Slashdot Mirror
How To Choose Archival CD/DVD Media
An anonymous reader tips us to an article by Patrick McFarland, the well-known Free Software Magazine author, going into great detail on CD/DVD media. McFarland covers the history of these media from CDs through recordable DVDs, explaining the various formats and their strengths and drawbacks. The heart of the article is an essay on the DVD-R vs. DVD+R recording standards, leading to McFarland's recommendation for which media he buys for archival storage. Spoiler: it's Taiyo Yuden DVD+R all the way. From the article: "Unlike pressed CDs/DVDs, 'burnt' CDs/DVDs can eventually 'fade,' due to five things that affect the quality of CD media: sealing method, reflective layer, organic dye makeup, where it was manufactured, and your storage practices (please keep all media out of direct sunlight, in a nice cool dry dark place, in acid-free plastic containers; this will triple the lifetime of any media)."
I'm surprised to hear that consumer media can last so long. I was under the impression that consumer media would only last at most 20 years. Good to know it is longer.
Can I bum a sig?
I always thought Matsui "Gold" and "Silver" were the top-rated media. At least for CD-Rs (though I thought they were held in high regard for DVD blank media, too). I used to mail-order un-branded blanks them by the spool.
Method of processing duck feet
Cheap but adequate DVD-R media costs $200 for 1000 discs, about 4TB capacity. And a cheap DVD-R changer jukebox costs under $500, about 800GB per load.
Why not just burn a few copies of the archive to a bunch of DVD sets? The DVDs will get defects, but shuffling the chunks across the discs just a little will probably ensure that the random distribution of specific defects will not hit every copy of a given bit, against the odds a low defect rate will produce.
How about a pair of those archivers, which fire up every few years just to transfer the aging DVDs to fresh new ones? For another $1000, that's another 5 cycles of DVDs, 800GB per cycle. Another $1000 gets a pair of backup jukeboxes.
For higher capacities than 800GB, there are pricier pro jukeboxes, but with dual drives for the retranscription cycle (and faster restores). But the architecture is the same. Why try to make the media more reliable, when there's cheaper/easier solutions that just accept unreliable media, and move on?
--
make install -not war
I have some movies on laserdisc that're pushing 20 years, and I haven't had a problem with them yet!
Much Madness is divinest Sense --
To a discerning Eye --
Much Sense -- the starkest Madness
I now have some no-name-brand CDs burned in 1998 that are still good. I have never had a good, name-brand CD fail for any reason. The only failure I have ever had was the top layer peeling off some el-cheapo CDs which were stored in plastic sleeves, not jewel cases. One BIG key the article does not mention is to store the disc where the burned surface is not touching anything, such as in a jewel case -- the article should have mentioned that. Do not put in plastic sleeves or cases with slide-in sleeves. Odd that the article is a sales pitch for that T-Y brand -- what about RiData? That's what I use for DVD archival storage. I haven't been using DVD-R long enough to comment on how long they'll last. I have always found the alarmist idea that CDs will spontaneously self-destruct to be sort of over-the-top. CDs seem much more reliable for archiving than any other medium like diskette, hard disk, USB flash, or tape. Flash is more reliable, but has to be refreshed or it will disappear.
One Maxell DVD-R I burned in Sept. 2003 went bad within 3 years, despite every detail of the burning, readback, handling, and storage being in accord with the advice I've seen posted. An email to Maxell support on this issue had the reply: "The media if stored properly will have a life of at least 50 years."
- spots.html
Possibly relevant, I noticed an internal pattern of small spots visible with a loupe or macro lens (on order of 10 microns in size; much larger than the data pits). You can read more about it here: http://www.bealecorner.com/trv900/DVD/Maxell-DVDR
Maxell America agreed to take back this DVD for analysis. As instructed I sent it to their Fair Lawn, NJ site. It was received Oct. 5 2006 and Maxell acknowledged receipt. They have apparently done nothing with it since, despite several emails to them in the ensuing two months.
The reason being that DVD+/-R has the recording surface sandwiched between the two layers of plastic. CD-Rs have the recording surface on top, which can flake off unless you handle it very carefully.
Sure, you can handle the CD-Rs carefully and avoid this problem. But wouldn't you rather use a more reliable medium in the first place?
AccountKiller
Here's my two cents:
Stay away from Taiyo Yuden 16x media. I'm using a BenQ 1620 for all my DVD burning needs, and the PI/PO tests done on T-Y 16x media using DVDInfo Pro have always resulted in low quality burns. I am currently using 8x Verbatim DatalifePlus DVD+R media, and burning them at 4x. The results are truly unbelievable. The media code on the 8x Verbatims is MCC 003. I've heard through the grapevine that T-Y changed their media somehow from their 8x sets to their 16x sets, which has resulted in that the 16x DVD+Rs aren't as good. If you can get your hands on 8x T-Y DVD+Rs, then go for it; otherwise, stick with the Verbatim DatalifePlus series.
-BB
I have found this dvd media quality guide to be extremely informative. Yes, Taiyo Yuden is always ranked at the top (and is what I use), but they are not readily available at local retailers. It really helps to have a detailed comparison of various media instead of just saying "brand X is best".
Actually, the pits have a depth of 1/4 wavelength of the laser, so that the light that is reflected from the bottom of the pit travels 1/2 of the wavelength longer and cancels out the light that is reflected from the land.
That would be about 263157894s. or about 3046 days - more than 8 years.
Additionally, I really don't get his argument about the ATIP and wobble. If your DVD-R has degraded so badly that you can't read the ATIP before burning it, you probably don't want to use the disc for long term archival anyway. He goes on about the error correction of the data in the ATIP, but as far as I know, the ATIP is only used to determine the recording strategy, and should be of no relevance to reading the DVD - after all, the relevant part shoud follow the same standard as pressed DVDs so that the "new" DVD-R and DVD+R media are compatible the old drives that predate the recordable media. The same compatibility argument holds for the encoding of the ATIP data itself - if it differs from the pressed DVDs, it can't really be important for reading the medium.
Well, if you're going to be that pedantic, the data still goes through an encoding layer called "Eight to fourteen modulation", which guarantees that a pit / land will be at least x bits and no more than y bits long, so the laser won't lose tracking due to not finding any transisions in a long time.
Oh, I'm so tired of these articles. Everyone concentrates on dye fading, because I guess it's easy to measure and quantify. If dye fading were the failure mechanism for these disks, they'd last twenty to two hundred years... according to vendors and researchers.
Everyone says "I've never had any trouble with brand ABC," but the thing is, ABC varies depending on what you read or who you talk to. Some people insist they've never had any trouble with the cheapest generic products they buy at Staples. Some say any name brand is OK. Some say Verbatim is good. Some say to stay away from Verbatim. The more sophisticated will tell you not to use anything but phtalocy- pthalocy- pffthal- the Mitsui stuff. Others (like this guy) are partial to other dyes. Some say you're a fool to use anything but Mitsui Gold... some say they're an overpriced waste of money.
It's all authoritative sounding talk, talk, talk and no two experts say the same thing.
In reality, I don't think anyone understands very well what actually causes these disks to fail in the real world. I've had disks fail in less than two years--maybe only a couple-three in many hundreds, but certainly not zero--and I've never seen any obvious pattern as to which of them fail.
The thing that really bothers me is that drives and/or their accompanying software drivers never give you any indication of what the signal quality of a particular disk is. If they did, you could detect that a disk was deteriorating before it failed, and make a copy. As it is, they just keep silently keep correcting errors behind your back and you have no warning until there is utter, catastrophic failure.
"How to Do Nothing," kids activities, back in print!
That realy depends today.
.002 cents.
If you have a large organization and you're backing up terabytes+ daily, then yes with incremental. You can probably afford the $37,000+ for a TB storage solution.
A good bit of small businesses really don't have more than a couple hundred gigs that need to be backed up and the nightly stuff is probably under a gig unless you're in the media business.
Tape backup for archival is a horrible solution. You're dependant upon the media and the media player and in the case of Microsoft, the OS as well. BackupExec issues anyone?
How many of you had that backup you made to tape take forever to retrieve where that happenstance copy you made to another server work just fine?
Tape backup for `backup` is a fine solution but not for long term storage.
When marketing companies are finished with projects and they need to be removed from the server, they are archived on CD/DVD media.
If is probably cheaper to buy about 20 500 Gig Hard Drives and run a backup scheme using that media than to spend $2,000 on a 120Gig DLT tape drive that will take 6 hours to make a backup.
I'm not really discounting the usefulness of tape backup, I just question if it is relevant today when disk storage is cheaper.
My
if you steal from one source, that is plagiarism, if you steal from many, well, that's just research.
Of course, if a disk gets a read error in the sector holding the FAT, I'm toast. :-)
Maybe not. PAR2 files store the filenames as part of the recovery data. As long as the TOC track (innermost track) isn't kaput, you can recover the data even if both the UDF and older 8.3 file tables are blitzed. If the TOC is busted, you'll have to get a professional DVD reader or go to a recovery service.
(The "how to" is over on the QuickPar forums. Basically, you rip the disk at the sector level to a pair of files. Rename the one with a PAR2 extension, then feed it the 2nd file as source data. QuickPar will find all of the data blocks and reconstitute the original files.)
Wolde you bothe eate your cake, and have your cake?
Why not just get a NAS that has RAID? That would make more sense. When a disc dies, you can replace it, rebuild your array, and everything is fine. PLUS, you could expand your archive over time.
How about NAS RAID and UDO all in one system, the Archive Appliance.
Am I the only one to be suspicious about the impartiality of this article? Check out the links. Quite a lot go through the weird ass domain name JDOQOCY.COM. Do a whois on this domain and you'll find the registrant is "Commission Junction". Hmm, impartial, NOT.