New processors and consumer devices are making big bets on HEVC, even as AVC remains ubiquitous. Despite some risk, HEVC can add significant value in certain use cases. In this presentation we present the forecast for HEVC uptake from an objective and reliable perspective. We also provide an update to our long-trusted recommendations on choice of codecs and architectures for popular applications in the context of ongoing trends such as 4K, virtual reality, and virtualization.

Choosing the number of streams in an adaptive group and configuring them is usually a subjective, touchy-feely exercise, with no way to really gauge the effectiveness and efficiency of the streams. However, by measuring stream quality via metrics such as PSNR, SSIM, and VQM, you can precisely assess the quality delivered by each stream and its relevancy to the adaptive group. This presentation identifies several key objective quality metrics, teaches how to apply them, and provides an objective framework for analyzing which streams are absolutely required in your adaptive group and their optimal configuration.

Streaming company-wide meetings, training, and other events through your internal network saves on time, travel costs and resources. This session looks at real-world architectures being used for distributing streams to all screens within an enterprise setting. You’ll learn about how peer-to-peer and multicast technologies fit in, and how to reach remote and mobile workers. Our panelists will recommend ways to increase quality and broadest audience reach, and share best practices and lessons learned.

Video content has seen significant improvement by the introduction of UHD video, high frame rate and high dynamic range. This presentation will showcase the improvements these features add to the video watching experience. It will also cover the video encoding techniques (encoding standards, bitrates and encoding parameters) used today to deliver high bitrate demanding video to the end consumer and challengers this process faces. Finally it will go through some recommendations for the encoding stage for both VOD and live streaming applications.

The video compression industry often encounters claims that some new encoding technology is X% better than previous state-of-the-art encoders, where “X” is some large number. In reality, arriving at a percentage gain is a complex process, depending on the reference (“baseline”) encoding, the encoding settings, the encoding bitrates, the video content being encoded, the method of comparison, and the comparison metrics. This presentation explains how to perform subjective evaluation of encoding technologies using a practical subjective test methodology that is credible and leads to meaningful results. The talk presents several examples of encoded bitstreams to be scored subjectively, shows how to score those bitstreams based on the artifacts present, and illustrates how to collect subjective scores and derive meaningful gain metrics from them.

Providing a good experience is key when providing millions of live video streams to tens of millions of viewers. This session dives into the technology needed to deliver an industry-leading live streaming experience on Periscope. Learn how Periscope, Twitter’s live broadcasting platform and the first to popularize live streaming, recently partnered with OpenTelly to improve its viewer experience through a series of technology migrations. Attendees will hear how Periscope improved the experience delivered to its viewers by migrating to HTML5 and low latency HLS.

This presentation covers how broadcasters can improve quality of service by making small changes to their ABR algorithms and analyzing via an effective AB testing workflow. We begin by offering practical feedback on how Dailymotion was able to improve QoS significantly via small player-side ABR iterations, explaining what metrics were used and the results obtained. In the second part of the talk, we delve into how to build an effective AB testing analytics stack, honing in on the tools needed and the appropriate environment to be able effectively analyze our actions and make changes in real time.

This presentation will focus on best practices around reliable live streaming of critical, high importance global events with multiple redundant signal pathways. Learn how to overcome challenging on-site environments like flakey internet, failing encoders, faulty cables and failing CDNs. Learn some best practice tips covering the gamut of on-site challenges (cameras, switchers, signal paths, ingest/uplink) to reliable end-user playback (transcoding/transmuxing, reliable playback engines, CDN selection). Find out how you can implement backups and redundancy at every possible level of your live workflow. The presenters will also cover best practices (and caveats) around successful leveraging Facebook Live - Facebook's native new entry into the live streaming category that is gaining enormous traction and engagement right now. Learn from our experiences (and the odd failure) to make sure your next event is a resounding success.

During this session, leading content owners and distributors will join top technology companies to examine the elements essential to making VR not just a reality but also a viable business opportunity. The expert panel will define VR video and discuss the workflow required to deliver a high-quality VR experience. Considering audio, video and CGI, the group will cover capture and production, processing, delivery and consumption. The session also will address the importance of standards in scaling the VR business, current technical challenges, and how the industry can address those challenges via more concerted efforts - such as those undertaken in the past by the DASH Industry Forum and the Ultra HD Forum - to drive the technology forward.

Almost all of the commercial cloud encoding services, and many of the largest streaming producers encode in the cloud using FFmpeg. It’s cheap, relatively simple, and highly effective. To accomplish this yourself, you need two basic skill sets; first how to encode with FFmpeg, and then how to automate, schedule, and manage the encoding processes. This presentation will detail how to produce H.264 files with FFmpeg, and describe the tools and techniques available to simplify the automation and management of these encoding jobs for both live and VOD encoding.

A lot of recent changes are going on in the area of online streaming formats. Apple announced at WWDC 2016 the support of fragmented MP4, a key feature of CMAF, and others have demonstrated its usage on all major browser platforms. Video players now support HLS in HTML5 via re-multiplexing in JavaScript, so that HLS can be played using the HTML Media Source Extensions. In parallel, MPEG-DASH got a lot of traction for DRM-protected premium content, as all major Web browsers support now the HTML5 Encrypted Media Extensions. This panel will provide an overview about MPEG-CMAF, and discuss whether it is an opportunity to actually harmonize deployments that currently have to support both (and separate from each) DASH and HLS. Finally, we will also discuss potential threats on existing deployments having its clear focus on MPEG-DASH and DASH-IF.