| Age | Commit message (Collapse) | Author |
|---|
|
Instead of treating Fermi GPU limits as default,
and overriding them for other devices,
we now nicely set them for each platform.
* Due to setting values for all platforms,
we don't have to offset the slot id for OpenCL anymore,
as the image manager wont add float images for OpenCL now.
* Bugfix: TEX_NUM_FLOAT_IMAGES was always 5, even for CPU,
so the code in svm_image.h clamped float textures with alpha on CPU after the 5th slot.
Reviewers: #cycles, brecht
Reviewed By: #cycles, brecht
Subscribers: brecht
Differential Revision: https://developer.blender.org/D1925
|
|
Seems particular CUDA implementations has some precision issues,
which made integer coordinate (which was expected to always be
positive) to go negative.
|
|
|
|
This reverts commit d91316dc672dc1ee69fbd24d2f00124a24b75c6b.
|
|
The improved Hosek / Wilkie model was added during my GSoC 2013 and the default since then.
The older model was kinda kept for compatibility, but after more than 2 years it's time to remove it.
The Hosek / Wilkie model is more realistic anyway, and people who really want a day / night transition can mix the Sky Shader with another one (e.g. color) and fade between the two.
|
|
Work around what appears to be a compiler bug.
|
|
Quite straightforward, main trick is happening in path_source_replace_includes().
Reviewers: brecht, dingto, lukasstockner97, juicyfruit
Differential Revision: https://developer.blender.org/D1794
|
|
The issue here was actually somewhere else - the attached scene from the report used a light falloff node in a sunlamp (aka distant light).
However, since distant lamps set the ray length to FLT_MAX and the light falloff node squares this value, it overflows and produces a NaN
weight, which propagates and leads to a NaN intensity, which is then clamped to zero and produces the black pixels.
To fix that issue, the smoothing part of the light falloff is just ignored if the smoothing term isn't finite (which makes sense since
the term should converge to 1 as the distance increases).
The reason for the different results on CPUs and GPUs is not perfectly clear, but probably can be explained with different handling of
Inf/NaN edge cases.
Also, to notice issues like these faster in the future, kernel_asserts were added that evaluate as false as soon as a non-finite intensity is produced.
|
|
|
|
Supports both smoke/fire and point density textures now.
Reduces number of textures available for sm_20 and sm_21, but you have
to compromise somewhere on such a limited hardware.
Currently limited to linear interpolation only, and decoupled ray
marching is not supported yet. Think those could be considered just a
further improvement.
Some quick example:
https://developer.blender.org/F282934
Code is minimal and we can fully consider it a fix for missing
support of 3D textures with CUDA.
Reviewers: lukasstockner97, brecht, juicyfruit, dingto
Reviewed By: brecht, juicyfruit, dingto
Subscribers: mib2berlin
Differential Revision: https://developer.blender.org/D1806
|
|
Using this paper:
http://graphics.pixar.com/library/ApproxBSSRDF/paper.pdf
This model gives less blurry results than the Cubic and Gaussian
we had implemented:
- Cubic: https://developer.blender.org/F279670
- Burley: https://developer.blender.org/F279671
The model is called "Christensen-Burley" in the interface, which
actually should be read as "Physically based" or "Realistic".
Reviewers: juicyfruit, dingto, lukasstockner97, brecht
Reviewed By: brecht, dingto
Subscribers: robocyte
Differential Revision: https://developer.blender.org/D1759
|
|
|
|
|
|
The goal is to make Experimental kernel closer in performance to the
official kernel, avoiding spills and such.
There should not be big impact on official kernel, own tests showed
few percent performance drop on laptop's GPU. CPU was always the
same speed on AVX, AVX2 and SSE4.1 CPUs i've been testing here.
This seems to be the last essential step before we can get rid of
Experimental kernel and enable SSS officially on GPU without causing
some major performance issues.
Surely some more tweaks are possibly required, but that we can do
for until cows go home anyway.
|
|
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. :)
|
|
This commit adds "Bands Saw" and "Rings Saw" to the options for the Wave texture node in Cycles, behaving similar to the Saw option in BI textures.
Requested by @cekuhnen on BA.
Reviewers: dingto, sergey
Subscribers: cekuhnen
Differential Revision: https://developer.blender.org/D1699
|
|
Vector mapping node was doing some weird mapping of both original and mapped
coordinates. Mapping of original coordinates was caused by the clamping nature
of the LUT generated from the node. Mapping of the mapped value again was quite
totally obscure -- one needed to constantly keep in mind that actual value will
be scaled up and moved down.
This commit makes it so values in the vector curve mapping are always absolute.
In fact, it is now behaving quite the same as RGB curve mapping node and the
code could be de-duplicated. Keeping the code duplicated for a bit so it's more
clear what exact parts of the node changed.
Reviewers: brecht
Subscribers: bassamk
Differential Revision: https://developer.blender.org/D1672
|
|
|
|
Previously RGB Curves node will clamp input to 0..1 which is rather useless
when one wants to use HDR image textures and do bit of correction on them.
Now kernel code supports extrapolation of baked LUT based on first/last two
table points and performs linear extrapolation.
The only tricky part is to guess the range to bake the LUT for. Currently
it's using simple approach -- minmax of the input curves. While this behaves
ok for the simple cases it's easy to trick the system up causing incorrect
results.
Not sure we can solve those issues in a general case and since the new code
is giving more expected results it's not that bad actually. In the worst
case artist migh always create explicit point to make sure LUT is created
for the needed HDR range.
Reviewers: brecht, juicyfruit
Subscribers: sebastian_k
Differential Revision: https://developer.blender.org/D1658
|
|
The issue was caused by non-continuous tangent space calculated for triangles.
This commit adds a Tangent input to Hair BSDF node which can be used to hook up
Tangent calculated form UV as an input to the node in order to make sure the
tangent space is continuous.
Doing this as an input instead of using default tangent layer from UV because of
several reasons:
- This way it's really easy to preserve compatibility with existing setups.
- Default UV map is not necessarily giving continuous space, one might want to
use other tangent space sources or distort the space for some artistic
decision.
Reviewers: juicyfruit, dingto
Reviewed By: dingto
Differential Revision: https://developer.blender.org/D1428
|
|
as volume input
|
|
It is now handled via texture extension type.
|
|
Added some extra seed to the hash, so it's now less likely to give repetitive
patters at values around zero.
This will change distribution of bricks for existing files. but it's something
inevitable.
|
|
This might confuse closure merger.
Spotted by Campbell Barton, thanks!
|
|
|
|
* Did not check data2, this partially fixes T45583.
* Initialize data2 in some closures to avoid potential problems.
Differential Revision: https://developer.blender.org/D1436
|
|
Basically this commit totally disables new SVN Voxel node, which solves some
of the compiler's issues.
|
|
|
|
The idea of this node is to sampling of 3D voxels at a given coordinate
supporting different mapping strategies (world space mapping, object
local space etc).
Currently not in use, it's a preparation step for supporting point density
textures.
|
|
Glass BSDF was doing some magic with copying weigths from initial closure
onto refraction one and the code was not checking properly for the number
of closures.
|
|
This will help figuring out cases when node was not properly handled by the SVM
by aborting execution on CPU, where all the nodes are expected to be supported.
|
|
This commits finishes initial selective nodes compilation into kernel, which
helps a lot performance-wise for AMD OpenCL kernels.
Split by node groups is based on statistics from simple scenes like BMW and
more complex scenes like mango and gooseberry production files. Further
tweaks are always possible, but it should be a good starting point.
TODO: Still need to ignore unused nodes when calculating requested shader
features.
|
|
|
|
For now it is unused in the kernel, actual usage will come with
the next commits.
|
|
Same as last commit, code is unused and this one actually would have required some fixes,
as these variants output values outside the 0-1 value range, which doesn't fit Cycles shader design.
|
|
Let's finally delete this code, after 4 years of being unused,
there really is no excuse anymore.
If we decide to extend the procedural textures in SVM, we can do this anytime in the future.
|
|
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
|
|
The goal is to be able to compile kernel with nodes which are actually needed
to render current scene, hence improving performance of the kernel,
The idea is:
- Have few node groups, starting with a group which contains nodes are used
really often, and then couple of groups which will be extension of this one.
- Have feature-based nodes disabling, so it's possible to disable nodes related
to features which are not used with the currently used nodes group.
This commit only lays down needed routines for this approach, actual split will
happen later after gathering statistics from bunch of production scenes.
|
|
Currently unused but will be needed soon for the split kernel work.
|
|
|
|
Now we calculate color in range 800..12000 using an approximation a/x+bx+c for R and G and ((at + b)t + c)t + d) for B.
Max absolute error for RGB for non-lut function is less than 0.0001, which is enough to get the same 8 bit/channel color as for OSL with a noticeable performance difference.
However there is a slight visible difference between previous non-OSL implementation because of lookup table interpolation and offset-by-one mistake.
The previous implementation gave black color outside of soft range (t > 12000), now it gives the same color as for 12000.
Also blackbody node without input connected is being converted to value input at shader compile time.
Reviewers: dingto, sergey
Reviewed By: dingto
Subscribers: nutel, brecht, juicyfruit
Differential Revision: https://developer.blender.org/D1280
|
|
This more a workaround for CUDA optimizer which can't optimize clamp(x, 0, 1)
into a single instruction and uses 4 instructions instead.
Original patch by @lockal with own modification:
Don't make changes outside of the kernel. They don't make any difference
anyway and term saturate() has a bit different meaning outside of kernel.
This gives around 2% of speedup in Barcelona file, but in more complex shader
setups with lots of math nodes with clamping speedup could be much nicer.
Subscribers: dingto
Projects: #cycles
Differential Revision: https://developer.blender.org/D1224
|
|
This way we can get rid of inefficient memory usage caused by BVH boundbox
part being unused by leaf nodes but still being allocated for them. Doing
such split allows to save 6 of float4 values for QBVH per leaf node and 3
of float4 values for regular BVH per leaf node.
This translates into following memory save using 01.01.01.G rendered
without hair:
Device memory size Device memory peak Global memory peak
Before the patch: 4957 5051 7668
With the patch: 4467 4562 7332
The measurements are done against current master. Still need to run speed tests
and it's hard to predict if it's faster or not: on the one hand leaf nodes are
now much more coherent in cache, on the other hand they're not so much coherent
with regular nodes anymore.
Reviewers: brecht, juicyfruit
Subscribers: venomgfx, eyecandy
Differential Revision: https://developer.blender.org/D1236
|
|
|
|
|
|
This inconsistency drove me totally crazy, it's really confusing
when it's inconsistent especially when you work on both Cycles and
Blender sides.
Shouldn;t cause merge PITA, it's whitespace changes only, Git should
be able to merge it nicely.
|
|
Issue was introduced in 01ee21f where i didn't notice *_setup()
function only doing partial initialization, and some of parameters
are expected to be initialized by callee function.
This was hitting only some setups, so tests with benchmark scenes
didn't unleash issues. Now it should all be fine.
This is to go to the 2.74 branch and we actually might re-AHOY.
|
|
|
|
This attribute missed derivatives calculation.
Not totally sure what's the proper approach for algebraic derivative
calculation, so calculating them by definition. This isn't fastest
way to do it in this case and could be replaced with some smarter magic
in the wireframe calculation loop.
At least currently implemented approach is better than nothing.
|
|
OpenCL doesn't let you to get address of vector components, which
is kinda annoying. On the other hand, maybe now compiler will have
more chances to optimize something out.
|
