Analyzing Alcohol 120% DPM

[Nexy]

Oh nice sarami, glad to see you working on this!

[Jackal]

Do we have anybody who would be able to extract a DPM "signature" from the executable? That way we would be able to get the exact timing ranges that the game is looking for.

[sarami]

Memo
----
. SecuROM DLL (Sintf16.dll, Sintf32.dll, SintfNT.dll) is combined in the exe of SecuROM 4.xx disc.
. These dlls are being compressed by petite (https://www.un4seen.com/petite/) and can decompress by enlarger v1.3 (http://www.angelfire.com/electronic/oldskool/stuff.htm)
. If someone can disassemble these dlls, it may be revealed what it is.

[Nexy]

Rib ran a patent search for SecuROM, document here https://patentimages.storage.googleapis … 2631A1.pdf

Rib/Wiggy2k found the one for CDilla/SafeDisc as well here https://patentimages.storage.googleapis … 353890.pdf

[Nemok]

Hello reentrant

Having also tried cdarchive for a cd, some 'ranges' appear to be missing. I was able to visually identify 4x25=100 'spikes' on the graph drawn by alcohol, but the program only saw 34 according to the ranges count. The 'data' values clearly differ from the example you had given when approaching one 'area' :

DPM Data: 0 1550 -1 1304
DPM Data: 0 1600 1 1305
DPM Data: 0 1650 1 1306
DPM Data: 1 1700 15 1321
DPM Data: 0 1750 17 1338
DPM Data: 0 1800 4 1342
DPM Data: 0 1850 1 1343

There's a second positive 'sector density difference' right after the first one here, so that could partially explain the issue? A resolution of 1 value every 50 sectors used by the high precision sampling is way more likely to have this case.

Maybe you already fixed it since the thread was created. By the way, have you found how to patch a MDS file?

[Nemok]

For anyone that still tries to get the best DPM approximation possible.

I have written a small bash script program that moves the DPM content into an old format MDS container (convert). This allows manual editing of the DPM values with Advanced MDS editor 0.5.5 and BWA edit 1.1 since these are incompatible with what the latest Alcohol releases create. The edited DPM content may then be put back inside its new format container (rebuild).

[Nemok]

I did my own research and wrote a program that analyzes DPM timings from MDS files, allowing to identify low quality dumps, and helps to find best performing drives.

DPM SCN supports :
- CD and DVD sampling rates
- curve quality score
- layout detection
- critical error detection
- reliability checks

Compatible with MDS files created by Alcohol v2.0 or later.

Windows : https://github.com/jonblau/dpmscn/relea … x86_64.zip
Linux : https://github.com/jonblau/dpmscn/relea … _64.tar.gz

[reentrant]

Hi Nemok. I didn't look at this stuff for ages. Nice work, btw. For a reference I am attaching src of dpm module from cd archive app. It's very simple, just a PoC...

[Nemok]

Thank you for sharing that code reentrant.

That's indeed the core logic involving the use of thresholds, obviously due to the analog nature of DPM. Even as is, it already worked quite well with DVDs because what you called 'readingTimeDifference' was spread across significantly less sectors than the length in sectors that's used for sampling (corresponding to the 'resolution') which made the detection of a 'range' in the same pass way more feasible. For CDs however, 2 to 3 samples are actually needed to describe that timing difference...

[sarami]

@Nemok, reentrant
Good work. If only I could figure out how to calculate the 4 byte enumeration in the DPM block, I could support mdf/mds...