NBN G.Fast “more hype than reality”, says Budde


news Veteran telecommunications analyst has described the G.Fast technology which the NBN company plans to deploy into its network from 2017 as “more hype than reality”, questioning whether it will actually be able to deliver on its close to a gigabit speed promises.

Last week the NBN company attended the Broadband World Forum in London. At the event, the company’s chief architect Tony Cross revealed that the company had been trialling the G.Fast standard at an office block in Melbourne. In the trial, the NBN company was able to achieve speeds of 600Mbps on a 100m stretch of copper than was more than 20 years old.

The company has also trialled G.Fast in its National Test Facility in Melbourne, achieving speeds of nearly 970Mbps over a 20m copper cable.

The G.Fast standard effectively gives telcos such as the NBN company who are upgrading existing copper networks a great deal of flexibility in terms of their infrastructure. Cross wrote in a blog post that the company could deploy G.Fast in apartment buildings simply by installing new equipment in their basement, or deploy it to a group of houses via a Fibre to the Distribution Point model, where fibre is extended further into the copper network.

“G.Fast really allows us to remove the need to actually enter premises to deliver ultra-fast speeds,” Cross wrote. The next step for the NBN company will be to trial G.Fast with retail ISPs, possibly looking to a launch date of around 2017 for the technology in the wild.

The technology is also being trialled by telcos such as BT in the UK, which already have widespread Fibre to the Node networks and are looking to upgrade those networks further. BT has delivered speeds of up to 1Gbps using the G.Fast standard.

However, in a post on his own blog this week, Paul Budde — one of Australia’s most senior telecommunications analysts — poured cold water over the NBN company’s plans.

Budde pointed out that G.Fast used a much larger frequency range of signal processing bandwidth than other copper-based broadband technologies, over a much shorter copper length. Some of that frequency range would overlap those used by commercial VDSL (which the NBN company is using for its Fibre to the Node network), Budde wrote, so running both technologies together caused interference.

Another issue, he wrote, was the extra power supply that was needed for the extra network nodes that would need to be deployed for G.Fast to work.

Budde’s view is that G.Fast is effectively “another rollout” on top of the existing rollouts that the NBN company has already committed to — with all the associated costs and time required. In this context, it is his view that it may be worth proceeding with the original Fibre to the Premises model instead, which offers a much less complex model, without the compromises involved in a broadband rollout based partially on copper.

“… the hardware industry, sensing the desperation of the telcos’ need to start improving their broadband networks, are talking up the G.Fast technology. However at this stage – and at least in a commercial sense – that technology is still more hype than reality,” Budde wrote.

What is my own opinion on G.Fast? I’m currently researching the situation and hope to provide an informed view at some point. It’s a relatively new standard, and the field trials are also new — so I am seeking to weigh the evidence before I come down on one side or the other on this one.

Image credit: Mikael412


  1. “I’m currently researching the situation and hope to provide an informed view at some point”
    Please seek out and find an technical expert. Not one of Australia’s “analysts” whose “technical” knowledge was gained from company PR and management and the few tech people they trot out to give selected highlights to hype their products.

    • The problem is, where do you get a technical expert from?

      Most of the experts in this field come from a vendor (eg Alcatel-Lucent) or a telco (eg Telstra or NBN Co) who actively want to promote the product. Well-meaning, but hardly objective.

      That’s why journalists quote analysts such as Budde so much. Because they have actually examined and compared the technology and spoken to lots of different sources to get context. However, they don’t owe any allegiance to a vendor or telco — they’re independent.

      Happy to take suggestions on who I should speak to :)

      • @Renai Budde “independence” is he’s not employed by a teco or equipment supplier. Such “independence” isn’t compromised in any other way (number articles critical of FTTH rollour?)?

        All of the technically capable are or have been connected to these companies, such skills have a high value in this area.

        Buddes’ comment re vdsl & g.fast being incompatible on the same copper is telling. This isn’t a deployment scenario.

        These unable to accept a scenario of continual upgrades, paid for from revenue generated by matching customer demands at significantly lower capex, keep pushing FTTH. Tellingly they never peovide figures; but always asserting cheaper.

        • Richard you already said the MTM won’t make money so they won’t be able to upgrade when demand. Yes lower capex traded for higher OPEX which in turn makes it more expensive.

          But when we provide S2 of the SR you dismiss it.

          • @jk again you misunderstand the financal basis. The cross subsidy model of MTM will make it fail, this was discussed in detail a couple of days ago. FTTN / B / dp is financially viable for most Australians, FTTH isn’t.

            Additonaly opex doesn’t come cloee to capex, again figures have been provided over at BS. You and Gregory continue to assert otherwise without providing any (his over hyped claims easily destroyed with realworld data).

            SR figures have been discussed often. It’s your contortion (comparing S2 with CP16) that is impossible to comprehend. We’ll continue to disagree.

          • From the CP16

            Management and the Board have not taken a view on assumptions beyond that time, and no better estimates exist than the assumptions applied in the Strategic Review dated December 2013.

            is it impossible to comprehend comparing last know figures if they had continued to roll out FTTP.

            But then you where claiming the y26-y28 was if they had contined.

            We have even now Verzon claiming there reason to drop FTTN for FTTP was OPEX but then you ignored the SR of MTM spending more the $30B in OPEX 7 years after its complete

        • Look Richard, I’ll put it this way.

          You have on the one hand a technology that that has been proven since the early 1970’s – there was a Fibre to the Premises roll out in the US in the early 70’s in California. FTTP has been put through over 40 years of every Devil’s Advocate skeptic engineer trying to rip it to bits. And that’s the way it should be in all science.

          Then you have G Fast, a still experimental, new technology that uses copper that can be of any length, in any state and have any level of interference.

          Even if you could argue that G Fast is cheaper and quicker, to choose it ahead of FTTP is like saying to some school kids, OK kids your education will be quicker and cheaper BUT one catch – throw a dice for whether you have to leave school in years 9, 10, 11 or 12.

          But I give up with you. You are not in the business of rational, evidence based discussion.

          • (But I give up with you. You are not in the business of rational, evidence based discussion.)
            That’s why I have never bother with him.

          • +1 MikeK..

            One can only bang one’s head against the wall of rationality for so long!

          • That’s very informative Derek…

            I always simply assumed it was the big flaccid penis hanging from the middle of their foreheads which impeded their vision of reality ;)

          • You guys have only just figured out that Richards arguments are all ideologically based and not based on fact???!

            That’s why he drinks the Turnbull Kool Aid on FttN and has faith in the cooked books the “new” nbn calls Corporate Plans….

          • No Tinman, we knew at ZD some 5 years ago… we are just ensuring those in doubt know too ;)

          • Excellent read Derek O,
            Most societies are divided into a party that wants change (the more liberal party) and one that is afraid of change (the conservatives). (The liberal party is generally more intellectual and the conservative party is more anti-intellectual.
            The conservative party is big on national defense and magnifies our perception of threat, whether of foreign aggressors, immigrants, terrorists, or invading ideologies like Communism. To a conservative, the world really is a frightening place.)
            This is the Coalition to a capital T,

          • (The amygdala is an almond-shaped structure deep in the brain that is active during states of fear and anxiety. )
            that’s no penis on Richards forehead its his swallen amygdala. Richard is scared, his fear leads him to irrational thoughts and ravings. It leads to delusions of grandeur and eventual madness. So keep rattling Richards cage and help him on his way.

          • MikeK…

            This explains why conservatives can’t see reality…

            A young conservative goes to the doctor with a small lump on his forehead and after testing …

            (Doctor) – I’m afraid you are growing an extra penis.

            (Conservative) – Well cut it off.

            (Doctor) – No I can’t do that, even conservatives have a brain and this is far too closely rooted to the brain. It would cause brain damage or death if we tried to remove it.

            (Conservative) – Well it’s ok as long as it doesn’t get any bigger

            (Doctor) – Oh it will grow to full size

            (Conservative) – So you mean I’ll have to go through life looking at a fully grown cock hanging from my forehead?

            (Doctor) – No of course not…

            … you’ll have two big balls covering your eyes.


      • Grab some of the white papers on the technology and contact the authors and the other sources of information in the bibliography directly.

      • Renai.

        I am pretty sure that when Peter Ferris gave the talk on “mythbusting the NBN” in 2011 that’s on youtube, he was asked in the Q&A about G Fast. He replied that NBN Co could not use any technology that was not standardized and was still experimental.

        If you could find Peter Ferris then he would know!

        • @at a sensible position in 2011,however G.9700 & G.9701 standards approved in 2014.

          • Richard, there are whole rafts of standards approved that may never go anywhere. There are 3 improvements that I know of off the top of my head that were going to improve ADSL2, none of which they could get working in the real world. For example, something that seems simple, variable bit rate ADSL2. Rather than drop sync when there is noise on the line the line was to slow down and then speed up once the noise margin improved. Never got it working properly.
            Even vectoring is an issue, BT in the UK are still trialling it after more than 3 years, trying to get it to the point that it works well enough to do a rollout. Now they looks like they are examining G.Fast as a solution to their problems with speeds falling as more people take up connections. Having a standard doesn’t mean a technology will work in real life. Especially when you have people touting it as a technology for general rollout. That was not what is being said by the people doing the actual development in their papers. They say it’s an alternative to FTTH for places that it’s very costly or impossible to get fibre all the way to the premises. That FTTH is best alternative if the fibre can get taken all the way to the premises, ie, like most houses in Australia. The reason was because it wasn’t significantly cheaper than FTTH. A study I posted the other day showed FTTdp was only 3-15% cheaper than FTTH. G.Fast has it’s uses, connecting a premises where the lead in conduit has collapsed, FTTB, that sort of thing.

          • Richard.


            Not that I normally quote a community based encyclopaedia as the perfect source of information, but fibre based standards have been around for quite some time too.

            And yet here we are. Hoping that a newly minted technology will somehow trump the end game; once again persisting in the same fruitless exercise as NBNco, despite the rest of the world continuing to struggle with ratified standards that don’t actually deliver in the real world.

            Vectoring is still an in-exact science. G.Fast is heavily reliant on vectoring, as is highly sensitive to cross-talk; so all of the ducks have to align.

            Or you can, you know, just use fibre instead.

          • @darren g.fast silicon exists, deployment happening.

            BT has withheld vectoring deployment becuase 1) exisitng speeds already capturing majority of demand & 2) higher speed trials indicate their support for g.fast.

            Please repost your link showing g.fast saving less than 15%, cost greater to enter a premise. Why is this point, laboured by every telco and equipment manufacture impossible to comprehend? Entering premises is very expensive. It is delussional to suggest otherwise.

            @bb you could use fibre, BUT AT A COST!

          • No commercial deployment Richard. Yes there is silicon. I’ve had silicon for all sorts of devices years before the final products are ready for commercial launch. But if the cost of the development silicon I’ve had is anything to go by I am sure paying 50 times the price of the final product to roll out early is very cost effective.

            The point mentioned by every vendor is that G.Fast is a technology to supplement a FTTH rollout to allow connect to premises that would otherwise cost huge amounts of money to get fibre to. That doesn’t include pulling a bit of fibre up existing conduits. It’s commentators and PR people getting all excited by G.Fast and claiming it’s viable.
            Here’s Huawei study on the costs of rolling out a FTTH network compared to FTTdp and FTTB G.Fast
            “Entering premises is very expensive. It is delussional to suggest otherwise”
            Yes, every time I open my front door it costs me thousands. I’ve not seen any customer charge NBNCo to enter their house. How does this fantasy that fitting equipment in the premises cost more than fitting in a pit work?

          • @Darren
            “Yes, every time I open my front door it costs me thousands. I’ve not seen any customer charge NBNCo to enter their house. How does this fantasy that fitting equipment in the premises cost more than fitting in a pit work?”

            The reason given is the trenching from the footpath through the yard.

          • Martian H that is only true if Telstra hasn’t maintained the ducts. But if there isn’t any ducts for the copper and direct buried like they are in the UK they are going to have problem when it rains.

            Richard has quoted that FTTP is only 9 sec slower than MTM which is pretty impressive for as you say digging up front yards vs connecting copper to a node.

          • @Martin, in the majority of cases there is already a conduit and it works fine. If there isn’t or it’s collapsed THEN drop a box in the pit for FTTdp. This is what G.Fast was intended for, not a general rollout because someone doesn’t want to use FTTH because of a political agenda.

      • Renai,
        I think all you can do is keep a close watch on the BT trial in the UK, BT having one of the largest FTTN roll outs in the world want to recoup that investment by offering higher speeds, only time will tell if it is not feasible, I don’t think anyone can make that call this early in testing.

        • Reality
          First BT has FTTdp we don’t since g.fast required 200m.
          Second the chips the run the gear can only handle 11 premises at this stage.

          So for us to run out FTTdp requires laying more fiber closer since MTM at $56B is only $8B less than a FTTP rollout NBN own figures it going to cost more doing this 1 2 3 step to sound fiber all the way then doing it the first time.

          • Jason K

            BT has FttdP because they are installing it, it doesn’t mean it won’t be deployed here if G. Fast trials in the UK and here look feasible.

            Currently the NBN Co is looking at deploying FttdP as well, but it may be only for those premises more than 1K from the node, you can rollout FttdP just to get minimum speed requirements on long node distances, it’s not just for G.Fast.


          • Again Reality NBN node to pillar are an average of 350M more than the distance for g.fast to even work.

            So they would need to extend more fibre which will cost more than $8B difference

            So since MTM cost only $8B less to rollout than FTTP how much more of this folly is going to cost becuase you realise that this approach is going to cost more than deploying FTTP.

          • Jason K

            “So they would need to extend more fibre which will cost more than $8B difference”

            How do you know what it will cost the NBN Co to extend the fibre to support G.Fast, it’s only being trialed in a small way here , no where near the size of the trial in the UK, the NBN Co have made no announcement on how many residences they plan to upgrade to G.Fast, if it happens at all.

            So why are BT bothering to trial G.Fast, they should have gone the easy route just asked you and rolled out FTTP instead.

            BT/Openreach in the UK are rolling both FTTP and FTTN, I think they know what they are doing and what the most cost effective option is area by area, they are one of the biggest listed blue chip companies on the UK stock exchange making good profits and paying healthy dividends to shareholders, they are not in the business to lose money.

          • Yet reality we are build a network costing $8B and when it comes time to upgrade again say 5 years after its complete it going to cost more than the $8B difference to do it right the first time.

          • You didn’t really answer the question why BT are wasting their time and investment in trialing G.Fast when they could have just chucked out FTTN and put in FTTP, because according to you to upgrade to G.Fast is the same cost as upgrading it to FTTP.

          • Reality It is a trail not investing yet due to its limited range 200M and limit of 11 premises the only practical use if for FTTB due to already small copper length. BT trailing it on dp which NBN isn’t doing yet but again only limited to 11 customers within 200m.

            But then are you saying rolling fibre to the premise is cheaper then what you are claiming. Wasn’t it the rolling fibre closer to the house is the most cost?

            To rollout g.fast on FTTN then a dp on every street block (200m and 11 premises). Can run it from the node as the average length of copper NBN is putting in from node to pillar is 350M longer than the limit of g.fast.

            I have said before FTTB is a great short to medium term becuase the copper is short and weather proof.

            But like you quoted before NBN doesn’t know what it’s going to do for premises out side a 1k from the node this still have made that decision in the SR they don’t start doing FTTdp as as y21 and as late as y25

      • How about you get an expert from both the FTTH department and FTTx Alcatel-Lucent sell both products so it would be good to get them in a room together. Love to know if those guys get along. :)

        • Turnbull tried to do his interview awhile back happy to talk about the issue of FTTP but then trying to stop them talking about the failing of FTTN lol couldn’t even edit it out.

          • Indeed Jason…

            And of course Mal has made up his own mind… he invests his own $’s overseas in FttP.

  2. The thing that annoys me about this discussion is that nobody seems to get that the speed and reliability of G.Fast drops off exponetially the further from the exchange you get.
    Its all great to say that you can get nearly gigabit speeds at 20 meters but when was the last time ANY house had a 20 meter run?

    I lived in a house in kalgoorlie that was right next to the exchange… I literally looked out my window (2nd floor townhouse) and looked into the exchanges back yard and THAT house had a 35 meter run.

    In suburbia you’re going to get hundreds of meters at LEAST.

    This is such an infuriatingly ulcer inducing political whitewash.

    • You’re definitely correct about all that, but as I understand the G.Fast standard the intention is to use it in combination with a FTTdp rollout – that is, you put a “shoe box” sized node within 20m of the house you’re connecting to and run fibre up to that, with the remain 20m using the existing copper.

      Now if that is the case, the obvious question becomes why the hell wouldn’t you just run Fibre all the way in if you’re going to be putting it literally within metres of the house anyway? Just build FTTP and be done with it.

      I’m no expert, so all I’ve just said could be completely wrong. But if it is correct, then I tend to agree with Mr. Budde. It’s essentially going to be an entirely new rollout to replace all the FTTN areas with FTTdp, which renders the whole G.Fast thing utterly pointless because for all that effort and money you may as well run FTTP and forget about it.

      • Why spend $11 billion buying something you are only going to use the last 2% of!

        It gets worse when you consider G.Fast lab vs NG PON2 Fibre in the real world.

        That extra 39km of cable distance is a massive thing in this country too given how large it is. Why we want to go down from 400m to 20m is anyones guess (guess my B-in-Law gets his own cabinet on his porch then lol).

        • @sm Conroy paid $11b, renegotiation of the contracted transfered the infrastructure for nothing. This is a popular misconception started by the ABC, a compliant has been lodged.

          The question is not whether fibre has considerably more bandwidth than copper (universally acknowledge as true), but at what cost and how fast speeds demanded by consumers can be delivered at a price theyre prepared to pay.

          Talking 10Ge to the premise is ridiculous. Transist, backhaul & servers unable to handle anywhere near such speeds. CVC pricing alone would be unacceptable.

          • “renegotiation of the contracted transfered the infrastructure for nothing”
            No it didn’t. It would be true if the only changes were in the transfer of the infrastructure. The extra cost for the infrastructure is at least the following.
            – NBNCo now have to pay for any remediation work on the pits, ducts and copper.
            – Telstra/Foxtel gets to use an upgraded and expanded HFC network for free for Foxtel services.
            – Telstra/Foxtel no longer hand to maintain the HFC network.

          • Yet Richard you don’t explain why peak funding on MTM is double the price of the capex while FTTP is less than half.

          • Talking 10Ge to the premise is ridiculous.

            Talking about a fibre capable of 10gbps does not necessarily mean delivering services capable of 10gbps. The only ridiculous thing would be the inability to understand that. Regardless as I’ve said in the past with fibre as the speeds increase due to new encoding etc the psychological barrier becomes bigger because people (usually ill-informed people) assume those are the speeds that are being proposed for the majority of end users.

          • “Yes most future opex costs transferred to NBNCo, but so was revenue (far greater).”
            Source? All government paid-for ‘reviews’ released to date have indicated that the MTM brings in significantly less revenue than the original FttP plan.

          • “Talking 10Ge to the premise is ridiculous.”

            Why? you want me to quote the massive amount of people over the years that have said x amount will always be enough?

            You realise that 10Ge (well 40) could be a 10 year or 20 year deal with fibre if its cost holding things back its just a matter of waiting until that reduces etc. Its not even remotely possible with Cu. (1Gbps isn’t possible with FttN currently)

            Also obviously you don’t start at the consumer end (puttin words in folks mouths there)!! FYI Optus just finished 100Ge upgrade to their intercap network.

            “how fast speeds demanded by consumers can be delivered at a price they’re prepared to pay.”

            Well under NBN Co we would have had up to 1Gbps (100Mbps for residential) services for the same pricing we have for ADSL now. with a ROI to NBN of 6-8%. If you want and can pay you can get it 100% guaranteed. (people moved suburbs to get onto the fibre in SA for their work/business).

            MTM its now a dice roll as to how close or far your node cabinet will be. What speed internet I will get is based on a random placement of a node from my house. Not what I am prepared to use (100/40) and or pay for (100/40) I get no choice, I’ll get an upper limit due to conditions+distance and have to pick somewhere within that, I could be stuck on 25/5 for all my luck. Reverse is also true, those only wanting 25/5 might be some of the few that can get the top speeds. This will impact the ARPU (which is shaky as is atm) more than I fear anyone’s bothered to think about. Bring on the next CR and Cu uptake figures.

      • TW,

        “Now if that is the case, the obvious question becomes why the hell wouldn’t you just run Fibre all the way in if you’re going to be putting it literally within metres of the house anyway?’

        Because as in BT’s case if you already have FTTN infrastructure in place it is obviously a cost effective speed upgrade, the G.Fast distribution box can be in ground or on a pole.

        “At this point, it seems likely that BT will choose G.fast as the successor to FTTC/VDSL, with a commercial roll-out in the UK pegged for around 2016/2017”


  3. Deployment classes will almost certainly mean any G.Fast deployment would be at the mercy of whether it’s classed above or below VDSL2 and Vectoring. Assuming that vectoring is in use. Do we know that answer yet? How can a product go on sale and it’s deployment class is still a black box?

    For example, ADSL2+ Annex M is currently at the mercy of Annex A. So an M based connection can be artificially constrained (usually at the port) to keep cross-talk within agreed terms (as set by ACMA, IIRC).

    If G.Fast crosstalk impacts the existing VDSL communications standards (has ACMA released anything on this either?) then it’ll be neutered. Regardless, it’s a bit premature to be spruiking “proof” that copper can go fast (of course it can, 10Gbit Ethernet, anyone?) when you’re still making a (ham-fisted? bacon?) meal of even getting VDSL out the door.

    I’m going to take the G.Fast message as it was intended “see, we could upgrade to this some time in the future, copper isn’t dead!”. Also isn’t G.Fast typically also used in the same sentace as FTTdp? Which isn’t what is being done, beyond multiple dwelling units?

    NBNco’s mission now is to ‘prove’ fibre isn’t the end game; even though it really is. So we see these announcements appear, entirely in a vacuum.

    • I’m going to take the G.Fast message as it was intended “see, we could upgrade to this some time in the future, copper isn’t dead!”. Also isn’t G.Fast typically also used in the same sentace as FTTdp? Which isn’t what is being done, beyond multiple dwelling units?

      NBNco’s mission now is to ‘prove’ fibre isn’t the end game; even though it really is. So we see these announcements appear, entirely in a vacuum.

      Exactly +1

        • Then Richard that includes Malcolm Turnbull!! He has said in the past that fibre is the end game.

        • No, they aren’t wrong at all, wasting time on various unnecessary intermediate steps in no way proves that. It’s just finding an excuse to delay it even further for political reasons.

          • Exactly HC…

            My new analogy is like travelling between say Sydney and Melbourne…

            You can book a flight, get a good price, fly and be there in no time (the FttP analogy)

            Or you can (as Richard obviously would with his MTM analogy) catch a bus to a, then hire a car to drive to b, jump on a train to c, perhaps a g.fast fucking hovercraft (like they do in Dorset, so it’s great) to d. and in the end he realises it actually took him heaps longer and he paid as much if not more.

            But having dat ego (there’s a blast from the past) although he surely must realise it… he just can’t quite man-up to admit, hey fuck, maybe I should have just flown.

          • Rizz don’t forget that the time you spent traveling is like the on going cost. Just time wasted going down the drain. Much like the OPEX

        • Fibre is the end game. Turnbull has freely admitted that fibre is the logical successor to copper.

          His mandate has always been cost, and a fractured, ludicrous policy pushed by a technologically bereft (ex)leader; it’s politics and lobbying, nothing more.

          G.Fast makes great sense in places where fibre is problematic (eg MDUs); but do not translate that to meaning it’s a replacement. It’s a solution to a problem, not the solution.

          Someone, eventually, has to foot the bill for a FTTdp/ FTTH upgrade. We had a shot to do it ‘now’; instead we’ll have to do it later.

        • “alternately those claiming fibre is the end game were wrong”
          Such as Turnbull and his paid-for ‘review’s that you usually love quoting obvious misrepresentations from?

        • And yet again just for you Richard…

          A fellow libertarian/conservative (from the always right Fox network ;) but a libertarian with personal wealth and profile way more up the ladder than you (you know that ladder you are on about the second or third wrung, but strangely believe nonetheless, you are looking down on us all here at Delimiter) so by your rules, his opinion is worth ooh, about 200 of yours ;)

          Feel free to send him your thoughts on wonder copper and saving a dollar or two, I’m sure he’ll laugh possily more than us here, if that is possible…


  4. Budde nails it as usual.

    Funny how all these new schemes rely on pushing fibre further, beyond FttN when we’ve been told that’s all we need. That length of copper that’s getting even shorter with each new announcement but don’t you dare think about putting that fibre into the home, we must half the distance with every iteration to prove the worth of the copper that we left each time, got to get our moneys worth from the expense from upgrading each time too. That length of copper right there that keeps getting shorter, that people is “The Hypocrite Zone”

    • If I am a fibre zealot or a fibre fetishist then this extreme fibre phobia sounds like a condition that needs therapy at Dr Phil’s Hacienda Clinic.

    • Funny thing is HC, after telling us all such FttP speeds are ridiculous and that even 100mbps was unnecessary because there are no apps, only a few years ago, these copper relics now boast about speeds available on (lab) copper and g.fast… that they themselves argued against…

      Much like Fraudband and err rolling out umm, err, fraudband… ROFLMFAO

      More backwards logic from the illogical/ideological.

      • This is true. They’re really saying “G.fast is great and it’s a big selling point of our network, but we won’t need it!” I am astonished at the reasoning demonstrated. People who demonstrated this kind of reasoning for something like a national infrastructure project used to live in mental institutions.

  5. I get the feeling that with these new FTTB type deployments of this kind of technology that ground floor apartments will be increase in value, since they’ll be able to get closer to the maximum speed, with the higher you go the slower the connection.

  6. Global telecoms is a massive industry, and there will be places where g.fast makes sense. But suburban Australia ? to get copper runs down to 100m means running fiber down every street. Then installing active electronics outside every second house. It seems very unlikely that this would be an economically viable thing to do compared with running fiber the last 100m.

    Good luck finding an expert. The issue in question is deployment, maintenance and operating costs compared with fiber. Until it is deployed at scale, there are no experts.

  7. G.Fast sounds like a fantastic fallback option for MDUs. It’s almost perfect as it provides near-fibre performance (asymmetry not withstanding) without the need to re-cable an entire building.

    It makes zero sense in the kind of FTTN deployment we’re seeing, with nodes up to 800m from premises; which I think is the crux of Budde’s article.

    • yeah 100% that G.Fast will probably do wonders for all those MDU’s eg that TPG are targeting. That is such a small % of the rollout however and totally won’t be helpful in most of the 38% area of Australia. I mean if we’re already talking a node cabinet between 2 or half dozen homes in wider burbs/hobby area’s it’d have to be cabinet on the porch kind of territory (because even the sheds probably too far away as well lol) it’ll cost too much so won’t happen (well I hope not).

    • Yes, it’s a great technology. As a supplement. As for applying supplements nationwide, it does seem like a stretch and yet G.fast is the great white hope, the low-cost cavalry that will save the day for the national deployment. It’s supposed to make up for FTTH and make it unnecessary. This casts a large shadow over that idea.

      Keep in mind that this good point from Paul Budde is coming from someone who regularly reports on complex factors associated with commercial deployments in telco markets around the world. So it’s anything but a vote of confidence in the MTM.

        • Nah, we have to keep hoping beyond hope that copper is going to have magical improvements over the coming decades that means we never need Fibre, because “reasons”.

          I see Vplus to VDSL as ADSL2 was to ADSL. Its an upgraded version, providing a better product, but it cannot get around the limits of copper.

          No matter what, over copper phone lines, without being within meters of the fibre, you won’t get the throughput you can get through fibre. It isn’t as if fibre research has stopped and isn’t going to improve just as quickly as copper…

      • I am sure Alcatal are not selling and developing VDSL2 vectoring, Vplus or G.Fast to lose revenue, despite the purveyors of doom & gloom the FTTP or bust brigade in Australia I am sure all products will sell well worldwide for many years to come.

        That was one of the main problems with the Labor FTTP decision, it locked the NBN Co out of looking at more viable economic decisions as new technology to speed up FTTN was developed, it also locked them out of the rational economic decision of using the existing HFC infrastructure for BB (which was being left up for Foxtel anyway) and its future upgrade to DOCSIS 3.1.

        FTTP was a blinkered one trick pony that was “too ambitious”.

        • Reality again talking about 3.1 upgrade while 40% of the country will have to pay more to get that same level of service. But then again you claimed we don’t need those speeds so why are we spending $B’s on upgrade to faster services when according to you is not required shouldn’t they pay for the upgraded service like the people on FTTN?

          • Jason K

            The NBN Co have stated they will consider upgrading to DOCSIS 3.1 if the demand is there, DOCSIS 3.0 will do for now, isn’t that the rational way to do it, the infrastructure is in place providing DOCSIS 3.0 and it’s a small incremental cost to DOCSIS 3.1 – IF IT IS REQUIRED.

            BTW How do you know the 40% of residences that are on a FTTN/B service (which is all of them when it is completed) want a immediate upgrade to HFC DOCSIS 3.1 like speeds?

          • “The NBN Co have stated they will consider upgrading to DOCSIS 3.1 if the demand is there, DOCSIS 3.0 will do for now, isn’t that the rational way to do it, the infrastructure is in place providing DOCSIS 3.0 and it’s a small incremental cost to DOCSIS 3.1 – IF IT IS REQUIRED.”

            More contradictions from the illogically ideological.

            “FttP is a waste because we/I won’t need such speeds…”

            “BUT if we do”… ROFL

            The much vaunted (by your twin) CBA said Aussie households will only require 15mbps by 2023, so why are my tax dollars being wasted on either these unnecessary FttN & HFC add ons for faster speeds or conversely why were my tax dollars wasted on a flawed ($1.5m) CBA that Ergas himself said could be done in a few days?

            Oh wait of course.

          • @Reality ” the infrastructure is in place providing DOCSIS 3.0 ”

            Its actually not all there you will have noticed recent news about Aussie HFC networks needing to be upgraded (By Telstra likely paid for by Taxpayers).

        • @reality considering there is a huge legacy base of FTTN built over the last 15 years why wouldn’t they develop VDSL line cards to replace the existing ADSL cards. In Germany alone there are 300,000 existing nodes to upgrade!

          Stop conflating installed base with new rollouts. The fact is almost no one is deploying new nodes to suburbs, most of the business is in upgrading existing nodes.

          • So sales of new Alcatel FTTN cabinets with VDSL2 vectoring gear inside has virtually come to a halt?, they are only keeping the manufacturing production lines open for Australia’s order, jeez that’s generous of them.

          • @reality nodes are just a standard outdoor rated cabinet with standard Telco grade components bolted in.

            There’s nothing special about nodes! Look in any Telstra rack in any exchange and you’ll see the same type of rectifiers, line cards, switches etc. The only Big difference is the ups configuration.

          • So you admit that Alcatel doesn’t need to do anything special to make nodes for NBN then …. Good now that we have cleared that up what’s next?

        • No, Alcatel know when they have the market cornered.

          They offer such products to desperate incumbents (or governments rolling out self confessed FRAUDBAND) who would otherwise be fucked, so as to profit from these copper owners desperately trying to wring every last cent from their antiquated copper.

          Surely that’s common sense to those with two eyes?

          You argue because a company makes a particular product for another company to keep utilising their antiquated products, it therefore legitimises the antiquated product? People can still buy film for cameras, it doesn’t therefore render such a device current or wise…FFS

          But no, I don’t blame Alcatel or the incumbents for profiting from their copper, from their own business perspective. But when customers, wherever they reside are held to ransom with obsolete copper, just to appease others profit margins, well one would have to be a goose to jump on the copper bandwagon IMO.

          Good thing you blokes weren’t around back when or the very copper you now cling to would have been “too ambitious and a one trick pony” and ironically, would never have even been built.

  8. I’m on several Body Corp committees besides the day-to day running of the scheme we look for different ways to save money and/or make our complexes a bit more attractive to buys and sellers. I would favor fibre over G-fast because of the year to year electricity cost and the favourable reception from buyers. The end result would depend on costs but if the costs was reflected in the increase of the unit sale price and the reduction in electricity of VDSL2+ or G-fast then I can’t see G-fast gaining a great deal of traction in FTTB schemes. We have just replaced all fluros with LED lighting and we will recoup the cost in 18 months. What would it cost to run a basement node for 54 units 24/7 year in year out. It might not be much but we look at things over time like 10 years or 20 years and who replaces modems in the basement, is that at our expensive ???? or NBNs.

    • You are looking at around 7watts per VDSL2 service and as high as 14watts per VDSL2+Vectoring Service.

      So assuming you have 50 apartments in a complex (I have no idea how many are in most MDU’s), then you have a 350 watt constant load for VDSL2 which will consume 8.4 kWh’s of power per day. With vectoring its more like 700 watts load and 16.8 kWh’s per day.

      To put this in perspective, the average house apparently uses around 20 kWh’s per day.

      • Thanks Derek O, I’ll make a copy of your reply for future reference and if you used G-fast your electricity consumption may well be higher than the average house hold per day.

        • No worries.

          Tbh I think g.Fast will be no higher than VDSL2 with vectoring as it’s the extra processing overheads of vectoring that drive up the power usage above normal VDSL2. Add in silicon process manufacturing shrinks for g.Fast chips and you have a similar power envelope.

    • Mitek,

      The other issue for a body corporate will be the radio/TV interference caused by G.Fast. What will the tenants say when they find that they can’t receive their favorite AM/FM or TV station or enjoy Shortwave or Amateur radio? Let alone find that a whole bunch of consumer devices (remote control, medical, etc) suddenly stop working?

      The technically illiterate will poo-hoo all this, but society as a whole is coming to realise that we pay a huge cost in thoughtlessly trashing our precious Radio spectrum. It’s got so bad in the USA that the FCC is planning to carry out a national survey to try and identify the many sources that have crept in over the years.

      Anyone who understands Electromagnetic Compliance or Interference Minimization will tell you that G.Fast will be a nightmare. You cannot feed broadband MF/HF/VHF/UHF signals down an un-shielded cable without jamming the operation of a whole range of consumer and medical products in the vicinity.

      If you want proof, try listening to AM radio anywhere near a VSDL modem. VSDL really only affects AM broadcast, but G.Fast will basically make any radio equipment unworkable in the immediate vicinity.

      Why would we want to feed this appalling (and totally unnecessary) crap into out homes?

      • The technically illiterate will poo-hoo all this

        I’ve brought this up many times. I’ve also posted that I don’t agree with how the current government is killing off the community radio/TV spectrum so they can sell it to Telstra. No one listens until they are affected though :o(

  9. None of the copper protocols are orthogonal, thus deploying multiple protocols / signalling in the same cabinet raises the noise level for all.

    Run gfast with other gear, you may find gfast works fine and others all suffer due to the increased noise. Depends on cable and patching separation.

    Doesnt matter if its fibre to the post box and 20m runs tho, fibre has zero crosstalk. You can put megawatts and milliwatts down parallel fibres with no issue (melting aside).

    • G.fast noise is considerable, it even interferes with FM/DAB radio.


      Each area will need to use different frequencies to avoid wiping out local radio, plus if they follow the standard, then they won’t get the speeds at the distances they’d like (or they need to reduce the distance a lot using something like FttDP).

      I guess that’s the trouble with hacks like G.fast, they are so close to the limits of what they can do, it starts effecting everything else.

        • Nothing very surprising there, G.Fast is ideal for MDU’s. It’s pointless for Nodes tho as they are too far away from the premises to deliver a worthwhile upgrade. Beyond 100m in cable length it’s really a waste of time!

        • Well the Communications Alliance doesn’t consider it a ‘hack’.

          Regardless, a hack is what it is:

          Noun: Hack
          Something that does not fix a problem but offers an alternative method to avoid it.

          It just offers an alternative method of getting around the issue (the issue being copper is very close to it’s limits).

          And I am not against the idea of using G.fast (it’s sensible to use in MDU’s and any area using FttDP) or HFC (as long as it’s upgrades to DOCSIS 3.1) in the area’s that it makes sense to use them.

          But a nation wide roll out of them from effectively scratch? That’s just plain ridiculous…

          • The NBN Co have not announced they are going to roll out G.Fast as a ‘nation wide rollout’.

          • @Reality (alain)..

            “The NBN Co have not announced they are going to roll out G.Fast as a ‘nation wide rollout’.”

            So WTF are you whining about and wanting to argue with everyone over then FFS?

            BTW it hasn’t been NBNCo for some 6 months… keeping your record of facts to 0%.

            And it only cost $700,000 to drop the Co… money well spent I’m sure you’d argue (even though you didn’t even fucking know) :/

            You’re welcome.

            I bet you start saying NBN instead of NBN Co and won’t thank me – AGAIN… like you did when I explained the RFP asked for FttN or FttP and you argued and argued (and I let you) then I posted the announcement and you tumbleweeded/disappeared tail between legs. Only to then resurface and start quoting that it was in fact FttN or FttP yourself… ROFL

            I repeat, you’re welcome, I like to help educate the uneducated even when they have their fingers tightly in ears and their one eye safely shut ;)

        • Sounds like a hack to me.

          And regardless of how many hacks are implemented eventually you have to replace the copper. That is a given. The question then becomes when do you do it and what to you replace it with.

          If you are a logical person concerned with waste and consider current and future needs you say: Let’s get on with the inevitable and replace it with Fibre now. Kill two birds with one stone.

          If you are GimpCo and the coalition clowns you say: More copper and let’s wait for a future unknown to try and squeeze more out of it.

          The most absurd thing about this is that they haven’t even started rolling out FttN in any significant form and they are already talking about rolling out G.fast, and what does that tell you?

          1. They lied about the speeds we’ll need.
          2. FttN is not up to the task, just like we said.

          • TBH I dont think calling G.Fast a Hack is really a fair portrayal … what it is is a great example of the “law of diminishing returns”. Twisted pair is now at the point where any improvements come at large costs (CapEx & Opex) and are not significant.

            Compare this with Fibre (eg xGPON) where gains come at modest CapEx cost, provide big leaps forward and reduce Opex.

        • So what your saying is we don’t actually need fast FttP speeds we just need err, G.fast speeds… which are just right (for a country who only needs 15 mpbs by 2023…LOL)

          Wow the illogically ideological are now taking on the goldilocks law… it’s just right, not too fast not too slow, juuuust right.

          ROFL, the dumb excuses get’s better (by better I mean fkn sillier) by the minute alain…!

          • So what your saying is we don’t actually need fast FttP speeds we just need err, G.fast speeds…

            That’s the funny thing isn’t Rizz, the speeds we need always magically align with what copper is capable of delivering in the coalition clown world. For years we heard them bleat relentlessly about FttN, “25mbps is more than enough” and the rest of the ill-informed rhetoric. Now they’ve discovered G.fast, so it’s the new cure all to distract everyone while they roll out a fraudband network that will need replacing regardless.

  10. So, we will need to have node spacings at 100 metres? That’s an awful lot more nodes than they are currently planning, just to get the (?) 600 Mb/s vapourware in maybe 5 years.

    Didn’t turnbull say that 25 Mb/s was “more than enough for a family”. If that is/was the case, why are NBN Co pushing so much on speed-envelope stories, with nodes at ever-decreasing distances?

    Sure it makes sense to deploy fibre ONCE, and enjoy the mult-terabit potential now, the next 50 years and at 40+ km in distance?

    Am I missing something here?

  11. Every time nbnco mention these upgrade paths, it is all smoke and mirrors. They don’t even have the money to finish the FTTN rollout. They are spending $50b rolling out nodes 800m from the premises – to upgrade to GFast or this unproven XGFast, will require even bigger rollouts to put nodes 100m from the premises. To service the same area as a VDSL2 node, the rollout would need an extra (at least) 63 nodes, all requiring power/cooling/extra fibre run to it. Currently planning for 50000 VDSL2 nodes, a GFast rollout would need over 3 million new nodes. How many 100s of billions would that cost, when we could roll out fibre now for a fraction more than the current FTTN rollout, and get 1Gbps now, and upgrade to 10Gbps by just upgrading the endpoints. Never mind all the other benefits of fibre over copper ie. immune to interference/heat/cold/corrosion/lightning, passive nodes etc.

  12. If HFC is good enough (according to Turnbull) for the majority of us in Sydney, why bother with FTTN at all? Just roll out an HFC network everywhere. It must be cost effective, because you can build the network, fully depreciate it over a period of nearly 20 years, and then sell it for big bucks to the taxpayer.

    • David,

      Actually you make a valid point, it is probably not cost effective in areas like regional cities where you have to start from scratch and where FTTN is more viable, but extending HFC out further to suburbs where the rollout stopped in the next suburb is a valid question to ask.

      I guess it all comes down to the cost of new HFC (you would want to rollout DOCSIS 3.1) and its supporting distribution infrastructure vs FTTN using existing copper.

      • “You then have to ask yourself: Should the taxpayer have to spend ($56) billion when there are so many infrastructure demands where there is a screaming need now”.

        I don’t want my tax payer dollars on such a white elephant network with unneeded gold plated Ferrari speeds? Especially because “12 megabits per second (Mbps) connection to the home is enough for anybody…”

        And “there isn’t much, if anything, you can do with 100 [Mbps] that you can’t do with 12 [Mbps].”

        See what I did there? How visionary eh?

      • If you were going to go through the trouble of employing labourers to actually run new cabling, then you might as well just run Fibre. Running new Copper/HFC now is ridiculous, unless the HFC is being used to just fill in an area currently underserved.

        • “unless the HFC is being used to just fill in an area currently underserved.”

          Which I think all that is going to happen, limited spot fill out of areas of suburbs, for example 95% of a postcode is already HFC but 5% missed out only because a date was picked and the HFC rollout was halted by both Optus and Telstra.

          • I remember someone telling me that he had HFC out the front of his place and no one used it, all it was good for was the pigeons.

            That was the argument used then to suggest FttP was doomed, like HFC.

            Now HFC is great…

            :/ unbelievable

          • So, Reality, you agree that if you’re running new cables in 2015, you wouldn’t run Copper or HFC?

            Wouldn’t that…. make Fibre the ultimate endgame? If in Greenfields you don’t run HFC or Copper, but Fibre, then eventually, even with advances in Copper tech, the length of the copper line gets shorter and shorter, to the point where it is 99% Fibre 1% Copper.

          • Yes I know you run FTTP in Greenfield’s that’s why the Coalition NBN Co will continue to run FTTP into these areas because there is no existing fixed line infrastructure.

            FTTN is for the existing Brownfield copper infrastructure, in terms of any new HFC rollout I think it will be very limited for the reason I explained.

          • @Reality, I guess you’re missing my point, if running Fibre in Greenfields is done, then there must be some kind of reason for that.

            If that reason is that using copper in 2015 onwards is ridiculous in new rollouts, it would seem to me that Fibre is the end game.

            If Fibre is the end game, then even in Brownfields, eventually, they will be Fibre too.

            If they are eventually Fibre too, then is it not better to do it once than do it piecemeal over multiple rollouts, costing multiple billions each time?

          • R0ninV3ph…

            No he’s not missing the point at all, he understands it fully and realises it disproves him, so he’ll either disappear or simply look for an angle to deflect…

            Our old friend alain (advocate, Node for me, Reality etc – the most banned poster in Delimiter history) has never come here for rational and friendly correspondence he’s here on a mission from (the ultra conservative) GODs… to push err, BS and that’s what he does and to give him credit, although I can see he is desperately FOS, he does a pretty good job at dismissing facts and deflecting negative articles and as such earns his few pieces of silver for disgracefully helping to sell Australia out…

      • But even in a new area, you can string HFC off power poles and a dozen guys can pass hundreds of premises in a day. Surely that’s going to be more cost effective than a FTTN solution that requires a node be within 100 metres of every end-point. Anyway, what’s the point in squabbling over whether the solution will deliver 25, 100, or 1000 Mbps, when the reality is that for less than a hundred bucks a month, during peak times you’re going to be lucky to average 2Mbps anyway?!?

  13. Two things. The NBN is a fraud and secondly arguing about the last mile is utterly moot.

    If we’re provisioning 50, 100, 1000mbps to the tails only to charge $20 per mbps at the PoI then no matter what way you calculate it the end-customer will simply not receive that bandwidth. Not even close to it. Literally a fraction. A contention ratio far worse then what we see with ADSL.

    If I buy a 100mbps tail from the NBN and expected 1:1 then this is some of the costs that my ISP would be charged are:

    CVC: 100mbps X $17.50 per mbps = $1750.
    ISP wholesale backhaul: 100mbps X $15 per mbps = $1500
    NBN Recurring Charges (rack, per tail fee, misc): Y
    ISP Expenses (Customer Service, plant, value added services network etc: X

    Total $3200 + probably another $500 per service for the other expenses.

    RRP that your paying = $89.95

    Total cost: $3700 / $89.95 = 41 – 41 one customers who would need to be paying $89.95 in order for the ISP to pay for the costs of a single 100mbps 1:1 service.

    Dividing 41 customers by a 100mbps of backhaul/CVC gives you a grand total of 2.43 mbps of bandwidth per customer. In order to use a reasonable say 30-40mbps of your 100mbps service you’d need to have at least 20 of your neighbours away at the time.

    The entire NBN is a fraud. Its a lie because of the way it was created. There are to fraudulent concepts that have been disseminate via the media to the educated public. Firstly that bandwidth/back haul is some sort of scare and hugely expense resource and secondly that digging a trench/laying a line costs tens of thousands of dollars per metre.

    First issue: This country is dripping in fibre. Our railway lines, powerlines and other utilities have fibre dug next to it. Our radio media companies own fibre and microwave transmission networks that they’d love to utilise for telecommunications (since they only use a fraction of the bandwidth) if it weren’t for cross media laws. Between Telstra, AAPT, Nextgen, Optus and Vocus between them have probably over a hundred thousand kilometres of fibre dug in. Along main roads across every city and town in this country.

    Second issue: Using a combination of the telegraph poles, power and gas conduits and trenches dug by a backhoe (about 5 minutes of work) we could lay fibre via nodes on your main road. Laying the cable via the pole would have been quick and painless, especially for sites with tricky access, problematic underground cable/issues and other problems.

    The NBN could have been rolled out years ago for a fraction of the cost whilst, most importantly, giving everyone a 1:1 internet connection.

    The government acquires houses to build critical infrastructure and yet in this case, in the biggest infrastructure project ever, they choose to overpay for network assets like HFC, that Telstra and Optus had written off and driven into the ground. They over paid massively for the copper and did so without a moments thought.

    The government was happy to give away tens of billions of dollars already, to their friends. Executives running Telstra and the other telcos standing to become immensely wealthy because of these acquisitions/sales.

    The telcos have no interest whatsoever in the NBN. They would have fought it had the NBN not put in the CVC.

    See the CVC ensures the sanctity of the wholesale, government and enterprise segments. See these customers pay $10-20k a month per fibre connection. Yes they get a few extra bits and bobs with it but basically all telecommunication services are driven over the top of a fibre connection.

    Had the NBN offered 1:1 $100 fibre to the house connectivity the corporate would have dumped their overpriced services and jumped right on board with a NBN service.

    So as I said in the beginning. The NBN is a fraud. The entire thing is about enriching friends, family and associates of the political parties. The laborious path taken to complete these project is a scandal of such proportions that both parties should be subjected to a royal commission and criminal charges.

    So arguing about the last mile is utterly pointless when you’ll never get to use a 100mbps.

    • The internet is based off a contention based model today – what’s being implemented doesn’t change that. Providers still need to manage contention in their networks no matter who the services are bought from. There is no magical solution to fix that & commercial reality always plays a part.

      Using existing infrastructure is normally a good idea. From observation though, the person using it needs to be confident it will be supported. Getting access to it does require time, effort & money – so can occasionally be easier to just build your own.

      • Michael

        Stop trotting out the same old crap. The CVC pricing component does not exist in the market today for carriers who own and operate their own xDSL networks.

        Backhaul is really cheap these days. When I was privy to wholesale backhaul rates it was $20 per mbps for gigabit sized orders, and that was a few years ago.

        The contention ratios have now become a means to create huge margins out of xDSL customers and have nothing to do with providing a fair way to share a limited resource as was the case when internet networks were first developed and it cost tens of thousands for a ISDN 64kbps link.

        The NBN’s CVC pricing component of $17 per mbps on top of backhaul costs and other expenses makes it impossible for anyone to get full throughput. We’ve been conditioned to accept mediocrity. Oh my 100mbps connection is only going 10mbps. I guess that’s good enough.

        And the point of this pricing construct – to purposely make the product expensive so as to protect the massive cost of a enterprise grade internet connection.

        I know exactly how much governments and large corporations pay for their internet services and its ridiculous.

        Its not 1985 and we don’t have 2mbps frame relay connections running the internet.

        The problem is the apologists who don’t realise just exactly how much money is being made here. You know when a large Telco pulls in $25b in revenue and claims that it costs $21b a year to run the business that money isn’t being used to pay for transmission (that was provisioned a decade plus ago). Its going into the pockets for executives, senior managers. Eye wateringly large bonuses.

        The industry makes bankers look like boy scouts. I’d dare say we put drug dealers to shame.

        Half the people commenting on these publications work in the industry. Misinformation disinformation, techno-babble laced with lies to confuse the punter into thinking that mediocrity is acceptable.

        Well its not. We have the fibre in this country to give every individual 1:1 internet (domestic that is), and why don’t we have it? Because the government is beholden to the sector, its greed and donations.

        Think about this – how much money has Ziggy, Broad and Turnbull raised for the liberal party. I’m acquainted with some of their family friends and have been told its a stupendous amount of money – though of course the AEC has no record of these. activities – that has effectively given them control of our telco policies and direction. And that’s just three individuals. There are dozens executives who have joined the party over the years.

        Whilst on the ALP side all of the telco unions have their own vested interest in running (as was the case in the 1970s and 80s) huge government owned telcos using it buy seats in areas by setting up call centres, employing thousands of unnecessary linesmen and so on. All of those stiffs being members/money for the machine men who run the ALP.

        Back to bandwidth. Even internationally backhaul didn’t have to be a limited resource either. The space sector (another incredibly corrupt and parasitic industry) should have given us a cheap launch platform to launch dozens of GEO sync sats instead of being forced to spend $20k a day on fibre laying ships.

        • CVC does exist elsewhere in the ICT industry, have you seen Telstra’s Velocity CVC pricing lately? It’s $70 per mbps!!!!

          That said I think the CVC/AVC model should be crapped and a better system found that doesnt penalise ISP’s for providing low contention ratio services to customers.

          • “CVC does exist elsewhere in the ICT industry, have you seen Telstra’s Velocity CVC pricing lately? It’s $70 per mbps!!!!”

            sure when you buying a L2TP tail sure your forced to buy Telstra transmission to connect to the AGVC.

            Hence why everyone went and installed DSLAMs into their exchanges.

            Though this move by the government with the xDSL crosstalk issue is a way to destroy the independent xDSL providers.

            I reckon they’ll make the NBN the universal service provider for xDSL services, forcing all of the ISPs to pay into the NBN.

            Nationalising communications.

    • Japan has significantly higher contention ratios than you’re describing, and I regularly manage to get well above 150Mbps down and 80Mbps up.

      They must be using some kind of Ninja Magic to make it happen.

      • Where is the published data on Japan’s contention ratios, or the architecture for their NBN style fibre connections?

        As far as I can see in the Japanese market internet is dirt cheap, there is fibre everywhere and the wholesale costs have no CVC style pricing component.

  14. What is the problem with fibre to the home now that NBN is putting it on the poles just like the HFC networks must be cheaper than trying to use the old copper only took 2 days to string it up for 50 homes in my street!

  15. When commenting on the first mention of G.fast back in 2013 https://delimiter.com.au/2013/08/02/so-is-g-fast-a-thing-or-not/ I got one thing wrong with my assessment and that was the deployment timeline. It is actually running 3 to 4 years ahead of where I would have expected, with first run chipsets and pilot systems appearing late last year, leading to competing vendor commercial product a year later.

    To one of the comments around VDSL2 compatibility – well it’s actually a matter of choice and both technologies allow for new psd definitions so G.fast being limited to a start frequency above 17MHz to allow the VDSL2 17MHz operation may not be as beneficial as restricting VDSL2 to 8MHz and allowing G.fast to operate all the way down to 8MHz.

    A blended bandplan to accomodate a mixed VDSL2/G.fast deployment has been identified within the ITU and over the “short” VDSL2 loops (that are “long” G.fast loops), the G.fast speed gains are impressive as the bandwidth from both transmissions are aggregated. A new node build with both technologies deployed would be complimentary in performance terms rather than compromised to protect VDSL2.

    The experts I have spoken to have given the impression that the technology is performing better than expected – remembering that the theroetical performance calculations are based on establishing a reference loop model that includes a range of bridge taps and other nasties that can be encountered in the real world, and perhaps these reference models have been overly conservative.

    This technology, as conceived, was really about avoiding the need to “dig the roses” to get fibre to the home from the street, so more like an extension of the cat 6 cable in the home, but as happens the boundaries get pushed and then the synergies with other technologies develop – in this case VDSL2.

    The delivery model is still unclear for G.fast – the self install model aligns it fully to FTTN with VDSL2, so the RSP foots the bill for the RGW, and lets face it the cost of the VDSL2 chipsets with vectoring compatability is not that great given the global volumes, but adding in G.fast will be expensive in the early years, so is self install palatable to the RSP funding the more exoensive WAN port?

    The integrated RGW with Wifi and LAN ports could be a crunch point cost for an RSP, but a two box model where the NBN provide the customer modem unnecessarily increases the total cost of ownership to the country, but arguably makes it look more like FTTH with an IAD being equivalent to the ONT from an architectural perspective.

  16. Since he doesn’t provide any evidence to support his assertions, he’s appealing to (his own) authority, expecting us to just accept his assertions, which leads to a logical fallacy.


    He and others make the assertion that FTTP will nearly always be cheaper, yet they never provide evidence that it is, contrary to the evidence that it isn’t.

    He’s being an FTTP zealot, as are many others.

    So here are the rules to being a zealot:

    (a) Don’t believe and don’t accept any evidence that is contrary to your beliefs, despite how credible the evidence and the people behind it is. Use logical fallacies to dismiss them.
    (b) Assert your own views are true. Use logical fallacies to support them.
    (c) Don’t seek objective evidence to support your views, because you might unintentionally find evidence that shows your views are wrong.

    Perhaps there is a lot of irony in people getting zealous about technologies that are a direct result of humans overcoming their zealotous tendencies and instead adopting a scientific (objective) approach.

    • He and others make the assertion that FTTP will nearly always be cheaper, yet they never provide evidence that it is, contrary to the evidence that it isn’t.

      yep, must be why Verizon isnt aggressively shutting down its copper PSTN & HFC wherever possible … oh wait they are!


      At the time of that article, there was concern that Verizon’s gamble on FTTP was not going to pay off for investors as FiOS was making only a minimal impact on revenues. However, it was already showing significant OPEX savings – a key factor in Verizon’s initial decision to deploy FTTP.

      yep, nothing to see here, move along please….

      • And we have Richard claiming the OPEX increase cost doesn’t equal the “saving” in CAPAX

      • Yet they’re also stopping their FTTP (FiOS) deployment!

        After Billions in Subsidies, The Final Verizon FiOS Map Is Bleak as Hell

        Did I just hear your head explode?

        • Yep, after passing more than 19 Million premises they decided they needed to spend more CapEx on improving their Mobile networks. This is what happens when you have Private companies cherry picking the profitable areas.

          This is in fact exactly why we needed the NBN to build near a universal FTTP network for Australia!

          That said, they are investing in upgrades to their existing FiOS Network – oh look, 10/10Gbps per wavelength over 40Kms with 80Gbps easily doable by adding another 7 colours. Makes G.Fast look kinda lame doesnt it!


          Oh and did you know you can add NG-PON2 to an existing base spec GPON network without impacting the services or speeds of the existing customers. This is entirely unlike ADSL + VDSL which interfere with each other if operated on the same copper bundle and force the VDSL services to run at a fraction of their potential speed as a result!


        • Verizon has about 10 million fixed line customers. They’ve moved most of the high-value 50% to fiber and they’re steadily migrating the rest over. In 10 years time there will be very few copper customers left.

          Verizon are very profitable. The fact that they are good at getting government subsidies – something many large companies are good at – says very little about their technology choices.

          Across most of the US, there is very limited choice for broadband. This is because fixed line telecoms is very close to a natural monopoly and it is rarely economical to build a network if you have to compete with some one else. This is why Verizon is not expanding their fiber network beyond the 19 million (!) premises it already passes.

          The map in that article just says that Verizon have deployed fiber where (a) they already had copper customers to migrate over and (b) most of the people live.

          The reality is that the whole world is steadily moving to fiber. There are niche applications for other technologies like g.fast. A niche in a market worth trillions is still a big opportunity, which is why companies are pursuing it.

          so sorry, no exploding heads.

    • “He and others make the assertion that FTTP will nearly always be cheaper, yet they never provide evidence that it is, contrary to the evidence that it isn’t.”

      It seems universally accepted that FTTP is the endgame solution for broadband. Copper cannot compete in regards to potential bandwidth nor in OPEX.
      The MTM therefore is only a stepping stone to FTTP for fixed line infrastructure.

      Some questions:
      1) How much more would it have cost to implement FTTP instead of the MTM?
      2) How much will it cost to upgrade those premises that receive FTTN/HFC to FTTP at a later date?
      3) what is the benefit to Australia as a whole in the intervening years between when MTM is completed vs when FTTP would have been completed?
      4) is the benefit in #3 greater than the cost for #2?

      With the delays and cost blowouts for the MTM, I cannot see how the benefits received by implementing the MTM a few/several years earlier than FTTP could exceed the cost to upgrade the MTM to FTTP.
      I do not have specific $$$ figures, it would be nice if they could be introduced into the equation.

  17. (a) Don’t believe and don’t accept any evidence that is contrary to your beliefs, despite how credible the evidence and the people behind it is. Use logical fallacies to dismiss them.
    (b) Assert your own views are true. Use logical fallacies to support them.
    (c) Don’t seek objective evidence to support your views, because you might unintentionally find evidence that shows your views are wrong.

    Copper Zealot Bible right here guys.

    • PS. I think Douglas Adams must have met a Conservative the day he wrote this:

      Protect me from knowing what I don’t need to know. Protect me from even knowing that there are things to know that I don’t know. Protect me from knowing that I decided not to know about the things that I decided not to know about. Amen.


    • (d) resort to calling people names if you can’t dispute their points of argument

      • Yep, the “Fibre Zealots” tag from TurnBull sure get’s bandied about a lot by those insisting on keeping legacy technologies and business models safe from the “scary” future!

        • “zealot
          a person who is fanatical and uncompromising in pursuit of their religious, political, or other ideals.”

          If you can’t accept the idea that FTTN is going to be the most cost effective option in a subset of situations, then you’re not being objective.

          • See that’s the difference between those advocating FTTP over legacy copper services, we use facts and real world examples to back up our points – you Copper loving folks rely on out of context quotes, legacy examples and faith for your arguments!

            An example is the “technology agnostic” claim – here in the real world, we assess the merits of the aviable technologies and select the right one for the job and FTTP is the right technology to service Australia’s needs for the next 60 years. Leagcy copper based services (FTTx in particular) are essentially obsolete now (except in very limited circumstances)!

          • Arguing that FTTN is the most cost effective option is akin to arguing that catching a taxi is the most cost effective option versus buying a car when you know you’re going to be driving in to work every day for the next twenty years.

            What’s worse, in this case, is that we’re going to blow tens of billions on taxi rides and then pay almost what we would have paid for the car anyway. Unless you think we’ll still be using copper at the end of the century?

          • cretin
            a stupid person

            Gee I can describe FttN lovers too.

            Anyone who thinks cost is the only gauge is not being objective or sensible (i.e. see above).

            Especially following the $15B blow out. And before alain says UPTO, if they thought it was going to be $4B or $5B they would have said upto $5B surely. The fact they are saying upto $15B is of course because…?

            So let’s consider that Mal said he had a fully costed plan at around $30B, but that blew out by half again… even before they even had a paying FttN customer. (iirc).. wow.

            So not even considering ROI, the fact that even Mal says fibre is the end game that dispels cheaper. So what about faster.

            2 years of no FttN, although again Mal’s plan was ready to go?

            Then there are people in our area and many others (even with the lesser FttP lag) who would have already had FttP…

            So no cheaper and no faster says the Zealot to the Cretin :)

          • If you can’t accept the idea that FTTN is going to be the most cost effective option in a subset of situations, then you’re not being objective.

            I can do that too:

            If you can’t accept the idea that FttP is going to be the most cost effective option in a subset of situations, then you’re not being objective.

            Thank for stopping by!

          • If you’d said FttB/dP then sure we’d be all singing along wishing labor had just bit the bullet and admitted it for MDU’s etc.

            you also said “cost effective” and not cheapest (and technically it is still $8billion cheaper atm).

            If we have to go from a->b->c but b costs $50b and c costs $70b

            if we are doing C anyway as the ultimate goal why step through b spend all that $$ only to change it to C for as much again … assuming the largest part of the cost (labour) hasn’t increased which it usually always does (and hence you end up spending far more to get to the same point just taking longer and spending more $).

            SA Govt build a 1 way Freeway because spending the extra $40million on a tourist attraction was more important or other reasons (cost diff of going dual carriageway). Cost today to get that road upgraded to a two freeway is 1.5-2x the original budget (many 100’s of millions). That doesn’t factor in having the road under construction again for another x years and the disruptions either.

      • (d) resort to calling people names if you can’t dispute their points of argument

        Awesome. Another Copper Zealot Bible verse. Keep them coming!

        • “(d) resort to calling people names if you can’t dispute their points of argument”


          More reverse wisdom from the illogically idelogical…lol

  18. Alcatels scammy technology is a marketing con. The issue is there is noise and interference to begin with ! No amount of fraud con will fix that.

    They have a product to sell and no doubt Turnbull has shares in the company.

  19. 600Mbps on a 100m stretch of copper is not worth skiting about. We still want FTTP not this 2nd rate crap.

  20. “… the hardware industry, sensing the desperation of the telcos’ need to start improving their broadband networks, are talking up the G.Fast technology. However at this stage – and at least in a commercial sense – that technology is still more hype than reality,” Budde wrote.

    Except where it isn’t.

    “Alcatel-Lucent and Chunghwa Telecom begin world’s first commercial G.fast deployment”


      • ” and at least in a commercial sense – that technology is still more hype than reality,”

        You let this go through to the keeper, understandably of course.

        • Point me to commercially active connections on G.Fast.
          Chunghwa Telecom and BT are of course going to activate some areas with G.Fast, why do you think they are only doing it in two areas?
          It’s still in trial phase to see how it will cope in a real world situation.

          • And with that… he disappears like a thief in the night, only to of course return later to claim the same already disproven (and he knows it) BS again and again…

            And thus, the circle of illogical ideological FUD continues…

          • Soth,

            I don’t have to point out anything, I just linked to a large Telco announcement that stated their G.Fast was the first commercial deployment, if you have info that states the contrary go for it.

          • @Reality
            And if you’d actually really read the link you would have realised Chunghwa are doing a FTTP roll out nationally and using G.fast for MDU’s…

          • Tinman_au

            So how does that differ from the NBN FTTB deployment here intended for 11% of residences under the estimated Coalition target and using G.Fast if the demand is there?

            If it’s FTTP as we know it as in fibre all the way to inside the residence why use G.Fast at all?

          • @Reality… He just said, it’s FTTB for MDU’s.

            Which is precisely NOT FTTP. What are you not grasping with his statement?

          • @Reality

            This is not the droid you’re looking for….

            Even when Mike was running NBNCo I thought, and posted, that FTTB was a great idea for MDU’s…

            I don’t hate everything about Malcolm’s MTM, there is a lot in it that could easily and cheaply get Australia ahead of the curve.

            But it is a huge waste of time and money starting a FTTN roll out in this day and age, when most other countries have already been there, done that, and are are switching over to FTTP.

          • ROninX3ph

            “@Reality… He just said, it’s FTTB for MDU’s.

            Which is precisely NOT FTTP. What are you not grasping with his statement?”

            No he didn’t.

            “you would have realised Chunghwa are doing a FTTP roll out nationally and using G.fast for MDU’s…”

          • @Reality, I feel like you’re purposefully being stubborn in understanding the statement about a nationwide FTTP rollout using G.Fast for MDU’s.

            Using FTTP for single premises then FTTB for MDU’s…… That is what that means.

    • “… the hardware industry, sensing the desperation of the telcos…”

      Pretty much how I answered you Reality (atop page) you while you keep on about Alcatel. Copy/paste-

      “No, Alcatel know when they have the market cornered.

      They offer such products to desperate incumbents (or governments rolling out self confessed FRAUDBAND) who would otherwise be fucked, so as to profit from these copper owners desperately trying to wring every last cent from their antiquated copper.

      Surely that’s common sense to those with two eyes?”

      But like FRAUDBAND you now embrace, I see you also embrace Alcatel?

      Ironically, the very company who were worthless according to you guys previously, because they had dodgy, if not criminal according to you lot, management (what was that blokes name from Alcatel you guys used to slander – oh that’s right Mike Quigley).

      Funny how every comment you make inevitably contradicts a previous comment, demonstrating no shame or ethics, in ones blind illogical, ideological crusade…

      Have a nice day :)

  21. It’s amusing and somewhat sad to see the same repetitive strategy used by some in the FTTP supporter camp, desperately trying to shutdown any comment that doesn’t join the club in the kick the Coalition NBN agenda with all sorts of bizarre diverting subject matter with a ‘we own this agenda get out’ mentality, while gleefully giving each other high fives on a fictional job well done.

    In the meantime in the real world the Coalition NBN rollout goes on unabated on a NBN platform that won them the election, FTTP supporters look back with nostalgia to the glory days of the before 2013 election and the Labor FTTP rollout, not ever coming to terms with the fact that they were not glory days in the first place.

    • You are totally full of it Reality, The ENTIRE Reason for MtM was to remove the costs associated with replacing everyone’s lead-ins with Fibre.

      If they were serious about this we’d see GPON being rolled down the streets with Micro-Nodes (VDSL2 and reverse powered) attached to the GPON Multi-ports. This would give everyone 100/40 quite easily and allow for really affordable FoD for those who need it.

      The fact the the coalition clowns havent done this and have also commissioned reports/reviews claiming we all only need 25mbps for the next 10 years shows their so called reasons to be total BS. The NBN is being destroyed to suit political/corporate agendas and nothing more. There is no valid technical or economic rationale to roll out new FTTN in 2015+ … end of story.

      PS. the HFC upgrades I can live with, at least DOCSIS 3.1 has some merit, FTTN has none!

    • Reality it that the same platform that won them the election so NBN 3 year plan is wrong we should all have the NBN by the end of next year and only cost $29B?

    • This is the sort of delusional comments expected before the 2013 election.

      Today even though they got the copper cookie they squealed for before the election they seem to have become even more delusional and shrill because it’s starting to crumble just as we predicted. Pride wont allow them to admit glaring faults. That copper is shiny, don’t shatter the dream for them.

      The $56 billion coalition clown faudband plan goes on “unabated” as if that were an indication of it’s merits. Regardless we wait for the garbage bins to hit our streets to offer us something we should have been impressed with ten years ago.

      427 days to go!

      • Indeed HC…

        To show the dishonesty of the cooper relics and their illogical ideology driven stupidity..

        Remember when FttP NBN was first announced and then underway, the comments from the perpetual naysayers?

        $43B for faster porn, gold plated Ferrari, white elephant, 12mbps is enough for anyone don’t use my taxes for your pirating, the $43B should be spent on hospitals, roads, schools etc with a cheaper broadband network provided (enter exhibit A Mal’s fully costed $29B FttN plan)… that’s what we need. Faster/cheaper than this $43B socialist monopoly. See FttP is already behind schedule by months.

        Fast forward to today…

        MTM/FttN already looking like being 5 years behind their completion promise, with some $15B cost blow out and the total now estimated at $56B…

        And the naysayers now say…

        “Yeah so…”

        • MTM/FttN already looking like being 5 years behind their completion promise, with some $15B cost blow out and the total now estimated at $56B…

          The really funny thing about that blow out? They haven’t even started building it yet :)

          I expect it to become even more expensive than what they’ve planned, especially as their planning is based off dodgy made up numbers from made-to-order CBAs/SRs/CPs

          • Tinman_au

            It’s maybe ‘up to $56B’, I appreciate why you and others ignore the correct way of stating the figure, it’s because you desperately need to beat up the difference, if misquoting is the way to do it so be it eh?

            Of course if the higher figure of the range is downgraded at any point that will be ignored and $56B will always be used.

            It also helps to totally ignore the constantly revised upward funding figures on the back of downgrades to rollout targets in the Labor NBN Business plans, but that was ok, they were always going to make up the shortfall in the next reporting period, except it never happened.

            You get so used to looking the other way pushing the all FTTP fixed line agenda it becomes second nature.

          • Reality, you are incorrect, the schedule slips under labor’s NBN Co never resulted in cost overruns as they were fixed price build contracts.

          • @Reality

            If the Labor FTTP had blown out double like FTTN has, then maybe I’d agree with you, but the facts of the matter are it didn’t…

          • UPTO $56B you say Mr pedantic/apologist?

            Well if it’s like UPTO 25mbps to 50mbps by 2016 for all Aussies perhaps it’s actually a shit load more.


          • LOL Rizz, I have to remember to not drink a beverage when I read your comments :-)

          • Derek O

            “Reality, you are incorrect, the schedule slips under labor’s NBN Co never resulted in cost overruns as they were fixed price build contracts.”

            I didn’t use the term ‘cost overruns’ , I used the term funding.

            The required funding was $40.9B in December 2010 raised to $44.9B in September 2013.

            BTW I love the use of the mild almost innocuous term ‘schedule slips’ to describe this. :)


          • Tinman

            “If the Labor FTTP had blown out double like FTTN has, then maybe I’d agree with you, but the facts of the matter are it didn’t…”

            Explain to me with actual figures how the FTTN has blown out double.

          • Reality
            Maybe you should tell Turnbull and NBN about upto like a statement I got from them
            “NBN expects that around 90% of FTTN end-users will be able to get access to wholesale download speeds of up to 50Mbps.”

            But then 1Mbps on a Min 25Mbps when it’s really an upto. As BT like to give a speed guarantee is the average speed of there slowest 2000 customers.

            But then $41b to $44B while they are building is still slot better than $29B to $41B to “upto” $56B would you agree.

            But you still haven’t explained why when you keep talking about the coalition election policy when there rollout target has double with out building anything and has upto double the price without building anything.

            Yet you still talk about g.fast or 3.1 for HFC when even you said we don’t need those speeds so while are we wasting taxpayer money on even trialing it should as the saying goes if they want the speed they should pay for it?

          • @Reality

            My bad, you are correct, that should have been “up to double, and quite probably beyond once they actually START building the damned thing…”

          • How about FttN costs have blown out UPTO 100000%?…

            That’s the good thing about the old trusty UPTO it means UPTO SFA…

            Again I reiterate if NBN thought the cost blow out would be $4b-$5B they’d have surely said UPTO $5B or if $9B-$10B they would’ve said UPTO $10B. But they said UPTO $15B so… ?

            And the humorous part alain is you still living in the past (even after your little spiel the other day…LOL) because MTM is so fucked with their UPTO $56B blow out… you have no reply and here you are making a big deal about the past AGAIN… and FttP going from $40.9B to $44.9B.

            Oh please… my sides are splitting with laughter. Keep up the good work.

    • “It’s amusing and somewhat sad to see the same repetitive strategy used by some in the FTTP supporter camp,…”

      LOL amusing indeed to us, not you, because it’s known as supplying the facts, hence repetitive… I.e. the facts don’t change. You ought try facts even once, it won’t hurt I promise.

      As opposed to the disgraceful, double somersault with triple pike and quadruple contradictions and lies you always used and still use (I’ll say it HC… “alain – explaining the mantle of the most banned Delimiter commentator, in history).

      “…desperately trying to shutdown any comment that doesn’t join the club…”

      No, refer to one again – FACTS. Your lies are easily disproved and the FACT that when someone disproves you, you either disappear (tumbleweed as one described it) rather than man-up or come here to sulk, as you have done here, so as to avoid the topic.

      I see nothing has changed, oh except your hero Abbott is now a leaner and personally I think Turnbull is a vast improvement…but then the neighbours dog would have been an improvement, so…

      Yes the MTM rolls on at snails pace many “years behind and $15B cost blow out” for a third rate, inferior, mish-mash network… brilliant.

      Have a nice (factual) day :)

    • In the meantime in the real world the Coalition NBN rollout goes on unabated ..

      Just like to point out that most of that ‘unabated’ work has been fibre, wireless and satellite. So actually it’s the NBNco rollout that has continued, based on a prior government’s commitment.

      If we stopped everything to deploy FTTN, then apart from ‘sky muster’ and some wireless work, there would be no progress.

      It’s funny, but the world doesn’t deal in absolutes. Confusing a technology for idealism, or policy, and then making sweeping generalisations – shows a marked lack of understanding.

      G.Fast has great promise for MDUs and potentially other locations where fibre is complicated to deploy. Honestly, it does. But much like how a “hammer” or “chainsaw” isn’t the best tool for every job, G.Fast isn’t the “end game” because we’re not (re)designing the copper network to be the end game.

      It’s really is as simple as that. The costs and deployment method may be argued back and forth, and almost certainly will be argued long after the heat-death of the universe. It doesn’t change the simple fact that fibre is the best replacement for copper.

      It has a long service life, does not degrade in the same fashion, and can be endlessly upgraded, much as copper has been.

      Pointing at bits of paper squealing about how much we’ve saved now and isn’t that fibre outrageously expensive and who needs that, crikey such a waste because of reasons that were once based on a little bit of factual data that really just ends up as a logical fallacy as soon as you consider the lifespans expected – ignores the price paid in future.

      Short changing the country by deploying something that will mostly work but won’t really have the same financial benefits isn’t to be congratulated. We can and should do better.

Comments are closed.