Intel Still Hoping To Have oneAPI/SYCL GPU Acceleration In Blender 3.3

To complement the AMD HIP, NVIDIA CUDA, and NVIDIA OptiX acceleration in Blender 3.2, Intel engineers are still hoping to have their Intel oneAPI SYCL support ready to premiere in Blender 3.3 for Intel GPU-based Cycles acceleration with Intel integrated and discrete graphics processors.

Intel has been working on Blender acceleration using SYCL with oneAPI that would work well with their compute stack on both Windows and Linux.

Intel GPU support is their principal focus while with SYCL being an open Khronos standard and there being various other SYCL software initiatives out there, other GPUs / compute stacks could end up making use of this Cycles target in the future.

The patch adding the Intel oneAPI SYCL support to Blender is over a five thousand line diff.

Intel's most recent review request for Cycles oneAPI/SYCL support can be found via this Blender ticket opened today that succeeds their earlier review request. At today's Blender Render and Cycles meeting it was noted by Intel their plan is to still see the code merged for Blender 3.3. There are though some issues to overcome regarding compilers, licensing, and implementation details. It's also a known issue right now that the Intel performance for pre Arc Graphics is rather slow with this back-end.

The Blender 3.3 schedule puts the bug fixing only "Bcon3" period beginning 27 July and the actual Blender 3.3 release to occur on 7 September, so we'll see if this oneAPI support is ready -- we're certainly crossing our fingers to see it in the next release.