Use Cases

• CI/CD for game developers

  Build, test and package game clients. Build and deploy container images for dedicated servers.

• CI/CD for plugin developers

  Build and test plugins against multiple Engine versions in parallel. Package plugins for distribution on the Unreal Marketplace.

• CI/CD for enterprise applications

  Build and deploy container images for streaming cloud applications using Pixel Streaming for Linux.

• CI/CD for simulation and machine learning

  Build container images for simulations and deploy them alongside machine learning models in the cloud.

Frequently Asked Questions

• Will Admiral support custom versions of the Unreal Engine or console platforms?

  The on-premises version of Admiral will use container images provided by the user, facilitating both the use of custom Unreal Engine versions and the inclusion of support for console platforms. However, the cloud-hosted (SaaS) version of Admiral will not support either of these features due to technical and legal restrictions.

• Will the SaaS version of Admiral include additional features such as fully managed Pixel Streaming?

  No, the cloud-hosted (SaaS) offering will not include additional managed services. Admiral is designed to act as a foundational automation platform for Unreal Engine projects and plugins, providing a flexible base which can be easily integrated with additional services. Admiral itself focuses purely on core CI/CD pipeline functionality, but allows users to run any arbitrary commands and hook into external services via any desired mechanism (e.g. webhooks, API calls, etc.) The Admiral API will also allow external services to trigger pipeline runs as needed, providing bi-directional integration for orchestrating more complex workflows.

  As an example, the workflow for a project using Pixel Streaming might look like so: the project's CI/CD pipeline first builds and packages the project inside Admiral's provided Pixel Streaming for Linux container image, and then builds a new container image to encapsulate the project's packaged files. This container image is then pushed to an external container registry and API or webook calls are made to notify the external deployment system (e.g. a Kubernetes cluster or a managed Pixel Streaming service) of the image digest so that the deployment configuration can be updated.

  This unopinionated approach allows Admiral to remain as simple and focussed on its core functionality as possible, whilst providing unlimited possibilities by facilitating deep integration with external services. TensorWorks will be announcing additional services to complement Admiral and the wider Unreal Engine ecosystem at a later date.

  If you are interested in exploring additional managed services (such as Pixel Streaming), please contact us so we can discuss your requirements. This information will help to guide our future development roadmap for the Admiral service ecosystem.

• What is the timeframe for release?

  We are currently planning to run the first round of beta testing for on-premises customers in Q3 2020 and the first round of beta testing for SaaS customers some months after that. If you would like to join the beta invite waiting list then be sure to register your interest today. Note that the dates listed here are provisional and may be subject to change in light of the ongoing global COVID-19 pandemic.