This is an excellent question. Unfortunately, color is not a useful CD-R quality indicator. The human eye sees color in the visible portion of the light spectrum, while CD-R lasers operate in the infrared part of the spectrum. Only the interaction between the infrared laser and the organic polymer dye layer is important to quality, not visible color such as gold or green. Read lasers operate at a very low power level that does not alter the dye. Writers switch laser power to much higher levels that alter the dye during recording. The matrix of CD-R dye types, writers, firmware revisions, and recording software are far more complex than for CD-ROM discs.

CD-R discs first used a cyanine dye developed in Japan by Taiyo Yuden. Discs initially had longevity problems until stabilizers were added. Current products have good longevity. The blue color of this dye and the gold reflective layer combine to produce an emerald green or blue-green color. A relatively high laser recording power of 6.5 milliwatts is required at 1X, while greater intensities must be used at higher speeds. The relation between recording speed (1X, 2X, 4X, 6X) and power is non-linear. Although some uncertainty exists regarding the recording process, it seems that the laser bleaches the dye or alters its molecular structure in a manner that simulates pits and lands. A long write strategy is required, since the length of the altered region does not significantly exceed the region illuminated by the laser pulse. Taiyo Yuden licensed their dye to Ricoh and TDK. These companies also manufacture discs for resale under other brand names.

Phthalocyanine has excellent longevity and was developed in Japan by Mitsui Toatsu as an alternative to cyanine. The dye itself has little color and combines with the gold reflective layer to produce a gold-green or rich yellowish color. Laser recording powers are lower, about 5.5 milliwatts at 1X. Although greater powers are required at higher speeds, the relationship is non- linear. The recording laser may actually form depressions in the phthalocyanine dye polymer. These are significantly larger than the region illuminated by the laser pulse, requiring the use of a short write strategy during recording. Mitsui Toatsu licensed their dye to Eastman Kodak and to Mitsubishi/Verbatim. All three companies may manufacture discs for resale by other companies.

Recently the NCC subsidiary of Mitsubishi developed a metallized azo, or metal chelate, dye that is a dark blue color. Use of a silver reflective layer provides an attractive background for a label and combines with the dye to produce an unmistakable blue color when viewed from the readout surface. Azo CD-R discs are also marketed by Verbatim.

Taiyo Yuden established stabilized cyanine dye as a de facto reference for CD-R recorders. The appearance of Mitsui Toatsu phthalocyanine required recorder manufacturers to develop alternative recording powers and write strategies. These dyes utilize proprietary and patented formulas that may be modified from time to time. High speed, 16X, 24X, and higher speed recorders as well as azo dye has produced new challenges. This matrix of dye type and speed has produced constantly changing firmware revisions in CD-R recorders. Updates are generated as more is learned about the properties of the different dye types. Not all recorders use the same approach to recording power and write strategy. Therefore, certain writers may work best with certain media. Some manufacturers will supply the names of approved vendors upon request.

Reliance upon one type of dye, or even one brand, does not always produce high quality one-offs , since dye properties can change because of design modifications or lot to lot variations. Dyes are often applied by spin coating followed by a curing step, therefore properties can also vary from ID to OD. Also, third party dyes have appeared on the market. These are not Taiyo Yuden cyanine or Mitsui Toatsu phthalocyanine dyes, although visible colors may be similar. Their response to the infrared recording beam may be different from that of licensed dyes, resulting in different recorded quality.

Although awareness of quality issues is important, it is not necessary for the user to acquire expertise in dye chemistry, recording power, and write strategies. Instead, focus on regular testing that assures high quality one-offs. This requires specialized in-house equipment or use of an independent testing laboratory. Functionality in just one drive is not an adequate test, since verification using sector reads at the DOS level is insensitive to defects that cause the disc to fail in other drives. CD-ROM manufacturers regularly utilize sophisticated quality tests to assure interchangeability of their products. Anyone who generates one-offs must perform similar tests if costly or embarrassing field problems and product returns are to be minimized. Rely on test results, not color or brand name.

Why does my recording software indicate different capacities for CD-R discs from various manufacturers?

There are at least three reasons for different capacities of both 63 minute and 74 minute CD-R discs. These reasons result in 74 minute capacities that vary between 73:50 and 74:50 for discs from different suppliers. User data capacity of a nominal CD-R is 74:12, or about 652 MB of mode 1 user data. Capacities of 63 minute discs vary between 63:02 and 64:02, with a nominal value of 63:18, or about 556 MB of mode 1 user data.

The first reason for different capacities is that disc manufacturers have certain tolerances in disc design. Two of these tolerances are the exact positions of start of lead-in and the last possible start of lead-out. The exact values chosen by the manufacturer are encoded as ATIP in the lead-in section of the wobbled pre-groove.

Standards only require that lead-in ends at a diameter of 49.6 mm to 50.0 mm, and that lead-out ends at a maximum diameter of 117 mm. Different choices by disc manufacturers can each satisfy the Standards, resulting in slightly different capacities. Although track spacing is fixed at 1.6 micrometers in the Standards, the 0.1 micrometer track space tolerance allows design variations that also can result in different capacities.

Recording software is a second source of variation. Various programs may indicate capacities that differ for the same disc. Disc capacity may be evaluated differently based upon assumptions about lead-in, post-gap, lead-out, and other elements of a recorded track. Lead-in usually occupies about 2.5 seconds but can vary by a few tenths of a second. Post-gap may be several seconds shorter or longer than the correct value of 2 seconds. Nominal 2 second lead-out regions may be a short as 1.5 seconds or can be much longer. Complex directory trees can consume many seconds of capacity, because sub-directories are recorded in the user data region together with data files. Large numbers of sub-directories will also lengthen the system area because root directory and path tables become larger.

Some software may report a smaller capacity for an unrecorded disc but will actually write larger amounts of information. Or the program may simply report a default capacity of 74 minutes, not the true capacity encoded by the disc manufacturer into the pre-groove of every disc. Certain software may report only available space for files, while other programs might include gaps and the system area in available capacity.

Conversion of recording time to bytes is the third source of variation. Sectors are always 2352 bytes long and repeat at a rate of 75 per second at 1X. User data contents of a sector can vary from 2048 bytes for Mode 1 and for XA Mode 2/Form 1 up to 2324 bytes for XA Mode 2/Form 2 and to 2352 bytes for CD-DA. This results in further capacity variations based upon format, even if maximum ATime does not vary. For example, the capacity of a 650 MB Mode 1 disc would be 737 MB for XA Mode 2/Form 2 and 746 MB for CD-DA.

Capacity variations are normal, and a slightly higher reported capacity does not mean that the disc is better. Discs having lower capacity may actually result in better recorded quality. Although some CD-R manufacturing methods can produce discs with uniformly high quality, other discs may have poor quality near the outer rim because of injection molding and spin-coating limitations. Heating effects during a long recording may also degrade quality at high values of ATime. It is always best to avoid recording computer files during the last four-to-eight minutes unless detailed test results for BLER, asymmetry, error rates, and other quality characteristics indicate that it is safe to record at maximum capacity. Only 35 MB to 70 MB of capacity will be unused, while quality may be significantly improved.

Another capacity risk is the appeal of 80 minute discs. Limited availability coupled with premium prices create a false impression of superiority. Avoid super-high capacity discs having capacities in excess of 74 minutes. These capacities can only be achieved by violating one or more requirements of the CD-R and CD-ROM Standards, such as track pitch. Such discs can be recorded in many drives, but not in all. More important, they will not be readable in all drives.