The NBN FTTdp option in detail


blog If you’ve been following Australia’s national broadband debate for some time, you’re probably familiar with most of the “fibre to the” terms. Fibre to the Premises is what Labor wanted to do with its National Broadband Network policy, Fibre to the Node is the watered down Coalition alternative and Fibre to the Basement is what most of the telcos want to build to apartment buildings. But what about Fibre to the Drop Point (FTTdp)? The concept, which would see fibre extended to the lead-in pit in front of Australian premises but the existing copper reused from that point on, has been explored in depth an extensive article published by the journal of the Telecommunications Society of Australia. A selection of the paper’s conclusion:

“The 2013 Australian federal election campaign presented an FTTP option against FTTN. The former has been criticised as too costly, while the latter may not provide a sufficient ‘future-proof’ capability. Following the lead of Gregory (2013), this paper presents the third option of FTTdp. It is suggested that FTTdp may provide network capability close to that of full FTTP, with cost similar to FTTN. FTTdp may provide optimal ability to meet the unknown, yet high, anticipated bandwidth demand of the future, with comparatively low initial capital outlay.”

I don’t know enough about the Fibre to the Drop Point option to be able to comment about its appropriateness in the context of the Coalition’s Broadband Network policy or Labor’s National Broadband Network policy. However, I do know that a lot of the effort involved in either policy is not in getting fibre down streets, but in getting fibre to actual premises. In this sense it seems that FTTdp may represent something of a worthy option that will allow fibre to be deployed along Australian streets while leaving much of the difficult work a little further down the track. Thoughts? And please read the journal article before commenting, or else I will not let you comment ;)

Image credit: Clix, royalty free


  1. I’ll take 3.

    Makes sense really – the householder can have good speed VDSL on short (presumably higher probability of quality) copper length, but be able to upgrade to full FTTP if they need to.


  2. In my books the further the fibre goes the better. That said if it reduces the time it takes for it to be deployed Australia with and it upgradable then it sounds great. Once this has been underway for a time NBN Co can go back and continue deploying the last section to give proper FTTP while also giving some upgrade to more premises sooner than if it went straight to FTTP. The fttdp equipment could then be reused where the first rollout is still underway.

    • Not from what I’ve read. Apparently the nodes involved in FTTdp are roughly the size of a shoe box – small enough to be attached to electricity poles. Also, FTTdp is like FTTN in that it still uses the copper, but it gets much closer to the premises – less than 200m at most. It also services fewer houses per node. Finally, it does offer a proper upgrade path to GPON FTTP (even on demand, which should make Mr Turnbull happy).

      If we can’t have FTTP, FTTdp seems like a reasonable enough compromise. I’m sure it still has its drawbacks (nothing at this point can truly compete with FTTP) but since the Coaltion’s Broadband Network demands we use the CAN, we may as well try to get the most out of it that we can. I can’t see MT backflipping on his anti-FTTP stance ever, but maybe, JUST MAYBE he can be persuaded that FTTdp is a wiser investment than FTTN. I’m not going to hold my breath but who knows.

    • Between FTT Node and FTT Premises, we have FTT Pit. It makes excellent sense from a technology/cost standpoint, and it removes a major bugbear — care and feeding of FTTN cabinets. No graffiti, no battery thefts, no separate power system, and a lot less copper involved in the overall run.

      Plus, I could probably afford fibre from the pit to the house as a homeowner; fibre from a node to the house would be potentially quite costly, and severely limit the options available to many. And given that I’m eyeing high-speed bandwidth as a major factor in the value of our house, I see an all-fibre option as being cost effective in increasing our property’s value. Fibre to the pit would make it easier for me to complete an all-fibre link.

      But it’s Telstra pits I see being upgraded near our home. Would we still have to deal with Telstra for that last link, or can we deal directly with one of the alternates, such as Optus or iiNet?

  3. So what about current aerial deployments? It would be easier to run a fibre cable into my house than to try and bolt a ntd to the pole!

  4. Still does not fix the problem that the copper is often in poor condition in the last section to houses.

    That being said, if it’s easier to upgrade to FTTP from FTTdp, then maybe houses with bad copper can get upgraded to FTTP right away and skip FTTdp. That seems like an efficient way to go about it.

    This might be a good compromise between the two sides. I would rather a compromise like this than the current utter chaos and uncertainty we have.

    • True, but at least replacing the copper for 200 metres is significantly cheaper than replacing it for a kilometre. It may even make the realm of paying for your own FTTP viable for many more.

  5. We don’t have much Cable TV in Australia, but image if cable had taken this approach: Run the cable down the street and then deliver the TV signal to each house via the existing copper. The technology was available decades ago, so why didn’t they do it?

    I suspect there were two main problems:

    (1) The cost of the box outside each house: If you are running the cable overhead (aerial) then it’s cheaper to use a splitter and run the coax all the way.

    (2) The bandwidth of the copper pair: This would limit you to one or two channels at a time.

    The same analysis should also apply to Fiber to the Drop Point.

    The crucial factor is running overhead cables. If you run the cables overhead, then even FTTH becomes much cheaper.

    Come to think of it, if it were practical to run coax to each house decades ago, how come it isn’t practical to run Fiber to each house today?

    Unless of course you are actually trying to LIMIT the bandwidth to each house.

    • Because the companies running HFC cherry picked the most profitable areas, so cost of installation was cheaper. There is a reason Telstra/Optus stopped running HFC everywhere, if they’d just rolled out to the whole of Australia like originally intended, the whole NBN debate probably wouldn’t be happening for a few more years.

      The Coalition don’t want to wear the cost of installing to every individual house, the cost of deployment increases exponentially per house as distance increases, again, there is a reason that the wired part of the NBN was originally intended reach 93% of the population, as the costs exponentially increased for every 1% beyond that (reducing it down to around 70% halved the cost!).

      That’s just factoring installation costs, which would be the main costs of the original NBN. Unfortunately the Coalition want to install just to a node and leave the copper in place, not taking into account the large maintenance costs involved with the copper and the expectation that Telstra will just gift them the copper for free.

  6. There are MANY problems with the costs outlined in this paper

    1) The cost stated as the last 200 m for FttP is Far Far too high the last cost per connection was around $2400 are they really saying that pulling 200m of cable is costing 63% of the total build cost

    2) Figure 3 states you need to add ethernet wiring to your home for FttP

    3) Figure 3 implies there is major construction work when all that is needed is to pull the fibre down the already existing infrastructure

    4) Says there are months of delays for FttP but no work is needed to supply Fttdp even though you still need to roll out fibre.

    So how can we believe the costs if there are a number of misleading statements already?

  7. The best way to deploy a system like this, just terminate fibre on an Ethernet port outside each house in a small metal post on the property boundary, and the householder or the ISP wire cat5e/cat6 to the post on the street.

        • Thats just silly, only 2% of Australian households are equiped with a gnome according to the latest statistics. Also we need to take into account not all gnomes are compaitible with this type of service, core filled gnomes will require exstensive remediation and the government would be liable for any damage.

  8. What is described is basically FTTP. They werent installing the final connection to your house until you signed up and payed for your connection anyway. As far as i am aware your ISP charges you for the final install from dp to your home. So it wont save the government any money it will cost the same as FTTP. But hey if it happens i wont complain they can call it whatever the FTck they want as long as they do it and soon!!

  9. This sounds so much like it could benefit FTTP.

    In my case, with a 6 y.o unit, there was a perfectly good 25mm conduit from the street pit under the long drive to my unit wall. NBN just pushed a rod thru and reported the distance to me (64m) and a couple of weeks later pushed a fibre in from the pit. Next January, they snip the Telstra copper, and pull it back out of the same conduit. Done. Not a blade of grass was harmed.

    Getting the fibre to the pit might help speed up those who do have the ability to switch to fibre due to the right infrastructure under the lawn.

  10. MT did say Mini-Nodes as a throw away line last year. See:

    “..If the Coalition wins government, Mr Turnbull would instruct NBN Co to guarantee a 25 megabit per second download speed for every Australian household, and if necessary the company would build an extra series of “mini nodes” closer to houses to achieve the speeds, he said.”

    But then he has since backed away from his guaranteed 25 megabit per second download speed for every Australian household to an “upto 25mbps”.

    The Prime Minister said:

    “Under the Coalition, by 2016…there will be minimum download speeds of 25 megabits… We will deliver a minimum of 25 megabits…by the end of our first term.”

    “..That promise didn’t last a year. The government announced it was breaking it just before Christmas.”

    Taken from:

    So seems unlikely we will get ubiquitous FTTRN or FTTdp rollout here in Oz.


  11. I’m a huge fan of this architecure and I think NBN should implement it for a whole host of reasons:
    – Unlike MTM, it actually implements LNP policy in terms of the speed it delivers and allowing reasonably priced fibre on demand.
    – It adderesses the core criticism of the NBN: that it’s “wasteful” by delivering fibre to everyone whether they need/want it or not.
    – The last 50m might not sound like much, but it represents ~40% of the FTTP NBN build cost.
    – Virtually none of the investment is wasted with the inevitable move to FTTP.
    – It almost certainly addresses the constitutional issues NBN presents, as end users – not governments or telcos – make the decision to disconnect the copper cable.
    – The difference between FTTN and FTTdp architecures is FTTdp includes the “local fibre” component, the communal cable in which everyone in the street gets a dedicated fibre to the distribution hub. That core piece of infrastructure effectively makes everyone a member of a (well cabled) MDU.
    – Provided there is appropriately open access to the fibres in the distribution cable (fibre from the “node” to the aggregation point/exchange – which all FTTx architecures have), or the distribution conduits, it should be possible for genuine facilities competition to exist, in the exact opposite way FTTN explicitly prevents it.
    – With the demarc at the distribution point, end users can deal directly with their choice of installer to do whatever remaining cabling or civil works required on their property – rather than having to deal with the RSP/NBN/Contractor/Sub-Contractor/Telstra/TIO/”I’m not responsible” logistical mess we have now.

    To me, FTTdp makes a lot of economic and practical sense, but more importantly actually solves the last mile infrastructure deficiency we have now in a way neither FTTP (too slow/expensive) or FTTN (too inadeqaute) delivers.

    But I don’t think it will happen. FTTdp is the antithesis of the telco industry, which is characterised by governments or ex-government monopolies telling “customers” what they do or don’t need. Giving people a real choice – apart from being not in a telco’s financial interests – seems to be an anathama.

    • I couldn’t agree more TTR,
      FTTdp is a really elegant solution that resolves so many issues that Malcolm’s MTM has created.

      It meets just about all of the Liberals user pays philosophy, it allows an easy upgrade path to FTTH which even Turnbull and Ziggy have acknowledged is the end game and which will need to commence before the FTTN build is completed. It gets rid of the bulk of the worst copper from the system.
      Some of us have been having this discussion on Whirlpool since last October:

      I really hope than everyone will have a look at that thread, and Renai: I do hope you can follow up on this article in more depth.

    • – The last 50m might not sound like much, but it represents ~40% of the FTTP NBN build cost.

      Can you provide a source for this?

      • > > – The last 50m might not sound like much, but it represents ~40% of the FTTP NBN build cost.
        > Can you provide a source for this?
        The Strategic Review ? :-).

        That suggests the drop is 50%-60% of the cost of installing the fixed line component of the network, thus 40-50% of the total build capex. The amounts vary across the recent experience, projections and corporate plan – but the ratios don’t change all that much.

        For all it’s faults that document did suggest the experience vs the corporate plan was all build costs were lower than anticipated, except civil works which were higher. And this extra cost falls disproportionately on the LIC/drop installation.

        If you look at it another way: to connect a user to the OLT, you need ~10km of dsitribution fibre shared among 2000-3000 users – or 4m/user. You need 1km of local fibre, shared among 50 or so – or 20m/user. The drop cable is 20-50m (depends on the architecure) not shared with anyone, but say 20m/user. That’s still 45% of the cost (of cable laying) per user in the drop alone.

        • That seems either wildly inflated or dishonest

          Is the cost of a NTD included in the cost for both estimates?

          • > That seems either wildly inflated or dishonest

            Probably a fair call given I was quoting the NBN Strategic Review. But regardless on whether or not the abolsute numbers are inflated, the ratios remain the roughly same: about 40% of the total last mile connection cost is in the customer’s connection.


    Fibre to the ship that has sailed.

    I hate it; but Turnbull wasn’t even willing to consider the same-cost Fibre/HFC build (option 4? or whatever) but instead chose his politically motivated FTTN. Being defeatist sucks, and I’d love for this kind of deployment to be considered an alternative. But we are about 2 elections too late to consider it.

  13. How can the technology can be close to the cost of FTTN given 1) it’s reverse powered, 2) at least a magnitude more pits to work, 3) scrapping CPE cost equivalent to FTTN?

    Good article pointing out the obvious (often denied) lead in costs experienced by NBNCo and the wildly underestimated time required. Their conclusion is heavily dependent on bandwidth growth expectations; unknown.

  14. Interesting article. Especially if you take into account that the NBN already has pre-signed contracts that allow them access to Telstra’s ducts, conduits, etc. You could essentially change legislation to make the network boundary at the DP, rather than at the premise, essentially removing any requirement for the NBN Co to maintain the copper between the DP and the CP, pushing that cost to the end user.

    Not sure how Telstra would feel on that, they’d lost less of their network, but still a decent portion.

  15. In view of todays major announcement:

    Hypocrisy: Turnbull approves MTM NBN without cost/benefit analysis

    In all seriousness, is there any chance that NBN Co will consider FTTdp and not just FTTB and FTTN given that this shareholder Ministers document states on page 2:

    “NBN Co will trial Fibre to the x (FTTx) network architectures to inform the Company’s planning and decisions”

    I’m trying to find something positive in all this.


  16. The idea has promise, and it’d be a lot cheaper to upgrade it compared to FttN. It’d also make Malcolm’s “Fibre on Demand” brain fart realistically doable.

    But because it does make sense, and because Malcolm would not be able to give more money to Telstra using this, I doubt they’d seriously consider it…

  17. Wow, FTTdp looks awesome. Is this related to the new micronode technology? i.e. Much smaller FTTN passively cooled (i.e. low power) nodes that are located much closer to the premises but still avoiding the massive cost of replacing the last section of copper from the pit to the house with fiber?

Comments are closed.