| Broadcast workflows have become infinitely more complex over the last decade, with media companies today typically operating in a multi-platform environment across hardware, software and cloud microservices. Workflows may be constructed using any number of separate vendor components and each of these requires an API (application program interface) to be able to communicate with other components in the workflow. Put simply, an API is like the intermediary translator that sits between separate applications enabling them to communicate requests and responses back and forth. | Read about real-world use cases and benefits of unified APIs: Download pdf |
Seamless integration of all these disparate components is a challenge for broadcasters and success is dependent on having a well-constructed and properly implemented API in place. The quality and efficacy of APIs also varies hugely, which only serves to add further complications when integrating separate workflow components. And even when APIs are well built and of a high quality, getting the implementation right still takes a lot of development time, and the more APIs involved, the more time is spent on this development process.
Throughout a typical modern workflow, video is encoded and decoded multiple times, and producers and broadcasters must choose whether to use software or hardware encoders and decoders at each of these stages. Using CPU-based software encoders offers higher quality and a broader feature set, particularly for production formats like 10-bit 4:2:2, while hardware-based GPU encoders provide higher processing performance and greater power efficiency.
It's not unusual for producers and broadcasters to want to harness the benefits of both software and hardware encoders and decoders within a single workflow. However, each codec implementation has its own API so integrating both hardware-based GPU encoding or decoding as well as software-based processing involves managing and implementing multiple SDKs and APIs from each vendor. This is a time consuming and complex process.
Major GPU manufacturers such as AMD, NVIDIA, and Intel each offer their own hardware encoders using various codecs that include AVC, HEVC, AV1, with each SDK having its own distinct API. This creates a significant integration challenge for broadcasters who need their workflows to support multiple platforms. For instance, to utilize Intel Quick Sync Video, NVIDIA NVENC, and AMD AMF (Advanced Media Framework) SDKs, separate integration work is required for each API implementation. This triplicates the development effort, leading to increased time and cost. When working with multiple SDKs and hardware APIs across different vendors, developers are having to write and maintain separate code paths for each implementation. This not only increases the complexity and adds to time and cost, but also increases the potential for errors.
The solution to these challenges lies in having a single interface that works across the most used hardware-based GPU encoders, namely AMD, NVIDIA, and Intel, as well as software-based encoders. MainConcept has worked alongside these technology providers to develop MainConcept Easy Video API (EVA), a unified API that simplifies the ability to work with multiple hardware and software codecs in a far more efficient and streamlined manner.
This unified API approach offers several significant benefits:
Software vendors supplying the broadcast industry are used to working with a lot of different APIs and broadcasters and producers are used to constructing workflows using separate components and juggling the multiple software and hardware APIs that different vendors operate. This is nothing new. What the MainConcept EVA solution brings to the table is a simplified and more efficient way to do it.
MainConcept EVA has a wide range of applications. Whether you’re a software vendor developing an ingest, transcode or live encoding product, application or service, or a broadcaster managing an ingest workflow, delivering a live production, or distributing content in lots of different formats ready to deliver to different devices, MainConcept’s unified API adds flexibility and saves time and effort.
By consolidating the disparate APIs of various hardware and software-based encoders into a single, cohesive interface, developers can overcome the complexities of multi-platform integration. This not only saves broadcasters time and money but also ensures a more robust and scalable workflow. As the industry continues to evolve, the demands and complexities of the modern video workflow are only going to increase. Being able to access GPU-based codec libraries alongside software-based codecs all though one simple interface will go a long way towards enhancing efficiency, streamlining operations and adding flexibility.