NVIDIA, Image Engine and Autodesk Showcase RTX Server Using Assets from Lost in Space

At GTC, experience faster content creation workflows and GPU-accelerated rendering with NVIDIA RTX Server and Quadro vDWS.
by Rick Champagne

Using real-world assets from Netflix’s award-winning show Lost in Space, NVIDIA this week at GTC is showcasing the flexibility and impressive rendering speed of NVIDIA RTX Server and Quadro Virtual Data Center Workstation (Quadro vDWS).

In collaboration with Autodesk and Image Engine, a Vancouver-based studio, NVIDIA is demonstrating how artists and studios can produce more visually rich content faster than ever with powerful GPU-accelerated rendering on industry-standard NVIDIA RTX-enabled Autodesk Arnold renderer.

Well known for its Oscar-nominated work on Neill Blomkamp’s District 9, Image Engine has created ultra-realistic characters, environments and simulations for television and film, such as Game of Thrones, Jurassic World: Fallen Kingdom and Fantastic Beasts: The Crimes of Grindelwald.

At GTC, NVIDIA is showing Image Engine’s attack robot from Lost in Space on an RTX Server virtual workstation running Autodesk Maya in real time. Autodesk Arnold rendered the assets — and reduced the time needed by 6x per frame on GPU when compared to CPU. Performance scales to 25x faster on a four-GPU render node over a typical CPU render node for entire shots or animations.

With NVIDIA RTX Server, artists can explore more creative options while creating higher quality effects that would ordinarily take too long to attempt. GPU-rendering also results in faster client review cycles with entire turntables or shots, rather than single frames or partial shots.

“We’re thrilled to collaborate with NVIDIA and Autodesk on the next generation of high-performance production rendering. Delivering the best creative results fast is critical to our success,” said Carsten Kolve, digital supervisor at Image Engine. “GPU rendering on NVIDIA RTX Server with Autodesk Arnold can deliver an exponential leap in speed we hope to utilize in all production departments.”

Courtesy of Image Engine. © NETFLIX

Higher Quality Meets Faster Rendering

Ultra-high-quality film and television content has exploded in volume. Jurassic Park introduced the world to groundbreaking computer-generated imagery with 63 visual effects shots. The modern blockbuster Avengers: Infinity Wars included a staggering 2,680 shots.

Streaming services like Netflix continuously raise the bar on visual quality. Design teams around the world increasingly collaborate and turn to new technology to create high-quality visual effects that meet audiences’ expectations.

With RTX Server and Quadro vDWS, designers can accelerate the entire rendering pipeline, as well as virtualize content creation for increased collaboration in geographically dispersed productions.

Design teams are turning to virtual workstations to create photorealistic visual effects with accelerated rendering power. The Turing architecture provides advancements in ray tracing, AI, advanced shading and simulation — all of which can be experienced from a virtual workstation on the RTX Server.

With Quadro vDWS, studios can have one, easy-to-manage server with multiple virtual workstations so designers can share GPU resources and securely access their work from any location.

“The ray-tracing compute capabilities of NVIDIA’s RTX GPUs and RTX Server, now with virtual workstations, are helping artists push the boundaries of content creation,” said Chris Vienneau, senior director of Media & Entertainment Products at Autodesk. “The Arnold GPU beta is now available to all our customers and we are excited to see how our users will take advantage of the unprecedented rendering speeds and amazing efficiency thanks to the powerful RT Cores, Tensor Cores and CUDA cores.”

Check out the NVIDIA booth at GTC to learn more about NVIDIA RTX Server and Quadro vDWS.

Learn more about Autodesk Arnold 5.3 with Arnold GPU in public beta.

Special thanks to Autodesk, Image Engine and Netflix for their partnership on this project.