[mythtv] diseqc

Yeasah Pell yeasah at schwide.com
Fri May 26 17:30:20 UTC 2006

>I think it is safe to assume that the device is not in timbuktu
>while mythbackend is running. DVBChannel is in the closed state,
>it is not deleted, when functioning in this mode; so you can keep
>the state there. I figure if anyone is using other programs with
>a MythTV device while mythbackend is running is playing with
>fire anyway... The backend machine shouldn't be restarted too
>often normally. When it is, it's good to assume nothing so that
>restarting the backend can function as a full reset.
Just so you know what we're talking about here (I'm not sure if you have 
a rotor and you might not know how fast they turn generally), worst case 
motor movement time assuming 2 degrees/sec would be 75 seconds, though 
you can cut that down somewhat for any given destination (as low as 37.5 
if you want to move to 0 degrees). I doubt anybody has a 150 degree view 
with active positions at the extremes, but you're talking 37.5-75 
seconds if you assume the worst case -- and some motors are slower than 
that even.

I'm not sure exactly what the signal monitor will be doing with the 
approximate position information. Presumably we'll want to display that 
on the OSD (several people seem excited about that idea), and presumably 
it will be used as part of tuning decisions as well. As long as the 
signal monitor doesn't do anything that actually prevents tuning of a 
signal for the 37.5-75 seconds that the position is unknown, I'm fine 
with leaving the persistent position feature out.

It might make sense for the signal monitor to treat a motor with unknown 
position as if there were no motor at all, thus falling back to the same 
behavior that exists right now during that window of startup?

More information about the mythtv-dev mailing list