back to article BT: G.fast is fast and we want it FASTER

BT has told a US broadband conference it's confident that G.fast will do what it says on the box – so long as ongoing standards work bears fruit. However, The Register notes with interest that the dependencies highlighted by BT Wholesale CTO Colin Bannon are many. As noted at LightReading, Bannon reckons G.fast is good for …

  1. Da Weezil

    Meanwhile another part of the BT group cannot sort out the capacity issues that drop my circa 70 meg line down to single digit speeds during large parts of the day only 8 months after our exchange went fibre.

    Id love to see them research how to ANTICIPATE rises in demand and then behave pro-actively instead of re-actively, that way when I am at home, I might see some of the service level I am paying for (my line does high 60smeg/low 70s on an upto 80 service I am happy with that level, if only BTw could provide it consistently), because for a large part of the day I am at sub 40 meg and that is a lower price plan.... any other industry would be in trouble for this sort of behaviour!

    My ISP tell me the next engineering update on this is today, Im sure there will be no good news while this shower fleece us for shoddy service.

    Faster speeds only mean more problems for customers on Wholesale as BT dont seem to want to invest in capacity until forced to, and usage is clearly rising with faster speeds.

    1. Anonymous Coward
      Anonymous Coward

      I'm not convince on that. BT seems to have invested a lot in their 21CN core recently..changing out P routers to the fastest models around. Further...isn't this story about them investing in better access?

      Dont mix up capex investment with opex investment needed to keep your current service running.

      1. Da Weezil

        Investing in better access is a moot point when the existing acess methods already have the network faltering. Its no use having a potential 300 meg service if the ACTUALITY of it is that for a lot of the say it will only do 10 meg.

        Cart before the horse... as usual BT are doing it arse about face but then I guess their view is that they can fleece us by charging for premium speeds that they don't then deliver for much of the time, as in my case.

  2. John Sager

    Yet more RFI further up the spectrum

    And they're *still* trying to sweat ancient copper (and Al) assets. Now we're down pretty much to the overhead drop from the DP. Why not just bite the bullet & run PON fibre from there?

    1. Anonymous Coward
      Anonymous Coward

      Re: Yet more RFI further up the spectrum

      John, I couldn't agree more.

      Wouldn't it be great if someone had a long term view of this? Dig the roads up now, lay some fibre and have infrastructure that'll last for 50 years.

      Rather than trying to get every last Mb out of the copper or aluminum.

    2. Anonymous Coward
      Anonymous Coward

      Re: Yet more RFI further up the spectrum

      Forget PON - go for full fiber and get rid off all bandwidth bottlenecks for the foreseeable future.

      1. FlatSpot

        Re: Yet more RFI further up the spectrum

        I expect BT will be saving as much money as possible to store up a war chest to defend against a Deutsche Telekom takeover

        1. Anonymous Coward
          Anonymous Coward

          Re: Yet more RFI further up the spectrum

          "I expect BT will be saving as much money as possible to store up a war chest to defend against a Deutsche Telekom takeover"

          But not too much. All the board will want is a war chest sufficient to mount a credible feint, such that DT increase the buy off payments to persuade the directors to recommend the offer. The verminous directors who sold Cadbury's to a maker of burger cheese have shown how this is done.

  3. Anonymous Coward
    Anonymous Coward

    G.fast is going to end up like supporting 5 versions of Windows -XP,Win7,Win8,Win8.1,Win10

    This sounds like its going to be a practical nightmare to support, to get theoretical max 300Mbps throughputs, just to keep using their huge copper (there's a lot of aluminum too) assets.

    BT are forgetting we need complex equipment on the premises that has to interact with what is likely to end up a support nightmare, the equivalent of supporting 5 Different Versions of Windows, along with 5 Different Versions of Macs. Attempting to update equipment, BT have no physical access to if an update fails to apply, then left communicating at the old protocol, producing unacceptable crosstalk.

    You move home, your old VSDL router connected at the equivalent protocol of 'Win10', (my analogy) the new home uses a 'Win7' protocol Cabinet (hasn't yet had an update). The subscriber uses it anyway, manages to connect. This interferes with the neighbours connection, causing crosstalk, but crosstalk due to connecting using the wrong protocol for that cab. Or the cab uses the other ECI design, not a Huawei one, which isn't 'exactly' compatible, giving very low throughput rates.

    BT is going to have to keep all their pre-installed managed Huawei/ECI VDSL Modems up to date, keep a complete inventory of models in use, which then allows BT a monopoly on the premises supplied equipment. We're back to the day's of using one telephone and one socket, circa 1970's supplied by BT.

    Can't see BT supporting the vast array of VDSL routers over time, i.e. Netgear D6400, or a ASUS DSL-N66U. The newly released Netgear firmware supposedly doesn't even support g.fast yet.

    I'm fine with all this if BT are paying but Taxpayer funded? NO WAY.

  4. Anonymous Coward
    Anonymous Coward

    Stuff G.fast!

    Can I just have something faster than ADSL2+ please BT?

    1. Anonymous Coward
      Anonymous Coward

      Re: Stuff G.fast!

      AC, if you live more than 500m by cable from the nearest FTTC cab (thats about 250-300m as the crow flies), none of these headline grabbing G.fast plans will ever reach you, the maths (costs) just doesn't add up.

      Its about time we started laying Fibre to you, by whatever means possible - doing the job once and doing it well - because its going to take a very long time to connect everyone to these sort of speeds.

      G.fast is only good for the people already getting 50+Mbps on FTTC. In a NotSpot with FTTC?, you'll be in a NotSpot for G.fast.

      It needs a vast rollout of FTTrNs + FTTC for this to be viable, a minimum of 8 FTTrN+1 FTTC, per 2km2, for partial coverage. 16 FTTrN+1 FTTC per 2km2 to get blanket coverage.

      Rurally, 'Ad hoc' Fibre bright yellow cabling left loose down roadsides, could do a better job and much cheaper.

      Over time, this could be placed into ducts, from subscription revenue, but would allow the network to be up and running quickly. Rurally, cabling is less likely to be subject to vandalism.

      1. Anonymous Coward
        Anonymous Coward

        Re: Stuff G.fast!

        "Rurally, cabling is less likely to be subject to vandalism."

        I'd have thought that the key challenge was accidental damage from vehicles parking on the verge, road accidents that chew the verge up, grass cutting, ditch clearance and drainage work, hedgerow cutting, and other causes like road repair plant and construction materials dumped on the verge.

        I'd guess that makes for a very high revenue cost for repairs, even if the cabling is modest lengths with plug and socket terminations.

        1. Anonymous Coward
          Anonymous Coward

          Re: Stuff G.fast!

          BT in the form of BT Openreach has done hardly any maintenance of its rural infrastructure themselves for years, its mostly been left to rack and ruin. There are plenty of telephone exchange buildings looking dilapidated/unpainted since built in the 1970s. The only time some manhole covers have been lifted in the last 30 years, is because of tax payer funded BDUK installs.

          So called 'Fibre rollout', but it turns out that 0.5mm copper is been replaced by 0.9mm copper to 'just' so meet outer targets, to prevent having to install real FTTP to those hamlets.

          Instead of G.fast, maybe we should drop a few test Fibre Optical cables between two or more villages in a roadside ditch or along a river embankment/edge and see how long it lasts?

          Its only got to last between 2 and 5 years, enough time to prioritise the areas where the cables need to properly installed, from revenues gained. The ones getting damaged you target for install. Do you think vast arrays of g.fast technology won't require maintenance?

          An Ad-hoc approach in Rural Wales for instance, is actually quite a good idea. Certain locations between towns, across farmland/hedgerows never see the light of day, by anyone. Public footpaths could be targeted, but there's always a chance a farmer could cut it, so its worth letting local farmers/local authorities know of the their existence.

          Better than G.fast, by miles. G.fast is a non starter rurally.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Other stories you might like