QuickTime format
use a compression codec (compressor/decompressor), an algorithm that reduces
file size by removing less important or duplicate information. Some codecs
compress spatially. This means in an area of flat color (like a sky or sea),
rather than recording all of the individual identical pixels, the codec treats
the area as one unit. Other codecs compress temporally by examining a series of
frames, recognizing the differences between them, and recording only that
information.
You to should
choose the value of your movies in five areas: codec, image size, frame rate,
data rate, and audio compressor.
Codec: The
Expert option offers a daunting range of codecs, from industry standard to
downright esoteric. However, 99 percent of you need only Sorenson, a temporal
compressor, and Cinepak, a spatial compressor. The Sorenson video codec
produces excellent Web and CD-ROM video that any Pentium or PowerMac can
download and play, provided it has QuickTime 3.0 or later. Compared with Cinepak,
Sorenson generally achieves higher image quality at a smaller file size.
However, it greater processing power, so if your movie will be viewed with
older computers that can't download QuickTime 3 and 4 - as unlikely as that may
be - you should probably compress in Cinepak.
Don't let the
daunting list intimidate you. Stick with small Sorenson or Cinepak, and
you’ll be just fine.
Frame Rate: A
frame is a still image, and frame rate is the number of per second (fps) in
your movie. North American video standards for 30 fps (actually 29.97 fps, to
be persnickety.) A Web movie can get by with 7.5 to 10 fps. An action video
requires a higher frame rate than a still interview.
Other Video
Compression Coding
JPEG
The Joint
Photographic Experts Group (JPEG) defines a set of compression for single
pictures. It is based on a lossy compression and reaches a ratio of 1:10 with a
fair quality. It also uses the run-length and Hoffman compression algorithms.
JPEG can be used for video sequences by compressing all frames. This is called
Motion-JPEG (M-JPEG) and ignores temporal redundancy.
H.261
This video
compression standard is compatible to PAL and NTSC and works on the video
standard YCrCb. It samples the Y component to 352 * 288 pixels according to the
Common Intermediate Format (CIF) or 176 * 144 pixels according to the Quarter
Common Intermediate Format (QCIF). The frame rate is 29.97 frames/s. H.261
makes use of interframe redundancy. It just codes the difference of the
predecessor frame and just sends in a certain interval a full frame.
H.263
Video
compression for low bit-rate communication. This is the enhancement of the
H.261 standard. The algorithms of H.261 had been fine tuned and enhanced so
that the compression increases by a factor 2 at the same quality. It works on
64 kbit/s bandwidth.
MPEG-1
MPEG-1 was
developed for digital video recording and playback. The standard works on a
quiet similar way like H.261. Except of the advantage searching forward and
backward in the frames. It is also a differential compression used. Each frame
has its unique sequence number and timestamp. This gives the ability of
receiving the frames in a different sequence that there are shown. It is
designed for a bandwidth of 1.5 Mbit/s and contains the audio standard MUSICAM.
The quality is comparable to VCR video quality.
MPEG-2
MPEG-2 is the
successor standard of MPEG-1. It covers digital TV (HDTV) standards and is
based on a 4 - 10 Mbit/s bandwidth. A major change to MPEG-1 is the ability of
working on interlaced frames. The sound supports quadro and surround options.