Nvidia announced that it has created added real-time ray tracing to its graphics processing units (GPUs) for workstations in a move that could make it much easier for media and entertainment professionals to create realistic animations quickly.
The Santa Clara, California company unveiled the Nvidia Quadro GV100 GPU with Nvidia’s RTX technology for real-time ray tracing, which is a more efficient way of animating a 3D graphics scene. It effectively uses rays of light to identify and animate the objects in a scene. Nvidia made a similar announcement for game engines last week at the Game Developers Conference.
Until now, it used too much computing power to do ray tracing in real time. Filmmakers could use it for special effects or animations, but it was the kind of task that the artists would leave running on their computers overnight. That’s why Nvidia calls real-time ray tracing the “biggest advance in computer graphics since the introduction of programmable shaders (which made it possible to create new kinds of surfaces in 3D images) nearly two decades ago.”
Now, Nvidia CEO Jensen Huang said at a keynote speech at the company’s GPU Technology Conference (GTC) 2018 event that computationally intensive ray tracing is possible in real time when running professional design and content creation applications.
On stage, Huang pointed out the details like the glint of a stormtrooper’s uniform and subsurface light scattering in Gummi bears and the reflections from a crystal glass. In animated movies, it’s hours to render a single scene, but Nvidia promises we are moving into an era of real-time lighting and shadows.
“It is so hard to compute, and that is why ray tracing has been the Holy Grail of compute science for the last 40 years,” Huang said. “Everything you see here is completely in real time.”
Media and entertainment professionals can see and interact with their creations with correct light and shadows, performing complex renders up to 10 times faster than with a central processing unit (CPU) alone. Product designers and architects can create interactive, photoreal visualizations of massive 3D models — all in real time, according to Nvidia.
“Nvidia has reinvented the workstation by taking ray-tracing technology optimized for our Volta architecture, and marrying it with the highest-performance hardware ever put in a workstation,” said Bob Pette, vice president of professional visualization at Nvidia, in a statement. “Artists and designers can simulate and interact with their creations in ways never before possible, which will fundamentally change workflows across many industries.”
Nvidia added to last week’s announcement by saying it is supported by more than two dozen of the world’s leading professional design and creative applications with a combined user base of more than 25 million customers.
The Quadro GV100 GPU, with 32GB of memory scalable to 64GB with multiple Quadro GPUs using Nvidia NVLink interconnect technology, is the highest-performance platform available for these applications.
Based on Nvidia’s Volta GPU architecture, the GV100 packs 7.4 teraflops of double-precision, 14.8 teraflops of single-precision, and 118.5 teraflops of deep learning performance. And the Nvidia OptiX AI-denoiser built into Nvidia RTX delivers almost 100 times the performance of CPUs for real-time, noise-free rendering.
“The availability of Nvidia RTX opens the door to make real-time ray tracing a reality. By making such powerful technology available to the game development community with the support of the new DirectX Raytracing API, Nvidia is the driving force behind the next generation of game and movie graphics,” said Kim Libreri, chief technology officer at Epic Games, in a statement.
Analyst Patrick Moorhead of Moor Insights & Strategy said in an email, “Ray tracing is one of the best ways to create completely realistic objects and scenes but has been hampered because it takes so much performance and bandwidth. Nvidia is addressing both these bottlenecks with the Quadro GV100. Nvidia is attacking the performance variable with Volta architecture compute and an AI denoiser and addressing bandwidth using the NVLink bus. This gets interesting for IBM Power architectures as it’s the only architecture that supports NVLink.”
The Quadro GV100 GPU is available now on nvidia.com, and starting in April from leading workstation manufacturers, including Dell EMC, HP, Lenovo, and Fujitsu, and authorized distribution partners, including PNY Technologies in North America and Europe, ELSA/Ryoyo in Japan, and Leadtek in Asia Pacific.
Huang said the RTX technology has been 10 years in the making. He said Nvidia estimates that a billion frames are rendered each year, and that will now go up by a factor of 10. That’s based on estimates of activity from 400 games, 500 films, and 12 million product designers.
“The more GPUs you buy, the more you save,” Huang joked.