RE: Real and Non-Real Time Traffic (was [WG-PIC:352] Re: FMM demo notes)

["Erik Dobbelsteijn" <>]

> >   ITU-T specifications cap mouth-to-ear delay at 150ms for 
> > toll quality voice.  How much of that can be allocated to 
> > transmission depends on the hardware being used for the 
> > terminal devices.  A guideline might be 100ms.

...and that goes for international calls! guidelines for short-distance
calls are even more constraining.

> The speed of light means that to circumnavigate half way 
> around the earth will take approx 200 ms. As long as jitter 
> is minimal, I have found that 300 ms is fine.

People have gotten used to long (>100ms) delays due to satellite calls
and especially mobile calls. ITU-members aren't able to meet their own
recommendations when letting their subscribers call from mobile phone to
mobile phone :)
IP pings crossing the Atlantic show that this section adds at least 80ms
to any type of traffic. still my collegue and I didn't experience much
of the long delay when calling from Indy to Holland (which usually leads
to not being able to respond accurately).

Erik

---------------------------------------------------------------wg-voip-+
For list utilities, archives, subscribe, unsubscribe, etc. please visit the
ListProc web interface at 

    http://archives.internet2.edu/

---------------------------------------------------------------wg-voip--