How live streaming works
- Capture: cameras and audio feed a production mixer
- Contribution: a live encoder compresses the feed and sends it to the platform, commonly over RTMP or SRT
- Transcoding: the platform creates several quality levels in real time
- Packaging: the stream is cut into short segments for HLS or DASH
- Delivery: a CDN distributes the segments to viewers
- Playback: the player fetches segments and adapts quality to the connection
Latency
Each stage adds delay. Standard segmented streaming runs roughly 10 to 30 seconds behind real time, which is fine for most events. Low-latency modes bring that down to a few seconds, for sport, betting and anything with live interaction.
Lower latency leaves the player less buffer to ride out network trouble, so it trades some resilience for immediacy.
Ways to earn from live
- Included in a subscription, as a reason to stay
- Pay-per-view tickets for single events
- Advertising in breaks
- Sponsorship of the broadcast
- Replays sold or added to the catalog afterward
What makes live harder than on-demand
There is no second take. Viewers arrive together, so sign-in, payment and delivery all peak at once, and a failure is seen by everyone. Test the whole chain beforehand, use a backup encoder and connection for important events, and size capacity for the peak rather than the average.
Last reviewed