Overview
V-Ray GPU can leverage CPU cores by running the CUDA engine on the CPU, in what is called CUDA-x86 mode. GPU cores, which also utilize the CUDA engine, can work together with the CPU cores, allowing every processor in a workstation to be used. The CGI community has nicknamed this rendering method Hybrid or XPU rendering, and it is also what is meant by total system performance.
When using both CUDA and CUDA-x86 engines, the contribution of the CPU is similar to that of adding another (typically smaller) GPU card to a multi-GPU configuration. The result is a boost in the performance of production and interactive rendering.
The CUDA-x86 mode supports all features of V-Ray GPU and delivers perceptually identical results.
Learn more in our blogpost about Understanding V-Ray Hybrid Rendering.
How to enable Hybrid rendering?
To get the full performance of your system, simply enable the C++/CPU device from the list of CUDA devices. As of V-Ray 6, this is the default behavior.
Learn more in our article about How to use the GPU Device Selector in V-Ray.
Testing
For our tests, we benchmarked 4 user-provided production scenes. We used a Ryzen 9 5950X 16-core CPU and an RTX 3090 24GB Founders Edition, unconnected to a monitor.
The results tell us that the addition of the CPU cores helped reduce render times by 21 - 34%.
The RTX engine doesn’t use hybrid rendering for technical reasons, but we've added its results in the graphs for comparison. In most cases, using the total system performance (CUDA + CUDA-x86) is faster than using the RTX engine. Additionally, CPUs with very high core counts (e.g., an AMD Threadripper) can often rival the speed of high-end GPUs:
Notes
-
If you run out of GPU memory, you can still render using CUDA-x86 mode using your system memory. In such a way, it is possible to use existing CPU farms to render GPU scenes for an easier transition.
Learn more about it in our article about Optimizing memory (VRAM) usage in V-Ray GPU.
- The CUDA-x86 mode doesn’t require any special drivers or Kernel compilation. You can use CUDA-x86 mode without having NVidia drivers installed and even without a GPU in your machine.
- The CUDA-x86 mode performance scales nearly linearly across CPU cores and clock speed, including Apple's M1, M2 & M3 processors.
-
While V-Ray GPU Hybrid rendering can use the CPUs and GPUs simultaneously, CPU cores and GPU cores are not identical. For example, a GPU with 2560 cores is not simply 320 times faster than an 8-core CPU. To determine the actual speed difference, check the V-Ray Benchmark results for real-world benchmark comparisons.