3 ms·

I can only provide a back-of-the-envelope estimation for the US: 120M households[1], 20% in rural areas [2], 40% of the people cannot select between multiple IS

by qayxc 6y ago

I can only provide a back-of-the-envelope estimation for the US: 120M households[1], 20% in rural areas [2], 40% of the people cannot select between multiple ISPs [3]. At the suggested $100/mo you'd get 120M * 0.2 * $100 = $2.4Bn/mo from rural areas plus 120M * 0.8 * 0.4 * $100 ≈ $3.8Bn/mo from urban areas. So in the US a reasonable cap would be about $6.2Bn/mo or $74.4Bn/yr. Operational costs are difficult to estimate, but using the current numbers, $50M/launch and $250k/sat seems reasonable. At 60 sats per launch and 12k sats fleet size, build-up cost would be $10Bn launch costs plus $3B for the sats, so about $13Bn total. With an average lifespan of 5yr/sat, 40 launches per year are required to maintain the fleet. That's an additional $2.1Bn/yr for replenishing old sats. Costs for ground operations come on top of that. Considering all this, even without a major game changer in launch costs (i.e. Starhip), Starlink would be very profitable even if they "only" reach %25 of the rural population in NA. [1] https://www.census.gov/quickfacts/fact/table/US/PST045219 https://www.census.gov/quickfacts/fact/table/US/PST045219 [2] https://www.worldatlas.com/articles/continents-of-the-world-by-degree-of-urbanization.html https://www.worldatlas.com/articles/continents-of-the-world-... [3] https://broadbandmap.fcc.gov/#/area-comparison?version=jun2019&tech=acf&speed=25_3&searchtype=county&searched=y https://broadbandmap.fcc.gov/#/area-comparison?version=jun20...

montenegrohugo 6y ago
Great napkin math. I'd like to also add that Starlink is a project that plays along really well with their main development: Starship. Starship will bring Starlink costs down, but that's not the real benefit: what they really needed was a product to increase Starship launches in the short term. They need something to prove it's reliability. Starship won't have a separate abort system, and yet Spacex wants to fly people on it. Not only do they want it human rated (NASA spec for this is >1:275 launch failure, which is still quite high), they _also_ want routine intercontinental civilian flights. New york to Singapur in 40mins. But again, to do this they need to prove they're as reliable as an airplane, and to do that they need volume. Traditional satellite market has proven very slow to move and take advantage of cheaper prices, so Spacex does the next best thing: create its own volume.

by qayxc 6y ago

montenegrohugo 6y ago

Great napkin math. I'd like to also add that Starlink is a project that plays along really well with their main development: Starship. Starship will bring Starlink costs down, but that's not the real benefit: what they really needed was a product to increase Starship launches in the short term. They need something to prove it's reliability. Starship won't have a separate abort system, and yet Spacex wants to fly people on it. Not only do they want it human rated (NASA spec for this is >1:275 launch failure, which is still quite high), they _also_ want routine intercontinental civilian flights. New york to Singapur in 40mins. But again, to do this they need to prove they're as reliable as an airplane, and to do that they need volume. Traditional satellite market has proven very slow to move and take advantage of cheaper prices, so Spacex does the next best thing: create its own volume.