A Sky, cable and digital tv forum. Digital TV Banter

If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.

Go Back   Home » Digital TV Banter forum » Digital TV Newsgroups » uk.tech.digital-tv (Digital TV - General)
Site Map Home Register Authors List Search Today's Posts Mark Forums Read Web Partners

uk.tech.digital-tv (Digital TV - General) (uk.tech.digital-tv) Discussion of all matters technical in origin related to the reception of digital television transmissions, be they via satellite, terrestrial or cable. Advertising is forbidden, with no exceptions.

DVB-T/T2 packet metadata parsing



 
 
Thread Tools Display Modes
  #1  
Old September 17th 13, 12:52 PM posted to uk.comp.os.linux,uk.tech.digital-tv
Jim Lesurf[_2_]
external usenet poster
 
Posts: 4,182
Default DVB-T/T2 packet metadata parsing

I'm currently finishing off some work on iplayer and plan to move on to a
fresh look at DVB-T2. I recorded some streams during the 'proms' and these
have prompted some things I want to investigate. I've also been talking to
someone I know who says there are more changes coming along to the details
stream specs, and I found some of the recent BBC RB 303 HD proms won't play
in an old VLC when others do play. So it seems some experiments have been
under way.

The point of saying the above is I'm wondering if anyone can point me at
software that already parses the NIT and other tables to give an analysis
and report. Covering all kinds of details like *other tables and TRD. I've
also been told some EIT data on DVB-T2 is compressed. But what documents I
have are fairly old, so I'll need to get up to date.

Slainte,

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

  #2  
Old September 17th 13, 03:11 PM posted to uk.comp.os.linux,uk.tech.digital-tv
Tony Houghton[_2_]
external usenet poster
 
Posts: 43
Default DVB-T/T2 packet metadata parsing

On Tue, 17 Sep 2013 13:52:19 +0100
Jim Lesurf wrote:

The point of saying the above is I'm wondering if anyone can point me at
software that already parses the NIT and other tables to give an analysis
and report. Covering all kinds of details like *other tables and TRD. I've
also been told some EIT data on DVB-T2 is compressed. But what documents I
have are fairly old, so I'll need to get up to date.


You should find dvbsnoop useful.

The four HD channels have compressed names and descriptions in the EIT,
using the same Huffman tables as Freesat, but on standard PIDs. It uses
the same algorithm as ATSC but with slightly different tables optimised
for the UK. I think support for it is standard in MythTV now, and you
can also check the code for VDR's eepg pluin or Freesat patch (you'll
probably be better off with the latter if you can find it), and in my
DVB stuff https://code.google.com/p/hdvb/.

The Freeview HD channel doesn't seem to be referenced in the NIT, unless
that's changed since last time I looked. Well, there are some DVB-T2
descriptors for it, but they don't carry the frequency etc. The only way
I know to find it is to do a full scan with the various parameters set
to AUTO.

--
TH * http://www.realh.co.uk
  #3  
Old September 17th 13, 07:40 PM posted to uk.comp.os.linux,uk.tech.digital-tv
Andy Furniss[_3_]
external usenet poster
 
Posts: 130
Default DVB-T/T2 packet metadata parsing

Jim Lesurf wrote:
I'm currently finishing off some work on iplayer and plan to move on
to a fresh look at DVB-T2. I recorded some streams during the 'proms'
and these have prompted some things I want to investigate. I've also
been talking to someone I know who says there are more changes coming
along to the details stream specs, and I found some of the recent BBC
RB 303 HD proms won't play in an old VLC when others do play. So it
seems some experiments have been under way.


Well historically at least IME VLC is far too fussy when it comes to ts.

IIRC VLC won't play sound without the PMT and though I can't recall
whether it was tzap or dvbstream I was using there have been occasions
where the recorder for whatever reason failed to include it. I guess you
can quickly check that one with pidsniffer type prog you have.

If you do have four streams (there should really be 6 but tzap doesn't
actually parse the PMT so you loose subs and AD) then you could always
try snipping a few hundred packets off the start.

dd if=in.ts of=out.ts bs=188 skip=300

I have occasionally seen players fail just because of an "unlucky
start". I did once write some code to seek out random access indicators,
which give an even cleaner start, but then got carried away and it
turned into a buggy mess of failed experiments :-)

The point of saying the above is I'm wondering if anyone can point me
at software that already parses the NIT and other tables to give an
analysis and report. Covering all kinds of details like *other tables
and TRD. I've also been told some EIT data on DVB-T2 is compressed.
But what documents I have are fairly old, so I'll need to get up to
date.


As I said before I now use tvheadend.org - which does EPG for T2, it's
quite big, but the code must be there somewhere.



  #4  
Old September 24th 13, 04:38 PM posted to uk.comp.os.linux,uk.tech.digital-tv
Andy Furniss[_3_]
external usenet poster
 
Posts: 130
Default DVB-T/T2 packet metadata parsing

Jim Lesurf wrote:
I'm currently finishing off some work on iplayer and plan to move on
to a fresh look at DVB-T2. I recorded some streams during the 'proms'
and these have prompted some things I want to investigate. I've also
been talking to someone I know who says there are more changes coming
along to the details stream specs


Did you get anywhere with this, did your source give anything specific
away about the nature of the changes?

Just curious.

Andy.
  #5  
Old September 25th 13, 08:01 AM posted to uk.comp.os.linux,uk.tech.digital-tv
Jim Lesurf[_2_]
external usenet poster
 
Posts: 4,182
Default DVB-T/T2 packet metadata parsing

In article , Andy
Furniss [email protected] wrote:
Jim Lesurf wrote:
I'm currently finishing off some work on iplayer and plan to move on
to a fresh look at DVB-T2. I recorded some streams during the 'proms'
and these have prompted some things I want to investigate. I've also
been talking to someone I know who says there are more changes coming
along to the details stream specs


Did you get anywhere with this, did your source give anything specific
away about the nature of the changes?


Sorry. Afraid I've been delayed/diverted by something else I'm still
working on!

A while ago I did a 'demo' of 'upsampling' audio to get user-control over
the process. Became interested in adding a 'causal' filter[1] example of
the process to the programs. Also thinking of trying to write an upsampling
player.

I've been trying to learn more about the packeting details, though. Alas,
they tend to be hard to grasp, and the multiple documents a bit of a maze,
so I keep going back to the audio as a relief! Designing audio
reconstruction filters is trivial in comparison!

I've been asking about the changes. But as yet not had a clear reply. Not
sure if this is because they are busy or because there seem to be a series
of changes.

Jim

[1] i.e. one that has no pre-ringing in the time domain and behaves like
fairly conventional analogue filters.

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

  #6  
Old September 25th 13, 11:59 PM posted to uk.comp.os.linux,uk.tech.digital-tv
Andy Furniss[_3_]
external usenet poster
 
Posts: 130
Default DVB-T/T2 packet metadata parsing

Jim Lesurf wrote:
In article , Andy
Furniss [email protected] wrote:
Jim Lesurf wrote:
I'm currently finishing off some work on iplayer and plan to move
on to a fresh look at DVB-T2. I recorded some streams during the
'proms' and these have prompted some things I want to
investigate. I've also been talking to someone I know who says
there are more changes coming along to the details stream specs


Did you get anywhere with this, did your source give anything
specific away about the nature of the changes?


Sorry. Afraid I've been delayed/diverted by something else I'm still
working on!

A while ago I did a 'demo' of 'upsampling' audio to get user-control
over the process. Became interested in adding a 'causal' filter[1]
example of the process to the programs. Also thinking of trying to
write an upsampling player.

I've been trying to learn more about the packeting details, though.
Alas, they tend to be hard to grasp, and the multiple documents a bit
of a maze, so I keep going back to the audio as a relief! Designing
audio reconstruction filters is trivial in comparison!


Trivial! I wish I thought sound was trivial, having recently resorted to
asking noob questions on surround sound lists about Db volume WRT
amplitude vs power, de-corrolation, phase ugh it's very confusing.

I have been doing a bit of downsampling I take it that sox is as good as
anything for 48 - 44.1?

The reason for the sudden interest is I've recently found ATK +
supercollider so on the fly I can use any of the cpic/listen hrtfs to
hear b-format ambisonic stuff.

It's also possible real time to feed in "normal" surround tracks, encode
with a virtual soundfield at the listening position of a 5.0 or 7.0
setup, then decode that to hrtf / uhj / stereo or proper ambisonic
setups (I don't have the speakers for that).

The databases are all 44.1k though so the sources need to be re-sampled.

Quite cool stuff - passes the time :-)

  #7  
Old September 26th 13, 08:39 AM posted to uk.comp.os.linux,uk.tech.digital-tv
Jim Lesurf[_2_]
external usenet poster
 
Posts: 4,182
Default DVB-T/T2 packet metadata parsing

In article , Andy
Furniss [email protected] wrote:


I have been doing a bit of downsampling I take it that sox is as good as
anything for 48 - 44.1?


Yes. In my view, sox is pretty good for conversions, etc. I certainly use
it quite happily. It has a fairly good rate convertor, and you can also
tweak the details a lot if you wish. Your life just needs to be long enough
to read its man pages and work out what they mean. Good rival for ffmpeg!
8-]

The reason for the sudden interest is I've recently found ATK +
supercollider


Not heard of that. (?)

so on the fly I can use any of the cpic/listen hrtfs to
hear b-format ambisonic stuff.


It's also possible real time to feed in "normal" surround tracks, encode
with a virtual soundfield at the listening position of a 5.0 or 7.0
setup, then decode that to hrtf / uhj / stereo or proper ambisonic
setups (I don't have the speakers for that).


The databases are all 44.1k though so the sources need to be re-sampled.


Quite cool stuff - passes the time :-)


Say more? :-)

FWIW I recently found a Pye LP "Quadraphonic Sampler" (Sansui QS)
QUAB1001. All 'easy listening' music, so pretty boring. Fairly typical of
ancient 'hifi demo LPs' in being tedious music very cleanly recorded. The
music's crap, but listen to that trombone! 8-]

However for 1 pound I couldn't resist it out of curiosity. And as I'd
guessed, the condition is good as it has probably only been played three
times, all with a decent stylus, in the past 40 years.

Does any of the above let you decode QS or SQ? I've been wondering about
that. I do have two or three ancient 'quad' LPs but this has prompted my
curiousity again. When played on my stereo the results are quite 'phasy'
and 'between the ears' in places.

FWIW I normally do 48k/24 recordings of LPs to make digital playing
versions, then convert them to 48k/16bit using sox to tweak the gain and
sometimes other alterations, and generate flac output. But for some items
like the above "Quad" disc I did a 96k/24 copy as I wanted more elbow room
during any processing.

Slainte,

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

  #8  
Old September 26th 13, 11:29 PM posted to uk.comp.os.linux,uk.tech.digital-tv
Andy Furniss[_3_]
external usenet poster
 
Posts: 130
Default DVB-T/T2 packet metadata parsing

Jim Lesurf wrote:

Yes. In my view, sox is pretty good for conversions, etc. I certainly
use it quite happily. It has a fairly good rate convertor, and you
can also tweak the details a lot if you wish. Your life just needs to
be long enough to read its man pages and work out what they mean.
Good rival for ffmpeg! 8-]


Good to know, thanks.

The reason for the sudden interest is I've recently found ATK +
supercollider


Not heard of that. (?)


Neither had I till very recently while generally reading about
ambisonics.

Supercollider its self provides an interpreted language for music
creation, and I don't really have a clue how to use it properly.
Luckily the ATK plugins have some examples to copy from so I can at
least load stuff into a buffer encode it (if needed) then decode it.

http://www.ambisonictoolkit.net/wiki/tiki-index.php

http://en.wikipedia.org/wiki/SuperCollider

http://www.ambisonictoolkit.net/wiki...?page=SC3+Help

Help like the above can be can be accessed/searched within supercollider


Quite cool stuff - passes the time :-)


Say more? :-)


Early days, I was impressed I managed to find something that lets me
play with different HRTFs, I never did much like alsa vdownmix so maybe
given enough time I'll find one that suits me better.

ATK can encode from below + sperical heads of various sizes.

So the time consuming thing is to try them all out with various content.

In the case of the listen database there are horizontal demos, so you
can just play the wavs with any player and get a shortlist to start with,
haven't found the same for cipic, and of course there's 3D to consider.

http://recherche.ircam.fr/equipes/sa...en/sounds.html

http://interface.cipic.ucdavis.edu/sound/hrtf.html


FWIW I recently found a Pye LP "Quadraphonic Sampler" (Sansui QS)
QUAB1001. All 'easy listening' music, so pretty boring. Fairly
typical of ancient 'hifi demo LPs' in being tedious music very
cleanly recorded. The music's crap, but listen to that trombone!
8-]

However for 1 pound I couldn't resist it out of curiosity. And as
I'd guessed, the condition is good as it has probably only been
played three times, all with a decent stylus, in the past 40 years.

Does any of the above let you decode QS or SQ? I've been wondering
about that. I do have two or three ancient 'quad' LPs but this has
prompted my curiousity again. When played on my stereo the results
are quite 'phasy' and 'between the ears' in places.


It won't dematrix but does have a quad with a virtual soundfield encoder
if you can find something decent to make four tracks.

I say something decent, because with a quick search I only managed to
find a Java app, which looks like it only takes 44.1/16 and needs some
jack type connecting to record the output. Maybe there is better somewhere.

http://www.hotto.de/software/quadrop...ixdecoder.html

Of course encoding 4.0 5.0 or 7.0 is not real ambisonics because the
content is pair panned, but I think the software "tries it's best" - I
don't know how well it does to a critical ear with music though, so far
I've been playing with channel checks.


 




Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


All times are GMT. The time now is 09:20 AM.


Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.SEO by vBSEO 2.4.0
Copyright 2004-2017 Digital TV Banter.
The comments are property of their posters.