Multiple error scans on same disc give slightly different results? Why?


#1

If I scan the exact same disc twice, at the same speed, I get slightly different results compared to each other.

Three scans at full speed result into these error counts:

C1 errors:

  • Scan 1: 3209
  • Scan 2: 3314
  • Scan 3: 2989

C2 errors:

  • Scan 1: 3.
  • Scan 2: 7.
  • Scan 3: 0.

Why does the result slightly vary?


#2

What exactly are you scanning? If you’re scanning red book audio CDs (AKA “Compact Disc Digital Audio”, or “CDDA”), that could be due to the lack of error correction in the standard. Basically, drives have to take a guess as to whether the data is being read correctly or not. Also, there may be instances where the drive starts and stops (such as when the computer lags), which is a problem because it’s very hard for the drive to guess where to resume reading the data from. On top of which some drives suffer from alignment problems and/or firmware bugs. (See this well-written article for more info.)

Of course, these issues only really affect red book audio disks. If you’re burning some other disc, you’re probably using a more error-resiliant file system, such as ISO 9660 or UDF. These file systems have error corrections and sector markings that prevent these issues.

Have you tried using different drives? Maybe the drive you’re using has firmware bugs, or maybe your drive is malfunctioning.

Also, are you using a laptop with a built-in drive? Or are you using an external drive (on any kind of computer, not just laptops)? If either of the above are true, maybe the drive is being physically moved too much (that is, moved from one location to another)? Make sure your drive/laptop is set on a stable surface (such as a desk), and leave it there until the scan is complete. Obviously this doesn’t really apply to desktop drives, unless you’re physically shaking the whole computer for some reason.


#3

The CD I scanned was Hiren’s Boot CD on CD-R.
So it has the ISO9660 file system among Joliet and RockRidge. Drive: TSST-WriteMaster S182.

This effect also happens in different drives. The count of C1/C2 errors is not exactly the same.

Are there some weather/environment factors?


#4

So if a RedBook CDDA has zero C2 errors, that does not guarantee 100% audio integrity?


#5

I think that’s how it works, although I’m not sure. It may depend on the drive. If in doubt, consider using EAC or CDParanoia.

From Wikipedia:

A CD drive can have extraction errors when the data on the disc is not readable due to scratches or smudges. The drive can compensate by supplying a “best guess” of what the missing data was, then supplying the missing data.

If there are any CDFreaks here who know more than me, feel free to correct me (unless I am right, of course).


#6

They have a few lacks unfortunately:

  • cdparanoia does not make use of C2 errors, according to german Wikipedia.
  • VCD/SVCD/CVD are unsupported despite they also have CDDA-like inferior error correction.

#7

Personally, I tend not to think about (S)VCDs, since they are hardly ever used anyway. IDK what CVD even is.

About CDParanoia, I would never have known that. Thanks for pointing it out.