| |
CD
formats
Compact discs
come in all kinds of flavors. Some CDs are audio CDs, some are CD-ROMs,
some are PhotoCDs, some are VideoCDs, and so on. The differences between
these various formats lies in the types of data they contain; the way in
which that data is encoded, stored, and read; and the way in which the appropriate
playback hardware treats that data.
When the various
formats were originally defined, Phillips published the technical specifications
in a series of books which have geeky-sounding titles like " CD-DA
IEC-908," which happens to be the technical name for the audio CD standard.
Fortunately, since each spec book in the series has a differently-colored
cover, it's common practice to refer to various formats by the color of
the book in which it's defined. Of the different formats, the most common
(and, the ones you need to be most concerned about as an audio professional)
are:
- Red Book -- audio
CD (CD-DA)
- Yellow Book --
data CD (CD-ROM)
- Orange Book --
Recordable CD (CD-MO, CD-R, and CD-RW).
- Blue Book -- CD
Extra
This would be a good
time to clear up some of the mystery surrounding the term "Red Book."
There is nothing special or magical about a so-called Red Book CD. A
Red Book CD is an audio CD (CD-DA), period. When a software package
claims to allow you to "burn Red Book audio CDs with precision,"
or that it is "fully compliant with Red Book specifications,"
that is merely a fancy way of saying that you can make an audio CD with
it -- plain and simple! It doesn't mean anything more than that. It usually
also means that you can independently set your own track IDs (sometimes
referred to as PQ editing), but that's not necessarily the case.
Because of marketing mumbo-jumbo like this, I've seen beginners needlessly
confused into asking questions like "How do I convert an audio CD
into Red Book format?" or "I have software that makes great
audio CDs, but I'm not sure if it's Red Book compliant." Thanks to
this article, you are now armed with the knowledge necessary to see through
sales hype like this!
Anther source of confusion is in the use of the term "Orange Book,"
which defines the technical specifications of CD-R and CD-RW, including
various recording methods that can be used with them such as TAO, DAO,
multisession, UDF (a.k.a. packet writing), and so on (don't worry, we'll
define these terms in Part II of this article). It's important to remember
that the Orange Book specification refers to recordable CDs in their blank,
"unrecorded" state. The spec doesn't preclude the fact that
it can be recorded as a Red Book, Yellow Book, or other format-compliant
disc.
More than Meets the Ear
From the explanation given so far, it would seem that recording and playing
back a CD is a simple matter of "ones and zeros in, ones and zeros
out." If it were only that easy!
The fact is that over two-thirds of the ones and zeros on a CD
are not part of the original raw data! This is because the incoming
bits are subject to a whole series of encoding processes before they can
be stored on CD in a robustly readable form. The encoding method used
on audio CDs is called CIRC, which stands for Cross-Interleaved Reed-Solomon
Code. CIRC adds a whole bunch of extra (but vitally important) information
to the final recorded CD. This information contains bits which provide
relative and absolute timing information, track and index placement, emphasis
and copy-protection bits, synchronization data, error prevention and correction
data, and merge bits to keep it all straight. Not only is a ton of extra
information added, but it all gets scrambled and spread around before
it's laid into the data track -- which makes error recovery a lot more
reliable.
CD Encoding
CIRC encodes
and stores audio data in small chunks, each of which contains 24 bytes
of audio data. The incoming raw audio data stream is first parsed into
individual packets of 6 stereo samples each, which is of course 192 bits
(24 bytes) of audio (6 samples x 16 bits/sample x 2 channels = 192 bits/24
bytes). By the time the encoder does its thing, those 192 raw bits have
become 588 encoded bits. Each 588-bit chunk is called a frame. Of course,
the encoding process is extremely fast -- a four-speed (4x) CD recorder
can encode over 29,000 frames in a single second!
The following
describes the nature of the bits enclosed in a single frame of CD audio:
| Sync word: |
24 bits |
| Subcode signal: |
14 bits |
| Audio data: |
336 bits |
| Parity: |
112 bits |
| Merge bits: |
102 bits |
| Total bits: |
588 bits |
Lead-In and Lead-Out
Besides the extra information added in the encoding process, there is
a lead-in and a lead-out area on every audio CD which contains
important subcode information (we'll talk about subcodes in a bit). The
lead-in is an area at the beginning of the CD where subcode information
such as the number of tracks, track start and stop points, total CD time,
and so on can be readily accessed by the player. The lead-in takes up
about 9 megabytes on the CD. The lead-out contains subcode data
that indicates where the CD ends, and is about 13 megabytes long.
The neat thing about all of this stuff is that, for the purposes of recording
an audio CD, everything is pretty transparent to you, the user. The encoding
is automatically taken care of in the recording / mastering process, and
the decoding is handled by the player. You don't have to parse the data
and do any of the hocus-pocus yourself.
That being said, why is it important to know what's going on behind the
curtain? Because I said so, that's why! And because most audio engineers
would agree that it's a good idea to have at least a basic understanding
of how audio is physically stored on a CD, how error correction works,
what subcode channels are for, and what all those other extra bits do.
I've tried to make the simplified explanations that follow as non-geekoid
and as user-friendly as possible, so don't be afraid to dive in. As you'll
soon see, there is a method to all the madness!
Go
to Page 4; Back to TOP
|