[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [AVT] RE: I-D ACTION:draft-lakaniemi-avt-rtp-evbr-00.txt



Hi

I agree we don't know all future needs, but the requirement to control the bit-rate, the wideband/superwideband mode and the mono/stereo mode, are well identified for codecs currently developed (At least in the ITU-T G.xxx series).

In addition the number of remaining PT for RTCP packets is not infinite, so it is worth working on a common, open, and extensible enveloppe for these codec commands.

I don't have a precise solution in mind, it is just to bring the issue to the WG and get feedback.

Aurélien

> -----Message d'origine-----
> De : ari.lakaniemi at nokia.com [mailto:ari.lakaniemi at nokia.com] 
> Envoyé : jeudi 29 novembre 2007 13:51
> À : SOLLAUD Aurelien RD-TECH-LAN; csp at csperkins.org
> Cc : avt at ietf.org
> Objet : RE: [AVT] RE: I-D ACTION:draft-lakaniemi-avt-rtp-evbr-00.txt 
> 
> Hi Aurélien,
> 
> I agree that it does not make too much sense to replicate 
> exactly the same bit-rate control mechansims for multiple payloads.
> 
> However, at least for EV-VBR the proposed mechanism is not 
> only for bit-rate control but the intention is to enable also 
> expressing receiver's (changed) preference regarding the 
> configuration of the (upcoming) bandwidth extension and 
> stereo options. I assume that future (layered) codecs with 
> more advanced features/options would need to be able to 
> express their preferences among even larger set of options 
> when changing the bit-rate/configuration. On the other hand, 
> I expect different codecs to offer different sets of options 
> (and different rules on combining the options). Thus, it 
> seems likely that (also future) codecs (will) have different 
> requirements for a such bit-rate/configuration control messages.
> 
> Regards,
> Ari
> 
> >-----Original Message-----
> >From: ext SOLLAUD Aurelien RD-TECH-LAN 
> >[mailto:aurelien.sollaud at orange-ftgroup.com]
> >Sent: 27 November, 2007 18:37
> >To: Lakaniemi Ari (Nokia-NRC/Helsinki); csp at csperkins.org
> >Cc: avt at ietf.org
> >Subject: RE: [AVT] RE: I-D ACTION:draft-lakaniemi-avt-rtp-evbr-00.txt
> >
> >> >Section 4 defines a new RTCP APP packet. I generally find
> >> APP packets
> >> >to be the wrong approach for anything other than
> >> application-specific
> >> >information. Either define a new RTCP packet type (with its
> >> own RTCP PT
> >> >value), or use an AVPF feedback packet type.
> >> >
> >> 
> >> This is obviously one of the open issues in the draft. RTCP APP 
> >> packet seemed like an easy (and therefore tempting) solution, 
> >> although I had some doubts about it already before this 
> discussion. I 
> >> will try to work on pros and cons of more appropriate solutions, 
> >> including (but not necessarily limited
> >> to) 1. New RTCP packet type 2. (New) AVPF FB packet type 3. 
> >> Usage of an existing mechanism, such as TIMBR
> >> 
> >
> >Hi
> >
> >I think we should define a generic, codec-independant way to 
> transport 
> >this type of codec bit-rate control command.
> >Actually, modern audio codecs developped in the ITU-T or 
> elsewhere tend 
> >to be scalable and layered.
> >It doesn't seem appropriate that every payload format define 
> a way to 
> >do the same thing.
> >As I understood it during the elaboration of the CCM draft, 
> it is not 
> >the goal of the current CCM TMMBR to perform the thing we 
> are talking 
> >about. Anyway, if it is done as a new kind of CCM, I suggest to open 
> >this kind of command to any profile, not only AVPF.
> >
> >Regards,
> >Aurélien
> >
> 

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