[mythtv] Current DVB problems rundown

John P Poet jppoet at gmail.com
Sun Sep 11 00:20:49 UTC 2005


On 9/9/05, Daniel Kristjansson <danielk at cuymedia.net> wrote:
> On Sat, 2005-09-10 at 05:01 +0200, Kenneth Aafløy wrote:
> > Lørdag 10 september 2005 04:37, skrev Kenneth Aafløy:
> > > Lørdag 10 september 2005 04:32, skrev Daniel Kristjansson:
> > > > I've talked to John Pullan about adding something like this to
> > > > the setup program as an "optimize frequency of each transport"
> > > > process. Though we would probably use a binary search in the
> > > > range on each side of the center frequency and then scan that
> > > > range at regular intervals, with time to tune as used to shape
> > > > the response curve estimate. With a lookup table to avoid
> > > > redundant tunes of course.
> > >
> > > That's a good idea.
> >
> > Since you won't respond, I'll do it myself.
> I'm not at the computer all the time.. esp after 7pm EST.
> 
> > What you are talking about is the zig-zag scan
> > that is incorporated into the linuxtv-core.
> Is there any way to reliably disable this feature ?
> That might be useful if we are trying to determine a better
> starting frequency. Not all frontends support getting the
> actual tuned params, plus I'd like more control.
> 
> 

This theory into my problem sounded really good.  Unfortunately I
don't think it is the cause.

The or51132 module defines:
step_size = 0;
max_drift = 0;

Those values set to zero effectively disables the zigzag scan.

Kenneth, do you have any other ideas?  I will keep digging into the
dvb drivers, but this is pretty new to me.

Thanks,

John


More information about the mythtv-dev mailing list