Break+Enter Builds VFX Workflow on AWS

Visual effects studio Break+Enter may not have started working in the cloud, but it got there as fast as it could.

The VFX team is an extension of multi-location creative studio Nice Shoes, which specializes in high impact short form content for brands. Longer form projects are Break+Enter’s sweet spot, including its debut credit, Wendy, a feature-length re-imagining of the Peter Pan story; and season two of Pennyworth, Warner Bros. TV’s series about the origins of Bruce Wayne’s butler, Alfred Pennyworth. After Wendy’s launch and as the team took on its next batch of projects, Break+Enter began to examine how it could enhance its VFX pipeline, ultimately opting to go all-in on Amazon Web Services (AWS).

“Nice Shoes has a history of using the cloud for remote services, so expanding our implementation to Break+Enter was a natural evolution,” said Dave Zeevalk, Break+Enter VFX Supervisor and Creative Director. “We successfully migrated our entire workflow to AWS mid-project on a feature and are now running exclusively on AWS. Much of our setup efforts focused on building out a custom front end, and that continues to be a priority. Since AWS now handles most of the technical backend, we’ve freed up resources to work on specialized areas that will benefit our artists.”

While envisioned as a virtual studio to give artists and clients the utmost flexibility, Break+Enter is formally headquartered in New York, which, like many locations, was subject to strict lockdowns throughout much of 2020. Initially, Teradici Cloud Access Software allowed artists to continue projects remotely by connecting to the machines in New York. Collaborating with AWS, the studio migrated their pipeline to the cloud. Now artists tap Linux-based virtual workstations running on Amazon Elastic Compute Cloud (EC2) G4dn instances for content creation, Amazon Simple Storage Service (S3) and Amazon FSx Lustre for storage, and Amazon EC2 Spot Instances for rendering.

“By building our infrastructure and workflow on AWS, our studio can be more dynamic,” noted Luiz Garrido, Cloud Architect/Head of Pipeline. “We’re able to quickly launch GPU powered virtual machines tailored for specific tasks and just as easily shut them down. We have the infrastructure we need, when we need it, so we can be location agnostic. It’s not a problem to launch a team in London, for example, and collapse it back down once the project wraps.”

After successfully testing its new workflow to deliver shots for two feature projects – one of which was done fully in the cloud – the Break+Enter team can scale at a moment’s notice. Additionally, by leveraging the services and solutions of AWS, the studio can allocate its development resources to targeted initiatives. Zeevalk explained, “The support we received from AWS is invaluable. AWS has been highly involved in M&E, and their emphasis on the industry and security earned our respect and confidence in their services.”

Looking forward, Harry Skopas, CTO, anticipates that Nice Shoes will take a page from the Break+Enter playbook and expand its own cloud usage throughout the rest of the studio’s teams; he shared, “Nice Shoes has leveraged the cloud for many years, mainly for rendering and media-based services, but given our long history in high-end color and finishing we have a substantial setup of high-performance on-premises hardware/software. Certain demanding applications, workflows, and user interactivity still benefit from a physical setup. We can have the best of both worlds by combining the power of our on-premises hardware with the scalability of the cloud thus implementing a hybrid-cloud solution. We’ll continue to evaluate our workflows, procedures, and performance demands as our on-premises gear ages out and take advantage of the appropriate cloud-based offerings. AWS is a leader in the space and offers amazing toolsets and flexibility; we look forward to further collaboration as both Nice Shoes and Break+Enter continue to grow and evolve.”

For more information on cloud-based content production with AWS, check out: