the primitives and such. This makes it impossible to do
simple check like dividing probability by number of BVH steps.
Other solution might have been to split BVH into fully independent
trees, but that will increase memory usage of all the static
objects in the scenes, which is also not something desirable.
For now used most simple but robust approach: store BVH primitives
time and test it in curve intersection functions. This solves the
regression, but has two downsides:
- Uses more memory.
which isn't surprising, and ANY solution to this problem will
use more memory.
What we still have to do is to avoid this memory increase for
cases when we don't use BVH motion steps.
- Reduces number of maximum available textures on pre-kepler cards.
There is not much we can do here, hardware gets old but we need
to move forward on more modern hardware..
We started to run out of bits there, so now we separate flags
which came from __object_flags and which are either runtime or
coming from __shader_flags.
Rule now is: SD_OBJECT_* flags are to be tested against new
object_flags field of ShaderData, all the rest flags are to
be tested against flags field of ShaderData.
There should be no user-visible changes, and time difference
should be minimal. In fact, from tests here can only see hardly
measurable difference and sometimes the new code is somewhat
faster (all within a noise floor, so hard to tell for sure).
Reviewers: brecht, dingto, juicyfruit, lukasstockner97, maiself
Differential Revision: https://developer.blender.org/D2428
Mimics how regular triangles are working and makes it more clear where
the stuff is located in the kernel.
Needed to have some forward declarations because of the current placement
of things in the kernel.
Basically, the problem here was that the transform that's used to bring texture coordinates
to world space is either fetched while setting up the shader (with Object Motion is enabled) or
fetched when needed (otherwise). That helps to save ShaderData memory on OpenCL when Object Motion isn't needed.
Now, if OM is enabled, the Lamp transform can just be stored inside the ShaderData as well. The original commit just assumed it is.
However, when it's not (on OpenCL by default, for example), there is no easy way to fetch it when needed, since the ShaderData doesn't
store the Lamp index.
So, for now the lamps just don't support local texture coordinates anymore when Object Motion is disabled.
To fix and support this properly, one of the following could be done:
- Just always pre-fetch the transform. Downside: Memory Usage increases when not using OM on OpenCL
- Add a variable to ShaderData that stores the Lamp ID to allow fetching it when needed
- Store the Lamp ID inside prim or object. Problem: Cycles currently checks these for whether an object was hit - these checks would need to be changed.
- Enable OM whenever a Texture Coordinate's Normal output is used. Downside: Might not actually be needed.
When using the Normal output of the Texture Coordinate node on Point and Spot lamps, the coordinates now depend on the rotation of the lamp.
On Area lamps, the Parametric output of the Geometry node now returns UV coordinates on the area lamp.
Credit for the Area lamp part goes to Stefan Werner (from D1995).
Basically don't use rcp() in areas which seems to be critical after
second look. Also disabled some multiplication operators, not sure
yet why they might be a problem.
Tomorrow will be setting up a full test with all cases which were
buggy in our farm to see if this fix is complete.
Several ideas here:
- Optimize calculation of near_{x,y,z} in a way that does not require
3 if() statements per update, which avoids negative effect of wrong
branch prediction.
- Optimization of direction clamping for BVH.
- Optimization of point/direction transform.
Brings ~1.5% speedup again depending on a scene (unfortunately, this
speedup can't be sum across all previous commits because speedup of
each of the changes varies from scene to scene, but it still seems to
be nice solid speedup of few percent on Linux and bigger speedup was
reported on Windows).
Once again ,thanks Maxym for inspiration!
Still TODO: We have multiple places where we need to calculate near
x,y,z indices in BVH, for now it's only done for main BVH traversal.
Will try to move this calculation to an utility function and see if
that can be easily re-used across all the BVH flavors.
Similar to the previous commit, avoid negative effect of bad branch prediction.
Gives measurable performance up to ~2% in tests here.
Once again, thanks to Maxym Dmytrychenko!
Similar to regular triangle intersection case. Gives about 3% speedup rendering
SSS object on my desktop,
Question: how to avoid such a code duplication in a nice way without speed loss?
Note that volume rendering is not supported yet, this is a step towards that.
Reviewed By: brecht
Differential Revision: https://developer.blender.org/D2299
This commit basically vectorizes existing code using AVX2 instructions
(without modifying algorithm itself). This gives quite nice speedups:
BMW: -8%
Classroom: -5%
Cat: -5%
Koro: +1%
Barcelona: -8%
That's on Linux machine, reported performance improvement on Windows
goes up to 20%.
Not currently sure why Koro is somewhat slower because it mainly uses
curve intersection tests, could be a time noise? Or osmething with the
cache utilization perhaps? In any case speedup in other scenes makes
me thinking that current state is acceptable for initial implementation.
This is again inspired by Maxym Dmytrychenko.
Mostly this is making inlining match CUDA 7.5 in a few performance critical
places. The end result is that performance is now better than before, possibly
due to less register spilling or other CUDA 8.0 compiler improvements.
On benchmarks scenes, there are 3% to 35% render time reductions. Stack memory
usage is reduced a little too.
Reviewed By: sergey
Differential Revision: https://developer.blender.org/D2269
Some of the files were wrongly attributing code to some other
organizations and in few places proper attribution was missing.
This is mainly either a copy-paste error (when new file was
created from an existing one and header wasn't updated) or due
to some refactor which split non-original-BF code with purely
BF code.
Should solve some confusion around.
Using ones complement for detecting if transform has been applied was confusing
and led to several bugs. With this proper checks are made.
Also added a few transforms where they were missing, mostly affecting baking
and displacement when `P` is used in the shader (previously `P` was in the
wrong space for these shaders)
Also removed `TIME_INVALID` as this may have resulted in incorrect
transforms in some cases.
Reviewed By: brecht
Differential Revision: https://developer.blender.org/D2192
Kernels can now be built without patch evaluation when not needed by the
scene (Catmull-Clark subdivision not in use), giving a performance boost
for some devices.
Enables Catmull-Clark subdivision meshes with support for creases and attribute
subdivision. Still waiting on OpenSubdiv to fully support face varying
interpolation for subdividing uv coordinates tho. Also there may be some
inconsistencies with Blender's subdivision which will be resolved at a
later time.
Code for reading patch tables and creating patch maps is borrowed
from OpenSubdiv.
Reviewed By: brecht
Differential Revision: https://developer.blender.org/D2111
Adds a descriptor for attributes that can easily be passed around and extended
to contain more data. Will be used for attributes on subdivision meshes.
Reviewed By: brecht
Differential Revision: https://developer.blender.org/D2110
