Starlink Satellite Internet Service Gets 500K Preorders 93
SpaceX has received more than 500,000 preorders for its Starlink satellite internet service and anticipates no technical problems meeting the demand, founder Elon Musk said on Tuesday. Reuters reports: "Only limitation is high density of users in urban areas," Musk tweeted, responding to a post from a CNBC reporter that said the $99 deposits SpaceX took for the service were fully refundable and did not guarantee service. SpaceX has not set a date for Starlink's service launch, but commercial service would not likely be offered in 2020 as it had previously planned. The company plans to eventually deploy 12,000 satellites in total and has said the Starlink constellation will cost it roughly $10 billion.
Cost and revenue (Score:3)
Re:Cost and revenue (Score:5, Interesting)
The 2020 article you link also ignores that in the Starlink (2016!!) application to the FCC Starlink states:
"Each satellite in the SpaceX System provides aggregate downlink capacity to users ranging from 17 to 23 Gbps, depending on the gain of the user terminal involved. Assuming an average of 20 Gbps, the 1600 satellites in the Initial Deployment would have a total aggregate capacity of 32 Tbps. SpaceX will periodically improve the satellites over the course of the multi-year deployment of the system, which may further increase capacity."
So they expect this downlink capacity to increase. It probably already has increased a lot. An example: Earlier this year they began launching satellites which features laser based satellite to satellite communications which means that satellites that are outside the range of a ground station can still transmit data. This is of course important over sea (when Starlink goes into that market) but it will also increase the available bandwidth per satellite over land. My guess is that they will be able to handle significantly more than 1.5M US subscribers when fully operational. But I guess they will keep the cards close to the vest regarding that as long as possible.
Re: (Score:2)
Would be nice to see some stats on how they are working now. Many people reporting 50-100Mbps downlink speeds, but I'd like to see some detail on the total bandwidth.
20Gbps is theoretical, you have to add in coding overhead and time slice management. Also it's 20Gbps total including upload. Will also be interesting to see what happens to latency as the number of users increases.
I don't see how a laser link between satellites will increase the bandwidth per satellite. That seems to be a limit of the customer
Re: (Score:2)
I could imagine the laser communication allowing more customer density.
Without the laser you have to have all of the data go into and then out of the satellite through the ground communication, but with satellite to satellite communication, a satellite over a denser area could dedicate all of its to the ground bandwidth communicating with customers and then flip to another satellite to get the customers to the internet.
I saw a screenshot today of someone getting 150/36 and 18ms ping using theirs. I pay $60,
Re: Cost and revenue (Score:2)
The question is if the uplink antenna on the sats can actually be used for transmission to customers?
Re: (Score:2)
Even if it can't.
If there's any overlap between the two (base station communication and customer communication) even in the design stage you could design for saturated customers to be more than saturated base station. I would think that maybe laser communication to another satellite would perhaps use less power for example.
At the extreme you could maybe even have sats with customer facing and laser communications exclusively that shoot over to others that are in less dense areas.
I'm skeptical of the overall
Re: (Score:3)
Re: (Score:1)
Re: (Score:2)
its $99 today, but could be tiered in future, and go up in price.
also the US DOD could buy global access as a premium user for $1b/year too.
Add in Africa, South America, Asia remote areas. India/Middle East.
Re: (Score:1)
Re: (Score:1)
How are people in Africa who earn less than $99 a year going to afford $99 a month internet, and you just siad its going to increase in price.
Re: Cost and revenue (Score:2)
Re: (Score:1)
Re: (Score:2)
Re: (Score:1)
Re: Cost and revenue (Score:2)
Re: (Score:2)
Re: Cost and revenue (Score:1)
Also ignores the fact that bandwidth demands roughly double every 3-4 years. So given you canâ(TM)t send out a tech to upgrade those things, launching more satellites is the only option. Also, they probably want more than just 1.5M subscribers, so itâ(TM)s a satellite per 10,000 subscribers. A city like NY would need the entire planned constellation for itself.
Satellite can be useful for some situations but itâ(TM)s never going to be feasible as your daily ISP.
Re: Cost and revenue (Score:2)
Re: (Score:1)
Which HughesNet and Iridium is already covering and neither of them aren't really financial successes.
Re: (Score:2)
Hughesnet: 25Mbps down/3 up, ping times typ 600-800ms, data-limited plans starting at 60$ for 10GB/mo.
Iridium: satellite capacity of 1.4 Mbit max, but this capability is net developed yet. Actual is about 700 kbit IIRC. Couldn't find prices. Iridium appears to be mostly call-focused. The handsets can be used for data it seems.
Re: (Score:1)
Yes, and there is a reason, which SpaceX also can't change the laws of physics. SpaceX' satellites are 1/10th as capable (25Gbps) as HughesNet flagship (200Gbps). Sure they will send more of them and they're slightly smaller, but that still increases the overall cost.
My point was that what SpaceX is currently technically infeasible without sending their entire currently planned constellations over every area they intend to serve. They want to serve hundreds of millions more than HughesNet with more, but les
Re: Cost and revenue (Score:2)
Re: (Score:1)
This was a recent headline:
SpaceX now plans for 5 million Starlink customers in US, up from 1 million. Note that their currently planned constellation is for global coverage.
What space law? (Score:3)
So what happens if Starlink starts selling the information of their customers? If the customer is in Europe is it then covered by the GDPR or are we looking at a relative free for all where Starlink can basically do what they want because it is outside the scope of current laws in the same way the Internet was or worse?
Re: (Score:3)
Re: (Score:2)
Starlink has ground stations in every country they operate, and thus must follow the same rules and regulations as any local ISP and radio operator. HughesNet [wikipedia.org] is an older example.
That is probably true. But from next year all Starlink satellites will also have satellite to satellite communication lasers. This means that they do not need as many ground stations in the future - probably skipping having ground stations in some smaller countries.
Re: (Score:1)
hot stock tip ITT (Score:3)
commercial service would not likely be offered in 2020 as it had previously planned
Going out on a limb with that prediction, aren'tcha?
Re:hot stock tip ITT (Score:5, Funny)
commercial service would not likely be offered in 2020 as it had previously planned
Going out on a limb with that prediction, aren'tcha?
Well hindsight is 2020 after all.
Re: (Score:2)
Almost 17 years to pay off (Score:3)
The planned constellation has an estimated cost of $10B. Let's ignore for a moment other fixed or operational costs. With 500K subscribers, each subscriber should pay $20K just to pay off the initial investment. With a $99 monthly payment (let's say $100) each subscriber should stay with the service for 202 months, or 16.8 years, to avoid losing money.
Since this is ignoring OPEX and other fixed costs (base stations), they will need to get many more subscribers to make it profitable on its own.
On the other hand, it is very likely they will meet the requirements for Government subsidies for providing Internet access in rural areas, not only in the US but in many, many countries when they expand the service. This might make it work.
Re: (Score:3)
Then again, there may be a fair number of "cool, I want it" preorders which will not materialize, so estimating the number of customers based on the current numbers is just a guessing game.
Re: (Score:3)
Preorders is not the same as customers. It is fair to assume that a lot of customers are waiting for the service to become operational so that they know what to expect before ordering it. After all, it is a $499 investment to even get started, so a lot of people will not commit before knowing what they will actually get.
Then again, there may be a fair number of "cool, I want it" preorders which will not materialize, so estimating the number of customers based on the current numbers is just a guessing game.
You do pay $99 to preorder (as far as I can see) so it does imply a certain willingness to subscribe to the service once operational. But otherwise I agree.
Re: (Score:1)
As always with Musk, those 99 bucks are refundable at any time.
Re: (Score:2)
Re: Almost 17 years to pay off (Score:1)
With AOL youâ(TM)re not just buying a name, you actually get a ton of infrastructure, real estate, employees, a captive customer base, historical data etc.
In this case however, youâ(TM)re literally building a product from scratch, a product that will never be viable as promised (which with Musk we are used to now, see Tesla, Hyperloop, Boring, ...)
Re: (Score:1)
Historical data about what people on dialup ? Who cares ?
WHo in their right mind is going to pay thousands for customers that pay $10 a month ?
Re: (Score:1)
If you're really thinking AOL is just dial-up service, you're grossly mistaken. They own ~50 other major websites including Netscape, a number of mobile ad brands, a number of social networks, a number of other ad-driven services. They are worth $10B because they're not just a dial-up service.
Re: (Score:1)
Re: (Score:1)
If you go on any social media website or YouTube, a chunk of the ads there will be served by an AOL subsidiary. AdTech, Verizon Media, LightningCast etc are all owned by AOL.
They're basically a large diversified holding company.
Re: (Score:1)
Re: (Score:1)
AdTech had close to $1B in profit alone last year.
Re: (Score:1)
Re: (Score:2)
I doubt the 500k number means very much in relation to the $10B number.
I believe the $10B is an estimate for the full deployment of around 42k satellites.
I doubt you'll have only 12 users per satellite, I think that's like two orders of magnitude off.
So they'll need much fewer satellites to serve the current preorders.
Also, I think the $10B number was pretty conservative since they were in early days of rocket reuse at that point.
They currently have rockets that have flown 8-9 times, and they're saying they
Re: (Score:2)
> This might make it work.
I don't know the numbers, but as long as it doesn't bankrupt them, it's working. They're pushing the technology forward, proving a communication technology that can be used in Mars.
Re: (Score:3)
Oh come on. Haven't you people realized what is going on here? Musk isn't launching internet enabled satellites. That is just what he "says" he is doing. Each satellite is actually an orbital death ray for when he plans to take over the planet and declare himself galactic overlord!.
Why else does he want to be King of Mars?
Re: (Score:2)
$100 could be optimistic if competition arises. I know OneWeb is a joke but I'd expect other companies, and particularly other states to start offering similar services. You can imagine low cost or free satellite internet access being a new form of soft power and influence, like long wave radio.
Re: (Score:2)
After all, "Radio Free Europe" has restarted transmissions for Romania, 20 years after the fall of Communism.
But, while you could receive "Radio Free Europe" with most radios with a long antenna (which were plenty and innocuous looking indeed), for fast satellite internet you need an expensive and conspicuous antenna (the already existing Iridium fits into what are recognizable as mobile phones - even though they too are specialized).
Re:Almost 17 years to pay off (Score:4, Interesting)
500k pre-orders are just that, pre-orders. And those are only from countries that are allowing Starlink access.
More to the point, they have not yet offered business plans yet. Once the laser links are active, I expect high frequency traders to purchase direct orbital links from NYC to other major stock markets. If that was not several hundred million dollars to several billion per year, I would be shocked. I plan on purchasing 40 plus Starlink dishes for my locations to serve as tertiary backups to the existing fiber and LTE backups. I also plan to at least price out essentially orbital equivalent of MPLS. The majority of sysadmins I know are planning on doing similar, eventually. We're willing to pay far more than $100 per month for backup satellite internet that will have little to no bandwidth under non-emergency circumstances.
An ROI of 17 years for a major infrastructure project isn't terrible. I suspect the ROI will be closer to 5-10 years once you factor in all of the costs, both opex and capex. Which is shockingly good for a major global infrastructure. Residential customers likely will cover the costs of the infrastructure, which is pretty normal for telecom service. Business and government users will be the main source of profits. Much higher margin, much lower costs per user. I imagine the bottleneck will be satellite and phased array antenna production.
Re:Almost 17 years to pay off (Score:4, Informative)
Starlink already received a $900m FCC subsidy for rural internet. That's 9% of the $10B.
Yes and no. It's not a lump sum just handed over. It's a per-eligible subscriber per month sum, and it does not get added to the standard $99/month retail price. It makes up a part of that price. SpaceX is expected to take a decade to collect it all, and as far as SpaceX's own finances go, it's distinguishable from standard retail income only in that it costs them slightly more to receive it, since they have to pay someone to file paperwork with the US federal government every year. As far as payoff calculations go, multiplying SpaceX's subscriber base by $99/month plus the one time $500 base station fee is the correct calculation.
Feel the love for Elon! (Score:5, Insightful)
Re: (Score:1)
I love C.D Reimer.
Re:Feel the love for Elon! (Score:4, Insightful)
And taking the state of the internet and subsequent adjustment of traditional media into account this surprises you why exactly?
You need look no further than Slashdot... granted, this has always been a problem here, but you get modded up or down to a large degree based on whether the people with mod points agree with your point of view rather than how factual your post is and how sound your logic.
This is especially true for topics like Trump and Corona.
In that climate, it is no wonder that somebody like Musk has become a defacto influencer who can even game the stock market with a single tweet.
I mean I almost feel presured to put a disclaimer in here of what I personally think about Musk... but that too I would bet money on being used against me in some way.
Re: (Score:3)
It's a shame because Starlink is a great service, but it also has limitations and it's become almost impossible to discuss them, or the wider issues around internet access.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
The horror of someone investing in something with design choices intended to benefit the districts and not the bigots in the capital!
Re: (Score:2)
For one they are already planning to expand the number of sats four fold to over 40k but for another that really only impacts densely populated areas. Those people have plenty of options and always have.
In the US, those people have maybe one and at most two real options. Your idea of "plenty" deviates rather far from mine. Those one or two options invariably behave like the worst possible monopolists, so I don't call that "plenty" of options at all.
Unfortunately because of density and sheer expense, Starlink won't be putting even a dent into those monopolies, so for urban and suburban dwellers, nothing much changes. Rural areas might see a population boom though. It becomes much more viable to migrate
Re: (Score:2)
Not just a population boom but an economic boom. Rural populations have a problem. They hold more traditional economic values and basically have a pocket economy wherein they are trading around a far more valuable dollar with few new dollars coming in. That would be fine if they had their own currency.
That is what I'd like to see. A crypto currency dedic
Charter sucks I hate it (Score:3)
Elon sure knows how to sell (Score:1, Flamebait)
For one of the best examples, see how many self driving cars he sold to date since 2016 when Tesla first introduced Full Self Driving option to all Teslas. Granted, selling and delivering are two different things, but Elon is definitely a genius salesman. He's also great at pacifying the masses who paid for stuff and didn't get it even years later. Your 2016 Full Self Driving car still doesn't drive itself home after it drops you off at the airport? Well, don't worry, your car can make fart noises on demand
Re: (Score:1)
The 1 million Tesla Robotaxis put on the road in 2020 prove that you are wrong!
Competition is good (Score:2)
The point is, ladies and gentleman, that competition, for lack of a better word, is good. competition is right, competition works. competition clarifies, cuts through, and captures the essence of the evolutionary spirit. Competition, in all of its forms; competition for life, for money, for love, knowledge has marked the upward surge of mankind.
Re: (Score:2)
Re: Competition is good (Score:2)
The success of open source shows what we can achieve when collaborating.
amazing (Score:5, Interesting)
That it's cheaper to put satellites up at $30M per launch than it is to run fiber along millions of existing power poles that are already delivering telephone and power to homes and businesses. That it's cheaper than rolling out 5G towers.
Amazing.
Re: (Score:2)
It's easier to maintain and have backups for satellites in space than for terrestrial lines out to nowhere. Just look how it sometimes takes weeks to put up every single power pole after a major event (storm, flood, freeze, ...).
(if you live somewhere common - like in a large city - everything is probably solved in a couple of days. If you don't, you're on the very tail of a long list of work orders).
Re: amazing (Score:1)
Re: (Score:2)
Poles are typically owned by the electric utility. Often places with municipal internet do it affordably because they also own the poles. It's mostly a red tape issue than cost. My issue with space based is it seems to me to be a lot more cost effective and easier to roll out a good 5G network to most people. Space is absurdly expensive up front. Seems to be 5G will do to this what broadband internet did to directv and dish.
Re: (Score:2)
Yes and that's why I'm hedging my bets by signing up for both StarLink and TMobile 5G. Neither one of them are actually available where I live but they have started to allow people to sign up for future service. My hope is that 5G is available first and I can try it out for a month or two and see if it is any good. I will be happy to get a consistent 150MB/sec download speed. Upload speeds are not nearly as important to me.
If I can get that I will cancel the StarLink and get my $99 back. I'm still not that
Re: amazing (Score:2)
This is very true. In one locality an ISP might have to pay more in permit fees than the cost of pulling a mile / a km of fiber in a different locale.
Re:amazing - $500K per sat vs $25K per mile (Score:5, Interesting)
60 sats/launch/$30M/launch = $500K per sat. Avg cost of arial fiber cable per mile = $25K. My rural county has thousands of miles of roads. 2K miles x $25K = $50M. One sat should be able to cover dozens if not hundreds of counties. https://www.ctcnet.us/CTCCosts... [ctcnet.us]
Re: (Score:2)
What this says to me is that the real impact of technology is a very different thing from having an idea or even a working implementation. What triggers real change is when two cost curves cross over each other.
Re: (Score:2)
Re: (Score:2)
I can answer your conundrum. There are no lawyers in space.
Re: (Score:1)
Re: (Score:2)
Re: (Score:2)
500K Deposits to get on the list, NOT Purchases! (Score:3)
The 500K count is the number of folks who paid a fully refundable $100 deposit to get on the Starlink Beta list. It is NOT the number of folks who have paid for or are using Starlink services.
I was on that beta list for a while. When I got the go-ahead to buy my Starlink system, I instead used that notification to beat up on my ISP to get more bandwidth for less money. Then I requested a refund from Starlink, which they instantly provided. That I promptly paid as a deposit for an Aptera. I must have a thing for lists like this...
Re: (Score:2)
Re: (Score:2)
I wish Starlink would release numbers for "Active Customer Installations". That's the only number that matters in the real world.
Re: (Score:2)