How many channels can the platform handle?

Channel count is limited by encoding capacity rather than the CMS, and scales horizontally. Lineups from a handful of channels to a full multi-genre grid are handled the same way.

What end-to-end latency should we expect?

Roughly 20-30 seconds with standard HLS, and under 10 seconds with low-latency HLS or DASH. The lower figure requires modern client devices and a CDN configured for chunked delivery.

Can viewers restart a programme that is already running?

Yes, with cloud DVR enabled - start-over and catch-up are served from the recorded copy of the linear channel rather than the live window.

live_tv

Linear TV

Manage the full linear lineup — News, Entertainment, Documentary, Sports, Kids and Music channels — with scheduled rows, featured content and live playout visible straight from the CMS.

Channel lineup

Multi-genre linear channels managed in one place.

Row scheduling

Drag-and-drop rows and reorder featured videos.

Live timeline

24-hour scrollable EPG per channel.

CMS-driven

Edit the front-end lineup without redeploying the app.

Playout and the live signal chain

A linear channel starts as a contribution feed - SRT, RTMP or an SDI signal handed to an encoder on site - which is transcoded into an ABR ladder and packaged as a rolling HLS or DASH live window. The live manifest advertises only the most recent segments, with the window length deciding how far back a viewer can scrub and how much storage the edge keeps warm.

Latency is a design decision rather than a fixed property. Standard HLS with 6-second segments and a three-segment buffer lands around 20-30 seconds behind the source, which is fine for general entertainment. Low-latency HLS or DASH with chunked transfer brings that under 10 seconds, at the cost of more requests, tighter CDN requirements and more fragile behaviour on older TV browsers - worth it for sport and live news, rarely worth it elsewhere.

Building the lineup in the CMS

Channels are grouped by genre - news, entertainment, documentary, sport, kids, music - and the on-screen ordering, featured rows and promoted content are edited in the CMS rather than hardcoded in each app. That matters more on TV than on web: shipping a change through six app stores takes days to weeks, so anything editorial has to be server-driven.

The same rows drive every client, so a schedule change or a new channel appears on Roku, Android TV, Tizen and webOS at once, without a release.

Redundancy and failover

Live has no second chance, so the parts that can fail alone are duplicated: two contribution paths from the venue or playout centre, two encoders, and an origin that can serve a slate or a backup feed if both drop. Players are given a manifest that can fail over without restarting playback.

The most common real-world outage is not the encoder but the contribution link, which is why a monitored backup path and an automatic slate are usually the highest-value redundancy to add first.

Inside the CMS

Linear channels with live scheduling timeline.

Frequently asked questions

How many channels can the platform handle?: Channel count is limited by encoding capacity rather than the CMS, and scales horizontally. Lineups from a handful of channels to a full multi-genre grid are handled the same way.
What end-to-end latency should we expect?: Roughly 20-30 seconds with standard HLS, and under 10 seconds with low-latency HLS or DASH. The lower figure requires modern client devices and a CDN configured for chunked delivery.
Can viewers restart a programme that is already running?: Yes, with cloud DVR enabled - start-over and catch-up are served from the recorded copy of the linear channel rather than the live window.

Discuss your projectarrow_forward