Fiber to the Home? Maybe Never, Says Comcast's Werner
Los Angeles — The question of when — or whether — cable should take fiber to the home doesn’t keep Comcast CTO Tony Werner up at night.
“I don’t see it as a necessity, almost ever,” Werner told me when met here at the Cable Show, after I asked when Comcast might see the need to deploy FTTH.
Continued Werner, “Ever is a very long time. But I spend no time awake at night on this.”
Comcast has conducted tests in its labs with DOCSIS 3.0 equipment bonding 28 downstream channels together, to deliver a whopping 1 Gigabit per second. “Running that fiber the last 100 feet to the home costs a lot of money,” Werner said. “I think we have lots of gas left in DOCSIS.”
In addition, Comcast is exploring using the spectrum in the 1-2 GHz range to tap into even more capacity. “That’s looking promising now,” Werner said.
Note, too, that CableLabs has initiated research into a new data-over-coax approach, which would do away with the 6-MHz channel divisions and use new MAC and PHY components to deliver ultra-high-speed data services even more efficiently (see CableLabs Next-Gen Broadband Project Envisions Multigigabit Speeds).
Christopher commented:
Great for downstream, but DOCSIS 3 offers little upstream. I'll take fiber and symmetry any day. Of course Comcast prefers its network to be one-way ... completing the loop when my money goes back to them.
Larry L commented:
Not at all surprising, Todd. The question may be should be "home-to-the-fiber" rather than FTTH, as the most likely way a consumer can realize FTTH is to buy or rent a home or MDU that is provisioned with FTTH.
Werner's comments are perfectly predictable. Until demand for bandwidth exceeds the technical capacity available over HFC plant, cable will never spend the capex to push FTTH. For the foreseeable future it simply doesn't make sense.
Cable is only now undertaking bonding due to increased demand for bandwidth arising from the demand for multiple streaming HD video, which is not served by a 15Mbps connection. Consumer demand multiple HD video streams will need about 50 Mbps to the home & cable can easily provision for this by bonding coax connection off a fiber node. Much more cost effective that pushing the fiber to customer premises.
Larry L commented:
Not at all surprising, Todd. The question may be should be "home-to-the-fiber" rather than FTTH, as the most likely way a consumer can realize FTTH is to buy or rent a home or MDU that is provisioned with FTTH.
Werner's comments are perfectly predictable. Until demand for bandwidth exceeds the technical capacity available over HFC plant, cable will never spend the capex to push FTTH. For the foreseeable future it simply doesn't make sense.
Cable is only now undertaking bonding due to increased demand for bandwidth arising from the demand for multiple streaming HD video, which is not served by a 15Mbps connection. Consumer demand multiple HD video streams will need about 50 Mbps to the home & cable can easily provision for this by bonding coax connection off a fiber node. Much more cost effective that pushing the fiber to customer premises.
Tom Burgmeier commented:
Great coverage Todd. Very interesting to hear what cable is thinking long term. Fascinating to me that cable didn't bond more channels sooner and blow DSL out of the water. Btw, purely for marketing purposes, consider replacing your picture on your blog with the one of Kyra Sedgwick that's now on MCN's front page. :)