[mythtv-users] Max number of DVB channels per usb tuner

[John Pilkington]

On 22/12/13 06:36, Mark Perkins wrote:
> Thanks everyone for the great responses. Am looking into it now and it
> looks promising.
>
>  > dmesg | grep dvb
>  > [ 10.662763] usbcore: registered new interface driver dvb_usb_af9015
>  > [ 10.663089] usb 1-3: dvb_usb_v2: found a 'TwinHan AzureWave
>  > AD-TU700(704J)' in warm state
>  > [ 10.880631] usb 1-3: dvb_usb_v2: will pass the complete MPEG2
>  > transport stream to the software demuxer
>  >
>  > I use that device only when the PCI cards are already in use. Project-X
>  > often reports packets out of order from it, and also blocks of this:
>  >
>  > [202034.369192] usb 1-3: dvb_usb_af9015: rc query failed=-71
>  > [202034.516313] usb 1-3: dvb_usb_v2: 2nd usb_bulk_msg() failed=-75
>  >
>  > So... some are more equal than others.
>  >
>  > John
>  >
>
> John, I get very similar messages from all my USB tuners. The rc query
> failed and usb_bulk_msg failed have always given me concern but I have
> never been able to put my finger on a problem that I can attribute to
> the errors so have let them go for now (that and I haven't been able to
> fix them...)
>
> But I have also often noticed that when fine tuning commflag markers and
> moving frame-by-frame that the picture often seems to jump backwards /
> forwards in an odd fashion. Again nothing that I notice while actually
> watching recordings but again has given me concern that all is not well.
>
> And then recently in digging into other problems I have noticed that
> when doing fifo transcodes with verbose all that mythtranscode seems to
> be continuously reordering packets.
>
> I wonder if this all ties together to suggest that I have a deeper, more
> intrinsic problem with my usb tuners?
>

Assuming that your WinFast USB device is the af9015-based one listed 
here, http://www.linuxtv.org/wiki/index.php/DVB-T_USB_Devices,
it suggests that that driver, or maybe the USB handling, isn't quite 
right; but the driver has been around for a long time now.  I don't 
remember seeing the 'failures' until quite recently. I haven't raised 
concerns in linux-media or done serious searches around it.  It wouldn't 
surprise me if the batch of new-kernel commits that break the PCTV 290e 
is also found to affect others, and muddying the waters at this stage 
may not help.  I might try swapping devices between systems, but at 
present neither has a serious problem and schedules are getting full :-)