This patch adds two new kernels: SORT_BUCKET_PASS and SORT_WRITE_PASS. These replace PREFIX_SUM and SORTED_PATHS_ARRAY on supported devices (currently implemented on Metal, but will be trivial to enable on the other backends). The new kernels exploit sort partitioning (see D15331) by sorting each partition separately using local atomics. This can give an overall render speedup of 2-3% depending on architecture. As before, we fall back to the original non-partitioned sorting when the shader count is "too high".
Reviewed By: brecht
Differential Revision: https://developer.blender.org/D16909
This patch tunes the integrator state sizing for Metal (`num_concurrent_states` and `num_concurrent_busy_states`).
On all GPUs architecture, we adjust the busy:total states ratio to be 1:4 which gives better rendering performance than the previous 1:16 ratio (independent of total state count). This gives a small performance uplift (e.g. 2-3% on M1 Ultra).
Additionally for M2 architectures, we double the overall state size if there is available headroom. Inclusive of the first change, we can expect uplift of close to 10% in future, as this results in larger dispatch sizes and minimises work submission overheads. In order to make an accurate determination of available headroom, we defer the calculation of `num_concurrent_states` and `num_concurrent_busy_states` until the time of integrator state allocation (i.e. after all of the scene data has been allocated). We also refactor `alloc_integrator_soa` to calculate an *exact* single-state-size in a first pass, right before allocating the integrator SoA buffers in a second pass.
Reviewed By: brecht
Differential Revision: https://developer.blender.org/D16313
This was added for Metal, but also gives good results with CUDA and OptiX.
Also enable it for future Apple GPUs instead of only M1 and M2, since this has
been shown to help across multiple GPUs so the better bet seems to enable
rather than disable it.
Also moves some of the logic outside of the Metal device code, and always
enables the code in the kernel since other devices don't do dynamic compile.
Time per sample with OptiX + RTX A6000:
new old
barbershop_interior 0.0730s 0.0727s
bmw27 0.0047s 0.0053s
classroom 0.0428s 0.0464s
fishy_cat 0.0102s 0.0108s
junkshop 0.0366s 0.0395s
koro 0.0567s 0.0578s
monster 0.0206s 0.0223s
pabellon 0.0158s 0.0174s
sponza 0.0088s 0.0100s
spring 0.1267s 0.1280s
victor 0.0524s 0.0531s
wdas_cloud 0.0817s 0.0816s
Ref D15331, T87836
This patch partitions the active indices into chunks prior to sorting by material in order to tradeoff some material coherence for better locality. On Apple Silicon GPUs (particularly higher end M1-family GPUs), we observe overall render time speedups of up to 15%. The partitioning is implemented by repeating the range of `shader_sort_key` for each partition, and encoding a "locator" key which distributes the indices into sorted chunks.
Reviewed By: brecht
Differential Revision: https://developer.blender.org/D15331
* Rename "texture" to "data array". This has not used textures for a long time,
there are just global memory arrays now. (On old CUDA GPUs there was a cache
for textures but not global memory, so we used to put all data in textures.)
* For CUDA and HIP, put globals in KernelParams struct like other devices.
* Drop __ prefix for data array names, no possibility for naming conflict now that
these are in a struct.
Move MNEE to own kernel, separate from shader ray-tracing. This does introduce
the limitation that a shader can't use both MNEE and AO/bevel, but that seems
like the better trade-off for now.
We can experiment with bigger kernel organization changes later.
Differential Revision: https://developer.blender.org/D15070
* Replace license text in headers with SPDX identifiers.
* Remove specific license info from outdated readme.txt, instead leave details
to the source files.
* Add list of SPDX license identifiers used, and corresponding license texts.
* Update copyright dates while we're at it.
Ref D14069, T95597
The root of the issue is caused by Cycles ignoring OpenGL limitation on
the maximum resolution of textures: Cycles was allocating texture of the
final render resolution. It was exceeding limitation on certain GPUs and
driver.
The idea is simple: use multiple textures for the display, each of which
will fit into OpenGL limitations.
There is some code which allows the display driver to know when to start
the new tile. Also added some code to allow force graphics interop to be
re-created. The latter one ended up not used in the final version of the
patch, but it might be helpful for other drivers implementation.
The tile size is limited to 8K now as it is the safest size for textures
on many GPUs and OpenGL drivers.
This is an updated fix with a workaround for freezing with the NVIDIA
driver on Linux.
Differential Revision: https://developer.blender.org/D13385
This reverts commit 5e37f70307.
It is leading to freezing of the entire desktop for a few seconds when stopping
3D viewport rendering on my Linux / NVIDIA system.
The root of the issue is caused by Cycles ignoring OpenGL limitation on
the maximum resolution of textures: Cycles was allocating texture of the
final render resolution. It was exceeding limitation on certain GPUs and
driver.
The idea is simple: use multiple textures for the display, each of which
will fit into OpenGL limitations.
There is some code which allows the display driver to know when to start
the new tile. Also added some code to allow force graphics interop to be
re-created. The latter one ended up not used in the final version of the
patch, but it might be helpful for other drivers implementation.
The tile size is limited to 8K now as it is the safest size for textures
on many GPUs and OpenGL drivers.
Differential Revision: https://developer.blender.org/D13385
This patch adds new arg-type parameters to `DeviceQueue::enqueue` and its overrides. This is in preparation for the Metal backend which needs this information for correct argument encoding.
Ref T92212
Reviewed By: brecht
Maniphest Tasks: T92212
Differential Revision: https://developer.blender.org/D13357
With the current code in master, scrambling distance is enabled on non-hardware accelerated ray tracing devices see a measurable performance decrease when compared scrambling distance on vs off. From testing, this performance decrease comes from the large tile sizes scheduled in `tile.cpp`.
This patch attempts to address the performance decrease by using different algorithms to calculate the tile size for devices with hardware accelerated ray traversal and devices without. Large tile sizes for hardware accelerated devices and small tile sizes for others.
Most of this code is based on proposals from @brecht and @leesonw
Reviewed By: brecht, leesonw
Differential Revision: https://developer.blender.org/D13042
The issue was caused by splitting happening twice.
Fixed by checking for split flag which is assigned to the both states
during split.
The tricky part was to write catcher data at the moment of split: the
transparency and shadow catcher sample count is to be accumulated at
that point. Now it is happening in the `intersect_closest` kernel.
The downside is that render buffer is to be passed to the kernel, but
the benefit is that extra split bounce check is not needed now.
Had to move the passes write to shadow catcher header, since include
of `film/passes.h` causes all the fun of requirement to have BSDF
data structures available.
Differential Revision: https://developer.blender.org/D13177
This patch exposes the sampling offset option to Blender. It is located in the "Sampling > Advanced" panel.
For example, this can be useful to parallelize rendering and distribute different chunks of samples for each computer to render.
---
I also had to add this option to `RenderWork` and `RenderScheduler` classes so that the sample count in the status string can be calculated correctly.
Reviewed By: leesonw
Differential Revision: https://developer.blender.org/D13086
Instead of printing debug flags listing various CPU and GPU settings that
may or may not be used, print when we are using them. This include CPU
kernel types, OptiX debugging and CUDA and HIP adaptive compilation. BVH
type was already printed.
Cycles:Distance Scrambling for Cycles Sobol Sampler
This option implements micro jittering an is based on the INRIA
research paper [[ https://hal.inria.fr/hal-01325702/document | on micro jittering ]]
and work by Lukas Stockner for implementing the scrambling distance.
It works by controlling the correlation between pixels by either using
a user supplied value or an adaptive algorithm to limit the maximum
deviation of the sample values between pixels.
This is a follow up of https://developer.blender.org/D12316
The PMJ version can be found here: https://developer.blender.org/D12511
Reviewed By: leesonw
Differential Revision: https://developer.blender.org/D12318
Remove prefix of filenames that is the same as the folder name. This used
to help when #includes were using individual files, but now they are always
relative to the cycles root directory and so the prefixes are redundant.
For patches and branches, git merge and rebase should be able to detect the
renames and move over code to the right file.
* Split render/ into scene/ and session/. The scene/ folder now contains the
scene and its nodes. The session/ folder contains the render session and
associated data structures like drivers and render buffers.
* Move top level kernel headers into new folders kernel/camera/, kernel/film/,
kernel/light/, kernel/sample/, kernel/util/
* Move integrator related kernel headers into kernel/integrator/
* Move OSL shaders from kernel/shaders/ to kernel/osl/shaders/
For patches and branches, git merge and rebase should be able to detect the
renames and move over code to the right file.
Add a Fast GI Method, either Replace for the existing behavior, or Add
to add ambient occlusion like the old world settings.
This replaces the old Ambient Occlusion settings in the world properties.
Similar to main path compaction that happens before adding work tiles, this
compacts shadow paths before launching kernels that may add shadow paths.
Only do it when more than 50% of space is wasted.
It's not a clear win in all scenes, some are up to 1.5% slower. Likely caused
by different order of scheduling kernels having an unpredictable performance
impact. Still feels like compaction is just the right thing to avoid cases
where a few shadow paths can hold up a lot of main paths.
Differential Revision: https://developer.blender.org/D12944
Easy now thanks to the main and shadow path decoupling. Doesn't help
in an benchmark scene except Spring, where it reduces render time by
maybe 2-3%.
Ref T87836
Taking advantage of the new decoupled main and shadow paths. For CPU we
just store two nested structs in the integrator state, one for direct light
shadows and one for AO. For the GPU we restrict the number of shade surface
states to be executed based on available space in the shadow paths queue.
This also helps improve performance in benchmark scenes with an AO pass,
since it is no longer needed to use the shader raytracing kernel there,
which has worse performance.
Differential Revision: https://developer.blender.org/D12900
The motivation for this is twofold. It improves performance (5-10% on most
benchmark scenes), and will help to bring back transparency support for the
ambient occlusion pass.
* Duplicate some members from the main path state in the shadow path state.
* Add shadow paths incrementally to the array similar to what we do for
the shadow catchers.
* For the scheduling, allow running shade surface and shade volume kernels
as long as there is enough space in the shadow paths array. If not, execute
shadow kernels until it is empty.
* Add IntegratorShadowState and ConstIntegratorShadowState typedefs that
can be different between CPU and GPU. For GPU both main and shadow paths
juse have an integer for SoA access. Bt with CPU it's a different pointer
type so we get type safety checks in code shared between CPU and GPU.
* For CPU, add a separate IntegratorShadowStateCPU struct embedded in
IntegratorShadowState.
* Update various functions to take the shadow state, and make SVM take either
type of state using templates.
Differential Revision: https://developer.blender.org/D12889
The assumption about absent shadow path was wrong.
The rest of the changes are to ensure shadow paths are finished prior
to the split, so that they write to the proper passes.
The issue was caught by running regression tests on OptiX.
Differential Revision: https://developer.blender.org/D12857
Make volume stack allocated conditionally, potentially based on the
actual nested level of objects in the scene.
Currently the nested level is estimated by number of volume objects.
This is a non-expensive check which is probably enough in practice
to get almost perfect memory usage and performance.
The conditional allocation is a bit tricky.
For the CPU we declare and define maximum possible volume stack,
because there are only that many integrator states on the CPU.
On the GPU we declare outer SoA to have all volume stack elements,
but only allocate actually needed ones. The actually used volume
stack size is passed as a pre-processor, which seems to be easiest
and fastest for the GPU state copy.
There seems to be no speed regression in the demo files on RTX6000.
Note that scenes with high nested level of volume will now be slower
but correct.
Differential Revision: https://developer.blender.org/D12759
Implement an overscan support for tiles, so that adaptive sampling can
rely on the pixels neighbourhood.
Differential Revision: https://developer.blender.org/D12599
Currently was only used for logging, but better to fix the size so
that it matches reality.
The issue was caused by decoupling number of shadow intersections
and using much higher number for CPU. This caused the total state
on GPU to be logged as 10s of gigabytes instead of 100s of megabytes.
Differential Revision: https://developer.blender.org/D12755
* Split GPUDisplay into two classes. PathTraceDisplay to implement the Cycles side,
and DisplayDriver to implement the host application side. The DisplayDriver is now
a fully abstract base class, embedded in the PathTraceDisplay.
* Move copy_pixels_to_texture implementation out of the host side into the Cycles side,
since it can be implemented in terms of the texture buffer mapping.
* Move definition of DeviceGraphicsInteropDestination into display driver header, so
that we do not need to expose private device headers in the public API.
* Add more detailed comments about how the DisplayDriver should be implemented.
The "driver" terminology might not be obvious, but is also used in other renderers.
Differential Revision: https://developer.blender.org/D12626
Noticed while looking into flickering issues in viewport.
Doesn't seem to solve the flicker issue for me, but is something
what is supposed to be happening anyway.
Differential Revision: https://developer.blender.org/D12673
This includes much improved GPU rendering performance, viewport interactivity,
new shadow catcher, revamped sampling settings, subsurface scattering anisotropy,
new GPU volume sampling, improved PMJ sampling pattern, and more.
Some features have also been removed or changed, breaking backwards compatibility.
Including the removal of the OpenCL backend, for which alternatives are under
development.
Release notes and code docs:
https://wiki.blender.org/wiki/Reference/Release_Notes/3.0/Cycleshttps://wiki.blender.org/wiki/Source/Render/Cycles
Credits:
* Sergey Sharybin
* Brecht Van Lommel
* Patrick Mours (OptiX backend)
* Christophe Hery (subsurface scattering anisotropy)
* William Leeson (PMJ sampling pattern)
* Alaska (various fixes and tweaks)
* Thomas Dinges (various fixes)
For the full commit history, see the cycles-x branch. This squashes together
all the changes since intermediate changes would often fail building or tests.
Ref T87839, T87837, T87836
Fixes T90734, T89353, T80267, T80267, T77185, T69800
