Individual members - don't forget to link your account to an IEEE account to access the SMPTE Journal, and any library discounts. Creating an account is free. Log in at My Account at smpte.org to get instructions.
Hot Button Discussion The Rise of Adaptive Streaming
In a recent presentation at SMPTE's Annual Technical Conference & Expo, Kilroy Hughes, a senior digital media architect for the Microsoft Corporation, made the observation that the Internet is experiencing an explosion in video traffic, and is well on its way to becoming overloaded and congested unless action is taken to dramatically improve and standardize the efficiency of video streaming. Hughes made his remarks during a Conference seminar titled "Optimal Encoding and Network Utilization for Live and Adaptive Streaming." He presented estimates about just how "overcrowded" the Internet is becoming—more than 40 million Internet-connected TV's will be sold in the coming year, more than 100 million video capable cell phones, and more than 250 million PCs. In a recent chat with SMPTE Newswatch, he was even more blunt.
"There is a huge change under way," Hughes said. "Estimates by Cisco and others project that, by 2014, over 90% of Internet traffic will be video. Right now, over 30% of primetime traffic in the U.S. is video, and that is just (largely) the contribution of two providers—YouTube and Netflix. So we are already in the middle of a huge transition that has largely been enabled by improvements in adaptive streaming technology to allow high-quality delivery of video over the Internet. If we don't develop standards to maximize adaptive streaming efficiency, the Internet could become severely overloaded."
In this regard, Hughes is involved as a contributor to the Digital Entertainment Content Ecosystem (DECE) project, which is a consortium of Hollywood studios, consumer electronics manufacturers and retailers, and various other hardware vendors and systems' integrators. The goal of DECE is to create media format standards that will allow content creators to create material that consumers can share seamlessly across all the various video playing, video downloading, and video streaming devices that are part of their household (hence the term "ecosystem"). DECE's work toward interoperability of a common file format that can be encoded once and then distributed or played on multiple devices has resulted in the UltraViolet™ consumer brand, and a standardized media format for video download and streaming.
Hughes said an UltraViolet™ standard will soon be published, and by the start of 2011, he expects some new consumer products to start carrying the brand's logo. "UltraViolet™ offers a way to encode content once to make it compatible with all retailers and all devices, similar to the way DVDs can be encoded once and delivered through all retailers and played on all logoed players," he said. "Right now, for video (streamed or downloaded from the Internet), a file has to be encoded many different ways at many different distribution points for many media formats, and everyone of those file copies has to be independently delivered and cached on the Internet. There is a huge inefficiency in doing that versus having one file that can be replicated on all networks for all services. DECE is publishing an UltraViolet™ spec for video encoding and packaging that manufacturers building consumer electronic products and computer products can rely on for interoperability, and publishers and networks can rely on for supply chain efficiency, and to make Internet delivery a thousand times more efficient."
As Hughes discussed at the SMPTE seminar, the so-called Common File Format (CFF) takes its basic structure and DRM-interoperability from Microsoft's Protected Interoperable File Format (PIFF). DECE is finalizing work on standardizing the media format, while MPEG is working to standardize a streaming protocol that will create a ubiquitous standard for Internet video delivery. Indeed, the CFF method of encapsulating data streams is designed for download on demand and live adaptive streaming capabilities. Other important features of UltraViolet™ include the adoption of the SMPTE 2052-1 Timed Text Format standard (awaiting publication approval) to add caption and sub-titles' data to Internet video, and several other improvements in video coding efficiency and random access playback.
All this, of course, is pointing toward the day when consumers, by and large, will rely on video that they have downloaded or are streaming, rather than using physical media such as DVDs, tape, and so forth. Hughes suggested that adaptive streaming is the logical path to take to get there over the Internet. He suggested that DECE's work to standardize the content format and MPEG's work to standardize the streaming protocol "will accelerate support and adoption dramatically."
"These trends are converging with the emergence of potential standards, both for the content format and the streaming protocol," he said. "Those trends will meet in the middle and cause companies involved in selling video to ramp up electronic distribution the way Netflix, Apple, and others have, and consider the new standards for video, subtitles, and streaming as a way to achieve the broadest reach with the best network and supply chain efficiency. Video delivery over the Internet is ready to explode next year, and we'll need the engineering improvements and efficiency increases in these new standards to keep from bringing the Internet to its knees from video congestion."
As technology unleashes more options, the concept of cloud computing and data storage becomes more interwoven with the lives of consumers, the work of businesses, even the functioning of governments. Along with this revolution, of course, has come the issue of security, and this topic is one of huge interest in the technology sector, so much so that various organizations, companies, committees, and consortiums are hotly studying these issues and debating various options at an almost fever pitch. The non-profit Cloud Security Alliance, for example, has been examining various architectures and researching potential standards since 2008. Now, Intel and several technology partners have launched a new initiative called Cloud 2015 with an agenda to make cloud computing more interoperable and simple while assuring maximum security at the same time. Among their goals is a plan to pursue what they are calling a "federated" cloud that would allow organizations or groups to share data across different internal and external clouds, and an automated network to allow the secure movement of applications and resources in a more energy-efficient way inside data centers, and also PC- and mobile-device connected clouds that are programmed to know their users preferences, applications, and command structure. Read more about the initiative here.
Meanwhile, the recent U.S. elections may end up impacting research funding for development of IT technology in the United States for the foreseeable future. This is an issue that recently received wide attention when it was announced that China had surpassed the United States in terms of building the world's fastest super-computer. Many bloggers across the Internet are suggesting America's failure to keep pace in the super-computer race is indicative of a larger problem involving computer-science research funding cuts across the board that have accelerated since 2008, and are now expected to be even more severe given the likelihood of Congressional austerity efforts in the wake of the elections. An ongoing series of postings at the Computer Research Policy Blog have been covering this trend recently, and offers up the latest news, developments, and opinions about this issue.
As the 3D paradigm shift moves further into the broadcast world, manufacturers and industry watchers are conducting various kinds of research to figure out if the format is making a big enough impact on viewers to justify the additional costs required of content creators and broadcasters to broadcast 3D programs, and the costs and effort required by consumers to buy 3D televisions, wear 3D glasses, and so forth. One interesting study was recently conducted by ESPN, which made a big effort with its 3D World Cup Soccer coverage in 2010, and then did everything it possibly could to analyze the response of viewers, not only to the games themselves, but to the 3D advertising that accompanied them. Interestingly, a report at the Television Broadcast (TVB) website hints at mixed results. The report suggests viewer attention to advertisements did, indeed, go up with 3D programming, but at the same time, the network continues to examine how its 3D channel (ESPN 3D) will perform in the long-term. Check out the report and follow this debate at the TVB site here.
SMPTE will be holding its 2nd Annual SMPTE International Conference on Stereoscopic 3D for Media and Entertainment on 21-22 June 2011 and is seeking proposals for scientific, academic, and highly technical papers for the event. Click here for details.
Let us know what you think! Provide suggestions and feedback on improving this newsletter by completing this brief survey.