On Saturday 04 November 2006 17:51, David Engel wrote:
> On Sat, Nov 04, 2006 at 09:23:20AM +0100, Hans Verkuil wrote:
> > Hmm. Just checking: you are using MythTV 0.20, right?
>
> MythTV SVN as of noon yesterday (Nov. 3).
>
> > Are you using a PVR350? If so, make sure you do not use the TV-out.
> > It's not yet supported (I'll be working on that today).
>
> One of the devices is a 350, the other is a 250. Both are fairly old
> (>2 years). I'm not using the TV out. I see the same behavior with
> both devices.
>
> > Are these recording made while watching LiveTV? (I've been testing
> > with that and the recordings look fine to me).
>
> Scheduled recordings.
>
> > If you use the vbi test tool, is the CC correct? If you
> > run 'cat /dev/videoX >/dev/null' at the same time, does the CC
> > remain correct?
>
> Even without catting the /dev/videoX device, I see missed characters.
>
> > Regarding the VIM_RST messages: can you edit ivtv-irq.c, search for
> > VIM_RST and uncomment this line:
> >
> > //ivtv_vapi(itv, CX2341X_ENC_REFRESH_INPUT, 0);
> >
> > Try again and see if it makes a difference for CC. I think not, but
> > CC remains difficult to test for me.
>
> There might be slightly fewer characters missed with that line
> uncommented, but not enough to make a real difference.
Can you mail me an example of the vbi output? I assume that the
SLICED_VBI_PIO define in ivtv-irq.c is set to 1? (It should be).
Can you also do a test with the SLICED_VBI_PIO define set to 0 (again
just test with the vbi tool, nothing else).
Thanks,
Hans
_______________________________________________
ivtv-devel mailing list
[email protected]
http://ivtvdriver.org/mailman/listinfo/ivtv-devel
