[mythtv] Solution to CableCARD problem?

[Tim Moore]

> > So, the solution is not to make decryption part of a MythTV system,
> >
> 
> Sort of like the old "budget" DVB cards, that had
> their own MPEG decoder and video output on them,
> except that they would also have their own disk
> space for recordings?

I'm not familiar with those, but sounds about right...

> 
> 
> > but to make MythTV a control solution for a bare-bones DVR that is
> > installed inside the PC chassis.
> >
> Or even outside it, if you are talking about the
> MythTV PC having to superimpose OSD, EPG, et c.
> It could just be an external FireWire STB, with
> the HD video output looped through whatever
> special card does the video overlaying.

Yes.  The difference between a standard STB and one of these cards is that
there is an electrical interface which provides a much more enhanced API
compared to trying to remotely control a device that was never intended to
be integrated with.  We should be able to query the content and status of
the device quite easily.


> 
> 
> 
> The other "snags" I see are:
> 
> 1) Disk space is fixed and finite. Once this CableLabs
> DVR "walled garden" fills up, you are either forced to
> delete things, or to change the hard disk and start over.

Yes, that is a limitation.  I don't yet know what the maximum disk size is
that these Broadcom cards support.  I believe that it will be a tolerable
limitation for most people though, and one that will be shared across most
other DVR systems.

> 
> 2) Doing the video overlays might be quite hard.
> It is possible that the a DVR chip allows write
> access to an overlay buffer, but unlikely.

Yes, there may also be complications with copy protection in the signal that
could prevent an external graphics card doing the overlay.  This could be a
serious problem if so.  Further research is obviously required.  I'm trying
to get some more detailed specs of the cards.

> 
> 3) What happens if, a few years down the road,
> CableLabs changes their mind?
> (e.g. DRM cert 5.2 says users are not
>        allowed to keep recordings at all)

True, but the best we can ever do is live with what we know.  This approach
could provide a current solution to the problem.  We might even find that
the FCC forces CableLabs to reduce/drop the certification charge when they
realize that it is working against their intention of the CableCARD scheme
(we can hope!).

There are some unknowns that could hinder this solution, and I will endeavor
to find some answers and report back.

Cheers,
Tim