"Cinavia"/Verance BluRay audio watermark protection [Archive] - Page 13

xenex

19th January 2011, 10:25

Although I am (personally) not affected by Cinavia, I find it "technically" interesting. So, for those who are interested, I rented "Takers" which I understand to be the first actual DVD (not Blu-Ray) which contains Cinavia. I make a few observations:

* "Takers" DVD is CSS and RCE protected, but has no bad sectors and no apparent Arccos/Ripguard/special-protection at all.

* "Takers" DVD has an "AACS" folder - never seen that before! - which has two files:

AACS->|
|-> CSS_CONTENT_CERT.AACS (76 bytes)
|-> CSS_CONTENT_HASH_TABLE.AACS (116696 bytes)

* Quick Google of these files brought me to this:

http://www.aacsla.com/specifications/AACS_Spec_Signed_CSS_Final_0.951.pdf

So, it appears that the files in the AACS folder consist of one hash table file that hashes the contents of all VOB files on the DVD (in order to determine CSS presence), and a second AES-signed file that authenticates the original hash table file.

I've also read the final AACS agreements, and if I am correct in my reading, which I believe I am, I see the "rules" as:

1> AACS-Licensed devices must always attempt to read the Cinavia watermark from ALL media.
2> If present, the watermark is one of two possible flags - NO_HOME_USE or TRUSTED_SOURCE.
3> If NO_HOME_USE stop play immediately and return Cinavia message 1.
4> If TRUSTED_SOURCE check if it is a trusted source. (Otherwise, return Cinavia message 3.)

Now, on item 4, there is in fact more than one "trusted source." Check the AACS license. For the purposes of Cinavia on DVD, just see this PDF, page 20:

http://bit.ly/fEM8WB

If you can't be bothered to check the link, it says:

*New Signed CSS book defines format for CSS [sic - should be AACS] certificates for use with DVD
*Allows Content Owners to place AACS Watermarks on content which is only protected with CSS
*If an AACS player sees the Trusted Source watermark on an optical disc which does not include the certificate, the player will mute the audio

Hopefully, this was all clear. It's not speculation on my part, it is directly from the AACS specifications. I recommend reading them, it is interesting stuff.

Ghitulescu

19th January 2011, 10:44

As I see it today, since almost every player is manufactured in China, if China refuses to implement them, the content industry would be "somewhere over the rainbow". It would be difficult to say how much time they can hang on ...

setarip_old

19th January 2011, 11:14

@xenex

Hi!

Extremely interesting stuff.

However, I'm curious to know why the second link you provided is to (after redirection) CPTWG.ORG, Copy Protection Technical Working Group that, per CNet, is a committee formed by consumer electronics and computer industry companies to recommend DVD and DTV copy-protection protocols - and the presented information is, in fact, a July 22, 2009 presentation made by AACS to CPTWG.

Is there not a final version of this information available from actual AACS documentation? I ask because things may have changed since 7/22/09...

xenex

19th January 2011, 11:33

@setarip_old

That was the first thing I could quickly find. I also found these:

www.aacsla.com/license/AACS_Adopter_Agrmt_090605.pdf
www.aacsla.com/license/AACS_Content_Provider_Agrmt_090619.pdf
www.aacsla.com/license/AACS_Content_Participant_Agrmt_090605.pdf

But it takes a while to read through this stuff! I'm only human. But I bet you will find what you look for in the above.

rotty

19th January 2011, 11:55

These film companies are wasting too much money on all this "copyright" protection.

They certainly are.

Although, they dont lose out with that cost, we have to pay extra for the disks to cover that.

They DO however loose out greatly on sales numbers because of the higher cost of the disks.

And, ironically, thats what gives the pirates their market.

kurkosdr

19th January 2011, 15:52

The studios dont seem to have any logic at all, they are still paying high costs to have BD+ / AACS on their disks EVEN when they know full well there is no point. Especially with versions that are well and truly useless.

Would any of you pay a Company for the protection on your disk when you know it NOT doing that at all. I think theres a legal phrase for that eh.

If you head over to the sites of the companies that provide such tech (ex macrovision) you won't believe the lies and exaggerations they tell to convince the studios to buy into their tech.

Some i noticed are:

1) It takes a skilled computer user to find a program to defeat ripguard and other structure protection scemes (yeah right)

2) Most devices respect CGMS-A (as regards standalones maybe, but when talking about tv cards no)

3)BD+ is uncrackable

4)It wont cause playback problems, trust us

Since most studio owners are old people who believe computers are some kind of miracle device from planet Vulcan, they buy into everything these companies tell them, and eventualy end up irritating only their legit customers.

Anyway, i think this is going to end up just like Cactus Data Shield. It will turn people away from BD players and entice them to buy dedicated MKV players (like CDS turned people away from cds and towards mp3 players). So, people wont be able to buy any of their content even if they wanted to. And eventually hollywood will understand its pointless to go against your own customers, and release their stuff in mkv, but only after they ve lost millions in the process.

Ghitulescu

19th January 2011, 16:32

The studios dont seem to have any logic at all, they are still paying high costs to have BD+ / AACS on their disks EVEN when they know full well there is no point. Especially with versions that are well and truly useless.

Would any of you pay a Company for the protection on your disk when you know it NOT doing that at all. I think theres a legal phrase for that eh.

The logic is very simple and I sincerely don't think anyone has failed to see it.

The protection needs not to be 100% bulletproof - that is almost impossible to attain and it would cost huge amounts of money to develop and to implement.

The protection needs only to be strong enough that the Average Joe not be able to crack it (at least to survive the attack with normal tools, like Nero & co.). To crack it, Average Joe needs to find a proper tool. Where?

Here comes however the second part - the law. The amis invented the DMCA and lobbied with such a frenzy that we have a similar directive here, too. Under this directive it's simply enough for a DVD to ask a password/PIN (be it 0000) to be considered as a prevention means against unauthorised access, and therefore any software that automatically fills the 0000 for you be banned.

So, by setting a reasonably strong protection and "bending the law", the studios managed to have the control of the user actions.

Why Elby and similar moved to fiscal paradises, while Nero & co suddenly could not copy commercial DVDs? Why doom9 is the only forum I know (in one of the languages I know, seven of them) that still has a sub-forum of decrypting?

This is the logic.

xenex

19th January 2011, 16:35

I don't want to get off-topic, but one thing I have noticed is that the new "Takers" DVD (and most new DVD's) no longer contain the old analog protection previously just known as "Macrovision."

I suppose the studios feel that people making VHS tape copies of DVDs are now largely just an anachronism and no longer a threat. From what I have noticed, this old protection is becoming less and less common on newly released DVDs.

Ghitulescu

19th January 2011, 17:17

To answer this slightly off-topic observation:

AFAIK JVC, the last manufacturer, officially withdrew its support for VHS and any flavour of it at the end of 2008. VHS gear is no longer manufactured. Neither its media. And in 2012 most if not all analog connections (and signals) will become history, too.

So, why paying royalties to Macrovision any longer?

rotty

19th January 2011, 19:07

Since most studio owners are old people who believe computers are some kind of miracle device from planet Vulcan, they buy into everything these companies tell them, and eventualy end up irritating only their legit customers.

Hi kurkosdr

You would think that some bright spark younger Executive at the studio(s) would walk into the board room with a laptop with AnyDVD running, put a Blu Ray disk from their studio with the extreemly expensive "uncrackable" protection on it into the drive and say "the files im now showing you are completely unprotected, all I have had to do is put the disk in the drive, its all done for me"

Surely that has been done in the studios at some point, its been going on from DVD days.

It was more complicated to remove Macrovision from VHS, you dont need to do anything yourself with disk media and its gone.

Just put the disk in the drive, are they saying that that is beyond Joe Average.

It must be they dont want to admit defeat, and the fact as I said before that it is driving up the price of the disk to us and actually fueling piracy.

setarip_old

19th January 2011, 20:01

I, for one, would like to thank "fengtao" and the DVDFab crew for making the effort to provide intermediate Cinavia-related solutions that, when coupled with the hack of the PS3, surely require the Verance/AACS to veer off track in their effort to provide the movie/video industry with the "perfect protection solution".

Several months ago, it was observed in this thread and elsewhere that substituting the DVD audio track for the original audio track of a Cinavia-modified Blu-ray disc is a viable concept.

With the release of "The Takers" watermarked DVD, it is apparent that Verance felt compelled to redirect at least some of its assets to addressing this capability.

I have little doubt that Verance will, likewise, be compelled to address this initial professional "non-removal solution" - that I hope is part of an ongoing "tit-for-tat" response regarding Cinavia.

Again, rather than giving short shrift to this "fengtao"/DVDFab discovery and intermediate fix, I congratulate them for making the effort ;>}
Based on the recent postings to this thread, it's my opinion that the confirmed hacking of the PS3 (SONY is suing "GeoHot" for having successfully done this) is not being fully appreciated for its potential impact in the effort to overcome Cinavia...

SamuriHL

19th January 2011, 20:08

With the release of "The Takers" watermarked DVD, it is apparent that Verance felt compelled to redirect at least some of its assets to addressing this capability.

I thought that, too, until someone pointed out that Cinavia on DVD was always part of the spec. This disc is simply the first implementation of an already existing spec. They didn't have to redirect anything. They simply had to use what was already available and implement it. So, the tit-for-tat that you're looking for has yet to materialize. The ball is in Verance's court to respond to the aforementioned approach that DVDFab took and the known audio settings workaround on the PS3. We shall see what, if anything, they do about it now. :)

EDIT: I know you like when people quote their sources. :)

http://forum.doom9.org/showthread.php?p=1472584#post1472584

That's one of the places I read that Cinavia on DVD was part of the spec.

setarip_old

19th January 2011, 20:39

DrinkLyeAndDie

19th January 2011, 21:54

Whether part of an existing spec or not, it didn't have to be implemented until now - and implementation requires the concerted effort of AACS-LA as an organization, Verance, fabrication plants, the individual studio(s) chosen to or choosing to float the trial balloon of this additional effort to stymie legitimate purchasers' ability to make a backup copy of a disc. This certainly qualifies as a redirection of at least some assets...

Given the infinitesimally small number of Cinavia titles I can't see that they redirected anything. It's in the specs for Cinavia on DVD and they decided to implement it with the Takers DVD release. If not this release then when? There always has to be a first time for anything. It was intended from the beginning and as standalone DVD players die off and and more standalone Blu-Ray players are manufactured that do support Cinavia then the industry in theory has the bases covered to hinder piracy for both DVDs and Blu-Ray using the same technology. DVDs won't be disappearing anytime soon.

Now we just have to wait until someone finds a real complete and permanent long-term Cinavia removal solution.

dirio49

19th January 2011, 22:54

setarip_old

19th January 2011, 23:22

If not this release then when? There always has to be a first time for anything.Thank you for reinforcing my point ;>}

I'd suggest to you that it wasn't until DVD-to-Blu-ray audiostream switching was confirmed to be a valid and working concept that the Verance/AACS-LA consortium deemed it necessary to actually implement this second stage "fix" to their "perfect solution". Until the determination that switching audiostreams was a functional solution, there was no reason for it to be activated. Therefore, this represents the first example of Verance/AACS-LA being compelled to react/respond to the users in this, the most recent copy protection "cat and mouse game".

If, in fact, any protection-defeating software company has chosen to commit its resources solely to finding "a real complete and permanent long-term Cinavia removal solution", that is certainly that company's prerogative - and the sooner it's developed, the better. However, I sincerely believe that until such developments, it's extremely shortsighted to allow this protection to continue unfettered, as doing so may convince the consortium that now's the time for a widespread application of Cinavia to many, many new releases. If nothing else, short term solutions will make it obvious that this is not and will not be the case.

SamuriHL

19th January 2011, 23:23

Curious.
So do the studious with watermarking DVDs
have also messed up coping dvd for us in the end.
or at least you cannot shrink the dvd. you can only have a 1:1 copy, because of the hash check.

what do you guys think?

later

Can't even have that. The hash is done against each VOB with CSS applied to it. Once you remove it, even a full backup isn't going to help. Obviously this only affects playback on a Cinavia infected device, but yes, this does indeed affect DVD backups now, as well.

SamuriHL

19th January 2011, 23:27

Thank you for reinforcing my point ;>}

I'd suggest to you that it wasn't until DVD-to-Blu-ray audiostream switching was confirmed to be a valid and working concept that the Verance/AACS-LA consortium deemed it necessary to actually implement this second stage "fix" to their "perfect solution". Until the determination that switching audiostreams was a functional solution, there was no reason for it to be activated. Therefore, this represents the first example of Verance/AACS-LA being compelled to react/respond to the users in this, the most recent copy protection "cat and mouse game".

If, in fact, any protection-defeating software company has chosen to commit its resources solely to finding "a real complete and permanent long-term Cinavia removal solution", that is certainly that company's prerogative - and the sooner it's developed, the better. However, I sincerely believe that until such developments, it's extremely shortsighted to allow this protection to continue unfettered, as doing so may convince the consortium that now's the time for a widespread application of Cinavia to many, many new releases. If nothing else, short term solutions will make it obvious that this is not and will not be the case.

You're welcome, then, as I was the one who authored the DVD track from The Losers into a BD-R/E backup and confirmed there was no Cinavia detection. :D However, the inverse also worked, which is scary. Meaning, taking the BD's DTS core track, and adding it to the DVD in place of the AC3 track triggered Cinavia. So, any media that contains a Cinavia audio track *MUST* have a valid corresponding AACS hash. DVD, BD, etc. Matters not as long as the AACS hash is there. Good times.

setarip_old

20th January 2011, 00:07

@SamuriHL You're welcome, then...In my post #618 I quoted "DrinkLyeAndDye's" post #616. How could you possibly think I was thanking you for reinforcing my point? (Perhaps we're both taking reading lessons from the same place ;>})

SamuriHL

20th January 2011, 00:27

@SamuriHL In my post #618 I quoted "DrinkLyeAndDye's" post #616. How could you possibly think I was thanking you for reinforcing my point? (Perhaps we're both taking reading lessons from the same place ;>})

I know you weren't. But you should, since I did an insane amount of testing on The Losers when it first came out. Including taking the DVD audio track and verifying that it could be used perfectly well with the BD image. You're suggesting that those tests are what got Verance to start protecting DVD's. So, again, I say, you're welcome. :)

mike20021969

20th January 2011, 00:47

I've been reading the latest posts here, and cannot keep track of the scores at all.
Who's winning?? :rolleyes:

SamuriHL

20th January 2011, 00:49

I've been reading the latest posts here, and cannot keep track of the scores at all.
Who's winning?? :rolleyes:

Verance. :D Unless, of course, you don't own any of the discs that they put their stuff on and/or a player that cares. :D

setarip_old

20th January 2011, 01:53

@SamuriHL I know you weren't. But you should, since I did an insane amount of testing on The Losers when it first came out.

The only title that the portion of Post #616 I quoted and my Post #618 refer specifically to the "Takers" BD and DVD releases.

Your earlier efforts regarding "The Losers" have no doubt been noted by those who comprise the more than 77,000 views of this thread.

SamuriHL

20th January 2011, 02:05

@SamuriHL

The only title that the portion of Post #616 I quoted and my Post #618 refer specifically to the "Takers" BD and DVD releases.

Your earlier efforts regarding "The Losers" have no doubt been noted by those who comprise the more than 77,000 views of this thread.

Yes I realize that. However, you said this:

I'd suggest to you that it wasn't until DVD-to-Blu-ray audiostream switching was confirmed to be a valid and working concept that the Verance/AACS-LA consortium deemed it necessary to actually implement this second stage "fix" to their "perfect solution". Until the determination that switching audiostreams was a functional solution, there was no reason for it to be activated. Therefore, this represents the first example of Verance/AACS-LA being compelled to react/respond to the users in this, the most recent copy protection "cat and mouse game".

Clearly what I bolded did not refer to the Takers DVD and/or Blu-ray. It must be referencing the earlier efforts on discs such as The Losers. And that's what my comment of "you're welcome" was in reference to. :)

A.VOID

20th January 2011, 03:22

Can an encode contain a flag to tell the player
"TRUSTED_SOURCE"?

SamuriHL

20th January 2011, 03:26

Can an encode contain a flag to tell the player
"TRUSTED_SOURCE"?

No. TRUSTED_SOURCE is only meaningful to the Cinavia checking routine. It means it has a valid AACS certificate. Which a backup won't have.

A.VOID

20th January 2011, 03:34

No. TRUSTED_SOURCE is only meaningful to the Cinavia checking routine. It means it has a valid AACS certificate. Which a backup won't have.

I understand that, but I was suggesting this as a workaround/ solution.
Can we write the TRUSTED_SOURCE flag into the encode?

SamuriHL

20th January 2011, 03:37

I understand that, but I was suggesting this as a workaround/ solution.
Can we write the TRUSTED_SOURCE flag into the encode?

No, clearly you don't. It's not ANYTHING that's part of the encode AT ALL. It's something the Cinavia code sets internally based on what it finds on the disc after doing several AACS checks.

xenex

20th January 2011, 04:28

Mug Funky

20th January 2011, 07:49

it is driving up the price of the disk to us and actually fueling piracy.

well... actually... as far as has been determined (apologies for not providing the link, but it can be found in this thread), the only information we've got from industry people working with this is that at least at one point the watermark can be applied during the sound mix stage. this is in post-production, and long before it hits the DVD studio and any audio encoders.

in this case, it would clearly be a producer's decision to put the protection in.

this means the cost will not be passed to the consumer. if there's any additional costs, it will be to the DVD distributor who license the film from the studios. in order to remain competitive, they will not change the price of their discs - they'll just spend a bit less time making nice menus and fancy functions. or if the licensing cost hasn't changed at all, then there'll be no additional cost (indeed, companies like Sony have licensing sort of vertically integrated as they author their own titles).

implementation requires the concerted effort of AACS-LA as an organization, Verance, fabrication plants,...

there's no conspiracy here. all that's required is a trivial change to the authoring workflow. the uncompressed audio will arrive with cinavia already burnt into it, and the extra AACS folder just means the disc is mastered with DVD Data (they are just files like any other). replicators will pass a master with this, though they potentially could flag it with the authoring house ("hey, guys, there's an extra folder here... just go ahead and master it?... okay").

it helps that all these stages, at least at Sony, are under the same company. i await a time when i'm asked to learn some extra crap because a licensor wants the illusion of a little more safety. so far it hasn't happened yet for BD or DVD, but on the BD front it's quite possible that it could slip through the net unnoticed if it's already in the audio and our testing stations don't respond (i can see the PS3 test triggering cinavia unless we were to compromise that unit, but then it'd be useless as a testing machine).

as far as cinavia appearing on DVDs, it could well be because the DVD usually comes from the same master as the BD.

it may in fact be that the AACS folder is not an attempt to break copying, but in fact an attempt to not break playback on a cinavia equipped device. if a player finds cinavia on a DVD with no hashtable, it must stop playback. otherwise removing it from an infected disc would be as easy as not burning that extra folder.

if that's the case, i'll get back to you when the first one rolls in the door and the nervous phonecalls start. i would hate to forget the last step where i use some NDA'd program to generate the AACS folder, and as a consequence thousands of discs have to be recalled and shredded because they wont play.

with a little luck, cinavia will be confined to studio titles and not indie titles. if this is the case, there could be legal ramifications for the studios - if small distributors are locked out of potential business unless we buy an unnecessary program from them under NDA, or have to "upgrade" our authoring software (and re-train staff or buy new machines even), then that is clearly anti-competitive and probably illegal.

tl;dr - if this trickle becomes a flood, shit's gonna get real in the business world as well as with consumers.

[edit]

just thinking... the hashtable would probably use the CSS keys that are pressed into a separate physical part of the disc (DVD-R's don't have them). this means it would have to be generated at the glass-mastering stage, after scrambling.

this means that even if it's reverse-engineered and new hashtables are created, the disc will not have any CSS keys to compare against and the disc wont play.

looks like the only way around at this point is compromising players or identifying and jamming the watermark itself.

xenex

20th January 2011, 09:00

@Mug Funky

I have read the specs over and over, and I think you are mistaken on at least a couple of things, I'll explain why I think so.

The spec says there are two possible flagging-types of Cinavia - NO_HOME_USE and TRUSTED_SOURCE. So, if the Cinavia was to be added in post-production, they would have to at least create two audio tracks, one for each flag. The NO_HOME_USE is for theatrical releases, screeners, etc. And it OVERRIDES the TRUSTED_SOURCE mark.

So, if a pressed Blu-Ray or DVD contained the NO_HOME_USE mark, it wouldn't be able to play at all! According to everything I have read so far, if NO_HOME_USE mark is found, playback stops. Period.

it may in fact be that the AACS folder is not an attempt to break copying, but in fact an attempt to not break playback on a cinavia equipped device. if a player finds cinavia on a DVD with no hashtable, it must stop playback. otherwise removing it from an infected disc would be as easy as not burning that extra folder.
just thinking... the hashtable would probably use the CSS keys that are pressed into a separate physical part of the disc (DVD-R's don't have them). this means it would have to be generated at the glass-mastering stage, after scrambling.

The hashtable doesn't concern itself with CSS keys at all - it hashes the VOB files on the disc - all of them - without regard to whether or not they are CSS encrypted. BUT - if a copy is made where CSS encryption is removed, then the hashes will no longer match. And the player is supposed to detect if they match, at least some percentage of them.

And although it seems a "new" hashtable could be created - I believe it can - the AACS CERT file hashes the hashtable! And is signed using AES public/private key crypto. using a special AACSLA private key that no one knows.

looks like the only way around at this point is compromising players or identifying and jamming the watermark itself.

Yes, I agree with that. At least for those own/use Cinavia-enabled playback gear.

EDIT TO ADD:

I have to say, on a purely technical level, I'm really impressed with the way AACSLA is going. Not that I like them, but they are thinking ahead. I could not originally foresee how they could add Cinavia to DVD, since the DVDCCA agreements could not force it. But, who makes mere DVD players now?! Hardly anyone. So, they are using the power of the AACS license to force makers of Blu-Ray gear to check for Cinavia on EVERYTHING, even DVD, as part of the terms of manufacturing a licensed device. And it probably won't be long (year or two) before the only disc-playback device that one can even find to purchase is an AACS-licensed Blu-Ray device.

Ghitulescu

20th January 2011, 09:53

Not that I like them, but they are thinking ahead. I could not originally foresee how they could add Cinavia to DVD, since the DVDCCA agreements could not force it. But, who makes mere DVD players now?! Hardly anyone. So, they are using the power of the AACS license to force makers of Blu-Ray gear to check for Cinavia on EVERYTHING, even DVD, as part of the terms of manufacturing a licensed device. And it probably won't be long (year or two) before the only disc-playback device that one can even find to purchase is an AACS-licensed Blu-Ray device.

I said it before....
who stands behind both licencing bodies?
So why wonder?

rotty

20th January 2011, 09:57

well... actually... as far as has been determined (apologies for not providing the link, but it can be found in this thread), the only information we've got from industry people working with this is that at least at one point the watermark can be applied during the sound mix stage. this is in post-production, and long before it hits the DVD studio and any audio encoders.

Hi Mug Funky

I was talking about protection in general, the studios (sorry we) do pay for that.

As for Cinavia, I dont know but im sure someone does.

derbeDeus

20th January 2011, 09:58

rotty

20th January 2011, 10:08

Two supermarkets near me started a blu ray section in the with the DVD's about two years ago.

It has got slightly larger up until about three months ago, and now they are defiantly getting smaller and smaller in both supermarkets (Tesco and ASDA).

I’m not saying that anything in particular has caused this and I don’t know it that is meaningful but I was surprised.

Has anyone else seen this.

SamuriHL

20th January 2011, 14:47

Just a FYI, if you go to this page:

http://www.aacsla.com/license/

And download the PDF named AACS Final Content Participant Agreement (it's the 2nd link) and go to page 150, you will see "Table W" which lists the various DRM's that can be considered TRUSTED_SOURCE. There appear to be twelve trusted DRM methods so far, it seems.

Nice find! Thanks for posting that. Very interesting info.

EDIT: Oh man, number 5 worries me, although I'm not sure how Cinavia could be implemented on it. I've actually been wondering quite a bit whether they would try. For those not familiar with FairPlay, it's Apple's DRM for iTunes content. Yes, that means all the digital copies you get on your new BD's. I doubt they'd add Cinavia to FairPlay protected titles simply because they're (in theory) not playable on anything but apple hardware/software anyway. However, it is one of the TRUSTED_SOURCES listed in the AACS content agreement, so, um, hmmmm.

SamuriHL

20th January 2011, 15:17

And although it seems a "new" hashtable could be created - I believe it can - the AACS CERT file hashes the hashtable! And is signed using AES public/private key crypto. using a special AACSLA private key that no one knows.

That part I didn't know. That actually makes sense and helps to explain why it's not as vulnerable as it appears. If you can't recreate the AACS Cert, then even being able to recreate the hashtable with non-CSS crippled VOBS wouldn't do you any good. And good luck breaking AES encryption and finding the private key to sign it with. Maybe someone's got one of them new fangled quantum toys they wanna let us borrow for a few seconds? :D

SamuriHL

20th January 2011, 15:19

Thanks for that. Looking at table W, AACS Content can be in the form of "Prerecorded Video, Prepared Video or Recordable Video". So if PS3 bans the new solution in a future firmware, they'll break the AACS license. Good to know ;)

ROFLMAO! That is indeed one way to look at it! :)

kurkosdr

20th January 2011, 16:01

Don't want to be a nuisance, but one last silly question (please): Has anyone actually tried to diff the dts track from a non cinavia watermarked DVD and the cinavia watermarked dts track of the Bluray? There is nothing to lose.

Anyway, in the worst case scenario, just admit your ps3 won't play any back ups and go buy a dedicated MKV player (just make sure it reads usb drives formatted with ntfs and ext3 and can downscale dts to stereo, and if english is not your first language, also check it can read srt's with your language's encoding). Problem solved. Thats what I did. Cinavia opened a bad can of worms IMO, where the studios have the ability to control what you play on your device, even he it's a file you produced yourself. And until they get the lawsuit they deserve, i dont want to be part of this.

SamuriHL

20th January 2011, 16:32

kurkosdr

20th January 2011, 17:44

Yea, if you read through this thread it's been tried before. The problem is that the tracks wouldn't be identical even without Cinavia, so, finding the "cinavia stuff" is kinda like the proverbial needle in the haystack. Ideally, the best approach to Cinavia would be to rip apart the algorithm that checks for the signal, see what it's looking for, and then you can do several things with that. First and foremost, you can write your own detector so you can identify Cinavia infected tracks. Second if you know what they're looking for, you might be able to create a clever mechanism for "jamming" it and breaking the detection routine. Of course the trick is to do so inaudibly and not breaking the audio further than Cinavia already does. Finally, if you know the algorithm and what it's looking for, you should be able to isolate the signal embedded in the audio track and then begin to work on removing it. That's the final, most permanent solution. Of course, this sounds so easy in theory. Sure, just find a player like the PS3 that's willing to give up the algorithm and you're all set. Yea, um, no. :) Even if you had the algorithm given to you, you'd still have a *LOT* of work to do. So far, no one I know of has even that much.
Thanks for the bunch of info. I think we should make a "Cinavia files" thread were everyone will dump diffed tracks, leaked firmwares (if that last one is against the rules, then just post the model whose firmware was leaked), pdfs and generally any relevant files. So that new members willing to take up the difficult task of cracking cinavia wont have to start from scratch everytime.

Anyway, i believe this thing is like somewhat like cactus data shield. It does not need to be cracked. People will hate it so much that they ll buy MKV players, and then offset the cost by pirating all their stuff and not buying/renting anything original. This is how the real world works, from my short experience in it.

Peer van Heuen

20th January 2011, 18:06

Anyway, i believe this thing is like somewhat like cactus data shield. It does not need to be cracked. People will hate it so much that they ll buy MKV players, and then offset the cost by pirating all their stuff and not buying/renting anything original. This is how the real world works, from my short experience in it.

No it doesn't. Not the very least bit. Take that from my probably significantly longer experience in that same world ;)

derbeDeus

20th January 2011, 18:18

And good luck breaking AES encryption and finding the private key to sign it with. AES is a symmetric algo, you don't sign with it. It's ECDSA that is used for content signing (with AACS_CC key).

Don't want to be a nuisance, but one last silly question (please): Has anyone actually tried to diff the dts track from a non cinavia watermarked DVD and the cinavia watermarked dts track of the Bluray? There is nothing to lose.

Well, yes there is... time. I'm sorry, but this is the sort of naïve approach that will lead nowhere. Such watermarks are not a black spot on an image, they have several layers and each encodes its information in a different way. You diff first layer, then what? You're still in the woods.

Fact: Cinavia it's more complicated than DVD-Audio watermark. So to start talking about removing Cinavia and be useful, a fair understanding of DVD-Audio is recommended, I'd say.

SamuriHL

20th January 2011, 18:21

I have to agree with Peer on that one. I think the majority of people are like me. We invest a lot of money in the equipment to play the discs. Then we spend a lot of money purchasing the discs we want to own every month. To protect that investment, we want to backup that large collection. Also, it's nice having it all ripped to a hard drive so your collection is easily accessable without fumbling around with discs. Back them up to ISO, mount them, and play them like the original. So far the software players aren't Cinavia infected, but, they all will be at some point or another. That's going to cause me a large headache for a while. Yes, I personally backup most of my discs as MKV and play them in an OSS player that ignores Cinavia, but, I also watch a lot of rented discs that I don't ever rip. And I *ALWAYS* have AnyDVD running when I watch ANY disc. God, SlyPlayer can't come out fast enough. :)

SamuriHL

20th January 2011, 18:22

AES is a symmetric algo, you don't sign with it. It's ECDSA that is used for content signing (with AACS_CC key).

My mistake. I'm not overly well versed in encryption. Just enough to be dangerous but clearly not enough to understand all facets of it. Thanks for the clarification. I'm guessing that's not any easier to break to find the private key to sign with? :D

derbeDeus

20th January 2011, 22:20

I'm guessing that's not any easier to break to find the private key to sign with? :D

It all depends on the implementation. While normally I should say "not in our lifetime", considering the latest mistakes made by Sony in PS3... it could be "in less than a minute" ;)

Only that AACSLA is not Sony, so they should be safe on this one. Btw: ECDSA seems to be the "latest fashion" for public crypto :p

SamuriHL

20th January 2011, 22:27

It all depends on the implementation. While normally I should say "not in our lifetime", considering the latest mistakes made by Sony in PS3... it could be "in less than a minute" ;)

Only that AACSLA is not Sony, so they should be safe on this one. Btw: ECDSA seems to be the "latest fashion" for public crypto :p

I admit to being behind on that one. Guess it's time to go look at it. I had been keeping up on all the "latest" encryption standards for a while but have slacked off lately. Cinavia seems to have upped the ante a little bit. As for your Sony comment, yeaaaaa, I'd say that's pretty accurate. If you can't beat em, sue em. ;)

xenex

22nd January 2011, 12:32

Just for my own interest, and testing I wanted to verify some of the AACS stuff on the "Takers" DVD. I first found the SHA-1 hash of the CSS_CONTENT_HASH_TABLE.AACS file:

xenex@MBOXL:~/Desktop$ openssl dgst -sha1 CSS_CONTENT_HASH_TABLE.AACS
SHA1(CSS_CONTENT_HASH_TABLE.AACS)= 21b2d3371de6a1aa167d53e5683a5c84051cee75

And I also made a small program to read the CSS_CONTENT_CERT file:

xenex@MBOXL:~/Desktop/aacschk$ ./a.out /dev/sr0

total_blocks = 2976742
AACS Folder Found!

FN: css_content_cert.aacs Found CERT LSN:2972640

FN: css_content_hash_table.aacs Found HASH LSN:2972641

Certificate Type : 80
Reserved : 00
Total # of Hash Units : 000038fa
Total # of Layers : 01
Layer_Number : 00
Number of Hash Units : 000038fa
Number of Digests : 0001
Applicant ID : 0002
Content Sequence # : c0020e80
Reserved : 0000
Reserved : 0000
Len. Format Spec. Sec.: 0000
Reserved : 0000
Content HTDigest : 68 3a 5c 84 05 1c ee 75
Signature Data : 66 83 32 0d 19 6c d8 a3 1c 11 e8 22 11 28 b7 49 9d e9 62 dd 72 79 bf 86 cc 9f 74 25 1b 31 c3 05 49 ba f3 6b 78 b1 33 1d

So, at least I notice that the final 8 bytes from "Content HTDigest" match the final 8 bytes of the SHA-1 hash of the CSS_CONTENT_HASH_TABLE.AACS

Nothing genius here, ofc, and I doubt the "Signature Data" has the "SONY mistake." But there is only the one "Takers" DVD to look at so far. Be interesting to look at the next few that come out.