Low latency streaming is always a compromise, but what can be done to keep QOE high?
This on-demand webinar looks at CMAF and presents some real-world data on this low latency technique. The webinar starts by explaining that CMAF is a low-latency streaming technology similar to HLS and other streaming protocols where the idea is to deliver the video as small files. Olivier and Alain from Harmonic explain how this is done and look at some of the trade-offs and compromises that are needed and introduce techniques to keep QOE high. They also look at deployment in cloud vs. on premise.
Pieter-Jan Speelmans talks about play tradeoffs and optimisations within the player. CMAF allows the buffer to be reduced and whilst a bad network may mean you buffer is similar to ‘normal’, but in good networks, this buffer can be brought down significantly. He also talks about how ABR switching is impacted by GOP length even in CMAF.
Viaccess-Orca explains how DRM works with CMAF and looks at some of the challenges including licences acquisition time and overloading licence servers at the beginning of events. Akamai’s Will Law explains some benefits of CMAF and the near-real-time of chunk-based transfer (HTTP 1.1) and how downloading chunks at full speed leads to problems when the same broadband link is used by several clients.
There are lots of good talks on CMAF, but this is one of the few which talks about CMAF not as theory, but as is deployable today.
CMAF brings low latency streams of less than 4 seconds into the realms of possibility, WebRTC pushes that below a second – but which is the right technology for you?
Date: June 12th 2019 Time: 11am PST / 2pm EST / 19:00 BST
CMAF represents an evolution of the tried and tested technologies HLS and DASH. With massive scalability and built upon the well-worn tenants of HTTP, Netflix and a whole industry was born and is thriving on these still-evolving technologies. The push to reduce latency further and further has resulted in CMAF which can be used to deliver streams with five to ten times lower latencies.
WebRTC is a Google-backed streaming protocol with the traditional meaning of streaming; it pushes a stream to you a opposed to the HLS-style methods of making small files available for download and reassembly into a stream. One benefit of this is extremely low bitrates of 1 second or less. Used widely by Google Hangouts and Facebook messenger, WebRTC is increasingly an option for more broadcast-style streaming services from live sports & music to gaming and gambling.
Both have advantages and draw-backs so Wowza’s Barry Owen and Anne Balistreri are here to help navigate the ins and outs of both technologies plus answer your questions.
Everyone has a go-to program or three they use for problem solving. Here is a review of a whole swathe of diagnosis programs out there for live streaming.
There are known favourites like Wireshark, FFPlay and MediaInfo, free applications such as Eyevinn Technology’s Segment Analyser and the open source YUView. And this also covers paid programs like Elecard’s Stream Analyser and Telestream Switch.
This talk by David Hassoun CEO of RealEyes media is well worth a look because there is bound to be something there you didn’t know about – and who knows how useful that will be to you!
Server-Side Ad Insertion (SSAI) it’s the best defence against ad-blockers, but switching in an ad at source can be tricky particularly in low latency streams. This talk at the OTT Leadership Summit at Streaming Media East brings together leaders in the field to explain where they’re up to in delivering this technology and the benefits they see.
Magnus Svensson tells us about the instrumental role Eyevinn Technology, the consultancy who run the technical conference Streaming Tech Sweden , is played in Sweden creating an open standard for all the broadcasters to work to in order to agree how to track SSAI allowing the correct payments to be made. Magnus also talks about aligning SCTE insertion with MPEG structure and the importance of correct preparation of the source video.
Tony Brown from Newsy talks about the centralised nature of SSAI making management easier and gives ana overview of decisioning bringing together buys and sellers of ads. Tony also discusses other analytics such as adjacency and targeting.
Jason Justman of Sinclair Broadcasting Group, explains SCTE insertion and talks about the technical difficulties in reacting to live changes in programming.
Geir Magnusson, Jr. from fuboTV covers the difficulties of preparing the ads quickly enough for thousands or millions of streams to get customised, SSAI ads at the same time and discusses his strategy to start pre-fetching ads from the ad server to prepare them ahead of time. Geir also highlights the misunderstanding that can exist where streaming provides the same video and programme experience as traditional broadcast but ad buyers don’t all understand how much more targeting is possible – even with SSAI.
Most online video streaming uses HTTP to deliver the video to the player in the same way web pages are delivered to the browser. So QUIC – a replacement for HTTP – will affect us professionally and personally.
This video explains how HTTP works and takes us on the journey to seeing why QUIC (which should eventually be called HTTP/3) speeds up the process of requesting and delivering files. Simply put there are ways to reduce the number of times messages have to be passed between the player and the server which reduces overall overhead. But one big win is its move away from TCP to UDP.
Robin Marx delivers these explanations by reference to superheroes and has very clear diagrams leading to this low-level topic being pleasantly accessible and interesting.
There are plenty of examples which show easy-to-see gains website speed using QUIC over both HTTP and HTTP/2 but QUIC’s worth in the realm of live streaming is not yet clear. There are studies showing it makes streaming worse, but also ones showing it helps. Video players have a lot of logic in them and are the result of much analysis, so it wouldn’t surprise me at all to see the state of the art move forward, for players to optimise for QUIC delivery and then all tests to show an improvement with QUIC streaming.
QUIC is coming, one way or another, so find out more. Watch now!
Web Performance Researcher,
Adaptive bitrate, ABR, is vital in effective delivery of video to the home where bandwidth varies over time. It requires creating several different renditions of your content at various bitrates, resolutions and even frame rate. These multiple encodes put a computational burden on the transcode stage.
Lowell Winger explains ways of optimising ABR encodes to reduce the computation needed to create these different versions. He explains ways to use encoding decisions from one version and use them in other encodes. This has a benefit of being able to use decisions made on high-resolution versions – which are benefiting from high definition to inform the decision in detail – on low-resolution content where the decision would otherwise be made with a lot less information.
This talk is the type of deep dive into encoding techniques that you would expect from the Video Engineering Summit which happens at Streaming Media East.
Mux’s Justin Sanford explains the difference between Quality of Service and Quality of Experience; the latter being about the entire viewer experience. Justin looks at ‘Startup time’ showing that it’s a combination of an number of factors which can include loading a web page showing the dependence of your player on the whole ecosystem.
Justin discusses rebuffering and what ‘quality’ is when we talk about streaming. Quality is a combination of encoding quality, resolution but also whether the playback judders.
“Not every optimisation is a tradeoff, however startup time vs. rebuffering is a canonical tradeoff.”
Finally we look at ways of dealing with this, including gathering analytics, standards for measuring quality of experience, and understanding the types of issues your viewers care most about.
There are two main modern approaches to low-latency live streaming, one is CMAF which used fragmented MP4s to allow frame by frame delivery of chunks of data. Similar to HLS, this is becoming a common ‘next step’ for companies already using HLS. Keeping the chunk size down reduces latency, but it remains doubtful if sub-second streaming is practical in real world situations.
Steve Miller Jones from Limelight explains the WebRTC solution to this problem. Being a protocol which is streamed from the source to the destination, this is capable of sub-second latency, too, and seems a better fit. Limelight differentiate themselves on offering a scalable WebRTC streaming service with Adaptive Bitrate (ABR). ABR is traditionally not available with WebRTC and Steve Miller Jones uses this as an example of where Limelight is helping this technology achieve its true potential.
Comparing and contrasting Limelight’s solution with HLS and CMAF, we can see the benefit of WebRTC and that it’s equally capable of supporting features like encryption, Geoblocking and the like.
Ultimately, the importance of latency and the scalability you require may be the biggest factor in deciding which way to go with your sub-second live streaming.
Streaming on the net relies on delivering video at a bandwidth you can handle. Called ‘Adaptive Bitrate’ or ABR, it’s hardly possible to think of streaming without it. While the idea might seem simple initially – just send several versions of your video – it quickly gets nuanced.
Streaming experts Streamroot take us through how ABR works at Streaming Media East from 2016. While the talk is a few years old, the facts are still the same so this remains a useful talk which not only introduces the topic but goes into detail on how to implement ABR.
The most common streaming format is HLS which relies on the player downloading the video in sections – small files – each representing around 3 to 10 seconds of video. For HLS and similar technologies, the idea is simply to allow the player, when it’s time to download the next part of the video, to choose from a selection of files each with the same video content but each at a different bitrate.
Allowing a player to choose which chunk it downloads means it can adapt to changing network conditions but does imply that each file has contain exactly the same frames of video else there would be a jump when the next file is played. So we have met our first complication. Furthermore, each encoded stream needs to be segmented in the same way and in MPEG, where you can only cut files on I-frame boundaries, it means the encoders need to synchronise their GOP structure giving us our second complication.
These difficulties, many more and Streamroot’s solutions are presented by Erica Beavers and Nikolay Rodionov including experiments and proofs of concept they have carried out to demonstrate the efficacy.
MUX is a very pro-active company pushing forward streaming technology. At NAB 2019 they have announced Audience Adaptive Encoding which is offers encodes tailored to both your content but also the typical bitrate of your viewing demographic. Underpinning this technology is machine learning and their Per-title encoding technology which was released last year.
This talk with Nick Chadwick looks at what per-title encoding is, how you can work out which resolutions and bitrates to encode at and how to deliver this as a useful product.
Nick takes some time to explain MUX’s ‘convex hulls’ which give a shape to the content’s performance at different bitrates and helps visualise the optimum encoding parameters the content. Moreover we see that using this technique, we see some surprising circumstances when it makes sense to start at high resolutions, even for low bitrates.
Looking then at how to actually work out on a title-by-title basis, Nick explains the pros and cons of the different approaches going on to explain how MUX used machine learning to generate the model they created to make this work.
Finishing off with an extensive Q&A, this talk is a great overview on how to pick great encoding parameters, manually or otherwise.