11.9 Media Issues

There's been a lot of nonsense written about CD-R media. One person swears that gold/green discs are great and silver/blue discs worthless, and another says the opposite. The truth is that there are distinct differences in media, but no absolutes. Disc A may work perfectly in Drive A and not work at all in Drive B, and Disc B may work perfectly in Drive B and not at all in Drive A. That situation is less common with recent drives than it was with older models, but some drives still show a strong preference or dislike for particular disc types. Discs differ as follows.

Current CD-R blanks use one of these reflective layers:

Gold

The metal used in early CD-R discs, and still commonly used in current production discs. The advantage of gold is that it is stable. The disadvantage is that 24K gold is expensive, even in the microscopically thin layers used in CD-R discs. As the price of CD-R discs continues to plummet, the cost of gold becomes an increasingly important factor, which has led some manufacturers to substitute silver. By early 2002, many disc manufacturers, including Kodak, had discontinued their gold-based products or limited gold reflective layers to their premium or "professional" lines. Note that some CD-R discs that appear gold in fact use little or no gold in their formulations.

Silver alloy

The advantages of silver alloy relative to gold are that it is relatively inexpensive and actually has better reflective characteristics across a wide spectrum. The chief disadvantage of silver is that it corrodes. Even a tiny crack in the label layer can allow airborne pollutants, particularly sulfides, to gain access to and react with the silver layer. If this occurs, the CD become unreadable.

Most current CD-R blanks use one of these dyes (although to avoid patent infringement, some disc manufactures use similar but not identical dyes):

Cyanine

As the name indicates, this dye is cyan (bluish-green) in color. Used with a gold reflective layer, cyanine-based discs appear green on the clear side and gold on the label side (the so-called "green/gold" discs). Used with a silver reflective layer, cyanine discs appear light blue/silver. Cyanine was the first dye used to make CD-R discs, and quickly developed a reputation for stability problems. That has been overcome in current discs by using metal-stabilized cyanine. The advantage of cyanine is that it is more sensitive to light than other dyes, which means that cyanine discs tolerate a wider range of laser power settings, making them more likely to be compatible with any given drive and firmware revision. For example, a typical cyanine disc can be written with a laser power setting between 5.5 and 7.5 milliwatts, while a typical pthalocyanine disc requires 5.0 to 6.0 milliwatts. On the downside, accelerated aging tests show expected lifetimes for cyanine-based discs of "only" 50 years or so, significantly less than other dyes. In practical terms, no one 50 years from now will have a drive capable of reading any CD, so this matters little.

Pthalocyanine

This dye is a very pale yellow-green color, approaching colorless. With a gold reflective layer, discs appear gold/gold or greenish-gold/gold. With a silver layer, they appear light yellow-green/silver or even silver/silver. Because cyanine-based CD-R discs were protected by patent, other manufacturers developed alternative dyes, including pthalocyanine. The one thing to be said in favor of pthalocyanine is that its lower light sensitivity relative to cyanine gives it estimated archival stability of nearly 100 years. Lower light sensitivity also means that pthalocyanine-based CD-R discs are less likely to be compatible with any given CD-R drive, particularly an older model. All recent CD-R drives support pthalocyanine-based discs, and most older models can accommodate them with a firmware upgrade.

Azo

This dye is a deep blue color. Commonly available azo discs, notably those from Verbatim, use a silver reflective layer, which gives the data side a very deep blue color. Azo is even less sensitive to light than pthalocyanine, which gives azo-based discs expected archival stability of more than 100 years, and makes drive support even more problematic than pthalocyanine. All recent CD-R drives support azo-based discs, and most older models can accommodate them with a firmware upgrade. Due to their insensitivity, some early azo-based discs did not support writing faster than 1X or 2X, but current azo-based discs are usable at moderately high recording speeds.

There is no single answer to the question, "Which is best?" There are numerous variables in the CD-R manufacturing process. The thickness and density of the dye layer vary, as does that of the reflective layer. Some manufacturers have begun using dyes that resemble those listed above, but have different characteristics. There is no way to tell by appearance alone which dye a disc uses. The physical groove structure of different CD-R blanks may differ, to optimize that disc for different laser powers, writing speeds, and schemes. But the bottom line is that all CD-R blanks, including the no-name ones, are much better than they were a few years ago. The primary measure of CD-R disc quality, Block Error Rate (BLER), is much lower now than in the past.

But the real problem is that, although most CD-R disc manufacturers make every effort to comply with strict Orange Book specifications, some CD writer manufacturers do not. The upshot is that different writers vary widely in which discs they can use successfully. In general, more compromises are made with inexpensive writers, which are typically much more media-sensitive than are more expensive models. A $175 Plextor burner is likely to work with just about any media you attempt to use. An $75 no-name ATAPI burner may be very choosy indeed about which blanks it is willing to write. Also, do not overlook the importance of keeping your drive firmware up to date. As disc manufacturers change formulations and new disc types come on the market, you may need to update your firmware to enable your drive to use those new blanks.

Your sole criterion should be which media work properly in your recorder. The best starting point is to use the disc(s) recommended by the drive manufacturer. Most manufacturers provide such a list on their web sites. Those lists are updated as new media types become available, and assume you have the latest firmware.

Whichever blanks you settle on, recognize that any CD-R disc should be treated more carefully than a pressed CD. In particular, avoid exposing blank or recorded CD-R discs to sunlight or heat unnecessarily. Leaving an unprotected CD-R disc on the dashboard of your car is a quick way to destroy it.

11.9.1 Writable CD Capacities

Standard CD-R discs are available in 63-, 74-, and 80-minute lengths, which respectively store about 550, 650, and 700 MB of data in ISO 9660 format. For example, a nominal 74-minute CD-R disc stores (74 min x 60 sec/min x 150 KB/sec) = 681,984,000 bytes, 666,000 KB, or 650.390625 MB. CD data formats use 2 KB sectors, which means that a standard 74-minute disc contains (666,000 KB / 2 KB) = 333,000 sectors. Most blanks actually contain more than the required number of sectors. For example, a nominal 74-minute blank may contain as many as 344,250 sectors, which translates to 76.5 minutes. This "extra" space permits overburning (writing more data to a disc than its nominal capacity) if the CD writer and software both support overburning.

Think twice before you try overburning. If your CD writer doesn't support it, attempting to overburn may physically damage the drive beyond repair and void the warranty. Many CD-ROM drives, even those that read ordinary CD-R discs, cannot read overburned CDs. Our advice about overburning is simple: unless there is no other way to do the job, don't use it.

Remember that when you write data to a CD recorder, the 2 KB logical sectors are actually written to 2,352 byte physical sectors, with the remaining space used for ECC code. That means that a standard 74-minute CD-R blank actually stores (333,000 sectors x 2,352 bytes/sector) = 783,216,000 bytes or about 747 MB. Accordingly, you may find 74-minute discs with identical true capacities advertised as having nominal capacity from 650 to 780 MB, although something in the 650 to 680 MB range is most common.

When selecting CD-R discs, use the lowest-capacity discs big enough for your data. In practical terms, that means using 74-minute CD-R blanks rather than 80-minute blanks whenever possible. The 63-minute blanks have now all but disappeared from the market. That's a pity, because few burned CDs contain more than 63 minutes (550 MB), and CDs recorded on 63-minute blanks are in every respect superior to those recorded on 74-minute or 80-minute blanks. They are more likely to be readable on more CD-ROM drives and CD players, and are less likely to generate read errors. However, 63-minute blanks fell prey to the pervasive "more is better" way of thinking. Most people, given the choice of a 63- or 74-minute blank for the same price, chose the latter. The 63-minute blanks became harder and harder to find, and their increasingly limited distribution meant they eventually cost more than 74-minute blanks, and so disappeared from the retail channels. Alas.

You may not have the choice of 74- versus 80-minute blanks, depending on your CD writer. Not all CD writers can use 80 minute blanks, and even if the drive supports 80-minute blanks, your CD burning software may not.

In late 2001, some vendors began selling so-called 90-minute or 99-minute CD-R blanks. This increased capacity is achieved by tightening the spiral track and using overburning. Technically, these are not CD-R blanks at all, because they are completely non-compliant with CD-R standards. Despite that, they work in some burners and some players, although both recording and playback problems are common with them. Their actual capacity depends not just on the blank itself, but on the burner, the software you use to write the blank, and the player or drive you use to read it. Not all burners can write these extended blanks, and not all drives and players can read them. If your burner writes them at all, it will likely limit writing to between 87:59 and 99:05. For example, our Plextor PlexWriter 24-10-40A writes at most 94:59, and our PlexWriter 8-2-20S at most 89:59. Also note that even burners that can write these oversized blanks cannot necessarily write them at all speeds the writer supports. For example, our PlexWriter 24-10-40A writes these discs only at 2X, 4X, and 8X, while the 8-2-20S writes at all supported speeds other than 1X. If you want to play with these discs for recording music, fine, but we recommend you not count on them for anything that matters.

11.9.2 CD-R Media Manufacturers and Brands

When it comes to choosing CD-R(W) media, the factory that makes it is more important than the brand name that's on it. In our (albeit limited) experience, some factories consistently turn out high-quality media that produce few coasters and have few compatibility problems with diverse drives. Other factories seem to consistently turn out lower-quality media that produce more frequent coasters and/or have more compatibility problems. Table 11-1 lists current CD-R(W) disc manufacturers, our own experience with at least some of their products, and some of the brand names that their discs have been sold under. Note that many manufacturers resell their discs to others who sell them under a different name. Also, some manufacturers both produce and sell discs under their own brand name, and resell discs that are produced by other manufacturers under their own brand name.

In the table, CD-R(W) manufacturer lists the actual company that produces the media. Rating lists our own opinion of that factory's products, using + for Above Average, 0 for Average, and - for Below Average. If we have no experience with a particular disc, we leave the rating column blank. Our rating is a composite one, based on reliability (number of coasters); how compatible we have found the discs to be, both with a variety of burners and with various CD-ROM drives and players; and how stable we have found the discs to be based on ad hoc testing. Note that our opinions are based on testing a limited number of discs in a limited number of drives, so your experience may vary. A comprehensive list of CD-R media manufacturers is available at http://www.world.sony.com/Electronics/CD-R/index.html.

Table 11-1. CD-R(W) media makers

CD-R(W) manufacturer	Rating	Brand/product names
Auvistar Industry Co., Ltd.		[unknown]
CMC Magnetics Corporation	-	BASF, CMC, Imation
Customer Pressing Oosterhout		[unknown]
Doremi Media Company, Ltd.		[unknown]
Fornet International PTE Ltd.	-	Fornet & Magnex
Fuji Photo Film Company, Ltd.	0	FujiFilm
GigaStorage Corporation	-	GigaStorage, Amadis, Cursor, Maxmax, Mozart
Hitachi Maxell, Ltd.	0	Maxell
King Pro Mediatek Inc.		[unknown]
Kodak Japan Limited	+	BASF, Kodak
Lead Data, Inc.	-	Lead Data, Lenco, Targa
Mitsubishi Chemical Corporation	0	Sony, Traxdata, Verbatim
Mitsui Toatsu Chemicals, Inc.	+	Hewlett-Packard, Mitsui, Philips, Sony
Multi Media Masters & Machinery SA		Polaroid
Opti. Me. S. S.p.A.		[unknown]
Optical Disc Manufacturing Equipment		[unknown]
Pioneer Video Corporation	+	Pioneer
Plasmon Data Systems Ltd.		[unknown]
Postech Corporation		Laser
Princo Corporation	-	BTC, Princo, KingTech
Prodisc Technology Inc.	-	Prodisc, Smart & Friendly
Ricoh Company Limited	+	KAO, Ricoh
Ritek Company	-	Arita, BASF, Dysan, FujiFilm, Hi-Space, Intenso, JTEC, Memorex, MegaData, MMore, Onkyo, Philips, Pony, Ricoh, Rimax, Samsung, Targa, TDK, Traxdata, Waitec & War
Seantram Technology, Inc.		[unknown]
SKC Company, Ltd.		[unknown]
Sony Corporation		Sony
Taeil Media Company, Ltd.		[unknown]
Taiyo Yuden Company, Ltd.	+	3M, Imation, Philips, Sony, Taiyo Yuden
TDK Corporation	+	3M, Pioneer, TDK, Yamaha
Xcitek, Inc.		BestMedia

11.9.3 Identifying Unknown CD-R and CD-RW Media

Many CD-R and CD-RW discs are unlabeled or relabeled, and it may sometimes be useful to identify the actual source of such discs. Frank Wolf's CDR Media Code Identifier utility, shown in Figure 11-7, does just that. (See http://www.gum.de/it/download/english.htm.) Based on data the manufacturer hardcodes into the ATIP (absolute time in pregroove) of the discs themselves, this utility reports who made the disc, the Type number and dye used, the media type (CD-R versus CD-RW), the nominal capacity of the disc, and (infrequently) the range of burning speeds for which the disc is suited. In this case, although the disc was labeled as a Smart & Friendly 4X CD-R disc, it was actually produced by Prodisc Technology Inc.

Figure 11-7. The CDR Media Code Identifier utility identifies the actual source of a disc

Although this utility works on most CD-R(W) discs whether or not they have been written to, it usually requires reading the disc in a CD-R(W) drive rather than a CD-ROM drive. And not just any CD-R(W) drive, because many cannot read the ATIP data upon which this utility bases its reports. We've tried using this utility with many CD-ROM and DVD-ROM drives, and few work properly. Most, like the Plextor PX-40TW shown in Figure 11-7, simply return an Invalid Command-Operation Code error message. Some return partial or garbled information. A few, including the Hitachi GD-5000 (labeled "Compaq" in the figure) do work properly, so it's worth trying the program with your CD/DVD-ROM drive if your CD-R(W) drives don't support it. The utility is free and well worth downloading and using if you have a drive that supports it.

An unlabeled CD-R or CD-RW disc is not necessarily an inferior "bulk" disc. Some first-rate disc manufacturers (Taiyo Yuden, for example) sell the same disc with or without printed logos. Even some resellers who should know better describe such unlabeled blanks as "no-name," "bulk," or "replicator" products, implying that they are somehow of lower quality than discs with the manufacturer logo printed on them. That may or may not be true. Some manufacturers do make a "premium" line of discs that is of higher quality than their "value" line. But the presence or absence of a silk-screened logo is no evidence either way.

11.9.4 Archival Stability of CD-R(W) Media

So how long can you expect that CD-R disc you just burned to last? There's been a lot of nonsense written on that topic, but ultimately the answer is that no one knows. Projected lifetimes of 50, 100, and even 200 years are casually tossed around, as though they had any meaning.

Here's the truth. The only way to know for sure if a disc will last 100 years is to burn one, wait 100 years, and try to read it. Obviously, that's not a practical solution, so media manufacturers use various testing methods to estimate archival stability. All of those testing methods depend on accelerated aging, achieved by storing the disc at much higher than normal temperatures, often in conjunction with high humidity and high ultraviolet flux intended to simulate bright sunlight.

The rule of thumb in chemistry class says that the rate of most chemical reactions approximately doubles for each 10°C (18°F) increase in temperature. So, in theory, if one assumes that normal storage temperature for a CD-R disc is 20°C, a disc stored at 30°C ages twice as fast as normal, one stored at 40°C ages four times as fast, one stored at 50°C ages eight times as fast, and so on. The trouble is that this rule of thumb is just that—a rule of thumb—and that the reaction rate slope is linear over only a limited range of temperatures.

The well-meaning chemist, being told that his Thanksgiving turkey should be cooked for 8 hours at 325°F might reasonably conclude that he could decrease cooking time to four hours by setting his oven to 343°F, to two hours at 361°F, to one hour at 379°F, to 30 minutes at 397°F, and to 15 minutes at 415°F. We'll leave it to the reader to imagine the results of a turkey cooked 15 minutes at 415°F. (Robert was the well-meaning chemist, and he also didn't realize that he was supposed to remove the feathers. We are not making this up.)

So manufacturers project CD-R disc lifetimes based on testing them at high temperatures and in bright sunlight. But all that really determines is how long that CD-R disc is likely to last when stored at high temperatures in bright sunlight. Most people, of course, store their CD-R discs at room temperature and in the dark, or nearly so.

The real issue is not the archival stability of media, but the future availability of drives that can read the discs you're currently writing. Any CD you write today on a good-quality blank is almost certain to last at least 20 years, if not 100. But fast-forward to the year 2012. Finding a drive that can read a CD written in 2002 won't be a problem. In 2022, it may be a bit harder, but you should be able to find such a drive. But in 2102? Good luck. Try finding a drive today that can play back a wax cylinder audio recording made in 1900. Finding a drive to play a 100-year-old CD won't be any easier.

11.9.5 Labeling CD-R(W) Media

Finally, a few words about labeling. Drive and media manufacturers vary in what they recommend, and the best course is to follow those recommendations. However, some guidelines are nearly universal:

• Never use a standard sticky label on a CD-R disc. The adhesive may damage the label side of the disc, causing it to degenerate rapidly and become unreadable. Also, the small weight of that label is sufficient to imbalance the disc in some high-speed players. At best, such an imbalance may cause read errors. At worst, it may destroy the disc or even the drive.

• The circular labels designed for CD-Rs are generally safe, but even they have been known to imbalance a disc if not applied perfectly centered. If you use such labels, never attempt to peel one off. Doing so may cause the top layer of the CD-R to separate, destroying the disc.

• Do not use a hard-tip marker, which may score the label layer. We usually label CD-R discs with a Sharpie® soft-tip permanent marker. Some sources recommend not using a permanent solvent-based pen because it may etch the label layer. Other sources, including some CD-R manufacturers, do recommend using a solvent-based permanent marker. Although we've never had a problem using the Sharpie®, for maximum safety choose your marker according to the recommendations of the disc manufacturer.

When hand labeling is not neat enough, e.g., for discs you plan to distribute outside your company, consider using printable discs, which have a surface that may be printed with an inkjet and/or thermal printer. These are available in various types, and background colors, and generally work well if you follow the manufacturer's instructions carefully. Note, however, that you can't print these discs in just any printer. They require a special printer designed to print CDs, and such printers are not inexpensive. Many service bureaus have such printers, and will be happy to label short runs of CD-Rs for you at reasonable cost.