We encountered the following problem while experimenting with VRay Hybrid Rendering in CUDA Mode:
When choosing CUDA mode for rendering a scene, GPU usage during the rendering process is minimal (5-10%, measured in Windows 10 Task Manager). There does not seem to be any kind of render time improvement between rendering CPU only, rendering with multiple GPUs, or any combination of CPU and GPU.
The following hardware configurations were tested:
Windows 7 64bit
Intel i7-5960X
64 GB RAM
NVIDIA TITAN Xp
Maya 2017
Vray 3.60
Windows 10 64bit
Intel i7-7700K
32 GB RAM
2x NVIDIA 1080Ti
Maya 2017
Vray 3.60
What we have tried so far:
- Selecting different hardware configurations in the vray render device selection
- Changing the ray bundle size to high and low values
- Rendering in CUDA as well as OpenCL
None of our efforts have been able to utilize the full power of neither a single GPU, nor the dual setup, with the dual setup barely using the second GPU at all.
Are we missing something here?
When choosing CUDA mode for rendering a scene, GPU usage during the rendering process is minimal (5-10%, measured in Windows 10 Task Manager). There does not seem to be any kind of render time improvement between rendering CPU only, rendering with multiple GPUs, or any combination of CPU and GPU.
The following hardware configurations were tested:
Windows 7 64bit
Intel i7-5960X
64 GB RAM
NVIDIA TITAN Xp
Maya 2017
Vray 3.60
Windows 10 64bit
Intel i7-7700K
32 GB RAM
2x NVIDIA 1080Ti
Maya 2017
Vray 3.60
What we have tried so far:
- Selecting different hardware configurations in the vray render device selection
- Changing the ray bundle size to high and low values
- Rendering in CUDA as well as OpenCL
None of our efforts have been able to utilize the full power of neither a single GPU, nor the dual setup, with the dual setup barely using the second GPU at all.
Are we missing something here?
Comment