3. Lossy File Coding Formats
Objectives:
- Identify some of the lossy formats currently in use
SOME LOSSY CODING SYSTEMS
MPEG
The Moving Pictures Expert Group formed in 1988 to devise
compression techniques for audio and video.
there have been 2 standards produced. The standards
define three "layers" of coding, each with different applications.
MPEG-1
The standard was completed in 1992.
Specifically developed to support audio and video coding
for playback within the CD's bandwidth of 1.41 Mbps.
Supports encoding of 32, 44.1, or 48 kHz at rates of
64 - 448 kbps for stereo channels.
Rates < 256k are used when full screen motion quality
is critical; rates > 256k are used for higher audio quality but partial
video images.
MPEG-1 Layer 3 (MP3)
- Most common form of "lossy" compression
- File Structure
- Files are divided into "frames"; each frame contains a 32 bit metadata header, and the audio data
- Constant Bit Rate: all frames are the same size
- Variable Bit Rate: frames containing lower frequencies use less data than other frames
- Contains ID3 tags for metadata
- History of the mp3: http://www.iis.fraunhofer.de/amm/techinf/layer3/
MPEG-2
- Can handle multichannel audio
- Compression ratio not as high as MP3
MPEG-4 / Advanced Audio Coding (MP4, AAC)
- AAC is the codec that is used under the MPEG-4 standard
http://www.apple.com/quicktime/technologies/aac/
- Newer than MP3; most popular "new
generation' lossy coding scheme
- Very broad audio/video standard; designed for multimedia, not just music
- Allegedly produces higher quality than MP3 given a compression ratio
- More advanced copy protection
Dolby Digital AC-3
- Dolby Labs system for multichannel audio. The data
is stored in the space between sprocket holes on a film. The space can only
accommodate a limited amount of data - much less than needed for six channels
of audio.
- AC-3 will encode 48K sampling rates with bit depth
as high as 24 bits.
- Using AC-3 lossy coding, 382 Mb/s reduced to
320 kb/s - sufficient for storing on the film.
- Supports multichannel audio as well as mono and stereo
(up to 5.1)
- Decoders downmix channels when necessary; downmix
is not stored with multichannel audio
http://www.dolby.com/consumer/technology/dolby_live.html
DTS
- DTS is an alternative coding system for multichannel
audio for motion pictures. The data is stored on an accompanying CD that
is time-locked to the film.
- DTS encodes up to 8 channels of audio ranging from
8 to 192kHz, bit depth of 16 to 24 bits, using a nominal bit rate of 1.4
Mbps ( the rate for standard CDs).
In-depth articles:
Compatibility:
- Virtually all current DAWs can decode MP3s, allowing for quick import (ex. ProTools)
- MP4 & AAC not yet supported in most DAWS without the addition of a decoding plug-in