[mythtv-users] HD-PVR: Encoding Errors Running 0.22

[David Engel]

On Fri, Nov 27, 2009 at 01:11:11PM -0500, Jeremy Thornhill wrote:
> I believe the 'device not running' stuff is MythTV's signal monitor
> detecting that the HD-PVR hardware decided to stop recording. I don't
> pretend to know the details of why, but it seems that sometimes the
> hdpvr just mysteriously fails, and when mythtv detects this it tells
> the HD-PVR to start recording again.
> 
> So, those 'gaps' that people are seeing are spots where the hd-pvr is
> cutting out. Once mythtv restarts the device, the recording picks back
> up.

I thought everyone already knew this, but just in case...  Mythtv has
a workaround for the problem for the problem when the hdpvr stops
recording.  When mythtv hasn't received data in some period of time,
it assumes the hdpvr stopped and restarts the recording.  If mythtv
didn't do this, the recording would end at the firts gap you see, the
backend would be wedged affecting later recordings and there would be
a lot more unhappy hdpvr users.

> I'm horribly perplexed by my situation though. My device seems to just
> arbitrarily fail to start recording properly - once it's started, it's
> fine. The first time I start recording with it after a boot, it's
> fine. But after that it's a total crap shoot. Sometimes the next
> recording is fine, sometimes it will fail 2 times and then start,
> sometimes it will fail 20 times in a row!

This sounds like an overheating problem...

> Even more bizarre, if I rmmod/modprobe the hdpvr module, it will
> always record properly immediately thereafter.

...and this sounds like a driver problem.

> I'm at a loss. Is my hdpvr defective? Perhaps something subtle is
> wrong with my kernel config? The debug messages from the hdpvr module
> tell me almost nothing. I'm wondering if I could patch the mythtv
> source to rmmod / modprobe the module when a signal error is detected,
> as that seems to almost always resolve my problems.

I don't know about your hdpvr, but I doubt it's your kernel config.
Yes, the driver debug is so terse it's of little use, at least for
this problem.

My guess is it's one of two things.  1) It's a firmware problem.  Has
anyone been brave enough to try the new 1.5.6.1 firmware yet to see if
it improves anything?  2) It's a subtle driver problem that doesn't
affect the driver developers, and consequently, is not getting enough
attention.  John P Poet is doing what he can, but the original hdpvr
driver developer has been conspicuously abesnt from this whole
discussion.

David
-- 
David Engel
david at istwok.net