| CD
/ DVD Data Rot
How
reliable are CD's and DVD's as backup? Well... somewhat.
I had
to restore media from way back, all stored on CD and DVD before (external)
500GB+ harddrives became affordable. I've been copying almost all of the
material back to HDD, and here are my results... This isn't too scientific
:-) just my personal experience with several brands...
Update:
I found another batch of discs, and am in the process of adding the results
to the table below.
Drives
First,
I am using four different DVD / CD drives to read the disks:
- external
USB Liteon eTAU108
-
external USB MaxCube CD-SL07-U2S / Teac DC-W28S-R
-
internal IDE Sony Optiarc DVD RW AD-5170 ATA
-
internal SATA (reported as SCSI) Hitachi (Dell) HL-DT-ST DVD+-RW GSA-H31N
Sometimes
one drive would not be able to read a disc but then one of the other two
drives would stil read it. I couldn't figure out if there was a relation
between the drive that had been used to burn the discs versus the drive
used to read them. Could be head alignment, could be simply incompability,
I do not know.
Crashes
Though
Windows 7 is fairly stable, I've managed to crash the CD / DVD device drivers
more than once. The PC was still working, but sometimes wilst reading a
(failing) disc the specific drive would become locked up and would no longer
be accessible to Windows 7 until the next reboot.
Storage
conditions
Note:
most of these discs have been stored in a living room for up to 10 years,
followed by some years in the attic, including some 35+ degrees summers,
wet and dry weather, and several soft winters (I guess the attick dropped
to around zero degrees Celsius but probably not below).
Results
The
table below lists brand, year, and success versus fail numbers. All cases
where I could recover ALL files (even if I had to move the disc from drive
to drive to get all files) I counted as successes. All discs where I could
NOT recover ALL files I considered fails.
Most
discs were the cheapest I could find, burned in the period 2002 to 2006.
I haven't seperated CD's and DVD's, as over time CD's were phased out and
replaced by DVD's. I have been trying to read them a number of years later,
as marked in the header of the table below.
DVD's
have a worse track record than CD's. As some of those contained movies
and were handled by my kids that doesn't necessarily indicate DVD's are
by definition worse over time (though I suspect they are, or are at least
more sensitive to damage). The majority of these disks have seen little
to no use.
The
LifeTec CD's and Mmore CD's were mostly fine, but the reflective layer
was becoming brown and transparent. Not a good sign...
I've
added one line at the bottom to indicate factory printed CD's / DVD's that
failed as well, but please note these often suffered scratches from (poor)
handling so these are not representative of 'aging' as such. Though I have
a few original factory prints that are slowly becoming more and more 'transparant'.
Good candidates for early retirement :-(
| Brand |
15+ yrs |
15 yrs |
14 yrs |
13 yrs |
12 yrs |
11 yrs |
10 yrs |
9 yrs |
8 yrs |
| Nobrand / printable |
- |
- |
7
/ 1 |
6
/
2 |
- |
1
/
1 |
1
/
0 |
4
/
0 |
- |
| Nobrand / non-printable |
- |
2
/
0 |
3
/
0 |
2
/
0 |
4
/
0 |
- |
- |
- |
- |
| Aterra Media |
- |
- |
3
/
0 |
2
/
0 |
1
/
0 |
- |
- |
- |
- |
| Bruna |
- |
- |
2
/
0 |
- |
- |
- |
- |
- |
- |
| Double Diamond Digital |
- |
- |
2
/
0 |
3
/
0 |
- |
- |
- |
- |
- |
| Getcopy |
- |
- |
3
/
0 |
- |
- |
- |
- |
- |
- |
| Imation |
2
/
0 |
5
/
0 |
8
/
0 |
7
/
0 |
1
/
0 |
- |
- |
- |
- |
| Lenco |
- |
- |
4
/
0 |
- |
- |
- |
- |
- |
- |
| Lifetec |
5
/
0 |
7
/
0 |
33
/
0 |
8
/
0 |
- |
- |
- |
1
/
0 |
- |
| Maxell |
- |
- |
3
/
0 |
2
/
0 |
1
/ 0 |
5
/ 0 |
- |
- |
- |
| Memorex |
- |
- |
- |
2
/
0 |
- |
3
/
0 |
- |
- |
- |
| Mmore |
2
/
0 |
- |
34
/
1 |
20
/
1 |
1
/
0 |
- |
- |
- |
- |
| Nashua |
- |
- |
- |
- |
2
/
0 |
- |
- |
- |
- |
| Octron |
- |
- |
11
/
0 |
6
/ 1 |
2
/
0 |
- |
- |
- |
- |
| Philips |
- |
- |
- |
2
/
0 |
4
/ 0 |
- |
2
/
0 |
- |
- |
| Platinum Plus |
- |
3
/
0 |
- |
2
/
0 |
14
/ 1 |
- |
- |
- |
- |
| Precision |
- |
2
/
0 |
2
/
0 |
2
/
0 |
- |
- |
- |
- |
- |
| Pleomax (Samsung) |
- |
- |
- |
- |
2
/ 0 |
- |
5
/
0 |
1
/
0 |
- |
| Samsung (non-Pleomax) |
- |
- |
6
/
0 |
6
/
0 |
- |
- |
2
/
1 |
- |
- |
| Sony |
- |
- |
- |
- |
- |
- |
- |
1
/
0 |
- |
| TDK |
- |
2
/
0 |
1
/
0 |
1
/
0 |
- |
- |
- |
- |
- |
| That's Wite |
- |
- |
- |
- |
- |
- |
1
/
0 |
- |
- |
| Thing Xtra |
- |
- |
23
/
1 |
9
/
0 |
- |
- |
- |
- |
- |
| Verbatim |
- |
- |
- |
8
/
0 |
- |
- |
- |
3
/
0 |
- |
| Original
Audio CD (not a CD ROM!) |
x
/ 2 |
- |
- |
- |
- |
- |
- |
- |
- |
Conclusion
1.
This is obviously not the most scientific test :-)
2.
Average failure rate < 2%.
3.
Age and storage time did not seem to matter much... yet
4.
Read speeds varied, but many disks would only read 100% at slower speeds
5.
No brands that stood out
6.
Even original factory printed audio CD's suffer (but probably mostly from
handling)
|
