This site may earn affiliate commissions from the links on this page. Terms of use.

AMD appear a new line of GPUs today aimed at the increasingly lucrative virtual GPU business. Up until now, that's been generally Nvidia'south turf, and Squad Green has released several iterations of its Nvidia Filigree platform over the past few years. (Notation: Nvidia uses the same "Grid" term to refer to both its hardware virtualization business organization and its deject gaming service.The gaming version is "GeForce Filigree").  The idea behind GPU virtualization is that companies don't necessarily demand to purchase a detached card for every single user or desktop. Instead, companies can centralize their infrastructure and provide admission to virtual systems. By deploying multiple GPUs in a single server, companies tin can provide the benefits of a discrete GPU (up to and including loftier-end workstation-class rendering platforms) while simultaneously cut costs.

AMD is claiming that its new Multiuser GPU family volition support up to fifteen users on a unmarried GPU. Typically this is done past dividing the GPU's total processing power across the number of users, though Nvidia has released Grid cards with multiple GPUs on a unmarried PCB to enable optimal scaling per-user. "The AMD graphics cards are uniquely equipped with AMD Multiuser GPU engineering embedded into the GPU delivering consistent and predictable performance," said Sean Burke, AMD corporate vice president and general manager, Professional Graphics. "When these AMD GPUs are appropriately configured to the needs of an organisation, end users get the aforementioned admission to the GPU no thing their workload. Each user is provided with the virtualized performance to design, create and execute their workflows without any ane user tying upward the entire GPU."

VDI-Clients

AMD's Multiuser GPU uses the SR-IOV (Single Root I/O Virtualization) standard adult by PCI SIG, and the visitor claims information technology'due south explicitly designed for both OpenCL and graphics performance, with OCL 2.0 back up, OpenGL 4.4, and DirectX 12 all supported as well.

Placeholder art or new GPU? We don't know just yet.

Placeholder art or new GPU? Nosotros don't know only still.

With relatively few details, it'due south difficult to say how AMD's Multiuser GPU family volition compare to Nvidia's Grid. The single 3D render of a Multiuser GPU shows a unmarried-slot, adequately long PCB without whatsoever apparent libation. It's not unusual for these kinds of systems to exist cooled by a system fan, but the GPU hardware running underneath that heatsink must be adequately graceful to fit in a unmarried-slot form factor. A bullet point in AMD's PDF implies that these may non be distinctly different cards at all.

AMD-MultiUserFeatures

Given that AMD has appear no hardware specs, information technology'due south possible that Multiuser is a brand proper name for software functionality that'south rolling out to existing hardware. AMD would withal have to practice enough of software work to go these capabilities rolling in its FirePro cards, merely the exact situation is unclear. It's also unclear why AMD's PDF characterizes Nvidia Grid the way information technology does, by listing it equally a "Software" solution compared to AMD's "Hardware" option and past claiming that the maximum number of users supported is eight per GPU.

NV-Grid

Nvidia's own webpages claim that Nvidia Filigree is a hardware-based solution with full GPU laissez passer-through enabled to let local clients to interface properly with remote virtualization stations. It also notes that the maximum number of simultaneous users on a Grid GPU is up to 32 in sure cases. We've reached out to AMD to analyze these questions and will update this story when we hear back. What's probable is that AMD chose some preconditions for its table that aren't listed within it — allocating a certain amount of RAM per user, or supporting sure resolutions could knock dorsum the number of simultaneous users that a Grid GPU tin can support, for example.