Blender 2.78b performance update

Started by Stefan, February 10, 2017, 08:12:34 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.


Blender 2.78b is a performance update release which mainly includes speedups for Cycles rendering.


  • Multithreaded shader compilation.
  • Various optimizations for emission meshes.
  • Corrected light sampling for Branched Path Tracing, reduces noise leading to faster convergency.
  • Improvements in constant folding algorithm, removing shading nodes which are not used for final evaluaiton.
  • Improvements to some procedural textures (brick)
  • Early light ray termination based on contribution of light to the result.
  • Fix for background MIS with textures having small bright spots, which was causing fireflies.
  • Different seed for subframes and different stereo views, allowing "slow-motion" effect and improves VR experience.
  • Distance culling for objects to remove objects which are too far away from the camera.
  • Fix for undesirable threads affinity change on Windows.
  • Avoid various numerical issues in the kernel, solving fireflies.
  • Fixes for deformation motion blur combined with autosplit.
  • Better remainedtime estimation.
  • Various optimizations for deformation motion blur.
  • Added time steps BVH option to speed up rendering of scenes with motion blur.
Cycles: GPU 

  • Use XDG folder for cache on Linux and OSX, which avoids having per-Blender version folder with all the OpenCL/CUDA kernels built.
  • Ability to enable/disable individual GPUs as opposite to old behavior with pre-defined combinations only. Not only this helps to some artists, but also makes it possible to have flexible benchmark scripting.
Cycles: CPU 

  • Use more global SSE optimizations for SSE4.1+ kernels.
  • Multiple improvements for the latest AVX2 CPUS:

    • Optimized various math utilities (cross-products, dot-products, min-axis-selection and others).
    • Faster version of triangle intersection function.
    • Optimization of various steps in BVH traversal algorithm (including both construction and traversal).
Cycles: OpenCL 

  • Added 3D textures support for OpenCL
  • Make it conditional to compile-in transparent shadows support. Depending on a scene and hardware gives really major render time improvements.
Dependency graph 

  • Solve various race conditions (conflicts between threads which are working on the same data).
  • Optimization of dependency graph construction time.
  • Fix missing or wrong relation links, which were causing hard-to-reproduce bone flickering.
  • Fix various crashes related on linked data blocks.
  • Fixes for missing animation updates.

  • Fixed regression crash when adding texture node to compositor.
  • Fixed crash of inverse kinematics on 32 bit Windows platforms
  • Improve multi-threaded usage of fluid simulation
  • Return correct alpha for environment map in GLSL