What is Per-Title Encoding?
Per-title encoding analyzes an individual video and constructs a bitrate and resolution ladder suited to its visual complexity. It replaces a fixed ladder applied uniformly to every title.
How Per-Title Encoding works
Per-title encoding measures the rate-distortion behavior of a particular source instead of applying a universal set of rungs. Trial encodes or content models estimate which combinations of resolution, bitrate, and codec settings deliver useful quality steps. The selected ladder is then packaged for adaptive playback like a conventional ladder, but its shape may differ from every other title. This optimization belongs between source analysis and large-scale encoding, and it must still respect device, network, and business constraints.
Key facts
- Low-motion animation and grainy live action can require very different bitrates at the same resolution and nominal quality, which fixed ladders cannot express well.
- Ladder construction must preserve monotonically useful choices; a higher-bitrate rendition that looks no better wastes storage, CDN bytes, and player options.
- Per-title analysis adds computation before final encoding, but its decisions can reduce downstream transfer when the catalog is viewed often enough.
When Per-Title Encoding matters
Use it when bandwidth efficiency justifies analyzing each source before encoding. A poorly selected quality target can remove useful renditions or spend extra bits without a meaningful visual benefit.