said by Bengie25:
In the lab, they can get COAX up to 10Gb over a few miles. In the lab, they can get fiber at almost 1Tb over 800 miles and well over 1Tb in the sub 2mile range(for cheap to). By cheap, I mean upwards of 12Tb/s fiber that is cheap enough to integrate into your cell-phone and has upwards of 2Km range. IBM is working with many companies to start using this new tech.
The difference between me getting 10Gb fiber and you getting 10Gb cable is I get all 10Gb to myself between me and my ISP, you have to share.
If money were no object, then of course, you go for the best technology available.
The problem with FTTP is that the bonds start accruing interest immediately, whereas you start out with a take rate of 0% and have to work for years to get it up to the targeted value. That's why Verizon stopped their FiOS deployments and concentrated on improving take rate. If they already spent the money, it's a sunk cost. But if they haven't deployed fiber, it's just not worth it to spend the money.
Whereas the cablecos have the ability to upgrade incrementally
until they hit that 10 Gbps barrier. The cablecos haven't even hit the 10 Gbps limit for a 250-household node -- much less 10 Gbps per household. So, the cablecos have many years of runway left.
Eventually, of course, they run out of bandwidth and will have no choice but to deploy fiber. But the key is that they do not incur the capital costs of a fiber deployment until they actually need it.