First, Mark... thanks for two superb replies.
I'll get to your first one later today, and possibly address some of the issue's you've brought up on the FCTF board.
As for the 64k DS0 rate, versus the use of compression, I see some room for interesting things here. Yes, understood that the upstream in HFC/coaxial plant systems must economize. This, I've always felt, was one of the drivers for the MSOs to be looking seriously at VoIP/IPtel. But in other venues, we will <allegedly> be seeing bandwidth gluts over time. I don't subscribe to this in a panicky way, but I do acknowledge that more bandwidth will be available per unit of message than we've had to date.
All of which leads me to wonder why compression will be attractive at all over packet voice, given that we are moving away from a minimalist view of bandwidth in the greater view of networking venues, and the other penalties that compression carries in processing overhead and latency.
Also, one of the assumed benefits of future IP/ATM (packet) voice is a "higher" (CD-) quality signal, possibly in stereo at some point, than what we perceive to be the case on POTS over the PSTN today. All of which would suggest that future high quality voice might even use super-64k, as opposed to <64k. Indeed, there are currently some IP PBX vendors whose packets exploit the full 64k, as well.
Just some thoughts on compression.
FAC |
