1 (edited by scsi_wuzzy 2020-07-15 16:06:40)

I'm doing some troubleshooting on a disc that for some reason is nearly unreadable on all my Plextor drives (despite it being in what appears to be perfectly fine condition). It dumps just fine with Alcohol on my LG drives, but the PX-760A and both the PX-716AL drives just rattle and can't read it. The Premium can read it, but it struggles a lot and often crashes DIC.

The disc is a copy of the PC game "Rollcage." I found it second hand ages ago, and I don't even know what region it is. It indicates on the disc that it's in English, so it's probably not a North American release (most of my CDs are), but it also doesn't seem to match the Europe copies in the database. It has 9 tracks like the ASUS OEM release, and the data track matches, but none of the audio tracks match and the mastering ring / serial is different.

I was finally able to get a dump from the Premium. Since it struggled severely, I wanted to be paranoid and check the dumped audio against what the LG dumped using CUERipper. So, I converted the first audio track from CUERipper to raw signed 16 bit audio (via ffmpeg), and I opened up both the DIC bin output of the first audio track and my converted file in HxD. Based on the leading zeroes in both files (as seen in HxD), I used the hex editor to manually adjust the offset in the CUERipper track to match the DIC one. I also truncated the CUERipper track to the length of the track output by DIC (the lengths of the tracks were different both before and after padding to correct the offset). After that, the files matched (compared with fc.exe).

Cool, looks like probably the first audio track was dumped correctly despite the Plextor's struggles. So I move on to the second, expecting to go through the same set of steps. However, when I load up the second audio track from DIC in HxD, there's obviously samples from the first audio track at the beginning of the second one, before the 2 second silence. Importing the raw audio in Audacity shows that, indeed, there's a very short click of audio on the track before the 2 second silence. The CUERipper files don't seem to have any of this overlap (but it looks like there's instead overlap of the pregap).

I expected the offsets to be different between CUERipper and DIC (though, honestly, I'm not sure why except I gather it has something to do with offsetting the audio relative to the data track, whereas CUERipper is just concerned with offsetting the audio samples so that it matches other drives when dumping audio data). However, I didn't expect to find samples from the end of the previous audio track at the beginning of the next. Is this normal? Is it something I need to investigate further, or is it maybe just a common mastering error where some number of samples are misaligned?

Forgive me if this is an extremely obvious question to the more experienced rippers, but I've been dumping PC games and relatively few of them have been "enhanced" CDs, so I haven't looked into it previously. I may try to follow the old dumping guide to see if I get the same results as DIC (http://wiki.redump.org/index.php?title= … acks_(Old)), but all the images in the guide are dead so some of it isn't entirely clear to me.

2 (edited by reentrant 2020-07-15 22:31:36)

I have few discs which have misaligned audio data. You see it because Plextor with help of D8 command can detect data offset within disc (remember that all cds are just audio with "pasted" data + ecc on top of it) and it properly aligns audio data following data track.

For non Plextor drives depending on read offset you might see it or not. Drives with bigger / smaller read offets can mask it so you won't see it.

Anyway looks like "mastering issue"...

reentrant wrote:

I have few discs which have misaligned audio data. You see it because Plextor with help of D8 command can detect data offset within disc (remember that all cds are just audio with "pasted" data + ecc on top of it) and it properly aligns audio data following data track.

For non Plextor drives depending on read offset you might see it or not. Drives with bigger / smaller read offets can mask it so you won't see it.

Anyway looks like "mastering issue"...

Thanks, I kinda suspected something along those lines. I ended up comparing the audio by concatenating all the raw files from CUERipper and offsetting it to where the audio is located in the .img file. That made the track offset problem disappear.

It turned out the audio matched except for a couple bits at the very end, and I suspect that's a drive offset issue (probably one of the drives dropped off the last few samples due to the sample offset). So, looks like it's a good dump.

The disc itself definitely seems like it's not the highest quality master. The mastering ring doesn't have anything except the text "IFPI.CO-02" -- strangest ring I've seen.

All my pirate discs have such odd looking rings smile Maybe it's a pirate disc?