back to article A 5G day may come when the courage of cable and DSL fails ... but it is not this day

5G has the potential to make cable and DSL as antiquated and pointless as using a horse and a cart to drive to the supermarket. And it's already here. A few weeks ago, Verizon switched on commercial 5G FWA (fixed wireless access) to the public. Come, ditch the most hated company in America, it whispered. But how serious are …

  1. Baldrickk Silver badge
    Black Helicopters

    Dead birds?

    I hope they are looking out for dead birds around the new 5G sites.

    Or not... https://www.snopes.com/fact-check/5g-cellular-test-birds/

    These tin foil hatters get around, don't they?

    What that snopes article doesn't say is that a couple of years ago there was another sizeable mass bird death - that one was, upon examination of the bodies attributed to toxins from a nearby building site getting into the water the birds drank from (e.g. puddles).

    I wouldn't be surprised if something similar was the cause this time too.

    1. DougS Silver badge

      Re: Dead birds?

      There will always be crazies claiming their ailment or pet cause is affected by a new technology. 5G doesn't use higher power levels than LTE, and the new frequencies that are being opened up are mostly those being used for satellite uplinks for C band and Ka band - which operate at much higher power levels than cellular towers.

      So if "5G" caused bird deaths, there would be giant piles of dead birds around every satellite uplink facility!

      1. Loyal Commenter Silver badge

        Re: Dead birds?

        Simple solution for those who get mysterious ailments 'caused by' mobile masts / wi-fi / bad energy, etc.:

        1) Put up a great big dummy mast somewhere obvious.

        2) Wait for all the loonies to be attracted to it

        3) Meanwhile install the real masts in peace and quiet.

        4) Publicly announce that the mast that all the loonies have been protesting about and camping outside is in fact a dummy mast and that their illnesses are all psychogenic.

        1. DougS Silver badge

          Re: Dead birds?

          I read something a couple years ago about this happening, though not by design. A new cell tower was put up, and one of the big telcos like AT&T or Verizon had trucks and a crane on site for days getting antennas placed. Then the inevitable electrosensitive whining from a couple people in the area claiming all sorts of ailments at a county government meeting that begun the moment the trucks left and the tower was operational. The government people said they'd try to get someone from the telco out at the next meeting.

          The telco sent a representative out to the next one, and who had to hold back a grin when he told them they are still waiting on the electric utility to connect power to the site so it hadn't even been tested yet, let alone become operational. The whiners got laughed out of the room by the rest of the audience, as they deserve.

        2. Mayday Silver badge
          Facepalm

          "Put up a mast"

          This reminds me of some protest that happened in Aussie once.

          A telco was going to install a tower in an area due to coverage reasons, so all these locals protested about how bad it was to have a tower etc at the site. Needless to say, just about all of them took their mobiles out at some stage to check something/text/call etc. Probably will also complain to the telco about lack of coverage too.

        3. JohnFen Silver badge

          Re: Dead birds?

          "4) Publicly announce that the mast that all the loonies have been protesting about and camping outside is in fact a dummy mast and that their illnesses are all psychogenic."

          That would accomplish nothing at all. It won't change any minds or reduce the unwarranted fears even a little.

          However, if you changed step 4 to "Remove the dummy mast with a lot of fanfare and moaning about how much it hurts to do so" would likely work. Everyone nearby would feel better (both physically and emotionally) and you'd still have the real, functioning sites.

  2. ivan5

    Something they are always omitting - cost. If this costs more than a fixed line internet then they can, as far as I am concerned, forget it.

    A couple of months ago while on holiday, I had need of getting on line using a 4G modem. It was limited in the amount I could download and the cost was more than double my fixed line internet.

    It might be of use for the occasional connection via a phone but not for serious SOHO use.

    1. Oddlegs

      Did you actually read the article?

      The other thing 5G-FWA has is cost. It's much cheaper and easier to deploy and that's a saving that can be passed on to a household almost immediately

      1. ivan5

        @oddlegs,

        They might say that but there is absolutely no guarantee that it will be passed on. In fact it will most probably be more expensive because it is 'faster'.

      2. Roland6 Silver badge

        Re: Did you actually read the article?

        Cost is not the same as price to the consumer. Something Ovum obviously missed as they compared the £240-a-year line rental (price to the householder) with the cost to deploy (cost to the telco).

        Likewise, whilst the 'saving' could be passed on to consumers, we know that this is unlikely to happen in any meaningful way. Particularly given the cost of entry into the 5G market gives little opportunity for new players.

      3. JohnFen Silver badge

        But that statement gives no indication that it will be cheaper for the end user. Anyone who thinks (in the US, anyway) that any of the telecoms would actually reduce their prices because they have reduced costs are people who have never actually dealt with telecoms.

    2. JetSetJim Silver badge

      > If this costs more than a fixed line internet then they can, as far as I am concerned, forget it.

      Given that the article mentions:

      For example, there's no need for £240-a-year line rental. Overall Ovum estimates it's almost 50 per cent cheaper to deploy.

      One would expect it to be roughly on a par with current fixed line costs (once everything settles down, at least - early adopters liable to be gouged). For example, swapping your current land line for a £20/mo LTE contract can currently get you a usage of 20GB/mo with EE, or even up to 100GB with Three. So get a SIM only contract and a SIM enabled router, preferably one that will take an external antenna. If you live in a decently covered area (YMMV), then you can already lose the BT copper connection.

      1. DJ Smiley

        Any success actually getting 20Gb down in a month, or even hahahaha 100Gb?

        1. JetSetJim Silver badge

          > Any success actually getting 20Gb down in a month, or even hahahaha 100Gb?

          I've not tried, more than happy with a 1GB allowance considering most of the time I'm in range of 100mbps fibre-powered wifi.

          I can understand how folks in little coverage can get frustrated, but I guess I'm fortunate in that the times I have clicked on a YouTube video while out and about I've not had any trouble. But for sure, fixing not-spots should be a priority

      2. eldakka Silver badge

        Given that the article mentions:

        For example, there's no need for £240-a-year line rental. Overall Ovum estimates it's almost 50 per cent cheaper to deploy.

        The article only mentions installation costs and line rental costs and deployment costs for the telco.

        It does not mention data allowance costs or in fact whether there are tiered speed costs.

        For example, swapping your current land line for a £20/mo LTE contract can currently get you a usage of 20GB/mo with EE, or even up to 100GB with Three.

        20GB/month? For £20/mo? Seriously? That's better than or equivalent to landline? Even 100GB/mo?

        Rubbish.

        For my landline I pay about £60/month and get around 500GB/month. Note that is how much I use per month, the plan is actually unlimited, but I thought giving actual usage numbers versus hypothetical was more useful.

        Unless they can match or better £60/month for 500GB/mo today, not in 2 years time when it'd probably have to be 750GB/mo or more likely 1TB/mo, then not interested.

        1. JetSetJim Silver badge

          > 20GB/month? For £20/mo? Seriously? That's better than or equivalent to landline? Even 100GB/mo?

          > Rubbish.

          Go check the price plans - Three has a SIM only deal for £27 with unlimited data, which includes tethering (aka personal hotspot)

          I'm not shilling for Three, I'm with EE at the mo - but with those prices one has to start thinking about switching

          1. Roland6 Silver badge

            >Go check the price plans - Three has a SIM only deal for £27 with unlimited data, which includes tethering (aka personal hotspot)

            Looks like everyone missed last month's 'announcement' from Three:

            Three UK Quietly Removes Tethering Caps from 4G Mobile Plans

            However, what does need to be clarified is whether Three have also lifted their restriction on using a phone SIM in a mobile broadband only devices.

            1. JetSetJim Silver badge

              > whether Three have also lifted their restriction on using a phone SIM in a mobile broadband only devices.

              err, if you have a SIM only contract, I'm sure you can sue them into oblivion(*) if they don't let you put it into a device of your choosing. I admit they can tell by querying the IMEI returned via the authentication and encryption routines, but you do at least still have the option of setting your phone to be the "broadband device" by turning on the mobile hotspot function (admitedly probably with limited range in comparison with a proper router). But equally, from your link:

              > Since we opened our investigation in March 2018, Three has confirmed that it has already:

              > ...

              > withdrawn restrictions on the use of handset SIMs in dongles and mifis

              (*) YMMV, IANAL

              1. Roland6 Silver badge
                Pint

                > withdrawn restrictions on the use of handset SIMs in dongles and mifis

                Missed that - thanks.

                I like the fact that on the Three website they are still clearly labelling SIMs as being for Phones or Mobile Broadband, having used Three for many years now, I naturally assumed they still restricted the use of phone SIMs in mobile broadband devices.

                However, I suspect Three still use different APN's for their phone and broadband networks - they used to apply different firewall/content filtering policies and connection/traffic profiles. So you may need to change the default phone APN to the broadband APN get the full broadband service.

                Suspect Three have brought online more capacity and now wish to grow subscriber numbers again.

    3. Lee D Silver badge

      Three SIM

      Pay Monthly "contract" (i.e. month to month but there are 12/18/24 month options available to make it cheaper).

      4G.

      30Gb / month, not counting TVPlayer and Netflix traffic.

      30Mbps or thereabouts in peak times.

      Vodafone do similar - 40Gb, available for £35 a month with a pass to let all kinds of junk that don't count towards your data limit (YouTube, Facebook, WhatsApp, Netflix, etc.).

      EE do up to 100's of GB but they get expensive.

      If you wanted to do it now, today, then it's a £50 Wifi 4G router (tiny little box), a £22 a month contract or £30 if you don't want to be tied in.

      Just an aside - I have had that for at least a year as my only Internet connection because BT etc. generally charge at least £19 a month line rental on top of the broadband prices, so it's cheaper to 4G, and I have no choice for other providers in my area.

      5G being even cheaper - more than possible.

      Try a quick google search for "Mobile Broadband SIM" before you start throwing around assertions.

    4. D@v3

      Cost

      While the article may say

      "It's much cheaper and easier to deploy and that's a saving that can be passed on to a household almost immediately. For example, there's no need for £240-a-year line rental"

      I personally just read that as,

      "it costs us less to deploy, but because it is 'faster' we can charge even more for it, so double profit for us. yay."

      1. yoganmahew

        Re: Cost

        Unless my sums are wrong, two-forty knicker a year is a score a month. That appears to be less than the cheapest fivegee whiffy at twenty-two a month?

  3. phuzz Silver badge
    Unhappy

    Cityfolk

    "Ovum believes 5G can help achieve a doubling in the broadband speeds many customers see today, for not everyone is in metropolitan nirvana."

    And yet all of the test sites for 5G are cities, and while there's plenty of talk about signal loss inside buildings, there's noticeably nothing about longer ranges, or impediments like hills and forests that are problems outside of cities.

    1. katrinab Silver badge

      Re: Cityfolk

      Basically, you want to live in my house, in a fairly low-density suburb of a large town, with a line-of sight view of the phone mast about 30 meters away. My 4G is the same speed as my ADSL.

      Cable options are not so good though.

  4. Spazturtle Silver badge

    The way we handle mobile networks is outdated and inefficient. More and more of the spectrum is being eaten by mobile networks because each MNO needs to operate on different frequencies, and they each need multiple frequencies. And then you have 5 towers near each other because each MNO has their own tower. There should only be a single MNO and existing MVOs like O2, EE, Three, ect should become MVNOs. Who upgrades and maintains the towers can be bid on like the national grid is.

    1. JetSetJim Silver badge

      I give you MBNL. There's also an agreement between Vodafone and O2/Telefonica, but that looks to be in renegotiation in London due to 5G trials.

      True, it's not your nirvana of "one infrastructure network to rule them all", but it's something.

    2. Anonymous Coward
      Anonymous Coward

      "We're going to be sitting on a huge amount of spare capacity on which the incremental cost is extremely low," says Three CEO David Dyson recently.

      Translation: "5G is going to cost us a ton of money which our existing customers won't cover, so we're desperately hunting for other ways to sell it"

      1. Mage Silver badge

        5G application

        Basically stadiums only. Replaces multiple WiFi. Can use wide spectrum Line of Sight only bands, which are useless for normal mobile.

        Also 5G spec isn't even needed for Fixed Wireless (FWA). That uses Line Of Sight and has been in use for over 12 years at high performance. It needs roof top aerials and Line Of Sight, as does ANY 5G FWA better than existing mobile. FWA is only cost effective for rural Line Of Sight that can't be fibred. It's REALLY expensive to install. Also can't deliver the speed & capacity of 200Mbps Hybrid Fibre cable in urban. It only beats rural DSL, but that is actually cheaper to fibre now than FWA if you want more than low contention 50Mbps! The Rural fibre is 1Gbps & lower contention.

        1. Ian Michael Gumby Silver badge

          @Mage Re: 5G application

          Try living in a densely populated city where during rush hour you have tons of vehicles using wi-fi. When you consider the future of making each of them a 'hot spot' more data.

          So in dense populated areas 5G can help.

          W.R.T rural settings... I'd either need a line of sight to a Cell tower or I'd have to set up a 100 ft. tall mast for line of sight to the local telco (~20-30 miles away) and then put in expensive equipment on both ends. 100mb/s symmetrical links would be nice for an IoT based ag outfit.

    3. katrinab Silver badge

      Or like BT Openreach. I'm not sure that's a good idea.

  5. A Non e-mouse Silver badge

    Points to consider

    * It may be possible to achieve 20,30 or 40Mb/s over 5G. But it that for all users covered by that cell at once, or just one lucky punter?

    * "The average consumer uses 8GB a month". That may be the case on 4G data, but on home broadband (which they're saying 5G can replace) my kids are currently eating 200GB a month on Netflix, Youtube, Snapchat, Facebook, et al. That doesn't include my or my partner's use for work.

    * Indoor coverage is getting more important. The problem is that the carriers don't see that as their problem and expect you to pay for your own indoor solution. For an average building, we're seeing quotes of £250k for setup fees. (This is a proper in build solution, not the mickey mouse femto boxes the carriers fob their consumers off with which barely work in corporate environments).

    In the past, WiFi was used as a fill-in for 3G. I think that's going to flip now: 4G (&5G) will be the fill in for WiFi.

    1. Headley_Grange Silver badge

      Re: Points to consider

      I see 5 bars of 4g when I'm out and about but it can still take minutes to download a page of Google map because I'm sharing the bandwidth with so many other users. Will this 5g be different? Does it restrict the number of users so they can all achieve decent bandwidth, or will I still be lost because the bloke at the next table is watching Netflix?

      1. vtcodger Silver badge

        Re: Points to consider

        Trust me on this. If 5g actually gives you more bandwidth, Google will increase the size of their map pages proportionally because they know that you secretly enjoy sitting around waiting for their pages to load.

        (I think it's something in the drinking water in Silicon Valley ...).

    2. JetSetJim Silver badge
      Boffin

      Re: Points to consider

      > It may be possible to achieve 20,30 or 40Mb/s over 5G. But it that for all users covered by that cell at once, or just one lucky punter?

      For everyone (*). The whole "massive MIMO" thing effectively means that each user has their own dedicated cell. Funky things go on at the lower protocol layers that steer the beams to each user, and this is only really limited by the antenna complexity installed. This is also why the 3.5GHz radios, operating at the same power, will have approximately the same coverage - they are more tightly focussed. Fixed wireless replacement services will also have the advantage in that the "mobile" being targetted by the beam steering is actually a house, and therefore not normally that mobile, so once you have a lock, you don't need to do much more steering (except when leaves grow on trees to interfere, and what not).

      LTE systems (and UMTS, and even GSM), operate in a different manner in that the spectrum is not focussed on a user, but broadcast more widely for everyone to share - hence while an LTE cell can offer 150mbps (for example), the rates experienced will be load dependent.

      (*) Up to a limit, YMMV

      1. vtcodger Silver badge

        Re: Points to consider

        For everyone (*). The whole "massive MIMO" thing effectively means that each user has their own dedicated cell. ... Fixed wireless replacement services will also have the advantage in that the "mobile" being targetted by the beam steering is actually a house, and therefore not normally that mobile

        That'll be nifty if it actually works. And it sounds like a nightmare if it doesn't work so well and users interfere with each other in weird and unpredictable ways. Time will tell I suppose.

    3. EastFinchleyite

      Re: Points to consider

      Don't worry about bandwidth. The MNOs will handle that by making cells smaller and having many more base stations. Like one per house :).

      Also to remember is why this whole idea came up in the first place; the lack of a high speed network out to customers. So who is going to provide the high speed back haul network required to service the myriad of 5G base stations?. In the rush to push Mobile as "Good", Fixed is "Bad" most evangelists forget that most of a mobile network is fixed land based transmission and switching. Only the last few hundred metres is radio.

    4. John Brown (no body) Silver badge

      Re: Points to consider

      "* Indoor coverage is getting more important. The problem is that the carriers don't see that as their problem and expect you to pay for your own indoor solution."

      Not to mention that many new builds using new insulation products often means insulation panals with a layer of tinfoil on them. I'd imagine that would be like living in a Faraday cage.

  6. Aladdin Sane Silver badge

    Don't expect these two to be in the vanguard

    Maybe they should take a long, hard look at Kodak and Polaroid.

  7. Blank Reg

    46.2Mbps fiber?

    What kind of crappy fiber are they using? I can get 1Gbps on cable, and they have just started running fiber to the home in my area so it could still improve further.

    1. JetSetJim Silver badge
      Facepalm

      Re: 46.2Mbps fiber?

      Crappy BT "superfast fibre", which is not FTTP, but "Fibre To The Green Box Somewhere Else And Then Copper For The Last Mile Because We Don't Invest In Infrastructure", or FTTGBSEATCFTLMBWDIII for short (BT call it "FTTC", but I think mine's snappier)

      1. eldakka Silver badge

        Re: 46.2Mbps fiber?

        Crappy BT "superfast fibre", which is not FTTP, but "Fibre To The Green Box Somewhere Else And Then Copper For The Last Mile Because We Don't Invest In Infrastructure", or FTTGBSEATCFTLMBWDIII for short (BT call it "FTTC", but I think mine's snappier)

        <pedant>

        FTTC (Fibre to the Curb) is where the fibre goes past the premises in the street out front, and only the last 10-50 meters is copper.

        A green cabinet in the street and a mile of copper is Fibre to the Node (FTTN), not FTTC.

        </pedant>

        1. JetSetJim Silver badge

          Re: 46.2Mbps fiber? @eldakka

          Sorry eldakka, BT Infinity is a Fibre-to-the-Cabinet network, not Fibre-to-the-Curb. Terminology is not consistent around the world, and in the UK it's meant as Cabinet as that is typically where the fibre is terminated in the BT network.

    2. Lee D Silver badge

      Re: 46.2Mbps fiber?

      Fibre = VDSL in the UK because OFCOM are pathetic and let it slide.

      VDSL = ADSL + some knobs on. Max is about 75Mbps in general.

      VDSL2 gets up to 200-300Mbps.

      1. Crypto Monad

        Re: 46.2Mbps fiber?

        > VDSL2 gets up to 200-300Mbps.

        Actually VDSL2 is what we use in the UK, but because we use profile 17a, the maximum speed is 100M (capped to 80M by OpenReach)

        Some countries use profile 35b, which could do up to 300M in the best case. Unfortunately, OpenReach decided to do G.fast instead.

        G.fast is crippled by skipping over the VDSL2 17a lower frequency bands, to avoid interference. But those are the frequencies which propagate better over longer distances. As a result, beyond about 500m, G.fast is actually *slower* than VDSL2.

        Plus: because there are LLU providers with their own ADSL modems in exchanges, OpenReach run VDSL2 with a reduced power level to avoid ADSL interference. Again that reduces the speeds obtainable on VDSL2.

  8. James 51 Silver badge

    Not one mention of the ping times, I'm bad enough on line but to be killed waiting for packets to load would be infuriating.

    1. JetSetJim Silver badge
      Mushroom

      4G latency is currently averaging at a bit under 50ms. 5G is *aiming* at 1ms (I believe this is a requirement set by the GSMA) - frag away

    2. Lee D Silver badge

      I game over 4G.

      It's fine. 30-50ms usually. Of course it spikes, but then if you game over wireless you're getting bigger spikes all the time and I bet you didn't notice those (seriously... run a constant "ping -t" to your wireless router IP in the background, go and play a game, come back and I bet it's not "1ms" constantly as it would be on a wired connection).

      5G will lower it further but it's already more viable than most people's broadband.

      (P.S. I run game servers, and everyone always used to moan about how low my ping was "because it's in his back room"... er... no... my servers were in France, I was in the UK... I just don't have junky wireless, I QoS my gaming packets, I have a good, well-managed connection, and other users even on my local network blasting the connection into oblivion don't get to stomp over my gaming ping... I was going through an ordinary ISP, to a foreign server, same as everyone else, and getting sub-10/sub-20 pings all the time. So, trust me, I'd notice if 4G pings were atrocious, because that's all I have nowadays.).

  9. Anonymous Coward
    Anonymous Coward

    >> Of the nine commercial trials Ovum lists, Verizon is potentially the most disruptive – in a broadband

    >> market in desperate need of competition.

    Is this referring to the US market or UK market?

  10. folbec

    Question : 100Mb/s but at what latency ?

  11. folbec

    latency?

    Question : 100Mb/s but at what latency ?

  12. -dp-

    could replace leased line options for sure.

    We are in the market for fast business broadband - leased line is too expensive for us as the installation costs are seriously high. This sounds ideal - especially if we can achieve upload speeds of 100mb + - we do not care upload latency.

    1. DJ Smiley

      Re: could replace leased line options for sure.

      I can tell you for sure. You wont.

  13. Oddlegs

    This can't come soon enough. Fast internet is becoming a necessity and much as I'd love to see FTTP everywhere the costs would be exorbitant. Mobile broadband will be the future.

    It's noticeable though that all of the test sites are in large cities which already have good 4G coverage while more rural areas still struggle to get any mobile signal whatsoever. That makes perfect economic sense however steps should be taken now to ensure we don't end up with a two tier system where some lucky punters have a choice between FTTC, cable and 5G while others are lucky to get a 1Mb/s ADSL line. The 5G networks should be forced to sign up to offering coverage to 100% of the UK population. This needn't require putting masts absolutely everywhere. It would be stupid to put a mast in the middle of the Scottish highlands to cover only a single house but the networks should contribute to a fund to allow all properties to receive a decent minimum speed broadband at a reasonable price whatever the mechanism may be.

    1. Roland6 Silver badge

      Re: "Fast internet is becoming a necessity"

      But how fast? and where?

  14. AndrueC Silver badge
    Meh

    5G has the potential to make cable and DSL as antiquated and pointless as using a horse and a cart to drive to the supermarket. And it's already here.

    I'm not convinced. 5G might offer 'as much bandwidth as most people currently need' therefore making a wired connection no longer a necessity but it can't ever beat a wired solution for throughput.

    Radio waves have limited bandwidth and unless you use beam shaping that bandwidth is shared by the entire catchment area. I can't find any information on what bandwidth is available from a single 5G mast but I'd guess it's a few Gb/s at most. Say it's 5Gb/s. Now put that transmitter in a town so that it covers 5,000 properties. That's a paltry 1Mb/s per property. Now to be fair that's not really as bad as it sounds because it's rare for every property to be downloading at the same time. But still, in this day of Netflix et al. I'd question if an average of 1Mb/s per property was really adequate. By contrast an FTTP roll-out in the same area gives a potential of several Gb/s per property. Of course that's backhaul and ISP dependant but then that's true of a mast as well.

    You could increase transmitter density but there are limits on that (planning permits for one but also just avoiding frequency overlapping). And anyway each mast uses a wired connection so eventually you're just approaching one mast per property which is the same as a wired solution.

    So radio will only replace wired solutions if/when end user bandwidth requirements cease to grow and if they do so at a point where several hundred properties sharing a single mast can assure that level of bandwidth per property. I'm not saying it'll never happen but I remain sceptical.

    1. JetSetJim Silver badge

      > Radio waves have limited bandwidth and unless you use beam shaping that bandwidth is shared by the entire catchment area. I can't find any information on what bandwidth is available from a single 5G mast but I'd guess it's a few Gb/s at most. Say it's 5Gb/s. Now put that transmitter in a town so that it covers 5,000 properties. That's a paltry 1Mb/s per property.

      It isn't shared, 5G uses a lot of beam steering - see here for what looks like a prototype, but lots of work is being done on this now.

      Naturally it won't be able to always have the situation where each user gets their own full capacity beam, but it'll be a lot better than sharing the bandwidth.

      1. Lee D Silver badge

        Do you think that every street cabinet has a 5Gb dedicated line back to BT?

        I highly doubt it.

        Contention is an inherent part of Internet provision anyway. Doing it over wireless media doesn't change that.

        The difference is - that street cabinet likely serves several streets (dozens of properties). The 5G likely serves a thousand people using no-data-at-all and a few dozen houses at peak periods. Anywhere where you have more people, you need more bandwidth / coverage / poles anyway.

        1. Anonymous Coward
          Anonymous Coward

          Do you think that every street cabinet has a 5Gb dedicated line back to BT?

          Yes, when necessary.

          A typical FTTC cabinet covers an area of 300-400 properties, and maybe 40-60% take FTTC service.

          If 1G is enough for peak demand then that's what they'll provision, but if it's not enough, OpenReach can easily upgrade it just by swapping out the interfaces at each end. The fibre itself does not need to change, and has effectively unlimited capacity.

          It's in OpenReach's interest to increase the available bandwidth as consumption grows, because that in turn requires ISPs to upgrade the bandwidth of their interconnects (which they pay for)

          1. Roland6 Silver badge

            Do you think that every street cabinet has a 5Gb dedicated line back to BT?

            I think you are looking the wrong way and overlooking the role of Edge technologies - remember the BT FTTC cabinet unlike the old PTO cabinet, isn't just a patch panel.

            5G-FWA effectively aims to put a 5G mast on top of that street cabinet. So if that street cabinet currently supports 300 properties, the majority with FTTC then the 5G mast will need to provide an equivalent amount of bandwidth, potentially that could be 30~40Gbps, just so that 300 houses can watch "The News at 10".

      2. AndrueC Silver badge
        Boffin

        It isn't shared, 5G uses a lot of beam steering - see here for what looks like a prototype, but lots of work is being done on this now.

        Naturally it won't be able to always have the situation where each user gets their own full capacity beam, but it'll be a lot better than sharing the bandwidth.

        Interesting. Do you have any idea how many discrete beams can be created and/or how big an area they cover? As another poster mentioned most modern wired solutions have an element of sharing. Currently all roll-outs are TPON so there is an aggregation node which is shared like a mast is. However aggregation nodes can be upgraded just by lighting/blowing more fibre or upgrading the receivers.

        1. JetSetJim Silver badge

          > Interesting. Do you have any idea how many discrete beams can be created and/or how big an area they cover?

          How much money do you have? Current stuff I'm aware of supports 16. In terms of Fixed Wireless, these would pretty much get nailed in place, and they may well not be just for an individual's use, and so resources can be shared with others in the footprint of the beam. For comparison, an LTE site would typically be sectorised into 3x120degrees. If a 16 beam antenna were used, then this would probably be 48 beams. As for area, make them narrower and they get longer. Make them fatter and they get shorter. They probably can do the same range as current 4G sites with a narrow beam width.

        2. eldakka Silver badge

          Radio waves have limited bandwidth and unless you use beam shaping that bandwidth is shared by the entire catchment area.

          Beamforming is a requirement of 5G, as is Massive-MIMO.

          Do you have any idea how many discrete beams can be created and/or how big an area they cover?

          In a 2017 article, Everything You Need to Know About 5G, from the IEEE (which answers a lot of your questions), they say 100 ports, where each port can be an antenna for either Tx or Rx. However, they are working on full-duplex for 5G, where each port is simultaneously both Tx and Rx (instead of either). Since the article is from 2017, it's likely the port-estimate has gone up.

          1. AndrueC Silver badge
            Thumb Up

            A very interesting read, thank you. It's interesting to note that the article suggests a massive increase in the number of masts needed which seems to be to be a tacit admission that the technology itself doesn't actually address the shared bandwidth problem. Also one wonders how roll-out is going to be cheap if we have to install ten times as many masts. And what feeds each 'mast-on-a-lampost'? Fibre? The cost of getting fibre to a lamp post is probably 80%+ of the cost of getting fibre to the homes the lampost lights so why not just go FTTP and be done with it? A mast per roadside cabinet would avoid that problem of course but it's not entirely clear from that article if that density would be enough.

            Mind you even more 'interesting' are some of the comments on that article. It's clear that some people are still scared about radio waves. Oh and some religious nut job thinks that 5G is the beginning of the End Times.

            1. John Brown (no body) Silver badge
              Joke

              Oh and some religious nut job thinks that 5G is the beginning of the End Times."

              Oh, I think the End Times are while away yet. One of the harbingers will be 666G which is quite a ways after a measly 5G.

      3. Roland6 Silver badge

        >5G uses a lot of beam steering - see here for what looks like a prototype

        Strange, because once you start to consider a typical cell with a few hundred (or even dozen) client devices in it even fully digital (ie. no moving parts) beam steering will begin to fail - as you will be having to re-steer the beam every few microseconds...

        1. JetSetJim Silver badge

          > as you will be having to re-steer the beam every few microseconds...

          Not sure why you'd resteer the beam that covers a house (or a few houses) that frequently? The house ain't going anywhere

          1. Roland6 Silver badge

            >Not sure why you'd resteer the beam that covers a house (or a few houses) that frequently?

            It's a shared medium, thus it sends a packet to you, then it beam steers and sends a packet to someone else and so on before it gets around to sending you another packet.

            1. AndrueC Silver badge
              Meh

              I don't think they are proposing beam shaping down to that level. From the sound of it it's just going to divide the mast coverage area into discrete sections. So instead of the mast talking to a single group of 12,000 properties it deals with a dozen groups of 1000 properties. That still involves sharing bandwidth but isn't that far removed from what TPON, FTTC and cable are doing so it's improving.

    2. fredds

      can't wait

      I would love to have 5G. Currently living on the outskirts of a rural town in Oz. ADSL2 speed has dropped to 5Mb/s due to crappy lines, NBN will never be seen in my street, fixed wireless is 5Km away, but there are a million trees in between, so no signal, and satellite is a joke. BUT, I do have a mobile tower 300m away, and can get 50Mb/s on a good day, with about 35ms ping. Can't wait.

  15. Mage Silver badge
    Coffee/keyboard

    potential to make cable and DSL as antiquated and pointless

    Sorry, but this nonsense on so many levels.

    5G is just Mobile Wireless. Fixed Wireless can deliver up to 16x performance but is MORE expensive than Fibre unless it's Rural 20km Line Of Sight. Fixed Wireless needs rooftop aerials and LOS to get the performance. Mobile, NO MATTER WHAT xG, is limited to roughly 0.8GHz to 2.3GHz frequencies (Physics).

    5G doesn't change the mathematics & Physics of Wireless. EVERY other lamp post would need to be a fibre fed femto base station.

    Only fibre can make DSL and pure cable obsolete. Hybrid Fibre Cable beats Fibre to cabinet + copper pair (VDSL etc) because cable can manage over 1GHz of bandwidth. A Copper pair struggles to do 0.03GHz of spectrum and slows rapidly with distance over 100m and crosstalk. Cable can manage a 1GHz of bandwidth (certainly over 250Mbps per user) at over a 1km, however for new installs / routes Fibre to the premises is better. It can go anywhere with existing water, electric, gas, sewerage, poles or ditches.

    The 5G is an excuse for Infrastructure companies to sell upgrades, Mobile companies to try and replace WiFi Hotspots (fed with Fibre, cable or DSL) with mobile femto cells fed by fibre.

    All decent Mobile bases or Femto cells need fibre backhaul anyway.

    The 5G is more about integration of services, not faster or higher capacity Mobile, which can only be achieved by treble or x10 number of base stations, no Return on Investment to give better performance as it won't bring in more customers or revenue, only increase Capex and running costs.

    1. JetSetJim Silver badge

      Re: potential to make cable and DSL as antiquated and pointless

      > 5G is just Mobile Wireless. Fixed Wireless can deliver up to 16x performance but is MORE expensive than Fibre unless it's Rural 20km Line Of Sight. Fixed Wireless needs rooftop aerials and LOS to get the performance. Mobile, NO MATTER WHAT xG, is limited to roughly 0.8GHz to 2.3GHz frequencies (Physics).

      If the antennas on the base stations were the regular omni/sectorised ones, I might agree with you, however with highly directional beam steering to fixed sites, there is much more antenna gain, leading to an increased range. Also, perhaps people will install rooftop aerials too - seems simpler than running fibre to the premises if there's no existing connection.

      Agreed that LOS helps a lot, but it isn't the only option. Again, with a highly directional steered beam, there is lots of opportunity to exploit beam reflections to maintain a decent signal without LOS. This would be particularly applicable in urban environments with all that nice shiny glass to reflect signals off of.

      Not entirely sure why you're limiting mobiles to 2.3GHz when there are already 2.6GHz deployments in the UK (Vodafone & EE, for example). 3.5GHz isn't that big a leap up, particularly when you're looking at fixed wireless devices, rather than phones. Sure, with a traditional masthead antenna that is not very directional, these would have limited range, and building penetration probably sucks, but that's not what these base stations are deployed for. You might be interested to note that 3GPP defines a few LTE bands in the 5GHz range (bands 46 and 47), even if I can't find anyone using them (probably for obvious reasons with current 4G equipment).

      > The 5G is an excuse for Infrastructure companies to sell upgrades

      Not sure I disagree with you here :) I'm sure there are lots of ways these new providers can screw-up or screw-over the punters, and I'd like to be damned sure I'm going to have a decent connection speed before changing to such a provision, but then I'm already on FTTP

  16. IGnatius T Foobar !

    And the pendulum swings again.

    Television went from antenna/aerial (wireless) to cable (wired) then to satellite (wireless) and then to fiber/fibre (wired) ... looks like data could be going there as well. For subscribers who need absolutely solid Internet all the time, such as telecommuters, that old satellite mast on the roof will now have a 5G transponder on it, wired into the home router.

    (For those of us with fiber to the home, though ... we're already good, thanks)

  17. Anonymous Coward
    Anonymous Coward

    But if you have already got cable why do you need 5G

    If you already have cable in your house, why would you need 5G.

    I have 350Mb/s now and this is already overkill for most applications, only real use is downloading massive GB size files but for streaming/;web browsing this is already too much. And Virgin has capacity in their fibre to allow gigabit speeds to each home

    5G may be useful for areas that are not covered by fibre but doubt these places will see 5G any time soon either and when everyone using it, speeds will be dramatically lower (just like 3G and 4G)

    PS The mobile manufactures are having trouble with heat dissipation on 5G modems at fast speeds so this could also change the thin phones to a bit chunkier due to heat sinks.

    Just more marketing presently to sell new phones, the same as 4G phones.

  18. 89724102172714182892114I7551670349743096734346773478647892349863592355648544996312855148587659264921

    I use a SIM router when Virgin Media Fibre craps out and I must say it's nippy even on 4G, reliable too. Also very cheap at £10 a month for "Unlimited" - 6gb full speed (network sharing allowed), the rest at 256k (no network sharing, which is when the SIM goes into the laptop), which is fine for email and general browsing - from Giffgaff (O2 will pull the plug on their Giffgaff brand if it threatens their offerings too much)

  19. Luiz Abdala
    Stop

    I call bullshit... on the economical level for the end user.

    They said all of those wonders of 4G... and nope.

    They charge 4G through the nose, for a paltry 5GB per MONTH on my mobile connection. If you burn through it, you're out until next month.

    On my cabled DSL I get to watch Netflix, or someone at home does, for several hours on end every day. 8GB is barely a single HD movie, (well maybe two, to be honest). I'm watching in a 1080p TV, so it is bound to use all of the needed bandwidth.

    Another concern is lag, because I game. Indeed, my home wifi is not 1ms-stable, but it is in the mid-20s, so are the gaming services I use, WHILE someone is watching Netflix. with a dsl line to the fiber on the street, 40mpbs on speedtest.net being frequent, and 30mbps being the minimum.

    I highly doubt they will ever release 5G on unlimited load plans like my cabled connections, but instead use the 4G model.

    So, yeah, it may be great, low latency, great bandwidth, but they will charge through the nose anyway to get an antenna somewhere on the top of the building (mine does not allow individual ones) and then again, cable the whole thing through inside the building and then AGAIN using my home 802.11 AC wifi to provide the last mile... not practical.

    Either it comes through the wall with great penetration directly to my stuff, or not at all. Even if it does, what about the rest of the stuff? Everything else uses wifi, would they provide 5G adapters to EVERYTHING? Would I have a 5G receiver in the most innard portion of the house to pipe it on my wifi? Nobody thought this through...?

    TLDR won't work for me.

  20. JohnFen Silver badge

    The more I learn about 5G

    The more I learn about 5G, the more it seems that it really only works for densely populated areas. A question for those who know this stuff: am I right?

    I hope so, because if it's true then it means I can completely ignore all this 5G stuff.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Biting the hand that feeds IT © 1998–2019