» ISP News » 

BT Openreach Start Inviting ISPs to 500Mb NGA2 G.fast Broadband Trial

Friday, July 17th, 2015 (2:16 pm) - Score 5,073

BTOpenreach has told ISPreview.co.uk that the first large-scale residential trial of their future G.fast broadband technology (details), which the telecoms operator expects to sell via the Next Generation Access 2 (NGA2) tag, will begin next month in Huntingdon (Cambridgeshire) and ISPs are now being invited.

As a quick recap, Openreach hopes to begin the decade long commercial deployment of G.fast (ITU G.9701) technology in 2016/17 and its eventual aim is to make broadband download speeds of ‘up to’ 500Mbps available to “most homes” across the United Kingdom (initially the top speed may be capped at a lower level, such as 300Mbps like FTTP).

G.fast itself works in a roughly similar way to the operators current 80Mbps capable Fibre-to-the-Cabinet (FTTC) service, except that it requires significantly more radio spectrum (FTTC = 17MHz vs G.fast 106MHz+) and must thus operate over a much shorter run of copper cable (ideally less than 350 metres).

The service can reach many homes by simply being installed inside / alongside an existing street cabinet, although in other situations (i.e. where the property resides further away) then BT’s high capacity fibre optic lines will need to be moved even closer. At this point the fibre may be taken to a smaller remote node or distribution point (FTTdp), which can also be built on top of a telegraph pole or possibly even put underground.

g.fast broadband bt network diagram

The initial trial locations are no secret and we’ve covered them before (here), with the first one in Huntingdon (Cambridgeshire) expected to involve around 2,000 premises. A similar trial will also take place in Gosforth (Newcastle) and Openreach plans a third / smaller “technical trial” for Swansea (Wales); some FTTP and FTTP on Demand (FoD) upgrades may also be tested.

Openreach has now started inviting ISPs to the first “free trial” in Huntingdon and we’re told that a number of different providers have already confirmed their participation. The related ISPs should begin inviting some of their customers in the relevant area(s) over the next few weeks.

As you’d expect the purpose of the first trial is simply to test the deployment processes and the technology itself, as well as the speeds and experiences it delivers for customers. We’re unlikely to see any preliminary pricing details or results until well after the summer, but we will of course aim to keep our readers in the loop as soon as anything leaks out.

Mind you G.fast isn’t without its pitfalls. The technology suffers a big performance hit when forced to work alongside existing FTTC (VDSL2) services and the upgrade itself could still carry a fairly big cost. On top of that nobody has clarified whether “most homes” means UK coverage of a little over 50% or significantly more (we suspect it may reflect something akin to FTTC’s second 66%UK deployment target, but that’s a guess).

The little bit of copper in G.fast’s diet is of course still the biggest hindrance to performance (i.e. the signal degrades rapidly over distance) and reliability, but even so it looks likely to deliver a big boost over current FTTC services. BT’s prior closed trials have reportedly been pleasantly smooth, which hopefully bodes well for future deployments.

Leave a Comment
39 Responses
  1. Avatar AndyH says:

    Small correction, but it’s the number of homes/premises passed rather than those actually involved. It’s expected there will be 400-500 g.fast and 50-80 FoD trialists in Huntingdon.

    1. Mark Jackson Mark Jackson says:

      Hi Andy,

      Do you have a verifiable source for that? The documents from Neil McRae’s recent presentation only states “Around 2000 premises will be covered in each pilot area”.

    2. Avatar AndyH says:

      This info has been briefed to ISPs a number of times now from Openreach. Ian Boothman (senior NGA product manager) being one of the main briefers.

    3. Avatar NGA for all says:

      @Andy – with a fibre bundle being delivered to a DP will customers have the choice of G.FAST or ordering FTTP?
      With fibre being delivered that far is there any reason why G.FAST and FTTP cannot be ordered from the same DP?

    4. Avatar AndyH says:

      Of the 120 or so postcodes covered by the trial in Huntingdon, around half will have the option of chosing g.fast or fod. Around 20 post codes will be FoD only and the other 40 will be g.fast only.

      The trial will test the deployment of g.fast units at existing fibre DSLAMs and also deeper in the network.

    5. Avatar MikeW says:

      “With fibre being delivered that far is there any reason why G.FAST and FTTP cannot be ordered from the same DP?”

      When you mention FTTP in that post, are you generically asking about the technology, or are you asking about the current native FTTP product (compared with some future FTTPoD product variant)?

      The difference between the two products becomes the installation cost, and the timing of deployment of some of the physical infrastructure.

      Most of the work in a full FTTP rollout comes in the final lead-in to the property … i.e. the part from the DP. So even if the fibre has reached the DP, we should expect there to still be a significant cost to going further.

      In practical terms, the extra infrastructure would (by the existing architecture) require BFT tubing to a pole-mounted, or chamber-sited manifold; this would be pre-installed for native FTTP, but would be deployed on-demand with an FTTPoD variant. Both would need the subsequent BFT lead-in/drop-wire.

      Any reason an FTTP product could not be ordered would then be down to cost.

      However, you start with an assumption that fibre will have reached the DP; I think that is an invalid assumption. From Sckipio, we know BT are targeting longer range for G.fast, suggesting that they want the nodes to be further away than the DP.

    6. Avatar NGA for all says:

      @MikeW I was attempting to find the answer.
      The customer will be paying the final connection charge anyway.
      I guess I am still struggling with putting a potential point of failure so deep in the network, when demand aggregation ought to dictate these installs.

    7. Avatar NGA for all says:

      @AndyH thanks for confirming one or the other. I assume any deployment will be require demand registration in advance of a deployment. Would customers get a choice G.FAST or FTTP?

    8. Avatar MikeW says:

      From what I’ve seen, participation in this trial is free to the end-users.

      If there is no cost impact to the end-user, then you might find that, with a free choice of G.fast or FTTP, they’d opt for FTTP.

      While the main driver for the trials is BT’s wish to discover the parameters for a G.fast rollout – with a secondary driver to discover the costs of an FTTPoD extension – I suspect that BT will have to enforce the choice for most locations.

  2. Avatar adslmax Real says:

    Don’t care! Don’t think everyones on FTTC will get G.fast on all FTTC cabinets in UK by the year 2016/17. No chance. There will be few select choice by BT Openreach. Not everyones will get it.

    1. Avatar FibreFred says:

      Correct, no-one said it would, it will START in 2016/17

      Good to see the trial moving on

    2. Avatar Ignition says:

      Don’t think anyone even said that all people on FTTC cabinets will get G.fast full stop?

    3. Avatar Ignition says:

      If it helps with your not caring I seriously doubt that you’ll be in the early phases of the G.fast rollout so you don’t have to worry as Openreach probably won’t especially care about Cuckoo Oak.

      It was at the back end of the commercial deployment and take up hasn’t exactly been amazing.

  3. Avatar james says:

    Why is it always in these tooty snooty places? Why not give the rest of us a chance!

    1. Avatar adslmax Real says:

      That’s BT for u! They don’t care!

    2. Avatar Ignition says:

      Swansea is ‘tooty snooty’?

      There are places a whole lot more ‘tooty snooty’ than these. If they wanted to go full-on ‘tooty snooty’ they’d deploy in Westminster or Kensington and Chelsea, or outside London Windsor, Weybridge or Sevenoaks, perhaps Henley on Thames.

      In a country with over 27 million premises trials covering 4,000 and change are going to leave a quite large proportion of the country out.

  4. Avatar MikeW says:

    @Ignition’s recent comment – on rate of take-up of one exchange – raises an interesting question here…

    What will Openreach’s rollout strategy be?

    The FTTC rollout has been governed by a lot of hidden factors, but essentially it seems to have been led by Openreach choices over making the rollout cheap and efficient.

    But with FTTC largely out there already, and largely satisfying current speed demands, will a similar rollout policy be best for G.Fast? Will Openreach aim for doing work in a cost-efficient order, and hope that take-up follows?

    Or will they choose to go where FTTC hasn’t worked well – at the fringes of coverage?

    Or will it be a demand-led thing? Register with 15 neighbours? Or perhaps FTTdp-on-demand?

    Or perhaps target VM’s expansion? Or rollout of Docsis 3.1?

    I’m not sure what they’ll go for, but I can’t help feeling that something demand-led will be best.

    1. Avatar Ignition says:

      Demand-led would probably be good rather than the approach taken with FTTC which appears to have been deploying at lowest cost per premises passed or for political reasons.

    2. Avatar Sunil Sood says:

      At a guess, any deployment will be a combination of 4 things:

      1. protecting itself against VM + their expansion plans;
      2. in BDUK areas where they need to reach coverage targets;
      3. depend led commercial rollout;
      4. the costs of deploying it in a particular location

      with probably 2/4 and 1 taking priority..

    3. Avatar DTMark says:

      How much more money has Ed Vaizey already promised BT? Since people are talking about a G.Fast rollout. There isn’t going to be any meaningful rollout without public money. And I’m wondering why anyone is talking about one.

      Even the BT guy said recently that no plans at all have been announced, yet this discussion might imply that some semblance of a plan exists.

      Areas to be targeted commercially would be rather restricted to business parks with alt nets robbing BT of their cash cows (bespoke circuits) since making some money is probably better than making no money.

      It would be tempting for BT to deploy in some cabled areas, though I wonder if they got somewhat stung rolling out VDSL to cabled areas because we don’t have any idea of the take-up figures and they may be extremely low.

    4. Avatar FibreFred says:

      “There isn’t going to be any meaningful rollout without public money”

      Why not BT rolled out FTTC with its own money?

    5. Avatar MikeW says:

      “Even the BT guy said recently that no plans at all have been announced, yet this discussion might imply that some semblance of a plan exists.”

      *This discussion* might imply some semblance of a plan exists?

      I think you imagine that the online ramblings of a bunch of enthusiasts has somewhat greater power than reality.

      On the other hand, there must be enough of a semblance of a germ of a plan for them to decide it looks like a cost-effective option. Enough for the CEO to make a strategic announcement, anyway.

      IMO, I imagine that they have created no distinct plans as yet (on a nationwide basis, at least). But I imagine they have a strategy in mind, from which a plan could be generated once more of the variables have been honed down. I guess BT will hope that some answers will come from the trials. And that some answers will come from chipset manufacturers trying to extract longer range from G.fast.

      I suspect we’re a long way from a real plan.

    6. Avatar Ignition says:

      I imagine it would use the same guarantee that VM’s Project Lightning is using.

      As far as DOCSIS 3.1 goes, VM have no need to deploy that to deliver speeds superior to the 500Mb BT have mooted on G.fast.

    7. Avatar MikeW says:

      “How much more money has Ed Vaizey already promised BT? Since people are talking about a G.Fast rollout. There isn’t going to be any meaningful rollout without public money. And I’m wondering why anyone is talking about one.”

      I think you’re wrong, because you’re looking at this in the wrong direction.

      You could be 100% right that BT may never deploy a meaningful G.fast rollout without public money … But you’re wrong in then assuming that public money will therefore fund BT to perform that rollout.

      The simple answer here is that Virgin are expanding to two-thirds coverage, and will soon increase their highest headline speed option to 300Mbps … and that ignores an upgrade to Docsis 3.1 that they could choose to employ later.

      With those plans in play, you can be sure that the government will not be able to subsidise any part of a BT rollout for two-thirds of the UK, if that rollout is of broadly similar speeds. The EC just wouldn’t allow it – and VM would use the courts to stall things just as much as happened with ultrafast in Birmingham.

      If BT want to continue to compete in that part of the market – which has previously been the most competitive because it is also the easiest to reach – then they will have to self-fund a rollout.

      Will they? No idea. But if it happens, the money won’t be coming from any Ed Vaizey promises for state aid.

      Use of the government’s “UK Guarantee Scheme”, however, might be included; this was said after an earlier budget that mentioned the VM expansion:

      UK Guarantees Scheme for Virgin Media

      Government can also act directly to ensure that it enables the rollout of ultrafast broadband. One example is through the UK Guarantees Scheme, which is in place to help infrastructure projects raise debt finance; the government has announced that it has prequalified the expansion of Virgin Media’s ultrafast broadband network for the UK Guarantees Scheme, supporting Virgin’s proposed £3 billion investment.

      There is capacity within the £40 billion scheme to support significant further investment, and we are actively engaging with UK broadband operators to explore how the UK Guarantees Scheme can be used to support and accelerate their respective investment programmes.

    8. Avatar MikeW says:


      Good point on the use of G.fast for BDUK infill to reach targets.

      It would certainly help do that, but it would be better-placed to do so with reverse-power in play.

      I think the reverse-power feature is going to be slower to mature than G.fast itself, so I suspect that FTTdp nodes will appear a year or two too late to help BDUK coverage.

    9. Avatar DTMark says:

      If that coverage is for meeting minimum targets specified, this must leave BT bound to deliver FTTP as G.Fast won’t be ready in time?

      G.Fast could never have been a proposed option for any BDUK projects because it’s only just being tested now after all the contracts were signed.

      So without that, there must be the money to run FTTP as that necessarily is what BT bid to roll out to meet those targets.

      Which is better anyway. So BDUK would seem to be an irrelevance in this context.

    10. Avatar MikeW says:

      Isn’t BT’s solution-of-choice for BDUK infill known as FTTRN? Mini VDSL2 DSLAMs.

      This wasn’t deployable as a widespread solution last autumn because of power issues. Who knows where they are with it now.

    11. Avatar MikeW says:

      I’ll add that SFNY signed their phase 2 contract with BT knowing that FTTRN wasn’t, at that stage, a viable technology. But that it *might* end up being used.

      The phase 2 budget, and the plans announced so far, allow for a high percentage of FTTP – and a consequent low number of premises passed. If BT makes a go of FTTRN in time, then SFNY expect a lot more premises to be passed.

  5. Avatar themanstan says:

    From the view point of technical compatibility and current technical issues, replacement of the ECI equipped cabs would have a rational priority… why spend time and money swapping out technically dead end M41 cards, when taking the next technology step would bring hardware “harmony”.

    1. Avatar Ignition says:

      Or just don’t swap out technically dead end M41s if they are working okay and achieving what the company / BDUK contracts require of them.

    2. Avatar themanstan says:

      Depends whether BT takes the reactive or proactive strategy.
      These are the most “vulnerable” of the commercial roll-out (no BDUK), those cabs in VM areas will see performance degradation through crosstalk as the cabinets continue to populate. The lack of compatibility with vectoring, even G.INP means that they present a likely customer loss situation to VM… particularly if they take a reactive pose.

    3. Avatar Carl says:

      VDSL is an odd choice full stop if VM is available and a customer is disappointed by the loss of a few Mb.

      The estimates now incorporate a substantial amount of segregation due to crosstalk. Even with 400 lines worth of potential crosstalk I remained in the middle of mine.

    4. Avatar themanstan says:


      Your speaking as a reasonably well informed individual… plenty of people only consider headline speeds, rather than sustained.

      Also, plenty of impacted lines out there…

    5. Avatar MikeW says:


      There’s a commercial imperative at work too…

      If a cabinet gets somewhat impacted by crosstalk, then BT can spend money upgrading to vectored cards – but at no additional income.

      However, if they add a G.fast node, then *some* of the people will upgrade to G.fast to overcome the crosstalk issue, which (presumably) triggers more income. It also reduces crosstalk in the FTTC cab. A double-win!

      That observation is true for a customer motivated by advertised headline speed. The Ofcom tests show that VM speed loss is worst at peak hours, and worst for 152Mbps lines, suggesting a fair impact from congestion … so such a person might complain on that side instead.

      From my perspective (which appears to match a limited set of people), I’m driven more by the problems that such congestion causes to latency … even driven by just the risk that this can happen. Horses for courses, I guess.

    6. Avatar DTMark says:

      “However, if they add a G.fast node, then *some* of the people will upgrade to G.fast to overcome the crosstalk issue, which (presumably) triggers more income. It also reduces crosstalk in the FTTC cab. A double-win!”

      Or, they could just go back to cable and leave BT’s circuit dangling in the wind for another decade. 4G is another threat. I did consider ordering VDSL to see if it was adequate, if I did and it fell very far short I’d be inclined to charge back the money paid and go back to 4G, not give BT any more money. That damn competition, limited though it is, eh?

      The other dynamic that’s not been discussed is the difficulty in marketing.

      How do you market G.Fast when any deployment will be highly localised?

      How do you market the HD TV services? “You can have it if the line length is 400m or less to one of our vectored cabs”.

      It’s so sporadic and leaves BT with the same sort of marketing and visibility issues that alt nets have.

    7. Avatar FibreFred says:

      So by your logic DTMark, everyone in a VM area should be using VM and nothing else, not ADSL2 or FTTC.

      But.. that doesn’t happen does it

    8. Avatar DTMark says:

      Anyone who wanted speed probably migrated to cable years ago. That will be one of the difficulties BT have in appearing to be relevant to people and useful when they gave up with it a decade ago and how have products which remain inferior in speed terms to what they already have. Unless they’re right next to a VDSL cabinet in which case they might be able to get a few more meg upstream if they’re lucky at the expense of losing half the downstream.

      The other dynamic which begins to come into play: people who tried VDSL, saw out the 18 month contract, and then/now move back to cable. Would be interested to see those figures 😉

    9. Avatar MikeW says:

      By this picture, there’s probably only about 10-15% of the VM subscribers who would fall into your group of “anyone who wanted speed” who probably migrated to cable:
      That group will probably account for 3-4% of all broadband subscribers.

      “Right next to a VDSL cabinet” is an interesting term. 80 Mbps seems to be available, in real life, out to 400m – which means around 40% qualify as being “right next to”. Speed, distance, and percentage all improve if they ever go for vectoring.

      Another perspective is that BDUK projects seem to be reporting an average speed of 50Mbps; that speed would appear to appeal to 65%, possibly up to 80%, of VM users.

      On balance, however, I’d say that those FTTC vs VM curves suggest the two markets aren’t currently out of step with each other, or particularly out of step with consumer demands.

      4G is indeed competition. It will be the obvious choice for a few, like yourself. But a last resort for many too (I reckon this group is bigger), kinda like satellite. Sharing will always be the bugbear for 4G, just as it is for satellite.

    10. Avatar FibreFred says:

      So DTMark, why do VM cable customers migrate to FTTC? Or does that simply not happen?

Comments are closed.

Comments RSS Feed

Javascript must be enabled to post (most browsers do this automatically)

Privacy Notice: Please note that news comments are anonymous, which means that we do NOT require you to enter any real personal details to post a message. By clicking to submit a post you agree to storing your comment content, display name, IP, email and / or website details in our database, for as long as the post remains live.

Only the submitted name and comment will be displayed in public, while the rest will be kept private (we will never share this outside of ISPreview, regardless of whether the data is real or fake). This comment system uses submitted IP, email and website address data to spot abuse and spammers. All data is transferred via an encrypted (https secure) session.

NOTE 1: Sometimes your comment might not appear immediately due to site cache (this is cleared every few hours) or it may be caught by automated moderation / anti-spam.

NOTE 2: Comments that break our rules, spam, troll or post via known fake IP/proxy servers may be blocked or removed.
Cheapest Superfast ISPs
  • NOW £22.00 (*32.00)
    Speed 36Mbps, Unlimited
    Gift: None
  • TalkTalk £22.00 (*29.95)
    Speed 38Mbps, Unlimited
    Gift: None
  • Vodafone £22.00 (*25.00)
    Speed 35Mbps, Unlimited
    Gift: None
  • Hyperoptic £22.00
    Speed 50Mbps, Unlimited
    Gift: Promo Code: EASTERSALE
  • xln telecom £22.74 (*47.94)
    Speed 66Mbps, Unlimited
    Gift: None
Large Availability | View All
Cheapest Ultrafast ISPs
  • Vodafone £26.00 (*29.00)
    Speed: 100Mbps, Unlimited
    Gift: None
  • Virgin Media £28.00 (*44.00)
    Speed: 108Mbps, Unlimited
    Gift: None
  • Hyperoptic £29.00 (*35.00)
    Speed: 150Mbps, Unlimited
    Gift: Promo Code: EASTERSALE
  • TalkTalk £29.95 (*39.95)
    Speed: 145Mbps, Unlimited
    Gift: None
  • Leetline £34.99 - 43.99
    Speed: 100Mbps, Unlimited
    Gift: None
Large Availability | View All
The Top 20 Category Tags
  1. FTTP (3154)
  2. BT (2907)
  3. FTTC (1843)
  4. Building Digital UK (1842)
  5. Politics (1820)
  6. Openreach (1736)
  7. Business (1576)
  8. Mobile Broadband (1366)
  9. FTTH (1355)
  10. Statistics (1335)
  11. 4G (1175)
  12. Fibre Optic (1121)
  13. Wireless Internet (1097)
  14. Virgin Media (1089)
  15. Ofcom Regulation (1086)
  16. EE (775)
  17. Vodafone (767)
  18. TalkTalk (729)
  19. Sky Broadband (711)
  20. 5G (655)
Helpful ISP Guides and Tips

Copyright © 1999 to Present - ISPreview.co.uk - All Rights Reserved - Terms , Privacy and Cookie Policy , Links , Website Rules , Contact