Re: Pitfalls of G.fast
AndrueC, I did (try to) make that point in my first post:
"It's not a cheap alternative to real Fibre, its just an alternative that makes use of BT's legacy copper (in the interests of BT, and only BT). It will get rolled out iniitally as 1:1 mapping with existing FTTC Cabinets, so notspots are still notspots"
'notspots remain notspots'
(maybe 1:1 mapping wasn't the best way to describe it, but it means it will only go into places where existing FTTC cabinets exist, the technology is mapped in, one for one, replacing one existing FTTC line card with a newer G.fast line card).
Also a point often missed, the quality of PSUs in existing FTTC Cabinets will need be substantially improved to cater for the PSU noise sensitivities of G.fast too.
Britain's Broadband gets an minor incremental upgrade (BT will certainly lobby to go the route of further handouts/taxpayer subsidies) to enable exactly the same people that were getting 'upto' 80Mbps beforehand, to get 'upto' G.fast speeds, which I'll be generous and say, 'upto' 100Mbps-200Mbps* (on a good day). As said, notspots remain notspots.
(*as always 'upto' dependent on cabling alu/copper, site issues, crosstalk, low frequency pump/induction noise, distance from the FTTC by cable length, Backhaul/ISP traffic management/network management/congestion - restrictions permitting - ofcom's idea of regulation heaven, keeps them in a job, hence their unremitting support for G.fast)