You would've thought people would have worked this one out by now.
Who fancies running a HN pool on how long until the root keys are public? (I'm serious - how funny would that be. We just need someone with enough trust from the community to handle the money. And someone who lives where online gambling is not regulated...)
And its not that tech runs just from Intel to AMD.
x64 is AMD tech that got adopted by Intel. Monolith multi-core technology is also something Intel adopted from AMD.
Intel wouldn't dare destroying AMD. Since AMD got created for sole reason to prevent Intel from being torn apart. But Intel will do everything to keep AMD's market share somewhere comfortably below 20%.
And it remains entirely at Netflix's discretion whom they trust. I agree it's a Bad Thing, but the content industry is entirely predicated on artificial monopolies.
This is actually the problem with Netflix on Android - Android doesn't have any DRM standard, so it's up to the hardware manufacturer to include it.
My point still stands though - there is absolutely no way this doesn't get cracked.
As long as watchable images are displayed on a screen, someone will figure out a way to "archive" them for later.
Presuming this is the case, how would key management happen? Would every chip have its own unique keypair? If not, an attack similar to the one linked to here by JohhnieCache would render this scheme useless. If each keypair is unique, how would a particular content provider know that a public key was actually one paired with a private key only contained in Intel's silicon? Would Intel provide a registry of all valid public keys to content providers? If there is no such central registry, would the public keys themselves be signed by Intel so that content providers could be certain of their origin?
Please correct any wrong assumptions on my part. I know little of cryptography compared to many here, but I am frustrated by how little technical commentary has accompanied the articles I have read about this new processor family.
[0] Mobile "broadband" doesn't count. In most places, it is entirely unsuitable for streaming high-def video and will be for several years.
My TV provider offers films on demand at £2-4 per viewing. The convenience of being able to watch immediately, and cheaply outweighs the choice of buying the DVD/BR and being able to watch multiple times.
Quick cost comparison:
1) Stream it for £4. Film was rubbish. Total cost = £4.
2) Stream it for £4. Film was great. Wait a few months until price falls. Purchase DVD at discount for £5-£10. Total cost = £9-£14
3) After release purchase for £15. Total cost = £15.
Of course this depends on the economics of your specific location and the cost differences, but personally I think I spend less in total on films because I very rarely buy physical media until the prices fall.
Streaming to a PC fits a very similar market profile, the only difference being it caters to those who build their entertainment system around a PC rather than a regular TV, and I think the lines between those as two separate systems are going to increasing blur in the coming years.
Right, that's why Netflix Instant is such a failure.
The DIVX revolution has been killed.
My guess is that CPU microcode designed to detect that h264 decoding being offloaded to the GPU was from a Usenet post instead of from Hollywood would be difficult, if not impossible, to implement. (My point: the CPU doesn't even see bits from the video to decode anymore. So it probably can't fuck them up, even if it wanted to, which it doesn't.)
If a human can view it, you can always, always, always copy it, reproduce it, and share it. No matter how much HDCP you throw at it, no matter what specialized hardware you devote to it...in the words of Mr. Universe..."you can't stop the signal."
(And all the King's horses, and all the King's men, couldn't put the industry's shitty paranoid ineffective wannabe DRM back together again.)
> For Alex, the impossibility of making digital information copy-proof is a central truth of our age: something to be explained, and then re-explained, to judges, reporters, and businesspeople, in amicus curiae briefs and interviews on NPR. For me, it follows from the fact that the set of n-bit strings constitutes an orthogonal basis for Hilbert space.
1/24/2007 10:13 PM Defeat any copy protection on video/audio: play the content on a certified software player in a virtual machine, copy material from the virtual screen/loudspeaker.
This is night and day different from what Hollywood wants to do, and what Intel is trying to enable them to do: force consumers to buy physical discs, or force consumers to rent from a hardware-enforced DRM'd digital kiosk.
And even so, this is just another DRM tool that Netflix will use, unless they opt to only use Intel's built in DRM (which would be strange given the wide competition Intel's hardware DRM faces from TPMs and the like.
Is Intel supposed to just implement it for them or something?
Or are they supposed to get to their own terms with them studios?
Especially since these are completely different markets who do most stuff differently. Thats what I wanted to point out - if they want their chips to play DRM content they'll have to negotiate it themselves with the Hollywood. I also strongly suspect that this is not an exclusive pact.
Since nowadays content producers are very well aware that more and more (probably the majority?) of content is consumed via - non x86 architecture devices.
Note the ready availability of HDCP-strippers, for example.