If this is your first visit, be sure to
check out the FAQ by clicking the
link above. You may have to register
before you can post: click the register link above to proceed. To start viewing messages,
select the forum that you want to visit from the selection below.
Exciting News: Chaos acquires EvolveLAB = AI-Powered Design.
To learn more, please visit this page!
New! You can now log in to the forums with your chaos.com account as well as your forum account.
Announcement
Collapse
No announcement yet.
CPU/GPU Hybrid Using/Choosing CPU Only Render Node
To connect both machines you can use Distributed rendering. Just make sure that you have enabled the GPU and the CPU no the node as well. You can do that with the ocldeviceselect.exe tool which comes with V-Ray installation.
The render node is a Linux headless using Centos7, I have logged in and enabled the C++/CPU in ocldeviceselect. There is a message saying the OPENCL in not isntalled, does this have to be installed for it to work?
The render node is a Linux headless using Centos7, I have logged in and enabled the C++/CPU in ocldeviceselect. There is a message saying the OPENCL in not isntalled, does this have to be installed for it to work?
Thanks
Oh, okay.
Just set environment variable VRAY_OPENCL_PLATFORMS_x64=c++/cpu;nvidia
For the current terminal process you have to just "export RAY_OPENCL_PLATFORMS_x64=c++/cpu;nvidia". I am not sure how to export environment variable for the whole OS though (but there is a way).
I've found using Distributed rendering on 11 dual xeon machines to be absolutely amazing the more complicated and longer the frames are. A one minute render is going to have some overhead and not be a good representation of the speed increase. Give it something more to do and it should pull ahead of the other render quite a bit.
No issues with just CPU DR which I?ve been using since v2. The very quick test was specifically concerning the Hybrid in 3.6. I?ll throw a more complex scene at it when I have time to try it it out.
Comment