Re: [AVT] Audio Level Indicators - new versions submitted

[Stefan Sayer <[e-mail]>]

Hello,

o Emil Ivov [10/22/09 11:25]:
> All,
> 
> As requested by the chairs several weeks ago, Jonathan, Enrico and I
> have submitted new versions of the client-to-mixer [1] and
> mixer-to-client [2] audio level indication drafts. Both documents
> reflect comments received on the mailing list such as:
> 
> * using a uniform way of encoding level
> * add additional considerations/design choices sections
> * change tittles so that they would be easier to distinguish from one
> another
> * others
> 
> Needless to say further comments are most welcome.
in [1] 3. Protocol Operation the levels are described as having values 
from 0 to 255, and referencing section 4 (which should be section 5):

    Each instance of this header contains a list of one-octet audio level
    values (see Section 4).  Such values indicate audio level on a 0 to
    255 scale where 0 is silence (i.e. same as omitting the corresponding
    source id from the CSRC list) and 255 corresponds to a threshold
    accepted by the mixer implementation as the maximum audio level that
    a participant is likely to reach during a conference.

while in section 5, the levels are described like in the comfort noise 
specification with values 0..127 being 0..-127 dBov, where 0 would be 
the maximum audio level threshold (not 255).

In [2], section 3,  the MSB of the level field is in the Figure 1 of the 
extension block:

           0                   1                   2
           0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3
          +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
          |  ID   | len=1 |0| level       |V|  reserved   |
          +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

So it may make sense to add this to Figure 2 in [1] as well :
        0                   1                   2                   3
        0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
       |  ID   |  len  |    level 1    |    level 2    |    level 3   ...
       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

              Figure 2: Audio level indicators extension format

would become:
        0                   1                   2                   3
        0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
       |  ID   |  len  |0|  level 1    |0|  level 2    |0|  level 3   ...
       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

              Figure 2: Audio level indicators extension format

(or, alternatively remove it from Figure 1 in [2], but I think that it 
is good to have the MSB set to 0 in the Figure.)

Regards
Stefan

> 
> Cheers,
> Emil
> 
> [1] - http://tools.ietf.org/html/draft-lennox-avt-rtp-audio-level-exthdr
> 
> [2] - http://tools.ietf.org/html/draft-ivov-avt-slic
> _______________________________________________
> Audio/Video Transport Working Group
> [e-mail]
> https://www.ietf.org/mailman/listinfo/avt

-- 
Stefan Sayer
VoIP Services

[e-mail]
www.iptego.com

IPTEGO GmbH
Wittenbergplatz 1
10789 Berlin
Germany

Amtsgericht Charlottenburg, HRB 101010
Geschaeftsfuehrer: Alexander Hoffmann