Server 2012 R2 (GUI, and Hyper-V enabled). Enhanced Session Mode Policy set to "Allow..." on Server and "Use if..." on User. Hardware is Asus P9X79, socket 2011 i7, 24G, etc. RemoteFX 3D Video Adapter set to show Maximum number of monitors = [2]. Server is not part of a domain (i.e., it is WORKGROUP).
Graphics card is GeForce GTX 650 with current NVIDIA driver (tried the Beta but that didn't fetch.) The NVIDIA control panel (on the server) shows correct monitor identifiers (they are not the same). "Screen Resolution" on the server shows both monitors and their correct identifiers.
All drivers (including motherboard and Intel for the i7) are up to date.
Guest is Windows 8.1 (Gen I volume).
<symptoms>Whenever I "Connect..", the "Screen Resolution" dialog shows a single monitor with an incorrect identifier. The "Adapter Type" says Microsoft Basic Display Adapter identified with an IBM SVGA BIOS "Chip type". The mobo has *NO* video hardware on it. Nada. None. (I thought the Windows 8.1 Device Manager was supposed to show the display adapter to be "Microsoft RemotFX Graphics Device - WDDM".) I even tried installing the NVIDA drivers in the VM - but the installation program couldn't find any NVIDA devices.</symptoms>
All I want to do is learn NVIDIA CUDA and MSFT AMP while I'm at home. I'm a MSFT partner so this is not my first dance and the company I work for is more concerned with storage and networking so I can't put NVIDIA GPUs on any of my other Hyper-v servers. Since I can't get a Windows 8.1 VM configured for two monitors, I've got a feeling my chances of becoming familiar with GPGPU programming are between slim and none. HELP!