I've got a Hyper-V 2016 host passing a P2000 thru to a 2016 VM doing hardware encode/transcode for Plex.
My Holy Grail! I was under the impression from some online articles that this wasn't possible with Server 2016 and Hyper-V. I'm hoping to start sharing my server with some family/friends so making sure I can offload transcoding to a GPU inside of my Plex VM would be a huge benefit for me.
If you don't mind, I do have some questions:
1) Are there any issues you've encountered running this on your Hyper-V VM?
2) I've seen multiple Plex users recommend the P2000 and P4000 for the GPU. Do you know if there are any functional/performance differences between these two if we're only using them for transcoding Plex streams?
3) I'm assuming this approach should still work with Server 2019 since DDA is still supported?
4) Any thoughts on Microsoft's planned replacement for RemoteFX vGPU that they are calling GPU-P/GPU-PV ? I had bookmarked the following article a while back thinking I would need to wait for this technology to accomplish what you've been able to do via DDA -
RemoteFX vGPU put out to pasture as Microsoft RDP grows up
Once again, thanks for this great news and the link you posted. Before picking up one of the nVidia cards, I've got to confirm that my SuperMicro X9DR3-LN4F+ board and Xeon E5-2680 V2 CPU's meet the requirements they discuss in that link.