AACS Hack Blamed on Bad Player Implementation

seriouslywtf writes "The AACS LA, those responsible for the AACS protection used by HD DVD and Blu-ray, has issued a statement claiming that AACS has not been compromised. Instead, they blame the implementation of AACS on specific players and claim that the makers of those players should follow the Compliance and Robustness Rules. 'It's not us, it's them!' This, however, does not appear to be the entire truth. From the Ars Technica article: 'This is an curious accusation because, according to the AACS documentation reviewed by Ars Technica, the AACS specification does not, in fact, account for this attack vector. ... We believe the AACS LA may be able to stop this particular hack. While little is truly known about how effective the key revocation system in AACS is, in theory it should be possible for the AACS LA to identify the players responsible for the breach and prevent later pressings of discs from playing back on those players until they are updated. As such, if the hole can be patched in the players, the leak of volume keys could be limited to essentially what is already on the market. That is, until another hole is found.'"

Comments

To be expected

[Iphtashu Fitz]

Did anybody really expect the AACS LA to say anything other than what they did? (Besides, maybe "we give up"?)

Re:To be expected

[purpledinoz]

I wonder what they're going to say when it's brutally apparent that ALL software players can be compromised. From what I can see, they have a few options, and none of them are pretty.

- play the cat and mouse game, and have the keys updated on the players while revoking the old keys.

- disallow software players all together.

- admit defeat and forget about revoking keys.

Re:To be expected

[MoxFulder]

I wonder what they're going to say when it's brutally apparent that ALL software players can be compromised.

In my mind, we're already there :-) The logical next step is to allow only hardware and partial-hardware players. For a PC, this would mean having some kind of "trusted" chip on your motherboard which can encrypt and decrypt data using keys that are hard-wired in.

Of course, hardware solutions can be broken too. I can envision a couple of ways this will happen:

• If the keys are truly embedded in the "trusted" ASIC: Making custom chips is expensive. There are substantial setup costs for each new mask, so there will be enormous economic pressure to only have one or a few versions of the chip. This means once one version gets cracked, millions of computers will be freed. What will it take to read the keys off an ASIC? A scanning electron microscope, that's what. As a bored physics grad student currently sitting 10 feet away from an SEM, I can tell you it'll happen :-)
  
• If the keys are somehow individualized to each computer, they'll be stored on a flash-based FPGA, or in some kind of microcontroller's flash memory. Manufacturers of such flash-based devices go to great lengths to make it so that the code stored in flash can't be read off of the device, but this is nothing more than the same ol, same ol security through obscurity... figure out the magic voltage that you need to apply to pin 12, and oops there goes the security. Smart card hackers have already figured out ways around the protection in the common PIC16C84 microcontroller.
  

Bottom line: DRM is futile because it requires the distribution of a SECRET PIECE OF DATA (the decryption keys) in UNENCRYPTED form (the keys themselves must of necessity be unencrypted). All the crap interposed between the user and the keys is merely security through obscurity. QED.

Re:To be expected

[dr_labrat]

yup, and there it is folks.

For the uninitiated, (i.e. non-security chaps), fundimentally when it boils down to it, its irrelevant how good the encryption mechanism if someone is sitting over your shoulder reading the information.

I really wish the DRM happy crowd would understand that if it gets to be decrypted by a bit of kit that can be in "hostile" hands it is not going to be "secure" for more than 2 months (see DeCSS, Fairplay, Microsofts thingy, BlueRay, um.... Wait... all DRM thus far has been cracked in less than 2 months.).

Frankly its absurd. You employ a team of 50 programmers to make the next greatest hack proof DRM schema, however you are (if you make anything worth viewing/listening to/playing) up against at least 1,000 times that in terms of people that are interested in breaking it.

The worst thing is: The crackers only need to find one way to break it.

Hey ho. The reality of the situation is that DRM is costing the media conglomerates more to implement than the potential losses.

Its like putting a $200 lock on a $20 bike.

If I like I buy. If I can I take. If had taken, it doesn't mean I would have bought it.

If I like something I have taken I will buy it.

Re:To be expected

[Lumpy]

And how will that stop a well equipped Hacker or a 12th year grad student hacker at the MIT electronics engineering labs from ploping a hardware player on the desk and reading the contents of the ram directly?

Hacking a software player is only a bit easier if you have the tools. Hacking hardware players is as simple if you are equiped with the right analyzers and equipment.

Hell if they fudged up and used sram you can halt a processor and read the contents of ram between each processing cycle pretty easily.. DRAM is a bit more difficult you simply need to supply the refreshes between processor haltings.

HDDVD and BluRay are big enough targets that the hacking community will start taking these advanced approaches.

Re:To be expected

[alienw]

First, ASICs are not expensive. They are in fact extremely cheap to produce, and the development costs are not that high and are easily justified in a mass-market application. Nobody in their right mind would use an FPGA in a consumer application -- they are far too expensive.

Second, I don't think you will be able to read off keys with any kind of microscope. I don't think you'd be able to find out the key even if you had a complete wall-poster-size plot of the chip. I don't think you quite appreciate the complexity of a chip. Even low-end ASICs push millions of transistors these days. About the only method that can be used to steal keys is wafer probing, and that's pretty hard to do with modern chip densities.

Reading data from a flash EEPROM is even harder. Engineers who design chips are generally much smarter than people who try to break them, and there are plenty of tamperproof chips available. Most tamper-resistant chips now incorporate self-destruct features that erase the data when you try to probe the chip or screw around with its supply voltages or clocks. The industry has come a long way since the 16C84, which wasn't even intended to be tamperproof.

I am also not sure what your point is with regard to keys. Any secure system ultimately depends on the security of its keys.

Re:To be expected

[Anonymous Coward]

Yes !!! right!!!
Who is gonna rush and by these new MB's ???
I don't think MB manufacturers are going to change the production
line for the movie industries.

Re:To be expected

[MemoryDragon]

Good morning mc fly

almost every pc sold in the last three years has this chip, it is called TCPA and
one of the key areas this chip should be used for was BlueRay and HD-DVD
it is just thatthere are lots of PCs nowadays which do not have those chips.

Re:To be expected

[C32]

Yes! The unfounded optimism of halfbaked ee/compsci slashdotters never fails to annoy me.. :=)

Re:To be expected

[makomk]

First, ASICs are not expensive. They are in fact extremely cheap to produce, and the development costs are not that high and are easily justified in a mass-market application. Nobody in their right mind would use an FPGA in a consumer application -- they are far too expensive.

Funnily enough, apparently one of the early standalone HD-DVD players does have a FPGA on board, as well as a CPLD.

Re:To be expected

[smallfries]

ASICs are not expensive if you're designing a high-priced piece of consumer electroncis where you can absorb the cost into your fat generous margins. If you're aiming at the disc player market then you're competing against cheap imports. DVD players are now so cheap here that you can't give them away (about £30 last time I looked).

But we're not talking about a common ASIC for each player - you've twisted the GPs point. We're talking about a unique ASIC for each player, and making runs of 1 ASIC would be unimaginably expensive. Hence the FPGA route would need to be taken to avoid a single key across the players.

Reading keys off with a microscope has been done. That is how the 2048bit Xbox private key was compromised. Of course the gradstudent that did it couldn't tell anyone what it was, and had a Microsoft goon at each one of his seminars, but it still prooves that it can be done.

Nobody has ever made a tamper-proof device. There are many approximations on the market - things that will resist X amount of tampering before they fail, but any tamperproof box will fall to a determined adversary. When tamperproof casing are designed, the measure used is how much effort / cost can we force the adversary to use before they gain access.

The GPs point was that, by necessity, DRM requires unencrypted information to be hidden in plain sight. Furthermore, this "secret" is common. So there is a single point of attack in the system, which when breached compromises the entire system. This is his point with keys that yuo missed. DRM cannot work unless the the secret keys are available in plaintext. Hence the system is always screwed, by design.

Re:To be expected

[MoxFulder]

Thank you! You've clarified everything in my original post, much better than I could have.

I guess for the manufacturers of "trusted" chips, there's a tradeoff between uniqueness of keys and ease of finding them:

• If the keys are hard-encoded into the mask of the chip, then they will be difficult to locate and recover. But cost considerations (e.g. the pressure to make $20 DVD players as you pointed out) will mean that the number of unique keys is quite small.
  
• If the keys are stored in some mutable memory, then they will be easier to locate thanks to semiconductor processing considerations. The process for making flash memory is different from that for making CMOS logic, so the key will be in an isolated and distinct part of the chip, maybe a separate die. But on the other hand, there could be a unique key for each player.
  
• ???? Maybe there's some intermediate case I haven't though of ????

Re:To be expected

[alienw]

DVD players are now so cheap here that you can't give them away (about £30 last time I looked).

DVD players are that cheap BECAUSE they use ASICs and ASSPs. A DVD player typically consists of one or two application-specific VLSI chips, each of which costs 1-3 dollars to produce. If you tried to implement this using off-the-shelf DSPs and general-purpose chips, a DVD player would cost thousands of dollars.

But we're not talking about a common ASIC for each player - you've twisted the GPs point. We're talking about a unique ASIC for each player, and making runs of 1 ASIC would be unimaginably expensive.

Who in their right mind would do that? You can easily make a chip with OTP fuses or a secure EEPROM area if you want to give each one a unique key. But even if all of them have the same key, it would be pretty much impossible to extract.

Reading keys off with a microscope has been done. That is how the 2048bit Xbox private key was compromised.

Uh, no. The Xbox was cracked by snooping on unencrypted bus traffic on the LDT bus. The guy did try to decap the chip, but could not see anything other than the top metallization layer (which was mostly power distribution lines).

Nobody has ever made a tamper-proof device.

How many cable boxes, smartcards, or satellite TV/radio systems have been seriously compromised? I am not counting the primitive card-cloning/emulation schemes, but rather obtaining the actual encryption keys.

DRM cannot work unless the the secret keys are available in plaintext.

Embedded within a chip is not exactly plaintext. I don't think a single hardware encryption chip has ever been compromised. They have been bypassed, and software players have been cracked, but I don't ever remember someone extracting keys from hardware.

Re:To be expected

[alienw]

The only reason those players use FPGAs and CPLDs is because they were basically rushed to market and cost was not a consideration. The 2-3 years it takes for the players to reach the $100-$150 level is basically the time it takes to replace all the standard chips with specialized ones. Of course, these days, it is much easier to transition an FPGA design to an ASIC -- you can recompile it with ASIC tools, add boundary scan capabilities, redo the verification steps, and send it off to a fab.

Re:To be expected

[smallfries]

My memory may be off on this. I was pretty sure that I recalled seeing a talk from a student at Cambridge who had scraped the key directly out of the chip. The only references that I can find on google are for the bus snooping attack - so maybe I've remembered it wrongly.

There is no compelling reason that a "tamper-proof" chip cannot be broken. The only limit is how much it costs. The bus-snooping paper about the xbox mentions that taking the key directly was one considered approach that they didn't go with due to cost. The 7-layer process takes some specialised equipment to decode.

As for whether or not keys have been compromised in the wild - I would assume that they have been, and I would assume that the information has not been released publically. There is no reason to assume that they are secure. If it can be done, and there is a motive for some to do it...

Re:To be expected

[alienw]

There is no compelling reason that a "tamper-proof" chip cannot be broken. The only limit is how much it costs.

Cannot be broken by anyone? Maybe not. Cannot be broken without spending a few million bucks? Sure. Really, the chip is pretty much unbreakable if it costs more to break it than it does to just steal the keys directly from the manufacturer. Industrial espionage is often a much more practical solution.

The bus-snooping paper about the xbox mentions that taking the key directly was one considered approach that they didn't go with due to cost.

I am sure they considered the approach, but just saying it was due to "cost" doesn't even begin to describe it. Really, I don't think it would even be possible. Even a high resolution 10-foot wide plot of a digital chip looks like a jumbled mess. Those things are automatically synthesized and routed. I've seen them and it's not pretty -- even the designer probably couldn't find the key by looking at the layout. If you think you'll see anything useful with an electron microscope, you are deluding yourself.

As for whether or not keys have been compromised in the wild - I would assume that they have been, and I would assume that the information has not been released publically.

I fail to see the basis for your assumptions. There is no reason for anyone to crack satellite or cable receivers and not tell anyone about it.

Re:To be expected

[smallfries]

I am sure they considered the approach, but just saying it was due to "cost" doesn't even begin to describe it. Really, I don't think it would even be possible. Even a high resolution 10-foot wide plot of a digital chip looks like a jumbled mess. Those things are automatically synthesized and routed. I've seen them and it's not pretty -- even the designer probably couldn't find the key by looking at the layout. If you think you'll see anything useful with an electron microscope, you are deluding yourself.

From Huang's paper:

The direct approach of decapping the MCPX southbridge ASIC was rejected be-cause this ASIC appears to be manufactured in a 0.13 process with perhaps 6 or 7 metal layers (gure 2). Extracting the bootblock from this ASIC would require a de-
layering facility and access to an electron microscope. While there are companies such as Chipworks that specialize in these kinds of services, it is a difcult, expensive, and time-consuming task

It is possible, so I am not the one who is deluded on the subject.

I fail to see the basis for your assumptions. There is no reason for anyone to crack satellite or cable receivers and not tell anyone about it.

Information is value, if you can't see the point in cracking pay-TV and with-holding the information then I'm not going to explain it to you. But you should probably read Tamper Resistance - A cautionary note and Low cost attacks on tamper resistant devices before you make any further gross assumptions.

Re:To be expected

[alienw]

I really don't think Chipworks would risk a multibillion dollar DMCA lawsuit, so it is highly unlikely that they would have been able to help Huang, even if he did manage to raise a few million bucks. In addition, I don't think they would have the capability to extract the key from a well-protected circuit. The main deterrent is that it would have cost a few million bucks and that it is completely illegal.

Information is value, if you can't see the point in cracking pay-TV and with-holding the information then I'm not going to explain it to you

The only purpose for cracking a pay-TV system is to steal service. If the information is not widely available, then the integrity of the system has not really been breached.

As for the two attacks you linked to: they are highly theoretical, apply largely to smartcards, and are trivial to guard against. For example, the processor could be reset or even destroyed by a supervisor circuit if the supply voltages or clocks are out of specification. Tilt bits and hardware checksums could be used to guard against data corruption due to external influences. Design measures could be used to effectively guard against microprobing or surface attacks.

Re:To be expected

[jackbird]

Cannot be broken by anyone? Maybe not. Cannot be broken without spending a few million bucks? Sure. Really, the chip is pretty much unbreakable if it costs more to break it than it does to just steal the keys directly from the manufacturer. Industrial espionage is often a much more practical solution.

Hmm, who has a vested interest and a few million bucks? Oh, that's right, the mafia does...

I'm mixed on this.

[grub]

Part of me wants them to find a proper fix for these holes. My CableCo phoned me because I've already gone way over my quota this month.

Re:I'm mixed on this.

[Sheltem The Guardian]

I am curious. If we've, let's say, pwned a hardware player. And AACS LA revokes original keys. They don't want buyers to kill them, so they have to reissue some keys. But if we've already pwned device, can't we happily receive these freshly-reissued keys, pretending we're this device? We might know every cryptokey this device holds at this moment!

Re:I'm mixed on this.

[ThePiMan2003]

Actually that is part of the spec. They can kill your hardware player, and then blame it on a poorly made hardware and you the end user are SOL.

Re:I'm mixed on this.

[Sheltem The Guardian]

I guess that'll be a big bang if they're ever recall keys for a large and still selling generation of players. Hardware makers didn't understand that they've gave hollywood just enough rope.

Re:I'm mixed on this.

[The Warlock]

Well, the solution to that is easy. Rip keys from a very prolific hardware player.

Imagine if the keys that got leaked came from, say, the PS3. Can you imagine the shitstorm that Sony would throw if the first million or two buyers couldn't play Blu-Ray movies anymore? Those keys would never get revoked.

Re:I'm mixed on this.

[ThePengwin]

Those keys would never get revoked.
Or would they..... it would be quite a humorous predicament if The Xbox 360 and the PS3 had a feature forcibly removed from them :P

Re:I'm mixed on this.

[GWBasic]

Well, the solution to that is easy. Rip keys from a very prolific hardware player. Imagine if the keys that got leaked came from, say, the PS3. Can you imagine the shitstorm that Sony would throw if the first million or two buyers couldn't play Blu-Ray movies anymore? Those keys would never get revoked.

I can hardly wait until there's a SETI-style distributed brute-force attack on AACS to get every single player-specific key.

Of course not, dear...

[bhamlin]

Of course it's not your fault. Your highly paid engineers are WAY smarter than anyone else.

DRM is silly

[tfinniga]

You give them the lock.

You give them the key.

You hope that they can't figure out how to put one into the other.

High fives.

Re:DRM is silly

[Abnormal Coward]

I agree. The only way to show that this DRM protected is shite is for people not to buy. Copying media in my option has never been a problem, I've had a a lot of tape copys from people and went and brought the cd/tape because I really like the music. Same with movies and TV, I've brought DVD's and TV boxed sets after downloading DIVX copys from the 'net. If the boys at the top (RIAA/MPAA) ensure there music is cheap enough its a no brainer. The real battle is here is that 'they' want to tell you want to buy and set any price they like. Its all about control (time to put on your tin hat). Well fuck them, where the consumers we should decide what to buy, and what is an accecptable price. So back to my orginal point, the only way to show is with your wallet ....

Re:DRM is silly

[CoughDropAddict]

Putting the lock deep in silicon, where no software can touch it (or only specifically authenticated/authorized software), does not count as "giving them the key." This is the direction DRM is moving.

Re:DRM is silly

[et764]

Still, the machines are made up of electrical pulses moving across the chip. These electrical pulses can be observed and manipulated. As long as you have physical access to the playback device, which won't go away as long as you can use your media at home, there exists some way to get the hardware or software to reveal the key. It may take a whole lot of creativity, trial and error, but it can be done.

Re:DRM is silly

[Anonymous Coward]

If there's one thing history has proven, it's that encryption is an absolute, unbreakable method of keeping people out of things you don't want them in.

Period.

Re:DRM is silly

[Anonymous Coward]

Except you have to give people a key to decrypt the encryption, or no one anywhere will see what's inside. Dumbass.

Re:DRM is silly

[Anonymous Coward]

Whoosh.

Re:DRM is silly

[xlsior]

The only way to show that this DRM protected is shite is for people not to buy

Of course, that's only part of it:

MPAA boardroom meeting: "Hey guys, this years sales figures continue to drop... What's going on here?"
"Must be evil pirates again stealing our movies! Lets slap on some more DRM, that'll stop them!"

While it's always good to vote with your feet, it probably won't do a whole lot of good unless they also hear a lot of noise about WHY people choose to no longer rush out and buy the latest DRM-infested offerings.

Just a thought.

Re:DRM is silly

[kripkenstein]

You give them the lock.
You give them the key.
You hope that they can't figure out how to put one into the other.
High fives.

Yes, any DRM method can be circumvented, as you indicate. But they still work. Thing is, DRM isn't meant to protect against Slashdot-reader-type people; we will always find a way to get our fair use rights back. But for 90% of the public, DRM works perfectly - they never heard of DeCSS, they won't know how to download this particular AACS hack.

Why do the DRM people even try to get DRM to work for everyone, that is, to prevent it being hacked by power users? Publicity, mostly. Yes, it's a losing battle, and they surely know that it is, but they can't let it be heard that 'DRM has failed'. So they will band-aid every breach, always a few steps behind the hackers. They can lose this battle, they have already won the war.

There isn't much we can do about this. I help out my non-geek friends with this sort of thing, but the majority of non-geeks probably just live with DRM. The only real solution to this would be political (like Norway, but against DRM in general, not Apple in particular).

Re:DRM is silly

[elrous0]

The only ones who are really winning are the software engineers and contractors who sell companies on these dumb-ass DRM schemes to begin with (and also the true pirates who know it won't effect them anyway). The biggest loser is the average consumer who just wants his video to play without a bunch of hassle.

-Eric

Blame Canada

[euri.ca]

It's a widely known fact that Canada is responsible for 50% of the HD DVD piracy.

Even worse, the AACS specification does not, in fact, account for this large sparsely populated country.

Re:Blame Canada

[grub]

It's a widely known fact that Canada is responsible for 50% of the HD DVD piracy.

The problem is all those HD-DVD bootlegs are cams done in Montreal.

Re:Blame Canada

[Anonymous Coward]

Unlikely. That story tells us that camcording is illegal in Canada. That the theaters are using night vision to catch people doing it. That Canadian police are arresting and convicting people who do it. And that anyone can rent the movie at blockbuster and copy it - in the US (which has ten times the population) as well as Canada.

This little statistic was almost certainly made up, to pressure Canada into taking away more fair use rights.

On the other hand, word has it that the guy who bypassed HD DVD's AACS encryption is a Canadian programmer. (With possible help from a NZ cryptographer. Google "My first experience with HD content being blocked", and compare his story with that of Muslix64.)

Re:Blame Canada

[PitaBred]

Holy balls are you really that obtuse?

Dig it

Re:Blame Canada

[Jesus_666]

It's a widely known fact that Canada is responsible for 50% of the HD DVD piracy.

Indeed! Out of the five HD-DVDs in circulation, three are illegal copies. Two were made in Canada and one in China, where people have met the booming HD-DVD market with irrational exuberance.

Never!

[Troed]

if the hole can be patched in the players

It cannot, ever, unless they disallow software players from any platform not running on Trusted Computing enabled hardware and a Trusted Computing enabled operating system.

Until then, no DRM scheme works.

None.

It's that simple.

Re:Never!

[CrystalFalcon]

It cannot, ever, unless they disallow software players from any platform not running on Trusted Computing enabled hardware and a Trusted Computing enabled operating system.

And at that point, virtualization kits will become commonplace that run Windows in a sandbox so that Windows thinks it's in a Palladium environment, but where it's really not.

If it can be played, it can be copied. Playing is copying. Any manipulation of digital data is copying it. Trying to make bits not copyable is trying to make water not wet.

Re:Never!

[purpledinoz]

So the marketing department is telling the engineers to do something impossible. Sounds like a scene from a Dilbert comic.

Re:Never!

[AuMatar]

Dilbert? I call that Friday afternoon.

Re:Never!

[mugnyte]

There are ways combat this - like requiring timing that only hardware can satisfy, but virtualization is a tough thing to hide from. In the end, it will require an dual-key system from each piece of hardware that the system accepts. You cannot write a virtual one because you cannot provide a valid key. Yes, yes, I know this is a terrible design.

  Then you degrade the problem to a Man in the Middle, where your microcode simulates a processor and performs some operations before/after sending to same/different hardware. Microcode is the standard for many OS's now.

  In these cases, the OS sends a public key to the hardware, and receives one in return, you can capture them but cannot mimic these pieces. Then, each buffer in the pipeline ends up encrypted, leaving you to decipher.

  This is the gist of the whole architecture: locking down anywhere one could put custom code. The problem is, in a heterogeneous environment, there's no much stability with asking a whole market to obey these specs. Someone is going to write hardware that conforms, but has an unencrypted out channel. In fact, companies will simple comply to the Trusted Computing program but sell this out at a high price. MS creates a valued market out of it's security scheme, losing both the anti-piracy initiative, and the content providers' trust, eventually.

  Until these phases come to pass, the market moves slowly to adjust to the new formats and pricing. Content providers pour into the channels believing the issue is "solved". Then, suddenly, an unrevokeable layer is compromised (as in: you cannot re-stamp all the discs already on the market) and much of the content appears in black market format. The market floods easily because people do not believe the cost of the model is worth the output (like music today).

  If you think I'm speculating, all of this has happened before. Hacking in all it's forms has never had any different lesson.

Re:Never!

[morgan_greywolf]

And at that point, virtualization kits will become commonplace that run Windows in a sandbox so that Windows thinks it's in a Palladium environment, but where it's really not.

But will these virtualization kits have the TCPM-enabled BIOS? What about open source virtualization kits? IIRC, Mac OS X can run in a virtualized environment only once the checks for the hardware DRM are removed by a hack. It seems like there would be copyright or patent issues that would prevent a FOSS virtualization kit from running Windows in a Palladium environment.

Re:Never!

[Rich0]

Mac OS X can run in a virtualized environment only once the checks for the hardware DRM are removed by a hack,

And this only works because MacOS X needs to be able to run offline.

Once your HD-DVD player is required to be hooked up to the internet to work, that hack will stop working. The key to your DVD won't be stored on the DVD - instead you'll download it each time you play it. However, the server hosting the key won't provide it unless you pass a TCPM check - if the hashes of the bios/OS/player aren't trusted the key won't be provided. Without defeating the physical security on the TCPM chip you won't be able to simulate/virtualize/emulate/whatever this handshake - it ultimately relies on a private key buried in the chip that never leaves it.

Re:Never!

[Anonymous Coward]

Trying to make bits not copyable is trying to make water not wet.

Pff, that's easy. Just freeze it!

Re:Never!

[eMbry00s]

So... what if we reduce these bits to 0 Kelvin?

You just gave the TCPA engineers a huge lead on what to do now! D:

Re:Never!

[Anonymous Coward]

No matter how ridiculous you make your DRM scheme, its still going to fail. Thats because DRM is fundamentally flawed.
You CANNOT prevent video or audio from being copied, and that's all there is to it.
Audio: 3.5mm->3.5mm cable and any kind of audio recorder. You can't send data to speakers without converting it to an analog signal at some point.
Video: You cannot send data to someone's eyes without converting it to an analog signal at some point, same as audio. HD video camera+TV+color correction=DRM pwned.

It simply can't be done perfectly. Once someone gets a copy of it in an unprotected format whether by the above techniques, by a leak, or by simply breaking the DRM mechanism, its over. Its on bittorrent, posted to thepiratebay, and anyone can download it.

I don't illegally download things, and I don't want to be restricted by bullshit when I want to watch the movies that I own, that I bought with MY money.

Re:Never!

[BryanL]

Like ice?

sceurity holes

[stocke2]

all the windows security problems are caused by the computers these companies build to run it, I mean windows is perfectly secure so long as it sits in it's box there on the shelf....if these darn companies would just stop installing it on computers there would be no problem

Re:sceurity holes

[cheater512]

Oh its secure when installed a computer. Super secure even.

Just as long as that computer has no input or output devices that is.

No AACS, Blu-ray, HD-DVD for me.

[Anonymous Coward]

Since July of last year I have basically cut out the mass media from my life. I sold my TV, gave away my DVD player, and donated my CDs and DVDs to a charity auction. For entertainment, I've taken up a number of sports, including basketball and skiing. I also now listen to local bands live at pubs and restaurants, rather than listening to the radio or CDs. I never had any gaming consoles to begin with, and I uninstalled and gave away the few computer games I do have. I do rely on the BBC for news, but even that's become limited these days.

I'm glad I made that decision. All this new crap involving DRM and frivolous from the entertainment industry just goes to show you how full of horseshit they are. I'm very pleased that my money does not go to them. They don't deserve it. Not only that, but now that I play sports rather than just watching them on TV, I've become much more fit and far healthier. Getting away from the mainstream media was one of the best things I've ever done.

Re:No AACS, Blu-ray, HD-DVD for me.

[Anonymous Coward]

Good for you. That means the only reason you would be reading an article about those things you "gave up" was so you could blow your own horn. Guess what? We don't give a shit.

Re:No AACS, Blu-ray, HD-DVD for me.

[nuzak]

How about taking the next step and cutting out slashdot from your new life?

Re:No AACS, Blu-ray, HD-DVD for me.

[Loconut1389]

Yes, yay to the courageous anonymous coward.

Re:No AACS, Blu-ray, HD-DVD for me.

[eln]

Sure you have a more fulfilling life, but you're generating massive amounts of smug, which is highly toxic to the environment.

Re:No AACS, Blu-ray, HD-DVD for me.

[SEAL]

Yes, yay to the courageous anonymous coward./i. ... because a Slashdot user-id is so much LESS anonymous?

Re:No AACS, Blu-ray, HD-DVD for me.

[Anonymous Coward]

You no longer belong here.

You are not one of us
You are one of them

Kill the outsider!!

Re:No AACS, Blu-ray, HD-DVD for me.

[Anonymous Coward]

Have you ever tried listening to a live band on your commute to work?
How about skiing on a lazy, summer afternoon?

Good luck with your new lifestyle.

Re:No AACS, Blu-ray, HD-DVD for me.

[MrNaz]

Well done. The only "media" that I "consume" is the Internet, and the two way nature of that really differentiates it from movies, TV and radio IMHO. It is an interactive activity as opposed to a mind-numbing method to kill time. I don't have a TV in my home at all. However I occasionally cheat and watch a movie on my laptop in bed. (I anticipate lots of jokes about porn right about now. This is /. after all, a thread without toilet humour would be out of place.) What sports have you taken up? I chose all the non-standard ones, skipping soccer and football for things like rock climbing, hiking, scuba diving and snowboarding. I wish you well, fellow unplugger!

Re:No AACS, Blu-ray, HD-DVD for me.

[Loconut1389]

I was making a joke about: courageous and coward being used in series.

Re:No AACS, Blu-ray, HD-DVD for me.

[Anonymous Coward]

Sure you have a more fulfilling life, but you're generating massive amounts of smug, which is highly toxic to the environment.

Nah, it's only toxic to people with unfulfilling lives... funny how that works.

Why not go to the real source?

[kad77]

AACS LA executive decision maker:

A. Test their player key detection procedures (prove they can't reverse Volume Unique Key)

B. Read the public forums that have trumpeted the players and methods involved directly.

Ed Felten writes about an economic model...

[Saint Aardvark]

...for this fight at freedom-to-tinker.com. The whole series on AACS is worth reading, as is every single thing he posts.

Ahh... the fun begins!

[monopole]

If they are really going to use the device revocation option, things are going to get way fun.
Players which will only play certain discs and not others, instant obsolescence for entire classes of $1000 players.
This makes the format wars look like a sales promotion!

Re:Ahh... the fun begins!

[Sircus]

I'm no fan of the content mafia, but all they're talking about at the moment is disabling certain software players which the publishers could easily offer free updates for. The current crack isn't applicable to hardware players.

Re:Ahh... the fun begins!

[H0ek]

This is also a sure-fire way to kill a format. Usually technology is promoted via word-of-mouth, and when the drive of the early adopters begin to fail, the word will spread that you can't trust either Blu-Ray or HD-DVD.

In short, AACS is doomed if it does, doomed if it doesn't.

Re:Ahh... the fun begins!

[sdo1]

Agreed. If one of my friends asks me about these formats (and they do, knowing what a home theater and media junkie I am), I roll through all of the DRM hoops that they'll have to jump through in order to play the things the way they want it.

The industry NEEDS the word-of-mouth. And as it stands, that word-of-mouth is negative. It's "yea, the picture is great, but then there's all this other stuff you'll have to deal with." That's not going to fly.

-S

Re:Ahh... the fun begins!

[Rich0]

The current crack isn't applicable to hardware players.

Well, sure it is - but it would be a lot harder to pull off.

Re:Ahh... the fun begins!

[Jeff DeMaagd]

No, not entire classes. Single players can be revoked, and other players of the same model won't be.

Re:Ahh... the fun begins!

[cheater512]

You just know someone is going to get sued for putting a feature like that in.

Player Keys

[Anonymous Coward]

Just wait until they start finding some *player* keys (not volume keys).

Gather enough of those and you can screw revocation by subverting the master key authority. Hopefully, they'll quietly hack the player key, get them to issue a new one, hack that and....

Re:Player Keys

[Goaway]

No, you can't. Hardware players use individual key sets and can be individually revoked, and player keys are revoked every six months anyway.

Re:Player Keys

[Anonymous Coward]

Wrong. Using a little linear algebra and those player keys, you can find the key used by the master key authority and subvert it. So long as they don't start using a new master key or something, there's not much they can do about it, even if they decide to revoke each player key the second it's released.

There's a proof of this floating around on the internet. Of course, I think they said it'd take something like 42 player keys, worst case.

Re:Player Keys

[Watson Ladd]

The way AACS works is players are arranged in a tree. Each player has the keys that generate the other branch of the tree all the way up to the top. If you break 2 players, seperated a distance n completely, you break n other players as they have keys you can find out.

Re:Player Keys

[Goaway]

That was a crack of HDCP, not AACS. Try to actually know what you are talking about before accusing others of being wrong.

Re:Player Keys

[Goaway]

References for that?

Yeahbut...

[Iphtashu Fitz]

Don't forget that if you drink the *AA koolaid and bend over properly for them then you've hooked up your DRM infested HD-DVD or Blu-Ray player to the internet not only so that they can track what you watch but so that the players can automagically download updated keys so that you'll never even see any non-functioning disks. That, and monkeys may fly out of their respective butts.

Updated? Battle of the Rootkits!

[Tackhead]

> While little is truly known about how effective the key revocation system in AACS is, in theory it should be possible for the AACS LA to identify the players responsible for the breach and prevent later pressings of discs from playing back on those players until they are updated. As such, if the hole can be patched in the players [emphasis added], the leak of volume keys could be limited to essentially what is already on the market.

If the players are non-patchable:

1) We will live in a universe in which, every year or so, an unknown number of players will play discs produced up to, but not after, a certain date.

Consider the sales/support implications of customers selecting products for Christmas 2008: "Well, sir, this Foobar-1000 plays discs up produced in 2006-2007, a Foobar-1130 plays discs produced from 2006-2008, and a Fonybaz-1900 plays discs produced from 2006 to August 2008."

If the players are patchable, it's even worse for the industry:

1) Your Foobar 1000 will play discs produced in 2006 and 2007. It ceases to work for discs produced between February 2007 until you buy a disc produced a few months later that happens to contains some code that query the player whether it's a Foobar 1000... and if so, to automatically/silently patch the firmware. Then all your discs work again.

That's a good thing for the user, and a bad thing for the industry, because as soon as you've got a firmware patch on a DVD, the obvious thing for an enterprising hacker to do is to put his own firmware patch on his own DVD, and your Foobar 1000, all of a sudden, ceases to implement the DRMish crap which the MPAA crammed onto it...

...until, of course, a few months after that hack, where the firmware-updating discs are modified to downgrade any hacked players to MPAA-compliant revisions of the firmware (or even to self-destruct)...

...and someone else comes up with a better hack to make the hacked firmware indistinguishable from the "approved" firmware...

In short, if players can be patched in the field (and this applies to both hardware/firmware-based players in embedded systems and to PC-based disc-playing software), it's a long-term battle of the rootkits, and that's a battle that MPAA is likely to lose.

I thought the player key hadn't been revealed?

[Jartan]

Why is Ars saying they believe they can stop this hack by revoking the player key? The original person who cracked it specifically didn't release the key I thought and was only releasing TITLE keys which will be much more dangerous to revoke yes?

Not that it matters much either way because this attack vector will always exist for any kind of system they come up with. Since it will always exist someone will rip it and post the movie on bittorrent.

They are actually probably pretty happy that this is the only possible hack anyways since it isn't anywhere near as useful as DeCSS.

Re:I thought the player key hadn't been revealed?

[Iphtashu Fitz]

Since it will always exist someone will rip it and post the movie on bittorrent.

And that's the lynchpin to the whole DRM debacle. All it takes is one individual somewhere on the planet to manage to crack or circumvent the encryption on any given movie to make it available to everybody. It may take some time but it's likely to happen eventually.

Re:I thought the player key hadn't been revealed?

[nuzak]

All it takes is one individual somewhere on the planet to manage to crack or circumvent the encryption on any given movie to make it available to everybody.

And, unlike the disc you legally purchased, the cracked version is pretty much guaranteed to actually play on your hardware.

Re:I thought the player key hadn't been revealed?

[Pootie Tang]

The hack extracts the title key by grabbing out of memory. It's in memory because the (software) player puts it there, at least temporarily.

If they revoke the key for that player on future discs then (that particular) software player won't be able to extract the title key, thus the hack can't access it.

In theory this should work, but not for any discs already created, only going forward. Of course all that really accomplishes is making someone have to compromise a different player, it's not exactly checkmate.

Re:I thought the player key hadn't been revealed?

[cheater512]

Would a website which was basically just a database full of constantly updated title keys be legal if it wasnt hosted in the US?

FOSS HD-DVD/Blu-ray players could just query it to play any movie of their choice.

Re:I thought the player key hadn't been revealed?

[Dunbal]

Would a website which was basically just a database full of constantly updated title keys be legal if it wasnt hosted in the US?

      Probably not. Remember how the *AA's can somehow strongarm foreign police forces to do their dirty work for them. Legality doesn't come into it anywhere.

Re:I thought the player key hadn't been revealed?

[gatzke]

The easy fix for them is to revoke the keys for the player they assume is at fault. Easy to do, but the movies pressed so far will be in the open.

The other fix I read about is to release a single movie with hundreds of different keys. Instead of "pirates" releasing one cracked key per movie, they would have to get all of the hundreds of keys out there and match them as well...

What you need is a 1080p LCD with HDMI HDCP attached to a 1080p scanner... Like the good old days where you put your boom box next to another boom box to dub cassette tapes... Love the 80s!!

bwa.ha.ha.

[geekoid]

Dear consumer:
Please check our website so you can download a patch and intall it on your DVD player.

BWahahaha..

That will go over like a lead balloon.
as will a machine that no longer playing new movies every few months so you have to buy a new player.

Which is good. DRM is just causing more consumer frustration and less value.

they can't patch these holes away

[idlake]

You can play around with keys so that the same player won't play both old and new discs, but that doesn't change the fact that the old software will continue to be out there and will continue to be able to play old discs. And next time someone screws up, all the discs up to that point will be compromised, and on and on.

Revocation is pointless

[asc99c]

I don't understand the point of revoking a hacked key. Now the key has been found and discs have been hacked, the output of the process is an unencrypted file with no key. Until something like AnyDVD comes out that just silently and automatically strips encryption on the fly, the primary use of the program will be to get unencrypted content onto P2P networks.

Why bother revoking the key? I must be missing something. Sure, don't use the same key on future discs, but pirated copies will have no encryption - key revocation only seems to affect legitimate users of the disc.

Oh yeah, I'd forgotten, DRM isn't about piracy...

Re:Revocation is pointless

[Wesley Felter]

Not everyone can afford to download 20GB files. Plenty of people rip DVDs that they own/rent even though the same movies are available for download, and those same people will want tools to locally rip HD-DVD and Blu-ray discs.

Re:Revocation is pointless

[discordia666]

Well that's a damn good point. But you may be mixing up your keys. Easy to do. But hacked player keys in the wild will allow people to decrypt any disk. Hacked volume and title keys are the ones tied to the disk itself. No mechanism to revoke those, until new disks are manuafactured. But by that time your point comes into play and it doesn't matter.

Re:Revocation is pointless

[asc99c]

I occasionally do the opposite of this for my media centre - download a 1.4GB AVI because it's easier than ripping the DVD. To rip the DVD, I've got to either just rip the raw files which takes a lot of HD space, or I've got to know a lot more about the encoding options - my rips from DVD to XVID have always been poor quality compared to what is available online. Looking at the BBC HD stuff, it seems HD can be compressed to ~ 3-4 GB / hour while retaining most of the quality. The 1080 demos available from the divx.com website are also stunning at those sort of bitrates.

Unfortunately I've not yet seen any 20Mb/sec 'full' bitrate HD stuff yet. I've just bought a 1080p TV and I will buy in to the Blu-Ray / HD-DVD stuff, but not to the DRM side (same as for DVD). I am against piracy as it is clearly not morally OK, but defeating the hideously annoying uses to which DRM is being put does give me a nice warm feeling. While I can agree the copyright owner technically can impose whatever restrictions they like, I don't seem to remember ever agreeing to such restrictions when buying DVDs, and I don't think anyone can have any moral objection to format shifting and skipping copyright warnings.

The last straw for me on DRM was trying to watch my new Donnie Darko DVD while on the exercise bike. I got to see an unskippable copyright warning in English, then French, German, Italian, Spanish and even Suomi (Finnish!!! - at least I could understand one or two words from the other languages), each one shown for what seemed the best part of a minute. Following that there was a long video featuring a pirate with a branding iron that must have been a couple more minutes long, followed by scrolling notices and warnings from FACT. When the film eventually started, the timer on the exercise bike was already showing 10 minutes and 30 seconds and I was feeling quite tired and very annoyed.

Re:Revocation is pointless

[Dunbal]

the primary use of the program will be to get unencrypted content onto P2P networks. Why bother revoking the key? I must be missing something.

      What you are missing is that the primary goal of copy protection has ALWAYS been to annoy the legitimate user, never the pirate. That copy protection stops piracy is but a smokescreen for the masses. Go on, user, do your bit to fight piracy and buy a new player/DVD with updated keys... wow, you helped change the world, don't you feel proud that you're a legitimate user and not a lowly pirate?

Re:Revocation is pointless

[dtfinch]

They thought of that.
http://en.wikipedia.org/wiki/ROM-Mark

I'm not sure if HD-DVD has a similar feature, or if this is Blu-ray only.

Re:Revocation is pointless

[asc99c]

I'd mod that interesting if I wasn't already posting on this thread :)

I can't find an awful lot of information on how it's implemented, but it seems designed to stop wholesale bypass of the DRM scheme (i.e. copy every bit of the disc from source to target, which gives you a copy with full DRM etc. without ever breaking the DRM). My guess would be that certain details of the volume key would tell the player to expect certain details in the ROM Mark.

I can't imagine this working on a disc with no DRM, simply because that would stop the disc being used for HD camcorders etc or just for archiving your own footage from other sources. Particularly with Blu-Ray debuting as a rewritable format, this is obviously a market they are after.

TPM is anti-virtualization

[tepples]

And at that point, virtualization kits will become commonplace that run Windows in a sandbox so that Windows thinks it's in a Palladium environment, but where it's really not.

The express purpose of "Trusted" Computing is to distinguish an OS running on bare hardware from a virtualized OS. The virtualized Trusted Platform Module is issued not from a recognized mainboard manufacturer's keyspace but from VMware's.

Re:TPM is anti-virtualization

[theelectron]

Could you not alter the TPM key the VMware server uses? Also, is this true for other virtualization, Xen etc., as well?

Re:TPM is anti-virtualization

[AuMatar]

Then people will just write open source virtualization kits, that fake it from whatever keyspace we want.

If we own the physical hardware, DRM is never going to work. Period.

Re:TPM is anti-virtualization

[CrystalFalcon]

And would you bet money on the impossibility of spoofing a specific motherboard identity?

Similar things have been done before in so many different scenarios... Just to take a trivial example, MAC addresses were supposed to be unique for each network card, too.

Re:TPM is anti-virtualization

[JesseMcDonald]

Sure, but the whole point is that you can't access the keys the "trusted" mainboard manufacturers encode into the hardware. You can program the emulator with any key you want, but it won't be one of the "trusted" keys. The keys are stored and used entirely within a single IC; the only way to extract one would be, in theory, to examine the IC directly (with an STM, for example), or somehow gain access to the master copy held by the manufacturer (and risk violating trade-secret laws).

IMHO this raises interesting legal issues, since it would tend to allow holders of one form of monopoly monopoly (copyright) to influence market shares in another industry (computer hardware). With TC the priviledged holders of media monopolies would be free to determine which hardware manufacturers succeed and which ones fail. Might not the RIAA/MPAA find themselves on the receiving end of an antitrust suit as a result of this cross-industry influence? (I don't support antitrust regulations myself, but I'm not the one they have to worry about.)

Re:TPM is anti-virtualization

[theelectron]

I'm not completely familiar with the TPMs, but would it be practical for me to 'guess and check' keys until I got something in a trusted namespace? How big are the keys?

Re:TPM is anti-virtualization

[tonigonenstein]

Sure, but the whole point is that you can't access the keys the "trusted" mainboard manufacturers encode into the hardware.

I am not familiar with the TPM spec, but wouldn't it be possible for the virtual TPM to relay requests to the real TPM ? Then you could access the software player's state by querying the VM.

Re:TPM is anti-virtualization

[Jherek Carnelian]

Sure, but the whole point is that you can't access the keys the "trusted" mainboard manufacturers encode into the hardware. You can program the emulator with any key you want, but it won't be one of the "trusted" keys. The keys are stored and used entirely within a single IC;

What is to stop a guy with a real TPM system and a virtual environment from just proxying any TPM requests/responses from the virtualized system to the real TPM module?

The TPM is like a black box right? Nobody can see inside it, all anyone can do - including a "trusted OS" is send it inputs and read the output. So, there should be no way for the virtualized OS to tell the difference between a proxyed TPM and a "directly connected" one.

So, now you've got a fully virtualized system that thinks it is running not-virtualized. Its memory, even its cpu registers, are ripe for harvesting supposedly protected information. If the system is going to depend on the TPM to do the actual decryption without exposing any keys, you still have easy access to the decrypted data that comes out of the TPM.

Re:TPM is anti-virtualization

[Rich0]

They would use SSL. Most likely 1024/2048 bit keys. You won't guess them.

And the manufacturer wouldn't know your key either. Most likely the chip will generate its own keypair, store it in flash, give the manufacturer a CSR, which would then be signed and returned to the chip as a certificate. At this point the only copy of the private key is in the chip - at best the vendor knows the public key, which is no good for bypassing TPM.

Now, what you could do is get the manufacturer's signing key and make your own certificates. That would certainly work. However, it hasn't really happened yet in the SSL world, and there is no reason to think that it will happen in the future - those keys would be kept under close guard.

Re:TPM is anti-virtualization

[Rich0]

Yes, but MAC addresses aren't kept secret.

The private key for your motherboard will be - it will never leave a single chip. Sure, if you have the hardware you can in theory obtain it, but this will require stuff like electron microscopes.

You can't make it impossible - but you can make it REALLY hard.

Re:TPM is anti-virtualization

[Mr2001]

The keys are stored and used entirely within a single IC; the only way to extract one would be, in theory, to examine the IC directly (with an STM, for example), or somehow gain access to the master copy held by the manufacturer (and risk violating trade-secret laws). And as long as you're risking violating the trade secret laws, why not go all the way? I'd love to see a few dozen guys with machine guns just break down the TCPA's front door and steal the damn keys the old-fashioned way.

Re:TPM is anti-virtualization

[tepples]

I am not familiar with the TPM spec, but wouldn't it be possible for the virtual TPM to relay requests to the real TPM ? Then you could access the software player's state by querying the VM.

No. The TPM hardware is designed to watch the boot process. It would report the presence of the virtual machine, signed using the private key hidden inside the TPM hardware.

Re:TPM is anti-virtualization

[tepples]

Then people will just write open source virtualization kits, that fake it from whatever keyspace we want.

But we do not know the 2,048 bit private keys of the mainboard manufacturers, and faking it from another keyspace is easy to detect because the keys don't match.

Re:TPM is anti-virtualization

[tepples]

The TPM is like a black box right? Nobody can see inside it, all anyone can do - including a "trusted OS" is send it inputs and read the output.

The inputs include what kernel was loaded. If the kernel loaded by the mainboard does not match the kernel loaded by the VM, the publisher's software will be able to detect this situation.

Re:TPM is anti-virtualization

[cheater512]

Shh! Your giving them ideas!
Next we'll be 'licencing' our computers instead of owning them.

Re:TPM is anti-virtualization

[paeanblack]

The private key for your motherboard will be - it will never leave a single chip. Sure, if you have the hardware you can in theory obtain it, but this will require stuff like electron microscopes.

How do you account for this hole:

1) Asus' servers get "hacked".
2) The keys to all Asus motherboards get posted on the web
3) Sales of Asus motherboards skyrocket.
4) Asus issues a press release to the effect of: "It was the fault of those damn dirty hackers. We have no idea how this happened. Excuse us; we must return to sifting through this mountain of cash".

The hardware manufacturers have no incentive to play nice with the Trusted Computing scheme. This is just a repeat of DVD Region Coding. The manufacturers just started producing players that ignore the region code, because they outsold the locked players. Of course the first few on the market were "accidents", "mistakes", and "test designs".

In a Trusted Computing world, machines with a broken TC implementation will be cheaper to make and command a higher price in stores. What do you think will prevail?

Re:TPM is anti-virtualization

[gutnor]

Yes that may be possible if your VM application is a Trusted Application itself.

Before anybody else ask, as you may imagine, they have gone far enough with TPM so that making a trusted application is a bit more difficult than switching a property in the config file.

There is no *obvious* flaw in TPM ( there may be flaws, but nothing that you could think in 5 sec: like running a VM, changing the registry, blinking your eyes real fast ).

Actually in theory, TPM is really cool and could mean the end of viruses, trojan, phishing, ...
Unfortunalty that's not often the good guys that have the big guns.

Re:TPM is anti-virtualization

[John Hasler]

1) Asus' servers get "hacked".
2) The keys to all Asus motherboards get posted on the web
3) Those keys are revoked.

Re:TPM is anti-virtualization

[Jherek Carnelian]

The inputs include what kernel was loaded. If the kernel loaded by the mainboard does not match the kernel loaded by the VM, the publisher's software will be able to detect this situation.

I don't see how that makes a difference - in a VM the kernel is not modified, that's kind of the whole point of a VM, you do not need the cooperation of the hosted OS.

Re:TPM is anti-virtualization

[Anonymous Coward]

But we do not know the 2,048 bit private keys of the mainboard manufacturers

That's a dangerous assumption.

Re:TPM is anti-virtualization

[Chyeld]

Sure, but the whole point is that you can't access the keys the "trusted" mainboard manufacturers encode into the hardware. You can program the emulator with any key you want, but it won't be one of the "trusted" keys. The keys are stored and used entirely within a single IC; the only way to extract one would be, in theory, to examine the IC directly (with an STM, for example), or somehow gain access to the master copy held by the manufacturer (and risk violating trade-secret laws).

You forget the third, possibly not completely possible right now, but certainly concievable in the near future, option of obtaining the key. Brute force.

It wasn't that long ago (in the timeframe of video formats) that RC5-56 was considered 'secure' enough. It might not be around the corner, but there is certainly the possibility that CPU power could continue to ramp up quickly enough that the keys themselves can be brute forced through a botnet version of distributed.net. And once that cat is out of the bag, it'll be out forever.

Re:TPM is anti-virtualization

[nuzak]

The kernel in the VM has different state than the one loaded by the mainboard, including such things as being loaded at a different physical address. For this purpose, that makes it different.

Re:TPM is anti-virtualization

[Cheesey]

>> But we do not know the 2,048 bit private keys of the mainboard manufacturers
> That's a dangerous assumption.

What about the private key for signing XBox 360 games? Or the one for Playstation games? Or the ones used by Verisign to validate SSL certificates? Or Microsoft's Windows Update key? These things have always remained secret - it's unlikely that they'll ever be public. It's not hard to keep them secret as they never need to be distributed.

That said, there is a private key locked inside each TCPA module. If you can get at that, then the TCPA system breaks down - you can fake trust as required.

Re:TPM is anti-virtualization

[nuzak]

> If we own the physical hardware, DRM is never going to work. Period.

What on earth makes you think you really "own" the physical hardware now? That's what the DRM endgame is all about: closing off the hardware to you as well.

Re:TPM is anti-virtualization

[Cheesey]

The TPM is built into the CPU, so a man-in-the-middle attack like the one you suggest is very difficult. It has access to all sorts of CPU data that you can't really fake.

Although perhaps you could fake that information by loading new microcode into the CPU...

Re:TPM is anti-virtualization

[Anonymous Coward]

4) All of the DRM-related stuff on PCs with Asus motherboards stops working.
5) At best, Asus' customers are locked out of all of their DRMed content, such as movies and music that they played for. They may also be locked out of documents and files they created, or even locked out of all of their data (if they use BitLocker, or similar).
6) Asus' customers file a class-action lawsuit for some totally insane sum.

Re:TPM is anti-virtualization

[cibyr]

Couldn't you man-in-the middle it with a virtualisation layer that passes the decryption requests from the official software to the TPM hardware and then grabs the "plaintext" (in this case video) on the way back?

If it hasn't been said enough yet, this is why DRM can't ever work.

Re:TPM is anti-virtualization

[Jherek Carnelian]

The kernel in the VM has different state than the one loaded by the mainboard, including such things as being loaded at a different physical address. For this purpose, that makes it different.

I still don't see how that matters - for two reasons -

1) All those things you list can easily be handled by a VM designed to do so - the virtualized kernel doesn't know its actual physical address, it only knows the one in the VM, same thing for any other sort of state.

2) Who says the TPM has to be in a machine running a "trusted" OS in the first place, boot it non-trusted and just feed the TPM the appropriate inputs to make it think the system is booting. Failing that, run the TPM by itself - yank it off the motherboard and host it on your own PCB with an FPGA programmed to provide a full "VM" for the TPM itself.

Re:TPM is anti-virtualization

[Jherek Carnelian]

The TPM is built into the CPU

Pretty sure that's not true. Feel free to identify such a CPU with specs to prove it.

Re:TPM is anti-virtualization

[Anonymous Coward]

7) PROFIT!

Re:TPM is anti-virtualization

[Cheesey]

Well, there's the PPC chip in the XBox 360, for one. That's a full TCPA system.

Please bear in mind that I'm only arguing this point because I think it's important that people are well informed about what we're up against here. It's not going to be easy to get around TCPA, really it isn't. Virtualisation and man-in-the-middle attacks are exactly what TCPA is intended to prevent, and it's been designed by people who understood what sort of work would need to be done to enforce DRM as required by the entertainment industry.

However, citations. Anderson says that current (2003) TCPA chips are on the motherboard, not the CPU, but:

However, in a few years, the Fritz chip may disappear inside the main processor - let's call it the `Hexium' - and things will get a lot harder. Really serious, well funded opponents will still be able to crack it. But it's likely to go on getting more difficult and expensive.

He also notes that some portions of TCPA are already in your CPU:

The operating system security kernel (the `Nexus') bridges the gap between the Fritz chip and the application security components (the `NCAs')... Finally, the Nexus works together with new `curtained memory' features in the CPU to stop any TC app from reading or writing another TC app's data. These new features are called `Lagrande Technology' (LT) for the Intel CPUs and `TrustZone' for the ARM.

With the chip on your motherboard, yes, you can do a MITM attack on the bus lines. That and cost saving is exactly why it'll be part of your CPU, if it isn't already.

Re:TPM is anti-virtualization

[Dunbal]

Asus' customers file a class-action lawsuit for some totally insane sum.


      What was that number again? Oh yeah! 11 trillion dollars sounds about right... Glad I have an asus motherboard ;)

Re:TPM is anti-virtualization

[Dunbal]

Next we'll be 'licencing' our computers instead of owning them.


      Unless you write your own software, you already do. At least according to everyone else, who feel they have the right to a) install "extra" features that phone home whenever they want b) fail to uninstall 'nagware' when you remove a program c) trash your registry.

        Surely a computer is not just the pile of sillicon, glass, metal and plastic - but rather the software you use. What can you honestly run on your computer now without clicking "I agree" on an EULA? Your computer was pwned a long time ago...

Re:TPM is anti-virtualization

[Jherek Carnelian]

Well, there's the PPC chip in the XBox 360, for one. That's a full TCPA system.

Pretty sure that's wrong. Infineon makes the TPM chip for the xbox 360.
One of many mentions of the chip - no part number, but then I only spent 30 seconds googling for confirmation.

He also notes that some portions of TCPA are already in your CPU

While necessary to implement TCPA, those are not at all functions of the TPM.

That and cost saving is exactly why it'll be part of your CPU, if it isn't already.

Sounds like the party line. As long as the current TPM modules are supported, it won't matter if others eventually are integrated into CPUs - somebody could put a "TPM server" on the net and support thousands of users - maybe not all simultaneously, but the need for such a server would not be constant - only long enough to extract the protected information from the virtualized "trusted system."

Re:TPM is anti-virtualization

[Cheesey]

One thing is for certain - this is going to take quite a bit of hacking.

I do think there are ways to attack this, but I personally reckon that the main chance lies in extracting the private key from the TC module. An easier approach would be welcome.

Re:TPM is anti-virtualization

[The Warlock]

Doesn't matter. If a piracy group cracks one key, they can turn any movies into an unencrypted format, and then that's it. Once that one copy has been FXPed and BitTorrented and etc., it's over; there's no putting that cat back in the bag.

Re:TPM is anti-virtualization

[Anonymous Coward]

doesn't that mean that using windows virtualized within windows (same kernel) would compromise the scheme?

Re:TPM is anti-virtualization

[thisissilly]

Actually, MAC addresses had to be software settable, in order to be able to run DECnet. That's because DECnet didn't have ARP. Instead, MAC addresses are computable from DECnet node addresses (and I'm showing my age by even posting this).

Re:TPM is anti-virtualization

[tepples]

What can you honestly run on your computer now without clicking "I agree" on an EULA?

Free software, such as LOCKJAW.

Re:TPM is anti-virtualization

[tepples]

doesn't that mean that using windows virtualized within windows (same kernel) would compromise the scheme?

The Windows kernel is loaded twice, once by the bootloader and once by "an application". TPM detects and reports this.

Re:TPM is anti-virtualization

[tepples]

Couldn't you man-in-the middle it with a virtualisation layer

And watch the TPM dutifully record the presence of libmitm.

Re:TPM is anti-virtualization

[pyrrhonist]

Free software, such as LOCKJAW.

Wait a minute - NO EULA? HOLY FSCK! You just removed the GPL license restrictions and placed your software under the public domain! You read it here first, folks. LOCKJAW is now PUBLIC DOMAIN SOFTWARE. SWEET!!!!

What?!? Oh. You mean without literally click on, "I agree". I see.

Nevermind.

Re:TPM is anti-virtualization

[ink]

Right, but what are the odds on all of the software that is signed with those public keys being bullet-proof? It's so complicated, that (maybe) they could get it right after four or five generations. There are also attack vectors in all of the trusted computing peripherals, most notably the screen.

I highly suspect that it'll be a waste of the content producers' time; and that it will be great fun for hackers. DeCSSv2.

Re:TPM is anti-virtualization

[Anonymous Coward]

Sorry, no.

If they use 128 bit keys there will never (in our life time) be enough processing power to brute force through all possibilities. 2^128 is Just Too Big - do the math.
If each core could somehow magically try 10^9 (giga) keys per second it would mean about 10^22 core years, i.e. 10^22 cores for a year or one core for that many years.

Re:TPM is anti-virtualization

[Albert Sandberg]

I find your lack of faith disturbing

Take a look into quantum computing.

Re:TPM is anti-virtualization

[zcat_NZ]

The Free Software Foundation's position (I mailed them and asked) is that once you have a copy of the software you OWN it, the way you own a book or a TV or a car, copyright (and thus the GPL) only applies when you make a copy to give to someone else.

I don't have to agree to the GPL to use GPL software.

Re:TPM is anti-virtualization

[Anonymous Coward]

I'd love to see a few dozen guys with machine guns just break down the TCPA's front door and steal the damn keys the old-fashioned way.

The Canadians already thought about that.

Re:TPM is anti-virtualization

[Rich0]

Right, but what are the odds on all of the software that is signed with those public keys being bullet-proof?

No software will be signed with any of those keys. The certificate only certifies that the chip implementing TPM is genuine.

The logic is that on bootup the TPM chip will hash the BIOS and store this has, and will provide a signed attestation upon request that this BIOS was booted.

The BIOS will then hash the OS that it boots and provide its hash upon request. The OS will do the same for a piece of running software.

A remote website will ask a piece of software for a chain of trust. The software will ask the OS for its hash, and the OS will ask the BIOS for its hash, and the BIOS will ask the TPM chip for its hash. All of these signed hashes will get sent to the remote website. The remote website will check all the hashes and decide whether to provide the software with a decryption key.

If the software is found to have a vulnerability it could be revoked at the server level. Obviously this will be a pain for anybody who owns that software, but TC isn't designed to make user's lives easy.

I agree that there are a bunch of issues with TC, but it will make extracting protected content a real pain. It might also make it harder for you to open your documents in open-source software. While you could always download an unprotected torrent of the latest movie release, you won't be able to find an unlocked torrent for the spreadsheet you created in MS Excel the other day.

My feeling is that we need legislation requiring the disclosure to computer buyers of ALL keys stored within them, and any related-keys that are needed to access features on those computers (such as any signing keys needed to flash the BIOS). And by disclosure I mean the keys themselves - not just the fact that they're there. Computer owners could use TC to secure their computers against hackers/viruses/etc, but 3rd parties couldn't use TC to secure computers against their legal owners.

Re:TPM is anti-virtualization

[ink]

The software will ask the OS for its hash, and the OS will ask the BIOS for its hash, and the BIOS will ask the TPM chip for its hash. All of these signed hashes will get sent to the remote website. The remote website will check all the hashes and decide whether to provide the software with a decryption key.

They will be revoking software ALL the time then. Any software that relies on this system will also stop functioning after so many days (also don't forget about company firewalls with only proxy-access, which would trigger a MIM alert for the remote website). I don't think consumers will put up with the fallout that will be ultimately caused by buggy software that has been signed and revoked by the same folks.

And congressmen are also consumers. The first time they can't open an Excel document because it wasn't signed.. well, you can guess what will happen on the political front.

It's doomed to fail; only the NSA and other anal information agencies will use it fully.

Re:TPM is anti-virtualization

[asuffield]

The hardware manufacturers have no incentive to play nice with the Trusted Computing scheme. This is just a repeat of DVD Region Coding. The manufacturers just started producing players that ignore the region code, because they outsold the locked players. Of course the first few on the market were "accidents", "mistakes", and "test designs".

It's a little more subtle than that.

In the first round, all the "major" manufacturers produce compliant devices (modulo bugs), which are locked down.

Then the Asian bootleggers get in on the business. Their friends in the Asian device production plants that make all these motherboards slip them copies of the current keys. Mod-chips and entire motherboards start appearing on the grey market, on the streets of Hong Kong and Seoul. Not to be outdone, Japanese importers start grabbing up these devices and they appear in the back-street stores in Akihabara.

Slow to catch on, the TCPA consortium revokes the offending keys, and the major motherboard producers are forced, at great expense, to recall all the previously sold boards and offer free replacements to anybody who wants their copy of Vista to keep working (it's impossible to securely issue a software update for this problem - the update would be equally applicable to the bootlegged devices, since there's no way to authenticate the 'genuine' ones when they're all using the same keys).

The morning after the keys are revoked, the keys for the new devices are available on the internet (because those production plants are still run by the same people, who really don't give a damn about the demands of the American corporations). This pattern continues for a couple of months, while the corporations shuffle their staff in the production facilities - and discover that there isn't anybody they can hire in those countries who is going to run the operation securely enough to matter. Frantic board meetings are held.

Meanwhile, alerted by media reports of the product recalls, western importers start getting hold of the bootleg devices. They begin to appear for sale in the US and Europe, via ebay and dedicated sites. The TCPA consortium flails about a bit, a bunch of stuff on ebay gets delisted, but there are too many importers and not enough time to sue them all.

The board meetings of most of the major motherboard manufacturers come to this conclusion: "TCPA is costing us money from having to change the keys all the time, there's no way that us *and all our competitors* are going to be able to secure all our production facilities any time soon - and worst of all, we're losing sales to this bootlegged hardware, because our customers want to download videos from thepiratebay. Screw this. We're going to start selling a product that people want to buy."

The second round of motherboards are rather less secure. Much like DVD region coding, the boards look like they do what they're supposed to at first glance, but actually there are ways to persuade the chips to give up their keys, or just sign anything you hand them. These are initially blamed on "test designs", etc. Not every manufacturer will do it at first - but those that don't will take a heavy hit in the market. Do not underestimate the desire of Americans for free porn and free violent movies.

TCPA is now dead.

This is basically what happened to DVD region coding - the major western production houses, faced with decss/dvdcss on the one hand and eastern import hardware eating into their sales on the other hand, quickly realised that siding with the DVD consortium was ultimately going to lose them a lot of money. The only way that TCPA could avoid this is if somehow every single approved motherboard manufacturer could manage to make their security watertight - and that just is not going to happen.

Of course, non-Vista platforms will be buried in a legal quagmire for years, as we have been with libdvdcss - it's not strictly legal, maybe, but it's the only way we'll ever have. This is perhaps the objective of the entire TCPA concept.

Re:TPM is anti-virtualization

[kernelistic]

That is incorrect. The most commonly used TPM chip is the Infineon one, and it sits squarely on your motherboard. Have a look at http://en.wikipedia.org/wiki/Trusted_Platform_Modu le and http://www.infineon.com/tpm/ .

Re:TPM is anti-virtualization

[Rich0]

They will be revoking software ALL the time then.

Just like they do already. Ever try to use an older version of Itunes?

Re:TPM is anti-virtualization

[r3m0t]

I was impressed by your analysis at first, but then I realised:

"Their friends in the Asian device production plants that make all these motherboards slip them copies of the current keys."

Each (individual) motherboard gets a different key.

Re:TPM is anti-virtualization

[Rich0]

It's doomed to fail; only the NSA and other anal information agencies will use it fully.

Oh, on a side note - I'd use it. I love the idea of hardware-enforced security. If I could upload my own certificates into the TPM modules I could implement a number of features:

1. Run a rootkit-hunter that could reliably detect rootkits running in memory.
2. I could keep my ssl keys / etc locked up tight - no need for boot-up passwords, and yet no worries about people getting them off the drive.
3. I could use remote attestation to detect viruses/etc running on other PCs in my house. Maybe somebody in the house managed to install some malware (even without admin privs - local escalation hole or something like that) - I could cut that off at the firewall level.
4. If I were a business I could ensure computers browsing the internet had full patches.

All of this stuff is a good reason for trusted computing. As the owner of the computer and the holder of the keys I'm in control. The problem comes when I'm not the holder of the keys, and somebody else is doing the attestation.

I'd use TC in a second - but it needs to serve me - not some 4-letter association.

Re:TPM is anti-virtualization

[asuffield]

Right, and that's an essential part of the problem (I should have gone into it, but the post was long enough already). Each "official" motherboard gets a different key.

The thing which is stolen from the production plant is the set of 50,000 keys for the current batch of motherboards (even if it's not supposed to be retained, the production software/equipment will be quietly modified so that it does get retained).

You have to realise: there is absolutely no reason why every bootleg device needs to have a different key. So in fact, all the ones on the street have the *same* key, and the bootleggers make one new key available for download every day. The manufacturer is now faced with a problem: kill one key a day and have no effect, kill them all and face a product recall, or forget the whole thing.

Re:TPM is anti-virtualization

[ink]

Amen. I'd use it in that case as well.

Looks the same

[Anonymous Coward]

This is the hole security expects have been telling them about since they started using cryptography in DRM.
Currently all encrypted content and keys must exist in unencrypted from at some point in memory.
Also, as long as the contents of memory are viewable this hole will exist. The strength of the
crypto doesn't matter if the user can get the key. Currently scanning the contents of memory and
trying each group bits against something known, like an encrypted MPEG headers is reasonably quick and effective.

Re:Looks the same

[Anonymous Coward]

Precisely why lots of new PCs (Intel Apple Macs, for example)... come with TPMs. To make sure you, as the owner of the machine, can't see what a "trusted app" is doing. The fact that they also gives the likes of Gates, Jobs and whichever suit is running IBM at the moment, access to undreamed of levels of control of the customer is also a benefit.

Re:Looks the same

[Dunbal]

TPMs. To make sure you, as the owner of the machine, can't see what a "trusted app" is doing.

      I'm positive someone will find a way around THAT, too. Even if it means applying a soldering iron to a motherboard. Some people are very creative. And the fun part is, you only ever have to hack it ONCE, and the internet does the rest...

Re:Looks the same

[Anonymous Coward]

and when the TPM is inside the processor... as it will be when newer Intel and AMD processors ship?

Re:Looks the same

[Dunbal]

and when the TPM is inside the processor...

      I've been left behind by technology a long time ago, but I think a processor still relies on the information it receives to make the call. For the processor to decide if there's a violation or not, there have to be other variables involved - be it in the BIOS - which can be reflashed, or the hard drives/boot sectors,
etc.

I refuse to believe in the unhackable computer. Remember the old saying - "garbage in, garbage out". If the CPU is fed the right info, it will act the way you want it.

Re:Looks the same

[Anonymous Coward]

I've been left behind by technology a long time ago, but I think a processor still relies on the information it receives to make the call.

The keys/encryption engine are all in the TPM... nothing sensitive and unencrypted gets outside the processor. Read this. Essentially, the environment the TPM starts in is hashed, and used with the TPM's secure key to verify that everything (software/hardware) is 'trustworthy' (i.e. approved by central authority). It wasn't designed by idiots, but rather by paranoid control freaks who imagine that what the world needs is a "fritz" chip in every computing device that decides which software and hardwares get approved for use.

Re:Looks the same

[Anonymous Coward]

That should have said: Read this

Hmmm i swear thats not the way i read it....

[jkerman]

"to identify the players responsible for the breach and prevent later pressings of discs from playing back on those players until they are updated."

wasnt this attack based on being able to extract the title-key from the disc, then run it through stock AACS decryption libraries? they could revoke whatever keys they wanted, but wouldnt the existing un-retractably released software still have to read the key (making it visible, unencrypted, in ram....) before it could deny playing it?

The way i understand it the "player" gets the title key, normal AACS libraries are used to decrypt into a format (no-drm) that any player will play back. it has no key to block!

Re:Hmmm i swear thats not the way i read it....

[Anonymous Coward]

wasnt this attack based on being able to extract the title-key from the disc, then run it through stock AACS decryption libraries?

It was, but the title keys are encrypted with the disk key. There are lots of copies of the disk key in a single file on the disk, all encrypted with one of the many player keys. If a player key is revoked, there will simply be no copy of the disk key that this player will be able to decrypt on any future disks.

And Don't Forget..

[JohnnyOpcode]

Just because some 'software player' has been compromised and will get revoked, there are a few hardware models out there that also have 'weak' firmware and a JTAG port to jack into. I just wish this hacking business would get me some hotter lookin' chicks like in the movies..the ones we can decrypt now..and forever!

I hail our new hacker overloads!

Something they seem to be ignoring

[Iphtashu Fitz]

All the focus, and for good reasons, has been on software-based DVD players. They're easy for any hacker to play around with. However there are plenty of people out there who happen to be hardware hackers as well. I wonder how long (probably just a matter of time) before some hardware/firmware hacker disects a standalone HD player and is able to extract keys from that. Hardware hacking hasn't been as glamourous as software hacking in recent years, but a mere 20 years ago it was all about hardware hacking. Read a book like the Cuckoos Egg - a sysadmin physically tapped into communication lines and directed the output to line printers so that a hacker he'd been hunting wouldn't know he was being tracked. I'd be willing to bet that some hardware/firmware gurus with the right tools would be able to hack a standalone HD player if they had the desire to do it. And if they can pull that off it'd be a LOT harder for the AACS LA to plug that hole.

I would like to suggest a better solution

[Anonymous Coward]

How about adapting to the market, recognizing that Fair Use is allowed by Copyright Law, and arriving at the logical conclusion:

Given that Fair Use is a right protected by law (it's NOT a defense as some love to claim) in exchange for a limited monopoly on distribution of the work in entirety, make the product more attractive to paying customers. Don't try to make money on both ends; recognize that timeshifting and formatshifting and copying of excerpts for use in reviews, parodies, etc. are all allowed in exchange for your limited monopoly, and improve your packaging, product quality, and pricing. Also: recognize that you are selling a commodity product identical in every way to a book; first sale doctrine applies; it is NOT licensed. If you do not offer a refund after the transaction is made PRIOR TO acceptance of any such license, then it is clear that it is not licensed, but a commodity good sold as is, and as such, the purchaser can do with it as he or she damn well pleases after the point of sale. Otherwise, offer refunds and back your product with a warranty.

Also, when a DVD/Blu-Ray Disc/HD-DVD delaminates or becomes too scratched to use or breaks, happily replace it for the customer, or at least provide a mechanism for making a 1-for-1 backup, as provided for under Fair Use and many court precedents.

Also, you should also quit conveniently selectively reading and interpreting the DMCA: recall the exclusions allowing for interoperability, and that cracking it for exercising Fair Use such as format shifting and viewing on alternate platforms and devices is explicitly allowed even under the draconian DMCA.

I reiterate: The Best solution is to drop DRM and increase value for legitimate paying customers. Give them a reason to WANT to buy your product.

I usually buy at least five DVDs per month, usually closer to 15 (my DVD collection is rapidly approaching 500). This month I bought "only" 3 (well, seven, if you consider that the extended edition of The Lion, The Witch, and The Wardrobe is four DVDs, and another movie I bought was a two-DVD collectors edition) because I incurred large expenses (a vacation, new PDA and other electronic equipment, and promptly ripped 5 DVDs for viewing on the PDA while traveling - interoperability format shifting provided for by DMCA exclusions and Fair Use I might add).

I occasionally download in a "try before I buy" scenario. I purchased THX1138 and T3 and many other movies after I decided they were worth buying (honestly, there hasn't been anything I downloaded that I didn't end up deciding to buy, except for material which falls into public domain). Some I won't pay more than $7 for and look for used DVDs, but I buy most DVDs new.

I was holding out on HD-DVD and Blu-Ray pending cracks to enable Fair Use and Linux interoperability, but now I am going to wait longer to see how it pans out. Will they actually revoke the keys? Will downsampling be turned on? If so, then I'll stick with DVD's 720x480 resolution, and use Blu-Ray solely as a read-write medium for backing up data, and buy exactly zero high-definition discs. It is not as though the increased resolution improves poor writing, and makes a bad movie worth watching (although in some cases, such as Plan 9 from Outer Space, or Starship Troopers, it is possible that a movie is so horribly bad that it enters the must-own/must-see category). Futurama or Stargate (ignoring the seasons with the Ori) or Seinfeld is equally entertaining to watch at 320x240 (roughly VHS resolution), 720x480 (DVD), or 1920x1080p (high def's top resolution).

Posting anon for the benefit of litigious MPAA assholes. MPAA: FUCK YOU.

Re:I would like to suggest a better solution

[Anonymous Coward]

promptly ripped 5 DVDs for viewing on the PDA while traveling - interoperability format shifting provided for by DMCA exclusions and Fair Use I might add

Really? I bet when you ripped those DVDs you did so using some variety/implementation of DeCSS. I also assume that you possessed a copy of said implementation. Where'd you get it? Did you download it from an offshore or overseas website? Did you get it from someone else distributing it? If so, you may be guilty of 'importing' or 'trafficking in' a circumvention device. Under the DMCA, you're in a gray area of the law - although you are using the device for fair use (or so you assert, your assertion has not been vetted in court) the device itself is certainly used for far more than that, and as such you might still be liable.

Selective keying using the whole .exe from memory.

[russ1337]

They talk about this on Security Now, Episode #76 (http://www.grc.com/securitynow.htm)

It seems muslix64 just had a snapshot of the entire .exe running in memory, then used selective keying - serially trying bytes 1-4, then 2-5, 3-6 etc as the keys until the mpeg frame decrypted. (which, of course this is much faster than a pure brute force attack, and took only seconds).

So as long as a software player has the key in the clear and is loaded in memory 'somewhere', this type of attack will continue to work.

AACS is still 'unbroken' but like many failed encryption schemes, it was circumvented due to poor implementation.

Re:Another blow struck for free entertainment

[HomelessInLaJolla]

You would make sense if a money map of the industry didn't show that the vast majority of the profit goes to CxOs, VPs, board directors, and career stock investors who have little or no real interest in the actual entertainment content.

When you can separate honest entertainment interest from pure and erated business interest then you may pull your head from your backside.

Individual players are revoked, not classes

[Wesley Felter]

If one hacker's player gets revoked, it won't affect regular users at all. And the hacker will probably just buy another one.

Re:Individual players are revoked, not classes

[Rich0]

How many keys does the system support? They'd need an awful lot of them if they're going to encode every disc with one unique key per copy of software / hardware sold.

A billion keys isn't all that unrealistic a number. Sure, I guess it could be done, but it sure is an interesting approach.

Plus, in this particular case they wouldn't know the key for the piece of hacked hardware - they didn't disclose the software key - only the media keys.

Re:Individual players are revoked, not classes

[Joe5678]

They don't generate unique decryption keys for each and every player. Individual players are NOT revoked. It is at least a class of players (in this case we're only dealing with software players as a sibling post has pointed out) and I wouldn't be surprised if they keys were only unique on the manufacturer level (i.e. Sony has one key for all the players it makes).

Even if each player did have a unique decryption key though, they would have no way of knowing which key to revoke. This is the reason the person who decrypted the data hasn't said "I used player X" since he doesn't want to make it easy for them to revoke the key for "player X".

Re:Individual players are revoked, not classes

[Wesley Felter]

Yes, AACS supports billions of keys.

Obviously AACS LA will have to know which players to revoke, but I suspect hackers will start leaking player keys soon, since they're more useful than title keys.

Re:Individual players are revoked, not classes

[Wesley Felter]

They don't generate unique decryption keys for each and every player.

My interpretation of the spec is that every individual player has unique keys. Software players may be a little more relaxed, though.

It is at least a class of players (in this case we're only dealing with software players as a sibling post has pointed out) and I wouldn't be surprised if they keys were only unique on the manufacturer level (i.e. Sony has one key for all the players it makes).

The software players may have one set of keys for each app, but then it's easy to update software to change out keys. After several such updates, the software player developers may end up implementing individual keys using online activation to reduce the hassle. Either way, the collateral damage would appear to be minimal.

Even if each player did have a unique decryption key though, they would have no way of knowing which key to revoke.

IMO it's only a matter of time until someone releases a DeCSS-style crack with player keys included. ...the person who decrypted the data hasn't said "I used player X" since he doesn't want to make it easy for them to revoke the key for "player X".

Plenty of people on the Doom9 forums have admitted to using WinDVD 8 Japanse edition. Besides, there are only two software HD-DVD/Blu-ray players anyway, so AACS LA could just revoke both.

Re:Individual players are revoked, not classes

[Jeff DeMaagd]

A paper describing the hardware revocation used to be findable for keywords "AACS player revocation". You don't brick entire brands, just single models, and it's easily possible to do that with the new method. The graphical description looks kind of like a binary search tree.

Re:Individual players are revoked, not classes

[Jeff DeMaagd]

You don't brick entire brands, just single models, and it's easily possible to do that with the new method

I meant to say something like this:

"You don't brick entire brands or even just single models, you brick single players. It's easily possible to do that efficiently with the new method."

Re:Individual players are revoked, not classes

[cheater512]

Each model of hardware player has the same key. They dont all have unique keys.

If I break one hardware dvd player and they decide to revoke its key then all of the players break which are the same model.

Re:Individual players are revoked, not classes

[Talchas]

And if an entire brand or class of players has an easily to extract key, they'll have to brick the entire group or people will just grab keys off each player.

Re:Individual players are revoked, not classes

[RespekMyAthorati]

I've read the spec and I think you are wrong.
Every player has what amounts to a serial number in rom, and a corresponing key.
The whole idea was to prevent someone buying a player, removing the AACS chip, and cloning it over and over without paying royalties.
By revoking this particular chip, all the fake players using it would be revoked. Anyone who bargain-hunted by buying one of them would be screwed.

Re:Individual players are revoked, not classes

[r3m0t]

"I've read the spec and I think you are wrong.
Every player has what amounts to a serial number in rom, and a corresponing key. "

Yet again... the spec doesn't reflect current practice - for software players.

I don't know whether all hardware players (of a certain model) share the player key (because it's difficult to see on the outside) but all the discs/downloads of a certain version of WinDVD and PowerDVD share the same player key.

One day, maybe AACS will move to individual keys for every player. Until then, they will face difficulties revoking player keys.

I need to buy, rip, and store the content

[sdo1]

Open letter to the MPAA: I hope a true "CSS" style hack is found. Otherwise, I'm remaining on the sidelines and I won't be buying any HD-DVD or Blu-Ray discs.

Hear that, MPAA!?!?! I said BUYING. You claim piracy costs sales, but you MUST then subtract the lost sales due to your overbearing copy protection. I have about 2000 CDs and about 600 DVDs in my collection. I have no HD-DVD or Blu-Ray discs. And I don't plan on it either unless things change.

It's a new world. And in this new world, I have an expectation of device portability. That means when I buy a 5" media-containing silver platter, I expect to be able to store it on a server in my house to stream it to my living room or my computer or my bedroom. I expect to be able to re-compress it for my laptop or my ipod (or -like device) for watching when traveling. I have no desire to be tied to a specific (and expensive) playback device in a specific location. You're terrified of future storage capacity that will reach into the terrabytes on small devices, but to me, that's the thing that's keeping me interested at the moment in the stuff you have to sell... the knowledge that I can have that portability in movies and TV the same way I have it for the music that I've collected over the years. The RIAA freaked out when MP3's came along, but to be honest, my interest in music had waned significantly. But now, with so much available at my fingertips, I'm VERY interested in hearing new things and I'm buying probably more than ever before (though none through the DRM-crippled iTunes store).

I will gladly buy the media, but I expect that at that point, our relationship is OVER. Thanks, goodbye. Now if I want to extract images from the movie, print them out, and wall-paper my room with them, that's MY business, not yours.

-S

Re:I need to buy, rip, and store the content

[suv4x4]

Hear that, MPAA!?!?! I said BUYING.

I think MPAA just pissed its pants.

Re:I need to buy, rip, and store the content

[guzzirider]

Funny thing about DRM schemes,

Now not like this new information but historically when some kind of a secret code was used to keep a secret message secret the intent of the code was to prevent the 'un-trusted' from reading it. When the message is / if delivered the trusted party they would have the means to decipher it.

Now 'secret messages' are being used where the content is delivered to the 'un-trusted'. ..

If it was not tragic it would be pretty damn funny that now we have terms like 'trusted computing' WTFO ? I mean now I am supposed to buy something from a manufacture, by laying down my hard earned green, for a product where I the customer is not to be trusted. ?? What next, ya' buy a house and don't get the keys ??

I have been called a lot of things, maybe some of them deserving but 'un-trusted'?

I guess the only power I have here (which is minuscule) is to not lay down my green. I can only hope others will join me.

Re:I need to buy, rip, and store the content

[ClamIAm]

when I buy a 5" media-containing silver platter

They hand you the keys on a silver platter...

Re:I need to buy, rip, and store the content

[Dunbal]

You claim piracy costs sales, but you MUST then subtract the lost sales due to your overbearing copy protection.

      Sheesh, between all those losses from piracy, and now all the losses from people like you - no wonder Hollywood is completely bust and unable to turn a profit on a film despite hundreds of millions per week in gross... /sarcasm

Re:I need to buy, rip, and store the content

[fraudrogic]

dammit, I had mod points yesterday. Wish I had them for your post... This is the key. I am INTERESTED in the content, but that's it. If I have to jump through hoops to get it in a format of my choosing, then I'll find other ways to watch it. There are a TON of ways to get the content I want. To be honest I can do without the cable TV, because if I really try, I can find everything I want via the web. If I can't, then well, there is a ton of other content that will grab my short attention span. I love "The Office". I love talking about it with my coworkers and recreating the funny stuff in our own office (someone put my freakin' stapler in a jello mold for god's sake!). But if I lost bit torrent, cable, and they DRM'ed the shit out of the media they sold it on, I really could do without it. This super inflated sense of "I gotta have it" is created by the *iaa's. We don't NEED it. It's fun, but I don't "demand" it. In other words. Fuck DRM. I don't want there shit SO BAD that I would subject myself to the hoops they would like me to jump through.

The mere fact

[diff2uni]

That AACS protection can be easily cracked because some player manufacturer did not make the player right goes to show just how ridiculous the whole proposition is in the first place.

Re:Another blow struck for free entertainment

[drinkypoo]

It warms my heart to know that there are people out there watching out for my right to free entertainment. If it weren't for them, the people who invested in, assisted with, created, and distributed my entertainment would be getting their greedy little mitts on my money. Now I don't have to worry about that happening, and I can have the massive entertainment collection I deserve.

It warms my heart to know that there are people out there watching out for my fair use rights. If it weren't for them, the people who (blah blah blah) my entertainment would be able to prevent me from taking actions which are supposedly explicitly protected by law, based on legislation which they bought and paid for. Now I don't have to worry about that happening, and I can do the things I'm supposed to be able to do with my entertainment collection.

There, fixed that for you.

I bet you are in favor of banning water since it's possible to drown someone in it, too.

Bring it on!

[nobodyman]

This is starting to get interesting.

In theory it should be possible for the AACS LA to identify the players responsible for the breach and prevent later pressings of discs from playing back on those players until they are updated.

Personally, I can't wait for this key revocation to happen. The thing is, 95% of consumers have no idea what the hell DRM is. I'd wager that 95% of the people that own a hi-def player are blissfully unaware of the implications of key revocation are. Send out the key revocation lists and all that is about to change.

So magine the shit-storm when customers start flooding the Best Buy customer support aisle thinking that their machine is broken, when if fact it "works" just fine and the movie industry has shut down your player because some hacker is using its AACS key.

I can't wait.

Re:Bring it on!

[markbt73]

It'll play right into the MPAA's hands, though. "See? Those filthy pirates stole the key for YOUR player! It's their fault, not ours!"

Re:Bring it on!

[HairyCanary]

Won't matter. The average consumer will stop at "It does not work, give me my money back, NOW." They will not really care how it got that way.

Re:Bring it on!

[rblancarte]

I don't know if I would agree w/ this. The average consumer is quite naive. With the /. community, I would say they are knowledgeable. But most users don't really know what is going on in the black box. How does a DVD play? They don't know.

Regardless of what is going on, HD-DVD Cracked, flawed, etc. They are still going to use the technology.

RonB

Re:Bring it on!

[KKlaus]

Well its not quite there yet. This is a software player compromise, so all it will mean is that users are forced to patch, which they are unlikely to get too upset about. Now when hardware player keys start getting revoked... People get upset when they find out they payed 1 grand for a fancy brick. And as always, the pirates won't have to put up with any of this rediculous crap. Way to sell following your rules MPAA.

Re:Bring it on!

[r3m0t]

Consumers know that pirates can't send stuff down the power lines (or change pressed discs) to stop their movies from playing. They will want a refund, but I suspect (if a player is ever actually revoked) they will only receive refunds for the discs, not the players. They are also likely to just receive equally useless replacements.

Re:Bring it on!

[evilviper]

So magine the shit-storm when customers start flooding the Best Buy customer support aisle thinking that their machine is broken,

It's only one software player in question, and the updated version will probably be automatically download, or such, so few people even notice.

And when a hardware device key is revoked, you can bet the geniuses at Best Buy will be only too happy to explain either that their player is an "old model" which can't play newer discs, or that their player is old and worn out, so that it doesn't work anymore...

There will be NO mention of the fact that the industry intentionally sabotaged their equipment.

And in other news:

[dpbsmith]

The Hindenburg did not catch fire, it was merely the hydrogen in the Hindenburg that caught fire.

The Titanic did not sink, it was just that Captain Smith did not adhere to the specifications as to how the Titanic should be operated (it says clearly on page 216, "Do not allow icebergs to rip open more than four of the water-tight compartments.")

And talk of "blunders" in the Battle of Balaclava are hogwash.

Re:And in other news:

[cant_get_a_good_nick]

I know you meant this as a sarcastic comment, but..

The Hindenburg did not catch fire, it was merely the hydrogen in the Hindenburg that caught fire.

The thing that made the hindenburg so dangerous actually was the skin; hydrogen was just an aid. They took a small piece of the skin (very small, since it's historical item now) tried to light it on fire, and it went up like it was doused in gas. Since that was the skin, i guess you could say the Hindenburg did catch on fire.

I agree with your main point though. Their statement was pretty silly.

Re:And in other news:

[Anonymous Coward]

Don't cite theories as facts. I think the mythbusters would disagree with you.

Re:And in other news:

[evilviper]

The thing that made the hindenburg so dangerous actually was the skin;

According to ONE NASA scientists theory. Everyone else in the world says it was the hydrogen.

"New" theories always get serious press, in particular when the "old" theory is well-establish and otherwise universally accepted. That it gets press coverage doesn't mean it's any more sound than the old tried and true theory.

early adopters, instantly obsolete!

[swschrad]

quoting the weasels,

"in theory it should be possible for the AACS LA to identify the players responsible for the breach and prevent later pressings of discs from playing back on those players until they are updated. As such, if the hole can be patched in the players.."

which roughly translated into English means

"anybody hacks the Belchfire player, we just lock out new titles. Belchfire fixes it for you, or you pound tar, customers. Bwa-ha-ha-ha-ha!"

just what we all needed to hear to make us want to run out and dump a mortgage payment down the rathole on one of these things. thanks for clearing it all up for us, industry!

Re:early adopters, instantly obsolete!

[PitaBred]

I'll bet Belchfire will be much less willing to keep making players, too. Way to alienate even your cohorts, MPAA!

Re:Updated? Battle of the Rootkits!

[Goaway]

AACS can revoke individual player keys. No need to revoke a whole line of players.

Their only logical option

[Jugalator]

If they admitted this was in fact a miss in the AACS specification about protecting the keys, AACS LA could have their algorithm face a quite severe dent in its reputation. By blaming it on player implementations, it's not their problem. However, the real problem still remains despite whatever they say -- it's the end result that matters, not whom's fault it is.

Re:Their only logical option

[Anonymous Coward]

not whom's fault it is

not whose fault it is

The Grammar Nazi returns! Zieg heil!

Re:Another blow struck for free entertainment

[optikSmoke]

I think you miss the point. By the very nature of how any of these encryption schemes HAVE to be used (ie, within someone's home with key and encrypted data all existing on some device), they can be broken. Always. Therefore, the pirates will break them. The more difficult the studios make it, the more sport it will become to the pirates. I honestly don't see a way they can solve the piracy problem this way.

On the other hand, I purchase my movies (currently on DVD) and am very happy that DeCSS exists. It allows me to watch legally-purchased DVDs on linux, and possibly back them up if I wish. I hope a similarly useful scheme for breaking AACS exists if/when I ever get a high-def drive for my computer (could be a big if).

At the same time, the only people really detrimentally affected by these schemes are legitimate consumers. Once hardware players start getting keys revoked, this is going to become even more apparent. The studios will achieve:

a) Not stopping piracy (and possibly adding to the "fun" by making it marginally more challenging)
b) Starting an ongoing war of revoke keys / break keys / revoke / break / revoke / ...
c) Monumentally frustrating their legitimate customers. Even more so than the few linux/etc users annoyed by CSS.

Not when Apple does it!

[mattgreen]

Then it is Not That Bad Because I Can Waste Time Burning And Reripping. Don't forget about this important exception!

No more player software

[Anonymous Coward]

So there simply won't be any more PC-based player software. The outcome of this will be that no more software-based HD players will be licenced. Only stand-alone boxes, and sealed black boxes that sit between your PC/Mac and an HDCP-compliant monitor.

Otherwise the next time a programmer complains to a cryptographer that his DVDs won't play, a bypass will be found. Google "My first experience with HD content being blocked" to see this in action - about 8 days later, AACS was bypassed.

Re:No more player software

[Dunbal]

Only stand-alone boxes, and sealed black boxes that sit between your PC/Mac and an HDCP-compliant monitor.


      Oh, and don't forget to have those sealed boxes connect to a phone line, so that it can call home and report tampering attempts, as well as your viewing habits. Oh goodness me, how hard would it be to "expire" your movies after a pre-set number of plays with such a system...

Vicious circle of blame

[ThePhilips]

AACS hack is blamed on bad player implementation

As programmer, I can tell that it work both ways. Any deficiency (or bug) can be blamed on poor implementation. At the same time, big companies which actually looked and benchmarked development process (e.g. IBM) claim that 75% bugs are caused by erroneous specifications.

IOW, players were implemented as good as AACS has told what/how to implement.

Somehow, I doubt that documentation from AACS would be much better than that of Microsoft.

Re:Selective keying using the whole .exe from memo

[3.14159265]

Now, see, this is the part I don't understand.
Even if the key is not loaded in memory, isn't reverse-engineering the exe enough to find that key, wherever it is?

Re:Updated? Battle of the Rootkits!

[Todd Knarr]

Except that this is a "class break": it affects all players of the same type. If they revoke the cracker's player's individual key, the next 100 crackers continue blithely along unaffected. They have to revoke the keys for all players of the vulnerable type to stop the break. Which will always affect more legitimate customers than crackers.

fuckwits (n/t)

[Anonymous Coward]

What more can I say.

rgds

What a smart industry would have done

[zygotic mitosis]

A smart industry would have learned from CSS and required a PCI hardware decoder for playback.

Re:What a smart industry would have done

[pv2b]

An even smarter industry would just have put the decoding and decryption chip in the disc reader itself.

Though I guess that means they wouldn't be able to cash in on licensing fees for multiple pieces of hardware, but it would make a lot of engineering sense.

Re:Another blow struck for free entertainment

[cheater512]

Yeah if you put up walls then someone will see if they can tear them down.

I for example enjoy messing with certain MMORPGs as a hobby.
Its very theraputic. :)

Thankyou (parent is right)

[Cheesey]

Virtualisation does not save us from trusted computing - as the parent says, TCPA was designed with virtualisation in mind.

Every time a thread about DRM comes up, TCPA is mentioned, and a whole bunch of people get modded +5 Insightful for saying that they'll circumvent it using VMware or similar. But to do that, you have to make your own TCPA keys, which won't be signed by a trusted third party. Online services that require remote attestation will require you to use a key that has been signed in that way.

The key in your TCPA module will have been signed, but you can't get at that key by design. You can't use it to sign programs in your VM. That's the idea. They know that virtualisation is a hole. They are as smart as you.

However, perhaps we can get at the key in the TCPA module by getting the module to repeatedly sign something while monitoring its power consumption. This technique, differential power analysis, is apparently very hard to defeat. You can use it to get keys out of smart cards, given enough time: perhaps you can use it to get keys out of your own processor. The price of freedom in the future?

Get informed about TCPA here. http://www.cl.cam.ac.uk/~rja14/tcpa-faq.html

It's not really a "bad implementation" or bug

[Anonymous Coward]

The reason the title keys were retrieved is that the soft DVD player made it very easy to find the keys.

AACS (or any s/w based scheme) have an inherent flaw. Somehow/somewhere the decryption keys need to be brought into memory as clear text in order to decrypt the content. How do you protect them from prying eyes? One option is have the player play some kind of shell game, or try some form of obsfucation with the keys. Of course a determined hacker can track this down as well. You should never hide the secret in the content (see Kerckhoff's Principle) http://en.wikipedia.org/wiki/Kerckhoffs'_principle

The problem

[dtfinch]

Obviously, the player wasn't as defective as it was designed to be.

It only takes one time

[Danathar]

To crack a movie. Once it's been cracked and posted on Bittorrent fixing the key in the DVD player that was the culpret might keep the same hack from being used again, but it's toast for that movie.

Wrong audience, pal

[ruiner13]

I'm fairly certain that if at some point the **AAs ever visited slashdot that it didn't take long to figure out that this isn't the place for them to visit. Why don't you try actually sending them your thoughts DIRECTLY, as I have done in the past. If more people did, maybe they wouldn't think that the public actually wants DRM. Otherwise, you're just doing what the network exec in South Park said "please direct any further complaints to the brick wall over there". You're being just as effective.

Re:Wrong audience, pal

[sdo1]

>Why don't you try actually sending them your thoughts DIRECTLY, as I have done in the past.

As have I... regularly. I've even sent them pictures of my media collection just to prove I'm not BS'ing them on the amount of media I buy. But I'm pretty much fed up with them (the **AA's. I buy almost 100% independent artists now. I subscribe to eMusic. I shop at Newbury Comics (a local music chain that specializes in independent music, though they do carry all of the other stuff too).

Everything the **AA's are doing now are the death throws of a wounded animal. They're becoming increasingly irrelevant and probably will be a footnote in history 25 years from now.

-S

Making life hard for customers doesn't mean more $

[BillGatesLoveChild]

DRM *is* a pain the ass. Even on DVDs, with copies you don't have to sit through those annoying ads and logos or the annoying main menu (which always leads to the movie). On the real-McCoy you must suffer. How many people with legal copies of Windows are using volume keys just because they don't want to call up Microsoft for permission whenever they change their config?

The MPAA (and Microsoft) are fighting the way their enemy fights best. If you make DRM inconvenient, and it *is* inconvenient, hackers will find a way around it. If you overcharge, or having play-one-time-only restrictions, people won't use it. If you make any system harder to use than what is out there already, people will go around it! And I'd bet my money on a bunch of teenager hackers over any boring, Microsoft wage serf.

My suggestion: make movies cheaper and drop DRM altogether. PC game companies are realising this. My Oblivion DVD says 'we didn't include any copy protection so please don't copy this'... and I didn't. They've got my goodwill. Some hackers probably did copy it, but DRM doesn't make it any more or less likely. Maybe even more?

Keys Site

[Anonymous Coward]

hdkeys.com/

Good luck

[tepples]

Who says the TPM has to be in a machine running a "trusted" OS in the first place, boot it non-trusted and just feed the TPM the appropriate inputs to make it think the system is booting. Failing that, run the TPM by itself - yank it off the motherboard and host it on your own PCB with an FPGA programmed to provide a full "VM" for the TPM itself.

I await the Slashdot story introducing your implementation of this concept ;-)

Re:Good luck

[Anonymous Coward]

I await the Slashdot story introducing your implementation of this concept ;-)

Ah, the make fun because I've run out of counter-arguments approach!
He doesn't have to do it, but all it takes is one guy to do it and the jig is up.
Especially if it takes a few years to get there, by then there will be way too much hardware in the field to change course.

Re:Updated? Battle of the Rootkits!

[Goaway]

What, exactly is "this"? Certainly nothing referred to in the parent posts that I can see.

Re:Updated? Battle of the Rootkits!

[pyite]

Your parent's point is that if you obtain the player key for HDVision-1000 serial number ABCDE, just revoking the key for serial number ABCDE is not enough. Since you can obtain the key from one HDVision-1000, you can easily do it to any other amount of the same model, thus they keys for ALL of that model must be reversed, since the design* has been compromised.

Suffice it to say, the design of all of them is flawed from the get-go, so whatever.

Re:Selective keying using the whole .exe from memo

[Fat Cow]

It's probably not in plaintext in the exe, but they do have to have it in plaintext before they use it.

Re:Selective keying using the whole .exe from memo

[pyite]

Yes, but to be clever, the player might do something like encrypt the player key with some other key that it stores in plaintext. But, the bottom line is that *at some point* the player key will be in memory. I did basically the same thing for a project in an information security class. Our task was to locate a key in a 1MB memory dump to decrypt a GIF which was encrypted with AES-128. Java's crypto library was able to brute force test a million keys against the first 128 bits of the image (enough to see if the first 3 bytes are ASCII GIF) in about a minute and a half on 1 GHz G4. Decrypting one frame with all the keys in a memory dump can be quick. Even if you dumped all 1GB of your memory and tried it, thats 1500 minutes using the technique mentioned above. Not super quick, but you can improve that by only searching high entropy areas of memory (this could help a lot). In any event, this is most likely simpler than reverse engineering the EXE.

Well, Is that so? Not!

[hAckz0r]

Give me any HD-DVD or Blue-Ray hardware player using AACS and any old cheap logic analyzer and I could (but don't bother asking) hand you any hardware or volume key you want. DRM does not work because the whole concept of DRM is flawed. If you give someone the data, and also give them the key so they can play it, then they can copy it. Period. Any "magic" that is applied to keep you from knowing the key is merely a speed bump to an average geek.

All you need is one very pissed-off average geek that can't watch their bought-n-paid-for movie and the whole non-DRM'ed movie is likely going to be out there for everyone else, that can't watch their own copy, to download it. In fact, the more players that they "revoke" the keys for, then the more pissed-off geeks there will be, and the more movies that will likely be available for download. Its a loosing proposition any way you look at it. With DRM the "fix" becomes "the problem". The only people that win are the ones writing the DRM and spoon feeding the Board room executives that don't know that DRM can't work.

When will they ever learn that you can't solve a SOCIAL PROBLEM using technology of any kind. In fact they should wise up and realize that its the professionals that build specialized hardware that copy the "protected" disk bit-by-bit, then burn a thousand copies, and are making big bucks off of all the boot-leg copies. Those are the ones they should go after, not the average people that paid for the movie and just want to watch what they paid for, when and where they want to. So, RIAA/MPAA, take it from a security geek, know thy enemy! You can't fix a problem if you don't even try to understand what the problem is!

Re:Well, Is that so? Not!

[angulion]

To me it seems they know what the problem is quite well and they are getting closer to solve it.
The problem is that you (consumer) can buy a (HD-)DVD or CD and transfer it to your ipod/phone/something instead of buying it *again* in a different format. *That* is the point of DRM, it never was about piracy, blaming piracy just sounds so much more legimate/better with the media.

Re:Selective keying using the whole .exe from memo

[Burdell]

It doesn't have to end up in the process memory in the clear; it could be kept in a register. Those register contents will end up on a kernel stack somewhere when you hit a process switch though. That could be limited by getting everything set up to decrypt, yeilding control back to the kernel (not sure exactly how) so that when you next run, you have a good chance at running a certain amount before being interrupted by another process. If you can decrypt in that time, the key still doesn't end up in RAM.

It is hard (and maybe impossible in this case) to do, but it could happen.

Malware, and why they made this statement

[Myria]

Two separate but important points:

1. The most devastating attack that can be done against software players would be to use malware to extract keys. There are many, many zombies out there. The malware could search for installed HD-DVD/Blu-Ray player software on the victims' machines that it knows how to break, extract the unique key from such software, and send to the malware author. There would then be enough keys known that only revocation of the entire product line's keys could get around the problem. I wonder whether they've considered this scenario. (However, one mitigating factor is that malware is done for profit, and this wouldn't be profitable. For-profit pirates just copy disks outright without bothering to decrypt.)

2. The reason the AACS made that wording about the players not following the "Compliance and Robustness Rules" is probably so that they can invoke the parts of the contract allowing them to fine the licensee millions of dollars.

Business Economics 101

[Anonymous Coward]

My suggestion: make movies cheaper and drop DRM altogether.

Do you suggest that they pay big stars less money? Big directors? Big producers?
Should they just make less profit? Should they require theaters to raise the price of popcorn?

Ridiculous!

I say they hire more lawyers, spend more on R&D, and keep changing DRM. That's the only way they can keep making the money they deserve. If all else fails, raise prices to make up for the loss of volume.

Re:Business Economics 101

[Archangel_Azazel]

--Do you suggest that they pay big stars less money? Big directors? Big producers?
Should they just make less profit? Should they require theaters to raise the price of popcorn?--

In order: Yes. Yes. Yes. Yes. and for the love of all the Gods in existance NO! $4 for a $1.50 box of Sno-Caps is more than plenty thank you.

--I say they hire more lawyers, spend more on R&D, and keep changing DRM. That's the only way they can keep making the money they deserve. If all else fails, raise prices to make up for the loss of volume.--

"The money they deserve" ? Do you seriously think that making a movie is worth Millions? Billions? People who perform heart surgery don't get nearly that much and these people have made a living SAVING PEOPLE'S LIVES.

You'll forgive me as I sit here in my run-down house, working a job I hate for next to no money because it's the only one I can find and NOT shed a tear for these poor destitute movie makers.

My 2 cents,
A.A

Re:Business Economics 101

[level_headed_midwest]

The parent supposed that if the studios lowered the price of the films and dropped the DRM, they would end up with higher overall revenues and profits. You know, sell each unit for a little less, spend less on DRM, and sell a whole lot more units = higher profits.

I see that you slept through Econ 101...

Re:Business Economics 101

[Zoolander]

*Swooooshhh*
-Did you hear that joke that just flew by? -Huh?

what about a privilage escalation type of attack?

[Anonymous Coward]

I'll be honest, I don't have a perfect idea of how this circumvention worked, but I had some thoughts about it and I would like if someone would be kind enough to access their validity.

What this article and the others related to this story basically show that a user can gain some control over this encryption and the other comments I see seem to indicate that a user can gain local control of most drm devices in question.

But what would happen if a greater degree of control was obtained though detailed analyses over time of the software and hardware. Say in six months, an organized group determines how revoke keys or force tainted upgrades on the user. Could you see electronics manufactures being forced to pay protection money to stop virus, or piles of HD players (I imagine they have decent computing power) being used to power a botnet?

Keyjacking will be possible in future

[ratboy666]

But that is what DRM must do -- hide the "secret" in the content. The soft DVD player did NOT make it easy to find the key. However, HD has to work pretty much "to the metal". The key was not "identified", in the sense of being reversed.

The attack is simpler. We know the key must be in memory SOMEWHERE. So, try decrypting with all possible memory bytes. For each 1MB of memory the player takes, the keyspace is effectively only 20 bits. The key is "brute forced" across the player. The key of interest is not in the read-only portion of memory (not a player key), which reduces the search space.

How to protect against this: the key may be broken up and hidden in its construction. Instead of storing the bytes in sequence, they can be (say) separated. If the key is known to be in a 1MB region of memory, this can provide up to 20 bits additional keyspace for each byte. But, the pattern of access can be determined by watching the processor, so this would be defeated. Basically, the attack would be to watch all data accesses, and determine which is incoming encrypted. All other access would be potential keyspace, which can the be "bruted". I would estimate that only up to 12 key bits per byte would be added (watching page access is trivial). Which makes the brute force attack possible again. The AES crypto core itself could come under attack -- not to break AES encryption, but simply to determine where its key is coming from.

This attack cannot really be defended against. Except if we either (1) move the key storage to somewhere more secure, or (2) modify the "Operating System" to not allow content-vendor untrusted software. The soft player would need to survive all possible attacks that may result in other code to be executed (thus allowing injection of monitoring software). Which means the player software needs to be either "perfect" in a security sense, or that the OS cannot allow any "untrusted" drivers (ring 0 code) to be run WHILE the player is running, AND not permit any application access to the player (to prevent fuzzing attacks, etc.), AND not allow direct data recovery from (say) screen buffers (which means, in turn, that the entire GUI has to be hardened, as well as the fancy 3D drivers, etc.).

Much easier to just "sandbox" trusted applications by forcing them to run separately (in a sense) from all untrusted applications. Since a possible attack is to recover frame data from the (say texture memory) graphics subsystem, no display of other applications can be allowed.

The computer would need to be converted into a simple HD content player while the HD content is playing.

So, you are correct. Key-jacking will be possible and practical in the future (at least as long as it is needed)

Re:Making life hard for customers doesn't mean mor

[MrZilla]

My suggestion: make movies cheaper and drop DRM altogether. PC game companies are realising this. My Oblivion DVD says 'we didn't include any copy protection so please don't copy this'... and I didn't.

Actually, the only thing I can see that DRM (or copy-protection) could help with is so called 'casual copying', i.e. the thing I know at least I did as a kid:

"Hey, you've got the new Doom II? Cool! I have a bunch of blank disks in my backpack, could I get a copy?"

But other than that, no, they will never stop piracy, only inconvenience their customers. Some companies understand that, like ID Software who eventually dropped the CD check in Quake 3, others do not, like Blizzard who recently made it so you can no longer play on Battle.Net if you are using any form of No-CD patch, regardless of your bought and valid CD Key.

Re:Making life hard for customers doesn't mean mor

[BillGatesLoveChild]

> Blizzard ... can no longer play on Battle.Net if you are using any form of No-CD patch,
> regardless of your bought and valid CD Key.

I use no-cd on all my games, because I hate fumbling for disks from a stack whenever
I want to play something. I only have one drive you know. Some laptops have none!

Thanks for the warning. There are so many games out there now days and such little time
that you can afford to be choosy. If companies do this, lets name them and shame them
so at least buyers know they'll spend a week of their life this next year shuffling CDs.

Blizzard, you're evicted!!!

Quantum computing

[Joce640k]

Quantum computers are only useful for a small set of problems, breaking AES isn't one of them.

There's a much simpler argument to show that 128 bit keys are big enough and that's to calculate the amount of energy needed to try all keys. Even if a key can be tried with the energy of a single electron transition then you still need to suck all the energy out of a couple of sun to do it.

2^128 is a *big* number!

Re:Quantum computing

[Chyeld]

The problem with this assumption is twofold.

A. It assumes that the key will be the last possible one in the key space.
B. It assumes that the only method used will be 'pure' brute force.

A. is almost certainly not true. And while it might be optimistic, it's quite possible that it'll be discovered that due to some brain dead maneuver the keys themselves have been generated weakly in a fashion where all 128 bits don't really come into play.

B. might be true for now, but I refuse to believe that there aren't already people out there working on more elegant methods of brute forcing the keys which would allow the space to be narrowed down to specific areas 'quickly'. I also refuse to believe there isn't one.

Re:Quantum computing

[numbski]

I'd mod you up, but my points all just expired.

That's the fun thing about brute force...there's nothing that says that the first key you try won't work, other than odds and common sense. You could quite possibly break it on the first try. :)

So I say, let the games begin. Right a screen saver to do a distributed brute force on the key, and give it a try. You say we don't have enough CPU power, I say we don't have to have "enough" do do them all, only "enough" to get the right one. :)

You are wrong!

[Dion]

AACS has such a large keyspace that every single player can get its own individual key.

AACS also allows cheap and secure revokation of just a single player key.

If the entire production run was flawed then all 10000 playerkeys can be revoked at the same cost as revoking a single player.

There is (sadly) no hostage situation possible with AACS.

Re:You are wrong!

[r3m0t]

"AACS has such a large keyspace that every single player can get its own individual key.

AACS also allows cheap and secure revokation of just a single player key.

If the entire production run was flawed then all 10000 playerkeys can be revoked at the same cost as revoking a single player.

There is (sadly) no hostage situation possible with AACS."

It's true that every player manufactured (hardware, CD, or download) could be given a different key, but currently, each version of the software and each production revision of the hardware gets a different key. This *does* create a hostage situation.

Besides, there are serious practical problems with revoking keys - whether it's one individual player sold, a production run of players that all share the same key, or a production run of players with different keys. The only "easy" thing to deal with is software players.

If AACS thinks their system is so secure (see their recent statement), why don't they revoke the player key of WinDVD? As it is, they haven't taken that action, so new HD-DVDs and BluRay discs can still be copied.

Re:You are wrong!

[Dion]

What are the large practical problems with revoking player keys?

What makes you think that the windvd keys havn't been revoked?

Do you have a movie that was mastered after the crack was made public?

The only thing that happens when a player key is revoked is that new movies get encrypted so that player key doesn't work any more, it's not like all windvd installations will magicly stop working.

I once read an article that proposed that all software players should share the same key and that key would be revoked about quite often (once a month was mentioned), thus forcing all users of software players to upgrade often to be able to play new movies.

I don't know if this is what they do at the moment, but there is really nothing to stop them.

Re:You are wrong!

[r3m0t]

"What are the large practical problems with revoking player keys?"

A product recall and replacement, including a publicity campaign to advertise the recall. Either that, or risk facing a class action lawsuit and, of course, tarnishing your image. (You know those horror stories about AACS LA bricking your player? The people who write them would be falling over themselves if it actually happened once.)

"What makes you think that the windvd keys havn't been revoked? Do you have a movie that was mastered after the crack was made public?"

I thought that if the WinDVD keys were revoked, AACS LA (and Slashdot) (and Arstechnica) would have said so. That's far more interesting news than "AACS LA denies that AACS has a fundamental flaw." Also, I haven't heard about WinDVD offering a patch, although technically they don't need to offer any patch until new remastered discs are published.

"I don't know if this is what they do at the moment, but there is really nothing to stop them."

They haven't done that yet because no player was compromised. I agree that it is possible, and I suppose that the people who can afford HD discs are very likely to have a good connection to the internet once a month. (I'm not sure once a month is often enough, though.)

The problem with it (for Hollywood) is that if an old version is revoked, there is still a massive library of discs whose title keys can be extracted. As you said, the software won't suddenly stop working, it will just stop playing new discs. I also imagine that the cost of remastering all (or most) of the discs that are *currently* out and being sold, every month, is high.

Perhaps software players will build in the "functionality" of bricking themselves (for all discs) as soon as they encounter a disc whose title key they cannot decode. They will also refuse to run under VMs (if they don't already). However, I know of some stuff which can restore your disc to a known state with a reboot. (That allows somebody with a bricked software player to unbrick it and try a different disc.)

Basically, DRM cannot stop piracy for another 3 years... quote me on it. :-)

Re:You are wrong!

[Dion]

Bricking a software player is virtually free, if the player can self update.

Otherwise I'm pretty sure the content mafiaa is pretty content bricking the individual player that was compromised.

Bricking an entire model of hardware player is going to be quite a different matter, I'll give you that.

I don't think that the cost of remastering a disc is all that high, they might even remaster for every single pressing of the disk, just to be mean.

I quite agree that DRM will never stop piracy, scraping the bits out of the LCD and the amplifier is never going to be impossible.

DRM will make it hard to write good players that will play unaltered disks, so in the end it will mean that people like me will never buy a legal disk and in stead sink that money into a fatter Internet line.

There is an alternative though.

[Dion]

Someone with a valid player key could set up a server that will decrypt title keys that HD customers send it.

That way the player key is never in the clear and the content mafia will not know what player key to revoke.

The tricky bit is that it's quite possible for the mafia to figure out what player key it is, because they can generate all the revocation trees they want and use that to perform a binary search for the player key.

So when someone does this they will need to make sure that the system refuses to even work on key blocks with strange revocation trees (previously unknown) that come from untrusted sources.

Alternatively the server could simulate that its key is everywhere in a huge part of the keyspace where there are no revoked keys, so if the attacker tries a keyblock that has any key in that entire range revoked, then the server will act as though it couldn't decrypt the keyblock and thus keep the attacker guessing.

Re:Selective keying using the whole .exe from memo

[kruhft]

This can be foiled by 'encrypting' the key by swapping the bytes and using a bit of assembly to 'decrypt' the key in a register before use and making sure the key never leaves the register at any time. Not really encryption I know, but it's not difficult (if you know the arcane art of assemly) to foil this type of attack.

What about unrelated drivers used to break TPM?

[Anonymous Coward]

I am no expert in these matters, so please correct my assumptions if they are wrong:

To play back, say, a movie stored on an encrypted HD-DVD, you need:

- The data on the HD-DVD
- A "known secure motherboard" with TPM-Chip
- A "known secure OS booted off and verified" by this motherboard
- Trusted drivers for all kernel-mode peripherals, signed off by OS-Vendor
- A software player recognized as "known secure and not yet revoked" by the HD-DVD and
    launched under the OS

Now, let us assume that I program a kernel-mode peripheral such as a graphics card and put a backdoor in my driver that neither my employer nor microsoft notices.

Since a graphics card needs to handle uncompressed and unencrypted content, an image of each frame of the movie is in my memory.
With a few extra cpu cycles, i can re-encode it in plain jpeg or mpeg and write it at low bandwidth as digital noise to, say, a secondary LCD-Display connector, where another PC records it and writes it to disk.

Note that this works with any motherboard, any software player and any OS version up the point that my backdoor is detected.

Now i can:
a) Rip all movies released up to that point
b) Rip all later movies that do not expressly reject all prior OS versions

Since only one bad guy needs to do this to be able to pirate the movie, b) effectively means that any legitimate user cannot watch new HD-DVDs he bought, unless he periodically connects to the internet to keep his OS-Version up to date.

The interesting thing here is, that for driver development, there must also be developer hardware and software that does not enforce trusted drivers, because I cannot have them verified and signed off by the OS-Vendor between compile-cycles in development and debugging in any meaningful way.

I could possibly see how Apple, which controls drivers and Hardware for their boxes, could hope to pull off such a complicated scheme with at least a tiny chance of success.

But for Microsoft, relying on thousands of third party system and component vendors, this seems practically impossible to me.

-

Re:Updated? Battle of the Rootkits!

[Goaway]

No, you can most likely not "easily" do it. Usually, this kind of thing takes scanning electron microscopes and multi-million dollar equipment. The keys are not going to be sitting in some externally accessible memory you can just dump.

Re:Updated? Battle of the Rootkits!

[pyite]

You underestimate the amount of people with skills to use an SEM and plentiful access to them. I'm within walking distance of two of them and while I couldn't use them, grad students are poor and would love to make a few bucks.

Re:Updated? Battle of the Rootkits!

[Goaway]

Who'd pay for a key that will only be revoked as soon as it is used? And it's not like the problem is the skills to use an SEM - that's just the basic tools. You need to be able to actually take the chip apart layer by layer, scan it, and then the real work starts, trying to reverse engineer it.

Granted, once you've done the reverse engineering once, doing the same thing again is significantly easier, but in the end, all you get is a key that will be revoked as soon as you start using it.

Re:Making life hard for customers doesn't mean mor

[aaronl]

Daemon Tools still seems to work fine for me, as does a mounted image and Cedega.

Re:Updated? Battle of the Rootkits!

[pyite]

Granted, once you've done the reverse engineering once, doing the same thing again is significantly easier, but in the end, all you get is a key that will be revoked as soon as you start using it.

This is certainly true. However, if it becomes common practice for player keys to be revoked, people will stop releasing them to the public and just release volume keys. While less useful, you really only need one volume key to unlock a lot of copies of a title.

Re:Making life hard for customers doesn't mean mor

[BillGatesLoveChild]

True. Just saves you have to keep the ISO on your HDD. (BTW wikipedia says latest version of Daemon tools has spyware you need to deselect on install)

This is a good place for NO-CD patches. *Seem* virus/spyware free. Legal caveats apply:

http://gamecopyworld.com/

GPL and EULAs in US law vs. UK law

[tepples]

The Free Software Foundation's position (I mailed them and asked) is that once you have a copy of the software you OWN it, the way you own a book or a TV or a car, copyright (and thus the GPL) only applies when you make a copy to give to someone else.

You're right. Unlike the typical proprietary software EULA, the GNU General Public License does not take away material rights of the owner of a copy, such as rights under Title 17, United States Code, sections 109 (right to resell a copy; right for nonprofit libraries to lend a copy) and 117 (right to copy computer software into RAM and to make backups). But unfortunately, United Kingdom law does not have anything directly corresponding to 117, so in that country, the GPL applies to plain users.

Re:Smart Cards Already Do This.

[mpapet]

The logical next step is to allow only hardware and partial-hardware players.
Correct. Welcome to trusted platform computing. It's coming and Microsoft will be forcing it down your throat. See WMP 10's drm, Vista's DRM. Tip of the iceberg.

f the keys are truly embedded in the "trusted" ASIC: Making custom chips is expensive
Wrong wrong wrong! It's called a smart card module. A surface mount is simply another package.

If the keys are somehow individualized to each computer....
Smart card modules that are actually microprocessor are specifically designed for this kind of activity and very difficult to compromise.

The smart card module will make it sufficiently difficult to freely copy digital content very few will do it. That's the objective, not 100% air-tight DRM.

Re:Updated? Battle of the Rootkits!

[r3m0t]

"Your Foobar 1000 will play discs produced in 2006 and 2007. It ceases to work for discs produced between February 2007 until you buy a disc produced a few months later that happens to contains some code that query the player whether it's a Foobar 1000... and if so, to automatically/silently patch the firmware. Then all your discs work again.

That's a good thing for the user, and a bad thing for the industry, because as soon as you've got a firmware patch on a DVD, the obvious thing for an enterprising hacker to do is to put his own firmware patch on his own DVD"

An enterprising hacker *with the private keys of the AACS group* in order to sign their new firmware. Unlikely.

Look at BD+.

Re:Selective keying using the whole .exe from memo

[r3m0t]

Now that we know some real title keys, somebody can play 20 different movies, take 20 memory dumps, and compare them to find the title keys.

Re:Updated? Battle of the Rootkits!

[Goaway]

This is certainly true, but it also means you are limited to the discs you can personally get your hands on. And while maybe you feel good about yourself for helping others by releasing a handful of keys, the effort required to reverse-engineer an entire microchip is maybe not worth it.

No, that's not going to fly.

[Kadin2048]

An even smarter industry would just have put the decoding and decryption chip in the disc reader itself.

I'm sure they've thought of this. The problem is that if you put the decryption there, then necessarily the reader must output a decrypted stream to the computer, which is anathema to the whole concept. If the drive just spits out a decrypted stream, why bother making the content on the disc encrypted at all?

No, the whole point is to carry it in encrypted form as far along the signal chain as possible. This is why the decryption is done in the computer/player, rather than in the drive itself. They want the computer/player to authenticate the rest of the downstream signal chain (the HDCP compliant monitor, generally) and only then decrypt the video content and send it onwards (and even then, re-encrypt it specifically for the display, so that you can't just attach some alligator clips to the HDMI cable and get the unencrypted hi-def version).

In the optimal arrangement, the decryption would be done as far "down" the signal chain as possible (where you, the viewer, are at the absolute bottom, where you belong), preferably in the display, just before the last-minute conversion to analog signals of some sort. However, for various technical (and political) reasons, this isn't practical, and so the decryption of the disc content is done by the player, which then decides what level of content to send out to the display device depending on its credentials.

Re:Updated? Battle of the Rootkits!

[Briareos]

Consider the sales/support implications of customers selecting products for Christmas 2008: "Well, sir, this Foobar-1000 plays discs up produced in 2006-2007, a Foobar-1130 plays discs produced from 2006-2008, and a Fonybaz-1900 plays discs produced from 2006 to August 2008."

And, of course, the Foobar 2000 will only play audio.

We don't need another bloated media player - WinAmp was horrible enough... :P

np: Radiohead - Morning Bell (I Might Be Wrong - Live Recordings)

Nobody likes a dongle.

[bensch128]

Sure, it's possible to bury the key inside a dongle distributed with each software player.
However, it's annoying to work with and difficult to distribute

The video game industry and the mid-high end design industry tried dongles and both failed.
The only place I've seen it work was at a company which made textile design software
which was extremely specific and extremely expensive.

I seriously doubt any mass-media software will be tolerable with dongles.

Just my $0.02
Ben