[mythtv-users] HD-3000 recording corruption

[Steven Ihde]

On Wednesday 29 March 2006 07:22, Craig Partin wrote:
> This is the first time I have been stumped in over two years of mythtv
> use.  I got my HD-3000 working just fine tuning QAM stations over
> cable in myth.  The system works flawlessly and reliably except that
> QAM recordings get trashed when anything else is happening on the
> system.  It records beautifully when nothing else is happening.  This
> happens on all channels, HD and SD.  I can playback 1080i content with
> XVMC just fine, even with bob.

I've been debugging a similiar problem with my HD-3000.  In my case it appears 
to have something to do with the video card.  When I play video (with 
anything -- xine, mythtv, mplayer), any QAM recording in progress will get 
trashed.

The interesting thing is that it doesn't appear to have anything to do with 
data transfer to the hard drive.  Rather the HD-3000 actually loses the 
signal lock.  I can verify this by shutting everything down (mythbackend, 
etc.), and running "azap" to tune a channel.  As azap prints status lines, 
they all show excellent signal strength in the range fd00-ff00, and 
FE_HAS_LOCK.  As soon as I start playing any video with xine, the signal 
strength drops, and FE_HAS_LOCK disappears.  The larger the xine window, the 
worse the problem.  If I shrink the xine window down to a tiny size, 
FE_HAS_LOCK comes back and the signal strength improves.  The amount of 
signal loss seems directly related to the size of the video window.  It's not 
related to data transfer that I can see -- it behaves the same whether I 
simply tune the card without recording, or if I use "azap -r" and 
"cat /dev/dvb/adapter0/dvr0" to record.

My card is an nVidia 6200 AGP.  I switched back to my old Radeon 9200SE, and 
found that it did not have this issue!  Signal strength stays essentially 
constant even when playing video, and the QAM recording is crystal clear.  
However I can't really use the Radeon because of the TV-out fiasco with 
Radeon, and besides my CPU is too slow to do the job without XvMC, which the 
Radeon doesn't support.

Anyway, I don't have a solution, but since elsewhere in the thread you 
indicated that the PCI latency fix didn't work for you, I wonder if you might 
be having a similar issue to me.

BTW, my mobo has a Via KT600 chipset.  I'm planning on upgrading the whole 
system to a mobo with an Intel chipset and a CPU that can do the job without 
XvMC at some point.

-Steve