Musk's Starlink rockets to 4 million subscribers

Starlink's subscriber count is accelerating and has passed the 4 million milestone, up from the 2.3 million it claimed in its 2023 progress report. According to astronomer Jonathan McDowell, there are currently 6,426 Starlink satellites orbiting the Earth, of which 6,371 are operational. However, considering the rate at which …

[heyrick]

While the company has taken steps to reduce the visible brightness of the satellites

Very long exposure with a mobile phone, so not exactly "serious" astronomy.

There's one. There's another. Oh look over here. And here. Oh FFS...

The way it's going, the world is going to have to book "openings" to fling their own space junk around these things.

[HamsterNet]

Re: While the company has taken steps to reduce the visible brightness of the satellites

How to show you have no idea about scale. its 6000 sats in an orbital plane 6E8 km2, if they where all at exactly the same height, which they are not.

That's 9.2E7 km2 larger than the total surface area of the Earth.

Interference with astronomy is an issue. Solved by getting telescopes off Earth. To which a sensible option is to charge constellation builders with having to provide free rides for large space based telescopes each year.

Another sensible option is to charge each company for orbital slots per year, including the sat and any debris from it. With the proceeds going to building and deploying telescopes in space / Tidying up junk. Would make doing increasingly dumb ass stuff, like blowing up sats with missiles - here's looking at you USA, China, Russia and India hopefully prohibitively expensive.

[DS999]

Re: While the company has taken steps to reduce the visible brightness of the satellites

its 6000 sats in an orbital plane 6E8 km2

IMHO the best way to explain it is to compare it to airplanes flying around the earth. There are probably in excess of 6000 planes (including private/military) in the air around the world at any given time. The "gaps" you have to hit to send something up are comparable to the gaps you'd have to hit to launch a rocket into orbit without looking out for planes.

Obviously in the real world you can't just launch rockets willy nilly for many many very good reasons, but if you did and gave no warning to the planes the odds of hitting one when launching a rocket from a random point on the earth would start with a decimal point and a lot of zeros...

[doublelayer]

Re: While the company has taken steps to reduce the visible brightness of the satellites

The problem with the analogy is that the largest risk isn't hitting something during launch. The greatest risk is hitting something some time during its life. Orbital changes to avoid collisions are already necessary from time to time with the number of things we have in orbit. The need to do that will only increase as tens of thousands more satellites are placed in orbit, and the risk grows that it might be necessary but not actually possible to do so, for example if more satellites suffer failures and no longer respond to commands to change orbit or begin reentry. As it becomes cheaper and more popular to operate large LEO constellations, this risk will increase and our attention to preventing or responding to it should as well.

[DS999]

Re: While the company has taken steps to reduce the visible brightness of the satellites

The airplane example works there too. We could probably get by without air traffic control if the planes that were flying right now at cruising altitude remained in flight for years and only a few were able to execute any significant changes in their flight path or altitude. There aren't any airports in LEO where satellites congregate.

[Baximelter]

Re: While the company has taken steps to reduce the visible brightness of the satellites

Hundreds of large earthbound telescopes provide viewing time for many thousands of astronomers who advance our knowledge of the cosmos. Our few orbiting instruments will never provide that kind of research in depth, not at present and not in the foreseeable future. Starlink is and will continue to be a disaster for astronomy. It is just another aspect of the damage this one man, Musk, has done to the human enterprise. Is his support for Trump a surprise to anyone?

[Anonymous Coward]

Global billboard?

Elon likes to think outside the box and the rules. For now he may be trying to cloak his satellites, but once he has enough coverage he can just turn them into a global billboard - light up the night sky as one huge display. Not all those communication laser are outside the visible range. No one will be able to escape his influence.

No worries, I'll stay safe with my tinfoil hat and night shades on.

[stiine]

Re: Global billboard?

I thought the Russian government was going to paint the moon red, and then president Nixon was going to paint a Coca-Cola logo on it.

Personally, I think all satellites should have collision avoidance lights, just like aeroplanes, or Starlink should make all of them that are over the US blink on July 4th every year.

[alisonken1]

Re: Global billboard?

Nah. That was Harrison in Heinlein's "The Man Who Sold The Moon" - although I don't think Coca-Cola was the logo that was going to be used.

Not sure about the Russian side of things.

[lightfighter]

Re: While the company has taken steps to reduce the visible brightness of the satellites

By “these things”, assume you mean Starlink sats.

“The world” already has a process to seek and gain approval beyond just country of origin for orbital positioning. Starlink is one of many, many, many leo occupants. (Many).

[Anonymous Coward]

How long...?

Before Elon demands mega sized price increases? He needs to recoup some of those billion dollar losses he's making on Twitter.

Asking for a friend....

[Flocke Kroes]

Re: How long...?

Check the news from last June. On top of that Musk has found alternatives to advertising to fund X, for example X gets investment from authoritarian regimes who want to silence dissent and from cryptocurrency scammers.

[Anonymous Coward]

Re: How long...?

Hopefully the satellites' skin is thicker than his.

[mostly average]

Did you know...

There are more hydrogen atoms in a single water molecule than there are stars in our entire solar system?

[Joe 37]

Re: Did you know...

What am I missing here? There are two hydrogen atoms in a water molecule. No more, no less.

[doublelayer]

Re: Did you know...

I'm not sure you're missing anything. There are two hydrogen atoms and only one star, so there are more hydrogen atoms. I think that's all the things to get.

[Ian Johnston]

Re: Did you know...

There are roughly as many molecules in a single average human breath as there are stars in the universe.

[heyrick]

Re: Did you know...

I'm curious as to how such a thing was even calculated.

[doublelayer]

Re: Did you know...

Molecules in a breath can be estimated by measuring a bunch of different people for their breath volume and humidity, then calculate the amount of gasses based on volume and pressure and assume that the non-gasses are mostly water.

As for estimating how many stars there are, I'll defer to the ESA on that one.

[John Brown (no body)]

Re: Did you know...

How deep a breath does one need to take? As we see deeper into the universe, we see even more galaxies than we saw last week :-)

[Anonymous Coward]

I'm not surprised - it works very well

As a customer I'm not surprised they are getting lots of customers. I moved into a house that could get 8mb ADLS or 1mb FTTC (with speed slower than ADSL due to distance from cabinet). With StarLink I get around 200mb down and 20mb ish up and I spend all day on conference calls with no issues.

(annon for the elon haters)

[James Hughes 1]

Re: I'm not surprised - it works very well

I guess the downvotes are for spelling anon wrong?

[elaar]

Re: I'm not surprised - it works very well

You're fortunate, we provide Starlink as a service to our customers and rarely get more than 10Mb up. Our equipment has to use 4G to "boost" the upload.

Very good for download, and latency is pretty impressive all things considered.

[Persona]

Re: I'm not surprised - it works very well

As of December 2023 78% of UK homes have access to full fiber. That means they are unlikely to become to become new Starlink customers as it would cost them almost three times as much per month, and they would have to buy the expensive user terminal too. The fiber roll out is going fast too so I expect the final figure for this year to be much higher than 78%.

From your comment I can guess you are in the UK, so you were one of the unlucky ones when you moved in, but unless you live in a truly remote spot you are very likely you get fiber in a year or two. It's even possible that it's already available for you but like most people in the UK despite it being available you haven't realized it was there or not got around to opting for it yet.

The US is of course a very different market with 20% of households being outside cities in a huge country served by expensive monopolistic network companies who have never had to fight for customers. A large majority of the 4 million Starlink customers are there despite having the most expensive subscriptions in the world and even charging "premium" prices to discourage customers in congested areas.

Most of Europe seems to be further along with broadband than the UK so Starlink is going to be a very niche solution in Europe. You can already see this is France, Italy and Spain where it only costs 40EUR per month and is very undersubscribed. This is good news for you as it suggests that even if you don't get fiber your monthly UK Starlink bill should drop by 50% as they struggle to get customers when fiber reaches more and more.

Curiously the availability of Starlink in remote areas and bandwidth to spare thanks to the lack of customers will make it a very good excuse for incumbent carriers not to roll out fiber to the last rural areas as people will be still be able to get fast broadband there albeit at a premium price.

[chivo243]

Can bots be subscribers?

See title!

[doublelayer]

Re: Can bots be subscribers?

Probably, in the sense that there are probably some industrial users who purchased Starlink connections to connect machinery that's not near a residence and may not be constantly managed. If you mean that the numbers have been exaggerated, it's always possible but I doubt they have been. Given Starlink's multinational coverage, four million doesn't sound like an implausible number to me.

[Jellied Eel]

Re: Can bots be subscribers?

Yep. Utilities have long been satellite users, so connecting things like SCADA for remote pumping stations etc. Cost comparison might get interesting. Traditional VSAT services could often be purchased as terminal kit + space segment capacity, so as long as you weren't planning on doing CCTV, low bandwith. Then could have dozens of terminals all sharing the same space segment. Also popular with regular gas/petrol stations, and in the UK, was used for a lot of Lottery terminals.

Starlink's pricing might not suit that given it's a monthly service fee per terminal, but then the additional capacity would allow for more CCTV. Especially as governments have become more interested in securing critical infrastructure. That gives utilities the challenge of how to do this for thousands of remote, unpersoned sites.

Oh, and then there's Ukraine and all the terminals being used on drones, especially as those are often single-use but there's still a minimum 12 month service contract..

[MachDiamond]

Re: Can bots be subscribers?

"Starlink's pricing might not suit that given it's a monthly service fee per terminal, but then the additional capacity would allow for more CCTV. "

The price isn't so high that it wouldn't be useful for a motorway services location where the shops all need credit card terminals, the majors will be sending and receiving inventory data and VOIP lines. In total, the bandwidth can be less than streaming a feature film and the landlord will be marking up the service as much as they can. The data isn't millisecond critical either so a few second delay won't be an issue.

[doublelayer]

Re: Can bots be subscribers?

Probably true, but right alongside a major road is one of the more likely locations to have a wire going to it. Not guaranteed, but it's not the same kind of situation as some machinery that's well off any major road and away from any network cables. Some of those places don't even have grid power. Chances are that, if there is a cable going to it, some people are going to opt for the cheaper service even if it is an old one and the service is slow. If it's a faster line, that will be pretty much everyone.

[MachDiamond]

Re: Can bots be subscribers?

"some people are going to opt for the cheaper service even if it is an old one and the service is slow."

Up to a point. If a shop has a slow data connection and it means more time to process each transaction, paying for faster service saves money. With so many people paying for things through digital services rather than cash, it just get worse. With EV's, charging stations don't have a cash option, which I see as a big issue. Not only do they need to process a transaction in a couple of stages, I'd not be surprised if even more data is being "shared".

[doublelayer]

Re: Can bots be subscribers?

That is true, but a slow connection has to be really slow to make that happen. Most of the time, electronic payment takes a few API calls, none of which are particularly large. Latency can be more important than bandwidth, but neither is very important. The main way for a connection to be the limiting factor is if that connection is unreliable and is constantly dropping packets. Then you can expect the transactions to begin failing. Unlike many uses, most payment terminals would not have any problem processing payments much faster than the people can. Of course, if you posit a single 128 kbps line shared by fifty shops, it could encounter that problem, but that's probably not the kind of alternative that something along a major road has access to.

[Casca]

Re: Can bots be subscribers?

And you managed to pry in some Ukraine hate even here. Good on you...

[mikus]

Helps when military adoption takes off for drones

Since finding these in use in Russian and only who knows what other types of intercontinental weaponry used globally, business is good for Musk.

[Anonymous Coward]

Re: Helps when military adoption takes off for drones

It's not really surprising that the Russians use them. Musk is a Trump fan and Trump is friends with (and probably owned by) Putin.

[harrys]

Nationalise it and put arthur scargill in charge of it

sorted

[Winkypop]

Any truth to the rumour

That His Muskiness’s satellites only travel to the right?

[Jellied Eel]

Re: Any truth to the rumour

I think they travel clockwise. If they travelled widershins, the fabric of reality might be torn asunder and the Great Old Ones emerge.

(strange the way our ancestors knew that the left was always sinister)

[stiine]

Re: Any truth to the rumour

Nope. Some are in polar orbits. Additionaly, very few satellites* travel in retrograde orbits.

* - not counting debris from collisions.

[Jellied Eel]

Re: Any truth to the rumour

Nope. Some are in polar orbits. Additionaly, very few satellites* travel in retrograde orbits.

But.. but.. that spoils the joke. Think my favorite orbit is the Molniya and I sometimes amuse myself trying to simulate some of the more eccentric orbits in KSP :p

[Persona]

Re: Any truth to the rumour

Be very very worried if countries start launching more retrograde satellites. You traditionally don't due to the small but useful effect from the latitudes rotation helping in the prograde direction. Retrograde satellites are scary as they are travelling a twice orbital velocity relative to prograde ones. Kinetic energy of a collision is proportional to that extremely high relative velocity squared.

[CrazyOldCatMan]

Re: Any truth to the rumour

ancestors knew that the left was always sinister

Sinister being the Latin word for "left" and all..

(And it's more the other way round, sinister cames to mean creepy because it was the word for left, not the other way round.. )

[bazza]

$6billion / 4million subscribers = $1500 per year.

$6billion / 6,426 = $933,706 revenue per satellite per year, or $4.7million per satellite in a 5 year lifetime.

One wonders what the cost of each satellite is...

[katrinab]

Somewhere between $25m and $52m depending on who you ask.

[MachDiamond]

"One wonders what the cost of each satellite is..."

The aspirational cost is $250,000 per. Might be more like 1/2 mil.

The full constellation is 42,000 birds and one also has to consider the ones that have already gone dark. I'd suggest that the cost to launch them is a couple of million at least with the program being in-house. If SpaceX were to spin off Starlink, it would make no sense for SX to provide launch services "at cost".

[bazza]

$0.5million to build, but I reckon that they're spending a few million per satellite to launch. It's a guestimate - say $40million internal cost to launch, with 20 satellites per launch; by that reckoning (which may be wrong) they're spending $2million per launch per satellite, for a total on-orbit cost of $2.5million. If they are indeed getting $4.7million revenue out of it, that's on the right side of the cost estimate.

Whether or not the market will continue to expand is questionable. Terrestrial comms continues to spread, and satcomms continues to be marshalled more and more into mobile applications (airliners, ships). Those can most efficiently be provisioned (still) by geo sats...). Having a massive fleet of LEOs supporting a comparatively small mobile market is a very expensive way of providing mobile sat comms.

[MachDiamond]

"If they are indeed getting $4.7million revenue out of it, that's on the right side of the cost estimate."

From a first approximation, sure. Now start accounting for failed sats, ground station costs, staff, constellation control, etc.

It does help that the satellites aren't an obvious loss on their own.

[John Brown (no body)]

El Reg really ought to look at other then "western" countries space programs

Good article, but taking about other constellations without looking at China might be a mistake.

Amongst other sources for space news, one I usually look at is Dong Fang Hour which is usually very interesting as he goes into some depth on what's happening in China. Marcus House usually mentions other countries launches", NASASpaceFlight covers them in a little more depth, but Dong Fang Hour goes into them in a lot more depth. China just launched it's first 18 sats of it's own StarLink-alike constellation and another is in planning. He does go on to wonder if their own re-usable rockets will be ready in time to get the costs manageable and queries if they even have the capacity to build that many sats yet. But I'd not bet against them doing it, especially with multiple Chinese companies working on reusable 1st stage launchers, two of which are at or past "Star Hopper" stage now. I'd not be surprised if Chinas constellation is larger than OneWebs in a years time.

[Anonymous Coward]

Re: El Reg really ought to look at other then "western" countries space programs

Very important