3 ms·

Hold onto your hat then: in the US alone there are at least 20M people without high-speed internet access [1]. Around 45% of the entire world's population [2] s

Hold onto your hat then: in the US alone there are at least 20M people without high-speed internet access [1]. Around 45% of the entire world's population [2] still doesn't have internet access - at all, not just broadband. They've said starship, when operational, will be able to deliver 400 starlink satellites to orbit at a time. The plan doesn't sound far fetched. [1] https://archive.is/do3Qp https://archive.is/do3Qp [2] https://en.unesco.org/news/new-report-global-broadband-access-underscores-urgent-need-reach-half-world-still-unconnected https://en.unesco.org/news/new-report-global-broadband-acces...

tsimionescu 5y ago
The vast majority of these people will not be able to afford a 1500$ satellite dish (that's how much it costs to build the thing, even though they sell it for 500$ for now) + 100$/month internet.
- ricardobeat 5y ago
  And why should one assume the cost once deployed in the millions will equal the cost while in beta with 60k users? There's another comment here already mentioning their prediction of a $300 cost for the dish in the future, you can only expect the subscription to go down as well. Plus, the places where Starlink is the most useful - rural areas - already have sky high costs. I just looked up what service looks like in Brazil: ~R$400/month (US$76) for 20mbps, with a 80GB data cap. When you have to buy the dish it costs over R$10k (~US$2000). This is the best you can get.
  - tsimionescu 5y ago
    According to them, the satellites have a maximum life span in orbit of 5 years. They can launch 60 satellites at a time, at a cost of 50M$ per launch. So they need to spend ~1.4 billion dollars every five years just to keep the constellation at its current size, ignoring the cost of the satellites, ground stations etc. If they have 500k customers, that means that each customer must pay ~50$/month just to pay for the cost of keeping the satellites in orbit. Add ground station costs, internet bandwidth, satellite production, dish production, customer service, personnel costs, R&D and you'll quickly see that the current prices are much below what is required to be profitable. Even if they have 5M customers (very hard to do in 5 years time), I wouldn't be hopeful for their prospects of not going under. Of course, they may well survive on government handouts. Edit to add: the bottom 45% who don't currently have access to high speed internet are unlikely to afford even a 300$ dish and an expensive data plan. The price of current plans is one of the main reasons why they don't have access in the first place.
    - regnerba 5y ago
      If they are able to get Strarship and the Super Heavy Booster working, which at this point in time I think is very probable, those economics will change a good bit as the cost of launch will come down a lot.
    - ricardobeat 5y ago
      $50M is the launch cost for SpaceX customers. Musk has said in an interview that the marginal cost of a reused Falcon 9 launch is $15 million. That means even the tiny subscriber base they have at the moment is already paying a significant chunk of the launches. That would make your numbers $420m every five years, or $84 million a year - slighly more than your average telecom CEO compensation. But the goal for starship is $2m per launch. Let's assume that is bullshit and go for $10M instead, and only 300 satellites per launch. That's already down to ~$10M/year to maintain the entire constellation. Now you have room for all the other costs you mention. If they reach the stated goals, it will cost peanuts... you can see where this is going. Starship will be a ridiculous leap in launch costs if it comes to fruition. The max capacity for the network, once at 42k satellites, is estimated to be around 50M people worldwide. Plus, you could have a downlink shared, say, among 50 people and spread the cost - 1-10mbps is still miles beyond 'no internet access'. You might think the plan is crazy, but the math checks out. If this whole thing works, they could charge $10/month and be in the green - the satellites themselves will be the cheapest part of the whole system. No rent, no right of way, no construction, no maintenance work, free energy and extremely low material costs.
      - ricardobeat 5y ago
        To add on the costs (yeah I have nothing better to do) - ground station: must be negligible compared to everything else. They are literally planting giant routers out in a farm, need to pay rent, power and cabling to an exchange. Zero other infrastructure costs - internet bandwidth: despite Netflix's attempts, ISPs have peering agreements that are usually free, as does Starlink - satellite production: nobody knows, but they have said the payload is cheaper than the launch, so somewhere between $250-500k each at start - customer service: online only, seems like the dish is self-diagnosing, will likely end up being much lower than traditional broadband
  - shaklee3 5y ago
    There is absolutely no basis for the cost to be that low. Nobody believes that. Not even Elon, since he's commented on it several times.
- beeboop 5y ago
  No but their community or local government can and will. Internet access has monumental positive impacts for an economy and $1500/$100month is such a small amount of money for a city or communtiy resource for the expected return. No, it's not going to be on every mud hut in the world. But starlink will definitely provide internet to millions who didn't have it before, and do a great deal of service for lifting some of those people out of poverty.

tsimionescu 5y ago

The vast majority of these people will not be able to afford a 1500$ satellite dish (that's how much it costs to build the thing, even though they sell it for 500$ for now) + 100$/month internet.

ricardobeat 5y ago
And why should one assume the cost once deployed in the millions will equal the cost while in beta with 60k users? There's another comment here already mentioning their prediction of a $300 cost for the dish in the future, you can only expect the subscription to go down as well. Plus, the places where Starlink is the most useful - rural areas - already have sky high costs. I just looked up what service looks like in Brazil: ~R$400/month (US$76) for 20mbps, with a 80GB data cap. When you have to buy the dish it costs over R$10k (~US$2000). This is the best you can get.
- tsimionescu 5y ago
  According to them, the satellites have a maximum life span in orbit of 5 years. They can launch 60 satellites at a time, at a cost of 50M$ per launch. So they need to spend ~1.4 billion dollars every five years just to keep the constellation at its current size, ignoring the cost of the satellites, ground stations etc. If they have 500k customers, that means that each customer must pay ~50$/month just to pay for the cost of keeping the satellites in orbit. Add ground station costs, internet bandwidth, satellite production, dish production, customer service, personnel costs, R&D and you'll quickly see that the current prices are much below what is required to be profitable. Even if they have 5M customers (very hard to do in 5 years time), I wouldn't be hopeful for their prospects of not going under. Of course, they may well survive on government handouts. Edit to add: the bottom 45% who don't currently have access to high speed internet are unlikely to afford even a 300$ dish and an expensive data plan. The price of current plans is one of the main reasons why they don't have access in the first place.
  - regnerba 5y ago
    If they are able to get Strarship and the Super Heavy Booster working, which at this point in time I think is very probable, those economics will change a good bit as the cost of launch will come down a lot.
  - ricardobeat 5y ago
    $50M is the launch cost for SpaceX customers. Musk has said in an interview that the marginal cost of a reused Falcon 9 launch is $15 million. That means even the tiny subscriber base they have at the moment is already paying a significant chunk of the launches. That would make your numbers $420m every five years, or $84 million a year - slighly more than your average telecom CEO compensation. But the goal for starship is $2m per launch. Let's assume that is bullshit and go for $10M instead, and only 300 satellites per launch. That's already down to ~$10M/year to maintain the entire constellation. Now you have room for all the other costs you mention. If they reach the stated goals, it will cost peanuts... you can see where this is going. Starship will be a ridiculous leap in launch costs if it comes to fruition. The max capacity for the network, once at 42k satellites, is estimated to be around 50M people worldwide. Plus, you could have a downlink shared, say, among 50 people and spread the cost - 1-10mbps is still miles beyond 'no internet access'. You might think the plan is crazy, but the math checks out. If this whole thing works, they could charge $10/month and be in the green - the satellites themselves will be the cheapest part of the whole system. No rent, no right of way, no construction, no maintenance work, free energy and extremely low material costs.
    - ricardobeat 5y ago
      To add on the costs (yeah I have nothing better to do) - ground station: must be negligible compared to everything else. They are literally planting giant routers out in a farm, need to pay rent, power and cabling to an exchange. Zero other infrastructure costs - internet bandwidth: despite Netflix's attempts, ISPs have peering agreements that are usually free, as does Starlink - satellite production: nobody knows, but they have said the payload is cheaper than the launch, so somewhere between $250-500k each at start - customer service: online only, seems like the dish is self-diagnosing, will likely end up being much lower than traditional broadband
- shaklee3 5y ago
  There is absolutely no basis for the cost to be that low. Nobody believes that. Not even Elon, since he's commented on it several times.
beeboop 5y ago
No but their community or local government can and will. Internet access has monumental positive impacts for an economy and $1500/$100month is such a small amount of money for a city or communtiy resource for the expected return. No, it's not going to be on every mud hut in the world. But starlink will definitely provide internet to millions who didn't have it before, and do a great deal of service for lifting some of those people out of poverty.