Chipmaker Nvidia is putting in a star performance this week at Siggraph in Anaheim, California. The company's demo spotlights its upcoming
chip, Tegra 5, code-named Project Logan. Due out in 2014, Tegra 5 is no small step for the company and for end users: Tegra 5 is based on NVIDIA's Kepler architecture, which is now going mobile.
The Santa Clara, California company's high-end Tesla graphics chips based on Kepler are being used in some of the world's fastest supercomputers, and now similar features will be available in mobile devices.
A mobile version of Kepler means mobile computing is leaping ahead when it comes to significantly more realistic graphics for mobile devices with lower consumption.
Low on Power
Nvidia's chips technology is recognized because of the way it handles multimedia in mobile devices and this week's demo signifies something more than just happier gaming. For the company, it means that Nvidia's latest graphics core is brought over to the mobile processor so that all graphics products are united on the same microarchitecture. That means its graphics roadmap is now converged.
For the end user, expect Tegra 5 to provide better graphics performance at the same power consumption levels. Smartphones and tablets can more easily deliver enviable multimedia features such as augmented reality and image recognition, yet will not be power guzzlers. Games are typically power-challenging applications on smartphones and tablets, so the power advantage could have major benefits for both end users and developers.
Beyond gaming, expect to see Kepler-gone-mobile as a milestone potentially spawning new applications outside of gaming.
Manufacturing, engineering, medical, and architectural firms will be the likely targets that tech innovators have in mind when working with computational imaging, computer vision, augmented reality and speech recognition applications for mobile devices.
Handling Finer Details
The chip will handle applications through ray-tracing, tessellation, advanced lighting and post processing. (Tessellation creates geometry dynamically on the GPU from high-level descriptions, sizing triangles optimally based on the user's viewpoint.)
Siggraph attendees this week are getting a birds-eye view of the next-gen mobile technology using Kepler by way of Ira, the almost-human face. Ira is startlingly realistic. One of the most detailed models of the human face ever created, Ira is a 3-D model that shows facial gestures, light refraction, wrinkles on the skin, and other tiny details.
The demo is a wow factor not only because Ira is so realistic but because the mobile processor shows the lifelike human face while consuming just two to three watts of power.
Nvidia's CEO Jen-Hsun Huang introduced Ira earlier this year at a GPU conference but it was running on a desktop PC with GeForce GTX Titan GPU. In this week's show, Ira is running on the Kepler GPU inside Logan.
The lesson learned from the demo this week is that features used in high-end GPUs can now be scaled down to fit into the power constraints of mobile devices.
"Logan will advance the capability of mobile graphics by over seven years," NVIDIA's Jonah Alben, senior vice president, GPU Engineering, wrote in a blog post on Wednesday.