back to article It's almost time for Australia's fibre fetishists to give up

Last week, BT and Alcatel Lucent let it be know that their experimental broadband-over-copper technology XG.fast had achieved 1.8 gigabits per second over a 100m copper strand. nbnTM, the company building Australia's national broadband network (NBN), has already trialled G.fast, the successor to VDSL and progenitor to XG.Fast …

  1. cantankerous swineherd

    most of oz is within 100 metres of an exchange I suppose, so yes, good plan.

    1. Simon Sharwood, Reg APAC Editor (Written by Reg staff)

      Yep 100m in the labs. Today. During early development.

      An optimist might suggest that's promising. Or even a remarkable achievement.

      1. dan1980

        @Simon

        "Or even a remarkable achievement."

        Do you know why it is a remarkable achievement? Because they have managed higher speeds on an medium inherently unsuited to it. No one is gaping at fibre achieving these speeds for a very good reason - it's designed to handle the fastest speeds that the equipment - whatever equipment - can throw at it.

        When it comes to it, the very reason this achievement is so remarkable is that it was done using a medium that is not designed for it. They are pushing the boundaries but you simply cannot push the boundaries forever.

        Further, that is cutting-edge stuff actually works against it for the purpose. VDSL2, is still new enough that there can be issues with interoperability and limited choice. Achieving these speeds over fibre is de rigueur and the equipment to do it is mature, stable and well understood.

        What this means is that to get copper to approach speeds that are by-the-by with fibre, you have to use the bleeding edge of technology with all the pitfalls of expense, lock-in, bugs and, of course, cost.

        Don't get me wrong - if I had 1Gbps to my home, that would be bleeding amazing but that really is not the point. It's about choosing the best technology/topology for the job and FTTP ticks all the boxes. The boxes that FTTN + copper ticks are price and speed of deployment but both of these are based on massive assumptions about the existing state of the current infrastructure as well as very selective reporting of on-going costs.

        Whatever the speed achieved by copper, the running costs of an active FTTN node will never be able to compete with the running costs of a passive fibre distribution node and nor will the maintenance costs of copper be able to compete with that of fibre.

      2. Myvekk

        But will they ever get around to updating the copper from the exchange/node to the user?

        I have to force my modem to use ADSL instead of ADSL2/2+ in order to achieve a reliable connection & it INCREASED my connection speed! At this rate, XG.fast will put me at 1200/75bps...

        (As also commented on by Dan1980 below.)

    2. Daniel Voyce

      I'm having a Sheldon Cooper moment. That was sarcasm right?

    3. A Dawson
      Meh

      Maybe Maybe not

      Umm .... not sure if sarcasm re closeness of exchanges .... plus rotting copper (yay)

      1. Fluffy Bunny
        Thumb Down

        Re: Maybe Maybe not

        The copper could only be deteriorating because Telstra is refusing to maintain it properly. Now that a mixed mode solution is on the cards, they are demanding the government pay for them to do the job they should have done in the first place.

    4. dan1980

      @cantankerous swineherd

      Appreciating the comment, it is important to note that the connection will not, in the majority of planned instances, be between the premises and the exchange but between the premises and the node.

      Thus the question is not how far away the exchange is but how far away the node is.

      These higher-bandwidth connections, like VDSL and VDSL2 and indeed like ADSL2 drop off sharply. Thus at about 1km (of cable) there is little difference but even 500m drastically lowers the benefit.

      The upshot is that to get an appreciable benefit from these higher-bandwidth technologies, you have to keep the distances short and thus deploy a lot of nodes. And, as I mentioned in my post below, each node is a collection of active devices, which need to be powered and cooled and maintained.

      For another look at one of the FTTN nodes, this close-up shows some of the DSLAM. Note the fans (there are more above, cooling the whole unit), the numerous switches and half-dozen or more different cable types - just in this one small section.

      An upgrade to technology requires that each of these nodes be upgraded, replacing hardware and reconfiguring. If lucky, that just means a straight swap but the new units may have different power needs that then requires upgrades to that section of the node as well. It's also more training for techs, which costs more. You'd also need a new gateway device at the premises. So yeah, there are possible future upgrade paths but they require time and money to implement.

      This idea of starting with lesser devices and technology and then upgrading as time goes on results in higher costs as hardware is replaced and a patchwork network of old and new technology, where new sites are installed with newer, faster equipment while older sites and therefore older customers are stuck with older, slower equipment.

      The older equipment is then selectively replaced, but only in those areas that are more 'commercially feasible'. Some locations will have the equipment only partially replaced, leading to nodes with a mixture of ports - some faster, some slower.

      The end result is a repeat of what happens right now with Telstra - which was something that a homogeneous, all-fibre network was going to rectify so that we no longer had this ridiculous situation where the bandwidth you are able to achieve is dependent not only on the suburb you live in, but sometimes down to the street or even the block. (Or the timing of the installation where there are no high speed ports left.)

      That's something that the new NBN doesn't adequately address and is a huge part of what was important about the original plan - it's not just about the max Mbps a given technology and medium can achieve.

      1. david 12 Silver badge

        a homogeneous, all-fibre network

        >That's something that the new NBN doesn't adequately address

        >and is a huge part of what was important about the original plan

        That's something the old NBN didn't adequately address either. The /fibre/ was going to be one homogeneous network, but the /active electronics/ were out of date almost as soon as the NBN was proposed, and we've already seen one upgrade from the original plans.

        Not that it wouldn't be an improvement, but there was a large degree of fantasy even in the original engineering proposal put forward by Telstra management.

      2. Anonymous Coward
        Anonymous Coward

        Which brings to mind the 'real' economics of the debate, and also a point (Fukuyama's actually.. go and read him) about the need for good governance to build up the radius of trust through transparency.

        National Wealth, if it is a component of individual well-being, is not measured by how hard we have to work just to stand still - fixing this, replacing that, monitoring everything all of the time and re-educating ourselves all the time. That might be great for GDP and profit generation but does not build up well-being for the population as a whole.

        For that you need to be watching how much really durable infrastructure you are putting in. Where by durable we mean both physical longevity and having the technological flexibility of applicationto keep it relevant and which has the lowest initial costs combined with long-term (50, 100, 200 years?) Total Costs of Holding. Get that right and we can go on year by year building real enduring value into each of our lives as we usefully utilise our efforts and skills to deliver more low cost services into our lives.

        It may not be good for GDP-leeches and, as the ancient song goes, it may not be good for the "dedicated followers of fashion" amongst us, but it will be good for our common wealth and individual well-being. That is true redistribution of income... the alternative just keeps your incomes depressed and the price of your gluttonous consumption continually inflating while destroying all stability in your daily life and your assurances of a better future.

        But what do you expect.... politicians and commerce would prefer that you only measure your well-being in terms of income and jobs so you can beaver away for them and feed their wealth rather than your own.

        In the end you get who and what you vote for.

    5. rtb61

      Don't forget all the copper is brand spanking new as are the conduits and all the connections. Wait no it's not, whole bunch of it is approaching half a century old, the copper is brittle, the insulation is failing and the conduits leak like sieves.

      Now I bet they did not test it on degraded copper network in the rain. I am 2000 metres from the node in an inner suburb, so more bullshit.

  2. Trigonoceps occipitalis

    ALUMINIUM

    How is it for Al "last mile"?

  3. Knoydart
    Stop

    So you think that Oz should pay the same as deploying a FTTP network and fix up the copper that's duff so that G.Fast and FTTN will work for a bit longer.

    Its already been said on this wonderful site that FTTP is the end game, why spend $53B or there abouts just to keep copper running?

    1. Simon Sharwood, Reg APAC Editor (Written by Reg staff)

      The point I am trying to make is that the likes of XG.fast suggest the life of copper is going to be rather longer than previously imagined. When we planned FTTP we did so because it was felt the useful life for copper wasn't long. Turns out it can probably do the job for quite some time yet.

      1. Knoydart

        I'm sorry but have to politely disagree with you Simon. If you are having to replace the copper in the ground to get these technologies to work, then that's a fail. FTTN has be sold as the cheaper option for nbn deployment but turns out its not the case by the nbn's own recent numbers.

        Once you get the fibre in the in the ground, you just replace the terminal equipment as technology improves and customers want faster speeds.

        Australia will be dealing with this decision for the next 20 plus years and a certain incumbent telco will be laughing all the way to the bank.

        1. mathew42
          FAIL

          Consumers don't want faster speeds

          The reality is that customers simply aren't prepared to pay for faster speeds. Every iteration of the NBNCo Corporate Plan (both Labor & LIberal) has predicted the most popular speed will be 12Mbps. The actual take-up figures published by NBNCo in July proved that consumers aren't prepared to pay with 35% on fibre connecting at 12Mbps and 42% connecting at 25Mbps.

          Labor's grand stupidity in introducing speed tiers on fibre has effectively crippled the network and meant that there is little difference for most people between the various technologies.

          1. Jasonk

            Re: Consumers don't want faster speeds

            And how does the MTM go becuase you need speed teirs because at certian distance it can only deliver certian speeds. But it's ok apparently people don't need faster speeds. Yet the MTM is now only $8B cheaper than FTTP and only finished 1 year sonner great saving right there. The we add in the less revenue because it can't deliver speed with demand makes a complete mockery.

      2. dan1980

        I also have to disagree.

        Repeating what I said in my (long) post below, you are not making the correct distinction, which is between copper as a medium and copper as something that is actually currently installed, as it is in Australia.

        The only way a FTTN deployment makes any kind of sense is by showing that you can save time and money by utilising as much of the existing infrastructure as possible. For bit of copper you have to replace, that's fibre that you could have installed.

        Working in IT, there are plenty of clients that I have who use ADSL2 connections that are around a 1km from the exchange but who are getting ADSL1 speeds, along with not-infrequent dropped packets and outright disconnections due to horrid attenuation and noise on the line.

        As you say, we need information about the state of copper and the cost-effectiveness of using copper is entirely dependent on not having to replace too much of it.

        Copper, as a medium, may well have a future, but that future is certainly not brighter than fibre so you are constantly trying to increase the abilities of copper with new technologies to overcome the deficiencies. This is great and interesting and very useful, but if you were starting from scratch, you would be mad to choose to install copper for a national network.

        Now, we certainly aren't starting from scratch but the more of that old copper we replace with new copper, the closer it comes to equalising and the longer the network stays in place, the more expensive the copper becomes as it requires more frequent replacement (if it is to keep up with or approach fibre) and the nodes themselves have not-insignificant running costs when they are deployed in the kind of numbers that are required to keep the distances short enough to see appreciable benefits from these new technologies.

        1. dan1980

          Continuing, you say:

          When we planned FTTP we did so because it was felt the useful life for copper wasn't long. Turns out it can probably do the job for quite some time yet.

          Again, what you really mean here is the useful life of copper as a medium which is only one part of the equation. Connecting increasingly able devices to lengths of copper does not do anything to extend the lifetime of that copper - it will need replacing at some point, if it doesn't already (and much does). And pure speed and network infrastructure life are not the only benefits of a FTTP NBN - we would get to unshackle from the existing providers, which is nothing to be scoffed at.

          As a final note, even if all the copper currently installed was in A1 condition, that doesn't come free because we still have to rent it from the providers and that is money that, when combined with running costs and upgrades will eventually exceed what a FTTP network would have cost.

          For once we had a real long-term plan (however poorly implemented) where, instead of selling off infrastructure for a quick dollar only to rent it back, we were going to actually invest in new, future-proof infrastructure that would work out cheaper (and better) in the long run - yes, even with the inefficiencies and contractor issues and overruns and blowouts.

          I have rambled again . . .

      3. Chris 155

        No, they don't.

        They yet again show that you can get incredible speed out of pristine copper over incredibly short distance with no interference. This isn't new or shiny or breathtaking. They've been getting incredibly high speeds out of uselessly short cable distances for years.

        No one has ever shown anything like that kind of speed on real world copper, even newly installed real world copper, let alone the crap that's in the ground.

        I've given up on having a land line because the copper in my neighborhood is so bad that you couldn't actually hear people. How am I going to go with FTTN next year? You reckon I'll get XG.Fast speeds? Do you reckon that the government is actually going to go out to every single node and upgrade it to G.Fast, let alone XG.Fast? Under FTTP you replace the units in the exchange and your home router and you're done.

  4. P. Lee

    Is el reg reduced to trolling for business?

    I know quite a few DC designers and telco's who prefer some well-tested fibre tech for WAN connections to some experimental copper LAN tech for, er, WAN links. Many of them like the easy x10 bandwidth fibre gives right now. Quite a few of them like the x40 bandwidth fibre gives right now, compared to production copper links. I think they'd still take a x20+ sized link over this new copper tech.

    Fetishists, obviously.

    What's with el reg? You could have reported this as some cool new LAN tech and I'd say "thanks for the news." Reporting it with name-calling and suggesting its useful for something it clearly isn't makes you look like prats.

    1. Simon Sharwood, Reg APAC Editor (Written by Reg staff)

      Re: Is el reg reduced to trolling for business?

      Prats? Hmmm ... not sure I want to wear that.

      I'm not suggesting that XG.fast is a WAN substitute. I'm suggesting that if even the most demanding, data-intensive, applications often don't need gigabit, homes and the average business need it because ... why exactly?

      By the time XG.Fast comes along, it will be more than mere theory. VDSL's already doing 100Mbps. G.fast's doing better.

      It all reminds me of when Nicholas Negroponte wrote off wireless as a carriage medium in Wired in about 1992. And then along came WiFi, 3G ... the rest his is history.

      There's so much copper, in so many places around the world that there's massive incentive to figure out how to get the most out of it.

      1. Diamandi Lucas

        Re: Is el reg reduced to trolling for business?

        "There's so much copper, in so many places around the world that there's massive incentive to figure out how to get the most out of it."

        How to get the most out of copper? Rip it out of the ground and sell it for scrap.

        The copper in the ground is in a terrible state in most places, it's been in the ground for decades and shows all the signs of it.

        "It all reminds me of when Nicholas Negroponte wrote off wireless as a carriage medium in Wired in about 1992. And then along came WiFi, 3G"

        The difference being that wireless doesn't rely on physical cables that can degrade due to time and environmental factors.

      2. dan1980
        Happy

        Re: Is el reg reduced to trolling for business?

        @Simon

        "Prats? Hmmm ... not sure I want to wear that."

        If you caper around the place wearing lycra then wearing the moniker of 'prat' is hardly unprecedented : )

        (For anyone from another country reading, the joke/good-nature is implied by our national genius for insulting each other as a form of affection*.)

        * - Not that the thought of Simon dodging buses down Parramatta road clad in skin tight apparel is the source of the 'affection' . . .

      3. dan1980

        Re: Is el reg reduced to trolling for business?

        @Simon

        "It all reminds me of when Nicholas Negroponte wrote off wireless as a carriage medium in Wired in about 1992. And then along came WiFi, 3G ... the rest his is history."

        An interesting point but there is one crucial difference: when you're dealing with radiowave-based wireless, there is no choice of medium: it's the air. There is no danger of choosing the wrong medium or upgrade cost to replace it. You just replace the equipment at either end. Further, with the exception of spectrum issues, you can run easily migrate over time from one technology to another - say from GPRS to 3G, and on to 4G and whatever comes next (5G . . .)

        This is manifestly not the case with wired connections because the choice of technology is bound up with the medium you have installed. You also have actual physical ports at exchanges/nodes that are connected, physically, to one type of device or another and to transition from one technology to another requires not only installing the new device but physically moving a section of of the physical medium from one port to another.

        But, tacking the spirit of the reference more generally, there is something you are missing. Yes, wireless is not very prevalent and some of the technologies can achieve respectable bandwidth. But it is not a direct competitor to wired connectivity because for any wireless technology, there is a faster wired technology.

        And this is important because the required bandwidth keeps increasing - we not only have HD streaming but ULTRA HD streaming. Sure that's overkill for many but the point is that data requirements generally don't go down.

        Now, while 4G may be able to handle a HD stream from Netflix, provided the coverage is decent, it is a shared medium so the more people using it for data-intensive applications, the slower it is for everyone. I recall that when iPhones were released in Australia, several of the networks were simply unable to sustain 3G connectivity in certain areas. Optus in North sydney for example (despite it being their headquarters at the time!)

        Again, that's focusing too specifically on wireless rather than the spirit of the quote but my point is that the nature of wireless technology as a whole - rather than any specific wireless protocol - has certain downsides.

        So, if you choose air as your medium over copper, you are bound by the inherent limitations of that medium, the main one being that it is a shared medium. With copper over fibre, you have bound yourself to a different set of limitations, being increased attenuation - which limits the lengths - and a more involved and frequent maintenance and replacement schedule.

        By the time copper catches up to where fibre is now, fibre will be further along.

      4. SImon Hobson Bronze badge

        Re: Is el reg reduced to trolling for business?

        > VDSL's already doing 100Mbps

        Err, you mean it can do "up to" 100M in ideal conditions. Meanwhile, back in the real world, not many people get that.

        Over here in Blighty, we already have reports that VSDL2 speeds drop when more than one subscriber is connected - due to crosstalk between pairs which were never designed as radio frequency conduits. I assume this newer tech is simply more of the same - go up the frequency band for more speed and accept the faster drop off with distance. So unless there is no sharing of cables, there's going to be massively more crosstalk - so you're buggered if you and a neighbour both share a cable from the joint box on the nearest pole to the node a couple of poles up the street.

        Not to mention, round out way, there will be a lot of houses that effectively would need their own node if 100m is the limit on distance to the node - and I imagine Australia with it's much larger space is similar. By the time you get to that sort of requirement, it's hard to see where the benefit is of stopping the fibre less than 100m from the house.

        At ${dayjob} we still have a lot of customers who cannot get decent connectivity at all. Some of them can't even get ADSL that works. Due to terrain, wireless systems aren't that easy to do; and due to tax implications*, some of the wireless systems that were around have gone anyway.

        * AIUI, our government in one of those "what can we screw up today" moves declared that such operators had to pay tax on all their infrastructure based on the potential for earnings - not any actual earnings. So each wireless repeater tower was suddenly going to get a big tax charge based on what it could earn if it's full potential capacity were sold !

        The same thing was supposed to apply to operators with dark fibre lying around.

        Needless to say, this wasn't applied to the incumbent operator, BT - which further skewed competition in BTs favour.

  5. Oengus
    Flame

    Copper isn't the answer

    At most there will be around 16 houses within a 100m run from the nodes. I also doubt that the existing copper that is run from the nodes will be in the same condition and quality as the copper in the lab tests.

    On a good day my existing connection gets 8Mbps. When the weather changes I drop to around 4Mbps and I am told that I am on a good line within reasonable distance of the exchange. The existing copper is degrading rapidly and there are quite a few places (including the second phone line into my house) where Telstra (in their infinite wisdom and to save a couple of dollars per household) rolled out "Pair-gain" technology which can't support ADSL2+ and I am sure will not support XG.Fast (or VDSL or G.Fast).

    If they have to replace the copper to give me XG.Fast capability wouldn't they be better off giving me fibre and saving the expense of future upgrades.

    1. Diogenes

      Re: Copper isn't the answer

      Your existing copper will be taken off the "pair gain" and put on the node, which is connected via optical, so what does "pair gain" not supporting adsl2 have to do with the price of fish in China ?

  6. Tannin

    What a load of rubbish,

    To get that performance, you'd need to lay a brand new copper line and get all the details exactly right - i.e, spend more than it takes simply to lay a fibre line. FAIL.

  7. Somone Unimportant

    I would be interested to see the speeds that these boffins get with their technology on our copper connection. 400 metres to the silver bullet, 3 kms to the exchange, and everytime it rains heavily our phone drops out because the copper is 50 years old and the shielding has rotted.

    Oh - and I am 7 kms from a CBD...

    While I respect pushing back the limits of what is possible, touting this as workable is as bad as telcos that tell me I should get up to 22 mbits per second because my line is ADSL 2+ capable.

  8. Snow Wombat
    Mushroom

    No..that's stupid.

    Screw off!

    We need to retool the country with fiber because hardly anyone lives within 100M of an exchange, and besides the Copper Oxide in the ground is so old, spliced, chopped up and bloody awful that you'll never get anything near that anyway.

    I have just managed to move into a place with NBN and it's utter fantastic. Stable, fast and works as it should.

    My previous place had a DSL link that was powered by kangaroos, and when the weather got hot and humid, it would go up and down like a yoyo because the wheezing old exchange it was running out of had a heat issue.

    I see reports like this all the damn time and they always herald that copper will live a little longer but they never actually deliver over any sort of useful distance. Facts are, that they are running up against the harsh, unforgiving wall of physics, and what copper is capable of.

    1.6Gbps is still pathetic when fiber can do 10Gbps with GPON-2(10G-PON), over a lot longer distances with no attenuation problems.

    1. mathew42
      FAIL

      Re: No..that's stupid.

      > 1.6Gbps is still pathetic when fiber can do 10Gbps with GPON-2(10G-PON), over a lot longer distances with no attenuation problems.

      Hypothetical numbers that have no relationship with reality when 35% connected to fibre selected 12Mbps and a further 42% selected 25Mbps. Labor forecast in the Corporate Plans (before losing office) that less than 1% would have 1Gbps in 2026.

      Sadly what Labor's spin is so far from the reality of the NBNCo Corporate Plans that the IT crowd in Australia should hang their heads in shame for supporting them and must now reap what they have sown or have enough cash for FoD.

      1. dan1980

        Re: No..that's stupid.

        @matthew24

        Actually, the potential of a network is indeed relevant, though not in any way the whole story.

        Many people who criticise the FTTP plan cite the figures you just have, expecting that these numbers are the only reason to implement such a network.

        Not so.

        Yes, fast speeds are important, but more important is that fast speeds are available everywhere. The reason being that we currently have a rich cousin/poor cousin situation where come areas are well served and others are poorly served. This matters because internet communications are a vital part of modern infrastructure, as important as postal services were in ye olde days. By that I mean that the ease and speed of information transfer is directly relevant to businesses and, with our population centres growing while regional areas stagnate, it is vital to start boosting growth in the less traditional locations.

        This was part of the point of a homogeneous network - to enable businesses and people to access adequate services no matter where they were, thus opening up pathways for investment in more areas. AS I have said more than once, I, personally, can vouch for clients who have either shelved expansion plans or changed branch office locations based on connectivity. That might sound far-fetched but when one location requires $15k for fibre install and ~8k/mo for 10/10mb while another site is able to get EFM at 20/20 for a fraction of that, well, that can and does influence decisions. I have a client that closed a branch office (2 pax) because connectivity was so poor that the database didn't work properly, resulting in a succession of people quitting because the system hampered their ability to make sales, which in turn reduced their earning potential.

        It's really not just a hypothetical - inadequate comms genuinely hurts 'the economy'.

        So, really, the question is not whether the average person will use X Mbps or Y Mbps but whether a user or business needing Z Mbps will be able to get that regardless of where they set up. If the answer is yes then that promotes growth and allows cheaper areas to attract investment. What good, after all, is saving $5,000/mo when the Internet connection costs $8,000/mo more than if you set up closer to the city?

        1. mathew42

          Re: No..that's stupid.

          > It's really not just a hypothetical - inadequate comms genuinely hurts 'the economy'.

          Totally agree and speed tiers is the perfect example of this.

          > So, really, the question is not whether the average person will use X Mbps or Y Mbps but whether a user or business needing Z Mbps will be able to get that regardless of where they set up.

          FoD suggests that the answer to that question is yes. Under Labor's FTTP plan unfortunately many businesses in smaller communities were going to be forced onto wireless rather than being given FTTN options.

          > What good, after all, is saving $5,000/mo when the Internet connection costs $8,000/mo more than if you set up closer to the city?

          I haven't seen any indication that once fibre is installed that additional monthly fees will be charged. The reality is that if you can afford the mythical 1Gbps speeds then you can probably afford the $5000 install cost. I say mythical because although NBNCo made 1Gbps available to RSPs in December 2013 we are yet to see an RSP offer the plan at a retail level.

  9. bep

    Well there's us and there's them over there

    We have fibre to the premises but they only have copper. Guess whose house is going to be worth more as a consequence in five years time? A nationally-owned network should be delivering a near-as-dammit the same network to everybody unless the distances are truly horrendous. As it is the NBN has now been reported running new fibre-optic cable in areas already served by fibre-optic networks open to other ISPs but at the same time it's using fibre to the node and copper elsewhere. Thanks but since I'm paying for all of it I'd prefer the original plan which was a good plan and it would seem no more expensive than the catastrophic mish-mash that Malcolm created.

    1. mathew42
      FAIL

      Re: Well there's us and there's them over there

      > Guess whose house is going to be worth more as a consequence in five years time?

      My guess is that more important factors like school zones, closeness to public transport and other factors will be more significant. A fibre on demand install should be less than $5000 which means it is cheaper than many of the renovations you might perform in preparing a house for sale.

      > Thanks but since I'm paying for all of it I'd prefer the original plan which was a good plan

      But you aren't paying for it. NBNCo is being run as a private company which is expected to deliver a return on investment. Don't you remember Labor's claims that it wouldn't cost taxpayers a cent? The reality is that most consumers (77%) aren't prepared to pay for speeds faster than what you find on FTTN, 4G, HFC, etc. This is simply down to Labor's stupid choice to impose speed tiers.

      1. Jasonk

        Re: Well there's us and there's them over there

        Sky mesh are offering a 100/100mpbs on fiber staring at $99 month. But the connection is really a 250/100 connection.

        Again you talk about FOD but why should some people have to pay to get better speeds FTTN which cost more than the cost of FTTP while other people on FTTP and HFC get that same connection for free.

  10. batfastad

    Sweat the assets!

    So how long will it take to get these sub-fibre speeds outside of a lab and down the typical copper line length of 0.5-3 miles? Yeah.

    Keep sweating that copper asset boys!

    Fibre. All over the place. 20 years ago.

  11. marky_boi

    Never been called a fetishist before

    I work in the industry in Au. The chance to update the copper to fibre was one that should have been taken. It resists water, corrosion, being submerged and all other manner of environmental stresses that Australian weather can throw at it. It was a value proposition, put all Aussies on an equal footing speedwise allow good uplink speeds [good for working from home, my pet reason] and generally falls into the do it once and do it right mentality. you cant sustain good speeds on crap copper PERIOD. the best thing about fibre is that to get a speed lift many years down the track, just swap out the ends, fibre stays in place,,,, Telcos been doing this for years. So in my considered opinion as an insider I call BULs%#T on this article. Why are countries like even lil ol NZ doing the right thing and our politicians ignoring it on purely political grounds.>. not my idea cant be good...........

  12. Perc

    Ongoing Costs = Large elephant in little room on footpath.

    I once read in an article about Turnbull's proposal for FTTN that it was estimated that the cost of providing power, servicing and ongoing maintenance would be $11B pa. In contrast the same ongoing cost for FTTP would be $2B.

    The burden for ongoing costs in any system will fall directly upon the consumer. If this figure is accurate, the cost of Turnbull's NBN will double in 5 years of trouble free running. Another way to look at it this is that a FTTN network could have been bought and paid for by its customers in five years if true.

    I don't know if the figures quoted are realistic or not. If only some tech-oriented media would pick up the cudgels.

    1. Tim99 Silver badge

      Re: Ongoing Costs = Large elephant in little room on footpath.

      <Smug Git> I have fibre NBN in my retirement village. </Smug Git >

      It is a hybrid system with fibre to the comms room in the main premises and VDSL2 to our 120+ units. As a result, the maximum speed to each household is "only" 100Mb down and 50Mb up. The longest cable distance from a unit to the comms room is about 150 m (well within the 500m sec for >100Mb). The main reason for using a hybrid system was so that we could have copper for the telephone emergency call service to each unit as, at the time, there was no way of offering this service through fibre. There has been one system failure in 2 years - The copper wiring to several units in a service pit was found to be under 20cm of water. The cowboy who had installed it, had twisted a number of wires together and "sealed" them with siloxane.

      I suspect that the LNPs real reasons for a hybrid system were:-

      1. It looked as though it would be popular with the electorate, so the LNP needed to say that it was too expensive, and that it was a typical example of waste by the Labor party.

      2. It was socialism, and we can't have taxpayers owning something and getting the benefits of having paid for it. This stops our corporate sponsors from being able to gouge you for ever.

      3. Rupert Murdoch, who still seems to be able to control much of the media here through News Corp, needed a delayed and a marginalized service, so that his Foxtel satellite monopoly could continue to be highly profitable. Who would invest in satellite when you can have media streamed through the NBN? The price of the Foxtel services plummeted when Stan and Netflix became more generally available.

      4. Many people who unwisely purchase Telstra shares are LNP supporters. Labor's agenda seemed to be that they would use the NBN to break up Telstra and effectively renationalize it - Telstra also are also joint owners of Foxtel (3).

      What me cynical - No, I have just lived a long time.

  13. Anonymous Coward
    Anonymous Coward

    Now flood the comms pits with water…

    … and see what happens.

    1. Simon Sharwood, Reg APAC Editor (Written by Reg staff)

      Re: Now flood the comms pits with water…

      The pits are indeed the pits. Mine were repaired twice last year. But are they the pits everywhere? We don't know. Hence my call for proper data on the state of copper

      1. Anonymous Coward
        Anonymous Coward

        Re: Now flood the comms pits with water…

        Well, some places there are comms pits with conduits running between, in other places, the phone line is buried bare, no conduit.

        My point being that, while it's great they've been able to squeeze high speeds out of copper pairs over 100m, possibly in controlled conditions, we don't know how well it will work in the real world.

        What happens when you run two of them in parallel, with the electrostatic and electromagnetic coupling that comes with such arrangements?

        Fibre isn't a guarantee either. If you've got reactive soil, then fibre is going to be a pain. Telstra tried putting a fibre across the Nullarbor plain years ago. They had to abandon it because soil movement kept breaking it.

        We seem to have less of a problem with this though, and there isn't the cross-talk issues that plague copper. It also doesn't have the line-of-sight or capacity issues of wireless, and for the most part, upgrading speed requires replacements of equipment at each end of the cable, not the cable in between.

        It's useful to have an extra tool in our arsenal, but I think it premature to declare one technology not-required so soon.

        1. A Dawson

          Re: Now flood the comms pits with water…

          Errrr Nullarbor has had fibre across it for a looooong time (Telstra and others)

  14. moonrakin

    Sweating the assets and resisting change instead of embracing enhancements to the system?

    Phone companies?

    Whodah thought eh?

    I remember back in the 80's a certain BT boss at an exhibition seeing some early web stuff and getting incensed that several web pages could be opened at the same time at no extra cost to the punter....

    Old habits die hard.

    1. Anonymous Coward
      Anonymous Coward

      Perhaps early 90's rather than 80's.

      https://en.wikipedia.org/wiki/World_Wide_Web#History

  15. Trixr

    What a ridiculous headline. No-one's cared that much about raw speed (not to mention numbers in a lab which will be lucky to be achieved within a factor of 100 in the real world). It's maintainability and future-proofing that are the advantages of fibre.

  16. shocking

    Copper degrades over time

    When I moved into my current residence, I could get 8032 kb/s. Now it's down to 3612. I was supposed to get the NBN in 2014, but we had a change of Government. My suburb's not on the list of the ones that are supposed to get the rollout by 2018.

  17. Anonymous Coward
    Anonymous Coward

    Rolling new copper = FAIL

    Who rolls new copper? No one. The only countries that keep pouring money down copper pits are the ones that already have lots of copper. No one else would even consider it. You put down fiber for backhaul and broadband and radio masts for cell and limited wireless "broadband." Not copper.

    Down where I live, nbn is offering fixed wireless "broadband." Please. It's radio waves and, like any radio-based tecnology faces problems from topography, vegetation, weather, and contention. You can only think that the "broadband" networks in Australia are fast if you've never been to countries with real broadband.

    Fiber. It's not a theoretical technology, it's a real-world technology that already works brilliantly. The rest of this stuff? Please. I remember when DSL/ADSL was still in the labs and word was leaking out. It was a *decade* before it was deployed in the field in any way that you could call reliable.

    A fundamental problem in Australia remains that the Howard government completely screweed up "privatizing" Telstra:

    1) They kept a huge tranche of shares and thus gave the (any) government a huge conflict of interest when it came to making sensible regulations.

    2) Wholesale/backhaul and commercial/consumer retail in one business? Who thought that would lead to anything but an entrenched monopoly that had zero motivation to innovate?

    It's not as though the Aussie government didn't have plenty of examples of how to sell off and regulate a formal soverign PTT service to consider. Heck, the Americans split up AT&T over 30 years ago.

    And, while dragging up old history, here was the Howard/Coonan plan for national broadband:

    That's a direct quote ;-) They had no plan. Helen Coonan was a very bright and able minister, but knew nothing about telecom. Howard was a clever politician, but knew nothing about tech. They had NO plan. So Labor's (somewhat odd) plan was streets ahead of anything else. The new plan? Meh. So far, not impressive.

  18. Bubba Von Braun

    Plain wrong

    Simon I have read many of your articles with interest this one I am afraid I have to throw the Bravo Sierra flag on this one.

    Yes I have fibre to the premise on the NBN network, prior to that I had crap copper so old the pits still had the PMG imprint on them. The copper inside was total rubbish and we all know that the copper network has not been maintained for the last 20+ years. (ADSL2 that was so slow it was ADSL speeds!!)

    After nearly 40 years in the IT business in AU I have seen it hamstrung by these sort of short sighted decisions, we all know lots of technologies that worked fine in the lab, but never make it out into the real world situations eg: v.fast wipes out ADSL when run in the same cable bundle!, using your logic we should all still be on acoustic couplers or heck lets all do wireless not like the airwaves are congested that was Malcom's last alternative...

    Sadly the decisions are driven by "I cant be seen to be doing what the other guys suggested, even if it is right" Reality is FTTN will benefit from the work done already to build the back haul network, the one thing that should stop is the deployment based on political considerations.

    I have FTTP because most of my locals voted Green for the last few elections, so thanks for that, but this was a chance for Australia to do something big and again we balked and kicked the can down the road in the "hope" it will be better and cheaper..

    Last I checked labor costs here have only gone in the other direction.

    Proudly a fibre fetishist as I was an Email, Internet, etc.. in the face of those arguing for expensive maintenance of the status quo. ;-)

  19. JJKing
    Flame

    Never seen an article's author defend their words as much as. Me thinks the lad doth protest too much.

    I was rolling out GB connections in businesses 10 years ago with, w a i t f o r i t.......100 metre copper. AND, it went 1GB in EACH direction. Will this fancy new "technology" go that fast in each direction over copper and as mentioned above, aluminium?

    In 1998 I was told about fibre being connected to an overseas business LAN. Guess their vision was better than the MFC retard brigade in this backwater country.

    Keep hammering away Mr Sharwood and you actually might convince yourself about what you preach.

  20. Diogenes

    FTTN switched on at GORO yesterday

    As it is one of the first let see what ACTUAL usage shows us.

    Many of my students are thrilled to actually be able to get broadband - some have had orders waiting for over a year waiting to be connected when a dslam port opened up. In the interest of fairness I must say that GORO was not the initial 5 year rollout plan for FTTP, so anything in less than 5 years is a welcome relief..

    Riddle me this - If the costs/times have blown out for the fftn model, where there was a publically available costing model - by how much more would they have blown out in the fttp model, given the original business case costings were literally done on a coaster on RAAF 1 ?

  21. Tannin

    Counterpoint

    I posted earlier to say that the article is nonsense - made much worse by the daft headline, but nonsense nevertheless. I have always regarded Simon as a very smart cookie and long since got into the habit of simply taking articles under his byline as holy writ. Sadly, I'll have to read more critically now - trust once broken is hard to repair.

    Having said all that, there are certainly some criticisms to be made of the original NBN rollout. In the details of its hurried implementation it was wasteful, chaotic and badly managed. No question of that. Try talking to anyone in the building trade: they regard NBNco with universal contempt: the organisation is chaotic, disorganised, arrogant, and you can't ever get hold of anyone who actually knows what's going on.

    Going beyond on-ground implementation to mid-level planning, it was just as bad. Take my case: my street has had an excellent HFC cable system in place for more than a decade. Sure it is technically inferior to genuine FTTP, but it works very well in practice, the infrastructure is still quite new and in good order, it has years of useful service life left in it, and it is plenty fast enough for almost all purposes. (My new-last-month NBN fibre connection is significantly slower, about half the speed on average, though that is more a matter of throttling to the lowish speed I'm prepared to pay for off-peak, and the backhaul network's inability to cope with the huge daily surge in Netfix movie traffic between 5 and 11. The HFC connection (through the same company) slowed too, but not as much as the NBN one.)

    What person with half a brain would prioritise NBN fibre service to a locality with a good quality existing HFC service over delivery of the new service to other parts of the same city with no HFC and a lousy, unreliable ADSL system running on ancient hardware? Completely daft! (No skin off my nose, of course, I'll never need to replace the new fibre and I can have it lots faster anytime I feel like paying the (much higher!) price for a better contract.Well, I will be able to, as soon as they fix the Netfix backhaul overload problem, which we assume they will sooner or later.)

    Nevertheless, the all-fibre NBN plan was, despite its many faults of on-ground implementation (e.g., doing least-needy suburbs first instead of last), unquestionably the right plan overall, and a vastly better plan than the even more wasteful, chaotic, and badly managed joke that Turnbull replaced it with. No fact in this article contradicts that. Lift your game, Simon.

  22. Neoc

    So... we have copper on the one hand: it is already laid, the state of it in most of Australia is crap and in need of serious repairs (I live less than 10km from Brisbane CBD and heaven help if it rains), and scientists are running out of ways of pumping more bits through it. (don't get me wrong - it's had a good run for its life-span).

    On the other hand we have fibre: It has to be laid but not fixed for a while, and as a new technology, scientists are still finding ways of pushing more bits per second through it.

    I feel like I'm back in time when Napoleon tore down a whole heap of buildings in Paris to make for wider roads, and nay-sayers poo-pooing the idea because the roads they had were good enough for the times. Yeah, they were... but it took a century or so for the new roads to reach capacity, allowing Paris to flourish.

    Copper advocates are giving us the same refrain - copper's good enough for what we need, we don't need this newfangled fibre.

    Except we will, soon enough.

  23. dan1980

    Simon . . .

    Mate.

    You are confusing two things: copper as a medium and copper as it actually is. So the question is not whether copper as a medium has a future but whether the copper that is currently running through the streets and bunched up in flooded pits and hung over trees and tangled together in masses of inadequately protected, tangled splices bundled together with duct tape is suitable for the future.

    The argument for FTTN being cheaper and quicker is not predicated on the medium and equipment being cheaper or easier to install; it's based on the assumption that much of the existing infrastructure will be REUSED.

    Unfortunately, as many know, there is a lot of very dodgy copper out there. And even if all that was replaced with nice new copper, the problem ill just recur.

    Even, however, assuming that these newer technologies really will provide - and continue to provide - the increasing bandwidth that is required even now and will definitely be required in the future, that's not the whole story because you also have the overhead of the N - the nodes.

    Part of the speed promised by these technologies, as with VDSL before them, is based on a relatively short distance because the faster the speed, the sharper the drop-off as you move further away. Most tests seem to agree that once you get past 1km from the exchange, there is little difference between VDSL2 and ADSL2. Some suggest that it doesn't equalize until about 2-3km but others show it happening much sooner. Seeing as the state of the copper has an effect, I think it would be prudent to take a lower measure.

    So, the nodes have to be built close to premises for there to be a real, meaningful benefit. and that is indeed the plan, but it does have a downside: a lot of nodes, each requiring active equipment that must be powered, cooled and maintained - which includes testing and replacing batteries.

    FTTP, on the other hand, does not require active equipment in the path and thus the distribution nodes can be smaller and less numerous as well as cheaper to install, cheaper to run (nothing) and cheaper to maintain.

    I linked some comparative images in some previous posts and I think I over did it that time so here are just a select few.

    FTTN nodes

    FTTP 'nodes'

    If you look at the second fibre distribution cabinet, they have started with one 1x12 splitter. To expand capacity, you drop in additional splitters into the slots in the bottom left and then cable up to a patch-point in the main section. These assemblies can be cabled up - or purchased ready cabled - before a tech goes out to install it. And, while that box is currently using a 1x12, you can get up to 1x32 units in a similar size, giving you 192 connections from 6 incoming lines in a very compact unit - one that's still larger than it really needs to be, as evidenced by the last image.

    All this complexity and need for power and cooling and maintenance costs money and so, the longer the network is in place, the cheaper a FTTP network become. I saw some conservative estimates (i.e. using the Coalition figures) that indicated parity of total cost by 2027 - 12 years. Every day after that, FTTP is cheaper.

    So to have an article that implies - if not outright states - that the case for fibre is now, or will soon be, moot is not necessarily giving the whole story. That's because it's not just about the raw bandwidth available over a given medium in a lab.

    The NBN project gave us the opportunity to replace the existing, aging copper network with a new, future-proof network and, in the process, uncouple us from Telstra and the problems that this has caused.

    It was ambitious and poorly planned when it came to the details and poorly costed and definitely poorly handled. BUT, the goal was not just to increase Mbps and so any comparison that focuses solely on this is missing the point. When that comes from the government, it is a deliberate attempt to mislead. From Simon I hope it's just him trying to salvage some good news by suggesting that at least that one aim - of increasing available bandwidth significantly - might not be completely lost.

    Even then, much still depends on the state of the copper and how this fits in to the mixed networks being proposed. And, further, those in poorly-serviced areas are likely to continually be poorly serviced.

  24. Winkypop Silver badge
    Meh

    "copper has a decent future"

    Much like the coal industry I guess.

    Just because we have it doesn't mean we should keep using it.

    I hear the gas lamp lighters union is reforming as well...

  25. Brian Scott

    Fibre? Copper?

    Wow. I'd very happy to have either copper or fibre.

    My NBN future (guessing at least the next 10 years) will be wireless delivery. I'm really looking forward to that like a good toothache! Of course at the moment I'm stuck on ADSL 1 unless I switch over to BigPong so maybe I shouldn't complain too much. Friends who have ADSL2 in the region tell me that they are going to be moved off that to wireless in the long term.

    A contact doing nbn installs suggests that they are really not very interested in anything other that wireless because it avoids playing in pits.

    I'm not sure where they would be bothering to install this stuff. It might just be Malcom Turmbull's place.

  26. Anonymous Coward
    Anonymous Coward

    The 1990s rang (Via pair-gain)

    Wanted to know why we are still using copper.....

    PS: Are the Liberals in again?

  27. adfh

    "Almost" "If"

    "almost time"

    "have a chance to kill off"

    "if it can be proven"

    "aren't in public view"

    So basically, copper is being pushed faster in lab conditions, but the higher the speed, the greater the distance drop off and our population density, well, it varies wildly.

    I do wonder, in the cost calculations, what accomodations are made for:

    * Cost of repairing and maintaining copper

    * Cost of maintaining and repairing nodes

    * Cost of powering/cooling nodes

    * Resilience of nodes to adverse/emergency weather conditions (heat waves, flooding, fire)

    * Space required for nodes

    * Ability for nodes to maintain service during grid failure

    The impression I get is that nodes will pretty much, depending on population density look like RIMs on steroids, requiring active power + cooling and potentially copping it in a flood.

    I have no doubt that installing fibre, where there was/is copper, is more expensive up front - but as population densities scale up, and bandwidth demands alongside that, it'd be interesting to see the projections over a longer term regarding cost.

    That is - even if copper is cheaper upfront, what lifespan is it expected to have as compared to fibre, and how would its costs compare over that lifespan? Surely there has to be some foreign market information on this on populations of varying size and density?

  28. Anonymous Coward
    Anonymous Coward

    Fibre does not get stolen and sold

    nuff said

  29. DanielR

    Until it rains and you have to wait weeks for a repair like you do currently. Businesses have had to wait months for line repairs. "Up to" is also a scam. You either get it 24/7 or you don't.

    I will be paying the extortion costs of $3000 to get fibre connected.

    They will be not replacing the telephone line, you know the one that causes most people's problems to the pits. They re-patch you at the pit to a non noisy line that goes to the pillar. I don't doubt for a second there is any usable lines left.

    Stop kidding around, unless we're talking ethernet cables as copper. 100 year old cables in my circumstances you have to be kidding me !

    Also to rejog your memory

    https://delimiter.com.au/2012/04/30/fttn-a-huge-mistake-says-ex-bt-cto/

    https://delimiter.com.au/2012/06/29/330mbps-bt-extends-fibre-from-node-to-premise/

    BT is moving to FTTP. Only 1% of users can get 75mpbs . JOKE ! Disruptor and economy killer !

  30. gxxh

    Interference and ligning damage anyone?

    I really don't understand why all the discussions never mention one of the main advantages of fibre:

    It does not conduct electricity.

    Two examples:

    I had a surge protector on the phone line in front of the ADSL router. A lightning strike took out the surge protector, the ADSL modem, and two network cards on the connected computers. With FTTN that is going to be still a problem.

    I had a faulty switching power supply in a TV. The TV was still working fine, but it took out ADSL completely every time I turned on the TV. And not only for me, but for my neighbor as well. I probably impacted the whole neighborhood. I spoke to a Telstra technician that told me about a fault where a faulty DVD player power supply that was never used and just plugged in and in standby was identified as the cause of interference - but it was not in the property that reported the fault, it was the 5th neighbour over, several hundred meters down the road. All the is going to be still a problem with FFTN.

    And I am really fed up that everyone is talking about the top speeds with FFTN. Look at the reality:

    http://fttn.mynbn.info/dash.php

    You will see that the average line length is in the range of 500m. But there is still a significant number of customers that are further away then that. And if you are in the unlucky position to be further away, no new technology can help you. And increasing the number of nodes is not going to be viable.

    And there are still customers that are so far from a node that their speed will never get better then 10MB/s.

    A lot of Australians live in rural or semirural areas with block sizes of 2.5, 5 or 10 acres. In these areas the distances are going to be big.That was always the promise of fibre to the premise that everyone will get the same speed, irrespective where they happen to be physically.

    1. Anonymous Coward
      Anonymous Coward

      Re: Interference and ligning damage anyone?

      Yep, had some house-hold equipment blown up by a lightning bolt striking a tree next door.

      The carnage was the ADSL router, some Ethernet switches and my radio transceiver. Earth potential rise is nasty, and my bet was the effective line voltage on the telephone line went from ~-50V to >1kV very bloody fast.

      It'd have been harder for an optic fibre to do the same.

  31. Richard Freeman
    Joke

    Really?

    " Let's assume that in the real world XG.fast delivers a gigabit per second."

    Why would you assume that?

    I mean you might achieve that on what? a perfectly balanced twisted pair, in Laboratory conditions, with no adjacent services also trying to achieve 1Gbps - oh, wait though, that is NOT the real world, and nowhere near the current state of the Copper infrastructure in many parts of Australia.

    1. Richard Freeman

      Re: Really?

      After all at the end of the Day the Shannon-Hartley still holds true, more services, more noise, less speed. This is why even under the current planned '100Mbps VDSL' roll-out services will be vectored to deliver no more than 15Mbps until such a time as NBN take-up in that area is complete at which time they will try and achieve higher speeds on a suck it and see basis....

  32. Richard Freeman

    and then range?

    While we are assuming, lets talk about the land of the 1/4 acre block, with typical 25 meter street frontage, so the 100 meter range gives us 4 houses up the street, and 4 houses down the street from your 'node'.

    Ah lets still be generous and assume the street isn't 15 meters wide and include the houses on the other side of the street as well, now we are up to 16 premises per node.

    Back-haul from the node is still likely to fiber, so it is difficult to see significant savings from fiber being dragged to a node every 16 houses, although I guess the node itself can be fairly compact and reverse power fed, Then maybe dropped down a cable pit, or stuck up a Power pole in the blazing hot sun.

  33. Anonymous Coward
    Anonymous Coward

    Why are we doing this then...

    http://www.zdnet.com/article/nbn-replacing-copper-between-node-and-pillar/

    If NBNCo are replacing copper between node and pillar when it's not up to scratch, surely it's a similar cost to replace copper with fibre. If I remember correctly, fibre installation is often cheaper than copper installs - partly due to the lower raw material cost (copper is now becoming an increasingly precious metal, as the signalling team at PTV can tell you when people nick the stuff). Fibre is also less prone to corrosion (it's plastic, it lasts 500,000,000,000,000 years according to the green lobby), and it has the added benefit of not needing to be replaced to upgrade the speeds.

    Yes I'm sure that XG.fast will provide longevity for existing copper, but this is being tested in a laboratory, under ideal conditions, with copper that doesn't look like it's been dragged through a combine harvester backwards whilst mud wrestling it's way out of a pit. Some of the Telstra cables look like this or worse.

    Verizon have now made a policy decision - yes a good ol' Murrican capitalist company - to only replace copper with fibre from now on. This makes sense; over time their maintenance costs will in theory reduce due to the lower maintenance requirement for fibre optic cable.

    So back to my question.....

    Why, when they replace copper cable, can they not simply replace it with fibre??

    Granted it wouldn't give FTTP immediately, but over time it would slowly give the same result.

  34. kyza

    This is all very lovely and wonderful, and I think the fact that someone is squeezing this much performance out of copper amazing.

    But in the context of public policy, and especially NBN, this is ultimately just kicking the can down the road. Irrespective of whatever estimates on traffic are NOW, as has been endlessly demonstrated, demand always surges to meet capacity. Just because we can't envisage what could possibly fill a 1gb pipe today doesn't mean a content creator won't tomorrow.

    Personally I'd just like to be getting more than 10mbits/s...but then I had a conversation with someone who has just moved into a new house in Upper Kedron and found out when they moved in that they won't even be getting landlines installed for another 12 months.

    Yup, Australia, the country where new build homes can be developed without ANY kind of infrastructure requirement on the developer.

    1. Anonymous Coward
      Anonymous Coward

      Yikes! Well, we told Campbell that it was a bad idea from the outset… I guess it's the military background that does it.

      For what it's worth, there's some new estates going in on the other side of the hill at The Gap where the old quarry was that will probably have a similar problem.

      I get a ~15Mbps service where I am (on a telephone line old enough to have had an extra 3 and 0 added to the start in the time we've had it). These estates are a good kilometre from where I am, so a decent run of cable in order to hook them to the telephone exchange off Settlement Road.

  35. Colin Tree

    watch the rabbit

    The best reason for fiber is the vastly reduced maintenance cost which pays off the upgrade in a small number of years.

    The cost of the VG fast modems will probably be extraordinary. Currently the standard for G fast MIGHT be finished last this year.

    Watch the magician pull a rabbit out of his hat, ah but wait it is a slight of hand, a distraction, from the real story which is.......... fiber, and........... data.

    In years to come, after wasting enormous amounts of money on old or transitional technologies fiber WILL be put in the ground.

    The providers have to give customers the required bandwidth and data limits for new applications right now.

  36. Anonymous Coward
    Anonymous Coward

    Experimental

    Experimental XG.fast achieving 1.8 gigabits/second over 100m isn't really all that noteworthy though since Alcatel Lucent/Bell had experimental fibre technology back in 2009 that achieved 15 terabits/second over 90 km (https://www.alcatel-lucent.com/press/2009/001797)

  37. NBNnigel

    Many bad tradeoffs

    It's not surprising they'd achieve such speeds on a SINGLE strand of pristine copper in lab conditions. Hell, they'd probably achieve that without the fancy g.fast/x.fast/whatever.fast technologies since most of the bandwidth gains come from clever algorithms that neutralise cross-talk between MULTIPLE stands of copper ('vectoring'), so that bandwidth is not eaten up by error correction. While it's very clever technology, it isn't free:

    - The nodes will need exponentially increasing processing power as more subscribers are added to a node. It becomes much harder to calculate the correct 'inverted phase' frequencies for each line (to cancel out the cross-talk). So variable costs from adding more and mroe processing power increases in proportion to uptake levels, doubly so because increased computation means higher node electricity consumption (and cooling).

    - Additional costs get pushed down to the end consumer. Vectoring only works if ALL modems on the end of the copper run are coordinating with the node; otherwise one 'noisy' strand messes up the inverted phase calculations and error rates jump. This means consumers will be forced to buy modems that have sufficient processing power to implement g.fast. Meaning higher fixed and ongoing costs for consumers, even if it's partially hidden in their quarterly electricity bill.

    - Vectoring, effectively fancy noise cancellation, only works when the 'noise' is constant or predictable. It doesn't work when the noise is variable or unpredictable. So when it rains, and Telstra's wonderfully maintained copper pits have water sloshing around in them, you'll start to see drop-outs or significant reductions in bandwidth as increase error correction is required. Or when a bird lands on your copper line. Or when there's a strong gust of wind...

    - Bandwidth gains from things like time-division multiplexing come at the cost of increased latency. This is very very bad, and puts Australia at a big competitive disadvantage when it comes to distributed computing, hybrid cloud etc. Simon, ask some of your colleagues who cover 'data centre' and 'networks' why this is bad if this doesn't make sense to you (especially in the context of the cloud computing and financial industries, who are increasingly 'latency sensitive')...

    So yes, it does squeeze more bandwidth out of our ageing copper, under certain conditions and with some bad trade-offs (and increased long-term costs). But at some point in the very near future, unless we figure out a way around the laws of physics, we'll hit a hard limit and have to rip out the copper and consumers will have to replace their modems. And in the meantime we'll have hobbled our economic productivity...

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