CD-R media longevity / durability / error tolerance TEST

[eiki]

Both my ASUS 32x12x40x and my CreativeLabs 48x speed drives seem to support C2 error detection. The ASUS wouldnt go though the test but the Creative drive passed with flying colours.

The methodolgy is beyond my scope. I'll let you smarties handle that one.

[nox]

By the way (if we want to use Feurio in the test): one thing is that Feurio "Test Device" report that drive can inform of C2-errors and another thing is that it is activated by default.
Most CD-ROM drives don't have Feurio ripping specific configuration, so Feurio use typical default configuration for almost all CD-ROMs, and C2-error reporting is disabled.

To enable it:
Feurio CD-Manager -> Program Parameters -> (Device Parameters ->) Special Parameters -> C2-error-message:
and enable "Report C2-errors" and let disabled "Show warning when C2-error occur" (so it doesn't stop extracting).
(When enabling C2-error reporting maybe it is necessary to decrease the number of sectors per each read block. Feurio would tell you if that is the case)

Now, insert a scratched CD, and rip it. (With "Read tracks individually" you get the number of c2 error for each track if there are c2 errors).



I have to say that if there isn't a way to avoid Feurio stopping extraction to send a message when an unreadable sector appears it would be a hell to extract when there are too many errors. (I'm not talking about C2 errors here but unreadable sectors. Most C2 errors are corrected by drive)
There are options to configure what to do when there are errors, but are dependant on drive).

[Pio2001]

I see that the usual Mitsui are missing in this test.
Mitsui gold are pretty uncommon. Philips, Teac and HP rebrands Mitsui silver.
I can send you some without problem.

[eiki]

Bump

[Kennyshin]

I see that the usual Mitsui are missing in this test.
Mitsui gold are pretty uncommon. Philips, Teac and HP rebrands Mitsui silver.
I can send you some without problem.

I bought one full box of Mitsui gold in jewel cases in 2000 and they were as good as Taiyo Yuden That's "Master of CD-R" discs.

BTW, I have been following this thread but isn't it better to make this "sticky"?

[eiki]

Are we any closer to finding a suitable methodology?

I was thinking that since we have two discs we could expose half of them to sunlight and the other half to corrosive fumes such as Ozone. I'm not sure where to get a hold of mass quantities of ozone except for car-exhaust, but that doesnt sound feasible. Just an idea. Ditch it if you want to.

[Pio2001]

It's already far-fetched to let them outside... My mother on the phone the other day had some arguments I couldn't refute :

"Why do you let the CDRs under the sun ? When they are stored, they are protected against the sun, therefore you're just going to test how different CDRs are sun-proof, and not how long they can last if they are kept away from the sun ! What's the point in knowing which CDR can resist the sun? We want to know which one can resist the time !"

Hmm...

Well...

OK, so... let's say we won't test them against Ozone, Chlorhydric acid, nor Pepsi-Cola...Just natural elements like heat (that is experienced in a CD ROM drive).

BTW I'm still ok for the test, we'll see what we shall see.

[Halc]

My apologies for not being able to participate in this thread. I've been a little busy for the past one and a half weeks.

Anyway, I think we should test for (in an ideal world):

- UV resistance
- infrared resistance (i.e. direct heat)
- temp changes (i.e. indirect temperature resistance)
- ozone resistance (i.e. increased oxidisation)
- humidity resistance (i.e. changes in relative humidity)
- physical wear & tear (i.e. scratch resistance)

However, doing all these in a lab controlled manner is not possible for us. All the above test separately is too time consuming.

Therefor I suggest that we test in direct sunlight, hopefully outside (hence greater humidity / temp changes) and leave out scratching and controlled ozone changes.

I think the discs that will handle sunlight badly will show their skin in a few days. I suspect Fuji will be out of comission in less than a week (maybe even in a day).

The good ones should hold on much, much longer (even several months) - even under heavy UV radiation.

Why to test like this?

While it would be nice to put the discs in a safe temp/humidity controlled environment and let the be there for 2000 years, and test every year or so, we don't have such luxury.

Hence, we will expose the discs to extreme levels of stress to roughly simulate accelerated aging of the discs. This includes UV, to which discs are often exposed to even in normal day-to-day handling.

Of couse it could be possible leave out the UV radiation test (i.e. no direct sunlight) or not put them outside (much less temp/relative humidity changes), but I don't think it would separate the weed from the chaff.

Hence, I think if we are going to find the most resistant and high quality disc in terms of long term durability, we will expose all the test candidates to as much as changes as possible without trying to physically break them immediately.

I sitll think the method shown by nox (having them outside air exposed to heat/humidity/rays) is the best compromise we have.

Unless somebody wants to buy us all a temp / humidity controlled sealed enclosure with a built-in UV lamp. These are usually used for something else than testing cd-r discs, though :)

cheers,
Halcyon

PS I still don't have all of your mailing addresses and can't send the discs. PLEASE SEND YOUR MAILING ADDRESS TO halcyon at myrealbox dot com IF YOU ARE WILLING TO PARTICIPATE IN THE TEST

[Halc]

Hello all would-be-testers,

I'm back from a short trip with some ideas on how to do the test.

I burned a test disc with a single track four times on the same disc. It's track number 2 from CD-Check and consists of c. 20 seconds of sine sweep tone and c. 18 minutes of silence. The four tracks make the disc about 70 minutes long (along with normal gaps).

I hang out two example CD-R discs (TDK made by Ritek and That's write made by Ritek) outside my window for the duration of my triop (7 days).

The discs were exposed to sunlight, temp changes of at least 12 degrees celcius, humidity, slight acidity (rain is not very contaminated in Finland) and even some dirt (what flies in a normal city air). No bird poo or other big dirt though :)

No scratches are apparent on the surface of the discs after one week and full water rinsing + drying.

TDK looks pretty much as I left it. That's Write looks like the printing on the label side has started to show through the reflective side.

I tried "Test and copy uncompressed" command with the TDK and came back with same CRCs, 100% quality and no errors lit up during the test.

I also tried the "Copy Tracks" feature in Feurio CD-Manager 1.65beta2 with C2 feature turned on with my Sony DDU1621. Feurio reported 66 C2 errors for the first track on the TDK disc.

The That's Write disc failed the test miserably showing huge amounts of errors on all tracks (in EAC), different CRCs, untolerably long test & compare times (in eac) and unability to play in my dvd and my normal audio cd audio players.

Feurio would just slow down to "0.01 times reading speed" when reading the That's write disc with the "copy tracks" feature. I didn't let it finish the copying as it became intolerably slow and Feurio would stop responding to my abort requests.

CONCLUSION
==========

I think it will be quite easy to separate wheat from chaff by this simple test alone.

Also, using Feurio for testing and additional playback testing in audio players should be enough.

I'm still not sure whether my test disc is optimal for this purpose, but at least it contains audible data on which it is EXTREMELY easy to spot audible errors.

If we want to show how quickly the discs detoriarate, I think we should test DAILY for the first week of the test and then revert to testing every three days or so after the first week (to save us from too much work).

Any comments before I make the test disc and start mailing out the sample test discs?

cheers,
halcy

[eiki]

Sounds about right to me.

[bmcswain]

Has this study been completed?

I am quite interested in finding out which CD's are most durable. I've read that they should last 100+ years. Yet am hearing recommendations that if you have CD's you should re-record them every 2 years.

So I have to ask what this study found:

What brands are most durable?

How do I identify those brands since a manufacturer's product may be relabeled to the "brand" printed on the box? (I need a way to stand at a retailer's shelf and determine what brand he is offering.)

Thanks!!!!
Bob
(Wanting to capture my 15 year old videotape to SVCD.)

[seafoodplate]

I was looking around the net when I came across this board. I hope you guys are still checking this because I think this is some information that needs to be known.

I am currently looking at purchased a safe to store my cd media in, but unless I just trust the safe companies, I cannot find any hard data on the durability of CD's, from heat, moisture, stock, and such.

Are you still working on this project or is it dead.

I would be glad to help in any way I can. Please email me at seafoodplate@yahoo.com

Thanks

[Halc]

I have to update this a bit.

I have personally postponed this test indefinitely.

I could make it a long explanation, but in truth it's a combination of having inadequate testing methodology (I'm improving on this as I write), inadequate testing equipment (also waiting for some writers to be released) and lack of time (this has improved considerably on my part).

So, the test at least for my part is on hiatus for some time at least.

I'm still very interested in this, but I'd like to do it in a more reliable manner, taking into account all sorts of comments and suggestions I've gathered from more knowledgeable folks on the way.

So, unfortunately no results yet and not for some time, but don't throw away your hope completely yet.

regards,
Halcyon

[Gooberslot]

This post has got me wanting to do my own test on the 2 types of media I have here, Kodak Gold-silver and Fuji(TY). Of course I don't want to do anything as complicated as Halc was proposing. I'll probably just wait till summer and leave them on the dashboard of my car or on a window sill or something. Anyway, I was wondering why you wanted to use audio cd's for your test instead of data?

[cfitz]

This post has got me wanting to do my own test on the 2 types of media I have here, Kodak Gold-silver and Fuji(TY). Of course I don't want to do anything as complicated as Halc was proposing. I'll probably just wait till summer and leave them on the dashboard of my car or on a window sill or something.

Yeah, I plan on doing something the same. Not very scientific, but it's the best I can do.

Anyway, I was wondering why you wanted to use audio cd's for your test instead of data?

I think because when this was first proposed (it was some time ago - this isn't a recent thread) there weren't widely available tools like WSES/CD Doctor/KProbe for testing low-level CD-R errors, and the checksum/binary comparisons of EAC and the like provided the best indicator of dics health. Also, keep in mind that data CD's contain an extra layer of error correction that masks low-level errors, a layer that audio CD's don't have. Thus, errors would begin to show on audio CD's before data CD's.

cfitz

[nox]

For me, the worst problem with this test is the media: it's changing constantly, so the test we could have made with 24x media could be useless nowadays... And the tests we can make now will be outdated in little time. (I think I could make some generalization about quality of brands for the future but it would be only a guess).

I don't have much time now to test all that media.

I was wondering why you wanted to use audio cd's for your test instead of data?

Apart from the extra correction layer in data CDs that cfitz mentioned, the only way I had to count C2 errors at that time, was Feurio (an only audio program).

Anyway, everybody can test his media easily with the tools available today.
Those WSES/CDDoctor tools must be great to test CDs, but I don't have a LiteOn drive. I guess those programs don't stop when finding critical errors: in that case, they are perfect...
Let's see if CDSpeed can support my LG to count C1/C2 errors...