Why everything you know about 5G is wrong

If you buy what the telcos are telling you, 5G will bring you gigabit per second speeds and sub-10-millisecond latency. It's a pity that you really won't see anything like that in the real world. Here's the real 5G story.

5g hype buzz over sell pitch announce bull horn megaphone
DNY59 / Getty

Maybe everything you've been told by your telcos about 5G isn't a lie, but most of what they've told you about 5G is wrong.

If you buy what the telcos are telling you, 5G will bring you gigabit-per-second-speeds and sub-10-millisecond latency. With speeds like that you could have 8K movies streaming to your smartphone. It's a pity that you won't actually see anything like that in the real world.

3 kinds of 5G

There are several reasons for that. First, 5G is not, never has been, and never will be a single technology. Instead it's an umbrella term. There are actually three different kinds of 5G. Only, the first, millimeter wave (mmWave) might – Might -- give you gigabit speeds.

[ Related: What’s next for 5G: A look at Release 17 ]

What Verizon wants to sell you, mmWave, under the tag line, "This is 5G built right," is only "built" today in a handful of places. It currently runs on 24 and 28 GHz bands and won't be spreading anytime soon. That's because its range isn't much more than Wi-Fi's. Sure, you can cover a stadium with it, but city blocks? Dream on.

Besides mmWave's limited range, it has no penetration to speak of. Your office's walls will block it. Heck even leaves and windows can seriously slow it down. So, sure outside while you're cheering on your team you may see Gigabit speeds -- so long as not too many of your fellow fans are using up the bandwidth -- but in your business? Forget about it. The only way you'll see 5G inside most buildings is the same way you get Wi-Fi in them: By filling them with access points.

Oh, and those new bandwidths mmWave will be getting soon? 37 GHz, 39 GHz and 47 GHz? While they're theoretically faster, they have even less penetration and range.

Or, let's say you're in a car and you're trying to use mmWave 5G. To maintain your 5G connection there will need to be a cell tower with a clear line of sight to your car every 150 meters or so instead of every one or two kilometers. As T-Mobile CTO Neville Ray wrote mmWave 5G  "will never materially scale beyond small pockets of 5G hotspots in dense urban environments." He has an axe to grind, but he's not wrong about that. Even with beamforming directing the signal right to your device, for practical purposes mmWave will only work in cities. On the road, in the suburbs, and the country, we'll never see mmWave.

Mind you, you'll only see the difference if you're watching closely. Thanks to adaptive beam switching your 5G device continuously keeps an eye on its signal quality. When it sees that it's about to lose mmWave -- and that's going to happen a lot -- it will transparently jump to a more reliable connection. But, nifty as that is, that connection is likely to be a good deal slower: Like back to 4G or slower.

So, yes, in ideal conditions, mmWave 5G will live up to its hype. In ordinary circumstances, you'll be using another slower, version of 5G or even 4G, for years to come.

Midband, which can run between 1 GHz  and 6 GHz has more coverage and penetration than mmWave. Sprint, which was the first to roll it out at 2.5 GHz, averages real-world speeds of over 100 megabits per second (Mbps) downloads. By comparison, 4G LTE averages just over 20 Mbps. Its range varies depending on exactly which frequency it uses, but it will be considerably more than mmWave and about half that of 4G LTE.

[ Related: Will 5G accelerate edge computing as a service? ]

So if it's so good, why do you hear so little about it? That's because, while it's already being deployed widely in China and South Korea, in the U.S., a lot of this bandwidth is already used by other networks.

The most significant chunk, the spectrum known as C-Band, between 3.7 GHz and 4.2 GHz, is used for satellite video providers and satellite phone service. They're none too eager to let go of their spectrum. While the lobbyists battle over who gets what of the C-Band, it appears it won't be divvied up until an FCC public auction sometime in 2020. In the meantime, in the States, midband 5G will only be growing slowly.

Low-band 5G, which T-Mobile launched on Dec. 6, 2019, lives on 600 MHz spectrum. If you're an old-school TV watcher of a certain age, that's where your UHF TV channels 38-51 live. They won't be for much longer though. By the end of the summer 2020, they'll be switching to new frequencies to make room for 5G.

This kind of 5G has far greater range than the others or 4G LTE. A single tower can cover hundreds of square miles. Its performance will vary, but it's usually at least as fast as 4G LTE's 20+ Mbps and can reach speeds of up to 250 Mbps.

That sounds great, and if you live in rural America, low-band 5G is going to be a game changer. If you live in many cities, due to how spectrum is divided up between the four (soon to be three with the T-Mobile/Sprint merger) mobile carriers, there's not enough spectrum for T-Mobile's 5G to show to its best advantage. It's usually, from early reports, still faster than 4G LTE, but sometimes it will be slower.

Real world speed bumps

Don't get too excited by all those high-speed numbers. In the real world, there are many reasons why you're not going to see these speeds. Frankly, I'm going to be happy if I see 40 Mbps 5G speeds, which is still 2 1/2-times what I see with 4G LTE. Here's why.

First, networking 101. A connection is only as fast as its slowest link. So, if the internet backbone for 5G is slow, your connection will be slow too no matter how fast 5G can theoretically go.

[ Related: Why is everybody talking about 6G?]

The corollary to this is if you have a lot of people all competing for the same resource, like bandwidth, there's less of it to go around. For now, with relatively few 5G devices, that's not too much of a problem. It will be though because I expect people will buy 5G phones faster than the providers will roll out 5G towers.

That's because you can't simply upgrade 4G towers to 5G. It's hardware and backbone must be upgraded. That's neither cheap nor easy.

And, when you're talking about mmWave 5G with its science-fiction speeds you need a lot more towers. Mobile Chief Technology Officer Neville Ray wrote in a blog post that millimeter-wave spectrum used for 5G "will never materially scale beyond small pockets of 5G hotspots in dense urban environments." That would seem to rule out the possibility of 5G's fastest speeds reaching rural areas or perhaps even suburbs.

As a result, 5G coverage is very spotty. Many users find themselves having a bumpy broadband ride as they move in and out of 5G range. This can be, shall we say, annoying.

Additional 5G annoyances

Surprise, surprise, 5G phones cost more than their 4G little brothers. For instance, as I write this in December 2019, Verizon's Samsung Galaxy S10 5G, costs a jaw-dropping $1,299.99. Meanwhile, it's older brother, the Samsung 4G Galaxy S10 is $899.99. Yes, there are other differences, most notably the 5G model comes with a 256 GB of storage instead of 128 GBs, but $400 worth of differences just for 5G? I don't think so.

Of course, to support the 5G infrastructure buildout the telecoms need more money? And, where will that cash come from? Why, from you of course. Only Verizon has announced a 5G surcharge so far of $10 more per device per month. But the others will charge more as well.  

5G Hype dreams

It doesn't help 5G in the long run that its providers hype it up so much. The worst example of this is AT&T with its 5G Evolution (5Ge). The problem here is that it's not 5G at all, it's just the latest and greatest version of 4G LTE, "LTE Advanced Pro." That's fast. I've seen it reach speeds of up to 40 Mbps, but it's not 5G fast.

And, then, there's all these claims of downloading 2-hour movies in under 4 seconds and sub-10 millisecond latency. These are simply not realistic numbers except for mmWave 5G. Which leads to another question: Just how likely are you to download say, an entire Netflix series in less than a minute while walking downtown? Sometimes faster really isn't better if there's no real use case for it.

The telcos know that, too. That's why you see them promoting 5G not for smartphones but for uses that don't exist yet. These include enabling self-driving cars to be more aware of each other. With this technology, C-V2X, cars could theoretically, negotiate the best way to handle four-way stops without human intervention.

Still others, like Randal Kenworthy, Cognizant, vice president of  corporate strategy, say 5G will transform the Internet of Things (IoT) as much as the jump "from typewriters to computers, because it enables a single-use device (e.g. a pressure sensor in a pipeline) to conduct digitally automated services (e.g. detecting a leak and sending a notification to a regulator, triggering an alert to a third party contractor for repair)."

Maybe 5G will give us smart cars and devices, but these are all years away. We've had similar automotive and IoT technology initiatives using Wi-Fi. They've made a difference, but is 5G all that much of an improvement over previous wireless technologies? I'm not convinced.

Finally, and this is where 5G mmWave might find a real home. As expensive as deploying 5G mmWave towers is, it's still potentially cheaper than dropping fiber to homes and businesses for gigabit internet. AT&T and Verizon are especially interesting in this market. At the same time, T-Mobile is exploring a home wireless ISP service first with 4G, and soon 5G, for less urban areas.

It sounds good, but as the Electronic Frontier Foundation (EFF) has pointed out, even as "AT&T and Verizon have discontinued their fiber-to-the-home efforts ... fiber connections to homes and businesses are, by far, the superior choice for the 21st century. It is not even close."

The telcos have also been, as the FCC has found, less than honest about both their coverage and speeds. For example, "Verizon, U.S. Cellular and T-Mobile likely overstated each provider’s actual coverage and did not reflect on-the-ground performance in many instances. Only 62.3 percent of staff drive tests achieved at least the minimum download speed predicted by the coverage maps — with U.S. Cellular achieving that speed in only 45 percent of such tests, T-Mobile in 63.2 percent of tests and Verizon in 64.3 percent of tests."

Hanging up on the hype

Will you be using 5G soon? I doubt it. By 2022, sure. But it won't be a magical high-speed technology that will transform the world.

In 2025, maybe – maybe -- we'll see 5G augmenting cars and devices. I'll believe it when I see it.

But, the bottom line is 5G will only slowly be deployed and it's not going to be a transformative, revolutionary technology you're being led to believe.

For now, save your money and don't invest in 5G until it becomes the new standard. That will happen, but there's no need to rush into a technology that's still being deployed piecemeal.