Version: 2008
  • On BNET: Is the Mac finally ready for the office?
Advanced Search
advertisement
advertisement
Click Here

Forum display:

Storage: Do burned CDs and DVDs have a shelf life?

by Lee Koo (ADMIN) Moderator - 2/13/09 2:55 PM
advertisement
Click Here
Post 136 of 257

About CD/DVD Lifespans

by mssusanf - 1/31/09 8:59 AM In reply to: Do burned CDs and DVDs have a shelf life? by Lee Koo (ADMIN) Moderator

Data Rot or Bit Rot are the terms used to describe this. In a way these are misleading as data itself has no physical properties to rot. However, the physical media data is stored upon can do exactly that - rot. It only takes a tiny degree of physical degradation, usually quite invisible to the human eye, to affect data. Older storage media such as floppy disks and magnetic tape were more prone to "rot" phenomena as they could also lose their magnetic alignment over time.

CDs and DVDs made of high quality material have little to no ability to "rot" but the usual lower end disks often can and do lose data over time. CDs can lose their reflective coating over time and thus become useless. This type of "rot" is called "bronzing" and is one that is visible to the human eye. The disks gradually turn brownish in color usually beginning at the outer edge of the cd and spreading toward the center almost like a fungus. This problem, thankfully, is not as common today as it was in the 1990s although older disks from that time period may yet become "bronzed" so if you have any, it would be a very good idea to make copies on newer disks.

CDs can lose their reflective coating in other ways too - oxydation, scratches, even melting in a hot car. A DVD on the other hand has an extra layer of plastic atop the reflective section thus is less prone to problems of this type. They are not without their own issues though. Many DVDs have proven to be prone to a problem called "delamination" which is oddly similar to the older "bronzing" problem of CDs. You can read more about the physical structure of CDs and DVDs and their similarities and differences at http://computer.howstuffworks.com/cd1.htm and http://electronics.howstuffworks.com/dvd.htm. Also, it is worth mentioning that CD-RW disks are quicker to degrade than non-rewriteable disks. Their material is easier to melt or become otherwise unreadable than the material used to make non-rewriteable disks.

The general consensus for extending CD and DVD life often comes down to how they are stored. It seems most claim that CDs and DVDs should be stored in jewel cases for optimum "shelf life". This allows the recordable surfaces of the disks to remain effectively untouched while stored. Paper jackets, envelopes, and CD binders are considered to be not so good as the disk surface is continually in contact with the storage device. However, such methods of storage for CDs and DVDs are considered better than no protection at all.

In verifying my facts for this response, I ran across something else I'd not seen before which seems worth mentioning here - Apparently some claim that specific data formats have better archival abilities than do others. The article states that the file type used can indeed matter and claims that the more stable formats are: ASCII text, PDF, RTF, PNG, TIFF, JPEG, MP3, WSAV, AIFF, MPEG, and DBF. It lists as unstable formats DOC, GIF, BMP, AU, WMV, RA, AVI, MOV, and all Excel and MS Access file formats. I believe they were also adding in to their measures of stability the likelihood of such formats being able to be read by tomorrows devices but this is something that bears looking into further. I found this blurb within a rather long document at: http://www.slais.ubc.ca/PEOPLE/students/student-projects/C_Hill/hill_libr516/print.htm - it is listed within the document under the heading of "The Future".

Wikipedia has a good article on CDs that includes much information on the various types of dyes used to produce the coating on CDs. Apparently the type of dye used directly affects the "shelf life" of a CD. This article can be found at: http://en.wikipedia.org/wiki/CD-R The same article itself echoes what seems to be most often written - that lower end CDs tend to have a useful life of anywhere from one to five years but that higher end CDs effectively have no specific limit to their lifespans as yet. Some manufacturers are claiming archival life of as much as 300 years with the more common claim being 100 years for archival quality CDs. MAM-A (http://www.mam-a.com/) claims a 300 year shelf life for their Gold Archive CD disks and a 100 year shelf life for their Archive Grade DVDs. Other brands known for longevity are Taiyo Yuden, Ricoh, TDK, and Verbatim. A caution here though - most of these manufacturers also make low end disks so be sure the disks you are buying state that they are archival quality.

I will end here before I write a book but want to mention one more thing: As with any and everything, if it matters to you, back it up, back it up, and back it up again! Anything important should have at least 3 copies with at least one stored in a separate location in case of disaster. To avoid the issue of degradation as much as possible, it is not at all a bad idea to take a day or two each year and re-backup your backups. In other words, copy all your archives to yet another, new CD or DVD. Copy your oldest copies each year and the degradation issue should not be a problem at all. If this is Just Too Much to do then you might also want to copy your CDs and DVDs onto an external hard drive or two and at least re-copy that every year. This of course would be an easier and less time consuming process as theoretically you could set the copy process in motion and leave it alone til done whereas with disks you need to be present to continually change them out.

Happy copying!

Post 137 of 257

'Data rot' and what not

by Culebra - 2/13/09 5:37 PM In reply to: About CD/DVD Lifespans by mssusanf

First, the bogus stuff: the claim that certain file formats are more prone to data loss based on degradation of the physical media is a pretty rotten notion in and of itself. As the bit patterns of digital information are practically random, there is nothing inherent in the data that would cause this. File formats which contain no magic headers or keys that are needed to read the entire file would have a better chance of being recovered given a random corruption that affected the header. ASCII text would be a main example of this, and mssusanf lists that as a file format as one that would MOST likely be lost instead of others. This is all poppycock, sorry to burst your bubble.

Apart from scratches that irrecoverably interrupt the serial data spiral of the CD from the center out, the main reason for data loss on CDs or DVDs of any kind is quite simple: oxidation (rust). CDs and DVDs which are not gold will oxidize (rust, i.e., 'rot') in around 10-15 years, depending completely on degree of exposure to factors which would accelerate the chemical reaction that is oxidation (rust). These factors are, of course, such things as heat, water, oxygen and catalysts (substances which speed up chemical reactions).

The only way to preserve data in the digital age is to have multiple copies of the data on more than one media type, AND -- regenerate/recopy the redundant backups onto new media periodically.

Post 138 of 257

Absolutely right on!

by vattan2005 - 2/13/09 9:05 PM In reply to: 'Data rot' and what not by Culebra

Yeah, I agree 110%!! File formats make no difference to digital technology. All the computer knows and reads are 1's and 0's. It couldn't care less what files formats reside on the medium.

Post 139 of 257

Does data format affect proneness to data loss?

by Robert de Strang - 2/14/09 8:55 AM In reply to: Absolutely right on! by vattan2005

I'm no expert in this area, but in principle I don't see how Vattan2005's posting can be correct because it disregards the effect of redundancy (the same would apply to Culebra's first paragraph). Some formats may have more redundancy built in than others and would therefore be inherently less prone to corruption.

I don't know enough about the actual formats to say more. However at the very least I would expect compressed formats to be particularly vulnerable, because they deliberately get rid of redundancy to save storage space.

I used to show my students how if one uses the Huffman-coding method for data compression, just one bit wrong can reduce, say, English-language text to gibberish for several words until the boundary between the variable-bit length representations of the characters happens to fall back into the right place.

Do post again if I'm missing something.

Post 140 of 257

data compression not the same thing as media compression

by Culebra - 2/14/09 12:14 PM In reply to: Does data format affect proneness to data loss? by Robert de Strang

Data formats do not have redundancy (which would imply multiple copies of data in a file. Any data compression technique which can restore the entire file with no data loss does not eliminate information, it simply encodes it by taking advantages of (arbitrary) sections of a data object which have repeating patterns (consecutive bits or bytes of the same value.

Lossless data compression is not the same thing as image or audio file compression in which image quality or audio fidelity is compromised in order to reduce the size of a streaming media file, in which case there is information loss coming out of the encoding before the file is written (or transmitted).

Recovering data from a compromised data file or object (in which the objective is to recover as much information as possible ) is not the same scenario as determining when a CD or DVD can or can not be read. Data written to a CD or DVD is done so either as a stream of bits (like tape), or structured into blocks and accessed by a file system also included when the media was written. Most application software, when encountering an area of data corruption in a stream written to CD or DVD as if it were a serial tape, fail and can not ( or do not ) continue. A single error is usually enough to corrupt a restore -- it is just a matter of at what point in the data stream the error occurs as to how much information can be restored prior to a failure.

If Hamming encoding or some other form of one-bit or two-bit error correction is employed (similar to the schemes used in RAM memory bit error recovery), a backup/restore application using this technique can potentially recover from a bit error in reading the stream. However, if the device driver is unable to continue to track the stream after a data error (failure in flag or sync bits or bytes used to synchronize frames or blocks ), full recovery would not be possible.

The randomness of media surface corruption from oxidation pays no attention to data formats, of course, and it is strictly a function of how pronounced the data loss in the serial stream and the heuristics of the driver as to whether or not the driver can continue to read the stream after an error, with or without data loss in the corrupted section.

Once again, we are not talking about mirrored data objects or other types of redundancy -- and most applications do not write redundant copies of data when they save data in application-specific file formats.

Post 141 of 257

Correction

by visualwarp - 2/13/09 9:38 PM In reply to: 'Data rot' and what not by Culebra

I think you misread the original message. ASCII was listed as one of the stable file formats. I believe the file formats listed as unstable are the ones vulnerable to data corruption regardless of media type or the cause of the corruption. I'm not saying it's all 100% correct - but I think you may have disregarded the information too hastily. There could be something to it so I'd keep an open mind.

Post 142 of 257

my bad if I flipped the bits

by Culebra - 2/14/09 12:45 PM In reply to: Correction by visualwarp

Fair enough, we agree that simple, unstructured ASCII text streams/files are the simplest to recover after an I/O error (be it on a block-structured file system, a streaming tape, or a comm line ) -- assuming that the device drivers, OS, and applications involved have the ability to continue after I/O errors. Given that, there would be a higher probabaility of recovering additional data after a data corruption for, say, ASCII data. But data lost would still be lost, even if additional information could be recovered after a loss.

My only point was that an incident of data loss caused by optical media corruption by oxidation is not specific or particular to any file format, and can occur anywhere in the data stream. As most applications will fail and give up with an I/O error, and many applications do not write using bit-level error-recovery encoding schemes, failure to recover or restore (completely) data from optical media due to media surface corruption is not correlated with file formats.

Post 143 of 257

Different programs

by wycca28 - 1/31/09 9:39 AM In reply to: Do burned CDs and DVDs have a shelf life? by Lee Koo (ADMIN) Moderator

I have been trying to find the answer to a similar problem.
What is the most secure method of storing documents and photographs?
I have discovered that some of my cds have info on them that was created by older programs I had on an old computer and now they won't read because that program is no longer available.
Should you change all pics to JPGs?
Any info greatly appreciated.
wycca28

Post 144 of 257

GOOD TYPE OF STARAGE FOR PHOTOS ETC.

by R1000 - 2/13/09 10:42 PM In reply to: Different programs by wycca28

Several items I have read in the past say that digigal memory, like a large thumb drive stores quite a bit of data and does not deterioate unless physically destroyed. As there are quite large units available, it may be feasable to go to this kind of storage for a more permanent situation. 8 gigs will store quite a lot of photos etc.(there are much larger units available.) They are small and easy to store. Would be quite interesting to know for sure if this is a better way or not. As the size goes up and the price goes down, it may be something to look into for a secondary storage just to see if it will work. As they haven't been out long enough to know for sure, would be interesting to find out..............
R1000

Post 145 of 257

CD shelf life

by photoguy38 - 1/31/09 10:00 AM In reply to: Do burned CDs and DVDs have a shelf life? by Lee Koo (ADMIN) Moderator

From my own experience, they do have a shelf life, but it's really more related to the amount of playback, and not length of time stored. I have burned many CD's and they play fine for a while, then begin to get corrupt. At that time, I burn a new copy. If they are stored after burning and not played, they play back months and even years later. The CD's and DVD R's you buy in the stores have a burning surface which is ink and that's what the data burns into. Evidently the ink degrades with use. Professionally made CD's have a different medium and are much more permanent.
I have a master external hard drive which I store my favorite files on, and burn CD's off it when I need them. I avoid using the external to run the files to preserve them and have never had a loss this way. And with the terabite hard drives available, you can store as many files as you want. Again, I just use burnt CD's as a temporary playback medium.

Post 146 of 257

Playback is a factor due to laser energy

by Culebra - 2/13/09 5:43 PM In reply to: CD shelf life by photoguy38

Playing a CD exposes the CD to the laser. However low-energy that laser is, it is a source of energy and heat. The reason that CDs and DVDs with metal surfaces will suffer dropouts in their encoded data eventually ( on the order of 10 years even without play ) is quite simple: the metal oxidizes (rusts). Only glass or gold (which does not oxidize) CDs avoid this problem which all other CDs have built in as a terminal illness.

LPs, on the other hand, will last a lot longer, because the vinyl does not oxidize. It could melt, or break, or turn brittle as the plastic polymer molecules lose their elasticity (e.g., if frozen). But ... they won't rust !

Post 147 of 257

I would say yes, they are limited.

by w3syt - 1/31/09 10:05 AM In reply to: Do burned CDs and DVDs have a shelf life? by Lee Koo (ADMIN) Moderator

I have certainly found failures in my transferred home movies after a very few years.

Post 148 of 257

They will fail.

by stevem291 - 1/31/09 10:27 AM In reply to: Do burned CDs and DVDs have a shelf life? by Lee Koo (ADMIN) Moderator

Your friend is right, they will fail. But the question of when is a big unknown. I too have researched the problem and found out a lot, and also from first hand experience. There are so many factors involved. Such as, quality of the disc, speed of burn, storage enviroment. I mainly deal with DVD's but some CD's also. I remember always reading disc will last 100 years, but you no longer see that in any info. Hard drives have come down in price to where it is as cheap as buying disc and cases to put them in. I have had several video disc to become unreadable within several years. These were stored in a climate controlled space, in a closed case. But I must also mention, these were cheaper off brand disc, I no longer use such.
You can google info such as dvd rot, http://hometheater.about.com/cs/dvdlaserdisc/a/aadvdrota.htm
Home burned disc are different than factory disc, and usually home burned are more likely to fail sooner. If its something you really want to keep for the long haul, use quality disc, burn at a slower speed, and back up to a hard drive. I have had the best luck with Verbatim disc, they seem more consistant with quality than other brands, even big name brands. More cd data here>
http://www.loc.gov/preserv/studyofCDlongevity.pdf
One more thing to note, cd's and dvd are different in that the top of the cd is very sensitive to sctatches, most people think the bottom, or burn side is. On a cd, the top is a reflective cover, much like a mirror, scratch the front of a mirror and you might be able to polish the scratch out, scratch the back of a mirror, and its ruined. Also, never ever put labels on a disc, and be careful with what you use to mark them. Hope this helps. I learned the hard way.

Post 149 of 257

The Simple Answer

by iscnitzan - 1/31/09 10:38 AM In reply to: Do burned CDs and DVDs have a shelf life? by Lee Koo (ADMIN) Moderator

Yes! they have a shelf life, most are 8 to10 years.

Post 150 of 257

Yes, CD/DVD-R's have a "shelf life".

by Bubba_Gump - 1/31/09 11:06 AM In reply to: Do burned CDs and DVDs have a shelf life? by Lee Koo (ADMIN) Moderator

The length of time the data on user-recorded CD's & DVD's varies greatly on a number of factors, but they do indeed have a "shelf life".

Assuming you are not talking about "avoidable" damage, such as fingerprints, scratches, dust, crayons, dog-slobber, etc, to the disc surface, yes, there ARE things you might need to be concerned with regarding disc longevity:

It helps to understand how CD-R's (and by this I include CD+R's as well as DVD's) work to understand how durable they are.

NON-Recordable discs have "pits" burned into the surface of the disk that deflect the laser while it spins. Whether the laser strikes the sensor or not determines if it reads a "1" or "0".

RECORDABLE discs use heat-sensitive colored dye sandwiched between layers of the disc. When you record, the laser changes the color of the dye in place of burning "pits". When the disk is read back, the points of different color reflect the laser differently, resulting in the same effect as noted above (to determine a "1" or "0").

Since the discs are "heat sensitive", storing your discs in a warm place can cause the dye color-change to "even out", making the data difficult to read ("Extreme" heat will ruin the disc itself long before the dye is affected.) Really cheap bargain discs are more susceptible to heat damage and/or use cheaper less-durable dye. It is important to keep all your CD-R's in a cool/dry place (even blank ones) when not in use (and why it's a bad idea to leave them in your hot PC when not in use.

"Heat" issues aside, simple "time" can cause the contrast of the ink to fade. And again, the cheaper the disc, the less durable the ink may be. Over time, the difference in contrast between "bits" on the disc become more difficult to discern, making it harder and harder to read the data on them.

But even low-quality discs, well-taken care of, should last years. I have some CD's I recorded nearly a decade ago that I can still read.

"High-speed" discs have more sensitive dye, which are more susceptible to the effects I mention above. But, if you are concerned and the data is super-important, you can counter this by recording at a lower speed than the disc is capable of (50% slower or so), ensuring a more "permanent" change in the dye color. Most good recording software will let you select a recording speed before burning.

Forum legend:
Locked Locked thread
Moderator Moderator
CNET staff CNET staff
Samsung staff Samsung staff
Norton Authorized Support team Norton Authorized Support team
AVG staff AVG staff
Windows Outreach team Windows Outreach team
Dell staff Dell staff
Intel staff Intel staff
Powered by Jive Software