[Asterisk-Users] Asterisk Codecs [G.729]

[Unavailable ID]

You're right.  It's symmetric so it only takes 83Kbits/sec for u-law.
IPTraf is confusing me :-)

----- Original Message ----- 
From: "John Fraizer" <tvo at enterzone.net>
To: <asterisk-users at lists.digium.com>
Sent: Wednesday, March 10, 2004 11:41 AM
Subject: Re: [Asterisk-Users] Asterisk Codecs [G.729]


>
> You need to decide if you're going to measure both sides of the call or
not.
>   ITU standard is 64Kbits/s.  That is correct.  It is a standard DS0.
But,
> guess what.  That DS0 goes both directions so, "measured bandwidth per
call"
> is 128Kbits/s using your logic.
>
> Only "consumer" grade DSL/Cable bandwidth is asymmetric. These "wanna-be"
> connections and the accompanying garbage that the sales/marketing monkeys
> spew (that the general public laps up as accurate) have done nothing but
> cause confusion to people who don't work the network side of the industry.
>
> [rent on]
> (Besides that, it makes their peering ratio so lopsided that there is no
way
> they're going to get any decent no-settlement bilateral peering and as
such,
> the price of those connections is going to remain artificially high since
> the providers have to PURCHASE transit instead of simply peering.) [rant
off]
>
> When I order an OC3, it's 155Mb/s in BOTH DIRECTIONS.  The industry
doesn't
> say "That's 310Mb/s of bandwidth" because the measure of circuit bandwidth
> is by maximum one-way FLOW.  You can (on paper) flow 155Mb/s in one
> direction on an OC3 (including encapsulation overhead).
>
> So, if you want to use accepted telco + IP metrics for measuring the flow
> (and thus the bandwidth), you look at the GREATER OF THE IN/OUT flow and
> that is how much bandwidth is required.
>
> ULAW = 83Kbits/s including encapsulation overhead.
>
> John
>
>
> wrote:
> > Absolutely agree,
> >
> > ITU standard is 64Kbits/sec.
> >
> > VoIP traffic with U-law per channel is 83Kbits/sec.
> >
> > VoIP traffic with U-law per call is 166Kbits/sec [measuring bandwidth
per
> > call]
> >
> > ----- Original Message ----- 
> > From: "Nicolas Bougues" <nbougues-listes at axialys.net>
> > To: <asterisk-users at lists.digium.com>
> > Sent: Wednesday, March 10, 2004 9:55 AM
> > Subject: Re: [Asterisk-Users] Asterisk Codecs [G.729]
> >
> >
> >
> >>On Wed, Mar 10, 2004 at 09:03:57AM -0800, <Unavailable ID> wrote:
> >>
> >>>Rich,
> >>>
> >>>In real world, using real tool, getting real number.  You don't expect
> >
> > to
> >
> >>>either talk only mode or listen only mode.  Per call must have Rx & Tx
> >
> > for
> >
> >>>inbound & outbound.
> >>>
> >>
> >>[...]
> >>
> >>
> >>>Engineering rate is per channel but to calculate the bandwidth
> >
> > consumsion,
> >
> >>>it must be in realtime full-duplex.
> >>>
> >>
> >>Sure, but in the real world, IP trafic is mostly carried over full
> >>duplex links (either serial or switched Ethernet), so you usually
> >>consider the trafic in one direction only (provided it's symetrical).
> >>
> >>If I consider an E1 to be 2 Mbps (2 inbound, 2 outbound, really), and
> >>Fast Ethernet to be 100 Mbps (again, 100+100 in full duplex), I shall
> >>consider U-law to be 83 kbits/s, not 166.
> >>
> >>-- 
> >>Nicolas Bougues
> >>Axialys Interactive
> >>_______________________________________________
> >>Asterisk-Users mailing list
> >>Asterisk-Users at lists.digium.com
> >>http://lists.digium.com/mailman/listinfo/asterisk-users
> >>To UNSUBSCRIBE or update options visit:
> >>   http://lists.digium.com/mailman/listinfo/asterisk-users
> >>
> >
> >
> > _______________________________________________
> > Asterisk-Users mailing list
> > Asterisk-Users at lists.digium.com
> > http://lists.digium.com/mailman/listinfo/asterisk-users
> > To UNSUBSCRIBE or update options visit:
> >    http://lists.digium.com/mailman/listinfo/asterisk-users
>
> _______________________________________________
> Asterisk-Users mailing list
> Asterisk-Users at lists.digium.com
> http://lists.digium.com/mailman/listinfo/asterisk-users
> To UNSUBSCRIBE or update options visit:
>    http://lists.digium.com/mailman/listinfo/asterisk-users
>