Nvidia and Mental Images create a RealityserverSupposedly a good thingBy Ed BerridgeWednesday, 21 October 2009, 15:18THE GREEN GOBLIN and its mate Mental Images have emerged from their smoke filled labs having created what they call a Realityserver platform for cloud computing.Nvidia's version of reality is a combination of GPUs and software that streams interactive, photorealistic 3D applications to any web-connected PC, laptop, netbook or smartphone.Under the bonnet is Nvidia's Tesla RS GPU-based server running Realityserver software from Mental Images. Using it, computer generated imagery that can take hours or days to create can be managed in streams of photorealistic images at rates approaching an interactive gaming experience, the companies claim.According to a press release that afforded the INQ team hours of amusement, Realityserver is the culmination of nearly 40 collective years of hardware and software engineering by Nvidia and Mental Images. Given that Nvidia has only been in business since 1993 it seems that time itself bends around the Green Goblin's press office.Anyway the software apparently enables developers to create a new generation of consumer and enterprise 3D web applications, all with remarkable levels of photorealism.This means that car product engineering teams will be able to securely share and visualize complex 3D models of cars under different lighting and environmental conditions.Architects can see architectural models, in different settings, including day or night, although we would have thought the night view could easily be rendered by unplugging the monitor.The press release also enthuses that online shoppers will be able to design home interiors, rearrange furniture, and view how fabrics will drape, all with perfectly accurate lighting.We would have thought that the killer application for this would be the porn industry, but the companies' press release somehow fails to mention that.Dan Vivoli, senior vice president, Nvidia said that Realityserver is "one giant leap closer to the goal of real-time photorealistic visual computing for the masses."The software uses Mental Images' Iray technology, which is a ray-tracing renderer that employs the CUDA architecture of Nvidia GPUs to create images. It manages to do this by simulating the physics of light on surfaces. Since ray tracing requires shedloads of processing, Iray uses the parallel computing capabilities of Tesla.Rolf Herken, CEO and CTO of Mental Images said that the technology will enable punters to experience the web in 3D.The funny thing about this pair is that while Nvidia insists on spelling its name in all capitals, Mental Images wants everything to appear in lower case. We don't observe either firm's marketing perversions.These quirks mean that the opening phrase of the Realityserver press release reads, "NVIDIA, inventor of the graphics processing unit (GPU), and mental images...." While we suspect that people have been having mental images for millennia before Nvidia came on the scene, let's hope that the Green Goblin and its new partner have not trademarked all of our thoughts. µ