Media applications use mm-renderer to play audio and video tracks. The applications attach mediastore files or playlists as the mm-renderer input and attach one or more hardware devices as the outputs. During playback, mm-renderer manages the media flow between the input and the outputs.
Figure 1. Media Playback
Configuring the rendering service
To play media content, an application must configure the mm-renderer service by defining a context and then attaching an input and one or more outputs to that context. For the input, the application must provide the URL of a track or playlist stored on an accessible mediastore. For each output, it must provide a URL that names an output device and lists the device configuration options. For an overview of the API calls required to set up playback, see the "Playing media" section in the Multimedia Renderer Developer's Guide.
Controlling playback
When an application issues the command to start playback, mm-renderer initiates the media flow between the input and outputs. Note that the rendering service doesn't parse the media files itself but instead uses lower-level mechanisms (e.g., HTTP streamers, file readers) to read and forward the media content to other utilities that send the audio and video components to the right drivers. The main purpose of mm-renderer during playback is to process commands such as changing the playback speed, skipping to a new track position, and stopping playback.
Outputting audio and video
Plugins within mm-renderer communicate with the Screen Graphics Subsystem for outputting video and with the io-audio utility for outputting audio. Screen is the windowing system that mm-renderer uses to render video to the display. The io-audio utility is a resource manager that dynamically loads and configures audio drivers; mm-renderer uses it to deliver the audio in media tracks to the appropriate output hardware (e.g., speakers).