Originally Posted by gsr
Further, when you copy that copy once content from your DVR to your DVHS VCR, what probably happens is similar to what W7 Media Center is doing: the DVR decrypts what it wrote to disk and sends it via firewire to the DVHS VCR (with the 5C flags maintained). The DVHS VCR recognizes the flag and then re-encrypts the content with a key that identifies that VCR - this would presumably prevent you from taking the tape and playing it on a different VCR.
This is a very interesting concept, that hadn't occurred to me before.
In light of the above discussions, and the story about Playready DRM encryption applied to the recorded copy-once WTV files (to the decrypted DES stream, probably using a key provided by the cablecard), it could certainly be very plausible that the JVC VCR is re-encrypting while it is recording to DVHS tape, using its own key. Actually I've had no problems playing DVHS tapes of "copy-once" content from TWC (offloaded from DVR, from these "copy-once channels" (meaning just about everything) on my other JVC DVHS machines. I simply could not play the on the Panny HD1000.
A long time ago, I had noticed that and written it off to some kind of electronic recording tape difference in the two VCRs. But the current discussion opens my eyes to what you're describing here and which sounds very reasonable, that the JVC VCR is doing its own encryption... as triggered by the "5C copy-once" flag. And the HD1000 simply has no decryption capability at all, much less the decryption key for the JVC family of machines. I don't know if the JVC-recorded tape could be playable on a Mitsubishi DVHS machine, but it certainly won't play on the HD1000.
Now, I'm still not certain about the decrypted data sent out the DVR to the VCR via firewire, even if it's re-encrypted by the receiving VCR on the recording. That firewire path can be relayed through any number of 2-port firewire devices along the way, with each one simply responsible for passing on the data stream. Any one of them could be a "digital pirate" (and of course this is precisely where an obvious vulnerability exists and would be latched onto, if the data on this out-in-the-open cable were actually non-encrypted). So I honestly cannot believe the data here is not encrypted.
Of course if that were the case, then unless the decryption key were already burned into the JVC VCR, how would that data get decrypted if it arrived at the VCR encrypted? This is an unanswered mystery at the moment, but I still cannot believe they would allow an unencrypted digital data stream of otherwise protected content out over firewire. Hey... you've got all kinds of Apple hackers who know firewire!
Maybe there's an opening handshake of some kind between the source and target device, where this key is somehow passed discretely. Again, the intermediate relay devices shoots down actually sending the key over firewire, but if the data is encrypted then the key must be somehow known by the receiver in order to decrypt the data (unless there was a manufacturer's agreement in which the keys were burned into the compliant devices).
I'm pretty sure the cable company DVR is writing a re-encrypted copy to the hard drive.
Ok, I'll buy that. The discussion about what WMC does, and Playready DRM re-encryption has convinced me that this is a reasonable idea.
I just honestly haven't bought in to sending the decrypted version out over firewire, having gone to such great lengths to encrypt it for protection while it was first arriving on the coax from the head-end, and then when storing it on hard drive as a recording. It just doesn't make sense to open it up on the firewire interface. I don't buy it. That data has to be encrypted.
So when you record a copy once program on it, it gets encrypted and the tape will play on VCR's that do the encryption / decryption thing, but not on VCR's that don't. I suspect the tape will probably play on other DT100U's or other VCR's that are fully compliant with all the protection crap. Basically, it's doing the same thing Windows Media Center does.
Again, as I agreed above, this now is a new concept to me... but one which I will subscribe to. Perhaps ALL DVHS VCRs use the same key if they subscribe to the 5C-compliant law, so that one of these tapes recorded on a JVC DVHS machine COULD be played on a Mits DVHS machine (both being 5C-compliant), but could not be played on the Panny HD1000 because it is not 5C-compliant and thus cannot decrypt the recording.
If you get a chance to run these additional tests, the results might prove to be interesting
Actually, I believe I have previously tried to use the Panny HD1000 as a recorder, connected to the DVR via firewire.
And as I recall it didn't work. But trying to analyze it now, maybe it was more an encryption thing... assuming the DVR is providing encrypted data over firewire as I suspect it is. If it's not caused by that, then maybe the DVR just refused to give data to the HD1000 because it determined the VCR to not be 5C-compliant.
I think that per your suggestion I should just run this test again... this time carefully trying both copy-freely channels as well as known copy-once channels, and see if there are any differences in the results.
There was one very interesting line item in my earlier tests that is very relevant here, and that is that when I had the HD1000 connected to the DT100U via firewire, and played a copy-once recording (made on the DT100U) on the HD1000 as the source device feeding the DT100U, that the video output from the DT100U on my HDTV was perfect. However the very same copy-once tape when played on the HD1000 connected to its "master DST50 receiver" (where the video outputs live) would NOT play successfully. This is most rationally explained by encryption on the tape (by either the DVR which provided the data or the JVC which did the recording).
In other words, as a playback->delivery-over-firewire system, the HD1000 simply read the data (which we feel was encrypted) and passed it out over firewire to the DT100U, disregarding its 5C copy-once flag. The DT100 received it and decrypted it successfully for display, knowing the proper JVC decryption key since it had also been used for the original encryption.
Note that the DT100U would DISPLAY the "5C copy-once" data stream delivered over firewire from the HD1000 just fine, applying the decryption key without complaint. But as soon as I pushed the RECORD button on the DT100U it threw up its hands and refused, obviously obeying the 5C flag... as it should.
But when the HD1000 sent the data over firewire to the DST50 receiver, it was useless and would not display, because the DST50 did not have the requisite JVC (or maybe all-DVHS manufacturer's) decryption key.
My other earlier test, using a copy-freely tape made on the DT100U, played without a problem from the HD1000 through the DST50 and on to the HDTV.
What I really need to do now, is re-try the recording from DVR to HD1000, both for copy-once and copy-freely programs, and see what happens. The earlier tests used tapes that had been recorded on the DT100U. I need to see if recordings from DVR to HD1000 can actually be made at all, and if so what are the (a) direct playback results, as well as (b) playback through the DT100U results.
But, I think this discussion about WMC functional blocks, and when/where the decryption and re-encryption occurs, etc., I think it's very worthwhile and enlightening. Definitely clarifying to me, anyway, and I hope others following along.
The key to "copy protected" from WMC does seem to most likely be triggered by just the "5C copy-once" flag.
And if that's on, the WTV recording is ALWAYS Playready DRM encrypted (regardless of whether the original datastream was itself clear-QAM or DES encrypted) using the Win7-provided key (after first decrypting the incoming DES encrypted datastream if necessary, using the cablecard-provided key I would guess). Period.
Pretty sure that makes sense now... except for the mystery about data from the DVR to the DVHS machine over firewire, and how it could possibly be allowed to be non-encrypted. The other side of the coin is that if it is in fact encrypted, then where is the key provided to the DVHS machine to decrypt it (unless again, it is part of a manufacturer's agreement).