5G Skeptic (tbray.org) 97
Tim Bray, writing in a blog post: When I was working at AWS, around 2017 we started getting excited pitches from companies who wanted to be part of the 5G build-out, saying that obviously there'd be lots of opportunities for public-cloud providers. But I never walked away convinced. Either I didn't believe the supposed customers really needed what 5G offered, or I didn't believe the opportunity was anywhere near big enough to justify the trillion-dollar build-out investment. Six years later, I still don't. This is a report on a little online survey I ran, looking for actual real-world 5G impact to see if I was wrong.
What is this nonsense? (Score:1)
This isn't news for nerds nor stuff that matters. It's one man's short-sighed view of something he very obviously scarcely comprehends.
The moronic incoherent libertarian contingent here is bad enough, now it's getting into ignorant "5G skepticism". FFS, read a kindergarten-level text about radio frequency.
Re: What is this nonsense? (Score:3)
Re: (Score:1)
Re:What is this nonsense? (Score:4, Insightful)
It's April Fool's Day, however if this isn't a joke
It isn't funny, so I doubt it is a joke.
Many Americans believe the rest of the world has cellular systems as dysfunctional as their own.
Re: What is this nonsense? (Score:3)
I was hoping for ponies again, but we got this instead
Re: (Score:3)
They didn't need to do anything this year. The site itself is a joke.
Not particularly nonsensical (Score:2)
Re: (Score:2)
Seriously, I was expecting an argument about 5G technology not delivering on its promises of bandwidth or ability to work in crowded areas .. but instead we got some BS about how 10 mbits ought to be good for everyone. I expected better from Tim Bray. I guess he must be glad that JSON is replacing bloated XML which needed more than 10mb/s bandwidth.
Re: (Score:2)
The moronic incoherent libertarian contingent here is bad enough, now it's getting into ignorant "5G skepticism". FFS, read a kindergarten-level text about radio frequency.
I just wish someone could explain 5G in a way that makes sense because I fundamentally don't get it.
From what I can piece together current 5G is the same modulation as LTE. The marketing promises of 5G in terms of density and speed are a function of using higher frequencies which trade performance for distance. Unless you are in a high density area with lots of micro towers everywhere you will be stuck with traditional towers /w 5G on low bands and this will be different from 4G how? More MIMO stacking
Re: (Score:2)
While I am also lacking technical details, my anecdotal experience is that it is allowing carriers to add more towers where they want them, in ways that were not viable with "4G". My condo now has a cell site for T-Mobile, along with another one just down the road.
Also anecdotally, I find it leads me to use cell service more than free WiFi when out and about as functionally it provides a better experience.
Otherwise pretty much a nothing-burger.
Re: (Score:2)
Re: (Score:2)
How is it possible to use the same frequencies and the same modulations and get substantially different results?
5G and 4G don't use the same spectrum. 5G uses substantially higher frequencies.
just replacing existing 4G towers with 5G towers would as far as I understand it result in no substantial difference.
If we replaced all the 4G towers with 5G towers, the difference would must assuredly be substantial -- we'd have almost no coverage! 4G towers have about a 10 mile range which is substantially more than the ~1000ft you get from a 5G tower. 5G needs a lot more infrastructure.
I just wish someone could explain 5G in a way that makes sense
The problem there is that 5G doesn't make any sense. The spectrum 5G uses seems completely unsuitable for its intended purpose. It's only good for ve
Re: (Score:2)
5G and 4G don't use the same spectrum. 5G uses substantially higher frequencies.
You seem to think that only 27ghz deployment is 5g, the 3500mhz 5G is also 5g but with significant differences in properties.
It's that there are real physical limits to the amount of information you can move over a chunk of spectrum at once. More bandwidth means you can handle more people. If you don't mind a little math, I have an example that's pretty easy to follow.
If you have directional signals and reuse the same spectrum spatially, that is one way to increase throughput. Which is the approach the lower frequency 5G uses at great computational cost.
Re: (Score:2)
Re: (Score:2)
5G and 4G don't use the same spectrum. 5G uses substantially higher frequencies.
If we replaced all the 4G towers with 5G towers, the difference would must assuredly be substantial -- we'd have almost no coverage! 4G towers have about a 10 mile range which is substantially more than the ~1000ft you get from a 5G tower. 5G needs a lot more infrastructure.
As far as I understand it this is not true. For example n71 is used for 5G and is only in the 600mhz range.
https://en.wikipedia.org/wiki/... [wikipedia.org]
Re:What is this nonsense? (Score:5, Informative)
So what is the deal with 5G in concrete terms? How is it possible to use the same frequencies and the same modulations and get substantially different results?
The answer there is dynamic beam forming, and it's what has led to a fair chunk of the complexity of 5g hardware. This will take some background.
Existing 4g telco panel antennas are effcetively internally arrays of dipoles. usually two or four arrays for the lowest band (700-900) and four or eight arrays for common 4g bands (1800-2600). Many dipoles are stacked vertically for each array, but usually only a single dipole wide.
The reason for them being in pairs is due to polarity, they have +-45 degree slants. Anyway, the way these function is the signal at the input is split on an internal phasing harness that at 0 e-tilt ensures that the signal reaches each dipole at the same time. Ensuring constructive interference along the boresight and getting you your directional beam. Most phase harnesses have provision for slightly altering the timing so the beam is electrically tilted downwards, but changes once set up are uncommon.
Enter 3500mhz 5g. While not significantly higher in frequency it's higher enough that the dipoles are smaller and you can fit more of them in the same area. Instead of being in fixed arrays with a common signal being split among them each has an individual connection to the remote radio unit.. which for not having a bunch of connectors is now usually built into the antenna although there are some exceptions.
5g sends out a reference signal, which through some nice direction finding finds out where the device to be spoken to is both in azimuth and verticle angle.
Instead of a fixed beam, it dynamically changes the phasing (this is computationally very intensive) of some of the array to direct a specific beam to the user. It can at the same time be doing the same thing for another user on the same frequency if the angle is sufficiently different.
When you aren't blasting it everywhere but only fixed targets spectrum reuse can be largely increased. Even in a single tower replacing 4g you get these benefits but the more it's reused and the more targeted the better the benefit.
I have made some simplifications and could write a bunch more on this, but this is probably enough to get the idea.
Re: (Score:2)
The answer there is dynamic beam forming, and it's what has led to a fair chunk of the complexity of 5g hardware. This will take some background.
This isn't exclusive to 5G. LTE has beamforming / MIMO (Up to 8x8 on LTE-A) too.
Is it just a function of higher frequencies = smaller waves = higher spatial resolution? Is it just tweaking parameters like stacking more antennas? Or asked another way what is LTE missing that is exclusive to 5G?
5g sends out a reference signal, which through some nice direction finding finds out where the device to be spoken to is both in azimuth and verticle angle.
Instead of a fixed beam, it dynamically changes the phasing (this is computationally very intensive) of
This does help I did manage to find some technical details on differences in reference signals between 5G and LTE yet this seems to be tweaks rather than substantial game changing differences.
If you simply took exis
Re: (Score:2)
This seems to be based on the premise LTE does not do beamforming.
Of the two thousand or so LTE sites I've seen, two have had dynamic beamforming. In actual deployment (at least in this country) it's pretty much non-existent.
Every single 3.5ghz 5G install I've seen so far has it though. Up to 64x64
Re: (Score:2)
I just wish someone could explain 5G in a way that makes sense because I fundamentally don't get it.
5G is a means of getting users to pay again for something they already have.
Yeah, that's about it. Can't really think of anything more to add.
Re: (Score:2)
Significantly less latency is one advantage with 5G, down to 1 ms. Compared to about 200 ms for 4G. Latency matters for a large number of possible applications.
Re: (Score:2)
Significantly less latency is one advantage with 5G, down to 1 ms. Compared to about 200 ms for 4G.
Your figures are totally incorrect. LTE is 10ms, LTE-A less than half that. To put it into perspective RF on a DOCSIS 3.1 modem adds more latency than current deployed 4G.
Currently getting 58ms real world pings to google.com over LTE-A.
Re:What is this nonsense? (Score:4, Insightful)
The thing they appear to be skeptical of is the dramatic claims of 5G changing the world, rather than just incremental improvement to the status quo.
There were claims of this enabling self driving cars and all sorts of things with about as much hype as your average NFT. None of those promised things have arrived. It's just a slightly faster cell network with relatively limited coverage.
Re: (Score:2)
I'm not convinced it's an improvement at all.
Re: (Score:1)
Re: (Score:2)
5G is just more bandwidth, but it's sold as the golden solution to everything.
So 5G will be the thing for people on the move watching videos and for ad services to pushing even more ads to your phone.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
No carrier has ever had coverage like that in the US. Not even 2G. Especially by land area.
Re: (Score:2)
Re: (Score:2)
You may be very young or very ignorant. Or, of course, both. If you bestirred yourself to find out a little about the author of TFA, you might have come to this: https://en.wikipedia.org/wiki/... [wikipedia.org]
It would be interesting to compare your qualifications and achievements with Tim Bray's. For a start, he has good manners.
WTF (Score:2)
Re: (Score:2)
I'm not going to read the blog ... (Score:2)
Re: (Score:2)
but it sounds like the question, "why would anybody need more than 640k?"
The article is backwards. Given an average data cap, you could now blow through it using 5g in a less than a second at the highest speeds, realistically under 3 minutes. It’s a Lamborghini with a shot glass gas tank. Given 5g lte speeds, this speed is useless and only shortens already very short human scale waiting. It’s hardly faster to most users who will likely notice the lack of even coverage and tendency to get easily blocked by anything. Does cell service cost less with 5g? Nope, if
Re: (Score:2)
It will cost more to add more antennas because it’s coverage limited to line of sight only, but this is overcome by the massive potential profits and revenue streams.
High-band 5G is pretty much limited to line-of-sight, but 5G in general can operate in a wide range of frequency bands, including the 850 MHz band, which penetrates buildings just fine. But because higher speeds necessitate wider bands, telcos won't be able to support nearly as many users at higher speeds in those lower bands.
Re: (Score:2)
High-band 5G is pretty much limited to line-of-sight, but 5G in general can operate in a wide range of frequency bands, including the 850 MHz band, which penetrates buildings just fine. But because higher speeds necessitate wider bands, telcos won't be able to support nearly as many users at higher speeds in those lower bands.
Most LTE uses sub 1Ghz and very few places use bands above 2.5GHz. There won’t be much of an improvement there compared to before, but if you are used to a 700 band and are switch to 2.5 it does lose significant performance in cluttered environments all else equal. What has been marketed to consumers is by far and away the speed, which is only achieved through higher frequencies for obvious reasons. Thus to get what is advertised you suffer from virtually no penetration at all. There really isn
Re: (Score:2)
High-band 5G is pretty much limited to line-of-sight, but 5G in general can operate in a wide range of frequency bands, including the 850 MHz band, which penetrates buildings just fine. But because higher speeds necessitate wider bands, telcos won't be able to support nearly as many users at higher speeds in those lower bands.
Most LTE uses sub 1Ghz and very few places use bands above 2.5GHz. There won’t be much of an improvement there compared to before, but if you are used to a 700 band and are switch to 2.5 it does lose significant performance in cluttered environments all else equal. What has been marketed to consumers is by far and away the speed, which is only achieved through higher frequencies for obvious reasons. Thus to get what is advertised you suffer from virtually no penetration at all. There really isn’t much of an upside at all to customers except perhaps if they suffer reduced performance in very high usage areas.
T-Mobile is deploying 5G as far down as the 600 MHz band. And unlike 3G, LTE and 5G can coexist in each frequency band. So if you're used to a 700 MHz band now, you'll probably have access to 5G service in that band as well, either through dynamic spectrum sharing with LTE or because your carrier dropped 3G service in that band and deployed LTE and 5G in it instead. The main difference is that your phone will be using a more efficient coding scheme that gives you somewhat better throughput than it did wh
Re: (Score:2)
Re: (Score:2)
It will cost more to add more antennas because it’s coverage limited to line of sight only
False. Nothing about 5G or many of the benefits is brings (including some of the ones you mentioned) requires it to use high frequencies, that is completely optional and used to increase speeds in some cases. The point of the standard goes far beyond bandwidth, the point of speed goes far beyond how much you can download in a given time.
Re: (Score:2)
Re: (Score:2)
It won't even pay off for that because the people passing through the area will either already be living in throttling hell because they burned through their month's limited fast networking in the first hour of the month or they hardly ever use it so they won't burn through their limit.
5G is an attempt to increase supply by 3 orders of magnitude while standing firm on price. But they'll be sure to advertise how you could watch mediocre movies while walking down the sidewalk if only you'd sell your house and
Re: (Score:2)
Re: (Score:2)
I'll complain if I want, thank you very much. They're using ever larger amounts of a limited public commons (the spectrum) for this mess.
Funny you mention WiFi. Everywhere the carriers are at all likely to densely deploy 5G, there is already WiFi everywhere. CArrier's way overcharging for unthrottled access means people will continue to flock to the WiFi rather than actually using the 5G capabilities.
Adding to that fun, that WiFi manages with a tiny little sliver of spectrum that was left open for unlicens
Re: (Score:1)
The original impetus to get everyone moved to 5G, came from the surveillance arm of the Chinese government; but they aren't getting what they wanted out of it, because most first-world countries elected NOT to use the (almost certainly backdoored) Chinese-government-subsidized Huawei equipment in their 5G networks. You can get a vague idea how big a deal this is, by reading the huge p
Re: (Score:1)
Not really considering the apocryphal nature of the question.
Perhaps Ken Olsen's quote would be more apt: "There is no reason anyone would want a computer in their home."
Re: (Score:2)
Re: (Score:2)
But the actual question asked in the blog is "Is the new shiny actually more than 640K?".
Home Service (Score:2)
5G has the potential for the cellular companies to be competitive in the ISP business. Of course, it will only be in urban areas, but that's where all the profit in the ISP business is anyway. They're solving the problem of wiring the last mile by making the last mile wireless.
Sure, power users like myself with 100 Gb/s or higher service won't consider changing, but for casual users, if it's good enough to stream TV on multiple devices and game, then it's good enough to compete.
Re:Home Service (Score:4, Insightful)
Try 5G 1 mile from the nearest tower. Let me know how that works out for you. More like solving the "last 0.25 mile problem".
Re: (Score:2)
That was the point. 4G wasn't working very well where there were large numbers of people. Since there is only so much spectrum available and improving the data rate within a given bandwidth gets you only so far, smaller cells was the only solution.
Re: (Score:2)
I am at least three miles or more from the nearest towers. I get 5G most of the time. How? Well, Until a month ago I had very poor service. I live in a mobile home with a metal roof. I am surrounded by pine forest on all sides. To get any service at all I had to put in a cell booster. Even then only my Moto 6 worked. My wife's LG Stylo 6 did not or did my father in law's Jitter Bug phone. We were on Boost Mobile.
One Sunday my brother in law and his wife came over to visit. They were talking about hav
Re: (Score:2)
Re: (Score:2)
Where do you live that the cell companies aren't already heavily in the ISP business? The main players here who offer 5G are also those who offer me cable / fibre.
Re: Home Service (Score:2)
100 Gb/s or was that a typo? (Truly curious and jealous if it wasnâ(TM)t).
Re: (Score:2)
April Fools.
It's actually 300 Mb/s, but I could upgrade to 1 Gb/s up and down.
Re: (Score:3)
Re: (Score:2)
Re: (Score:2)
5G has the potential for the cellular companies to be competitive in the ISP business. Of course, it will only be in urban areas, but that's where all the profit in the ISP business is anyway. They're solving the problem of wiring the last mile by making the last mile wireless.
This is what I don't understand. You're going to have to invest in fiber backhauls and your going to need a tower on basically every street so what is the point in deploying 5G when you could just string fiber?
Sure, power users like myself with 100 Gb/s or higher service won't consider changing, but for casual users, if it's good enough to stream TV on multiple devices and game, then it's good enough to compete.
If you don't the way things are being built out chances are someone else will take the initiative and at that point there is no competing with glass. It might be good enough for today or tomorrow but eventually you won't be able to compete.
same old whining.. same old whining.. (Score:2)
People never need the new thing when it come out.. the need builds up after time
Why do we need 3g?
https://www.theregister.com/20... [theregister.com]
Why LTE?
https://gizmodo.com/do-you-rea... [gizmodo.com]
People like the OP do not seem to understand that new technologies do not magically immediately create new serviced that need the power/speed. Yet some years later there are services that have popped up that make use of the new power/speed and would be really annoying to use on slower.
Thus some years in the future we will have the same w
Re: (Score:2)
Rumor has it the Ruskies need 3g for their encrypted radio communications. I can't believe they'd be that stupid, but then again...a month after they invade Ukraine and they haven't made progress except in the southeast, and they're apparently really pulling back from Kiev/Kyiv. So maybe it's true.
Re: (Score:2)
Why do we need 3g?
I don't know. AMPS was good enough for me. I could make calls from out in the woods (like in an emergency) which have no coverage now.
Re: (Score:2)
and ubiquity of WiFi constraining the value of more Gs
Do you mind explaining how one has anything to do with the other?
Re: (Score:2)
if the spectrum for wifi was given to commercial carriers
Good luck with that. WiFi is implemented in the ISM bands. WiFi has to share it with kids' RC toys, baby monitors and microwave ovens. There is no way for the FCC to 'get back' that spectrum.
trusting third parties with LAN traffic
Probably a telecom wet dream about shutting down WiFi and taking that function over. For a fee, of course. IMO, if the telecoms need to monopolize all communications as a part of their business plan, they need to give up and shine shoes for a living.
Comment removed (Score:4, Interesting)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
A wire
OK, pops. The cool people all have FTTH. ;-)
Re: (Score:2)
Re: (Score:2)
You must have a good near by cell tower. Not over here in my area. :(
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Fiber isn’t that expensive, depending where you live. From 2014 until 2018, I paid the equivalent of 58 bucks a month for a gigabit fiber connection in my Swiss apartment.
Then I moved to Denmark where I got a corporate-paid home internet at blazing speeds of 50Mbps /sigh
WRONG!!! (Score:2)
I have been living with wireless internet access since long before 5G and still do.
Parts of my security system use a cellular data WiFi bridge, and have done for several years. Other parts are still PSTN.
Cut my phone line, and the WiFi parts still work; jam my cellular data and the landline still works.
In addition, the WiFi is capable of streaming Netflix, although I don't usually run that on it.
The claim that 3G/4G cannot to that is either ignorance or propaganda. In either case, it is just plain wrong.
Cool story bruh. I think I'm done with /. (Score:2)
Seriously the stories are border-lining shit you read in the Daily Mail, often reposts, the editing is non-existent shit, half of what is interesting is behind paywalls, the trolls have largely taken over the comments section, and now that they nuked mod points you can't even filter out the good comments from the shit anymore.
The joke was always that I only came here for the comments, but now even that isn't worthwhile.
Seriously does anyone else think that someone is purposefully trying to kill Slashdot? Di
Re: (Score:2)
Yes, slashdot is mostly garbage and trolls and has been for a while. Seems like it took a turn for the worse very recently.
Re: (Score:2)
Many of the trolls seem to have gone silent 3 or 4 weeks ago (and not just on Slashdot). I've been seeing some better comments, too.
Re: (Score:2)
Iron curtain
Re: (Score:2)
I've been seeing some better comments, too.
I'd love to see better comments, but I'm not going to swim through a sea of shit that belongs at 0.
Slashdot's best feature was its moderation and filtering system, and that is gone. May as well go read a newspaper now like it's 1920s where News For Nerds was about Tesla not about Tesla.
The real benefit of 5G i for the Telcos (Score:2)
And not in the cynical "charge more for a service no one needs".
Is a simplification of operations, starting with the Opportunity to finnaly dump 2G and 3G. Is less electricity consumed per site (not only because you decomissioned 2G and 3G, but also, becuase 5G is the most effciient tech in the milliwats/bps department), the 2G and 3G networks were weird, in that there was a circuit switched and a packet switched netwoks, 4G and 5G are packet switched only, easier to operate. All of this reduces OPEX signif
Re: (Score:2)
The beneficts of 5G will be mostly for the telcos.
Which is why the smart users like my power company are not going to let go of their private mesh network for meter reading and SCADA. They don't need low latency or high bandwidth. They are quite used to sending a guy out in a truck to check substation breaker status. What they do need is something that will work through the rain, trees and other crap for miles. Because the telcos are never going to build out a high density network to serve low density customers. They didn't when they switched from 2G and 3
It would be nice if they finished a network (Score:2)
I live well within (only 10 miles or so from the center of) one of the 25 largest metro areas in the nation and neither LTE nor 3G are strong enough at my home to utilize. There is never more than a bar and calls are frequently dropped. If it weren't for my phone's WiFi connection through my cable provider, I'd have to have a landline. And it isn't just my home. I lose calls while driving along major boulevards throughout the area. I see no excuses. The land throughout the region is flat, and there are few
Re: (Score:2)
I live less than 1 mile from a major freeway in California, in a fairly dense neighborhood. I get no signal in my kitchen and 1 bar in the rest of my house. Without a micro/femto cell or wifi calling cell phones are useless in my house.
Re: (Score:2)
Correct me if I'm wrong, but 5G is not intended to replace 4G LTE in all locations. It's intended for high density use, like in a few metropolitan areas. The attenuation of the 5G bands is so bad that it is of practically no use unless there is a clear line of sight to the base station.
I get two bars of coverage on 4G and I can step out my front door and see the base station antennas about half a mile away. Through the trees. 5G? Not going to happen.
Re: (Score:2)
Correct. The first 4G LTE sunsets should be post-2030. What 5G does do, though, is siphon off investment in expanding basic coverage in favor of rolling 5G radios out into high density areas.
As to both of our coverage situations, the antennas used by the stations are highly directional. I also have antennas in sight that operate on the frequency of my phone. Regrettably, the antennas on the tower in sight are pointed lengthwise along a major roadway. I am well out of that tower's cones.
If you'd like to see
Re: (Score:2)
Coincidence? (Score:2)
Re: (Score:2)
Re: (Score:2)
You can't be serious (Score:2)
I pay my mobile provider more for extra data and do all my work via a hotspot on my aging Pixel 4.
Interestingly, the Marina also provides a WiFi signal which is pathetically slow and unreliable compared to the 4G data; the notion that WiFi is the gold standard for wireless Internet is pretty well over.
Why would you say such a stupid thing? He's literally using WiFi and it's faster than the WiFi they provide, so WiFi must be crap?
Re: (Score:2)
Re: (Score:2)
the notion that WiFi is the gold standard for wireless Internet is pretty well over.
This is where I stopped reading the article. That's such a stupid statement, you know that there is no useful information this person can possibly have for you.
The worst thing (Score:2)
Is they shut down 3g effectively reducing the range of devices. 5g implementations do not have the same range. In addition to that, many M2M and IOT devices have to be upgraded or abandoned. All this so the cell industry can give you another reason to buy a device or upgrade your plan
Re: (Score:2)