A Deep Dive Into The AMD/Intel CPU + NVIDIA GPU Performance With Blender 2.90

Written by Michael Larabel in Software on 3 September 2020 at 01:32 PM EDT. Page 3 of 6. 9 Comments.

Next up was making use of an AMD Ryzen Threadripper 3970X system with a NVIDIA GeForce TITAN RTX graphics card while carrying out the 2.82 vs. 2.90 comparison but looking at the various GPU acceleration methods rather than just the CPU-based rendering.

CPU Benchmarks - Blender 2.82 vs. 2.90
CPU Benchmarks - Blender 2.82 vs. 2.90
CPU Benchmarks - Blender 2.82 vs. 2.90
CPU Benchmarks - Blender 2.82 vs. 2.90
CPU Benchmarks - Blender 2.82 vs. 2.90

When looking at the CUDA performance, Blender 2.90 on this system was taking longer to render compared to Blender 2.82.

CPU Benchmarks - Blender 2.82 vs. 2.90
CPU Benchmarks - Blender 2.82 vs. 2.90
CPU Benchmarks - Blender 2.82 vs. 2.90
CPU Benchmarks - Blender 2.82 vs. 2.90
CPU Benchmarks - Blender 2.82 vs. 2.90

OpenCL rendering meanwhile saw Blender 2.90 generally faster than Blender 2.82.

CPU Benchmarks - Blender 2.82 vs. 2.90
CPU Benchmarks - Blender 2.82 vs. 2.90
CPU Benchmarks - Blender 2.82 vs. 2.90
CPU Benchmarks - Blender 2.82 vs. 2.90
CPU Benchmarks - Blender 2.82 vs. 2.90

Meanwhile the flagship NVIDIA back-end for Blender these days is the OptiX code-path. Originally the Blender OptiX functionality was just for RTX graphics cards but has now been extended to cover Maxwell, Pascal, and non-RTX Turing cards too. The OptiX code-path with Blender 2.90 was generally faster than the 2.82 release.


Related Articles
Intel Xeon Max Sees Some Performance Gains For OpenVINO & ONNX With Linux 6.9
Linux 6.9 Drives AMD 4th Gen EPYC Performance Even Higher For Some Workloads
Intel Core Ultra "Meteor Lake" Yields Faster Performance With Linux 6.9
Google Cloud's C3D Instances Provide Strong Performance Value For PingCAP's TiDB
Linux 6.9 Features: DM VDO, AMD Preferred Core, Intel FRED & Larger Console Fonts
LLVM Clang Shows Off Great Performance Advantage On NVIDIA GH200's Neoverse-V2 Cores