Age | Commit message (Collapse) | Author |
|
With upcoming light group passes, for them to sum up correctly to the combined
pass the clamping must be more fine grained.
This also has the advantage that if one light is particularly noisy, it does
not diminish the contribution from other lights which do not need as much
clamping.
Clamp values on existing scenes will need to be tweaked to get similar results,
there is no automatic conversion possible which would give the same results as
before.
Implemented by Lukas, with tweaks by Brecht.
Part of D4837
|
|
This will be used by Optix to help the compiler figure out scoping. It is not
used by other devices currently, but worth testing if it helps there too.
Ref D5363
|
|
Use a single loop to iterate over all lights, reducing divergence and amount
of code to generate. Moving ray intersection calls out of conditionals will
also help the Optix compiler.
Ref D5363
|
|
This makes little difference for CUDA and OpenCL, but will be helpful
for Optix.
|
|
|
|
Apply clang format as proposed in T53211.
For details on usage and instructions for migrating branches
without conflicts, see:
https://wiki.blender.org/wiki/Tools/ClangFormat
|
|
various parts of the CPU kernel
This commit adds a sample-based profiler that runs during CPU rendering and collects statistics on time spent in different parts of the kernel (ray intersection, shader evaluation etc.) as well as time spent per material and object.
The results are currently not exposed in the user interface or per Python yet, to see the stats on the console pass the "--cycles-print-stats" argument to Cycles (e.g. "./blender -- --cycles-print-stats").
Unfortunately, there is no clear way to extend this functionality to CUDA or OpenCL, so it is CPU-only for now.
Reviewers: brecht, sergey, swerner
Reviewed By: brecht, swerner
Differential Revision: https://developer.blender.org/D3892
|
|
|
|
Increasing the samplig dimensions like this is not optimal, I'm looking
into some deeper changes to reuse the random number and change the RR
probabilities, but this should fix the bug for now.
|
|
This is more important now that we will have tigther volume bounds that
we hit multiple times. It also avoids some noise due to RR previously
affecting these surfaces, which shouldn't have been the case and should
eventually be fixed for transparent BSDFs as well.
For non-volume scenes I found no performance impact on NVIDIA or AMD.
For volume scenes the noise decrease and fixed artifacts are worth the
little extra render time, when there is any.
|
|
This is done by storing only a subset of PathRadiance, and by storing
direct light immediately in the main PathRadiance. Saves about 10% of
CUDA stack memory, and simplifies subsurface indirect ray code.
|
|
Previously we used a 1D sequence to select a light, and another 2D sequence
to sample a point on the light. For multiple lights this meant each light
would get a random subset of a 2D stratified sequence, which is not
guaranteed to be stratified anymore.
Now we use only a 2D sequence, split into segments along the X axis, one for
each light. The samples that fall within a segment then each are a stratified
sequence, at least in the limit. So for example for two lights, we split up
the unit square into two segments [0,0.5[ x [0,1[ and [0.5,1[ x [0,1[.
This doesn't make much difference in most scenes, mainly helps if you have a
few large area lights or some types of HDR backgrounds.
|
|
Need to exit the volume stack when shadow ray laves the medium.
Thanks Brecht for review and help in troubleshooting!
|
|
Also pass by value and don't write back now that it is just a hash for seeding
and no longer an LCG state. Together this makes CUDA a tiny bit faster in my
tests, but mainly simplifies code.
|
|
OpenCL baking with SSS and Volume are not supported.
|
|
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!
|
|
This implements branched path tracing for the split kernel.
General approach is to store the ray state at a branch point, trace the
branched ray as normal, then restore the state as necessary before iterating
to the next part of the path. A state machine is used to advance the indirect
loop state, which avoids the need to add any new kernels. Each iteration the
state machine recreates as much state as possible from the stored ray to keep
overall storage down.
Its kind of hard to keep all the different integration loops in sync, so this
needs lots of testing to make sure everything is working correctly. We should
probably start trying to deduplicate the integration loops more now.
Nonbranched BMW is ~2% slower, while classroom is ~2% faster, other scenes
could use more testing still.
Reviewers: sergey, nirved
Reviewed By: nirved
Subscribers: Blendify, bliblubli
Differential Revision: https://developer.blender.org/D2611
|
|
The title says it all actually.
|
|
Simplifies code quite a bit, making it shorter and easier to extend.
Currently no functional changes for users, but is required for the
upcoming work of shadow catcher support with OpenCL.
|
|
It uses an idea of accumulating all possible light reachable across the
light path (without taking shadow blocked into account) and accumulating
total shaded light across the path. Dividing second figure by first one
seems to be giving good estimate of the shadow.
In fact, to my knowledge, it's something really similar to what is
happening in the denoising branch, so we are aligned here which is good.
The workflow is following:
- Create an object which matches real-life object on which shadow is
to be catched.
- Create approximate similar material on that object.
This is needed to make indirect light properly affecting CG objects
in the scene.
- Mark object as Shadow Catcher in the Object properties.
Ideally, after doing that it will be possible to render the image and
simply alpha-over it on top of real footage.
|
|
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.
|
|
|
|
their expected contribution
In scenes with many lights, some of them might have a very small contribution to some pixels, but the shadow rays are traced anyways.
To avoid that, this patch adds probabilistic termination to light samples - if the contribution before checking for shadowing is below a user-defined threshold, the sample will be discarded with probability (1 - (contribution / threshold)) and otherwise kept, but weighted more to remain unbiased.
This is the same approach that's also used in path termination based on length.
Note that the rendering remains unbiased with this option, it just adds a bit of noise - but if the setting is used moderately, the speedup gained easily outweighs the additional noise.
Reviewers: #cycles
Subscribers: sergey, brecht
Differential Revision: https://developer.blender.org/D2217
|
|
Path Tracing
The light sampling functions calculate light sampling PDF for the case that the light has been randomly selected out of all lights.
However, since BPT handles lamps and meshlights separately, this isn't the case. So, to avoid a wrong result, the code just included the 0.5 factor in the throughput.
In theory, however, the correction should be made to the sampling probability, which needs to be doubled. Now, for the regular calculation, that's no real difference since the throughput is divided by the pdf.
However, it does matter for the MIS calculation - it's unbiased both ways, but including the factor in the PDF instead of the throughput should give slightly better results.
Reviewers: sergey, brecht, dingto, juicyfruit
Differential Revision: https://developer.blender.org/D2258
|
|
Both spot and area light have large areas where they're not visible.
Therefore, this patch stops the light sampling code when one of these cases (outside of the spotlight cone or behind the area light) occurs, before the lamp shader is evaluated.
In the case of the area light, the solid angle sampling can also be skipped.
In a test scene with Sample All Lights and 18 Area lamps and 9 Spot lamps that all point away from the area that the camera sees, render time drops from 12sec to 5sec.
Reviewers: brecht, sergey, dingto, juicyfruit
Differential Revision: https://developer.blender.org/D2216
|
|
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).
|
|
57% less for path and 48% less for branched path.
|
|
Quite straightforward, main trick is happening in path_source_replace_includes().
Reviewers: brecht, dingto, lukasstockner97, juicyfruit
Differential Revision: https://developer.blender.org/D1794
|
|
|
|
Seems to be some compiler fault which leads to a wrong flag being used,
making it so wrong number of samples is used for the background.
This should in theory fix issue reported in T47213.
|
|
|
|
This commit changes the way how we pass bounce information to the Light
Path node. Instead of manualy copying the bounces into ShaderData, we now
directly pass PathState. This reduces the arguments that we need to pass
around and also makes it easier to extend the feature.
This commit also exposes the Transmission Bounce Depth to the Light Path
node. It works similar to the Transparent Depth Output: Replace a
Transmission lightpath after X bounces with another shader, e.g a Diffuse
one. This can be used to avoid black surfaces, due to low amount of max
bounces.
Reviewed by Sergey and Brecht, thanks for some hlp with this.
I tested compilation and usage on CPU (SVM and OSL), CUDA, OpenCL Split
and Mega kernel. Hopefully this covers all devices. :)
|
|
Ray direction is assumed to be normalized in such areas as scaling intersection
distance on instance push/pop when doing ray-scene intersection, but it was
possible that some closures wouldn't give normalized direction which could cause
wrong intersection checks.
Now normalization will happen on surface bounce, which could be a bit of a waste
if closure actually gives normalized direction, but currently only transparent
BSDF seems to give guaranteed normalized direction.
|
|
This commit contains all the work related on the AMD megakernel split work
which was mainly done by Varun Sundar, George Kyriazis and Lenny Wang, plus
some help from Sergey Sharybin, Martijn Berger, Thomas Dinges and likely
someone else which we're forgetting to mention.
Currently only AMD cards are enabled for the new split kernel, but it is
possible to force split opencl kernel to be used by setting the following
environment variable: CYCLES_OPENCL_SPLIT_KERNEL_TEST=1.
Not all the features are supported yet, and that being said no motion blur,
camera blur, SSS and volumetrics for now. Also transparent shadows are
disabled on AMD device because of some compiler bug.
This kernel is also only implements regular path tracing and supporting
branched one will take a bit. Branched path tracing is exposed to the
interface still, which is a bit misleading and will be hidden there soon.
More feature will be enabled once they're ported to the split kernel and
tested.
Neither regular CPU nor CUDA has any difference, they're generating the
same exact code, which means no regressions/improvements there.
Based on the research paper:
https://research.nvidia.com/sites/default/files/publications/laine2013hpg_paper.pdf
Here's the documentation:
https://docs.google.com/document/d/1LuXW-CV-sVJkQaEGZlMJ86jZ8FmoPfecaMdR-oiWbUY/edit
Design discussion of the patch:
https://developer.blender.org/T44197
Differential Revision: https://developer.blender.org/D1200
|
|
Ray length adjustment got lost in some refactor commit back to 2.71 days.
|
|
This was already mixed a bit, but the dot belongs there.
|
|
With this setting, we can limit the influence of a lamp to a certain amount of bounces.
0 = Only direct light contribution
1 = 1 light bounce
...
Differential revision: https://developer.blender.org/D860
You can find an example render in the release logs: http://wiki.blender.org/index.php/Dev:Ref/Release_Notes/2.73/Cycles
|
|
|
|
Differential Revision: https://developer.blender.org/D639
|
|
This has been fixed before, but somehow got reverted in d644753319b6.
|
|
* Volume multiple importace sampling support to combine equiangular and distance
sampling, for both homogeneous and heterogeneous volumes.
* Branched path "Sample All Direct Lights" and "Sample All Indirect Lights" now
apply to volumes as well as surfaces.
Implementation note:
For simplicity this is all done with decoupled ray marching, the only case we do
not use decoupled is for distance only sampling with one light sample. The
homogeneous case should still compile on the GPU because it only requires fixed
size storage, but the heterogeneous case will be trickier to get working.
|
|
|