Burn damage resistant CD/DVDs

[Guest Olof]

Advertisement (gone after registration)

any experiences with the following software ?

 

TrueDisc

[farnz]

Jan,

 

I'm not familiar with True Disc but it doesn't appear that they use phthalocyanine dye so, irrespective of their impressive physical robustness, the dye itself (probably AZO or cyanine) may not not last 10 years depending on how they're stored.

 

Pete.

[dist]

I gave the trial version a spin (saves a file up to 5 MB in size) and it did work (I applied some greasy fingerprints to the CD ) but I guess the show stopper is that you need a copy of the software to get it back off again. Well maybe not (at least that's what they say on the website; at least one copy is supposed to be stored in an ISO9660 file system) but I failed to see any evidence of this.

Which, I guess, is the major complaint here: they seem to be very secretive and feed you a lot of marketing mumbo jumbo but nothing tangible in terms of technical explanations, yet they expect for you to dish out 89 bucks (wait, it's discounted for a limited time - a real bargain at only $52!), and you don't really know anything about it.

Not at this time, thank you.

My current strategy is burning CD-Rs (Verbatim brand), store them under near ideal conditions (I once read you're supposed to treat a recorded CD-R like you would treat high quality chocolates: store in a cool dry place, away from light sources), plus replication of image files across several PCs, at least one of them off-site.

Cheers,

[Guest Olof]

i burne also a picture onto a disk, and than made 4 deep skratches into the disk.... no problem at all for TrueDisk

[iomatic]

If you really want to archive your stuff, use gold DVDs, make two copies in completely separate locations, use a hard drive backup, add an off-site enterprise-class secure server, and put tape backups in a bank vault. Or pick two of these options.

 

Remember, it's not a matter of if, but when your hard drive --and even your backup -- fails.

[SteveP]

Interesting - they may be using an approach similar to audio CD where the data is encoded redundantly. If so, I would suspect a reduction in the size of the amount of data you could store on a CD - possibly by as much as half.

 

If it does work, it could be paired with some of the "tough" recordable CDs for added insurance. But these methods really only protect against physical damage, which an archive is unlikely to experience in many cases. It won't resist fire, theft or being left on the train. And if the dyes fade, they fade. And of course one of the greatest threats is a lack of a reader/player as formats become obsolete and operating systems progress to the point they won't support old players. That's why I back up on 8-inch floppies :-)

[wparsonsgisnet]

Advertisement (gone after registration)

A friend of mine says there are 2 kinds of people in the world: those who back up and those who will.

[onitunes]

Hello,

 

I'm Erich Ocean, the developer of TrueDisc. I've drastically improved the technical information on the main page at http://www.truedisc.com, and gone into even more detail on the TrueDisc developer blog at http://blog.truedisc.com/?p=4.

 

I would be happy to answer any questions you may have. Many of the features currently in development are targeted specifically for photographers, and I sincererly appreciate your feedback.

 

Best, Erich

[dist]

Thanks Erich, that's a lot more like it! Digesting the info now

[farnz]

Hello,

 

I'm Erich Ocean, the developer of TrueDisc. I've drastically improved the technical information on the main page at TrueDisc, and gone into even more detail on the TrueDisc developer blog at TrueDisc Developer Blog » Blog Archive » The Technology Behind TrueDisc.

 

I would be happy to answer any questions you may have. Many of the features currently in development are targeted specifically for photographers, and I sincererly appreciate your feedback.

 

Best, Erich

 

Erich,

 

Welcome to the forum!

 

Truedisc looks like a novel and worthwhile approach to archiving but I need to be convinced that the dye will last too. Do Truediscs use phthalocyanine dye and, if not, what is the service life of the dye that is used?

 

Pete.

[SteveP]

It appears the product is the software. To quote from the website

 

"Do I need to buy a special burner or special discs to use TrueDisc™?

 

With TrueDisc™, there's no special hardware or media to buy; it works great with the burner and blank discs you already own. And to make it easy for you to share your TrueDiscs with friends and family, a free copy of the damage-correcting TrueDisc™ reader is automatically included on every disc you burn."

 

There is a variable (depending on degree of "recoverability" desired) reduction in the capacity of the media used.

 

There are also "robust" discs which are more resistant to physical damage available from other sources.

 

My spies suggest gold CDRs (never use CDRW for archiving) have the best chance of lasting a while. Archiving to DVDs, I'm told, is extremely risky.

 

There is a simple way of achieving redundancy. Buy CDRs in large quantities to reduce the unit cost and then burn multiple backups, which are then stored in physically separate locations. This has the benefit over a product like TrueDisc in that a single event (fire, flood, 3 yr old) will not destroy the archive. Of course, you could do both :-)

[Guest sirvine]

I've heard the opposite--that DVDs are more robust than CDs in the long term. This was explained as having to do with additional layers of plastic used on the exterior of DVDs, which might be utter nonsense for all I know. Or was your comment suggesting that DVDs are riskier because there is more data to lose per disc?

[SteveP]

I've heard the opposite--that DVDs are more robust than CDs in the long term. This was explained as having to do with additional layers of plastic used on the exterior of DVDs, which might be utter nonsense for all I know. Or was your comment suggesting that DVDs are riskier because there is more data to lose per disc?

 

CD standards have been around a long time and were one of the few technologies where the big players actually agreed to cooperate on from the start. The technology of storing and retrieving the data from CDs is well-proven.

 

DVDs, on the other hand, were born (like VHS and Beta) in the middle of a wrestling match between two divided camps. The technology is less than ten years old commercially and is essentially an outgrowth of CD technology.

 

The physical characteristics of polycarbonate, dye and reflective layers are much the same for both CD and DVD (single layer) formats. Obviously the data is much more densely packed in a DVD, so the machinery and processes which make them are more likely to cause faults. AFAIK, a DVD has exactly the same amount of plastic as a CD.

 

And keep in mind that while the "read" side of either disc can be scratched, it has more than half a mm of polycarbonate in front of the data. The back "label" side, however, has just some paint (maybe), lacquer and the reflective coating. Very easy to damage.

 

But that's just my pondering. Apparently, in archival circles, DVDs are a no-no.

[Guest sirvine]

Well, I guess I'll go back to archiving on CD's again! Good thing I bought 100 gold CDRs instead of the spindle of 50 DVDR I meant to order. Your logic seems sound--CD players are probably more robust at getting data out. I don't think it matters too much, though, since I'm always going to need a top-flight DVD player for my huge collection of movies.

[ennjott]

par2 is free software (free as in open source), you can generate parity data with it, similar to what a RAID volume uses. It's very stable and proven to work, I have been using it on 240 GBs of data at a time (and yes, I did test restores successfully).

That TrueDisc info page talks about a free reference implementation, so it should be ok to use once it's stable. I would not trust proprietary software or formats for data backups, ever.

 

If you regularly burn DVDs or CDs, something you will want to do is get a drive and software and that can report C2 errors resp. PO and PI errors (party inner / parity outer). Newly burnt DVDs can have a "PI sum of 8" error value of as little as 30, like a factory pressed DVD, or as much as 700 without you noticing any differences in usage. AFAIK the DVD standard allows 300.

There is a hard limit to how many low level errors a sector can have before the error correction (reed-solomon method) cannot handle it anymore, you will immediately get a read error once this limited is exceeded. When I still burned DVDs (switched to hard drives a while ago), I checked each batch for this.

 

Fact is, DVDs offer too little space and checking errors on them requires special soft- & firmware. Hard drives store diagnostics data that is easy to read out, among it the reallocated sector count and ECC recoveries. And since you won't have hundreds of discs like when using DVDs, you can easily store multiple copies at different places, and test these every now and then. Serial ATA mobile racks sell for <10 bucks and work great. My verdict: DVDs for data storage - don't bother unless it's only temporary.

[onitunes]

Pete, as other users have noticed, TrueDisc does not specify (or require) specific kinds of discs, dyes, etc. Of course, using "better" discs doesn't hurt and probably helps.

 

Steve wrote:

There is a simple way of achieving redundancy. Buy CDRs in large quantities to reduce the unit cost and then burn multiple backups, which are then stored in physically separate locations. This has the benefit over a product like TrueDisc in that a single event (fire, flood, 3 yr old) will not destroy the archive. Of course, you could do both :-)

 

TrueDisc has a much better form of redendancy than simply making copies. Depending on the specific scenario, you could require as many as 16 identical discs to reach the same level of damage protection TrueDisc achieves with one.

 

Steve's suggestion to do offsite backups (with TrueDisc) is right on; that's what I do.

 

ennjott wrote about various hardware-level detection strategies. Although I could make use of any of these to speed up the operation of TrueDisc on some machines, none of them are necessary in general and TrueDisc works fine without them. I have written TrueDisc to work well on lowest-common-denominator hardware.

 

Thanks for posting! Best, Erich

[ennjott]

I do think testing each batch is a good idea, even if you have protection through parity data. If you burn a whole bunch of discs with all of them near the maximum correctable block error count, the parity won't help you when after 1 or 2 years all of these discs have become completely unreadable.

 

par2 is already available as a stable implementation of an additional level of error correction through Reed-Solomon. Integrating this level into the structure of the disk and drive firmware sound promising. However, the technical info isn't specific enough and doesn't even include information about RS encoding. And you've probably heard the phrase: show us the code =).