Sigh. Here we go again. Samsung’s busy touting its “5G” network breakthrough, promising that we can “download movies in just one second“. Samsung, in other words, wants to sell networking gear, and that’s fine. What it does also risk doing is reigniting the tired old debate about fixed and wireless broadband, and whether we “need” the NBN at all.
Yes, Alan Jones, I’m looking at you. I mean, I don’t actually know if you’ve said anything on the topic as yet, but I’m sure it’s just a matter of time. You’ve got a lot of airtime to fill, and wireless-is-the-future is your favourite bit of tech gibberish to spout.
There’s more than a bit of coverage of Samsung’s announcement of its 5G speed breakthrough. Here’s Gizmodo Australia’s take. Here’s the SMH, via AFP. Here’s the ABC.
Now, lest I be accused of having a go at Samsung per se, that’s just not so. Well done Samsung, I say; if your new gear is as impressive as your hype indicates it is, then that’s a great step forwards for wireless technology generally. Samsung’s not alone in looking at next generation wireless, but that’s a case of the more the merrier in terms of advancing the state of the art.
From a surface examination, it might seem like a slam dunk home run for “5G”. 1Gbps over the air — that’s got to be better than spending billions on cable, right?
Not so. There are both issues relating to the physical differences, and a whole host of questions to answer about “5G” to start with. Before there are too many complaints, I’m using 5G as a lazy shorthand — as Samsung itself appears to have done. There’s no such standard to speak of, but then even 4G has become a mish-mash of differing networks worldwide. I can’t say I’m holding my breath for worldwide unified 5G any time soon.
But time is the first unanswered question in the 5G rollout portfolio. Samsung says it hopes to bring it to the commercial market “by 2020”. That’s a hopeful figure, obviously (but I wouldn’t expect anything else), but it’s worth bearing in mind that Samsung’s commercial market isn’t you and me.
It’s the carrier networks who might buy Samsung’s network gear. Then again, they might buy network gear from any of a number of other providers. That’s part of Samsung’s play right now; by building up the hype around its 5G solution, it might just score a few partners down the track willing to plunk down cold hard cash for its gear. That’d be in 2020 or thereabouts, but there’s no guarantee that it would be an instant installation.
Look at Huawei’s multi-billion-dollar rollout of new networking equipment for Vodafone Australia, for example. That’s been a slow, steady rollout of new equipment all over Australia — and it’s taken several years so far. Suddenly, that 2020 figure is looking more like, say, 2025. 4G is still in its infancy here in Australia, with Telstra having by far the largest group of 4G coverage zones across the country, Optus still only looking at capital cities and Vodafone yet to make its 4G network commercially available. These things take time, but moreover they take commercial realities into consideration as well. There’s a reason why 4G is available where it is, and why it’s still only 3G/2G elsewhere, as telco companies, above all, want to make money from their networks.
Then, there are the issues relating to spectrum, power and usability. We’ve only just rolled up the so-called “digital dividend” auction here in Australia. You can’t simply stick a tower in the air and claim gold rush stakes; there’s all sorts of red tape, safety considerations and interference issues to overcome.
The initial reports on Samsung’s 5G don’t have a whole lot of detail around that, which in one sense is perfectly fair for a product that’s seven years away from commercialisation. Still, for most high frequency networks, you’ve got to pump up the power output to manage decent ranges, which is a challenge all of its own. Samsung’s announcement talks of what seems like a great range for a 1Gbps transmission at 2 kilometres. That is, again, a great technical achievement. But again, the details aren’t clear. How many obstructions were in the way? I’m guessing very few. How many users on the network? I’m guessing one.
That’s the other issue that still plagues mobile networks; you’ve got to have both a good background connection to manage that 1Gpbs (like, say, a fixed line network), and everybody shares the same data pipe. I’ve had plenty of instances across both Telstra and Optus’ 4G networks where, despite getting what appears to be a solid “4G” signal, I’ve struggled to get something better than dial-up speeds. Typically, that’s when I’m in busy city business districts with thousands of other users. 5G may widen the size of the data pipe, but if you’re sharing a 1Gbps pipe with thousands of other users, it’s not quite as impressive… or as reliable.
That’s the other factor that makes a fixed line network — and while I don’t think I’ve been shy in making it clear which of the two alternate NBN proposals I reckon has more legs, this applies to either FTTP or FTTN — a better prospect. The value of the network is significantly greater when there’s ubiquitous connectivity between all parties, and wireless networking will always struggle with that to some extent, whether it’s because you’ve moved into a tunnel or a valley, or simply because the network itself is overloaded.
Now, there are wireless portions of the NBN being proposed by both sides of politics. In fact, it’s essentially the same proposal from both sides, but neither is pitching it (outside of strange appearances, again in front of Alan Jones) as a fixed replacement; it’s there to cover areas of vast distance for a very small fragment of the population where it’s not feasible to run quite that much fibre. Wireless has its place in the NBN infrastructure, and if in the future a 5G solution can bring even better speeds to those areas, that’s a win-win. But as a “replacement” or something that’ll make the NBN redundant? I don’t think so.