Upgrading & Repairing PCs: Optical Storage

This chapter is from the book 

CD-Based Optical Technology

The first type of optical storage that became a widespread computing standard is the CD-ROM. CD-ROM, or compact disc read-only memory, is an optical read-only storage medium based on the original CD-DA (digital audio) format first developed for audio CDs. Other formats, such as CD-R (CD-recordable) and CD-RW (CD-rewritable), expanded the compact disc's capabilities by making it writable.

Older CD-ROM discs held 74 minutes of high-fidelity audio in CD audio format or 650MiB (682MB) of data. However, the current CD-ROM standard is an 80-minute disc with a data capacity of 700MiB (737MB). When MP3, WMA, or similar compressed audio files are stored on CD, several hours of audio can be stored on a single disc (depending on the compression format and bit rate used). Music only, data only, or a combination of music and data (Enhanced CD) can be stored on one side (only the bottom is used) of a 120mm (4.72-inch) diameter, 1.2mm (0.047-inch) thick plastic disc.

CD-ROM has the same form factor (physical shape and layout) of the familiar CD-DA audio compact disc and can, in fact, be inserted into a normal audio player. Sometimes it isn't playable, though, because the player reads the subcode information for the track, which indicates that it is data and not audio. If it could be played, the result would be noise—unless audio tracks precede the data on the CD-ROM. (See the section "Blue Book—CD EXTRA," later in this chapter.)

Accessing data from a CD using a computer is much faster than from a floppy disk but slower than a modern hard drive.

CDs: A Brief History

In 1979, the Philips and Sony corporations joined forces to coproduce the CD-DA (Compact Disc-Digital Audio) standard. Philips had already developed commercial laserdisc players, and Sony had a decade of digital recording research under its belt. The two companies were poised for a battle—the introduction of potentially incompatible audio laser disc formats—when instead they came to terms on an agreement to formulate a single industry-standard digital audio technology.

Philips contributed most of the physical design, which was similar to the laserdisc format it had previously created with regards to using pits and lands on the disk that are read by a laser. Sony contributed the digital-to-analog circuitry, and especially the digital encoding and error-correction code designs.

In 1980, the companies announced the CD-DA standard, which has since been referred to as the Red Book format (so named because the cover of the published document was red). The Red Book included the specifications for recording, sampling, and—above all—the 120mm (4.72-inch) diameter physical format you live with today. This size was chosen, legend has it, because it could contain all of Beethoven's approximately 70-minute Ninth Symphony without interruption, compared to 23 minutes per side of the then-mainstream 33-rpm LP record.

After the specification was set, both manufacturers were in a race to introduce the first commercially available CD audio drive. Because of its greater experience with digital electronics, Sony won that race and beat Philips to market by one month, when on October 1, 1982 Sony introduced the CDP-101 player and the world's first commercial CD recording—Billy Joel's 52nd Street album. The player was introduced in Japan and then Europe; it wasn't available in the United States until early 1983. In 1984, Sony also introduced the first automobile and portable CD players.

Sony and Philips continued to collaborate on CD standards throughout the decade, and in 1983 they jointly released the Yellow Book CD-ROM standard. It turned the CD from a digital audio storage medium to one that could now store read-only data for use with a computer. The Yellow Book used the same physical format as audio CDs but modified the decoding electronics to allow data to be stored reliably. In fact, all subsequent CD standards (usually referred to by their colored book binders) have referred to the original Red Book standard for the physical parameters of the disc. With the advent of the Yellow Book standard (CD-ROM), what originally was designed to hold a symphony could now be used to hold practically any type of information or software.

For more information on the other CD book formats, see the section "CD Formats," later in this chapter.

CD Construction and Technology

A CD is made of a polycarbonate wafer, 120mm in diameter and 1.2mm thick, with a 15mm hole in the center. This wafer base is stamped or molded with a single physical track in a spiral configuration starting from the inside of the disc and spiraling outward. The track has a pitch, or spiral separation, of 1.6 microns (millionths of a meter, or thousandths of a millimeter). By comparison, an LP record has a physical track pitch of about 125 microns. When viewed from the reading side (the bottom), the disc rotates counterclockwise. If you examined the spiral track under a microscope, you would see that along the track are raised bumps, called pits, and flat areas between the pits, called lands. It seems strange to call a raised bump a pit, but that is because when the discs are pressed, the stamper works from the top side. So, from that perspective, the pits are actually depressions made in the plastic.

The laser used to read the disc would pass right through the clear plastic, so the stamped surface is coated with a reflective layer of metal (usually aluminum) to make it reflective. Then the aluminum is coated with a thin protective layer of acrylic lacquer, and finally a label or printing is added.

Mass-Producing CDs

Commercial mass-produced optical discs are stamped or pressed and not burned by a laser as many people believe (see Figure 11.1). Although a laser is used to etch data onto a glass master disc that has been coated with a photosensitive material, using a laser to directly burn discs would be impractical for the reproduction of hundreds or thousands of copies.

The steps in manufacturing CDs are as follows. (Use Figure 11.1 as a visual.)

1. Photoresist coating—A circular 240mm diameter piece of polished glass 6mm thick is spin-coated with a photoresist layer about 150 microns thick and then hardened by baking at 80°C (176°F) for 30 minutes.

2. Laser recording—A Laser Beam Recorder (LBR) fires pulses of blue/violet laser light to expose and soften portions of the photoresist layer on the glass master.

3. Master development—A sodium hydroxide solution is spun over the exposed glass master, which then dissolves the areas exposed to the laser, thus etching pits in the photoresist.

4. Electroforming—The developed master is then coated with a layer of nickel alloy through a process called electroforming. This creates a metal master called a father.

5. Master separation—The metal master father is then separated from the glass master. The father is a metal master that can be used to stamp discs, and for short runs, it may in fact be used that way. However, because the glass master is damaged when the father is separated, and because a stamper can produce only a limited number of discs before it wears out, the father often is electroformed to create several reverse image mothers. These mothers are then subsequently electroformed to create the actual stampers. This enables many more discs to be stamped without ever having to go through the glass mastering process again.

6. Disc-stamping operation—A metal stamper is used in an injection molding machine to press the data image (pits and lands) into approximately 18 grams of molten (350°C or 662°F) polycarbonate plastic with a force of about 20,000psi. Normally, one disc can be pressed every 2–3 seconds in a modern stamping machine.

7. Metalization—The clear stamped disc base is then sputter-coated with a thin (0.05–0.1 micron) layer of aluminum to make the surface reflective.

8. Protective coating—The metalized disc is then spin-coated with a thin (6–7 micron) layer of acrylic lacquer, which is then cured with UV (ultraviolet) light. This protects the aluminum from oxidation.

9. Finished product—Finally, a label is affixed or printing is screen-printed on the disc and cured with UV light.

Although the manufacturing process shown here was for CDs, the process is almost identical for other types of optical media.

Pits and Lands

Reading the information back from a disc is a matter of bouncing a low-powered laser beam off the reflective layer in the disc. The laser shines a focused beam on the underside of the disc, and a photosensitive receptor detects when the light is reflected back. When the light hits a land (flat spot) on the track, the light is reflected back; however, when the light hits a pit (raised bump), no light is reflected back.

As the disc rotates over the laser and receptor, the laser shines continuously while the receptor sees what is essentially a pattern of flashing light as the laser passes over pits and lands. Each time the laser passes over the edge of a pit, the light seen by the receptor changes in state from being reflected to not reflected, or vice versa. Each change in state of reflection caused by crossing the edge of a pit is translated into a 1 bit digitally. Microprocessors in the drive translate the light/dark and dark/light (pit edge) transitions into 1 bits, translate areas with no transitions into 0 bits, and then translate the bit patterns into actual data or sound.

The individual pits on a CD are 0.125 microns deep and 0.6 microns wide. Both the pits and lands vary in length from about 0.9 microns at their shortest to about 3.3 microns at their longest. The track is a spiral with 1.6 microns between adjacent turns (see Figure 11.2).

The height of the pits above the land is especially critical because it relates to the wavelength of the laser light used when reading the disc. The pit (bump) height is exactly 1/4 of the wavelength of the laser light used to read the disc. Therefore, the light striking a land travels 1/2 of a wavelength of light farther than light striking the top of a pit (1/4 + 1/4 = 1/2). This means the light reflected from a pit is 1/2 wavelength out of phase with the rest of the light being reflected from the disc. The out-of-phase waves cancel each other out, dramatically reducing the light that is reflected back and making the pit appear dark even though it is coated with the same reflective aluminum as the lands.

The read laser in a CD drive is a 780nm (nanometer) wavelength laser of about 1 milliwatt in power. The polycarbonate plastic used in the disc has a refractive index of 1.55, so light travels through the plastic 1.55 times more slowly than through the air around it. Because the frequency of the light passing through the plastic remains the same, this has the effect of shortening the wavelength inside the plastic by the same factor. Therefore, the 780nm light waves are now compressed to 500nm (780/1.55). One quarter of 500nm is 125nm, which is 0.125 microns—the specified height of the pit.

Drive Mechanical Operation

An optical drive operates by using a laser to reflect light off the bottom of the disc. A photo detector then reads the reflected light. The overall operation of an optical drive is as follows (see Figure 11.3):

1. The laser diode emits a low-energy infrared beam toward a reflecting mirror.
2. The servo motor, on command from the microprocessor, positions the beam onto the correct track on the disc by moving the reflecting mirror.
3. When the beam hits the disc, its refracted light is gathered and focused through the first lens beneath the platter, bounced off the mirror, and sent toward the beam splitter.
4. The beam splitter directs the returning laser light toward another focusing lens.
5. The last lens directs the light beam to a photo detector that converts the light into electric impulses.
6. These incoming impulses are decoded by the microprocessor and sent along to the host computer as data.

When introduced, CD-ROM drives were too expensive for widespread adoption. After the production costs of both drives and discs began to drop, however, CDs were rapidly assimilated into the PC world. This was particularly due to the ever-expanding size of PC applications. Virtually all software is now supplied on optical media, even if the disc doesn't contain data representing a tenth of its potential capacity.

Tracks and Sectors

On the traditional 74-minute CD, the pits are stamped into a single spiral track with a spacing of 1.6 microns between turns, corresponding to a track density of 625 turns per millimeter, or 15,875 turns per inch. This equates to a total of 22,188 turns for a typical 74-minute (650MiB) disc. Current 80-minute CDs gain their extra capacity by decreasing the spacing between turns. See Table 11.1 for more information about the differences between 74-minute and 80-minute CDs.

Table 11.1. CD Technical Parameters

Advertised CD capacity (MiB)	650	700
1x read speed (m/sec)	1.3	1.3
Laser wavelength (nm)	780	780
Numerical aperture (lens)	0.45	0.45
Media refractive index	1.55	1.55
Track (turn) spacing (um)	1.6	1.48
Turns per mm	625	676
Turns per inch	15,875	17,162
Total track length (m)	5,772	6,240
Total track length (feet)	18,937	20,472
Total track length (miles)	3.59	3.88
Pit width (um)	0.6	0.6
Pit depth (um)	0.125	0.125
Min. nominal pit length (um)	0.90	0.90
Max. nominal pit length (um)	3.31	3.31
Lead-in inner radius (mm)	23	23
Data zone inner radius (mm)	25	25
Data zone outer radius (mm)	58	58
Lead-out outer radius (mm)	58.5	58.5
Data zone width (mm)	33	33
Total track area width (mm)	35.5	35.5
Max. rotating speed 1x CLV (rpm)	540	540
Min. rotating speed 1x CLV (rpm)	212	212
Track revolutions (data zone)	20,625	22,297
Track revolutions (total)	22,188	23,986
B = Byte (8 bits) KB = Kilobyte (1,000 bytes) KiB = Kibibyte (1,024 bytes) MB = Megabyte (1,000,000 bytes) MiB = Mebibyte (1,048,576 bytes) m = Meters mm = Millimeters (thousandths of a meter) um = Micrometers = Microns (millionths of a meter) CLV = Constant linear velocity rpm = Revolutions per minute

The disc is divided into six main areas (discussed here and shown in Figure 11.4):

• Hub clamping area—The hub clamp area is just that: a part of the disc where the hub mechanism in the drive can grip the disc. No data or information is stored in that area.
• Power calibration area (PCA)—This is found only on writable discs and is used only by recordable drives to determine the laser power necessary to perform an optimum burn. A single CD-R or CD-RW disc can be tested this way up to 99 times.
• Program memory area (PMA)—This is found only on writable discs and is the area where the TOC (table of contents) is temporarily written until a recording session is closed. After the session is closed, the TOC information is written to the lead-in area.
• Lead-in—The lead-in area contains the disc (or session) TOC in the Q subcode channel. The TOC contains the start addresses and lengths of all tracks (songs or data), the total length of the program (data) area, and information about the individual recorded sessions. A single lead-in area exists on a disc recorded all at once (Disc At Once or DAO mode), or a lead-in area starts each session on a multisession disc. The lead-in takes up 4,500 sectors on the disc (1 minute if measured in time, or about 9.2MB worth of data). The lead-in also indicates whether the disc is multisession and what the next writable address on the disc is (if the disc isn't closed).
• Program (data) area—This area of the disc starts at a radius of 25mm from the center.
• Lead-out—The lead-out marks the end of the program (data) area or the end of the recording session on a multisession disc. No actual data is written in the lead-out; it is simply a marker. The first lead-out on a disc (or the only one if it is a single session or Disk At Once recording) is 6,750 sectors long (1.5 minutes if measured in time, or about 13.8MB worth of data). If the disc is a multisession disc, any subsequent lead-outs are 2,250 sectors long (0.5 minutes in time, or about 4.6MB worth of data).

The hub clamp, lead-in, program, and lead-out areas are found on all CDs, whereas only recordable CDs (such as CD-Rs and CD-RWs) have the additional power calibration area and program memory area at the start of the disc.

Figure 11.4 shows these areas in actual relative scale as they appear on a disc.

Officially, the spiral track of a standard CD starts with the lead-in area and ends at the finish of the lead-out area, which is 58.5mm from the center of the disc, or 1.5mm from the outer edge. This single spiral track is about 5.77 kilometers, or 3.59 miles, long. An interesting fact is that in a 56x CAV (constant angular velocity) drive, when the outer part of the track is being read, the data moves at an actual speed of 162.8 miles per hour (262km/h) past the laser. What is more amazing is that even when the data is traveling at that speed, the laser pickup can accurately read bits (pit/land transitions) spaced as little as only 0.9 microns (or 35.4 millionths of an inch) apart!

Table 11.1 shows some of the basic information about the two main CD capacities, which are 74 and 80 minutes. The CD standard originally was created around the 74-minute disc; the 80-minute versions were added later and basically stretch the standard by tightening the track spacing within the limitations of the original specification. A poorly performing or worn-out drive can have trouble reading the 80-minute discs.

The spiral track is divided into sectors that are stored at the rate of 75 sectors per second. On a disc that can hold a total of 74 minutes of information, that results in a maximum of 333,000 sectors. Each sector is then divided into 98 individual frames of information. Each frame contains 33 bytes: 24 bytes are audio data, 1 byte contains subcode information, and 8 bytes are used for parity/ECC (error correction code) information. Table 11.2 shows the sector, frame, and audio data calculations.

Table 11.2. CD Sector, Frame, and Audio Data Information

Advertised CD length (minutes)	74	80
Sectors/second	75	75
Frames/sector	98	98
Number of sectors	333,000	360,000
Sector length (mm)	17.33	17.33
Byte length (um)	5.36	5.36
Bit length (um)	0.67	0.67
Each Frame:
Subcode bytes	1	1
Data bytes	24	24
Q+P parity bytes	8	8
Total bytes/frame	33	33
Audio Data:
Audio sampling rate (Hz)	44,100	44,100
Samples per Hz (stereo)	2	2
Sample size (bytes)	2	2
Audio bytes per second	176,400	176,400
Sectors per second	75	75
Audio bytes per sector	2,352	2,352
Each Audio Sector (98 Frames):
Q+P parity bytes	784	784
Subcode bytes	98	98
Audio data bytes	2,352	2,352
Bytes/sector RAW (unencoded)	3,234	3,234
Hz = Hertz (cycles per second) mm = Millimeters (thousandths of a meter) um = Micrometers = Microns (millionths of a meter)

Sampling

When music is recorded on a CD, it is sampled at a rate of 44,100 times per second (Hz). Each music sample has a separate left and right channel (stereo) component, and each channel component is digitally converted into a 16-bit number. This allows for a resolution of 65,536 possible values, which represents the amplitude of the sound wave for that channel at that moment.

The sampling rate determines the range of audio frequencies that can be represented in the digital recording. The more samples of a wave that are taken per second, the closer the sampled result will be to the original. The Nyquist theorem (originally published by American physicist Harry Nyquist in 1928) states that the sampling rate must be at least twice the highest frequency present in the sample to reconstruct the original signal accurately. That explains why Philips and Sony intentionally chose the 44,100Hz sampling rate when developing the CD—that rate could be used to accurately reproduce sounds of up to 20,000Hz, which is the upper limit of human hearing.

Subcodes

Subcode bytes enable the drive to find songs (which are confusingly also called tracks) along the spiral track and contain or convey additional information about the disc in general. The subcode bytes are stored as 1 byte per frame, which results in 98 subcode bytes for each sector. Two of these bytes are used as start block and end block markers, leaving 96 bytes of subcode information. These are then divided into eight 12-byte subcode blocks, each of which is assigned a letter designation P-W. Each subcode channel can hold about 31.97MB of data across the disc, which is about 4% of the capacity of an audio disc. The interesting thing about the subcodes is that the data is woven continuously throughout the disc; in other words, subcode data is contained piecemeal in every sector on the disc.

The P and Q subcode blocks are used on all discs, and the R-W subcodes are used only on CD+G (graphics) or CD TEXT-type discs.

The P subcode identifies the start of the tracks on the CD. The Q subcode contains a multitude of information, including the following:

• Whether the sector data is audio or data. This prevents most players from trying to "play" CD data discs, which might damage speakers due to the resulting noise that would occur.
• Whether the audio data is two or four channel. Four channel is rarely if ever used.
• Whether digital copying is permitted. PC-based CD-R and RW drives ignore this; it was instituted to prevent copying to DAT (digital audio tape) or home audio optical drives.
• Whether the music is recorded with pre-emphasis. This is a hiss or noise reduction technique.
• The track (song) layout on the disc.
• The track (song) number.
• The minutes, seconds, and frame number from the start of the track (song).
• A countdown during an intertrack (intersong) pause.
• The minutes, seconds, and frames from the start of the first track (song).
• The barcode of the CD.
• The ISRC (International Standard Recording Code). This is unique to each track (song) on the disc.

The R-W subcodes are used on CD+G (graphics) discs to contain graphics and text. This enables a limited amount of graphics and text to be displayed while the music is being played. The most common use for CD+G media is karaoke "sing-along" media. These same subcodes are used on CD TEXT discs to store disc- and track-related information that is added to standard audio CDs for playback on compatible CD audio players. The CD TEXT information is stored as ASCII characters in the R-W channels in the lead-in and program areas of a CD. On a CD TEXT disc, the lead-in area subcodes contain text information about the entire disc, such as the album, track (song) titles, and artist names. The program area subcodes, on the other hand, contain text information for the current track (song), including track title, composer, performers, and so on. The CD TEXT data is repeated throughout each track to reduce the delay in retrieving the data.

CD TEXT–compatible players typically have a text display to show this information, ranging from a simple one- or two-line, 20-character display, such as on many newer RBDS (radio broadcast data system) automobile radio/CD players, to up to 21 lines of 40-color, alphanumeric or graphics characters on home- or computer-based players. The specification also allows for future additional data, such as Joint Photographic Experts Group (JPEG) images. Interactive menus also can be used for the selection of text for display.

Handling Read Errors

Handling errors when reading a disc was a big part of the original Red Book CD standard. CDs use parity and interleaving techniques called cross-interleave Reed-Solomon code (CIRC) to minimize the effects of errors on the disk. This works at the frame level. When being stored, the 24 data bytes in each frame are first run through a Reed-Solomon encoder to produce a 4-byte parity code called "Q" parity, which then is added to the 24 data bytes. The resulting 28 bytes are then run though another encoder that uses a different scheme to produce an additional 4-byte parity value called "P" parity. These are added to the 28 bytes from the previous encoding, resulting in 32 bytes (24 of the original data plus the Q and P parity bytes). An additional byte of subcode (tracking) information is then added, resulting in 33 bytes total for each frame. Note that the P and Q parity bytes are not related to the P and Q subcodes mentioned earlier.

To learn more about the concepts behind parity and error correction, which were originally used to guard against errors in memory and modem communications, see Chapter 6, "Memory," p. 345, and Chapter 16, "Internet Connectivity," p. 780.

To minimize the effects of a scratch or physical defect that would damage adjacent frames, several interleaves are added before the frames are actually written. Parts of 109 frames are cross-interleaved (stored in different frames and sectors) using delay lines. This scrambling decreases the likelihood of a scratch or defect affecting adjacent data because the data is actually written out of sequence.

With CDs, the CIRC scheme can correct errors up to 3,874 bits long (which would be 2.6mm in track length). In addition, for audio CDs, only the CIRC can also conceal (through interpolation) errors up to 13,282 bits long (8.9mm in track length). Interpolation is the process in which the data is estimated or averaged to restore what is missing. That would, of course, be unacceptable on a data CD, so this applies only to audio discs. The Red Book CD standard defines the block error rate (BLER) as the number of frames (98 per sector) per second that have any bad bits (averaged over 10 seconds) and requires that this be less than 220. This allows a maximum of up to about 3% of the frames to have errors, and yet the disc will still be functional.

An additional layer of error-detection and -correction circuitry is the key difference between audio CD players and data CD drives. Audio CDs convert the digital information stored on the disc into analog signals for a stereo amplifier to process. In this scheme, some imprecision is acceptable because it would be virtually impossible to hear in the music. Data CDs, however, can't tolerate imprecision. Each bit of data must be read accurately. For this reason, data CDs have a great deal of additional ECC information written to the disc along with the actual stored information. The ECC can detect and correct most minor errors, improving the reliability and precision to levels that are acceptable for data storage.

In the case of an audio CD, missing data can be interpolated—that is, the information follows a predictable pattern that enables the drive to guess the missing values. For example, if three values are stored on an audio disc (say, 10, 13, and 20 appearing in a series), and the middle value is missing—because of damage or dirt on the CD's surface—you could interpolate a middle value of 15, which is midway between 10 and 20. Although this might not be exactly correct, in the case of audio recording, it probably won't be noticeable to the listener. If those same three values appear on a data CD in an executable program, there is no way to guess at the correct value for the middle sample. Interpolation can't work because executable program instructions or data must be exact; otherwise, the program will crash or improperly read data needed for a calculation. Using the previous example with a data CD running an executable program, guessing 15 is not merely slightly off—it is completely wrong.

In a CD on which data is stored instead of audio information, additional information is added to each sector to detect and correct errors as well as to identify the location of data sectors more accurately. To accomplish this, 304 bytes are taken from the 2,352 that originally were used for audio data and are instead used for sync (synchronizing bits), ID (identification bits), ECC, and EDC information. This leaves 2,048 bytes for actual user data in each sector. Just as when reading an audio CD, on a 1x (standard speed) CD, sectors are read at a constant speed of 75 per second. This results in a standard CD transfer rate of 153,600 bytes per second (2,048x75), which is expressed as either 153.6KBps or 150KiBps.

CD Capacity

Each second of a CD contains 75 blocks of data containing 2,048 bytes per block. From this information, you can calculate the absolute maximum storage capacity of an 80-minute or 74-minute CD, as shown in Table 11.3. The table also shows the structure and layout of each sector on a CD on which data is stored.

Table 11.3. CD Sector Information and Capacity

Each Data Sector (Mode 1):	74-Minute	80-Minute
Each Data Sector (Mode 1):	74-Minute	80-Minute
Q+P parity bytes	784	784
Subcode bytes	98	98
Sync bytes	12	12
Header bytes	8	8
ECC/EDC bytes	284	284
Data bytes	2,048	2,048
Bytes/sector RAW (unencoded)	3,234	3,234
Actual CD Data Capacity:
B	681,984,000	737,280,000
KiB	666,000	720,000
KB	681,984	737,280
MiB	650.39	703.13
MB	681.98	737.28
B = Byte (8 bits) KB = Kilobyte (1,000 bytes) KiB = Kibibyte (1,024 bytes) MB = Megabyte (1,000,000 bytes) MiB = Mebibyte (1,048,576 bytes) ECC = Error correction code EDC = Error detection code

This information assumes the data is stored in Mode 1 format, which is used on virtually all data discs. You can learn more about the Mode 1/Mode 2 formats in the section on the Yellow Book and XA standards later in this chapter.

With data sectors, you can see that out of 3,234 actual bytes per sector, only 2,048 are user data. Most of the other 1,186 bytes are used for the intensive error-detection and -correction schemes to ensure error-free performance.

Data Encoding on the Disc

The final part of how data is actually written to the CD is very interesting. After all 98 frames are composed for a sector (whether audio or data), the information is then run through a final encoding process called eight to fourteen modulation (EFM). This scheme takes each byte (8 bits) and converts it into a 14-bit value for storage. The 14-bit conversion codes are designed so that there are never fewer than two or more than ten adjacent 0 bits. This is a form of Run Length Limited (RLL) encoding called RLL 2,10 (RLL x,y, where x equals the minimum and y equals the maximum run of 0s). This is designed to prevent long strings of 0s, which could more easily be misread, as well as to limit the minimum and maximum frequency of transitions actually placed on the recording media. With as few as two or as many as ten 0 bits separating 1 bits in the recording, the minimum distance between 1s is 3 bit time intervals (usually referred to as 3T), and the maximum spacing between 1s is 11 time intervals (11T).

Because some of the EFM codes start and end with a 1 or more than five 0s, three additional bits called merge bits are added between each 14-bit EFM value written to the disc. The merge bits usually are 0s but might contain a 1 if necessary to break a long string of adjacent 0s formed by the adjacent 14-bit EFM values. In addition to the now 17 bits created for each byte (EFM plus merge bits), a 24-bit sync word (plus three more merge bits) is added to the beginning of each frame. This results in a total of 588 bits (73.5 bytes) actually being stored on the disc for each frame. Multiply this for 98 frames per sector and you have 7,203 bytes actually being stored on the disc to represent each sector. An 80-minute disc, therefore, really has something like 2.6GB of actual data being written, which, after being fully decoded and stripped of error-correcting codes and other information, results in about 737MB (703MiB) of actual user data.

The calculations for EFM-encoded frames and sectors are shown in Table 11.4.

Table 11.4. EFM-Encoded Data Calculations

EFM-Encoded Frames:	74-Minute	80-Minute
Sync word bits	24	24
Subcode bits	14	14
Data bits	336	336
Q+P parity bits	112	112
Merge bits	102	102
EFM bits per frame	588	588
EFM-Encoded Sectors:
EFM bits per sector	57,624	57,624
EFM bytes per sector	7,203	7,203
Total EFM data on disc (MB)	2,399	2,593
B = Byte (8 bits) KB = Kilobyte (1,000 bytes) KiB = Kibibyte (1,024 bytes) MB = Megabyte (1,000,000 bytes) MiB = Mebibyte (1,048,576 bytes) EFM = Eight to fourteen modulation

To put this into perspective, see Table 11.5 for an example of how familiar data would actually be encoded when written to a CD. As an example, I'll use the letters N and O as they would be written on the disc.

Table 11.5. EFM Data Encoding on a CD

Character	N	O
ASCII decimal code	78	79
ASCII hexadecimal code	4E	4F
ASCII binary code	01001110	01001111
EFM code	00010001000100	00100001000100
ASCII = American Standard Code for Information Interchange EFM = Eight to fourteen modulation

Figure 11.5 shows how the encoded data would actually appear as pits and lands stamped into a CD.

The edges of the pits are translated into the binary 1 bits. As you can see, each 14-bit grouping represents a byte of actual EFM-encoded data on the disc, and each 14-bit EFM code is separated by three merge bits (all 0s in this example). The three pits produced by this example are 4T (4 transitions), 8T, and 4T long. The string of 1s and 0s on the top of the figure represents how the actual data would be read; note that a 1 is read wherever a pit-to-land transition occurs. It is interesting to note that this drawing is actually to scale, meaning the pits (raised bumps) would be about that long and wide relative to each other. If you could use a microscope to view the disc, this is what the word "NO" would look like as actually recorded.

Writable CDs

Optical disc recording has come a long way since 1988, when the first CD-R recording system was introduced at the cost of $50,000 (back then, they used a $35,000 Yamaha audio recording drive along with thousands of dollars of additional error correction and other circuitry for CD-ROM use), operated at 1x speed only, and was part of a subsystem that was the size of a washing machine! The blank discs also cost about $100 each—compared to less than 5 cents each in bulk cakebox form. (You provide your own jewel or slimline cases.) Originally, the main purpose for CD recording was to produce prototype CDs that could then be replicated via the standard stamping process.

In 1991, Philips introduced the first 2x recorder (the CDD 521), which was about the size of a stereo receiver and cost about $12,000. Sony in 1992 and then JVC in 1993 followed with their 2x recorders, and the JVC was the first drive that had the half-height 5 1/4-inch form factor that most desktop system drives still use today. In 1995, Yamaha released the first 4x recorder (the CDR100), which sold for $5,000. A breakthrough in pricing came in late 1995 when Hewlett-Packard released a 2x recorder (the 4020i, which was actually made for them by Philips) for less than $1,000. This proved to be exactly what the market was waiting for. With a surge in popularity after that, prices rapidly fell to below $500, and then down to $200 or less. In 1996, Ricoh introduced the first CD-RW drive.

Two main types of recordable CD discs are available, called CD-R (recordable) and CD-RW (rewritable). However, because the CD-RW discs are more expensive than CD-R discs, only half as fast (or less) as CD-R discs, and won't work in all CD audio or CD-ROM drives, people usually use CD-R media instead of CD-RW.

CD-R media is a WORM (write once, read many) media, meaning that after you fill a CD-R with data, it is permanently stored and can't be erased. The write-once limitation makes this type of disc less than ideal for system backups or other purposes in which it would be preferable to reuse the same media over and over. However, because of the low cost of CD-R media, you might find that making permanent backups to essentially disposable CD-R discs is as economically feasible as tape or other media.

CD-RW discs can be reused up to 1,000 times, making them suitable for almost any type of data storage task. The following sections examine these two standards and how you can use them for your own data storage needs.

CD-R

Once recorded, CD-R discs can be played back or read in any standard CD drive. CD-R discs are useful for archival storage and creating master CDs, which can be duplicated for distribution within a company.

CD-Rs function using the same principle as standard CD-ROMs.The main difference is that instead of being stamped or embossed into plastic as on regular CDs, CD-Rs have images of pits burned onto a raised groove instead. Therefore, the pits are not really raised bumps like on a standard CD, but instead are rendered as dark (burned) areas on the groove that reflect less light. Because the overall reflectivity of pit and land areas remains the same as on a stamped disc, normal CD drives can read CD-Rs exactly as if they were stamped discs.

Part of the recording process with CD-Rs starts before you even insert the disc into the drive. CD-R media is manufactured much like a standard CD—a stamper is used to mold a base of polycarbonate plastic. However, instead of stamping pits and lands, the stamper imprints a spiral groove (called a pre-groove) into the disc. From the perspective of the reading (and writing) laser underneath the disc, this groove is seen as a raised spiral ridge and not a depression.

The pre-groove (or ridge) is not perfectly straight; instead it has a slight wobble. The amplitude of the wobble is generally very small compared to the track pitch (spacing). The groove separation is 1.6 microns, but it wobbles only 0.030 microns from side to side. The wobble of a CD-R groove is modulated to carry supplemental information read by the drive. The signal contained in the wobble is called absolute time in pre-groove (ATIP) because it is modulated with time code and other data. The time code is the same minutes:seconds:frame format that will eventually be found in the Q-subcode of the frames after they are written to the disc. The ATIP enables the drive to locate positions on the disc before the frames are actually written. Technically, the wobble signal is frequency shift-keyed with a carrier frequency of 22.05KHz and a deviation of 1KHz. The wobble uses changes in frequency to carry information.

To complete the CD-R disc, an organic dye is evenly applied across the disc by a spin-coating process. Next, a gold or silver reflective layer is applied (some early low-cost media used aluminum), followed by a protective coat of UV-cured lacquer to protect the reflective and dye layers. Gold or silver is used in recent and current CD-R discs to get the reflectivity as high as possible (gold is used in archival CD-Rs designed for very long-term storage), and it was found that the organic dye tends to oxidize aluminum. Then, silk-screen printing is applied on top of the lacquer for identification and further protection. When seen from the underside, the laser used to read (or write) the disc first passes through the clear polycarbonate and the dye layer, hits the gold layer where it is reflected back through the dye layer and the plastic, and finally is picked up by the optical pickup sensor in the drive.

The dye and reflective layer together have the same reflective properties as a virgin CD. In other words, a CD reader would read the groove of an unrecorded CD-R disc as one long land. To record on a CD-R disc, a laser beam of the same wavelength (780nm) as is normally used to read the disc, but with 10 times the power, is used to heat up the dye. The laser is fired in a pulsed fashion at the top of the ridge (groove), heating the layer of organic dye to between 482°F and 572°F (250°–300°C). This temperature literally burns the organic dye, causing it to become opaque. When read, this prevents the light from passing through the dye layer to the gold and reflecting back, having the same effect of canceling the laser reflection that an actual raised pit would on a normal stamped CD.

Figure 11.6 shows the CD-R media layers, along with the pre-groove (raised ridge from the laser perspective) with burned pits.

The drive reading the disc is fooled into thinking a pit exists, but no actual pit exists—there's simply a spot of less-reflective material on the ridge. This use of heat to create the pits in the disc is why the recording process is often referred to as burning a CD. When burned, portions of the track change from a reflective to a nonreflective state. This change of state is permanent and can't be undone, which is why CD-R is considered a write-once medium.

CD-R Capacity

All CD-R drives can work with the original 650MiB (682MB) CD-R media (equal to 74 minutes of recorded music), as well as the now-standard higher-capacity 700MiB (737MB) CD-R blanks (equal to 80 minutes of recorded music).

Some drives and burning software are capable of overburning, whereby they write data partially into the lead-out area and essentially extend the data track. This is definitely risky as far as compatibility is concerned. Many drives, especially older ones, fail when reading near the end of an overburned disc. It's best to consider this form of overclocking CDs somewhat experimental. It might be useful for your own purposes if it works with your drives and software, but interchangeability will be problematic.

Some vendors sell 90-minute (790MiB) and 99-minute (870MiB) media to make overburning easier. Most standard CD-RW drives can reliably burn up to 89:59 of music onto the 90-minute media, and the resulting CD-R can be played on a variety of late-model auto and home electronics players.

CD-R Media Color

There has been some controversy over the years about which colors of CD-R media provide the best performance. Table 11.6 shows the most common color combinations, along with which brands use them and some technical information.

Table 11.6. CD-R Media Color and Its Effect on Recording

Media Color (First Color Is Reflective Layer; Second Is Dye Layer)	Brands	Technical Notes
Gold-gold	Mitsui, Kodak, Maxell, Ricoh	Phthalocyanine dye. Less tolerance for power variations. Has a rate life span of up to 100 years. Might be less likely to work in a variety of drives. Invented by Mitsui Toatsu Chemicals. Works best in drives that use a Long Write Strategy (longer laser pulse) to mark media.
Gold-green	Imation (nee 3M), Memorex, Kodak, BASF, TDK, Verbatim	Cyanine dye. More forgiving of disc-write and disc-read variations. Has a rated lifespan of 10 years (older media). Recent media has a rated lifespan of 20–50 years (silver/green). Color combination developed by Taiyo Yuden. Used in the development of the original CD-R standards. De facto standard for CD-R industry and was the original color-combination used during the development of CD-R technology. Works best in drives that use a Short Write Strategy (shorter laser pulse) to mark media.
Silver-blue	Verbatim, DataLifePlus, HiVal, Maxell, TDK	Process developed by Verbatim. Azo dye. Similar performance to green media, plus rated to last up to 100 years. A good choice for long-term archiving.

Some brands are listed with more than one color combination, due to production changes or different product lines. You should check color combinations whenever you purchase a new batch of CD-R media if you've found that particular color combinations work better for you in your applications.

Ultimately, although the various color combinations have their advantages, the best way to choose a media type is to try a major brand of media in your drive with both full-disc and small-disc recording jobs and then try the completed disc in as wide a range of drive brands and speeds as you can.

The perfect media for you will be the ones that offer you the following:

• High reliability in writing (check your drive model's list of recommended media)
• No dye or reflective surface dropouts (areas where the media won't record properly)
• Durability through normal handling (scratch-resistant coating on media surface)
• Compatibility across the widest range of CD drives
• Lowest unit cost

If you have problems recording reliably with certain types of media, or if you find that some brands with the same speed rating record much more slowly than others, contact your drive vendor for a firmware upgrade. Firmware upgrades can also help your drive recognize new types of faster media from different vendors.

CD-R Media Recording Speed Ratings

With CD-R mastering speeds ranging from 1x (now-discontinued first-generation units) up through the current 48x–52x rates, it's important to check the speed rating (x-rating) of your CD-R media.

Most branded media on the market today is rated to work successfully at up to 52x recording speeds (some are limited to 48x). Some brands indicate this specifically on their packaging, whereas you must check the websites for others to get this information. If necessary, install the latest firmware updates to reach maximum recording speed.

See "Updating the Firmware in an Optical Drive," p. 599 (this chapter).

If speed ratings are unavailable for your media, you might want to restrict your burning to 32x or lower for data. If you are burning audio CDs, you might find that some devices work better with media burned at 8x or lower speeds than with media burned at higher speeds.

CD-RW

Beginning in early 1996, an industry consortium that included Ricoh, Philips, Sony, Yamaha, Hewlett-Packard, and Mitsubishi Chemical Corporation announced the CD-RW format. The design was largely led by Ricoh, and it was the first manufacturer to introduce a CD-RW drive (in May 1996). This drive was the MP6200S, which was a 2/2/6 (2x record, 2x rewrite, 6x read) rated unit. At the same time, the Orange Book Part III was published, which officially defined the CD-RW standard.

CD-RW drives rapidly replaced CD-R-only drives, and although rewritable DVD drives have largely replaced CD-RW drives, any rewritable DVD drive can function as a CD-R/CD-RW drive. Some low-cost systems include DVD combo drives, which combine DVD-ROM and CD-R/CD-RW capabilities.

You can burn and write to CD-RW discs just like CD-Rs; the main difference is that you can erase and reburn CD-RWs again and again. They are very useful for prototyping a disc that will then be duplicated in less expensive CD-R or even stamped CDs for distribution. They can be rewritten at least 1,000 times or more. Additionally, with packet-writing software (software that supports the Universal Disk Format standard), CD-RWs can even be treated like giant floppy disks, where you can simply drag and drop or copy and delete files at will. Although CD-RW discs are about 1.5–2 times more expensive than CD-R media, CD-RWs are still far cheaper than optical cartridges and other removable formats. This makes CD-RW a viable technology for small-scale system backups, file archiving, and virtually any other data storage task where rewritable DVD is not suitable.

Four main differences exist between CD-RW and CD-R media. In a nutshell, CD-RW discs are

• Rewritable
• More expensive
• Slower when writing
• Less reflective

Besides the CD-RW media being rewritable and costing a bit more, it is writable at about half (or less) the speed of CD-R discs. This is because the laser needs more time to operate on a particular spot on the disk when writing. This media also has a lower reflectivity, which limits readability in older drives. Many older standard CD-ROM and CD-R drives can't read CD-RWs. However, MultiRead capability is now found in virtually all CD drives, enabling them to read CD-RWs without problems. In general, CD-DA drives—especially the car audio players—seem to have the most difficulty reading CD-RWs. So, for music recording or compatibility with older drives, you should probably stick to CD-R media. Check the drive or device specifications to determine compatibility with CD-RW media.

CD-RW drives and media use a phase-change process to create the illusion of pits on the disc. As with CD-R media, the disc starts out with the same polycarbonate base with a wobbled pre-groove molded in, which contains ATIP information. Then, on top of the base a special dielectric (insulating) layer is spin-coated, followed by the phase-change recording layer, another dielectric layer, an aluminum reflective layer, and finally a UV-cured lacquer protective layer (and optional screen printing). The dielectric layers above and below the recording layer are designed to insulate the polycarbonate and reflective layers from the intense heat used during the phase-change process.

Figure 11.7 shows the CD-RW media layers, along with the pre-groove (raised ridge from the laser perspective) with burned pits in the phase change layer.

Instead of burning an organic dye as with CD-R, the recording layer in a CD-RW disc is made up of a phase-change alloy consisting of silver, indium, antimony, and tellurium (Ag-In-Sb-Te). The reflective part of the recording layer is an aluminum alloy, the same as used in normal stamped discs. As a result, the recording side of CD-RW media looks like a mirror with a slight blue tint. The read/write laser works from the underside of the disk, where the groove again appears like a ridge, and the recording is made in the phase-change layer on top of this ridge. The recording layer of Ag-In-Sb-Te alloy normally has a polycrystalline structure that is about 20% reflective. When data is written to a CD-RW disc, the laser in the drive alternates between two power settings, called P-write and P-erase. The higher power setting (P-write) is used to heat the material in the recording layer to a temperature between 500°C and 700°C (932°–1,292°F), causing it to melt. In a liquid state the molecules of the material flow freely, losing their polycrystalline structure and taking what is called an amorphous (random) state. When the material then solidifies in this amorphous state, it is only about 5% reflective. When being read, these areas lower in reflectivity simulate the pits on a stamped CD-ROM disc.

To return the material back to a polycrystalline state, the laser is set to the lower-power P-erase mode. This heats the active material to approximately 200°C (392°F), which is well below the liquid melting point but high enough to soften the material. When the material is softened and allowed to cool more slowly, the molecules realign from a 5% reflective amorphous state back to a 20% reflective polycrystalline state. These higher reflective areas simulate the lands on a stamped CD-ROM disc.

Note that despite the name of the P-erase laser power setting, the disc is not ever explicitly "erased." Instead, CD-RW uses a recording technique called direct overwrite, in which a spot doesn't have to be erased to be rewritten; it is simply rewritten. In other words, when data is recorded, the laser remains on and pulses between the P-write and P-erase power levels to create amorphous and polycrystalline areas of low and high reflectivity, regardless of which state the areas were in prior. It is similar in many ways to writing data on a magnetic disk that also uses direct overwrite. Every sector already has data patterns, so when you write data, all you are really doing is writing new patterns. Sectors are never really erased; they are merely overwritten. The media in CD-RW discs is designed to be written and rewritten up to 1,000 times.

The original Orange Book Part III Volume 1 (CD-RW specification) allowed for CD-RW writing at up to 4x speeds. New developments in the media and drives were required to support speeds higher than that. So in May 2000, Part III Volume 2 was published, defining CD-RW recording at speeds from 4x to 10x. This revision of the CD-RW standard is called High-Speed Rewritable, and both the discs and drives capable of CD-RW speeds higher than 4x will indicate this via the logos printed on them. Part III Volume 3 was published in September 2002 and defines Ultra-Speed drives, which are CD-RW drives capable of recording speeds 8x–24x.

Because of the differences in High-Speed and Ultra-Speed media, High-Speed media can be used only in High-Speed and Ultra-Speed drives; Ultra-Speed Media can be used only in Ultra-Speed drives. Both High-Speed and Ultra-Speed drives can use standard 2x–4x media, enabling them to interchange data with computers that have standard-speed CD-RW drives. Thus, choosing the wrong media to interchange with another system can prevent the other system from reading the media. If you don't know which speed of CD-RW media the target computer supports, I recommend you either use standard 2x–4x media or create a CD-R.

Because of differences in the UDF standards used by the packet-writing software that drags and drops files to CD-RW drives, the need to install a UDF reader on systems with CD-ROM drives, and the incapability of older CD-ROM and first-generation DVD-ROM drives to read CD-RW media, I recommend using CD-RW media for personal backups and data transfer between your own computers. However, when you send CD data to another user, CD-R is universally readable, making it a better choice.

MultiRead Specifications

The original Red and Yellow Book CD standards specified that, on a CD, the lands should have a minimum reflectance value of about 70%, and the pits should have a maximum reflectance of about 28%. Therefore, the area of a disc that represents a land should reflect back no less than 70% of the laser light directed at it, whereas the pits should reflect no more than 28%. In the early 1980s when these standards were developed, the photodetector diodes used in the drives were relatively insensitive, and these minimum and maximum reflectance requirements were deliberately designed to create enough brightness and contrast between pits and lands to accommodate them.

On a CD-RW disc, the reflectance of a land is approximately 20% (plus or minus 5%) and the reflectivity of a pit is only 5%—obviously well below the original requirements. Fortunately, it was found that by the addition of a relatively simple AGC circuit, the ratio of amplification in the detector circuitry can be changed dynamically to allow for reading the lower-reflective CD-RW discs. Therefore, although CD-ROM drives were not initially capable of reading CD-RW discs, modifying the existing designs to enable them to do so wasn't difficult. Where you might encounter problems reading CD-RW discs is with CD audio drives, especially older ones. Because CD-RW first came out in 1996 (and took a year or more to become popular), most CD-ROM drives manufactured in 1997 or earlier have problems reading CD-RW discs.

DVDs also have some compatibility problems. With DVD, the problem isn't just simple reflectivity as it is an inherent incompatibility with the laser wavelength used for DVD versus CD. The problem in this case stems from the dyes used in the recording layer of CD-R and RW discs, which are very sensitive to the wavelength of light used to read them. At the proper CD laser wavelength of 780nm, they are very reflective, but at other wavelengths, the reflectivity falls off markedly. Normally, CD drives use a 780nm (infrared) laser to read the data, whereas DVD drives use a shorter wavelength 650nm (red) laser. Although the shorter wavelength laser works well for reading commercial CD-ROM discs because the aluminum reflective layer they use is equally reflective at the shorter DVD laser wavelength, it doesn't work well at all for reading CD-R or RW discs.

Fortunately, a solution was introduced by Sony and then similarly by all the other DVD drive manufacturers. This solution consists of a dual-laser pickup that incorporates both a 650nm (DVD) and 780nm (CD) laser. Some of these used two discrete pickup units with separate optics mounted to the same assembly, but they eventually changed to dual-laser units that use the same optics for both, making the pickup smaller and less expensive. Because most manufacturers wanted to make a variety of drives—including cheaper ones without the dual-laser pickup—a standard needed to be created so that someone purchasing a drive would know the drive's capabilities.

So how can you tell whether your CD or DVD drive is compatible with CD-R and RW discs? In the late 1990s, the OSTA created the MultiRead specifications to guarantee specific levels of compatibility:

• MultiRead—For CD drives
• MultiRead2—For DVD drives

In addition, a similar MultiPlay standard exists for consumer DVD-Video and CD-DA devices.

Table 11.7 shows the two levels of MultiRead capability that you can assign to drives and the types of media guaranteed to be readable in such drives.

Table 11.7. MultiRead and MultiRead2 Compatibility Standards for CD/DVD Drives

Media	MultiRead	MultiRead2	Media	MultiRead	MultiRead2
CD-DA	X	X	DVD-ROM	—	X
CD-ROM	X	X	DVD-Video	—	X
CD-R	X	X	DVD-Audio	—	X
CD-RW	X	X	DVD-RAM	—	X
X = Compatible; drive will read this media. —= Incompatible; drive won't read.

Note that MultiRead also indicates that the drive is capable of reading discs written in Packet Writing mode because this mode is now being used more commonly with both CD-R and DVD rewritable media.

If you use only rewritable CD or DVD drives, you don't need to worry about compatibility. However, if you still use nonrewritable drives, you should check compatibility with other types of media. Although the MultiRead and MultiRead2 logos shown in Figure 11.8 are not widely used today, you can determine a particular drive's compatibility with a given media type by viewing its specification sheet.

You can obtain the MultiRead specification (revision 1.11, October 23, 1997) and MultiRead 2 specification (revision 1.0, December 6, 1999) from the OSTA website.