[linux-dvb] Re: Flexcop (HD5000) dvr0 removes null pid?
smalenfant at gmail.com
Fri Sep 23 14:34:01 CEST 2005
The problem I'm talking about here is the the new HD5000 seems to be
generating "invalid" null pids when you disable the null pid filtering in
the flexcop chip. What I need to do to is receive the full transport stream,
which actually. I've contacted the manufacturer and we concluded that only
the new HD5000 based on the LG frontend is experiencing this. Looks like the
frontend is generating the extra null pids. This problem is only affecting
people that needs to capture the full transport stream. For those who are
just watching TV, that's not an issue.
For dvbsnoop, what I've done is only valid with this card and this is only
a workaround for my own stuff. I'm expecting that the LG frontend driver
will be modified to fix this issue and this code will not be necessary. If
you are looking for the output of the PCR transport bitrate calculation I've
made, that's a bit different. I've shared that with Reiner already, not sure
that's going to make any dvbsnoop release.
On 9/22/05, Johannes Stezenbach <js at linuxtv.org> wrote:
> On Thu, Sep 22, 2005 Steve Malenfant wrote:
> > I've tried this card with TSReader in windows, same thing happened.
> There is
> > added invalid NULL PID to the stream which doesn't allow me to measure
> > accuracy.
> > I've modified my little dvbsnoop variant and if I throw out the invalid
> > NULL PID, the PCR bitrate is fine.
> I don't get it. I thought the problem is that you don't get NULL packets
> (stuffing on PID 0x1fff, right?), now you say that NULL packets are
> added to the stream by the driver and you have to filter them out?
> Can you clarify?
> Also, would you mind sharing the dvbsnoop patch?
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the linux-dvb