US video platform provider Brightcove reminded Faultline at IBC 2022 that it is very much still a transcoding company – despite the numerous tentacles that have extended from its platform since acquiring the Zencoder business a decade ago.
Frustratingly for its rivals, Brightcove has a hankering for being ahead of the curve, and the encoding business is still going strong, having transcoded 20 million titles last month. Zencoder had embraced the concept of cloud video encoding before even Elemental was pioneering the space, an opportunity spotted by AWS three years later in 2015 – an example of the kind of forward-thinking reputation Brightcove is eager to sustain.
This is partly due to Brightcove’s R&D department, which the company’s IBC representatives confidently tells us is unrivaled. It may seem strange therefore that Brightcove is not betting big on next-generation codecs.
The only codec worth betting on is AVC, in Brightcove’s view, due to its unmatched global penetration.
Of course, Brightcove’s compression team isn’t totally ignorant of the fact newer codecs bring greater encoding efficiencies. Something Brightcove’s R&D department has been busy working on is a multi-codec streaming ladder layering AVC and HEVC, using its content aware encoding techniques.
With HEVC being mostly supported in Apple devices, deploying multi-codec streaming using the Brightcove Video Cloud has been shown to offer a considerable reduction in CDN costs, without compromising reach to legacy AVC devices.
Brightcove can identify device codec support and switch mid-stream, to optimize for the same bandwidth but with higher quality. Brightcove isn’t aware of any competitors in the space creating mixed HEVC and AVC profiles like it does in creative combinations, the company tells us.
Brightcove also had some critical words on LCEVC that caught us off-guard during IBC 2022.
Even former adversaries have been rallying around the MPEG-5 compression standard created by V-Nova, making Brightcove something of an anomaly here.
In fact, the whole LCEVC mission at trade shows post-pandemic has been about emphasizing the LCEVC ecosystem – collectively as 27 companies rather than V-Nova as a one-man band, as was the story of a few years ago. The concern is that Brightcove is a sizable anomaly.
These criticisms do not come without hard evidence. We are pointed towards some academic papers. The University of Kingston and TU Berlin have together evaluated LCEVC for live gaming video streaming applications. A recent paper compares LCEVC performance against H.264 and HEVC – concluding that LCEVC brings no meaningful mean opinion score improvements for HEVC (x265) against multiple bitrates.
V-Nova knows this. It backed this very paper and advised on encoder settings. We are told that operator error could not have been the issue here.
Another paper from revered research institute InterDigital (otherwise known as the former Technicolor research lab) has penned that VVC as a standalone codec is scalable. InterDigital compares performance of VVC single layer against VVC as two layers – Scalable-VVC for the first layer and a combination of VVC with LCEVC as the second layer.
Scalable VVC can be on par or outperform single-layer coding in specific conditions. The paper also addresses coding complexity and shows that customizing the encoding parameters can lead to large encoding time (and cost) savings.
Either Brightcove is behind the times, which as we have already stated would be unusual, or the US firm just has an astute nose for where and how to make money in this business, because it doesn’t believe LCEVC is it.
These are fine margins. We are talking about 0.5dB to 2dB PSNR differences in this context. Even to the untrained eye, these differences are visible. We are told that 2dB is about the difference between H.264 and HEVC.
And because encoding costs are just a fraction of overall delivery costs for video service providers, they can afford to run slower encoders to save a few bits. As bandwidth and storage costs also dominate Brightcove’s platform, its own positioning is therefore one reflective of the video service providers themselves – and in turn is reflective of the company’s growth.
Eventually, Brightcove does concede that for certain operators where reducing encoding costs is of higher importance, then maybe LCEVC is their solution.
As we said, our debate with the Brightcove team at IBC is a welcome reminder that Brightcove is in fact still an encoding company, but by the same token is a distraction from the broader Brightcove strategy – as an organization now with so many moving parts. We don’t mind this, as there is a chat with CEO Marc DeBevoise – who wasn’t available during the Amsterdam event – lined up for two weeks from now.