Tweet
Hi all,
I am comparing the speed between using Vray CPU vs Vray GPU in Modo.
The looks they generate are close enough that I am happy to use GPU, but the resource allocation is terrible. I have seen some other posts listing a similar issue here, but it's more specific to Mac drivers.
For example, for a 1080p image rendered using the Adaptive sampling method.
I'm using a Quadro P5000 graphics card on Windows 10 with Modo 13.2v1 and V-Ray Next.
CPU Rendering (CPU only) - 3m 24s.
GPU Rendering (GPU + CPU) - 11m 47s.
It takes over 3x as long. What am I missing here? So many people are swearing by the power of utilizing hybrid rendering that there must be something I'm missing.
The resource allocation in GPU + CPU rendering is shown below. GPU bounces around a little and CPU stays at roughly 24%.
Compared to CPU only rendering... CPU constantly at 100% and GPU not used at all (as expected).
Thanks for any advice. This is really slowing down my workflow and I'd love to understand this better to make use of the hardware I have available.
-Jeff
I am comparing the speed between using Vray CPU vs Vray GPU in Modo.
The looks they generate are close enough that I am happy to use GPU, but the resource allocation is terrible. I have seen some other posts listing a similar issue here, but it's more specific to Mac drivers.
For example, for a 1080p image rendered using the Adaptive sampling method.
I'm using a Quadro P5000 graphics card on Windows 10 with Modo 13.2v1 and V-Ray Next.
CPU Rendering (CPU only) - 3m 24s.
GPU Rendering (GPU + CPU) - 11m 47s.
It takes over 3x as long. What am I missing here? So many people are swearing by the power of utilizing hybrid rendering that there must be something I'm missing.
The resource allocation in GPU + CPU rendering is shown below. GPU bounces around a little and CPU stays at roughly 24%.
Compared to CPU only rendering... CPU constantly at 100% and GPU not used at all (as expected).
Thanks for any advice. This is really slowing down my workflow and I'd love to understand this better to make use of the hardware I have available.
-Jeff
