AccurateRip and protected CDs

vbimport

#1

Am I the only one, who is a bit of sceptical about the AccurateRip approach?

Maybe this has been voiced before, but how does the software actually separate good rips from bad rips?

Just statistics?

But what if (when) huge numbers of people who have no inclination to learn anything new about ripping start using it?

What about when most of the people will be using the same generic LiteOn rebadged drives?

Won’t then the results that are most plentiful become the ‘accurate rip’ standard in the AccurateRip result database? All other rips will be compared to this ‘standard’.

In the time of fast spreading copy protections (the biggest future thread to accurate DAE rips, imho), this becomes problematic.

Let’s say 100 people with practically the same LiteOn drives (I’m using them as an example, not because they are bad, but because they are hugely popular) rip the same copy protected CD with the same clicks and pops (I’m now assuming that the copy protection cannot be bypassed on these LiteOn drives just by putting a disc in and pressing the ‘let her rip’ button).

This 100 times ripped ‘accurate’ rip becomes the standard (now assuming that the errors produced by the copy protection are deterministic over various rips on the same drive using the same software).

Then somebody who has know how, a wide selection of drives, plenty of software and time to use his/her ears, comes up with a pristine rip. No clicks and no pops, but it has a different CRC from those 100 rips that share the same CRC with each other.

Now, this pristine rip with no pops and clicks will differ from the that of the 100 people who used the same drive and got the same (bad) results. Will it not be deemed inaccurate in AccurateRip?

Or to put it another way, if you are the 101st person to use that same LiteOn drive and get the same rip with the clicks and pops and same CRC, then will that not lead you to false sense of security that your rip is ‘accurate’?

What’s the point of this kind of rip accuracy testing when most of the record companies have said that they will go 100% copy protected CD audio releases in a year or a little more?

I mean, is there even a perfect rip in the time of copy protections, where the copy protection actually overwrites the actual data and let’s the drive interpolate?

Shouldn’t the only criteria of a good rip at that point be actual sound of the ripped file, at which point we lose the common benchmark for a good rip. After all, how good it sounds is very subjective and some will end up liking the linearly interpolated audio data produced by their cd-rom drives while others will end up liking the higher order polynomial interpolated results that some ripping program produces…

This may be a stupid question and it may have been answered before, so do not take offence.

I’m just interested in why this AccurateRip approach would actually be worthwile in the time of copy protected audio cds…

I truly see the worthiness of it for non-protected audio cds, but for copy protectd CDs I wonder.

best regards,
Halcyon


#2

You’ve got a point about protected CDs with CU errors (since C2 can be correctable, let’s call uncorrectable C2 “CU”).
In this case, AccurateRip can reject a rip that is actually better than the reference one.

If we assume that the C2 unrecoverable errors that we get in CDs with some copy protection are just coming from erroneous EFM, C1 or C2 data (and not sync errors (?), skipping (?), or even some smart trick like messing with the DSV), then, as different drive can use different level of error correction in their CIRC implementation, like correcting two wrong symbols in C2 frames, or trying to recover 4 of them, for example, different drives will not return errors at the same place (see http://perso.numericable.fr/~laguill2/dae/dae.htm, appendix 2 for hardcore details)
Some drive will return more errors than others. It has been experienced by the user Tigre : http://www.hydrogenaudio.org/index.php?act=ST&f=1&t=5370#entry54479 , and JeanLuc http://www.hydrogenaudio.org/index.php?act=ST&f=20&t=7628 , who have compared extractions of a CDS200 protected CD with different drives. We can note by the way that Tigre found regular patterns of differences, meaning that the errors position are not optimized for best interpolation, unlike what SafeAudio claimed to do.

For these CD, comparing two extractions, with EAC, remain the only way to get a secure rip, since there are C2 errors everywhere, which turns Plextool and Feurio unable to detect real errors, and different drives behave differently, which turns AccurateRip unable to compare CRCs.

The drive used seems very important since Plextor 121032A and PX-W2410A are reported to mute parts readable by other drives. Drives using higher CIRC error recovery should return less errors, and hifi Players, or drive capable of interpolating several contiguous samples should interpolate better.


#3

Here’s what Spoon, the maker of AccurateRip, answers :

Copy protected CDs are not ideal, I would have to say AccurateRip would not perform 100% on them, but then again that is what the disc is - damaged, different drives read different interpretations of it. I could possibly in the future update the database to look for and highlight to the user these copy protected CDs.

It is up to the consumer to force the issue on these discs (CD sales would drop if these things became the norm), if as Halcyon predicted they become widespread - people who rip, programs that rip and drives that rip are in for a hard time. Only time will tell.

PS please post on CD Freaks as I am not a member.

From http://www.hydrogenaudio.org/index.php?act=ST&f=20&t=8078