Re: [ippm] some notes on draft-morton-ippm-delay-var-as-03

Carsten,

Thanks for your comment.
> Dear authors, 
>
> I have recently read the ippm-delay-var-as
> draft and believe it is on the right way.
>
> For me it is clearly the most complete 
> comparison PDV vs. IPDV up to now and very
> helpful to understand the nitpicks of each definition.
>
> Let me add a few observations from my own work
> in this area here:
>
> a) as PDV is clearly a shifted version of the 
>    delay itself, its statistics can be directly
>    derived from the delay (OWD) statistics when
>    in addition the reference value is known. 
>   
Good remark. Will be added to the next version of the draft.
>    Even though I believe it has its own justification and use.
>
> b) For IPDV it would be nice to know, how symmetrical
>    it is in practice. If it is very symmetric in 
>    almost all cases then one might as well work 
>    with the stats applied to abs(IPDV) only.
>    If unsymmetric IPDV distributions appear then 
>    one should present a positive and a negative high 
>    percentile or something like inter-quartile range 
>    as resulting statistics (in addition to mean IPDV in any case)
>
>    Does anyone have results wrt. the symmetricity?
>   
I would also be interested in the results.
> c) Esp. for IPDV it is not clear to me the effect
>    of the packet selection function: Esp. the question:
>    Does the size of the gaps between selected packets have an 
>    influence on the distribution of the resulting IPDV values?
>   
Not sure. I advice my customers to use the same spacing as the service 
they try to monitor.
Example: 20 ms in voice.
>    (in theory yes (I could give an example), but in practice?)
>
> I will have a more detailed look at the draft in the next
> week and then write some specific comments as well.
>
> In general I'd say the draft should be concluded 
> with something like a table which recomends clearly 
> the use of PDV or IPDV and suggested stats
> per desired application.
>   
ok.

Regards, Benoit.
> Best regards,
> Carsten.
>
> --
>
> ### !! ATTENTION !! Change of email address! Please use in future only:
> carsten.schmoll@fokus.fraunhofer.de ###
>
> "The difference between theory and practice is that in theory theory and
> practice are the same but in practice they are not."
>
> Dipl.Ing. Carsten Schmoll              Fraunhofer Institute FOKUS
> carsten.schmoll@fokus.fraunhofer.de    National Research Institute
> Fraunhofer FOKUS / dept. NET           for Open Communication Systems
> Tel: +49-30-3463-7136                  Kaiserin-Augusta-Allee 31
> Fax: +49-30-3463-8136                  D-10589 Berlin, Germany 
>   
> ------------------------------------------------------------------------
>
> _______________________________________________
> ippm mailing list
> ippm@ietf.org 
> https://www1.ietf.org/mailman/listinfo/ippm
>