IP for Quality Video Panel
June 17, 2014, SMPTE Entertainment Technology in the Internet Age Conference, Stanford, CA—A panel considered the various aspects of video quality based on transmissions over the Internet. Colin Dixon from nScreenMedia moderated the panel comprised of Ian Blain from The Platform for Media, Rodolfo Vargas from EyeIO, Jim Helman from MovieLabs, Hui Zhang from Conviva, and Jean-Christophe Morizar from Dolby.
Problems?
Zhang mentioned that the number of viewers is increasing dramatically, resulting in more buffering and stalls. This waiting has increased to 10 minutes for the World Cup this year from less than 2 for the last Cup. The issue is in areas like encode/decode hardware bottlenecks and other infrastructure limitations rather than Internet bandwidth. The increased number of viewers, 500 times as many viewers this time, and the increasing percentage of streams is challenging viewers expectations for smooth video. One problem is that no one entity owns the whole delivery chain for quality issues, so no one works on any solutions.
Vargas stated that low bandwidth connectivity is an issue. Everyone considers Blu-Ray as the standard for video quality, but delivering this level of quality is a challenge over the Internet. The full production and delivery chain needs more integration and better standards. Compression using H.264 is capable of delivering HD at 3Mb/s.
Morizur offered high dynamic range (HDR) as an option for small screens. 4k video is not equal to human vision capabilities, so even more pixels will not make that big a difference. A 4k tv set needs to be over 55 inches, but smaller screens don’t benefit from the image quality. HDR and high frame rates plus wider color gamut will result in better image perception. The existing supply chain can deliver this level of quality but current practices like color grading, etc. degrades the images to some lowest common denominator. The content will have to be changed to be in HDR formats.
Helman added that UHD is not standardized yet. Previous technologies had existing standards so broadcast and cable had VHS and DVD before consumers. This time, there are no leaders. Technologies like HDR and increased color gamut need work on building blocks and standards for delivery across all platforms and through any pipe. Over the top distributors are leading the way, but are small and under-funded. The standards cover a wide range and will have to support HDR and metadata for the displays to use for conversions. The Web is a fast adaptation vehicle and cable has the largest pipes, so IP delivery will be the future. The broadcast and cable companies cannot change as quickly due to legacy issues. In addition, on-line delivery is also interactive, giving better feedback and analytics on viewers.
Blaine said that TV is now everywhere, and is pushing Comcast to use changing technologies for mobile to affect their cable deliveries from an IP stack. The changes are more about investments and economics than technology. Using for the front end allows the back end to be anything. Eventually, everything will be based on IP. The issue today is that companies have to support both ends and multiple protocols, but eventually everything will consolidate in IP.
As video moves from ATSC to on-line, we add apps and data to change the TV experience?
Vargas responded that the Web allows unlimited innovation. Users will have choices with additional capabilities through technologies like augmented reality. In addition, we will see reduced complexity in the infrastructure as the industry addresses legacy platforms.
Helman suggested the inclusion of interactivity will improve search and discovery. Data delivery will change play and the user interface.
Blaine noted that this is an international market, as Netflix moves to other countries. The incumbents need to change to address this issue of increased interactivity and deliver a mixture of video on demand and scheduled content. These changes are now a business requirement. The creative people will also have to change at their early stages to provide added value for interactivity to the consumer. The delivery companies will need to address issues like curation and search while learning to use the increasing quantities of data for system changes and targeted advertising.
Morizur stated that the user experience will evolve as the on-line capabilities move towards the Blu-Ray standards for 5.1 sound and video quality. Internet delivery will enable greater personalization and also facilitates experimentation. The audience wants accessibility for better personalization and a better user experience. For example, IP delivery can offer variable commentary, changed equalization, and platform independence.
Future impact of changing technologies, like Glass?
Zhang offered that Glass will change the nature of audio and video, but watching on Glass will increase distractions. The metadata management for other devices or for control are possible uses, but not for viewing.
Vargas agreed that Glass might be a second screen.
For 4k, is there capacity for bandwidth to consumers?
Zhang answered that it depends on how many users are online. The complexity is not in the individual components but in the integration of the infrastructure. The user experience needs software for innovation since hardware takes long to change. For example, adaptive bit rate streaming has to be in software within the infrastructure, which will be in very high speed hardware.
Blaine noted that 4k matters in home and on wire, since it is only viable on a big screen. It is not possible to deliver 4k on IP without multi-channel in docsis to get the necessary bandwidth.
DRM standards?
Blaine commented that the current leaders for DRM are Microsoft playready and Adobe primetime.
Helman offered the possibility to encrypt to a common scheme through different DRMs.
