[linux-dvb] Multiproto API/Driver Update

hermann pitton hermann-pitton at arcor.de
Tue Sep 9 03:17:50 CEST 2008

Hello Barry,

Am Montag, den 08.09.2008, 17:43 -0700 schrieb barry bouwsma:
> --- On Mon, 9/8/08, Georg Acher <acher at in.tum.de> wrote:
> > > Or is DAB/T-DMB too different from DVB and related technologies,
> > 
> > DAB itself is totally different to DVB, as it has no transport stream
> > equivalent. The different subchannels have different frame/packet lengths,
> First, many herzlichen dank for your reply.  Now I'm going to
> ask a question about something that I've just read about in
> the last minutes...
> Would this be somewhat comparable to the Generic Stream (GS)
> of DVB-S2, particularly the Continuous mode, as opposed to the
> packetised mode?
> If there is to be added DVB-S2/GS ability to the relevant API,
> which I suspect will be eventually necessary, as the Transport
> Stream mode is a compatibility mode with DVB-S, then I am
> wondering if that is reasonably close to how whatever DAB
> hardware would deliver the modulated multiplex data on its
> interface.
> Of course, I really do not know what I'm talking about, yet.
> > depending on their bitrate allocation and FEC. Also the service information
> > (FIGs) has no similarity to DVB.
> I would have to ask either someone familiar with the hardware
> which I have, or someone who has viewed its datastream, but
> if I understand what I've read so far, I should expect to need
> to write a demultiplexer in userspace in order to get the
> interesting data from the (1,5Mbit/sec-ish) multiplex datastream.
> Anything more than that would require extending whatever API.
> > closer inspection
> > also shows some differences in the allowed video and audio
> > codecs and how SI are handled.
> If I understand correctly, the difference between DAB and DAB+
> has nothing to do with the modulation -- unlike the newly-
> being-tested DVB-T2, so my hardware from an API perspective
> won't be magically obsoleted -- only the software player needs
> AAC audio decoding ability.
> Whereas, all my DVB-T devices (and DVB-C) will be obsolete in
> the event that the nearby countries of interest decide to
> introduce DVB-T2 -- which is not likely soon, but could happen
> with the introduction of terrestrial HDTV, should those countries
> actually do so, following the UK.

following your thoughts, you are likely right. DVB-T2 likely indicates
that you need new hardware, like it is for sure on DVB-S2.

I also agree, that DVB-S h.264 stuff, like we have it currently on
BBC-HD, might switch over to DVB-S2 in the near future. At least it is a
requirement for supported DVB receivers to be S2 capable too and they
stay open to it.

It seems to be guaranteed at least, that it will not be encrypted.
The ITV HD solution looks somehow mad to me currently and might change

French/German and others ARTE HD uses DVB-S2 now, so very likely they
will be all on it in 2010, when the major channels switch to HD too.

It is still not totally clear, but likely we don't come through it
without T2/S2 capable devices.

This could be a question of politics, but I'm currently not
interested ;)



More information about the linux-dvb mailing list