[Live-devel] Frame granularity

[Matthew Hodgson]

Ross Finlayson wrote:
> 
> FYI, RFC 2190 is currently being classified as 'historic'.  That 
> payload format should not be used any more.  

I am painfully aware of this - unfortunately, I have a bunch of legacy 
equipment and software which still insist on speaking it, hence my interest 
in adding support for it.  I assume this means that you wouldn't be 
interested in receiving patches for its support?

 > See
> <http://www.ietf.org/internet-drafts/draft-ietf-avt-rfc2190-to-historic-06.txt>. 
> Because of this, I won't be giving you much more help here after this 
> message - sorry.

Fair enough.  Thanks once again for the assistance that you have given. :)

>> Okay - but should one try to use the data stripped by the Source in the
>> Sink, if fSource->fSpecialHeaderBytes[] is available?
> 
> You shouldn't need to.  If you're writing a RTPSource->RTPSink relay, 
> then just pass the stripped data (a 'subpicture frame', to use your 
> terminology) to the RTPSink, and the special headers should get recomputed.

Okay - i'll recompute the special headers in the sink from a combination of 
querying the source with things like RTPSource::curPacketMarkerBit() and 
inspecting the payload of the discrete (sub)frame.

thanks,

Matthew.

-- 
______________________________________________________________
Matthew Hodgson   matthew at mxtelecom.com   Tel: +44 845 6667778
                 Systems Analyst, MX Telecom Ltd.