So how’s all this going to impact the conventional workstation market?

All of this begs the question as to whether these new and resurrected approaches from Sun and HP will be a boon or a drag to the existing graphics workstation market. Are we ready to see millions PCs and workstations disappearing from desktops, replaced by some thin client, while the horsepower moves back to a server or rackmount PC blades? Highly doubtful. But we will likely see such solutions successful in certain pockets of the market.

For example, in very high-end applications Sun’s Visualization System and HP’s Remote Graphics can serve as alternatives to multiple high-performance visualization centers which would otherwise be cost prohibitive. Or they can play in applications which demand effective interactive collaboration among geographically distributed team members, each using their existing deskside workstations to both render and display.

Alternatively, HP’s Blade Workstation fits a nice niche for those workstation applications where reliability, security, manageability and minimizing on-the-floor real estate and noise are priorities, and where rendering demand is relatively modest.

Ultimately, where the horsepower lives – centralized in the datacenter or distributed deskside – doesn’t have to be a big concern for those making their living selling hardware based on the status quo: Intel, AMD, Nvidia and the big system OEMs. Think about the Blade Workstation. There’s still (at least) one PC for each user, Nvidia or AMD is still selling a GPU, and Intel or AMD is still selling a CPU. It’s just that the PC containing those chips is a backroom blade, not a deskside tower.

Not only should hardware sales not suffer, optimistically the industry could be looking at a stimulus for more revenue. For distributed visual collaboration, the workstation remains the logical choice as the rendering and display workhorse. So if anything, an IT professional managing a collaborative environment might feel compelled to pay the premium for more powerful workstations, rather than skimping on performance. Even in the case of centralized visualization, the server probably isn’t replacing deskside workstations, as those workstations will likely still be in place as display nodes and for other computing tasks.

All told, the aggregate hardware sold – measured in chips, MIPS or FLOPS – isn’t likely to suffer, regardless of how much headway this new breed of remote rendering solutions can make. Rather, as these approaches take hold, centralized visualization and remote rendering should only add to the market’s demand for professional graphics hardware.