Slashdot Mirror

← Back to Stories (view on slashdot.org)

NIST Releases Study Of CD/DVD Longevity

Posted by timothy on from the save-save-save dept.

dirkin writes "The National Institute of Standards and Technology has released a preliminary study of the potential lifespan of CD-Rs and DVD-Rs. The PDF study is here. A good starting point for deciding what type of media to purchase to keep those backups and photos kicking around longer. (You DID buy the silver/gold alloy phthalocyanine CDs, didn't you?)"

5 of 425 comments (clear)

  1. but something is missing... by KingRamsis · · Score: 5, Interesting

    the speed in which the CDR is burned sometimes it makes a difference, for the highest reliability I think 1x is the best.

  2. Don't take it too seriously by tgv · · Score: 5, Interesting
    The online summary of the article says literally: "Uit onze steekproef blijkt dat er veel rommel op de markt is. We hebben cd-r's aangetroffen die nooit op de markt hadden mogen komen. Het gaat daarbij mogelijk om afgedankte partijen."

    Or, rather literally translated into English: "Our sample shows that there is a lot of junk on the market. We have found cd-rs that should never have been for sale. Possibly it concerns rejected batches." Which suggests to me that the correct heading of this article should be: CD-Rs are like everything else: you get what you pay for.

  3. i always knew... by Bill+the+Bilby · · Score: 5, Interesting

    that buying cheap crappy CD-Rs meant that your data died faster, but I had no idea how the degredation worked. What about the "armored" DVD-Rs from places like Datawrite? They're supposedly almost impossible to destroy. How well do they stand up?

  4. Re:As Usual... by DaveJay · · Score: 5, Interesting

    The trouble with this approach is the same as the trouble with preserving old videotape material, something I used to be involved in.

    I worked for a museum that preserved such materials, and always wanted to establish a program not just to copy older videotape to newer analog formats (which already existed) but to copy and RE-copy those digital tapes on a two-year cyclical program; the digital data wouldn't degrade during the transfer, and by essentially replacing the media containing the data on a regular basis, we'd have a good chance of saving the material long-term.

    I never got approval for a simple reason: tape stock is expensive, staff is expensive, and coordination of such an effort requires diligence. Similarly, you could use CD-R/DVD-R to back up your material and re-burn the discs on an ongoing rotation, but most people don't have that kind of discipline even if they have the money.

    For me personally, I've found the best approach is to maintain the data on a redundant RAID array, with occasional backup to DVD-R. This way, the data itself will outlive the death of individual local drives, while the DVD-R only needs to serve as a short-term disaster-recovery solution.

    Of course, once my critical dataset gets large enough to require more than a few DVD-Rs, I'll probably get lazy...aren't we about due for a new format by now? ;)

  5. Re:mobile fidelity... by sh00z · · Score: 5, Interesting
    used to include a study of glove compartment temperature cycles for their high end discs...
    Sounds like my personal study. The second CD I ever burned was at the fastest speed of my writer; I wrote on the label with a Sharpie; I store it in my glove compartment (in Texas). I listen to it once per quarter, and when it fails, I'll start replacing everything else (praying that temperature- and humidity-controlled environments are better for the media's health).