View Full Version : How does the TwinPeaks method work?

14.12.2002, 10:19

could someone explain how exactly the TwinPeaks method of making (maybe only temporally) working copies of SecuROM v4.8x works?
I think I understood how SecuROM works but I'd like to understand this also... :)


14.12.2002, 23:50
The TwinPeaks-Method simply inserts new Sectors, where the Speedchanges on Original-Disc occurs. F.E. you remember the BWA-Builder?
Ever noticed the "squares" that identify a original Securom-CD from normal CD? Twinpeaks inserts Sectors on the CD-R that contains the same data. Let me explain:

--> Read Direction:
f.e. Sectornumber 3000,3001,3002,3003,3004 and so on

--> Read Direction:
f.e. Sectornumber 3000,3001,3001,3002,3002,3003,3003,3004

However this is just an example, it shows the "trick" that let SecuROM "think" it's an Original CD: The additional Sectors are placed as good as it gets to fool the protection (time-delay).

As you see above, there are Sectors that contain the same data and the same Sectorheader. This is in fact a violation of the standards, but many cd-roms (burners, dvd's) just throw the following sectors away if same sectornumber is detected twice or more often. But Plextor-Drives f.e. detect them and aren't amused about :mrgreen: (which is in fact fully ok, as this is after all a non standard-disc anymore) and therefore you can't install those discs as readerrors appears on Plextor-Drives. This can or cannot maybe "cured" by Firmwareupdate or fooling the Plextor-Drives (I'm currently working on that)

The whole "trick" with the TwinSectors is (what a irony :mrgreen: ) is first developed by a company named "Thompson" and the protection was knowned as "Tagкs"--- I managed to copy such discs by manually generate Sectors with new Sectorheader/data

hope you now understand why it should be difficult (if not impossible) to generate 1:1 copy of new SecuROM

15.12.2002, 11:45

So if I understand this correctly, the varying structure on the disc only changes the time it takes to read out a sector. And TwinPeaks throws in the additional sectors to 'fake' that delay?
Now that's interesting...

I always confused the name TwinPeaks with the sometimes so-called TwinSector, that Mode2-sector in the Mode1-stream and wondered how manipulating it could fool SecuROM v4.8x...

What happens if e.g. the whole disc had a structure that would make it necessary to throw in EVERY sector at least twice? Wouldn't that create disc content being about 1.2GB in size?

Hope I understood all this correctly... :)


15.12.2002, 11:49
but a protection isn't going to read the entire disc, since this would take five or ten minutes, at lease that's what I think.

15.12.2002, 12:06
right... but apart from that, i understood that story right?

15.12.2002, 12:15
Yes, I believe what you say is correct, but i am no expert. Locutus seems to be more of an expert in this field :)

15.12.2002, 15:59

I thought about this a bit more... does this mean Alcohol's DPM does not really read the structure of the disc but only the delays between the "delivery" of the sectors? That would explain why the DPM process is so extremely time critical (as it is said in the Alcohol forums)...

for every answered questions, two new questions arise... ;)

Thanks for the information so far!


15.12.2002, 19:53
Yes, Sergei, you understand it right, and yes, that's why it's so time-critical (the dpm-function). But, on the other Hand, even SecuROM needs to distinguish original from copy this way (time-checks);
INSERT INTO `portal_posts_text` VALUES
because there is no "build-in" proof-function. And yes, the additional Twin-Sectors increases the amount of disc-space that is needed.

22.12.2002, 06:30
Which will make those copies very easy to identify - wait for next Securom version :P .