[mythtv-users] Time Warner & Firewire

Mark Knecht markknecht at gmail.com
Mon May 16 21:59:42 UTC 2005


On 5/16/05, Nathan Poznick <kraken at wang-fu.org> wrote:
> Thus spake Mark Knecht:
> > Hi,
> >    I've been following this thread sort of remotely. I'm an old 1394
> > guy who worked on a number of the original IEEE specs for that
> > technology so I have an interest. Besides I'm curious about what
> > results I might get here in the Bay Area.
> >
> >    Do you know anything specific about what is and what is not
> > actually on the 1394 bus? Is anything being transmitted? Has anyone
> > had an opportunity to stick a bus sniffer on the cable? Or is it
> > possible that the set top box is doing 5C content protection and the
> > Linux 1394 stack isn't responding appropriately?
> 
> I can provide a sample transport stream that it dumps if you'd like,
> along with the dvbsnoop output.  It appears that it's sending valid TS
> packets, but they don't contain video data.  I only get a few hundred kb
> worth of data when trying to record HD.  My guess is that it's trying to
> determine if the receiver is 5C enabled.

Exactly. Depending on the way the flags are set in the content the
transmitter is required to understand who it's talking to before
sending certain copyrighted material. One of the general requirements
of 5C is that base data (uncompressed and unprotected) cannot be
present anywhere in a system where it could be copied digitally.
(Except for the 'professional equipment' exemptions.).

Back in the days when the 5C specs were written and first introduced
to the community as a whole it was clear from Microsoft in the 1394b
committee meetings that they would not support applications on their
platform that would have access to unprotected 5C content. They pushed
pretty hard at the time to make sure that all hardware (and software)
decoders stayed on specific cards within the system to ensure that the
content remained protected.  I suspect that you know this already.

My thought here is that the Linux platform may be *too* open to meet
the basic requirements of 5C. (I'm assuming 5C or some form of it is
still the overriding specification as to control of copyrighted
material on the 1394 bus. I've been away from that for a couple of
years now.) If 5C is still what people are using but the Linux
environment handles software decoding of the material then the two may
be in conflict with each other and the set top box never sends any
data other than the strings required to set up the public/private
keys.

It would still be possible for a set top box to send fully digital
non-copyrighted or copy-freely data if it chose to but I don't know if
the cable folks are *required* to do that. Doing so may be optional.

Again, this is all just supposition and should not be acted upon in
any way as it could be completely bogus by now. This was just more for
my own curiosity and also to *possibly* save you some time trying to
find something you're never going to find anyway.

With best regards,
Mark


More information about the mythtv-users mailing list