Russell Trafford-Jones, Author at The Broadcast Knowledge

Video: VP9 Transcoding for Live eSports Broadcast

Posted on 6th March 2019 by Russell Trafford-Jones

VP9 is a well-known codec, but it hasn’t seen many high-profile, live deployments which makes Twitch’s move to deliver their platform using VP9 in preference over AVC all the more interesting.

Here, Yueshi Shen from Twitch, explains the rationale for VP9 by explaining the scale of Twitch and looking at their AVC bitrate demands. He explains the patent issues with HEVC and VP9 then looks at decoder support across devices and platforms. Importantly, encoder implementation is examined leading to Twitch’s choice of FPGA to provide live encoding.

Yueshi then looks at the potential of AV1 to Switch_Frame to provide low-latency broadcast at scale.

Watch now!

Speaker

Yueshi Shen
Principal (Level 7) Research Engineer & Engineering Manager,
Twitch

Video: Using CMAF to Cut Costs, Simplify Workflows & Reduce Latency

Posted on 5th March 2019 by Russell Trafford-Jones

There are two ways to stream video online, either pushing from the server to the device like WebRTC, MPEG transport streams and similar technologies, or allowing the receiving device to request chunks of the stream which is how the majority of internet streaming is done – using HLS and similar formats.

Chunk-based streaming is generally seen as more scalable of these two methods but suffers extra latency due to buffering several chunks each of which can represent between 1 and, typically, 10 seconds of video.

CMAF is one technology here to change that by allowing players to buffer less video. How does this achieve this? An, perhaps more important, can it really cut costs? Iraj Sodagar from NexTreams is here to explain how in this talk from Streaming Media West, 2018.

Iraj covers:

A brief history of CMAF (Common Media Format)
The core technologies (ISO BMFF, Codecs, captions etc.)
Media Data Object (Chunks, Fragments, Segments)
Different ways of video delivery
Switching Sets (for ABR)
Content Protection
CTA WAVE project
Wave content specifications
Live Linear Content with Wave & CMAF
Low-latency CMAF usage
HTTP 1.1 Chunked Transfer Encoding
MPEG DASH

Watch now!

Speaker

Iraj Sodagar
Independant Consultant
Multimedia System Architect, NexTreams

Video: P4 Tutorial

Posted on 4th March 2019 by Russell Trafford-Jones

P4 is a powerful programming language which runs on network switches themselves allowing realtime manipulation of the data traffic. In broadcast, this can be used to alter SMPTE 2110 video in real time as demonstrated by Thomas Edwards at the EBU Network Technology Seminar year and that can be seen in this short video. “This shows how even on an ethernet switch now, we can program it to make these switching decisions based on any header [including] the application layer of the broadcast data”

This video explains what P4 is and how it works taking us all the way from the core principles to ways of programming it and harnessing its power. Watching the beginning of the video is sufficient for most in order to get a feel for P4 and how it could be (and is) applied to broadcast.

The speakers, from Cisco and Barefoot Networks (who work with Thomas Edwards from Fox), cove these topics:

What is the Data plane
Software Defined Networking (SDN) & Openflow
Benefits of programming your own dataplane
Typical Applications of P4
Novel Applications
Basics of the P4 language
P4 Software tools

Watch now!

Speakers

	Antonin Bas Software Engineer, Barefoot Networks
	Andy Fingerhut Principal Engineer, Cisco Systems

Video: AMWA NMOS State of Play

Posted on 1st March 2019 by Russell Trafford-Jones

With all the talk of the SMPTE ST 2110 standards suite, it’s sometimes forgotten that it only deals with content. If you want a working system, you’ll need to do a few more things – find new devices on the network, work out what they can do, control them, guarantee the bandwidth and often deal with metadata that arrives separately like tallies.

This is what the AMWA NMOS specifications do. Peter Brightwell and Thomas Edwards have been heavily involved in creating them and in this video lead us through what each one does and how they are used.

Watch now!

Speakers