Re: death of the net predicted by deloitte -- film at 11
IP Multicast as a solution to video distribution is a non-starter. IP Multicast for the wide area is a failure. It assumes large numbers of people will watch the same content at the same time.
They do. Sure it degrades to effective unicast if too few people watch the same channel in the same area (so just use unicast for those channels), that doesn't mean it's no use for the popular channels that have millions of viewers.
The usage model that could work for it most mimics the broadcast environment before cable TV, when there were anywhere from three to ten channels to choose from, and everyone watched one of those. That model has not made sense in a long time. The proponents of IP Multicast seem to have failed to notice this.
10 or 1000 channels it's going to be better than not using it. I don't see the logic in using it for nothing because it's not good for some things. There are local factors that may mean some countries adopt it. In the UK all spectrum is sold, as we turn off analog it's not a given that the broadcasters will be able to buy that spectrum for HD. When we want 10 HD Olympics channels IPTV may be the only way for a large portion of the 20M or so viewers to get it.
The point is the more possible live content there is, the less multicast makes sense. Compounding this, fewer people care to watch live content, preferring instead to record and watch later on their own schedule, or be served on-demand. In this usage model, multicast is not helpful either.
Because they want to watch later doesn't make multicast no use. Who is going to pay for their time shift bandwidth use? Why would someone pay when a home device can do the time shift and make good use of the live multicast stream? They'll save the download cash for stuff that never was available live to them or they forgot to record, unless someone makes it appear to have no cost. brandon
10 or 1000 channels it's going to be better than not using it. I don't see the logic in using it for nothing because it's not good for some things.
Multicast isn't going to help the phoneco atm network. Whatever model emerges will only work if it works all the way to the end user. If you have a weak link in the chain then the chain breaks and right now that weak link is the last 2 miles. You can't pump gigE bandwidth speed over a DS3 to a dslam because you have 65 users watching HD content at 6pm. But if you accept that the average user only watches 3-6 hours of HDTV per day, you can spread the load out over 24 hours, the effects on available bandwidth can be reduced. The TIVO model appears to have an advantage for the viewer (a large archive to select from) and for the phoneco's and ISP's at the customer end. Geo.
nothing can help, or for that matter save, the phoneco atm network.
atm and frame relay do not need saving. they tend to be profitable. but the everything over mpls folk are managing to save them anyway, turning operating profit into capital expense to the vendors. brilliant. randy
On 12-Feb-2007, at 09:23, Brandon Butterworth wrote:
Sure it degrades to effective unicast if too few people watch the same channel in the same area (so just use unicast for those channels), that doesn't mean it's no use for the popular channels that have millions of viewers.
I think you're presupposing that the concept of "channels" is something that will persist. Joe
On Mon, Feb 12, 2007 at 06:42:06AM -0500, Joe Abley wrote:
On 12-Feb-2007, at 09:23, Brandon Butterworth wrote:
Sure it degrades to effective unicast if too few people watch the same channel in the same area (so just use unicast for those channels), that doesn't mean it's no use for the popular channels that have millions of viewers.
I think you're presupposing that the concept of "channels" is something that will persist.
Joe
perhaps you have to narrow a view of what a channel is? --bill
participants (6)
-
bmanning＠karoshi.com -
Brandon Butterworth -
Geo. -
Joe Abley -
Paul Vixie -
Randy Bush