No subject

3) delay variations that can not be handled by jitter buffer thusly sized
are packet losses.

Audible effects are caused by correlations of packet losses. Modern codecs
typically can handle single packet losses.

This argument suggests that the size of jitter buffer size depends on the
usage scenario.

As for measurements, ETSI EP TIPHON has standardized a measurement method
designed for VoIP and other periodic real-time streams. IETF is still
considering whether a similar method would be a good idea. The underlying
idea in this kind of measurement is to emulate VoIP media stream as to
packet sizes and packet transmission intervals.

 -- Vilho

> -----Original Message-----
> From: ext Randy Bush [mailto:randy at psg.com]
> Sent: 26 April 2001 07:32
> To: Cannara
> Cc: end2end-interest at postel.org
> Subject: Re: [e2e] Re: crippled Internet
> 
> 
> > There are ITU specs for jitter and delay in voice that have 
> been used for
> > years in standard telco system design.  I should think 
> these would be easily
> > accessible, but I only have a couple of graphs on paper.  
> The basic idea is
> > that 50mS or so of frequent variation in sample arrival 
> time is hard for
> > listeners, and 100mS roundtrip delay becomes annoying in 
> conversations.
> 
> in our universe, the modal delay is pretty much governed by 
> the speed of
> light in fiber and copper plus a few forwarding delays.  not 
> a lot we can
> do about it.  and i do not understand the meaning of 
> round-trip delay in
> the context of voip, as the phone calls i am on are 
> simultaneous but non-
> correlated one way chanels.
> 
> so the jitter characterization, "50ms or so of frequent 
> variation," would
> seem to be the interesting issue.  what is "frequent 
> variation," and how
> should it be measured?  i.e. if 3% of the samples are 
> outliers, how they
> are distributed in time would seem to be critical.
> 
> randy
>