This actually works somewhat now, although viewport rendering is broken and any
kind of network error or connection failure will kill Blender.
* Experimental WITH_CYCLES_NETWORK cmake option
* Networked Device is shown as an option next to CPU and GPU Compute
* Various updates to work with the latest Cycles code
* Locks and thread safety for RPC calls and tiles
* Refactored pointer mapping code
* Fix error in CPU brand string retrieval code
This includes work by Doug Gale, Martijn Berger and Brecht Van Lommel.
Reviewers: brecht
Differential Revision: http://developer.blender.org/D36
* Remove support for CUDA Toolkit 4.x, only Toolkit 5.0 and above are supported now.
* Remove support for sm_1x cards (< Fermi) for good. We didn't officially support those cards for a few releases already, now remove some special code that was still there.
use arrays instead of textures for general storage on this card (image textures
are still stored as texture). Textures were found to be faster on older cards,
but the limits on 1D texture size have not increased along with the memory size,
which meant that the full 6 GB could not be used.
The performance actually seems to be slightly better with arrays in some tests
on Titan. For older cards there seems to be a bit of a mix, some are better and
others not. We may change those to use arrays too, but more testing is needed,
only Titan and Tesla K20 (sm_35) is changed for now.
The fact that arrays are faster is a bit surprising, as others found textures
to be faster on Kepler. However even if they were, the memory limitation is
more important to solve anyway.
https://research.nvidia.com/publication/understanding-efficiency-ray-traversal-gpus-kepler-and-fermi-addendum
except for curves, that's still missing from the OpenColorIO GLSL shader.
The pixels are stored in a half float texture, converterd from full float with
native GPU instructions and SIMD on the CPU, so it should be pretty quick.
Using a GLSL shader is useful for GPU render because it avoids a copy through
CPU memory.
* GPU kernel can now be compiled without __NON_PROGRESSIVE__ again, was broken after my last commit. Also add a check for have_error(), in case the GPU kernel comes without Non-Progressive, to avoid a crash.
* Don't compile progressive kernel twice on CPU, if __NON_PROGRESSIVE__ would be disabled there.
* Non-Progressive integrator is now available on the GPU (CUDA, sm_20 and above).
Implementation details:
* kernel_path_trace() has been split up into two functions:
kernel_path_trace_non_progressive() and kernel_path_trace_progressive().
* We compile two CUDA kernel entry functions (in kernel.cu) for the two integrators, they are still inside one .cubin file but due to the kernel separation there should be no performance problem. I tested with the BMW file on my Geforce 540M and the render times were the same for 100 samples (1.57 min in my case).
This is part of my GSoC project, SVN merge of r59032 + manual merge of UI changes for this from my branch.
* Reshuffle SSE #ifdefs to try to avoid compilation errors enabling SSE on 32 bit.
* Remove CUDA kernel launch size exception on Mac, is not needed.
* Make OSL file compilation quiet like c/cpp files.
* Add CUDA compiler version detection to cmake/scons/runtime
* Remove noinline in kernel_shader.h and reenable --use_fast_math if CUDA 5.x
is used, these were workarounds for CUDA 4.2 bugs
* Change max number of registers to 32 for sm 2.x (based on performance tests
from Martijn Berger and confirmed here), and also for NVidia OpenCL.
Overall it seems that with these changes and the latest CUDA 5.0 download, that
performance is as good as or better than the 2.67b release with the scenes and
graphics cards I tested.
* Deprecate computing capability 1.3 (sm_13)
This commit disables auto build of sm_13 CUDA platform, which means that starting with Blender 2.67, we don't support sm_13 devices anymore. It has become difficult to support that and it was already feature incomplete (no render-passes, AO, Multi Closure etc).
It's still possible to manually enable sm_13 for own tests, but building might break in the future.
* CUDA: Make it more clear that sm_12 and below is not supported.
* OpenCL: __KERNEL_SHADING__ was declared twice for nvidia opencl device.
* Some reshuffle of defines in kernel_types.h. No functional changes.
precompiled cubins instead,
Logic here is following now:
- If there're precompiled cubins, assume CUDA compute is available,
otherwise
- If cuda toolkit found, assume CUDA compute is available
- In all other cases CUDA compute is not available
For windows there're still check for only precompiled binaries,
no runtime compilation is allowed.
Ended up with such decision after discussion with Brecht. The thing
is, if we'll support runtime compilation on windows we'll end up
having lots of reports about different aspects of something doesn't
work (you need particular toolkit version, msvc installed, environment
variables set properly and so) and giving feedback on such reports
will waste time.
This commit adds memory usage information while rendering.
It reports memory used by device, meaning:
- For CPU it'll report real memory consumption
- For GPU rendering it'll report GPU memory consumption, but it'll
also mean the same memory is used from host side.
This information displays information about memory requested by Cycles,
not memory really allocated on a device. Real memory usage might be
higher because of memory fragmentation or optimistic memory allocator.
There's really nothing we can do against this.
Also in contrast with blender internal's render cycles memory usage
does not include memory used by scene, only memory needed by cycles
itself will be displayed. So don't freak out if memory usage reported
by cycles would be much lower than blender internal's.
This commit also adds RenderEngine.update_memory_stats callback which
is used to tell memory consumption from external engine to blender.
This information is used to generate information line after rendering
is finished.
Just makes progressive refine :)
This means the whole image would be refined gradually using as much
threads as it's set in performance settings. Having enough tiles is
required to have this option working as it's expected.
Technically it's implemented by repeatedly computing next sample for
all the tiles before switching to next sample.
This works around 7-12% slower than regular tile-based rendering, so
use this option only if you really need it.
This commit also fixes progressive update of image when Save Buffers
option is enabled.
And one more thing this commit fixes is handling display buffer with
Save Buffers option enabled. If this option is enabled image buffer
wouldn't have neither byte nor float buffer until image is fully
rendered which could backfire in missing image while rendering in
cases color management cache became full.
This issue solved by allocating byte buffer for image buffer from
tile update callback.
Patch was reviewed by Brecht. He also made some minor edits to
original version to patch. Thanks, man!
When the scene is updated Cycles resets the renderer device, cancelling
all existing tasks. The main thread would wait for all running tasks to
finish before continuing. This is ok when tasks can actually cancel in a
timely fashion. For OSL however, this does not work, since the OSL
shader group optimization takes quite a bit of time and can not be
easily be cancelled once running (on my crappy machine in full debug
mode: ~0.12 seconds for simple node trees). This would lead to very
laggy UI behavior and make it difficult to accurately control elements
such as sliders.
This patch removes the wait condition from the device->task_cancel
method. Instead it just sets the do_cancel flag and returns. To avoid
backlog in the task pool of the device it will return early from the
BlenderSession::sync function while the reset is going on (tested in
Session::resetting). Once all existing tasks have finished the do_cancel
flag is finally cleared again (checked in TaskPool::num_decrease).
Care has to be taken to avoid race conditions on the do_cancel flag,
since it can now be modified outside the TaskPool::cancel function
itself. For this purpose the scope of the TaskPool::num_mutex locks has
been extended, in most cases the mutex is now locked by the TaskPool
itself before calling TaskScheduler methods, instead of only locking
inside the num_increase/num_decrease functions themselves. The only
occurrence of a lock outside of the TaskPool methods is in
TaskScheduler::thread_run.
This patch is most useful in combination with the OSL renderer mode, so
it can probably wait until after the 2.64 release. SVM tasks tend to be
cancelled quickly, so the effect is less noticeable.
Regular rendering now works tiled, and supports save buffers to save memory
during render and cache render results.
Brick texture node by Thomas.
http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Brick_Texture
Image texture Blended Box Mapping.
http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Textures#Image_Texturehttp://mango.blender.org/production/blended_box/
Various bug fixes by Sergey and Campbell.
* Fix for reading freed memory in some node setups.
* Fix incorrect memory read when synchronizing mesh motion.
* Fix crash appearing when direct light usage is different on different layers.
* Fix for vector pass gives wrong result in some circumstances.
* Fix for wrong resolution used for rendering Render Layer node.
* Option to cancel rendering when doing initial synchronization.
* No more texture limit when using CPU render.
* Many fixes for new tiled rendering.
other places, was mainly due to instancing not working, but also found
issues in procedural textures.
The problem was with --use_fast_math, this seems to now have way lower
precision for some operations. Disabled this flag and selectively use
fast math functions. Did not find performance regression on GTX 460 after
doing this.
The rendering device is now set in User Preferences > System, where you can
choose between OpenCL/CUDA and devices. Per scene you can then still choose
to use CPU or GPU rendering.
Load balancing still needs to be improved, now it just splits the entire
render in two, that will be done in a separate commit.
lower than 1.3, since we're not officially supporting these. We're already not
providing CUDA binaries for these, so better make it clear when compiling from
source too.
* Compile all of cycles with -ffast-math again
* Add scons compilation of cuda binaries, tested on mac/linux.
* Add UI option for supported/experimental features, to make it
more clear what is supported, opencl/subdivision is experimental.
* Remove cycles xml exporter, was just for testing.
* Add back option to bundle CUDA kernel binaries with builds.
* Disable runtime CUDA kernel compilation on Windows, couldn't get this working,
since it seems to depend on visual studio being installed, even though for
this particular case it shouldn't be needed. CMake only at the moment.
* Runtime compilation on linux/mac should now work if nvcc is not installed in
the default location, but available in PATH.
* It seems we have a problem compiling the CUDA kernel at runtime on Windows,
will need to investigate more how to solve this best, CPU render should go
fine though.
* Change OPENIMAGEIO to OPENIMAGEIO_ROOT_DIR on linux for consistency.
* Passes renamed to samples
* Camera lens radius renamed to aperature size/blades/rotation
* Glass and fresnel nodes input is now index of refraction
* Glossy and velvet fresnel socket removed
* Mix/add closure node renamed to mix/add shader node
* Blend weight node added for shader mixing weights
There is some version patching code for reading existing files, but it's not
perfect, so shaders may work a bit different.