Why do broadcasters suddenly care about IP - it’s not why you think!
summary
Explains why broadcasters move live production to IP by giving measurable engineering tradeoffs: handling tens or hundreds of gigabits of uncompressed 10-bit video, CRC checks that burn a quarter CPU core, software paths using net map/DPDK to batch packets and avoid syscall overhead, demonstrated 40 Mbps high‑bitrate feeds and 80 Gbps on a core with BBC research, workarounds for SDI quirks in ffmpeg, and the cost savings from remote production.
AI-written summary · about
chapters (16)
- 0:00Introduction to IP video
- 1:06Production process evolution
- 2:05Traditional hardware workflow
- 3:35Moving production to IP
- 4:46Software vs. hardware cultures
- 5:37Video contribution methods
- 7:08Protocols and timing models
- 8:45Latency and protocols
- 10:40Remote production benefits
- 11:55SDI and network scaling
- 13:50Mapping SDI to IP
- 15:42Pixel format challenges
- 17:32Optimizing network I/O
- 20:02NIC programming and timing
- 21:43Future standards and conclus
- 23:03Conclusion
related talks
22mSMPTE ST 2110 - Highest Quality, Lowest Latency, Pro Video over IP
07mSimon Frost interviewed at Velocity 2010
10mMoving Live Video Quality Control from the Broadcast Facility to the Living Room
05mWeb Performance, Why it Really Matters
22mProviding better video experiences for the next billion users.
18mDebunking the Myth of Video Delivery Monitoring on End-User Devices
20mYour TV Is Eating Your Frames
18mHow we accelerated Live 2 VOD delivery from 15 minutes to under a minute