[vdr] Re: dxr3/vdr problem

[Rantanen Teemu <teemu.rantanen@tekla.com>]

> From: Stefan Schluenss [mailto:Stefan.Schluenss@gmx.de]
> 
> Malcom is using DVB-T if i remember correctly and the currently
> implemented
> DVB-T systems seems to be more in an experimental state than in production
> state ;-)

I'd rather say that transport errors are more likely to happen on DVB-T then
on DVB-C (should 'never' happen) or on DVB-S (can happen on bad weather).

> The question is who (driver/vdr/dxr3-plugin) has to take care of corrupt
> data ?! In Kai's opinion the data has to be checked and rejected inside
> the
> driver and such corrcupt data should never reach upper layers (the dxr3-
> plugin).

Considering that full error checking is somewhat complex operation (correct
me if I'm wrong), I'd rather see that amount of code and CPU time not in
kernel, but in user space.

Anyway, I'm not expecting that someone announces tomorrow that there's a new
library that does all this, so I'd say we have to live with potential errors
for now.

> This is at least defined in the ISO/OSI-network model.
> Btw. the DXR3 is by itself very sensitive on corrupt data. It was never
> designed to work under such situations because it has normaly only deal
> with data
> from a DVD.

I really did try to crash the VDR with the latest DXR3 plugin (or the second
latest). I have an inside antenna so it can be moved around very easily. I
did move it into places that have bad reception inside my apartment, I did
point it to a wrong direction, I even disconnected the antenna. But I
couldn't get the DXR3 plugin to crash anymore. I only got really nasty
sounds and artifacts on picture...

I also haven't got a single broken recording anymore (= crash when
recording, or crashes when viewing the recording). I'd say the current
status is somewhat stable, at least stable enough for me (what VDR+DXR3 was
not a month ago or so).


					Teemu



-- 
Info:
To unsubscribe send a mail to listar@linuxtv.org with "unsubscribe vdr" as subject.