SpaceX's Starlink: Overhyped and underpowered to meet broadband needs of Rural America, say analysts SpaceX's Starlink has been described as the solution to dismal rural broadband. Like any project linked to Elon Musk, the satellite internet constellation is surrounded by a thick cloud of hype. But is it justified? Analyst house MoffetNathanson isn't sure. A new report published earlier this week expressed doubts about …
It's unlikely to be "no cap". Don't know how the Musk-idolizers, who are generally no-cap believers, are going to deal with the psychological contradictions of that. One solution would be a return to reading literature instead of watching movies & TV, and playing online low-bandwidth board games like checkers, backgammon, and chess, instead of the more modern multi-player action games. Those who feel that is too constraining can pack up and head out to Mars frontier, where there is likely be less demand for bandwidth.
This isn't as fast as running fibre to every last cabin in the woods
But since we can't seem to run non-lead water pipes to city centers, or transit to anywhere it's unlikely that rural America is going to become Singapore anytime soon
So yes, don't say we don't need any infrastructure spend because they can just use Starlink - but also don't block Starlink because it can't do 4k Netflix to everybody outside downtown SF
Note - downtown SF has terrible broadband, way worse than Starlink for many people. In fact, that is true for most of Slicon Valley as well - at best people have OK cable as their only provider. I would not use anywhere in the SF Bay Area as a benchmark for broadbad.
Most of the decent (aka fibre) infrastructure is in new-ish large subdivisions, most of which are not even in California....
I remember back in my days in the media reading about the situation in the USA. There was and presumably still is huge objection to satellite delivery. The National Association of Broadcasters (NAB) were horrified by the licensing of satellite radio companies XM and Sirius. They're also not keen on Dish and Echostar the satellite TV companies. The NAB represent terrestrial radio & TV companies and these upstarts were very unwelcome. They'd take away listeners/viewers from their members so they had to be brought down to earth (pun intended).
The radio services for example needed terrestrial repeaters to cover areas of no reception such as big cities with skyscrapers etc. These (after a lot of lobbying) had restrictions placed on them regarding local content* etc. Therefore I'm not surprised there's a group not keen on satellite broadband.
*Traffic, weather and so on.
There's been articles about the US gov adding a GNSS service to Starlink satellites. It seems that such an upgrade is trivial, the smart bit comes on the ground.
So doing the thing with Oneweb makes sense too.
It may not be a GPS clone, so a new receiver is likely needed. But then it's a matter of what needs a GPS clone in the first place, and does a new receiver matter? For critical government or infrastructure things, a new or second receiver is likely no big deal.
Like any project linked to Elon Musk, the satellite internet constellation is surrounded by a thick cloud of hype.
Well, here's another of his projects in the news...
"Neuralink video appears to show monkey controlling game paddle simply by thinking"
https://www.theguardian.com/technology/2021/apr/09/elon-musk-neuralink-monkey-video-game
Note: The picture accompanying the article is not a picture of an actual monkey with a Neuralink implant
Your feeling isn't accurate. What actually happens (and it does in those far-off pre-Musk days) is that groups of people (often called a "village") get a single installation and then distribute the connectivity to a school, community center, etc. And the fee for that single link is either subsidized or paid for by government or NGOs, so the actual out-of-pocket cost to a tribesman in, say, the SaHell is zero but the benefit is achieved.
Back in the "first world", my dad, who lives in the boondocks (almost half an hour from Oxford, so you know it's rural) only got actual broadband a couple of years ago; before that he had ISDN until BT cancelled that and then GEO satellite for a few years, which worked well but had (obviously) a bit of a latency problem. Large parts of Scotland, Wales and Northern/Western England have the same issue: there aren't enough people to justify BT OpenHand or the mobile phone operators building out infrastructure.
This post has been deleted by its author
I live in Oregon less than 5 miles from Interstate 5. I'm less than 2 miles from the city limits of the largest city in the County, and the County is the size of the state of Connecticut. We get (sometimes and only recently) 4G cell at my house which is on a ridgeline. Until 2 years ago I was the last house that could get DSL, my neighbor was out of range. We got 500K download, 128K upload speeds for $45/Month. Today, by a stroke of good fortune (being located along the path used by the local electric utility to reach one of their substations) I have fiber connectivity at 50Mbps for $50/Mo., but many of my neighbors (anyone off the main road) can't get it. No cable anywhere close (not that I would pay for cable in any case). The company I work for is even more excited, as our business is wildfire smoke detection, usually from currently unmanned forest fire lookout towers. (Can't get people to go spend time in a tower without their precious internet, and no there is no cell phone coverage, cable, or DSL (or in many cases utility power) out there. Currently we build long-haul microwave networks to link and serve the towers, which is expensive to build and to maintain, and any fault in a chain of towers can result in "blinding" the system over large areas - each tower covers up to 400 square miles. It will take a little work to run Starlink from an offline solar system, but the cost isn't prohibitive in light of reliability and bandwidth availability. I am sure there are many, many other applications that have similar requirements. Home users streaming 4K Netflix isn't the only user base out there. There is a whole lot more of America outside of urban services range than you think.
If we are talking about a user node, middle or nowhere, don't know where you get 100 watts from ?. A receiver and associated computing could be done for 20 watts or less, a single solar panel. How do you think portable sat phones work, same technology, a handful if watts ?...
This post has been deleted by its author
Intrigued as to why this has been downvoted twice - I'm not saying the monitoring kit only needs to be powered for a few minutes a day, but the comms link, particularly if it's relatively power hungry (I've not looked at whether the quoted 100W is typical, average, maximum or whatever).
This post has been deleted by its author
Forgive me if I don't consider reddit to be that much more reliable than wikipedia - to be honest it really doesn't matter when or how often it uses 100W, far too much to be permanently on anyway.
But your own source says that Starlink is quite happy to have power cut...
So you turn it off, and It uses no power.... Then you apply power and probably get the ability to transmit data within two minutes?
That's not too arduous for a system which is only ever reporting on emergencies over a wide area (and maybe a daily check-in)
Of course traditional sat technology could do the same job, since I am assuming that latency doesn't really matter, and the bandwidth required isn't likely to be extreme (a few photos of smoke and a direction indicator?)
This post has been deleted by its author
It might be reported in a lot of places - I've already declared that the actual power draw isn't relevant to the calculations, it's clearly too high to be always on for an application such as we were discussing.
> If used for remote monitoring, or emergency situations, 2 minutes for bootup just might matter.
> Emergency use isn't the primary function, was just an example.
Whilst the two minutes might matter, the likelihood of the fire being within two minutes of destroying the tower when it's watching hundreds of square kilometres of forest is... probably more remote than the tower.
This post has been deleted by its author
My first reply started from
“ The company I work for is even more excited, as our business is wildfire smoke detection, usually from currently unmanned forest fire lookout towers. (Can't get people to go spend time in a tower without their precious internet, and no there is no cell phone coverage, cable, or DSL (or in many cases utility power) out there. Currently we build long-haul microwave networks to link and serve the towers, which is expensive to build and to maintain, and any fault in a chain of towers can result in "blinding" the system over large areas - each tower covers up to 400 square miles. It will take a little work to run Starlink from an offline solar system, but the cost isn't prohibitive in light of reliability and bandwidth availability. I am sure there are many, many other applications that have similar requirements. Home users streaming 4K Netflix isn't the only user base out there.”
Call me sceptical, but why are people so confident that this is going to deliver a reliable service at the speeds people need?
If 30,000 satellites that are essentially disposable are needed to provide the service that is a massive overhead with absolutely no guarantee it can be maintained even assuming permission is granted.
That means that there will be a replenishment rate of 6,000 satellites per year. Currently there is capacity to launch 60 at a time with the supposedly increasing to 400.
That is 15 launches a year at best, or 100 at worst. It will probably end up being somewhere around 40 to 50 but that is a huge commitment.
Then you still need the ground infrastructure to use it. I think an awful lot of people are just assuming that this is going to be like WiFi or a mobile phone.
Is this really giving value for money and value for the resources that are being used to build the satellites. Satellites are generally stuffed full of exotic materials and the one thing we do need to be mindful of is our continued development of "disposable tech". These satellites are designed to be disposable and burn up. You cannot keep doing that indefinitely.
"If 30,000 satellites that are essentially disposable are needed to provide the service that is a massive overhead with absolutely no guarantee it can be maintained even assuming permission is granted."
The real money isn't in providing this service. It's actually a side-show when all is said and done
trans oceanic laser-linking outside the atmosphere at LEO is about 30% lower latency than the shortest submarine fibre cables. Financial houses will pay billions for that kind of advantage - quite literally that kind of money has been spent for 2-3ms advantage on entire submarine cable systems, let alone getting 10-12ms advanbtage
It's Musky, so of course it's overhyped. But I'm extremely skeptical that the average rural household actually needs 2-plus Mbps per second bandwidth just to watch some TV and support some work activities. Won't do 4K video? So what? Maybe rural users have to settle for moderate resolution. Or download the HiRes stuff in the middle of the night and view later. And at least the latency should be tolerable for most users.
You've got a resource limited by availability. Try allocating it intelligently (for a change). I know. I know. Applying intelligence isn't how we do things in this best of all possible worlds. But perhaps if we tried it, we'd find that it doesn't work all that badly.
But where the plot, casting, acting and direction are good HD does enhance the enjoyment of the work put in by the set designers, lighting designers, props and makeup teams etc.
It also makes certain sports much easier to follow when the ball isn't a single pixel in size.
I live in a reverse goldilocks of broadband, my city is too small for high end fiber or wireless options, but too large for rural broadband projects. I'm in a small city where my best option was Centurylink DSL. 12/0.8 speeds for around $130/ month (including landline, taxes, fees). Outside of town they're are areas with rural FTTH. Meanwhile every few years we hear about someone else getting rural broadband funds, which leads to a few years of more fiber running to municipal buildings and prospective cell base stations (still no FTTH or upgraded DSL in town)
I jumped at starlink as soon as I got an invite to the public beta. Now I get anywhere from 50-200 down, typically 20-25 up. It's still beta testing level of service, so there are outages and other hiccups, but I'm getting better service for less money. Plus, the service is stabilizing as more satellites come online.
No, it's not fiber. Yes, it costs $99. It's still easy better than what I had before. Plenty of people can't imagine the lack of quality internet options available in much of the US. I had it relatively good compared to many people.
Musk does hype. But, legacy providers (and prospective satellite Internet competitors) also spread FUD like crazy.
"All new wireless based systems look great before an economic number of users are added."
So ask yourself this question. Why are SpaceX/Starlink launching, if it's not economical/profitable in the long run? They have better access to the numbers than you do, yet you seem really confident it's not economical. So what do you know that they don't, which means they continue to spend huge amounts of money on satellites and launches for apparently no reason?
I think the point he's making is that while there is a small number of users (i.e. now) the service isn't making any money. At some point there will / should be enough users to make a profit, but that will be a very large number more than there are at present, and because wireless is a shared medium, peak speeds may continue to be high, but average speeds per user will drop (contention an' all that).
So it looks good right now, if you need that sort of thing (and the more I read about broadband provision in the 'states, the more I understand why Starlink is a viable option), but when half your neighbours are also connected, suddenly 200Mbps down becomes a sustainable 20Mbps (to pluck a number from thin air) and short of launching more / upgraded satellites, which costs money, meaning more subscribers are required, there's not a lot that can be done about it.
M.
Though of course I should have added that any connectivity is nearly always better than no connectivity, particularly if it is reliable which, power budgets aside (I didn't realise as mentioned above that it needs 100W) Starlink promises to be.
The question in my mind is how the economics will stack up long-term. Apart from swathes of the rural and semi-rural US I imagine that the highest densities of those who could most benefit from this are likely to be in the very poorest parts of the planet. Certainly the numbers of subscribers in heavily-populated Europe will be low.
I also didn't realise there weren't any polar orbits. That rather limits its marine use, as well as use in high northern (Scotland!) and southern climes. Does performance also drop off at the edge of coverage?
M.
We already have contention with cable or other internet services, with charges scaled to speed. So long as the total system bandwidth is adequate, contention should not cause any more problem than that with cable. Waving arms around is not fact. Radio is cheaper than cable for installation and upkeep, so however it's done, 5G, satelite, whatever, it is the long term future. Cable TV was quite popular in uk in the 1950's, right :-)...
Cable TV was quite popular in uk in the 1950's, right :-)...
There is a vast difference between "broadcast" - that is TV & radio - and the way one-to-one netoworks such as the internet work. In the case of TV there is no concept of contention at all.
In some parts of the UK, the popularity of (for example) the Rediffusion cable TV system was simply down to the lack of repeater stations, particularly once UHF launched as UHF required a lot more repeaters than VHF had. Rediffusion used above-ground cables and a local selector switch which was relatively quick and easy to install. If you know what to look for there are still places around here (South Wales valleys) where the remnants of the infrastructure are still visible - the house we moved from some 10 years ago still had the cable tacked around at eaves height, but cut off.
I take what you say about "total system bandwidth", and I admit it's not information I have regarding Starlink, however the claim of "up to 200Mbps" doesn't leave a lot of room for contention before average speeds are down to the level of standard ADSL2. I suppose it depends on whether that 200Mbps is per downlink or whether it is by aggregating several links, and whether each satellite has several downlinks, or whether if just the one it is possible to aggregate links from several in-view satellites. In either case it also depends on how many satellites are in-view at any one time and whether ground stations can choose satellites freely.
I suppose it boils down to a question of whether you and your closest couple of dozen neighbours are all likely to be using the same downlink or not. Perhaps I should do a bit more research.
M.
So it looks good right now, if you need that sort of thing (and the more I read about broadband provision in the 'states, the more I understand why Starlink is a viable option), but when half your neighbours are also connected, suddenly 200Mbps down becomes a sustainable 20Mbps (to pluck a number from thin air) and short of launching more / upgraded satellites, which costs money, meaning more subscribers are required, there's not a lot that can be done about it.
Whilst I'm always loathe to suggest that "the market will sort itself out", in this case it probably will.
The properly rural will get high speeds from relatively uncontended satellites.
Those in small communities or under-served rural areas will move in sufficient quantities to prompt network upgrades (the presence of actual competition - a rare thing in the land of the free). The price-sensitive will go with the cheapest option, regardless of speed. Speed-freaks will pick the most performant regardless(ish) of cost. If the contention gets too bad people will stick with the incumbent.
Even if it isn't giving you 200Mb down, it will either give you better service than the alternative or will provide a sound alternative that keeps the local incumbent (a bit more) honest.
In terms of income, profitability goes without saying, and the US is irrelevant. Beta testing has been in limited territories, but it's a global service (subject to spectrum licensing).
There will be communities in Canada, South America, Australia and Africa who will snap this up. Whilst some countries will not bear US pricing, it will be steady income for them. It will also be in high demand for maritime and aviation - cruise ships offering sensible wifi speeds should be a lucrative market for SpaceX, along with antennae built into airliners (which SpaceX have already done some preliminary testing with some help from the USAF).
Whilst there is very limited polar coverage right now, they've focussed on building their first shell out, to get usable service for customers under those tracks - rather than global coverage where you only get 10minutes per hour. Shells in polar orbits are coming.
I wouldn't be entirely surprised if - five years down the line - they could give their US residential users free service, underwritten entirely by international and commercial users (they won't, because this is paying for Musk's Mars ambitions, but they financially they could.
Meanwhile every few years we hear about someone else getting rural broadband funds, which leads to a few years of more fiber running to municipal buildings and prospective cell base stations (still no FTTH or upgraded DSL in town)
This is why I like muni-nets. So my concern is effective use of subsidies, and the huge slabs of pork being dangled as part of Biden's infrastructure upgrade plans. Turn those municipal buildings into PoPs, lay ducts any time there's a road upgrade, and gradually the town gets fibre.
Then it's the challenge at the service layer, but adding G.984 GPON gives 2/1Gbps support, or G.987 for 10/2Gbps for a standards-based distribution network. OLTs can potentially go in municipal buildings with the ONTs strategically placed around the town. Then potentially include NID provision, and service providers can interconnect at the OLTs & add routers to the NIDs to offer IP services. Or Ethernet.
Ok, so there's some complexities around building and managing that lot, but they're generally simpler at the optical level. Especially if this can be implemented at the county or state level. But that's what subsidies are for, making sure that money gets spent effectively, and not wasting it via overbuild. There'd be challenges, like avoiding overhead bloat, or dealing with competitors who may object to public sector bodies shifting services onto public infrastructure.. But that could also incentivise competitors at the service layer.
It's not really an alien concept given towns run services already, ie water/drains/power, and GPON should just be another service. But the old days of 'Build it and they will come' are long gone for telcos, so it's an opportunity for towns to build it instead.
Sure 5G won't cover all rural areas, but it will cover enough that SpaceX's potential market will be maybe a million people by the time 5G is fully deployed across the country in 3-4 years.
Even if they sold service to every one of those million people, that's only $100 million a month in income. That doesn't sound like nearly enough to support the service in the US, it will need to be subsidized by specialty commercial services (i.e. aircraft, offshore oil rigs, etc.) to be viable.
The satellites don't stop working just because they aren't over the US (or Canada?). In point of fact, anyplace on the planet between roughly 54S and 54N should have continuous coverage. That's roughly all the seriously inhabited parts of the planet except Alaska, Northern Europe and much of Russia. There may be ground station and frequency band allocation complications, but at least in concept, Elon should be able to sell his service to most of humanity. And that's hopefully without impacting service in the US.
though they do stop working in 5 years or less when they fall out of the sky.
if they really are going to have 30000 of them then that means launching about 6000 per year just to replace the ones that fall out of orbit. what an enormous waste of resources.
terrestrial fixed mobile would be a better solution
if they really are going to have 30000 of them then that means launching about 6000 per year just to replace the ones that fall out of orbit. what an enormous waste of resources.
Whilst it is wasteful, inefficient is better than no service.
Also, Ford build 50-80,000 Raptor pickups a month.
Now, leaving aside the fact that most people don't actually need a pickup (anyone who wants to whine about how it's necessary for doing DIY at the weekends needs to explain how their father and grandfather managed with small cars and station wagons!), it's definitely not necessary for people to change their car every 3 years. The volumes on automotive manufacturing are terrifying.
The inconvenient fact is that launching 6000 satellites a year with a life expectancy of 5 years doesn't move the needle on "waste" compared with the American and European churn of high-value goods bought on lease/credit (usually on 3-year terms). The difference is that Starlink will actually make a tangible and positive difference to people's lives - unlike changing a three-year-old car for the barely-indistinguishable new model.
If you think 5G will be deployed to most of rural America in 3-5 years, please pass the spliff. The big telcos have absolutely no intention of serving sparse rural neighborhoods, no matter how much money the feds shower on them. This little drama has been played out many times before, and it always ends with fatter shareholders, more yachts for suits, a slap on the wrist (if that) for the telcos, and rural customers still out in the cold.
"If you think 5G will be deployed to most of rural America in 3-5 years, please pass the spliff."
^-- this. I know plenty of people that barely have cell coverage at their house right now (as in: text them and wait for them to walk to the correct corner of the house where they get 3 bars of signal most of the time and they'll call you back) . None of them are wild off-grid backwoods recluse types either. Some of them would like to be able to work from home. Some have children who could have really used better access for remote schooling during covid. Some are retirees who would like to be able to do yoga class over zoom. None can with their current ISP options, all would be in great shape with starlink.
What do you think is happening to all the 3G spectrum that AT&T, Verizon and T-Mobile are phasing out in 2022? That's all going 5G - using the variant of 5G that allows for 5G/LTE coexistence so it can run either. So all those rural towers that never got upgraded to LTE will have to be upgraded when they drop 3G, and all of them will do 5G. There's also all that 600 MHz spectrum T-Mobile acquired, and has recently been freed up by moving TV channels to lower frequencies. They plan on building the base of their nationwide 5G network on that band, and using higher frequencies for more capacity in non-rural areas.
What's more, FIXED wireless works significantly better than your phone, because the antenna is 1) MUCH larger, 2) aimed properly for the best reception, and 3) can be outdoors and elevated if necessary. If you can get one bar in one room in your house or standing in your yard on a particular side of your house, then you can get a solid connection to fixed wireless cellular broadband. All most people need is 50-100 Mbps at a decently low latency (and 5G is VERY low latency) so even on the fringes of 5G reception areas it will work well thanks to the fixed/aimed antenna.
Will it cover everyone? Of course not, there are people in REALLY rural areas, mostly in the western half of the US, that don't have any cellular reception now and aren't likely to have any in 5 years because the telcos can't justify building a tower and running fiber to areas where only a handful of people live. That's the million household domestic market for Starlink.
And yes I'm aware that other countries have very different broadband situations so Starlink will be of much greater help there. But most people who live in those places can't afford anywhere near $99/month for broadband, either.
I've got 4G and Its £30 a month unlimited. Getting 70Mbps at the mo.I'd imagine there may be a lot of second hand 4g kit on its way abroad as we 5G up. Not that I think the average family in a rural area Musk is aiming at is going to be able to afford £30 a month or the money to run the generator.
How "rural" or poor do you imagine these people are? The US electrified rural areas decades ago, there is almost no one living in an actual house that doesn't have electricity. People actually making due on generators or with solar / batteries are often in temporary homes (i.e. on wheels) or have chosen to remain off grid.
Just about every rural household can afford £30/$50 a month for broadband. It is a matter of whether it is available to them and whether they think it is worth it if it is (i.e. GSO satellite broadband is very expensive for the slow, capped, and high latency service received, and it would cost them more than $50 a month)
"ow "rural" or poor do you imagine these people are? The US electrified rural areas decades ago, there is almost no one living in an actual house that doesn't have electricity. People actually making due on generators or with solar / batteries are often in temporary homes (i.e. on wheels) or have chosen to remain off grid.
Just about every rural household can afford £30/$50 a month for broadband. It is a matter of whether it is available to them and whether they think it is worth it if it is (i.e. GSO satellite broadband is very expensive for the slow, capped, and high latency service received, and it would cost them more than $50 a month)"
Says someone that does NOT live in a rural area.... Sock puppet, maybe?
This post has been deleted by its author
There was a footnote in one of the analyst reports that the peer networking between LEO satellites is significantly faster than fiber over long distances, so much so that it is competitve vs dedicated fiber for high-frequency trading.
Given how much traders spend on fast links, I suspect there is a pretty good profit for Starlink catering to that crowd....
Until a competitor realizes that they can make millions by jamming their signal for a few seconds at just the right time. Sure that's illegal, but when has that ever stopped Wall Streeters when it can make them money, especially since there's no way you could catch someone jamming you for such a brief window. You could do it via drone, helicopter, regular airplane, even a balloon, and the signal you'd have to broadcast would need very little power to blanket the area that included your competitor's Starlink antenna.
The issue with 5g is the limited footprint of the base stations, Actually, that is a problem with all GSM base stations, and the higher the bandwidth the greater the problem. An old 'analog' base station could cover square MILES . A 5g base station's range s measured in square meters...'Rural' 5G is an oxymoron.
NO, that's completely wrong. 5G base stations have the same footprint as LTE, 3G or 2G stations - AT THE SAME FREQUENCY.
The cell sizes measured in meters are for mmwave stuff, which will only be deployed in dense urban areas or stadiums. T-mobile is building a national 5G network in the 600 MHz band, and specifically chose that frequency because it would be available nationwide and could cover wide swathes of rural America.
Sure, you won't get multigigabit connections from that since you can't let a single customer grab several hundred MHz at once like you can with mmwave, but there's zero use case for phones or residential customers to have multi gigabit connections. Maybe someday in the far future if video is supplanted by holography, but in the meantime even 4K requires only 20-25 Mbps so unless you live with 40 people all watching 4K at once a gigabit is more than you'll ever need.
I love how he argues it's inefficient, yet his fiber broadband association won't even provide service AT ALL to a lot of places.
For example: the very same Florida Space Coast that is launching SpaceX rockets can not get fiber.
"Inefficient" is far, far better than "nothing"
We had fiber optic line running down our road for over 15 years and not a single home can connect to it.. phone company offers dsl, 3mb/s. Only preforms at 1mb/s on best days, currently use Hughes net, very limited on data, sometimes might hit 3 mb/s. Usually it around .7 mb/s. Cellular seems best but data amount is the worst. So it seems as if they are all failing. So why bash musk at least he's trying to give options. Maybe it is only a stop gap but who cares. Of all options, the fiber burns me the most because of money spent to install it yet no one gets to use it. And that guy ( bolton) complains about subsidies.
While it seems hopelessly starry-eyed to believe all the hype about Starlink, one has to admit that such pessimism is hardly warranted either. It's essential to remember here that SpaceX has launched only a fraction of their satellites so far, and only the last handful actually have the laser-links for inter-sat communication.These are still 'testing phase' level, not unlike the first attempts to roll telegraph cables across the Atlantic Ocean.
Lots of companies back then tried to make this 'transatlantic cables' idea stick, while others scoffed and just send letters by mail via ship as was common. Lots of transatlantic cable companies went bankrupt and service was incredibly spotty for decades, yet somehow the issues were resolved one by one, and transatlantic fiber-optic cables quite literally form the backbone of modern society, along with communication satellites.
As an engineering challenge, Starlink is fairly straightforward, as we didn't have to invent new technology to make it possible. It's just that nobody had tried to use all of these technologies, like Ka/Ku band transceivers, ion thrusters and laser-links in tens of thousands of sats zipping at only a few hundred km above the Earth's surface before.
Will Starlink work as advertised? Probably. They're realistic about how much bandwidth they can provide per cell (area covered by at least one satellite at any given time), and each new generation of Starlink satellite is likely to increase the bandwidth even more as lessons are learned and applied, and new manufacturing techniques make certain optimisations possible.
At the end of the day Starlink won't be the end-all, be-all of internet access. But providing solid broadband coverage to areas where wired broadband doesn't exist in any acceptable fashion? That seems eminently realistic.
If the article is correct then Musk has to launch 30,000 satellites every 5 years to maintain his constellation and that only provides low-bandwidth service. Notably, this is an on-going cost, not a capital one, so the cost of the service is unlikely to come down. It seems unlikely (to me) that rival technologies will not be able to reach nearly all of Musk's potential customers, with a significantly faster and cheaper service, with a combination of more down-to-earth tech like fibre backbones and 4G/5G radio links. Don't bother telling me that those aren't happening, because if *they* aren't commercially viable then Starlink certainly isn't.
Satellite broadband probably has a role in delivering emergency-level coverage to "just about anywhere". If Musk can deliver *that* for $100 a month then there are probably enough people (globally) to make that work. But as a daily driver internet service for residential customers? No way. No matter where you live.
Every new tech feat done frequently ends up with lower cost and you can trust Musk to drive cost down wherever possible. The project is its own best push to drive cost down. I myself am very curious to see what happens the upcoming years as the project gets established.
It's not.
5 years is the fall time of a failed LEO bird. They have ION drives and on-orbit lifespans considerably longer than that
Rural broadband is not the only market in any case.
Laser linking provides low-latency interconnection of trading hubs (1/2 the link latency of undersea fibre) which would pay for starlink many times over, making rural broadband essentially "pin money"
High speed connectivity of ships at sea and aircraft in flight at slightly less than current rates would also pay for Starlink several times over
It's smoke and mirrors.
Starlink exists to give SpaceX the reason/excuse to develop reusable heavy lift launch capability at breakneck speed. See Boca Chica Texas. Ostensibly this is also to reach and colonise Mars but the primary goal is to hide in plain sight the construction of massive orbital military capability.
It seems unlikely (to me) that rival technologies will not be able to reach nearly all of Musk's potential customers, with a significantly faster and cheaper service, with a combination of more down-to-earth tech like fibre backbones and 4G/5G radio links. Don't bother telling me that those aren't happening, because if *they* aren't commercially viable then Starlink certainly isn't.
While what you say may be technically feasible, from what I've read of the American broadband market the incumbent operators are more interested in milking their existing subscribers and pocketing federal subsidies than in building out any more infrastructure. They're following the American Dream and trying to litigate the competition rather than actually compete.
Also, don't underestimate the difficulty of running physical cable/fibre to a large and geographically sparse population.
You seem confused as to the difference between capital expenditure and operating expenditure. Capex has a depreciation lifetime, that’s what it does, this isn’t particularly short, nor particularly high.
Radio links and fibre optics require maintenance, replacement of line-cards with each new generation of technology also typically 5-10 years. Each of these satellites provides 30 Gbps for 5 years, at a cost of $0.5m, ie $100k per year. To put this in context: that’s not so different to what many *taxi drivers* depreciate their vehicle at.
Fibre costs maybe $10k per mile, and might easily have to trail 100 miles to the nearest trailhead. Costs $1 million to serve a small rural village. A single 5G tower probably gets you 5Gbps nowadays, depending on sectorisation. You reckon you are going to construct a whole 5G tower including its electronics for under $50k? So, six of those, and double or treble it for TCO including maintenance, power, back haul.
Am I the first to ponder upon the fact that Musk will be in complete control of at least a couple of thousands of working kinetic bombardment vehicles?
Nice business/government you have, it would be a sad thing if one of my decaying sattelites will happen to crash right there.
So who paid them for that survey ?. Hype or not, even if only partly successful, the more competition there is in broadband provision the better. Have cable here in Oxford, with 4g auto backup from BT, but have been experimenting with Netgear 4g modems / sim card as well, just to evaluate the tech. As for bandwidth, don't use streaming services and what is there already is more than enough. Good chance of success for a project like that and there are few who have the breadth of vision, resources and drive to take on such a project. Can still remember 56k and lower dial up modems years ago, glacial, but got the job done...
I can't read the paywalled report, but I doubt its conclusions.
Starlink will not be limited to "southern US" for its subscriber pool. Personal example: I am on Starlink from a suburban location on the San Francisco peninsula. I signed on with Starlink because of dissatisfaction with the only other supplier, Comcast. I'm overjoyed with Starlink so far (8 months in).
An interesting discussion of Starlink's economics is here: https://caseyhandmer.wordpress.com/2019/11/02/starlink-is-a-very-big-deal/ I mostly concur with his analysis. Starlink will be an enormous cash cow for Musk's extraterrestrial ambitions.
Starlink was an idea and the telecoms industry said nothing. Starlink was a couple of prototype satellites and the telecom industry said nothing. Starlink started providing rural broadband and applying for grants to provide more and the telecoms industry goes ape-shit. The first flung dung was that it would would be impossible for Starlink to meet the latency requirements when the theory said they could do better than fibre and real world measurements showed they had no problem at all.
If MoffetNathanson want to talk about efficiency then the measure is (Rural customers getting what the subsidy was granted for)/(subsidy). None of this moving the goal posts after the money has been paid or the regular reports of telcos getting caught lying about the figures. Both are evidence that the efficiency has been lousy. Musk statements are often optimistic (especially on delivery dates). Telco statements are far less credible.
Has this report really been published? The link goes to a page demanding username+password but no "create account" link. Looks like a Microsoft TCO report that misses out licensing fees and is only available after $40,000 and an NDA.
I live in a rural area in the desert southwest where even good solid cell phone service is iffy at best. We had internet through a 4g provider and we hit speeds of 10mbps down and 2mbps up for $150/month. Most people bought their basic package that was $50 but the speeds were too slow for Covid work at home solutions. Recently we switched to T-Mobile's 5g internet at home and it's far better than what we had before but there are a lot of issues, dropped 5G signals, lack of towers, poor 4G reception... as I said lots of issues. Reading the Starlink reddit, I signed up for the beta and am currently waiting for them to offer me a sat kit. Even at $100, if I can get a solid 40 mbps, it would be worth it. Oh and there are no other options for the neighborhood I live in. The phone company offers a slow DSL, but we live too far from the switches according to them. Oh, there is fiber in the area but only government buildings and the schools are getting to use it, so there's that.
homing pigeons. Lots of them.
Have a terabyte usb stick on their ankles. Send 2 for each message (redundancy) - as many movies as you can stomach - all in HD
Bandwidth sorted.
Latency is a bit poor - but natural selection can improve that with time.
Environmentally friendly.
having at least 2 means that you increase the bandwith as necessary
Breakfast sorted too, with an occasional pie for lunch.
There are paces in Virginia where a lot of the internet lives. Yet there are other places in the same state that connect to those same sites using 56k, and some have no internet connection. No internet connection, in Virginia, in 2021. Imagine your life today with no internet connection. The dude shoots rockets into space to launch satellites to beam that packet from earth into space and back down to earth to your house from that datacenter that's in the same state in milliseconds, and sadly that's more efficient than running a fucking wire on a pole. For $100 bucks a month, I can actually SMELL the competition sweating in this article, and it's about damn time.
Thanks for the downvote, dude, whoever you are :)
as it happens, there is a real RFC for carrier pigeon transfer - RFC1149
https://en.wikipedia.org/wiki/IP_over_Avian_Carriers
It has been used (in test only) several times.
The birds won over adsl up to 100KM - up to 2010. Except in Zealand - where they tried Kiwis.
This post has been deleted by its author
For anyone with good FTTP, FTTC or 5G coverage, Starlink is very pricey at the moment.
Well... why is that a surprise?
The trick here is extremely low latency for traders, and access to literally anywhere with a view of the sky.
No building out 150 miles of fibre to connect to anything meaning it's "not economically viable" to support this village, or that town.
"Although federal subsidies will make the product more widely accessible, SpaceX has faced opposition on this front, with the CEO of the Fiber Broadband Association, Gary Bolton, arguing that the product is an inefficient use of resources.
"As a matter of fact, a LEO [satellite] falls out of the sky in five years," Bolton told a telecom news site. "There is never a circumstance where satellite (LEO or geostationary) [broadband] should be subsidized with taxpayer dollars.""
Given that the fiber industry has left rural US of A on copper dial-up (literally in some cases...) I don't think they should be given a voice in the matter at all. Just DON'T use their propagandistic quotes in any format until they actually do something honorable. Mr. Musk may be a big-noter, but he backs up his brags with some ... uh, big notes. The fiber industry? If it ain't someone else's notes, it is no notes. And when given taxpayer dosh the CEO gets a raise, for TALKING about it, and NOT DOING SOMETHING.
For all the right pondian folks suggesting 5G that need some perspective for scale, go to your favorite map app and ask it for directions from Miami, FL to Atlanta, GA. Driving directions will be a touch over 600 miles. That's to get from close to the southern end of one state to about 2/3 of the way up the next state. (And not even the largest ones, 22nd and 24th by area.) That's like driving from Plymouth, UK to Inverness. That drive only made it through 2 of the 48 contiguous states.
And there's large portions of the middle of the country that are uninhabited or look like central Scotland in terms of density. 2/3 of the US population lives within 100 miles of the border. The continental US has about 5000 miles of coastline. That gives us about a half million square miles that 2/3 of the population lives in. The other 1/3 are spread out on the other 3.2 million square miles.
Up until about 6 months ago my parents lived in a rural house on a couple of acres in a very built up state with spotty cell phone service and only Hughes or other geosync satellite as an option. They finally got Internet due to the local electric utility deciding to offer service over the fiber they had run along their poles several years ago when building out the former farmland. And that's in an area between two 100K+ population areas 45-50 minutes either direction in a state averaging 160 people per square mile.
The US is big. Anything that relies on X people per mile is going to leave a LOT of the population outside major/minor cities out of luck. And I know I'm working to hopefully buy a few acres away from people before the Y2038 happens and we're all using abacuses again.
This post has been deleted by its author
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
