"By following a few steps, you can invite someone to connect to your computer. After he or she is connected, that person can view your computer screen and chat with you about what you both see. With your permission, your helper can even use his or her own mouse and keyboard to control your computer and show you how to fix a problem."
[0] - http://windows.microsoft.com/en-us/windows/what-is-windows-r...
in other words, teamviewer requires 0 user interaction
If it's a domain computer you don't even need that.
servers is one thing, user support is another animal.
https://www.opengl.org/discussion_boards/showthread.php/1981...
Looks like it works on almost any new GPUs now.
Question - Could I use hyper-V and RemoteFX to do something different - enable CUDA/GPU passthrough from a Windows host to a Linux VM?
I'd like to do some deep learning training on my Windows GPU machine without dual booting into Linux.
then you can passthrough hardware directly into a VM, theres some tricks and workarounds to it depending on your hypervisor/main OS
I'm not sure if non-MS clients (like rdesktop) are able to work with it, though.