[e2e] Re: crippled Internet

[David P. Reed]

At 08:36 PM 4/25/01 -0700, Cannara wrote:
>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.


This is true, but small numbers of delayed voice frames can be treated as 
dropped and turned into "interpolated noise bursts" or error corrected by 
the time they need to be played, as long as the frames around them are 
available.  Thus 50 ms of variation can be achieved by treating frames that 
arrive late as dropped.

You really have to design VoIP systems for the channel characteristics, 
which include loss and jitter, as I outlined in my previous note.  The ITU 
didn't consider this at all.

Don't take these comments to imply that I believe that VoIP doesn't need 
some kind of prioritization to work well in an environment with 
heterogeneous load.  But I don't believe that isochrony or admission 
control is needed.  Partly because human-human 2-way communications is 
going to be such a small part of the capacity of what gilder calls the 
fibersphere that a simple priority scheme is probably sufficient - 1-bit 
"low latency" vs. "variable latency".

If something like QoS takes 5 years to deploy, it should be designed for 
the cable plant of 5 years from now - voice is not growing, while 
human-machine and machine-machine is really zooming.