NBN not worried about 5G threat. It should be.

NBN Co’s CFO, Philip Knox, is unperturbed by the rise of 5G mobile broadband, claiming that it does not pose a material threat to the National Broadband Network’s (NBN) business model:

“We spend virtually zero time thinking about it or talking about it. We’ve always said it’s a complementary technology. It’s highly optimised for mobility and less so for fixed wireless”…

He based his confidence on the fact 5G relies on millimetre wave spectrum, which, while it has far superior capacity than 4G spectrum, travels only a few hundred metres and struggles to penetrate solid surfaces.

Philip Knox should be concerned. Telstra, TPG and Optus are each rolling out relatively cheap, fast 5G alternatives to the NBN.

As reported last month by Fairfax:

Telstra this week delivered peak download speeds of 4.2 gigabits per second (Gbps), four times faster than the fastest NBN plan available to homes, using 5G technology.

Meanwhile, Optus is gearing up to launch two 5G wireless services priced at $75 a month and $90 a month respectively that on paper can easily go toe-to-toe against fixed-line NBN plans in the market.

The $75 a month plan promises maximum download speeds of 100 megabits per second (Mbps) and an average download speed of 85Mbps between the busy 7pm-11pm time slot. The $90 a month plan, according to Optus, “currently delivers an average download speed of 214Mbps (between 7pm-11pm)”…

Anecdotally, I struggle to maintain 40 Mbps down and 15 Mbs up on my 50/20 NBN plan. Yet when I caught up with my mate last week in the park, he was getting 80 Mbps down on his Telstra mobile (checked via SpeedTest.net). And I don’t even think he’s on 5G!

Given the poor service I receive on the NBN, I would change to 5G mobile broadband in a heartbeat if it was cost competitive and reliable. I suspect many other Australians would do the same.

Regardless, as more services and technologies emerge to compete against the NBN’s fixed-line broadband monopoly, the likelihood of NBN Co ever hitting its subscriber targets and breaking a profit will recede.

The Parliamentary Budget Office in March reported that the “fair value” (or saleable value) of the NBN was just $8.7 billion, which is than one-third the federal government’s equity investment.

This valuation is looking more shaky by the day as more competing players emerge.

Sooner rather than later, the federal government will need to bite the bullet and heavily write its NBN investment down.

Unconventional Economist
Latest posts by Unconventional Economist (see all)


  1. They’re not concerned because the physics make it unsuitable for large numbers of high volume users. Sure people can get high peak speeds, but congestion will prevent it from being viable in populated areas.

    • They are pretending to be concerned so that the $7.10 per month broadband tax is extended to 5G.

    • Yeah, it’s self-limiting.

      The same way Melbourne/Sydneys’s road networks are only just barely adequate on a good day, BECAUSE the trains and trams exists, is the same way wireless works pretty well most of the time BECAUSE a fair chunk of the traffic is on the fixed network.

      Have a public transport strike and watch the traffic jams.

      They’re complimentary technologies. That’s not to say wireless won’t be perfectly adequate for many users – and there’s a real risk that the NBN business case that relies on less than ~20% moving exclusively wireless – but wireless will always fall over if there was a ‘public transport strike’ in the form of a national fixed line outage.

      The real question is the percentage. That’s unknown. If 25% of people can get by on wireless, NBN’s business case is blown out of the water.

      • macrofishMEMBER

        > They’re complimentary technologies.

        This is what most people dont get, to do 5g right and get any kind of decent cell density you need a large fixed network to plug it to…. kinda like the nbn

        • I just looked up the specs for 5g. It’s surprisingly quick per cell with up to 2.5gbit – making it equal to the current fibre NBN if only 32 users are on the cell. The fibre NBN can switch to 40gbit GPON at a moment’s notice though.

          However, one thing most people don’t know is that wireless modulations change constantly depending on the signal. If you’re close, it’ll use the fastest modulation for your share of the bandwidth, but if you’re far away it’ll drop to a lower speed modulation.

          If you want a truly mindbending read, try and understand QAM https://en.wikipedia.org/wiki/Quadrature_amplitude_modulation

          If I were to simplify it… There’s an animation that sort of explains it, but imagine a grid. 2×2. There are 4 possible grid references, A1, B1, A2, B2. Each of those can represent 2 binary digits. 00, 01, 10, 11. Land a constantly changing signal with the right clock timing in A1 and you get 00. That would be your very low speed QAM.

          Now look at a chess board. 8×8 64 squares. 64-QAM. You now get 6 bits (eg 101010). 3x the data per ‘instant’. The boundary (signal range) of the grid hasn’t changed. But the granularity of it has. You need a clean signal to operate reliably at 64-QAM.

          It keeps going to (so far) 4096 (64×64 – 12 bits per grid reference. 4x the bandwidth of 64-QAM and 6x the first example.) and to 32768-QAM (15bits, but that doesn’t seem to use a perfect grid). You need an even cleaner signal to get this – which is why it started out on cabled networks.

          There’s a rule of thumb in wireless “Any fancy new modulation technique gets 10x the benefit over a cable.”

          The signal is always clearer so you can push it harder.

          What might happen some time in the future is what as happened with audio bandwidth 20 years ago – there comes a point with video where extra bandwidth adds nothing the human eye can distinguish. 4k might be “it” but it could just as easily be 8k or 16k etc. At some point though, there will be a video standard that is just above the limits of human perception. When that bandwidth number is known for certain, you can reasonably guess that that number plus a bit of fat (say 20-50%) is “more than anyone could use.” THEN, since you’ll have the maximum bandwidth required per-user, it’s reasonably simple to make your wireless cell the right size for all the potential users. THEN, there might come a time when the modulation/interference workarounds/error correction, and resulting experience is so nailed down that wireless networks could entirely stand on their own without a cable to every home. They’ll still use cables back to the main internet since you can always get way more out of a cable with the same modulation technique, but the majority of the fixed-line assets will be stranded and fixed line internet will be the almost exclusive domain of big corporate offices once more.

          That day’s a long way off though, but it’s reasonable to expect a more creeping migration than a revolutionary death spiral.

          • macrofishMEMBER

            For the 5g fanbois, cable networks (HFC) was running the same level of QAM in the mid 2000s as 5g will run, current HFC will happily run at 4096.

            You also skipped over the bonkers clocking requirements to operate at high QAM that little handhelds struggle to deliver.

          • The real beauty of the signal transmission is not in the Modulation (QAM) but rather it’s in the FEC.
            High order QAM without FEC creates a system that is forever waiting for a resend when one or more bits get corrupted. Improved FEC is the real game changer.
            Yeah some of us actually know this shi#

    • At the moment? If 5G is on the way 6G can’t be that far behind, and whatever else. It all looks like ‘wired’ is going the way of the telegraph. Who even has a home phone anymore? Not long ago, everyone did.

      • macrofishMEMBER

        No forever.

        Any technological advancement that is delivered in 6g will work x1000 better of a fixed network.

      • There is a physical limit to the bandwidth available in the radio spectrum, as bandwidth goes up so does frequency, as frequency goes up distance drops by the inverse square law and penetration through objects also drops…

        Meanwhile on fibre optics…. current single fibre breakthrough

    • I could drive down the Gore Hill expressway and over the bridge at over 250klmh only slowing down to about 90klmh as I take the turn into the city… Wheels may be screaming a bit mid corner but its completely doable… Sure, it would rely on that whole section of road to have NO other cars on it and me to ignore all speed limits and lights… but it does prove we dont need to upgrade the train line over the bridge…

      Interesting that with a relatively low take up they are saying peak speeds of only 4.3Gbps. FTTP can deliver 100Gps with only the change of the technology on both ends….

      The real issue with the comparison of 5G to the NBN, is your comparing a system constrained by backend congestion issues which are outside the NBN design with a system that has a limited uptake at the moment.

      Anyone tried to make a phone call at the Football when the stadium is packed? 4G is not so flash in that situation… I remember going to a concert at the Sydney Football Stadium, could not call for a Taxi for nearly an hour after it ended, chose to walk to the City and wait.

      • > Anyone tried to make a phone call at the Football when the stadium is packed? 4G is not so flash in that situation… I remember going to a concert at the Sydney Football Stadium, could not call for a Taxi for nearly an hour after it ended, chose to walk to the City and wait.

        That’s exactly what the… ?24ghz? or thereabouts portion of the 5g band is for. I heard it gets stopped by glass, so penetrating a home is out of the question. It’s meant to be deployed like wifi base station in stadiums – one every 20-50m or so. With even 300 people per base station, and relatively high bandwidth per cell that problem should disappear.

        So, yeah, 5g is definitely going to improve some things.

        • My point was about how good the 4G is when congestion is not a factor is great but becomes unusable when congestion goes up.

          you would need over 100 base stations to cover that, and then you get channel and interference issues.. Sure it will be much better in that scenario but its still a Mobile technology being promoted as a replacement for a much more efficient fixed technology.

          • Oh, yeah. Your point is spot on.

            But, for those irritating situations, rolling out a shitload more infrastructure (could easily be 300+ base stations to cover a stadium like the MCG) 5g will actually work better than 4g could have.

          • Oh, yeah. Your point is spot on.

            But, for those irritating situations, rolling out a crapload more infrastructure (could easily be 300+ base stations to cover a stadium like the MCG) 5g will actually work better than 4g could have.

  2. As I understand it, the key ‘appeal’ of 5G is for the so-called Internet of Things – devices with Qualcomm chipsets, sensors & Zigbee transmitters that collect data to “help us better understand your needs.” Companies have invested big-time into it and part of the business model is that device manufacturers share the info with other parties and councils get to save on people hires by using surveillance/smart light bulbs instead. Aside from the fact that not everyone wants their dishwasher automatically connected to the cloud, there are privacy, health and security concerns.
    For example, do manufacturers need to get your consent to a listening device in your fridge? Or for the footage from your Ring door cam to be streamed back to a server where-ever? Health-wise, if you are trying to limit EMFs or are electrosensitive, do you have a right to know that your device goes online/switches on Bluetooth by remote? And as many have noted, there are significant security concerns – even with certain companies banned from the infrastructure, they still built the base structure. Ethernet is faster and more secure. I also like that if it is off, it’s off. Finally, one day landlines – phones with no need for microwave radiation warnings and which would operate in the event of a power outage (VOIP does not) will be re-appreciated as the revolutionary asset they were.

    • operate in the event of a power outage..
      ahh.. the power came down the line. a power outage at the exchange would be a problem.

  3. 5G kills bees and is responsible for COVID-19. Certain compounds in the The coming bill Gates vaccine will vibrate at certain frequency and turn them into lefty pooves like dr Smithy.

    • Is it old-fashioned of me to dislike the concept of beta-testing that involves 12,000 low-orbit satellites? Also, whose jurisdiction is this? Does a particular authority legislate the safety/legality of this is or is it all sweet if you’re a Tech-Bro/Bored Billionaire? I also note that astronomers are talking about the loss of cultural starlore and astral heritage due to the increasing amount of satellites AND physicists are concerned about the ionosphere.
      I just did the Ookla test and got 109 MBPS – Aussie Broadband ethernet – the original Rudd vision of NBN broadband was actually a good one. I wonder if whoever sabotaged it had any corporate interests with the companies now so keen on blanketing everything in emf.

  4. 5G is not a direct threat.

    My combined upload and download was over 1000GB on a 100/40 connection. (Work, Gaming, Streaming 4K). I would blow past any 5G plan in a week. Tesltra’s top 5G plan is 180GB at $115 not including discounts.

    My 100/40 plan is $109 unlimited with Aussiebroadband (who are the most expensive in the market away from the big Telcos) and I get no slowdowns at all during the day or night.

    Aussiebroadband also have up to 1000/50 for $149 unlimited per month, if you are connected via FTTP.

    5G looks impressive because of less users, once people pile on, the speed will diminish.

    If your 50/20 is shit, move to Aussiebroadband.

  5. Bit disingenuous their Leith
    Some readers told you many years ago that 4G/5G would absolutely kill NBN and you didn’t want to hear about it. oh our glorious NBN
    Fiber to the home or bust, was your response, nothing but fiber can ….etc etc
    Yet here we a few years later and we have a thoroughly F’ed NBN wanting to put the brakes on it’s users because ahm well, because they use the system. The whole NBN business structure is wrong from the technology through implementation to the service itself wrong, wrong, wrong.
    I suspect some apologies are in order!
    If you need a list of those that you brow beat with NBN techno-babble then I’ll see what I can find.