In the midst of the panic that the research for this article produced for me, I decided to make an evaluation of my personal archive of CD-R discs. I did not use a rigorous method, I do not offer a representative sample or aspire to establish benchmarks, but I found consistent results in a package of 140 discs, within which I found 11 with reading problems.
For the evaluation I used a fairly ordinary PC, equipped with Windows XP and a new drive of widely known brand. Each disc was subjected to two tests: copying of its contents using the operating system and subsequent analysis with CD-Check, a piece of software that allows to locate corrupt files on disks of any type. I thought it pertinent to establish a work cycle that would allow to rule out failures of the reading unit due to intense use or increase in temperature. That's why I limited the evaluation to five-disc packages with "rest" times of no less than 20 minutes. Most of the discs evaluated are stored in plastic cases, although about thirty have spent several years stacked in spindles.
I found the following:
- 9 of the 11 discs that failed are "generic" discs. The remaining two belong to a well-known brand.
- Most of the failed discs were "burned" at speeds above 16X, and none are more than two years old.
- The oldest records in my collection date back to 1993. None of these disks had read problems.
- A fact that can be useful: 10 of the 11 disks that failed could be read completely using software designed to recover data recorded on troubled media. My conclusion after this process is this: I'm going to start replicating the records that really matter to me. Eleven out of 140 equals almost 8%!
Pieces of software used for this evaluation:
CD-Check v. 3.1.0.1 free for non-commercial use) www.elpros.si/CDCheck/
IsoBuster v. 1.6 (limited version for free use) www.isobuster.com
