Is anyone trying to crack the device itself?

[n2kwo]

The huge bottleneck in this whole process appears to be that it 
takes forever to get the file out and then to convert it.

It would seem to me that if you could re-direct the filesystem to a 
network drive that most of the battle could then be fought on home 
ground. And that two units pointing to the same network resource 
would then see each others movies (record in one room, play in 
another). Although the storage limit would not necessarily go away 
(it is most likely a function of the Tivo code, not Linux) it would 
be very advantageous to have the files in a place where you could 
move them around very quickly.

To that end, has anyone tried to de-compile parts of the Tivo 
code? Has anyone tried to identify the ports that control the 
sampling chip? Or the Tuner, to select a channel?


I know there is a lot there. There must be hardware to generate 
the on-screen images, hardware to get IR commands, hardware to 
control the tuner ( or DirecTV ) and hardware to do the MPEGII 
compression (and de-compression for DirecTivo). And ports/commands 
to write too in order to make them do what you want. But it would 
seem that the REAL goldmine is either in re-directing the software to 
utilize a networked drive or re-engineering some/all of the code in 
it to utilize a normal Linux file system and store the recordings in 
a more easily manipulated format.

Has anyone replaced the processor with a more powerful one? That 
would allow you to run a cron job on the device and either feed it 
out to a network device as it is being created (and while you are 
away or otherwise occupied) OR to convert it and save it out in a 
more desirable format.

Go ahead, flame me. But I just had to ask.