[mythtv] UK Freeview "Playback"

David Matthews dm at prolingua.co.uk
Fri Jan 5 15:14:08 UTC 2007


Stuart Morgan wrote:
> On Friday 5 January 2007 12:19, Chris Birkinshaw wrote:
>> On 8/24/06, Stuart Auchterlonie <stuarta at squashedfrog.net> wrote:
>>> On Thu, Aug 24, 2006 at 04:37:32PM +0100, Martin Long wrote:
>>>> Has anyone managed to get their hands on the technical change
>>>> specifications for Freeview "Playback"? I know that to support some of
>>>> the changes there are plans to introduce some new metadata into the EPG
>>>> and possibly some MHEG stuff too.
>> Mythtv is almost compliant, aside from a few areas:
> 
> It's not a case of mythtv's compliance so much as requiring the specs for the 
> additional metadata.
> 
>> 1. Series linking and split events (I have seen people are already
>> working on this)
> 
> Series linking is already done in effect but without series identifiers in the 
> EIT it can't distinguish between different series or even accurately identify 
> repeats at times. This is the sort of thing which requires the information be 
> broadcast and which hopefully will be introduced by Freeview Playback - but 
> no one has seen the specs yet. 

There are almost certainly some documents specific to Freeview Playback 
that are only available within the DTG but some of this appears to be 
based on ETSI standards e.g. TS 102 323 and they available free from the 
ETSI site.  The programme ID and series ID that the BBC are currently 
transmitting in the EIT are based on that.  I've been using the 
programme ID information in particular for the last few weeks (see 
#2811) and it is much better at detecting duplicates than textual 
comparisons of the subtitle and/or description.  The series ID seems to 
need some improvement: generally a "series" seems to be a particular 
series at a particular time.  If it is repeated at a different time it 
seems to have a different series ID, although the programme IDs of 
repeated episodes will be the same.

>> 2. Recording back to back events should not register a conflict (this
>> was mentioned ages ago on this list - has it been looked at?)
> 
>> 3. Recording start and stop needs to be controlled by the EITpf
> 
> Ending the recording according to that is fine and Janne has done work in that 
> area, although it doesn't actually stop the recording but places cut marks. 
> Starting the recording according to event is more problematic. To start early 
> we would need to be monitoring ahead of the programme start which may not be 
> possible.

TS 102 323 defines several ways of detecting start and end, including 
EITp/f.  I have no idea how much of this the BBC are planning to 
implement but I would certainly be interested in trying out a patch that 
derived start and end from the EIT data.  It is always going to be 
necessary to tune the transport a little bit in advance and that may 
depend on having a tuner available.

David.


More information about the mythtv-dev mailing list