Rendering

Rendering can be thought of as having two stages: Source Interpretation and Sample Processing.

Source Interpretation is when an audio source’s spatial data is transformed into values that the Sample Processing stage will use to modify and output audio samples. The result of Interpretation is also called an interpretation. Interpretations vary in the type of data they hold as they are specifically for the target rendering format.

Interpretation does not interact with audio data, and Sample Processing does not directly interact with spatial data. Interpretations are intermediate data that sit between Interpretation and Sample Processing.

This separation of stages allows Interpretation to happen only when necessary. If an audio source’s properties are unchanged, it does not need to be re-interpreted (except in cases where the scene transforms relative to the listener, i.e. headtracking).