Synamedia Claims First Realtime 8K Encode Without Compromise

Synamedia, the video technology firm spun out of Cisco in 2018, has claimed a world first by achieving real time encoding at full 8K resolution without any comprise over quality.

The company applied its VIVID Compression platform incorporating AMD EP 7763 processors to reach this industry milestone.

The key to this success lay in avoiding the technology trade-offs that have dogged previous efforts, exploiting the full codec toolset. The company removed the need to split the 8K signal into 4K quadrants, or use some form of hardware acceleration such as dedicated GPU memory. These all eliminated issues with memory communication and throughput, as well as local video quality variations, according to Synamedia.

At the same time, Synamedia has added 8K resolution to its cloud-based video quality analysis package, Video-Quality-as-a-Service (VQaaS). This provides objective measurement and subjective visualization analysis.

“We are experiencing a technology shift here, combining our unique and one-of-a-kind compression algorithms, advanced video analytics, and proven expertise with the incredible speeds of the AMD EPYC 7763 processors,” said Elke Hungenaert, VP, product management, Synamedia. “The flexibility of using one CPU without a single technological compromise allows us to provide our customers with the high-quality video content they expect from us. By combining our team’s expertise with the performance of AMD EPYC processors, we can enable more 8K content to the market at a more affordable price to enable the 8K economy.”

The back story here is that Synamedia’s encoding and video quality measurement technology has taken a leap forward since the company poached the visionary Jan De Cock from Netflix in May 2020, not to be confused with the Belgian visual artist of the same name. De Cock was a key architect of Netflix’ Video Multi-method Assessment Fusion (VMAF), a benchmark video quality metric that enhances models of human visual perception with machine learning.

This project started as a research collaboration between Netflix, under De Cock, and the University of Southern California. At Synamedia, De Cock and his team have accelerated efforts to cut 8K down to size so that it becomes viable for streaming delivery and consumption on more widely available devices.

This does still beg the question of why 8K is needed at all given that few people can discern any quality gain over 4K unless they are so close to the screen that viewing is seriously disrupted. The answer is that some appealing use cases, notably zooming in on picture to highlight events within the field of overall view, such as a goal-line incident in association football, can benefit if the pictures are captured and transmitted in 8K. In that example, it means that say a quarter of the whole image blown up to full screen would still be rendered at 4K resolution. According to De Cock, Synamedia’s latest AI algorithms allied to the powerful AMD processor mean that with the additional help of the latest VVC (Versatile Video Codec), 8K is potentially viable for streaming delivery over broadband networks.

The other notable aspect of this 8K demonstration is the use of the generic AMD processors without need for a GPU or dedicated ASIC. It demonstrates that COTS (Common Off the Shelf) hardware can save money without diminishing performance. In the wider context, this can be seen as the latest swing in the Ying and Yan between dedicated and generic silicon. The swing goes towards generic silicon when advances in process design enable functions that were previously only executable in dedicated components such as ASICS can be run in a generic chip.

You might also like...

The Resolution Revolution

We can now capture video in much higher resolutions than we can transmit, distribute and display. But should we?

Microphones: Part 3 - Human Auditory System

To get the best out of a microphone it is important to understand how it differs from the human ear.

HDR Picture Fundamentals: Camera Technology

Understanding the terminology and technical theory of camera sensors & lenses is a key element of specifying systems to meet the consumer desire for High Dynamic Range.

IP Security For Broadcasters: Part 2 - The Problem To Be Solved

By assuming that IP must be made secure, we run the risk of missing a more fundamental question that is often overlooked: why is IP so insecure?

Standards: Part 22 - Inside AIFF Files

Compared with other popular standards in use, AIFF is ancient. The core functionality was stabilized over 30 years ago and remains unchanged.