This is also why we're not hosting full subchannels: It's almost impossible to obtain 100% clean unaltered subs.

Clean subs != unaltered subs: while it's relatively easy to obtain the 100% clean subs, it's hard to obtain the unaltered ones (leaving all the original errors and fixing all the read errors, including 1-bit ones).

Of course, because it's not "data", it's just a series of time measurements between reading certain sectors/areas of the disc. Since it depends on the drive's mechanics, the results may vary among different trials.

Yes, but only subQ channel uses CRC16. So about only subQ, it's possible to obtain clean subs.

I think nobody here ever really looked into the ways of preserving the Data Positioning Measurement data, because the format is proprietary/undocumented and the results seem to be different for each drive and even on the same drive if it's dumped multiple times or at different speeds. We could've gone the trurip way on this and preserve a single dump, but redump is about verifiable results. This is also why we're not hosting full subchannels: It's almost impossible to obtain 100% clean unaltered subs. Maybe this approach doesn't help preservation in the end, but since it's essentially metadata, it can always be added at a later time once a sensible preservation method is chosen. The SecuROM subchannel data was added 10 years later, but it only took us a couple months to recover more than half of the missing data (although some discs are missing the pregap sector).
It would be interesting to get some more info on exactly how the DPM data is retrieved and stored. If anyone has this info, feel free to let us know. Maybe a better way of preserving this would be to retrieve the accepted range of values from the game binaries in some way that the game checks for (good luck with that).

I notice when I dump it in Alcohol 120% that I get an *.mdf and a *.mds file, but the guide says I only need to keep the mds file. Is there a way I can attach the mds descriptor to the iso I dumped with isobuster? Or do I also have to keep the mdf file in my backup?

I've tried hex editing the mdf file and switching the *...m.d.f to *...i.s.o, but this just causes the resulting file to be unmountable.

also this might be a dumb question, but is there any possible way to get an 100% accurate descriptor file or is this just one of those things that our drives can't do 1:1?

edit: this is regarding CD SecuROM. I just dumped a DVD and apparently the mdf & iso I created match checksums and I can hex edit the mds to load it as iso without problems.

edit: So none of my DVDs are dumping properly. I've tried 8x and 1x speeds so far for the mds, but I keep getting error 1000 or unable to find original disc when mounted in alcohol. Is the plextor bad for DPM analysis?
secondedit: Yeah, dumped another mds at 8x on my laptop and it got through copy protection. Not sure what's up with my Plextor.