[e2e] Re: crippled Internet

[Cannara]

Craig, I believe these were perhaps updates to what you saw then, but I don't
have the reference for the graphs I have.  They do indicate one way.  I'll
look around more and may come up with the reference.

Alex


Craig Partridge wrote:
> 
> In message <3AE7AEF4.411F83C9 at attglobal.net>, Cannara writes:
> 
> >Randy, the ITU specs are in 6 categories, from "perfect" to "bad".  Packet
> >delays of under 200mS are rated "acceptable", higher are not.  Jitter
> >acceptance really depends on buffering at the ends as well as the delay.  So
> >with 100mS delay you can get away with more jitter than with longer delays.
> 
> Can you give us the citation for the ITU specs you're citing here?  Years
> ago when I looked at the literature, the standard was CCITT G.114 which
> had standards for *one-way* delay and a three rating standard (acceptable,
> may impact some applications, unnacceptable).  The one way delay values
> were up to 150ms (acceptable), 150 to 400 ms (may impact), above 400ms
> (unacceptable).
> 
> These were standards for voice and again, were one way.
> 
> Craig