Problems recording a gold disc
Recently, CD recorders have become very widespread. In this regard, the circle of people dissatisfied with these devices has also expanded dramatically. There was a firm belief that recording on “golden discs” is generally unreliable, “golden discs” cease to be read over time, and you will not find a good blank (“blank”) for recording during the day with fire. To deal with these issues, a series of tests were carried out in the research laboratory of the EPOS company. The tests used Samsung 32x and 40x CD-ROMs, TEAC 32x, Creative 48x and Hitachi 32x. Recording was carried out on blanks TDK, Fujifilm, BASF-Extra, BASF-Multispeed, TDK and cheap blanks from an unknown manufacturer. When recording, Yamaha 4416, Yamaha 100, Plasmon, Panasonic 7502 devices were used. Recording was performed at the maximum speed for each device. In addition, a series of discs were recorded on a faulty Panasonic 7502 device. This device was repaired after a period of operation with “classical” signs: the recorded discs are read on some CD-ROMs, but not on others, and recording is not possible on some blanks.
What is “readable” and what is “not readable”?
As it turned out, the question posed in the subtitle was perhaps the most difficult in these trials. How, for example, to assess a rather common nuisance: an attempt to copy the entire CD just burned to the hard drive leads to a read error, and copying in one directory is successful? How to assess the deterioration of “readability” over time? And in what numbers should the results be reflected so that any general analysis can be performed?
The fact is that CD-ROMs have very powerful error correction tools. CDs from the very beginning were intended for mass use. This means that they should not be expensive to manufacture and should not lose operability with careless handling (even a heavily scratched disk, most often, reads well). Therefore, when recording, two-level noise-resistant coding with data interleaving is used. As a result of this, the fact that after recording the disc is read, does not mean that it will be read after some time.
Minor errors in poor recording will be corrected when reading. However, part of the power of the error-correcting code will be used from the very beginning for this. If the disk initially contains many (even minor) errors, then the slightest damage to the surface, inevitable during operation, can lead to the fact that the disk stops reading. But due to the fact that minor errors are corrected on the fly, it is hardly possible to control their appearance. With an increase in the number of errors, the second mechanism for ensuring reliable readout of information comes into play: the drive reduces read speed. The more errors, the slower the speed. Thus, the read speed on various parts of the disk surface is a fairly stable sign of recording quality. This characteristic was used in the tests.