CD Life Expectancy
Thread Starter
CD Life Expectancy
I've just read an article about most CD-R/DVD-R/DVD+R discs becoming unreadable in as little as two or three years, due to the instability of the dye layer used.
Has anyone had experience of this, and are there any suggestions for storing data, eg photos, that you would like to keep for many years?
Has anyone had experience of this, and are there any suggestions for storing data, eg photos, that you would like to keep for many years?
Join Date: Dec 1998
Location: .
Posts: 2,997
Likes: 0
Received 0 Likes
on
0 Posts
Spoon PPRuNerist & Mad Inistrator
Re: CD Life Expectancy
I had reason to look into this not long ago from the perspective of archiving laboratory test results for clinical trials which have to be kept for at least 10 years (FDA regulations).
All the reputable manfrs. put the shelf life of an unused CD-R WORM (Write Once Read Many) in the region of 5 years plus, and the readable life of a CD-R (once created) in terms of decades.
The client was pretty comfortable with 2 CD-Rs created and stored in different locations in addition to the magnetic tape archive (DAT, IIRC) plus the on-line disk storage being able to guarantee data availability for a minimum of 15 years.
Carefully stored (not exposed to extremes of heat / cold, direct sunlight, moisture, physical damage) I am confident that data written to a branded CD-R will remain readable for at least 20 years - many manfrs suggest 50 years plus based on accelerated testing results.
A commercially pressed Alu CD-ROM is expected to last over a century, subject to storage provisos above (oxidation of the Alu layer is the killer, but gold is a tad expensive!) - however, that doesn't help if you can't write to it!
Now CD-RW is a different kettle of fish entirely and I believe that the lifespan is much less, though I haven't investigated it.
DVD disks are different again, due to the much higher data density and shorter wavelength of laser used. I would expect these to be significantly shorter-lived than equivalent CDs.
My advice is to archive data that you want to keep for your lifetime only onto CD-R media (branded) rather than DVD. It may only store 1/6 of the data of a DVD, but it will probably keep it 6-times longer, and don't use re-writable media for long-term storage.
SD
All the reputable manfrs. put the shelf life of an unused CD-R WORM (Write Once Read Many) in the region of 5 years plus, and the readable life of a CD-R (once created) in terms of decades.
The client was pretty comfortable with 2 CD-Rs created and stored in different locations in addition to the magnetic tape archive (DAT, IIRC) plus the on-line disk storage being able to guarantee data availability for a minimum of 15 years.
Carefully stored (not exposed to extremes of heat / cold, direct sunlight, moisture, physical damage) I am confident that data written to a branded CD-R will remain readable for at least 20 years - many manfrs suggest 50 years plus based on accelerated testing results.
A commercially pressed Alu CD-ROM is expected to last over a century, subject to storage provisos above (oxidation of the Alu layer is the killer, but gold is a tad expensive!) - however, that doesn't help if you can't write to it!
Now CD-RW is a different kettle of fish entirely and I believe that the lifespan is much less, though I haven't investigated it.
DVD disks are different again, due to the much higher data density and shorter wavelength of laser used. I would expect these to be significantly shorter-lived than equivalent CDs.
My advice is to archive data that you want to keep for your lifetime only onto CD-R media (branded) rather than DVD. It may only store 1/6 of the data of a DVD, but it will probably keep it 6-times longer, and don't use re-writable media for long-term storage.
SD
Join Date: Nov 2002
Location: 39N 77W
Posts: 1,630
Likes: 0
Received 0 Likes
on
0 Posts
Re: CD Life Expectancy
As you probably know, CDs of all sorts have a high intrinsic error rate. Fancy error-correcting codes and redundancy are used to give you an error-free result. As the CD ages, is slightly damaged, etc, etc, the error-correction has to work harder. Eventually the errors get too severe to be corrected. Few CD players give an indication of how hard the error correction has to work to achieve an error-free output.
If you really, really care, I'd make multiple "masters" and copy them every few years. You set the deterioration clock back to zero each time you make a copy since it records the error-corrected data onto a new disk. Of course do a data comparison between the copy and original.
I don't know the legal implications of the copies - you'd probably want to keep the originals as well.
I've heard that direct sunlight can wipe a recordable disk clean in a surprisingly short time. But sunlight in the UK isn't really strong very often.
I hope this is useful.
sc
If you really, really care, I'd make multiple "masters" and copy them every few years. You set the deterioration clock back to zero each time you make a copy since it records the error-corrected data onto a new disk. Of course do a data comparison between the copy and original.
I don't know the legal implications of the copies - you'd probably want to keep the originals as well.
I've heard that direct sunlight can wipe a recordable disk clean in a surprisingly short time. But sunlight in the UK isn't really strong very often.
I hope this is useful.
sc
). I have edited the link which had put 2 http's in 
Join Date: Nov 2002
Location: 39N 77W
Posts: 1,630
Likes: 0
Received 0 Likes
on
0 Posts
Re: CD Life Expectancy
I had a chance to talk to "my" CD Archival Characteristics fellow today. He is trying to help industry develop techniques so that they can rate unrecorded CD's for expected archival life. If successful, you'd find "30 year, 50 year, etc" CD-R s on the market. We are a long way from having these ratings at the present time. Even the same brands are inconsistent in their life.
One very useful factoid is that Roxio Easy Media Creator 8 now has a feature that allows one to find how hard the error correction is working when playing a CD. The parameter one cares about for CD-R's is BLER max (maximum block error rate), which has to be under about 220. For DVD's the parameter is PImax or PIEmax which must be under about 280. My expert suggested that one should keep a log of these readings vs time for each of your archive disks. This should help determine when it is important to copy the data onto a fresh disk. My expert pointed out that the same disk will give different BLER readings when played on different drives, etc.
I hope this us useful.
seacue
One very useful factoid is that Roxio Easy Media Creator 8 now has a feature that allows one to find how hard the error correction is working when playing a CD. The parameter one cares about for CD-R's is BLER max (maximum block error rate), which has to be under about 220. For DVD's the parameter is PImax or PIEmax which must be under about 280. My expert suggested that one should keep a log of these readings vs time for each of your archive disks. This should help determine when it is important to copy the data onto a fresh disk. My expert pointed out that the same disk will give different BLER readings when played on different drives, etc.
I hope this us useful.
seacue
Join Date: Jun 2004
Location: Portsmouth
Posts: 27
Likes: 0
Received 0 Likes
on
0 Posts
Re: CD Life Expectancy
One often overlooked aspect of long term archiving is whether the hardware necessary to read the media will be available in the distant future - consider the many tape backup types which are dead and gone! In many cases CD backup has been replaced by DVD ( more than one type -+R, +RW, -R, -RW, DVD-RAM) which is about to be superceded by Hi-density - so far 2 different standards fighting it out. Who has 8track audio tapes or even tapes as opposed to CDs for their audio, let alone Betamax or V2000 for video tapes!
MikeD
MikeD
Spoon PPRuNerist & Mad Inistrator
Re: CD Life Expectancy
MikeD -
I got my first cassette player about 35 years ago - while I'm not suggesting that all the tapes I ever bought or recorded are still playable, the format is still very much supported.
I have a collection of vinyl LPs that's even older and still possible to play - mind you my father's 78s have been a problem for quite some time now!
I believe that the ISO CD format will be the longest-lived of the optical storage media - it's been around for 20+ years already and showing no real signs of disappearing, partly because of the squabbling over the formats for DVD and next generation optical storage.
In the magnetic tape area I agree that you need to be rather careful - over the last 10 years there have been multiple tape formats, many of which are history already.
Perhaps it is no bad thing, due to the problems that magnetic tape has with de-lamination - I was reading recently about the awful problems that organisations have with their sound archives - oxide layer literally crumbling or peeling off the tape. Very sad - a lot of irreplaceable material is already lost.
I got my first cassette player about 35 years ago - while I'm not suggesting that all the tapes I ever bought or recorded are still playable, the format is still very much supported.
I have a collection of vinyl LPs that's even older and still possible to play - mind you my father's 78s have been a problem for quite some time now!
I believe that the ISO CD format will be the longest-lived of the optical storage media - it's been around for 20+ years already and showing no real signs of disappearing, partly because of the squabbling over the formats for DVD and next generation optical storage.
In the magnetic tape area I agree that you need to be rather careful - over the last 10 years there have been multiple tape formats, many of which are history already.
Perhaps it is no bad thing, due to the problems that magnetic tape has with de-lamination - I was reading recently about the awful problems that organisations have with their sound archives - oxide layer literally crumbling or peeling off the tape. Very sad - a lot of irreplaceable material is already lost.
Cunning Artificer
Join Date: Jun 2001
Location: The spiritual home of DeHavilland
Age: 76
Posts: 3,127
Likes: 0
Received 0 Likes
on
0 Posts
Re: CD Life Expectancy
Storing data is OK as long as you have the means of playing it back. Otherwise you should keep a hard copy of the original.
I have some of my gran's old 78 records (analogue data and hard copy all in one!). Fortunately I also have a playback device - the old wind up player she played them on. I managed to get some brand new needles on e-bay and have the mechanical skills needed to keep it running as long as I remain on the planet. I also have an original media lithograph (hard copy original) taken in 1865 of gran's mother as a baby, sitting on her grandmother's knee.
If you want to preserve digital data over a guaranteed lifetime, you're going to have to make multiple back-ups and restore and re-save them according to a regular schedule. There's no other way - but whatever you do, don't ditch the original copy if you have it!
I have some of my gran's old 78 records (analogue data and hard copy all in one!). Fortunately I also have a playback device - the old wind up player she played them on. I managed to get some brand new needles on e-bay and have the mechanical skills needed to keep it running as long as I remain on the planet. I also have an original media lithograph (hard copy original) taken in 1865 of gran's mother as a baby, sitting on her grandmother's knee.
If you want to preserve digital data over a guaranteed lifetime, you're going to have to make multiple back-ups and restore and re-save them according to a regular schedule. There's no other way - but whatever you do, don't ditch the original copy if you have it!
