RE: [AVT] Update (-09) of Timed Tex draft (WAS: RE:Commentsondraft-ietf-avt-rtp-3gpp-timed-text-08.txt)

[Jan vanderMeer <jan.vandermeer at philips.com>]

Jose,

Sorry, I meant the SD buffer, not the
SDIX buffer.

In my view, an SD is an element of the
timed text stream, and the SD buffer an internal attribute of the timed
text decoder. Within the SD buffer, each SD with a dynamic SDIX is stored
until it is invalidated. The SD buffer is transport independent.

I hope that answers your question.

Best regards,

Jan

"Jose Rey" <Jose.Rey at eu.panasonic.com> 15-04-2005 12:54

To:        Jan vanderMeer/EHV/IPS/PHILIPS at PHILIPS "Jose Rey" <Jose.Rey at eu.panasonic.com>         cc:        <avt at ietf.org> "Colin Perkins" <csp at csperkins.org> "Magnus Westerlund" <magnus.westerlund at ericsson.com> "Dave Singer" <singer at apple.com>         Subject:        RE: [AVT] Update (-09) of Timed Tex draft (WAS: RE:Commentsondraft-ietf-avt-rtp-3gpp-timed-text-08.txt)         Classification:

Jan,


Jose, 

> I.e., it cannot be that SDs be just discarded once they are used.

Right, but in my view they are stored in an SIDX buffer until they are
invalidated; when that is is not relevant. Or do I miss something ?
 
And where is the SD (not
SIDX) buffer? What I want to say is (cf. answer to Dave) that dynamic SDs
(and SIDX) are first introduced by this payload format and therefore codecs
that use them have to implement the dynamic SIDX wrap-up algo to know how
long to buffer SDs (at codec level).  This sounds a bit strange: a
payload format setting requirement on a codec and that's why I am a bit
doubtful this be OK(?) or am I missing something ??
 
Cheers,
José
 
 
 
Best regards,

Jan

 [] 
_______________________________________________
Audio/Video Transport Working Group
avt at ietf.org
https://www1.ietf.org/mailman/listinfo/avt