Controlling Dynamic Range
Dynamic range compression is a technique used to raise the level of soft sounds and lower the level of louder sounds, effectively reducing the loudness "swing" of an audio signal.
Classically, this has been done prior to transmission, but the Dolby Digital (AC-3) system uses a unique metadata approach where the gain reduction values are calculated during the encoding process but are not applied to the audio until after decoding. This means that the original audio can theoretically be passed untouched to the consumer who can then choose to listen to it with restricted, less restricted, or fully unrestricted dynamic range.
Setting the dynamic range control, or DRC, is as simple as picking from one of the five profiles included in all encoders: Film Standard, Film Light, Music Standard, Music Light, Speech, and none. Yes, that is actually six, but none does not count, right? Wrong. "None" doesn't necessarily mean that no dynamic range control will be generated. If the dialogue loudness parameter is not set correctly, protection DRC values will be generated to protect the two channel and mono listeners. If audio is downmixed, it could be possible to cause digital clipping when the channels are added together, and protection DRC values are generated to protect against this.
The DRC system only has a chance of working if all parameters are set correctly and the system is used as intended. While this is possible, it remains to be seen if it is truly practical.