Recovering audio from aging CD-Rs

On or about 8/16/06 4:34 PM, thus spake "Gregory Sigman" <sigmanohio.edu>:

> A colleague of mine has a number of old audio CD-Rs that no longer
> play well, and as the de facto local authority on all manner of
> digital audio, I was the expert he consulted for advice on
> resurrecting the non-playing tracks. These are rare recordings that
> would be difficult or impossible for him to replace, so he is willing
> to go to some trouble if necessary.
>
> I've never had to try such a thing, but I recommended he try the
> freeware "Max" <http://sbooth.org/Max/>

If these are truly audio CDs, and if Max didn't work in full paranoia mode
(and do try it on several different machines, since sometimes one CD reader
will have better luck than another), then the only thing that's going to
stand a better chance is EAC on a Windows machine. You have to use a Windows
machine because EAC there can take advantage of hardware-level features for
which there is no interface on a Mac.

It's important to prepare mentally for the possibility that the data is
unrecoverable. CDRs can "fade". I've seen a CDR go bad one month after
burning. The best known way to prevent this (in future) is to use excellent
media. The best media are said to be MAM-A:

<http://www.mam-a.com/Default.htm>

m.

--
matt neuburg, phd = matttidbits.com, http://www.tidbits.com/matt/
pantes anthropoi tou eidenai oregontai phusei
AppleScript: the Definitive Guide - Second Edition!
http://www.amazon.com/gp/product/0596102119
Take Control of Word 2004, Tiger, and more -
http://www.takecontrolbooks.com/tiger-customizing.html
Subscribe to TidBITS! It's free and smart. http://www.tidbits.com/

On 17/8/2006 11:34 AM, "Gregory Sigman" <sigmanohio.edu> spake thus:

> So I thought I'd kick the question out to the collective experience
> of the list. Any suggestions?

He may be out of luck. CD-Rs don't keep particularly well, even under ideal
conditions. The biggest determining factor in longevity seems to be the
quality of the media, but I doubt that many CD-Rs would last more than a few
years under typical storage conditions in any case. I've heard stories of
cheap discs spontaneously turning into coasters in less than a week.

From what I understand, most (all?) CD-R media are dye-based, and it's this
giving way that causes media failure. Quality of media is determined by
quality of dye used. Once it has died (ack, sorry) there's probably not much
that can resurrect it.

--
Nigel Stanger, Dunedin, NEW ZEALAND.
http://xri.net/=nigel.stanger

At 16:34 08/16/06 -0700, Gregory Sigman wrote:
>A colleague of mine has a number of old audio CD-Rs that no longer
>play well, and as the de facto local authority on all manner of
>digital audio, I was the expert he consulted for advice on
>resurrecting the non-playing tracks. These are rare recordings that
>would be difficult or impossible for him to replace, so he is willing
>to go to some trouble if necessary.

If he has a Windows system available, one possibility is Exact Audio Copy.
It may not do any better than the ones you mentioned, but it's another
possibility. It has lengthy retry logic and some people swear by it.

http://www.exactaudiocopy.de/

OTOH, I'm not sure I'd subject my failing CD-Rs to a lot of laser
stimulation, especially if they are from the early generations of CD-Rs.
It's light that releases the dye in the substrate, and excessive exposure
to sunlight is a known killer. I don't know whether the read laser is a
serious danger, but I'd want to find out.

Have you asked the various disk recovery companies whether they handle CDs?
This of course would be an expensive proposition even if they do.

Edward
Art works by Melynda Reid: http://paleo.org

On 16 Aug 2006, at 17:34 , Gregory Sigman wrote:
> A colleague of mine has a number of old audio CD-Rs that no longer
> play well, and as the de facto local authority on all manner of
> digital audio, I was the expert he consulted for advice on
> resurrecting the non-playing tracks. These are rare recordings that
> would be difficult or impossible for him to replace, so he is willing
> to go to some trouble if necessary.

Well, do you have an old Sony SCSI CD drive and some why to use it?
The older the better. A 1x would be ideal. If you can find a really
old triple beam drive (unlikely) those are amazing. I once had a
disc that was scratched up severely by a vacuum cleaner and was still
playable on a triple beam CD player, but that was 20 years ago, and
it was a stereo component, not a computer drive.

Then, clean the disc carefully, using radial motion (center to edge
in straight lines).

With most problem discs, that will be enough. The reason for the old
drive is that the lasers were stronger, so less prone to distraction
from slight flaws in the material.

> I've never had to try such a thing, but I recommended he try the
> freeware "Max" <http://sbooth.org/Max/> and also the demo of
> SubRosaSoft's "File Salvage"
> <http://www.subrosasoft.com/OSXSoftware/index.php?
> main_page=product_info&products_id=1>
> to see if he could get any helpful results, but no joy so far.

It's unlikely any software solution is going to help you. You most
likely need to work on getting the media itself repaired to the point
that it is readable.

A good quick test is hold the disk up to a bright light. If you see
pinholes, some of the data on the disc is irretrievably lost. If
not, chances are you can recover most, if not all, of the data.


--
Advance and attack! Attack and destroy! Destroy and rejoice!

At 13:10 08/18/06 -0700, Google Kreme wrote:
>Then, clean the disc carefully, using radial motion (center to edge
>in straight lines).

But ONLY THE UNDERSIDE. Do NOT so much as touch the upper (label) side,
much less rub anything on it. That's where the data live, and that's almost
certainly the part that's going bad unless the underside is obviously dirty.

Edward
Art works by Melynda Reid: http://paleo.org

I have found that some computer laser readers work on some disks and no on others...it is not predictable but one thing is for certain in my experience which is that readers do not work as well as write/read drives and secondly, often times a purpose built CD player from a stereo ( remember those?) have much better error correction than many computer burners. Go figure....In that case, one will have to go through digital to analog conversion and back again..big deal...better than nothing. For sure, try many burners or players as some work a lot better at discerning data than others and it might vary from disk to disk to complicate matters further. I would also try to copy the file rather than play the file if possible and do nothing else with the computer whilst attempting to recover the data.Use a fast machine as well and if there is a cache setting for the software you are using, I would suggest the largest cache it can handle relative to the machine RAM ALLOTMENT. Regarding CD-R disks being so fragile as not safe to use. It is all about storage. I transferred a number of image files made so long ago that the state of the art burner was a scsi laCie 4x burner...I had no trouble transferring the files so we are talking ten year life span here and there is no discernable damage on any of the fifty or so disks I transferred. They are kept in their cases, in a cool and dark environment...just like they recommended in the first place. I would never keep disks in paper envelopes nor in a warm and bright room...

At 09:50 08/24/06 -0700, Neil Fiertel wrote:
>often times a purpose built CD player from a stereo ( remember those?)
>have much better error correction than many computer burners.

Good point about trying standalone CD players. However, I'm pretty sure
that the reason is that a CD player will make a best effort at producing
(degraded) sound even in the face of errors, whereas a computer CD drive
usually refuses to return any data if there's any error. And in a case like
this, better degraded sound than none at all.

Edward
Art works by Melynda Reid: http://paleo.org

On 24 Aug 2006, at 10:50 , Neil Fiertel wrote:
> Regarding CD-R disks being so fragile as not safe to use. It is all
> about storage. I transferred a number of image files made so long
> ago that the state of the art burner was a scsi laCie 4x burner...I
> had no trouble transferring the files so we are talking ten year
> life span here and there is no discernable damage on any of the
> fifty or so disks I transferred.

It's not that simple. First off, discs were simply better built 10
years ago. My 1996 backup of my Mac still works fine, but many discs
from 2000-2001 don't work at all. Part of the problem is the ultra
cheap dye process that was developed (you know these discs because
they are a dark dark blue tending toward deep sea green. The dyes in
these are not stable, and the lifespan of these blanks is remarkably
short. Silver discs, on the other hand, as well as gold discs, use
stable inks as as long as the tops don't get dinged, they last a very
long time.

If you buy "Generic" blanks, it's not a bad idea to stick labels on
them, as this help protect the top of the disc, which is where the
data lives.


--
BUGS: There is no conversion specification for the phase of the
moon." strftime(3) man page

On Aug 26, 2006, at 3:46 PM, Google Kreme wrote:
> If you buy "Generic" blanks, it's not a bad idea to stick labels on
> them, as this help protect the top of the disc, which is where the
> data lives.

Be sure that the labels use an adhesive which won't damage the
surface of the disc, though! Some adhesives contain volatile
components which will affect many plastics. I can't comment on any
particular labels, and I don't know whether there are any bad labels
being sold for CDs and DVDs, but I would be surprised if there
aren't. I do know, though, that many discs come with a protective
layer to make it safer to label, write, or print on the disc. You're
safest if you know the discs you're using and the labels you're using.

Geoff

At 07:45 08/29/06 -0700, Geoffrey Odhner wrote:
>Be sure that the labels use an adhesive which won't damage the
>surface of the disc, though!

Yet another caution about labels: my informal observation is that CD-Rs
with paper labels left in my car (music CDs) deteriorate far faster than
unlabeled ones. My sample isn't large enough to state this with assurance,
but I do have a reasonable explanation: the coefficient of expansion with
temperature change may be different for the paper in the label than for the
polycarbonate in the CD, resulting in stress at the interface -- which is
exactly the location of the data.

Obviously an alternately hot and cold car is enormously more stressful than
even the moderately controlled environment of a house or office. However,
if my observation is borne out, it says that the combination of a label and
any temperature change is putting stress on the data layer, and might cause
trouble over a longer period of time.

So I would tentatively recommend that if you want the protection offered by
labels (which is real), then make sure you have a storage environment in
which the temperature is tightly controlled -- certainly more tightly
controlled than the average house and probably more so than the average office.

Edward
Art works by Melynda Reid: http://paleo.org

Apart from any designed differences in hardware or software, another reason to
try every reader you can - even multiple 'identical' ones - relates to centering.

The hole is supposed to be bang in the middle of the disk; the spindle is
supposed to be exactly aligned with the optics of the reader. In practise of
course neither is perfect, and the specs allow for some tolerances. But even
with brand new discs, we used to find that if you had a reader that was
sufficiently close to the edge of the acceptable tolerance, it would work with
some disks not others; with a disk that was a bit off, it would work in some
readers, not others. An off disk in an off reader was a doomed combination.

Presumably when there are other issues affecting readability, a drive that's a
bit off, but not enough to cause problems normally, is going to have even more
trouble; a drive which has better alignment is going to have a better chance.

- Ben

Yet another caution about labels: my informal observation is that CD-Rs with paper labels left in my car (music CDs) deteriorate far faster than unlabeled ones.

And yet another caution about labels. In my slot-loading CD changer in my car, the labels add just enough thickness that the CDs tend to get stuck in the changer, often requiring me to stick something (like the edge of another CD) in the changer while it's ejecting in order to jockey the CD into a position where it's ejectable. In looking at the documentation for the changer recently, I found it actually says not to use CDs with adhesive labels in the changer.

If storing materials on microfilm had mass appeal, there would be market incentives to make meeting these needs affordable but that's unlikely to happen.

There must be some market incentives, since the company I work for until recently made a lot of money microfilming documents (magazines, newspapers, dissertations, and older books). We still sell a fair amount of microfilmed content, although we're focusing on digitizing the same materials now (our microform products are now scanned first and then the digital files are converted to microform (film and fiche), at least dissertations are done that way). Yes, our microform orders have diminished in favor of digital documents, but we still sell quite a lot of microform.

There are advantages and disadvantages to each; digital is terrific for immediate access and searchability but requires regular attention to maintain accessibility, microform is ideal for long-term archival storage but is substantially more trouble and far less flexible for immediate access. Microform is also mainly useful for text documents, since it is indeed black and white, and even greyscale photos microfilm terribly.