I don't agree at all. I think the announced Rockwell technology is exactly what the market craves. At 1mps download to single individual, you are gonna nearly always be bottlenecked on either the backbone (actually, the hub points such as the one in Silcon Valley) on the server. Almost no one has that much speed available now DEDICATED TO A SINGLE USER. Half a Mbs would be enough, actually.
The issue is 1) telco implementation cost and 2) ISP cost 3) end user modem cost.
If this happens in next two-three years cable modems will be stopped dead in their tracks as any mass phenomenon. Actually, I think it is the anticipation of thinks like this coming that has kept the cable cos from undergoing the very considerable cost of converting their systems.
Also, what people fail to realize about cable solutions is that they are great for the early adopters, before there are many people on the line. But as soon as a whole city block, or whatever, is using cable modems the throughput drops dramatically...sometimes even below 56k.
The trouble is that although shielded cable inherantly has vastly greater bandwith capability, enormous numbers of people share the same cable running from a routing station to end users. Its like huge nos of people all on the same LAN. And most of the shielded cable's capacity is already being used by the enormous bandwidth hog of video...and hdtv is worse...
With telco twisted pair wiring, you (generally) have your own dedicated pair (actually, 4 wires per household, at least) goint to the switching station. Now that xDSL technology has figured out how to get lots more bandwidth out of those pairs, its the obvious way to go. Cable is only an interim solution for a few, in my opinion.
The main reason cable is moving forward is becasue the local telcos are being so slow....
And the main reason they are, is because they don't see the pricing incentives to go xDSL for the home. And in fact, are deathly afraid that the technology, which is similar to what is used to turn a couple of pairs into a T1 line to rent to business for $1k/month and UP UP UP, will end up canabalizing that market....
Which is probably why Gates bought into Comcast....as a hedge, and to get the telcos moving.
He probably figures that in all likelyhood, its gonna be a low return (and low risk) billion out the door. (But better than cash just sitting there.)...and can have the saluatory effect of scaring the telcos into acting....creating the necessary competition, even though he knows his Comcast side will probably end up loosing the Inet access battle....main even, as for Intel....is to get bandwidth up so that the demand for fatter software doesn't decrease, but increases....
Regards, Doug |
