back to article Relax, Hollywood, ARM's got your back: New chip 'thwarts' video pirates

ARM today touted the Mali-V500, its new graphics processing unit for gadgets, which apparently can protect 1080p video from pirates. The British design biz said the V500 can decode high-definition video without giving the operating system, nor the applications running above it, the chance to copy the footage in transit: in …

COMMENTS

This topic is closed for new posts.
  1. This post has been deleted by a moderator

    1. The BigYin

      Re: "ARM's TrustZone" - abuse of ENGLISH!

      Same with "SecureBoot". After all, only ne'erdowells would disable security. It's not like anyone else would ever want to do that and use their property in the way they see fit, not mandated by a freedom hating corporate.

    2. Vimes

      Re: "ARM's TrustZone" - abuse of ENGLISH!

      Even those that willingly pay have problems...

      http://theoatmeal.com/comics/game_of_thrones

    3. Dave 126 Silver badge

      Re: "ARM's TrustZone" - abuse of ENGLISH!

      >The marketers are taking a beneficent word - Trust - and turning it into a euphemism for something that removes the freedom of the user to use a computer or other possession how he wishes.

      I pay for it. I get to watch it- where is the betrayal of trust, Eadon? It sounds like a straight deal to me, one I can choose to take up or not.

      TrustZone can demand exclusive access to the hardware, of which this DRM scheme is just one application. Other applications include preventing memory-resident malware from sniffing PINs or passwords.

      Some people might wish to use their device to access a movie streaming service, and pay for the convenience.

      In any case, this doesn't nothing to prevent you from watching content from which you have previously stripped the DRM (or torrented), so I don't know what you're getting upset about. Many people can't be arsed with that sort of faffing about, and have the money to pay for convenience. To earn this money, they generally make themselves useful, by taking out your trash, tending to your illnesses, or generating the electricity that powers your Linux box.

    4. Jamie Jones Silver badge
      Unhappy

      Re: "ARM's TrustZone" - abuse of ENGLISH!

      " DRM forces the same user to buy the *same* content multiple times if he wants to see/hear it multiple times. And it forces them to watch annoying anti-piracy ads!"

      Surely this is against the law (First Sale Doctrine) - restricting reuse of something bought..

      I'd love to see a test case.

      Sad face, because I upvoted Eadon! :)

      1. Tom 35 Silver badge

        Re: "ARM's TrustZone" - abuse of ENGLISH!

        "Surely this is against the law (First Sale Doctrine) - restricting reuse of something bought.."

        No because they say you licence it, and the licence (page 17 of 43) said you have given up all your rights.

        There have been several services that sold* DRMed video and music who turned off their servers. Remember Plays for Sure? Walmart Video? ...

        There have been test cases with iTunes content that people wanted to resell.

    5. mike2R

      Re: "ARM's TrustZone" - abuse of ENGLISH!

      Leaving aside the linguistics and ethics, this: "In any case, video DRM always gets cracked sooner rather than later. One reason is that it is because you have to *display* the output" seems to indicate you skimmed the article.

      The whole point of the chip seems to be to solve this. Now I don't claim to understand how it works, but this isn't some MPAA guy saying "we've got some *technology* that will stop copying by using binary and encryption and other things you little people wouldn't understand." This is ARM, and if they say they've done it I'd think it warrants hearing them out at the very least.

      1. Richard 12 Silver badge
        FAIL

        Re: "ARM's TrustZone" - abuse of ENGLISH!

        What they've done is offer a way a device manufacturer can ensure that no application running on the ARM core of the same device can read back the output of the framebuffer. Quite how this can usefully stop an application popping it back into normal mode to read the framebuffer is unknown, given that users want nice transitions between 'play video' and 'not play video' and quite like having GUI elements like Play, Pause, Quit etc.

        But this feature isn't for the users. It's for devices that are genuinely frightened of the user.

        However, any given implementation of this feature may be flawed, and the output of the physical IC remains (and always must remain) available - both LVDS and LCD-RGB will forever remain unencrypted (so getting the image is easy if you pop the lid) and both HDMI and DisplayPort are crackable, if not cracked already.

        Fundamentally, the reason DRM as a concept cannot possibly work is because Bob, Eve and Mallory are all the same person.

        For the same reason, the overall effect is to only annoy legitimate users and damage reputations, because it prevents the legitimate user from watching the content they've paid for - both when it works and when it breaks down.

        In this case, if any issue (eg minor bug in playback app) causes the 'secure' GPU to get stuck in takeover mode, your device is a brick.

    6. N13L5
      FAIL

      Re: "ARM's TrustZone" - abuse of ENGLISH!

      Abuse of ownership principle too and a free society too.

      "permitted OS"...?

      Sick of people trying to sell you something and then tell you what you can install on it, and ARM supporting this kind of anti democratic fascist idea.

  2. Eponymous Cowherd
    Thumb Down

    Almost correct......

    "Unbreakable DRM would be so encumbering as to prevent sales; the trick is to make buying content easier than copying it, which is what ARM is hoping to facilitate with the Mali, but how exactly it works will be up to the manufacturers who using it."

    Current DRM systems already prevent sales. It isn't the buying of content that is the problem, it is making reasonable use of it once you have bought it.

    It works like this. Buy movie. Try to watch it, FAIL. Buy movie, try to watch it, FAIL. Download torrent, try to watch it, SUCCEED. The buying is easy. Its the watching that is the problem. Until the DRM obsessed media industries realise that DRM is a major driving force behind media piracy they will never beat it.

    1. Vladimir Plouzhnikov

      Re: Almost correct......

      The problem is that for you the goal is "watch" but for them it is "buy". As long as they trick you into paying they are happy, "watching" is not something they concern themselves with.

      1. Lord Elpuss Silver badge

        Re: Almost correct......

        @Vladimir

        Not true. Successful companies DO care about the 'watchability' i.e. end-user experience. A failure to make the experience enjoyable and successful would mean that punters will only ever buy a maximum of 1 item from them- before giving up and going somewhere else. Think how successful iTunes would have been if people had only ever bought one song.

        In a high transaction volume retail space, the measurement metrics that count are Willingness to Repurchase, and Willingness to Recommend.

        1. Dave 126 Silver badge

          Re: Almost correct......

          >How true, my daughter had a Sony CD player 'system'. The only way it could play Sony CDs was to copy them on a PC and play the copy. Totally stupid is not the half of the 'mu-sick' business.

          Curious. I have seen a Sony CD player refuse to play all tracks on a brand-new from HMV CD, Jurrassic 5's Power in Numbers. This must have been a different Sony scheme to the one that deliberately placed errors on a CD's TOC, errors that upset PC CD drives (and those car stereos that used the same drives), but not normal CD players. The idea was to prevent easy ripping. I seem to recall that such CDs didn't sport the traditional 'Compact Disc' badge on the cover, since they didn't conform to the Red Book standard.

        2. Eponymous Cowherd
          Thumb Up

          Re: Almost correct......

          "In a high transaction volume retail space, the measurement metrics that count are Willingness to Repurchase, and Willingness to Recommend."

          Indeed:

          Ultraviolet:

          Chances of me repurchasing: NIL

          Willingness to Recommend: NIL (unless to say "Avoid like the plague").

          1. Vladimir Plouzhnikov

            Re: Almost correct......

            Re: Ultraviolet.

            Yet, every time you buy a DVD or BD disc "bundled" with UV they are paid royalties, which they will report as a "great success" in attaining customer interest.

      2. Eponymous Cowherd
        Unhappy

        Re: Almost correct......

        "The problem is that for you the goal is "watch" but for them it is "buy". As long as they trick you into paying they are happy, "watching" is not something they concern themselves with."

        But how many time will people buy without being able to watch?

        1. Tom 35 Silver badge

          Re: Almost correct......

          Take the "digital copies" that come "free" with some DVD / BluRay disc.

          I only ever got one to work (after getting new codes from their support system twice) and the quality was crap. Now I toss the paper with the code in the trash with the shrink wrap, I'll make my own digital copy that I can play where I want if needed.

    2. Richard Jones 1
      Unhappy

      Re: Almost correct......

      How true, my daughter had a Sony CD player 'system'. The only way it could play Sony CDs was to copy them on a PC and play the copy. Totally stupid is not the half of the 'mu-sick' business.

      1. Suburban Inmate
        Pirate

        @ Richard and Dave

        I had the opposite problem and a bit of embarrassment when I burned a CDR (free to download and burn album) to take to test out some speakers I was thinking of buying. The shop's Arcam player wasn't havin' none of it! Had to resort to the Gnarls Barkley CD they had kicking around. When I got back home the CDR played perfectly in my 15 year old Technics.

        Additionally, SecuRom on F.E.A.R. wouldn't play nice with my drive, and I had to crack it just to play it, bringing the added bonus of not having to fart about with physical media.

  3. Ogi

    The thing about DRM...

    The problem is, no matter how hard they make it to copy the output, you only need one person to succeed and then it gets shared the world over.

    They are fighting a losing battle. Even if they made it totally impossible to access the bitstream, in the worst case someone can screen scrape frame by frame, reassmble into a movie file, and share it. The others will just download it.

    Funnily enough, I think that the harder they make the DRM to crack, the higher quality rips will be available. When everyone can click a button and rip a DVD, you get all sorts of rips, with varying audio/video quality/distortion, audio/video out of sync, etc...

    If it becomes really hard, then only those with the skills to do it will be able to release anything, and those people will probably also have a clue when it comes to normalising the audio correctly, and otherwise making sure everything works as intended. We'll get fewer rips of a movie/TV Series, but possibly better quality overall.

    1. JeffyPooh Silver badge
      Pint

      Re: The thing about DRM...

      "...you only need one person to succeed and then **it** gets shared the world over."

      You probably intended the 'it' to mean the media files stripped of DRM.

      The other option is 'it' being the piracy-enabling app that performs the technically-challenging task with one click.

      It's an obvious point, but I still see some clinging to the obsolete position that if something is technically difficult, then it helps to protect DRM'ed content. They overlook the obvious fact that computer processors exist in the wild.

      1. Ogi

        Re: The thing about DRM...

        Indeed that is true, but even if we assume that they somehow magically make 100% hack-proof DRM with Trusted module/path/execution, it can still be defeated by a determined individual (or group of individuals).

        Even if it meant they had to sit there and manually screen scrape the whole thing. The fact is that at some point in time the system will have to show the content to the user. And you only have to do this once, then the content can be distributed far and wide by the usual methods with no degredation of quality.

        DRM fails because it tries to deny access to the end user, while at the same time having to allow access to the end user. At some point in the line, it will be interceptable (unless they start embedding TPM modules in our brains to disable our audio/video senses if there is any unauthorised content around).

        1. Suburban Inmate
          Joke

          Re: "embedding TPM modules in our brains"

          STFU Ogi don't give them ideas!

  4. Anonymous Coward
    Anonymous Coward

    x86 > ARM

    Just sayin'.

    1. This post has been deleted by a moderator

      1. Anonymous Coward
        Anonymous Coward

        Re: x86 > ARM

        The main problem is that ARM architecture doesn't support x86 applications.

        x86 has been around for decades now, thus making things even more complicated.

        1. Tom 7 Silver badge

          Re: The main problem is that ARM architecture doesn't support x86 applications.

          Not a problem for 99.99% of applications - just MS office by the looks of it.

      2. This post has been deleted by its author

      3. JeffyPooh Silver badge
        Pint

        Re: x86 > ARM

        "In terms of chips sold these days, it's ARM > x86. Just "Sayin" (bah humbug!)"

        I guess that SMT resistors take the win then. In second place, SMT capacitors.

      4. Anonymous Coward
        Anonymous Coward

        Re: x86 > ARM

        "In terms of chips sold these days, it's ARM > x86."

        In terms of market segments where the technology is relevant, it's ARM >> 86 too.

        x86 is relevant in Window boxes and irrelevant everywhere else.

        ARM isn't relevant in Window boxes but in any other piece of consumer or professional electronics, there is little sensible alternative to ARM.

        1. Anonymous Coward
          Anonymous Coward

          Re: x86 > ARM

          >x86 is relevant in Window boxes and irrelevant everywhere else.

          What about Linux or Mac? What do they use?

          1. M Gale

            Re: x86 > ARM

            Well Mac is somewhat exclusive, but Linux? Well, there's x86, ARM, MIPS, PowerPC, 68k, Atmel oddities, your auntie's kitchen toaster...

            Linux is a whore, and we all love her for it.

            1. Anonymous Coward
              Anonymous Coward

              Re: x86 > ARM

              "Mac is somewhat exclusive"

              Yeah, that's right, just 68K, PowerPC, and currently x86 so far in IT kit (desktops and laptops). And what about iOtherStuff? How much ARM can it be?

              1. M Gale

                Re: x86 > ARM

                I don't see Apple Macs running on anything other than x86, and specifically Intel x86 at that. They dumped 68k and PowerPC faster than a fashion magazine dumping a model on her 29th birthday.

                Granted, I wouldn't want to try and run the latest Photoshop on a 68k machine, any more than I'd want to try and emulate a Core 2 on a 6502.

              2. Mark .

                Re: x86 > ARM

                OS X has never supported 68k, and PowerPC was ditched years ago. IOS is not OS X, nor even uses the "Mac" trademark (it's just a brand name - today's "Macs" are x86 PCs, with nothing but name in common with the original Macintoshes - and even the name isn't entirely the same).

                By your logic, Windows supports PowerPC, x86 and ARM.

                Talking about 68K and PowerPC are irrelevant anyway, since the OP didn't say only x86 was worse, he said "there is little sensible alternative to ARM", which would include PowerPC and 68K.

            2. Mark .

              Re: x86 > ARM

              So everyone here who uses Linux is doing it on ARM boxes, because "there is little sensible alternative to ARM"?

              I doubt it. These arguments are silly - ARM has its advantages, and is overall more popular (I believe it's been this way since the 90s) but doesn't come close to x86 for high end performance. The idea that the only reason for x86 is those people stuck with Windows x86 compatibility is not very well supported. (And anyhow, Windows now does support ARM - the only platform stuck on x86 is OS X.)

              1. Anonymous Coward
                Anonymous Coward

                Re: x86 > ARM

                "everyone here who uses Linux is doing it on ARM boxes, because "there is little sensible alternative to ARM"

                Don't be silly.

                The original sentence in full said "ARM isn't relevant in Window boxes but in any other piece of consumer or professional electronics, there is little sensible alternative to ARM."

                The chip industry isn't what it was ten years ago. x86 isn't what it was ten years ago. ARM isn't what it was ten years ago.

                Look around you, Where do you see x86 (or even MIPS or PowerPC or SPARC) in consumer electronics? You barely see x86 at all, or PowerPC or SPARC, and there's not much MIPS left.

                Time was when Intel and Microsoft were the 'goto people' if you wanted to build a DRM-encumbered content delivery platform based on "trusted computing'. Not any more. Even to the content companies, Intel and Microsoft have had it. Microsoft even lost the delivery platform for BT Vision's set top box (the Home Hub 3 is one of Broadcom's MIPS SoCs rather than one of Broadcom's ARM SoCs).

              2. Pookietoo

                Re: x86 > ARM

                "ARM has its advantages, and is overall more popular ... but doesn't come close to x86 for high end performance"

                Indeed, but we've reached the point where available x86 hardware is more powerful than is needed for web, mail, office, media consumption and other everyday uses. Media creators, programmers, scientists, PC gamers may want/need more performance, but a lot of people are seeing what they can do with Raspberry Pi, and finding that it handles many tasks with ease.

          2. Pookietoo

            Re: What about Linux or Mac? What do they use?

            In the case of Linux, a lot more than just x86, the main one being ARM of course but also PPC, m68k, SPARC, Alpha, MIPS ... I expect I've missed some.

  5. Dave 15 Silver badge

    In the good old days

    I would buy a record, tape, CV or DVD,. if the machine I used to play them broke I could buy a new machine and continue to enjoy the film/music I had purchased. It appears increasingly the case that technology companies and 'hollywood' are attempting to make it such that when the machine breaks I have to buy everything again... and the machines break after about 2 years tops because the battery is buggered.

    All in all eventually the public will just stop buying totally and rely entirely on ripped off copies on the internet

  6. Schultz

    I am not very motivated to...

    pay for the expensive development of a 'trusted zone' that will bring nothing but grieve to me. Let's hope that capitalism works as intended and I'll get the choice to buy something else.

  7. Lee D Silver badge

    It's TPM for ARM, basically.

    But, more interesting, is that this will never be the "only" device that such content is available on. As another caveat of the "analogue hole" method, if there's a single, unprotected (or poorly protected) 1080p digital content source out there, then all the rest don't really matter. Who's going to hack the 1080p on their smartphone (why 1080p on a smartphone anyway?) when they can just take the content from somewhere else?

    The thing about copyright infringement is: the original copyright infringers don't really care where the content has come from, and aren't the majority of your consumers. If anything, they are the tiniest, tiniest minority for whom even these sorts of measures aren't show-stoppers (see the linked article about the TrustZone compromises, for example - not saying that guy did it for copyright infringement, but there are people just that clever all over the net). Once a single, unlocked, digital content source exists, then a million people can copy the movie. It's that SINGLE copy ever coming about that you have to stop (without alienating your existing customers), not the casual smartphone user.

    And moving EVERYTHING to TPM, across all media, formats and manufacturers, is going to be a bit of an uphill struggle (consumers? Pah, they're happy to pay/rent on the contractual bases that already exist, for the most part, and might not like but generally tolerate things like "on X amount of devices only", etc.). For a start, it means the end of all DVD / Blu-ray -type formats, and that no one manufacturer ever makes a mistake (and, in fact, the TrustZone hack linked actually uses the Linux source code and some poor security in the code to compromise the device - which means that ANY NUMBER of similar devices might well be running that kind of code already).

    It's the next logical step in the TPM evolution. The problem is that it STILL doesn't stop "piracy". Hell, if it came to it, I'd just tap into the stream going to the LCD. Sure, it's awkward but it would work and obtain full 1080p 60fps perfect digital copies. It would mean raw frame captures (and thus recompressing them back into something MPEG-like) but that's a small price to pay if you're that determined. And once ONE guy has done it, that's game over for that content.

    I'd still place bets that the more profitable venture over the next decade or so would be to just sell un-DRM'd (but maybe tagged in some way, e.g. with steganographic subscriber name/numbers etc.) plain content for a decent price. But without something like that existing it's hard to prove that would be the case. I think things like Amazon MP3, though, kinda already proved that it would work without bringing an industry to its knees.

    1. Anonymous Coward
      Anonymous Coward

      "why 1080p on a smartphone anyway?"

      So you can use your trendy new smartphone with its decent software and hardware together with an HDMI cable to your large screen TV with its crap software and underpowered hardware. And you can upgrade the phone from time to time without having to upgrade the 42" screen. Smart TVs => dying species.

      1. Mark .

        Whilst I can see uses for Full HD on mobile devices, I don't think phones are a replacement for smart TVs.

        The last thing I want to do is get off the sofa and fiddle plugging my phone in with a cable! And then I either need a long cable trailing the living room, or have to get up everytime to change the video or volume! And what if I want to use my phone whilst I'm watching TV?

        Nor do I see the software is better - I'd rather sit on the sofa and use the remote on the TV's interface, than stand fiddling with a much smaller screen. In terms of functionality, Android doesn't come with "smart TV" functionallity - you could probably get the required stuff via 3rd party apps, but that means searching.

        If anything, smart TVs complement smart phones - I can use my phone as a remote for the TV; and I can play content from the phone to the TV without fussing with cables.

        And the upgrading issue can be solved by having upgradable parts (which some TVs allow). And if you want Android, just buy an Android smart TV USB box to put in it permanently, with dedicated remote - better than faffing with a phone.

        1. Marcelo Rodrigues
          Boffin

          "The last thing I want to do is get off the sofa and fiddle plugging my phone in with a cable! And then I either need a long cable trailing the living room, or have to get up everytime to change the video or volume! And what if I want to use my phone whilst I'm watching TV?"

          You don't have to. HDMI provides this functionality, built in the standard. Panasonic calls it "Viera Link". Sony says it is "BRAVIA Link" or "BRAVIA Sync" - and so on. I used it with my Sony mobile and Panasonic TV. I could play/pause and stop. Didn't check much more, but...

          (I)Relevant link:

          http://en.wikipedia.org/wiki/Hdmi#CEC

      2. Marcelo Rodrigues
        Boffin

        "...Smart TVs => dying species."

        I'm not sure about it. The reasoning is sound, but...

        Wouldn't it means smart TVs with more processor/memory? All in all, smart TVs are more expensive - hence, more profitable.

    2. Lord Elpuss Silver badge

      @Lee D

      "Hell, if it came to it, I'd just tap into the stream going to the LCD. Sure, it's awkward but it would work and obtain full 1080p 60fps perfect digital copies"

      The trouble is you'd have a perfect digital copy of a compressed frame (because it came from a compressed source) with artifacts and all. If you then tried to put this back into a compressed container, you would compound the artifacts and the resulting file would be measurably inferior to the original (double compression).

      I suspect that within a month or so of this technology becoming widespread the method you describe will have been done (and double-compressed rips will flood TPB) - but the pirates will not stop trying until they've cracked the original digital file. Say another couple of months work.

      1. Tony Haines
        Boffin

        Re: @Lee D

        //The trouble is you'd have a perfect digital copy of a compressed frame (because it came from a compressed source) with artifacts and all. If you then tried to put this back into a compressed container, you would compound the artifacts and the resulting file would be measurably inferior to the original (double compression).//

        While this is true for naive recompression, in theory it must be possible to regenerate the original compressed data from the uncompressed output.

        1. Lord Elpuss Silver badge
          Thumb Up

          Re: @Lee D

          @Tony Haines

          Interesting point. Reconstructing the uncompressed data from a compressed datastream is something that a lot of very clever people have spent many years trying to do (in fact that's the basis for any good compression algorithm), but I presume you're talking about reconstructing the original compression container and thereby rebuilding the original compressed file (but without double-compressing).

          I'm not aware of anybody having done this (and I've been in/around this industry for a while now) but equally I can't see any major reasons why it couldn't be done. In fact, if we could get hold of a couple of original uncompressed frames for comparison (maybe a scan from a movie poster?) then reconstructing the algorithm should be eminently doable. After that, putting the frames back in their container is a logical next step.

          Nice thinking.

    3. Suricou Raven

      "why 1080p on a smartphone anyway?"

      Google Glass, about five years from now.

  8. The BigYin
    Unhappy

    And it the W3C get their way...

    ...you'll soon need a CPU like this to be allowed to browse the Internet.

    Fuck Digital Repression Mechanisms in all their forms with the flaming stick!

  9. wolfetone Silver badge

    The V Chip

    Like the South Park version, this chip is designed to prevent you from doing naughty things.

  10. Tom 35 Silver badge

    They say it's to protect their video

    "compromised a TrustZone kernel on a Motorola device to unlock its bootloader."

    Then use it to lock you out of your own device. You can't trust the people pushing the "trusted platform" systems.

  11. heyrick Silver badge

    How will this make a dent in the fight against piracy?

    Unless the movie companies are going to ink a deal where their content will ONLY be available for devices with this particular hardware (aka the shoot-themselves approach), the obvious solution would be to find something already compromised and use that.

    It needs hundreds of thousands of phones and tablets and smart TVs to enforce content protection, but only ONE person to destroy it - is this really the best way to approach this problem?

  12. Anonymous Coward
    Anonymous Coward

    Ummm nope that doesn't work...

    ... it assumes ripping is done from the digital output, when it isn't.

    This does nothing to stop rippers reading the datastream from the source media, and breaking the encoding in memory, without ever bothering to send decoded data to screen.

    This is the problem all these DRM schemes have, they assume the purpose of the rip is to display the output, and so attempt to control the output, when in reality the rip is only about breaking the encoding to produce a usable file.

  13. TechnicianJack
    Pirate

    This system could be bypassed easily. My friend records gameplay footage of his Xbox games and uses a 'capture box' which plugs between the Xbox and the TV. Xbox -HDMI- capture box -HDMI- TV. All it does is record the sound/video stream to a hard drive as it passes through the box (In full HD). This would get around the chip stopping the software from reading the video straight from the computer. Admittedly it is extra hardware you need, but then again, the Anon who is ripping the videos is probably going to avoid buying or using hardware with the anti piracy chip in. The only way this system would work is to force manufacturers to include this chip in any graphics hardware they make. This is unlikely to happen (Unless the EU catch wind of the chip), but then you still have the problem that Mr Anon can still use his existing computer without the chip in (Unless you make it illegal to use video hardware without an anti piracy chip).

    1. Lord Elpuss Silver badge

      Yes but

      Recording the output stream (and then recompressing it) is not difficult. It's also not particularly a worry for content owners because the resulting double-compressed file is markedly inferior to the original. The trick is to capture the original bitstream before decompression - which is what this TrustZone is all about preventing.

      1. Anonymous Coward
        Anonymous Coward

        Re: Yes but

        "the resulting double-compressed file is markedly inferior to the original. "

        How can this be? Surely the industry have repeatedly assured the people who pay their wages that although the compressed data is smaller, any losses introduced by the smallering have no significant visible/audible effects? Therefore decompressing and then re-smallering can have no significant visible effects?

        You're not suggesting the industry haven't been entirely honest with the public are you?

        Btw: it's not possible to reverse engineer the (de)compression algorithm(s) from looking at a few displayed frames and their compressed equivalents. No chance.

      2. A J Stiles
        Boffin

        Re: Yes but

        That rather depends on the compression algorithm used.

        I don't think it is an intractable problem to identify, from the compression artefacts, how the original raw data was compressed. Lossy compression is just a many-to-one mapping, after all (the idea being that all members of the "many" which map onto the same "one" by compression should be barely distinguible by a human). Then applying the exact same lossy compression algorithm to the artefact-laden decompressed version ought not to produce any additional compression artefacts of its own.

        1. Anonymous Coward
          Anonymous Coward

          Re: Yes but

          "applying the exact same lossy compression algorithm to the artefact-laden decompressed version ought not to produce any additional compression artefacts of its own."

          Quite possibly so, but that's not the same as reproducing the original uncompressed data, which iirc was the earlier claim.

          1. A J Stiles
            Boffin

            Re: Yes but

            You don't need to be able to reproduce the original uncompressed data exactly; you just need to be able to reproduce the DRM-wrapped compressed data exactly.

    2. Anonymous Coward
      Anonymous Coward

      The idea around this bullshit is usually that the decoder (e.g. the console, it's CPU or whatever) will only permit a full HD connection to a view (e.g. TV) over an encrypted channel (the decoder and viewer will engage in some crypto hand-shake). And yes, this is a mandatory requirement for HDMI.

      Thus, even if the stream is intercepted, it makes no odds as said stream is encrypted (just not with H.264 or whatever). A bit like HTTPS really.

      Fundamentally, the decoder and viewer are there to service the whims of the content creators, not the person who bought the device or the license to view content on it. Further example of this is the recent patents to refuse to play a movie if there are too many eyeballs.

      Given the always-online nature of many things today, if the presence of the capture device were detected you would probably be automatically reported and then have to prove your innocence.

      In the case of your friend, playing a game is one thing; watching (and ripping) a Blu-Ray quite another.

      1. TechnicianJack
        Thumb Up

        Thanks for the info. I didn't realise the stream would be encrypted.

        1. Robin Bradshaw
          Pirate

          Re HDCP encryption of HDMI connections

          HDMI does indeed support encryption but the keys have been discovered:

          http://rudd-o.com/monopolies-of-the-mind/the-hdcp-master-key

          This is quite possibly how/why your friend magic box work.

      2. Shades

        "The idea around this bullshit is usually that the decoder (e.g. the console, it's CPU or whatever) will only permit a full HD connection to a view (e.g. TV) over an encrypted channel (the decoder and viewer will engage in some crypto hand-shake). And yes, this is a mandatory requirement for HDMI."

        You're talking about HDCP (High Definition Content Protection). There are many boxes out there that sit in the middle of a HDMI link (between the source and the actual display) which fool the source into thinking they are displays, take care of the HDCP encryption hand-shake and pass the unencrypted data onto the display device. These boxes didn't actually start life as merely a means to facilitate copying of content, they arose as a workaround to allow some early plasma/LCD TVs, which didn't support HDCP, to show content from devices which have HDCP always on (like the PS3 for instance).

        1. Anonymous Coward
          Anonymous Coward

          Well, all TVs should be up to spec so what other reason could there be for having one of these devices. Clearly they need to be outlawed, preferably under anti-terrorism legislation (and attack on IP is an attack on a country's assets).

          Once enough of these terrorists (pirates) are in jail for life without possibility of parole, we can enter a new utopian era of high quality production, export of IP and wealth.

          Oi! Stop sniggering at the back!

    3. Anonymous Coward
      Anonymous Coward

      Re: Ummm nope that doesn't work...

      "reading the datastream from the source media, and breaking the encoding in memory"

      I thought one of the many reasons Vista was as succesful as it was, was that it had anti-tamper mechanisms in the OS so that unless your app was suitably qualified with secret handshakes known only to the pigopolists (cryptographic signatures etc), it was not entitled to access the memory containing the protected data? End to end protection from BD drive or HD stream, through the 'computer', to the HDCP-supported display. You might want to look up "tilt bit".

      It worked really well for them, didn't it.

    4. streaky Silver badge

      "the Anon who is ripping the videos is probably going to avoid buying or using hardware with the anti piracy chip in"

      Doesn't actually work like this at all.

  14. Jamie Jones Silver badge
    FAIL

    DRM cannot be unbreakable.

    " Unbreakable or even tough DRM prevents "

    The point is, it doesn't matter how good your encryption is - ultimately, your viewing device has to decode the stream, so the "pass phrase" to decode it must be made available to the client.

    Even if there is only one stream - and it's for this ARM chip, you only need to reverse-engineer the protocol, and either extract the pass-phrase, or write 'emulation' code that instead of passing the decrypted stream to the decoder, it writes it to a file.

    This is quite tricky work, but ultimately, it's about "security through obscurity" - you don't need to 'crack' the encryption, just discover the decrypt key from the stream.

    The only alternative would be if the decrypt key was hardcoded in the ARM device itself - but how long would that remain a secret when world+dog starts to use the system

    1. John Sanders
      Pirate

      Re: DRM cannot be unbreakable.

      And the more unbreakable it is, the bigger the reward for cracking it.

      1. Jamie Jones Silver badge
        Thumb Up

        Re: DRM cannot be unbreakable.

        Yep! Crackers love a challenge!

  15. btrower

    Maybe force DRM out of the commons

    If we create and move to a 'free' network, we can force DRM out of the commons. Make it a condition of the 'on ramp' into the network that noxious habits such as hobbling systems with DRM are not allowed. You can't get your content into the 'free' network unless you play nice.

    Since I could write the code to do this myself, it is only a matter of time before most of us shift over to a new encrypted network that makes it impossible for bad players to use the commons. I am not that special. There are probably many thousands or even millions of programmers whose skills are greater than mine and this idea is hardly original. In fact, there is already a 'freenet' (https://freenetproject.org) and it already has a 'darknet' component. The only reason I don't use Freenet all the time myself is because "Files are automatically kept or deleted depending on how popular they are, with the least popular being discarded to make way for newer or more popular content." Unfortunately, most of what I would be interested in would be in the discards.

    Note that it is not really possible to outlaw DRM per se because at its root it is just a form of encryption and the ability to use encryption without restrictions is also necessary for a free network. What would be outlawed is the deliberate impairment of content destined for broadcast and 'cheating' by attempting to enforce over-reaching copyright restrictions.

    The only way bad players can stop this from happening is by outlawing the mechanisms necessary to build and use 'freenets' and 'darknets'. You can bet they are busy trying to do this right now, but a number of 'good' players are constantly on the lookout for such things and challenging them at every step.

    1. heyrick Silver badge

      Re: Maybe force DRM out of the commons

      Unfortunately your commons will contain trojans and viruses and child porn. Well, it doesn't need to be true, the politicians just need to convince each other of this to get them all fired up as to how it is A Really Bad Thing and locking down content with DRM helps protect children, blah blah blah.

      D'you think we got to where we are today without Big Media greasing the grubby paws of democracy? Idealism is nothing compared to FUD, spin, and cold hard cash.

      1. A J Stiles

        Re: Maybe force DRM out of the commons

        Ah, yes, wheeling out the old arguments -- "trojans, viruses and child porn".

        Those are social problems, not technological problems; and any hope of a solution lies in the social domain, not the technological one.

        If you can't make it technically impossible for someone to take over someone else's computer, then make it unattractive to do so. And if you can't stop people from looking at pictures, then make it harder for them to do anything else beyond that.

    2. streaky Silver badge
      Alien

      Re: Maybe force DRM out of the commons

      The "commons" can't work like that because the commoners want all the cool stuff that such a system can't provide because publishers won't support because they automatically think they're getting screwed - even if they aren't. Unless you want to spend your life buried in copyleft media obviously that doesn't work.

      Not for nothing but it's possible to outlaw *anything* if people in a country want that, there's even a case for damaging and platform restricting DRM in court cases if you've got the money to fight it.

  16. Anonymous Coward
    Anonymous Coward

    If man can view it ...

    man can copy it

  17. John Sanders
    Pirate

    If it can be watched and heard

    It can be copied. period.

    Having said that I buy what I like, and always download to check first. Back in the day when I could not download it was a friend lending me a tape to check stuff what lead me to buy.

    Most people who manage any disposable income behave like that.

  18. Major Variola

    Tap the LCD

    The display is receiving decoded pixels. Plug in a device to substitute for the LCD. Add internet-distribution, Game over.

    Also, the usual analog hole, without the popcorn or coughing.

  19. Neoc

    Translation...

    "That capability is built on ARM's TrustZone: a secure execution technology that allows hardware designers to temporarily shift the processor from running general-purpose software to a special ring-fenced area of trusted code that performs sensitive operations - this is software that may not even trust the OS running on the gadget"

    I may have misunderstood... but from what I can gather this would also require one hell of a firmware upgrade if CODECs changed, since the only codecs in the "trusted code" section would be those approved of by the MPAA *at the time of manufacture*. Because, of course, if I can easily flash the "trusted" firmware, I can easily disable it.

  20. streaky Silver badge
    Boffin

    Unbreakable..

    Ignoring the obvious Samuel L Jackson quotes:

    Does it stop you reading the panel pixel driving data from that little cable... *points at a ribbon that makes it's way into the back of the digital 1:1 panel that sends an unencrypted signal to drive the pixels*... There!

    ... think you might have missed a spot. Ooops.

    1. A J Stiles

      Re: Unbreakable..

      Yeah, back in the days of CRTs, this was my DRM-busting idea: A device that looks like a sawn-off CRT neck, replaces the picture tube in the TV set, and records the RGB stream from the grid drives together with the timing from the scan coil drives.

  21. Henry Wertz 1 Gold badge

    Not a big deal...

    So, if this is like the decoding on the older ARMs, it runs code on a DSP built onto the ARM that does the video decoding at almost 0 CPU load. so the lib that manages the DSP code is locked from access, and the video memory the video is written into would presumably be blocked as well. Nothing too draconian and invasive, I have not seen any Android app that would capture video like that anyway.

This topic is closed for new posts.

Biting the hand that feeds IT © 1998–2019