Glossary - A Practical Guide to Video and Audio Compression - page 720

Topic Description

3ivX Widely available encoder based on MPEG-4. Stands alone but

offers very good performance and a quality product. It compares
favorably with other Windows- and QuickTime-based MPEG-4
encoders. The compression efficiency is good. This codec is
available for a variety of platforms including Windows, Mac, and
Linux. In fact, it is supported on several variants of Linux and
some TV set-top boxes as well.

4:4:4 Luma and chroma sampled at full resolution.

4:2:2 Luma sampled at four times the sub-carrier frequency. Chroma

sampled at half the resolution on the horizontal axis and down-
sampled in the vertical axis.

4:1:1 Double-sized aperture cell to allow higher compression ratio for

chroma. 8:2:2 is a more functional description of the sampling.

4:2:0 Only one chroma sample for the whole block of luma samples.

4 × 3 Normal aspect ratio for TV. This is gradually being replaced by

widescreen TV. This is sometimes called 12 × 9.

70mm Top-end cinematic film-presentation format. The stock is used

vertically for 70mm film and horizontally for VistaVision and
IMAX presentations.

8mm A popular format for home movie production that was made obsolete

by the introduction of home video cameras. Film comes in standard
and Super 8mm formats. Sony adopted 8mm tape as a video format.

9.5mm An early home movie format that was popular before the

development of the more compact 8mm systems.

AAC Advanced Audio Coding is a format for compressing audio that

was introduced with the MPEG-4 standard. An enhanced variant
is known as AAC+.

AAC+ (AAC Plus) A more advanced form of audio coding based on the existing

AAC coding plus Spectral Band Replication. Also known as High
Efficiency Advanced Audio Coding (HE-AAC).

AAFA Advanced Authoring Format (AAF) Association.

AC Coefficient The first coefficient of a DCT transform is the DC component.

This is a constant average value across the whole macroblock. All
of the other coefficients are variable according to the level of detail
in the block. The frequency of the detail increases from left to right
and top to bottom. So this is the part of the DCT output that
describes the image detail in frequency terms and is non-zero if
the image in the block is not a solid area of the same color. The
values of the coefficients get smaller towards the right bottom
corner as the level of detail falls off.