[geeks] Re: [rescue] (no subject)

[Brian Hechinger]

> There are two problems with that plan.  First, there is the issue of
> listings.  The second is the issue of quality.  The TiVo uses some
> seriously nice hardware chips for doing it's video work.  Getting good
> decoding hardware, isn't hard, but encoding hardware tends to be
> expensive.

ok, we are WAY out of my realm here, but what about a SunBlade 100?  does
one of them have enough power, or could have a nice enough PCI card put
in it to work?

> One possibility would be to make it a dual proc machine with a lot of
> scratch space, then record mjpeg to disk, and have the second CPU pretty
> much dedicated to converting it to MPEG2.

sounds icky to me.

> The other issue is looks.  A TiVo look decent.  A PC based PVR is a PC in

that's why i like the SunBlade idea so much, they wouldn't look to terribly
out of place in the living room.

> the living room.  Bleh. Now, maybe an FDDI based video distribution system
> would be cool.  Stream video from netapp to TiVo, and vice versa.

this was my backup plan.  those TiVo boxes with networking, can they use the
network for storage (NFS or something)?  or is that just an effort to turn
yet another box into a "regular" Linux machine?

if so, is that a possability?  *my* thought was to build an imbedded device
with IDE on one side and FastE or FDDI on the other end.  emulate an IDE
drive and store everything on an NFS server.  now THAT would rule.

4TB TiVo anyone? :)

-brian