[pvrusb2] only few informations for udev available
Mike Isely
isely at isely.net
Sun Sep 20 17:45:44 CDT 2009
On Sun, 20 Sep 2009, Martin Dauskardt wrote:
> > Try the 20090919 snapshot just released and tell me what you think.
>
> sounds great!
>
> I am currently using a recent v4l-dvb snapshot. I think mixing it with modules
> built against the kernel sources will result in unresolved symbols.
> Could you integrate these patch in your hg repository?
I have every intention of doing just that; you've just gotten ahead of
me.
In the mean time you should be able to still safely build the standalone
driver against a recent v4l-dvb snapshot. Instead of just putting
KDIR=kernel/source/path on the make command line for the driver, also
include V4LDIR=v4l-dvb/snapshot/path on the same command line (you'll
need both). This will cause the standalone driver to be reconfigured in
a way that makes sense for the v4l-dvb snapshot you are using.
This method is a bit limited; since v4l-dvb snapshots are not versioned,
this is only intended to work for reasonably recent snapshots, i.e. I
keep making changes to the standalone driver's config for this to adapt
to the most recent changes in v4l-dvb, so trying it with an older
v4l-dvb snapshot might still not work.
>
> > Sorry this took so long. It's been a distracting summer. Hopefully
> > soon you'll see me knock out a few other niggling issues...
>
> I am anxious :-)
>
> I now also have a HVR 1900. You may remember - I asked some time ago the
> possibility for an "analoge-only" mode. Well, that is not necessary. I can
> use the adapter_nr and set this to a high number. vdr counts all found dvb
> devices and stops its search once a number is not used.
> This way I can use the analogue part. I noticed that channel switching is much
> smoother compared to the older PVRUSB2. I takes a litlle bit longer, but it
> blanks out the video and seems to start perfectly with an i-frame.
>
> I no longer close/re-open the pvrusb2 devices for channel switching. You know,
> I made this to stop and start the encoder, but too often I get encoder stuck
> errors, black screens and freezes. Now I only stop the read thread, tune and
> start the read thread again. There are still a few artefacts visible (encoder
> ist not cleared) and there is a loud audio crack, but this way the PVRUSB2
> works stable.
> The HVR 1900 does not have these problems. What is the difference? An other
> digitizer?
If you're comparing the HVR-1900 against a 24xxx device, the differences
really amount to just the additional stuff to support digital mode. In
other words, the HVR-1900 is essentialy a superset of the older 24xxx
device. If you're comparing the HVR-1900 against a 29xxx device, that
older 29xxx device used a different digitizer & audio processing chip -
saa7115+msp3400 for the 29xxx vs cx25840 for the 24xxx and
HVR-1900/HVR-1950 devices.
There have also been driver improvements that further stabilize stream
start / stop, but IIRC the majority of those changes were a
pre-requisite before hybrid devices could even be considered and it all
happened about 2 years ago (roughly the 2.6.22 kernel timeframe).
>
> By the way: Is there still analogue cable TV in the USA? Does it have a
> deadline? I remember that I read that at least the analogue terrestrial
> broadcasting ended some months ago.
The deadline came and went last summer. Analog over-the-air TV is
essentially dead here now. There are a few low power stations left
scattered around but I imagine they are an endangered species.
Analog TV is still possible with some (most?) cable systems. The
decision to go fully digital is up to the cable provider. Of course
unlike OTA TV, digital cable frequently includes various forms of
mandatory DRM inflicted by the provider, which is a major problem for
MythTV users (or for that matter any user of a PVR system not rented
from the provider).
-Mike
--
Mike Isely
isely @ isely (dot) net
PGP: 03 54 43 4D 75 E5 CC 92 71 16 01 E2 B5 F5 C1 E8
More information about the pvrusb2
mailing list