CDTOC tag for CD Extra discs is different between dBpoweramp and CUERipper

[Tenorfan]

I did some experiments with the CDTOC tag and by chance noticed that for some CDs dBpoweramp and CUERipper are in disagreement about the value for this tag. After some more digging it seems that only for CD Extra discs differences are seen.

As an example, this is the TOC of my copy of Sting's "Brand New Day".

TOC as logged by cdrecord

first: 1 last 11
track:   1 lba:         0 (        0) 00:02:00 adr: 1 control: 0 mode: -1
track:   2 lba:     26790 (   107160) 05:59:15 adr: 1 control: 0 mode: -1
track:   3 lba:     48237 (   192948) 10:45:12 adr: 1 control: 0 mode: -1
track:   4 lba:     71117 (   284468) 15:50:17 adr: 1 control: 0 mode: -1
track:   5 lba:     93850 (   375400) 20:53:25 adr: 1 control: 0 mode: -1
track:   6 lba:    118230 (   472920) 26:18:30 adr: 1 control: 0 mode: -1
track:   7 lba:    139885 (   559540) 31:07:10 adr: 1 control: 0 mode: -1
track:   8 lba:    141455 (   565820) 31:28:05 adr: 1 control: 0 mode: -1
track:   9 lba:    166835 (   667340) 37:06:35 adr: 1 control: 0 mode: -1
track:  10 lba:    191555 (   766220) 42:36:05 adr: 1 control: 0 mode: -1
track:  11 lba:    231470 (   925880) 51:28:20 adr: 1 control: 4 mode: 2
track:lout lba:    273527 (  1094108) 60:49:02 adr: 1 control: 4 mode: -1

TOC as logged by CUERipper

     Track |   Start  |  Length  | Start sector | End sector
    ---------------------------------------------------------
        1  |  0:00.00 |  5:57.15 |         0    |    26789  
        2  |  5:57.15 |  4:45.72 |     26790    |    48236  
        3  | 10:43.12 |  5:05.05 |     48237    |    71116  
        4  | 15:48.17 |  5:03.08 |     71117    |    93849  
        5  | 20:51.25 |  5:25.05 |     93850    |   118229  
        6  | 26:16.30 |  4:48.55 |    118230    |   139884  
        7  | 31:05.10 |  0:20.70 |    139885    |   141454  
        8  | 31:26.05 |  5:38.30 |    141455    |   166834  
        9  | 37:04.35 |  5:29.45 |    166835    |   191554  
       10  | 42:34.05 |  6:20.15 |    191555    |   220069  
       11  | 51:26.20 |  9:20.57 |    231470    |   273526

For this CD the rippers write the following CDTOCs:

• CUERipper: A+96+693C+BD03+11663+16F30+1CE6C+22303+22925+28C49 +2ECD9+388C4+42D0D
• dBPoweramp: A+96+693C+BD03+11663+16F30+1CE6C+22303+22925+28C49 +2EE05+35BA6+42D0D

After checking by hand, to me it seems the tag written by CUERipper is correct and the one by dBpoweramp, well, needs an explanation.

I repeated the rips with the respective tools and of course for both got exactly the same results. I checked some other CDs and for most the generated tags are identical, except the few other CD Extra discs I have. For these, the pattern for the differences seems to be:

• it's only the LBAs for the last audio track and for the data track that are different in the encoded CDTOC, all other tracks are ok
• the LBA for the last audio track is too high in the dBpoweramp tag by 300; this was the case for all CD Extra discs I could check
• the LBA for the data track is too low by 11550; this also was the case for all CD Extra discs I could check

All of this was checked using dBpoweramp releases 2024-11-04 and 2025-02-07, both are affected. All rips were done with the same disc in the same drive with correct offset setting, in case this would matter.

Let me know in case further examples would be helpful.

Cheers!

[Spoon]

• CUERipper: A+96+693C+BD03+11663+16F30+1CE6C+22303+22925+28C49 +2ECD9+388C4+42D0D
• dBPoweramp: A+96+693C+BD03+11663+16F30+1CE6C+22303+22925+28C49 +2EE05+35BA6+42D0D

These two differ before the datatrack, 2ECD9 vs 2EE05, that is track 10 start, and track 10 looks to be an audio track.

There are two ways of reading the TOC (LBA, or MSF), certain drives return different TOCs depending on which way is requested, and no it should never happen and is a bug in the drive firmware, we have seen this. It could be this.

[Tenorfan]

In the meantime, I looked at some more CD extra discs and also tried all of them on another drive.

The new results confirm what I wrote yesterday. To summarize:

• For all CD extra discs I have, dBpoweramp writes a CDTOC tag that is different than the one written by CUERipper or calculated by hand.
• Switching from a MATSHITA drive to a TEAC drive does not make a difference. For each of the CDs I checked, the two drives give the same results.
• In the CDTOC, it's always the 3rd and the 2nd entries from the end which are different, corresponding to the last audio track and the data track. The LBA for the last audio track is too high by 300, and the LBA for the data track is too low by 11550.
• The TOC listed in dBpoweramp's extraction log is correct. It's only the CDTOC tag that has these differences.

The good thing is that it really seems to occur only with CD extra discs. Up to now I've tested 5 and I might have 2 or 3 more, but that's all. For these I can fix the tags by hand.

[Spoon]

Will make a note to look at next update