[mythtv-users] TiVo (S1) as a MythTV Frontend

Brian Wood beww at beww.org
Tue Feb 6 11:25:02 UTC 2007

On Feb 6, 2007, at 4:11 AM, Tortise wrote:

> If I may Brian?
> Re: How do you plan to capture video? You need a device that has  
> Linux  drivers available and there is no such thing in a TiVo, nor
> is there  any obvious way to connect one or talk to it if you could.
> Is there a need to capture video in a TiVo/Myth Front-end?  TiVo's  
> can capture video, it may be possible to stream this to a Myth
> Server however I wouldn't see that as a development priority.  The  
> TiVo would only need to interpret / display a stream from the
> server.

Good point, I was thinking of a FE/BE. Sorry.

> From what you have said it seems clear that X is not good to run on  
> a TiVo front-end.  I note Pluto uses a cell phone front-end to
> control myth.  Perhaps there is a thin control version that could  
> run on the TiVo, along with displaying appropriate streams from
> the server.  Clearly the TiVo has the hardware to display digital  
> streams at PAL / NTSC levels as it does do it. (well)
> Re: And your "exception" is a huge one, as there is no PCI/AGP/PXI- 
> E or  other way to connect a supported video card, nor is there
> enough CPU  to run even X on it even if you could.
> I would not see the need for a VGA output, so long as there was  
> composite / RCA output (+/- RF) outputs.  I wonder if I was clear
> enough in my original proposal?

I was just thinking that you;d need some sort of video output device  
if you couldn't use the internal one, not necessarily VGA but what  
device would be available without major surgery to the board, except  
perhaps a USB device with all its limitations. I think the Series 1  
is USB 1.1 and the Series 2 is USB2. 1.1 would never handle video,  
and I doubt 2.0 could do it very well with the existing devices/drivers.

I've looked into USB video output devices for the NSLU2 but anything  
beyond text modes seems out  of the question at this point.

> Re: I wouldn't say it's impossible, but it's certainly impractical  
> and if  you could manage it you would have so much invested in
> time and   special equipment (surface-mount soldering gear for one)  
> as to make   it of only intellectual interest.
> This may of course still be true, however it is good to explore  
> this issue.  If it keeps coning up then that might perhaps suggest
> something like it is not going to go away?  If it can't be done one  
> way then it is good to look for other ways to achieve the same
> end.

OK, upon reflection I have this plan to accomplish what you want:

Step 1: Remove all those "parts" from the TiVo.

Step 2: Install a MediaMVP in the case.

I think this would be faster, cheaper, would look the same and would  
almost certainly work.

What I'm saying is that all the arguments against a TiVo as a  
frontend could apply to the MVP, so why is that working fine?

It must have to do with the info being available for the Video output  
chip(s). There's something for you to look into :-)

More information about the mythtv-users mailing list