From: Scott Morris (swm@emanon.com)
Date: Fri Aug 20 2004 - 01:08:26 GMT-3
Voice is sampled in 10ms "chunks". A standard voice packet will contain
20ms of voice (2 x chunks for every set of headers (50pps)).
If we change this to only 1 x 10ms chunk per packet, we need to have all of
the header information now for only half the data payload, thus taking more
bandwidth (100pps).
On the flip side, if we move to 30ms payload (3 x chunks), we are making the
ratio of header to payload byte counts better (33 pps)
HTH,
Scott Morris, CCIE4 (R&S/ISP-Dial/Security/Service Provider) #4713, CISSP,
JNCIP, et al.
IPExpert CCIE Program Manager
IPExpert Sr. Technical Instructor
swm@emanon.com/smorris@ipexpert.net
http://www.ipexpert.net
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
Patrick Torney
Sent: Thursday, August 19, 2004 11:27 PM
To: ccielab@groupstudy.com
Subject: new to voice
Hi All:
I'm trying to understand why something is so...
doc cd says...
Calculating Required Bandwidth
The bandwidth required for a voice call depends on the bandwidth of the
codec, the voice packetization overhead, and the voice frame payload size.
The ***smaller the voice frame payload size, the higher the bandwidth
required for the call***.
I'm not understanding the logic here. Can someone please illuminate me?
TIA.
Pat Torney
This archive was generated by hypermail 2.1.4 : Fri Sep 03 2004 - 07:02:46 GMT-3