Tangent Animation Scales Render Capacity and Competitive Footprint Using AWS

In animation, a model or scene is created in a computer program. It includes multiple layers of data that define geometry, color, light, shade, texture, viewpoint, and various other scene characteristics. That data file is then rendered by a program that produces a photorealistic image. Threading together each of those files creates a finished film.

Rendering is an intensive process that requires high compute capacity. For animation studios, rendering processes demand large capital expense for infrastructure and often result in production bottlenecks when executed on premises. Studios can better absorb those costs for small projects, but a feature-length film is another story. Animation rendering for full-length movies requires multimillion-dollar investments in infrastructure that would be used during production but would otherwise sit idle.

Tangent Animation’s technical team knew early on the film could not be completed on time or on budget using their on-premises render farm. There are 24 frames in 1 second of film. As a 97-minute film, Next Gen would consist of 139,680 frames and Tangent Animation was tasked to produce four deliverables—mono and stereo versions of the feature in both English and Mandarin. Those four versions equated to more than 3.5 million hours of rendering. The only way for Tangent Animation to meet looming deadlines and produce the required project deliverables was to turn to the Amazon Web Services (AWS) Cloud.

Tangent Animation reached out to AWS to explore using cloud capacity for the project. The goal was to use the company’s current render-management software, AWS Thinkbox Deadline, on Amazon Elastic Compute Cloud (Amazon EC2) Spot Instances. Amazon EC2 Spot Instances offer spare compute capacity at steep discounts compared to On-Demand Instances.

“For our team to get the most from cloud technologies, we needed advice on best practices to use in the cloud environment. With AWS and Thinkbox, we collaborated to work out the nuances and challenges so we could render more effectively," says Zorniak. During production, the team spun up 3,000 Amazon EC2 Spot Instances concurrently, which was 400 percent more capacity than the studio’s on-premises render farm.

Tangent Animation completed 3.5 million hours of rendering in two months, far outstripping the firm’s on-premises capacity, which would have required six to nine months. “If you look at what we were able to do with rendering on Next Gen, it was a godsend for us," says Jeff Bell, COO and vice president of Tangent Animation. “We had hard delivery deadlines that our on-premises machinery simply wouldn't allow us to meet. Using AWS was the best, most effective way to turn this project around so quickly.”

Tangent Animation discovered that the expanded compute capacity not only delivered time savings, but also allowed greater creative control over the finished film. The early stages of a project are spent refining the animation aesthetic—the look and feel of the final film. Early creative decisions impact all the work done after that point. If rendering takes too much time, at some point it becomes impossible for a director to make further creative changes and still meet project deadlines. "Faster rendering gave our artists far more creative freedom. They could make creative decisions and changes faster, getting much closer to the director’s ultimate vision for the final film," explains Zorniak.

AWS Cloud-based rendering capacity delivered greater creative control, and it showed Tangent Animation that it could more readily compete for and win larger projects. “Until we started working on the AWS Cloud, we didn't fully understand the possibilities and potential. Now that we've done it, it's part of our vocabulary and how we think when bidding on new projects,” says Bell. The success of Next Gen inspired the company to plan other strategic moves such as shifting its entire production pipeline to AWS, enabling more remote work, and selling render capacity to other animation companies.

"For us, going with AWS made business sense and creative sense," says Bell. “We now have a strong partner to help us grow the company going forward."