19 ms·

Launch capacity as a whole was never really a big bottleneck and is also driven by demand and supply. But even cutting down cost didn’t really meaningfully incr

Launch capacity as a whole was never really a big bottleneck and is also driven by demand and supply. But even cutting down cost didn’t really meaningfully increase demand, as Falcon showed. SpaceX is already in the spot they cannot utilize Falcon capacity - if you look at their earlier plans, they were expecting to lunch them more than once per week long time ago. Instead, even with a bulk of their launches being internal ones, they are still way below that goal. Assuming Starship can deliver on their design goals, we’ll see if another price cut will actually drive up the demand.

stetrain 5y ago
It's also likely that lowering launch cost does (and already has) increased demand. But the timescales are large. It took SpaceX less time to make a reusable rocket than it will take the industry to pivot to more and cheaper launches. But it is starting to have an affect already. There are multiple companies talking about launching and maintaining satellite "constellations", not just Starlink. Launching and maintaining that kind of network is difficult if you can only launch once a year and it costs what launches cost 10 years ago.
- ncmncm 5y ago
  Starlink is the only thing that will keep SpaceX from imploding while they wait for people to understand their value. Needing, and bankrolling, several hundred launches over several years will gradually get people used to the idea of bulk freight to orbit, and some will find other plausible uses for the capability, and shepherd the idea through fundraising channels. It will take time.
  - consumer451 5y ago
    I am curious how long it will take the currently low volume and high margin satellite production industry to re-align to the new paradigm which requires higher volumes and lower margins. I could see the incumbents being very reticent to do so.
    - wumpus 5y ago
      Any company that wishes to compete with Starlink needs to be careful of costs. One reason OneWeb ran out of money is that they ended up spending 2X/satellite than their plan, in their fancy high volume factory.
      - kortilla 5y ago
        Competing with starlink is likely difficult, but there will be lots of room to disrupt observation, satellite radio, etc etc etc.
        wumpus 5y ago
        People are already doing earth observation from larger numbers of small satellites and very large numbers of cube sats. https://www.planet.com/our-constellations/ https://www.planet.com/our-constellations/ They started with this strategy before Starlink was started.
    - rtpg 5y ago
      It’s not the new paradigm which requires higher volumes and lower margins, it’s SpaceX’s business model which requires it (or, I suppose, expects it) It is possible that there is a cap in how much stuff people want to put into space. (EDIT: I do think it’s possible for SpaceX to exist as an entity that is simply smaller and not wasting its time trying to make its CEO Baron Harkonnen , but hey corporate ownership law amirite?)
      - consumer451 5y ago
        > It’s not the new paradigm which requires higher volumes and lower margins, it’s SpaceX’s business model which requires it (or, I suppose, expects it) I used the wrong word there. I was thinking more in terms of: SpaceX enables higher volumes, etc.. I know it's not a perfect analogy but cheap and predictable global shipping enabled business models which were not possible before. Maybe cheap and predictable access to space will do the same.
      - 5y ago
  - baq 5y ago
    Starlink is much bigger than that. It isn’t life support. It’s the ultimate cash cow, profits from which are in at least some part expected to fund some part of a Mars mission. This is why SpaceX isn’t public - your wall street shareholders would suffocate Musk with lawsuits before he would be able to finish saying ‘Martian mission’.
    - ncmncm 5y ago
      Agreed that Starlink will be a huge cash cow, anyway until competition surfaces. $60B ARR? At issue is whether and when their planned launch capacity ever finds other customers. Once all 30k birds are up, will there be other uses for the launchers, besides going to Mars? Do they need more than a dozen or two, for the job?
      - baq 5y ago
        Starlink's orbit is so low, especially the large constellation, that it requires almost constant replenishment of birds due to extremely fast decay (order of 1-2 years of useful service life?) So yeah, if they ever launch the whole 30k, they'll be flying quite often; the first replenishment will likely happen before the constellation is complete if the start slow.
  - oh_sigh 5y ago
    Eh...SpaceX has investors knocking down doors trying to throw money at them, even before starlink was announced. They'd be fine without starlink, though I am excited for orbital internet.
    - ncmncm 5y ago
      Investors is one thing, customers another. It is possible customers will appear, but most uses don't seem to need many launches. Even a really big telescope doesn't seem to need more than 50 launches, $2.5B. A moonbase needs fuel depot runs in the tens per year, and the occasional crew exchange, all under $1B/y.
  - vkou 5y ago
    Alternative theory: People have a decent understanding of the value of satellites, and it's SpaceX who doesn't. If you lower the dollar cost of going to the dentist by 10x, you're not going to see people going to the dentist 10 times more often.
    - rbanffy 5y ago
      If someone made having my mouth professionally cleaned more convenient, I would change the schedule to every 3 months instead of every six. We can anticipate some businesses would benefit from having low-cost access to microgravity environments - right now you carefully select your experiments, because it's extremely costly, but, if it weren't, you wouldn't need to. While satellite/probe/crewed-craft building has an extensive industrial chain with multiple bottlenecks that would hold it back if launch costs plumeted, there are others that have few and that would be enabled. We'd probably need a lot of barriers so that people don't launch cubesats for $100 in volumes that make LEO dangerous.
      - vkou 5y ago
        And if someone had self-cleaning easier, you wouldn't be going in for cleaning at all. Nobody cares about satellites themselves. What they care about is using them for something useful. Communication and imaging. At some point, both markets will become saturated. I don't know what level of demand there is for them, but it's entirely conceivable that it will be long-term limited by something other than launch costs.
dylan604 5y ago
But isn't it SpaceX plan for getting to Mars to require something like 35 Starship sized lifts to get all of the supplies needed for one trip to Mars? Sounds like builtin needs for launch right there. Wasn't this "I can't get to space on my own so I'll sue everyone that can" Bezo's main critique in one of their attempts to make fun of SpaceX?
- rbanffy 5y ago
  But who'll be paying for that trip? Does NASA have enough astronauts to crew a Mars base in addition to a Moon base? What private companies would want to do with Mars that can't be done on Earth? I'm all in for astronaut selection to be less rigorous (since I wasn't flying supersonic jets and earning a PhD when I was 18 - seriously, the bar is waaaaay too high) and would be delighted to be able to work that remotely, but I'd expect to be paid to go and have a return ticket already paid for.
  - devit 5y ago
    As weird as it may seem, it looks like that there may be people willing to pay positive sums of money to go to Mars even without a return ticket.
    - vkou 5y ago
      Those people should try serving three tours on a submarine first, and a few winters in the Antarctic, after. My guess is, that short list of volunteers will get a lot shorter.
      - greedo 5y ago
        The difference isn't being stuck in a metal tube, it's being able to see space. See stars/constellations etc in person. On a sub you don't get to see anything. And I haven't heard much about the views from Antarctica.
        vkou 5y ago
TrainedMonkey 5y ago
I think you argument hinges on the fact that demand shows up immediately after the price is cut. That would require there is a backlog of payloads which are just waiting for the price to be low enough. That is obviously not true, I believe lower prices are dramatically accelerating the demand as evidenced by space sector investment (1), but the demand curve is lagging because building space vehicles is hard. (1) I could not find a good chart showing the historic Space Sector investment, but there were multiple articles talking about record amount of the investment: https://spacenews.com/space-industry-in-midst-of-transformation-following-record-private-and-public-investments/ https://spacenews.com/space-industry-in-midst-of-transformat...
- spfzero 5y ago
  >>building space vehicles is hard ...mostly because reliability must be so high and launching mass costs so much. If you can launch mass cheaply, and accept lower reliability, space vehicles should be much easier to build.
- rbanffy 5y ago
  > That would require there is a backlog of payloads which are just waiting for the price to be low enough. Not only that, but they'd need to integrate with Starship and, right now, we aren't even sure which way the payload needs to go once the target orbit is achieved (it looks like it's sideways, while every other launcher releases the payload forward).
wumpus 5y ago
Falcon 9 launched every 9 days for the first half of this year.
- rbanffy 5y ago
  It should be able to launch a lot more than that. Maybe the expendable second stage is the limiting factor here.
  - wumpus 5y ago
    It seems unlikely that there were more payloads sitting around waiting to launch. If there were, then second stages (always expended), fairings (imperfectly reused), and weather/pad/ship operations are all possible limiting factors.
cptaj 5y ago
The other issue is that a really aggressive price cut hasn't happened yet because SpaceX doesn't have any competitors that can cut prices to match. Why cut further when you're already the cheapest?
- rbanffy 5y ago
  Because they need demand to pay for the launches.
  - Robotbeat 5y ago
    But demand isn’t hyperelastic, is it? If anything, a quick reduction in price would reduce total revenue (and total profit even more) as demand wouldn’t immediately keep up. It takes time for mass demand to respond.
mlindner 5y ago
> Launch capacity as a whole was never really a big bottleneck and is also driven by demand and supply. But even cutting down cost didn’t really meaningfully increase demand, as Falcon showed. I disagree. The excess payload and cheapness of Falcon 9 created a huge market for smallsats and microsats as secondary payloads. If you look at the _number_ (not mass) of operational payloads since Falcon 9 has come online, the numbers have been shooting up dramatically (and that is excluding Starlink). There's been a 5-fold increase in the number of smaller satellites being launched (excluding Starlink) since the last decade. See (pdf): https://brycetech.com/reports/report-documents/Bryce_Smallsats_2021.pdf https://brycetech.com/reports/report-documents/Bryce_Smallsa...
ksdale 5y ago
I think it's hard to say there isn't demand since it was so ridiculously expensive prior to SpaceX. Getting things into orbit at this price point is quite a new thing. In a recent post on this blog, he talked about how before SpaceX, even basic things headed into orbit had to be redesigned entirely to satisfy grueling mass requirements. I wouldn't be surprised if it takes a few more years for people to realize that they're not subject to those restrictions anymore and to fully understand the import of that. (Indeed, the prior post I mentioned is called "Starship is Still Not Understood.)

stetrain 5y ago

It's also likely that lowering launch cost does (and already has) increased demand. But the timescales are large. It took SpaceX less time to make a reusable rocket than it will take the industry to pivot to more and cheaper launches. But it is starting to have an affect already. There are multiple companies talking about launching and maintaining satellite "constellations", not just Starlink. Launching and maintaining that kind of network is difficult if you can only launch once a year and it costs what launches cost 10 years ago.

ncmncm 5y ago
Starlink is the only thing that will keep SpaceX from imploding while they wait for people to understand their value. Needing, and bankrolling, several hundred launches over several years will gradually get people used to the idea of bulk freight to orbit, and some will find other plausible uses for the capability, and shepherd the idea through fundraising channels. It will take time.
- consumer451 5y ago
  I am curious how long it will take the currently low volume and high margin satellite production industry to re-align to the new paradigm which requires higher volumes and lower margins. I could see the incumbents being very reticent to do so.
  - wumpus 5y ago
    Any company that wishes to compete with Starlink needs to be careful of costs. One reason OneWeb ran out of money is that they ended up spending 2X/satellite than their plan, in their fancy high volume factory.
    - kortilla 5y ago
      Competing with starlink is likely difficult, but there will be lots of room to disrupt observation, satellite radio, etc etc etc.
      - wumpus 5y ago
        People are already doing earth observation from larger numbers of small satellites and very large numbers of cube sats. https://www.planet.com/our-constellations/ https://www.planet.com/our-constellations/ They started with this strategy before Starlink was started.
  - rtpg 5y ago
    It’s not the new paradigm which requires higher volumes and lower margins, it’s SpaceX’s business model which requires it (or, I suppose, expects it) It is possible that there is a cap in how much stuff people want to put into space. (EDIT: I do think it’s possible for SpaceX to exist as an entity that is simply smaller and not wasting its time trying to make its CEO Baron Harkonnen , but hey corporate ownership law amirite?)
    - consumer451 5y ago
      > It’s not the new paradigm which requires higher volumes and lower margins, it’s SpaceX’s business model which requires it (or, I suppose, expects it) I used the wrong word there. I was thinking more in terms of: SpaceX enables higher volumes, etc.. I know it's not a perfect analogy but cheap and predictable global shipping enabled business models which were not possible before. Maybe cheap and predictable access to space will do the same.
    - 5y ago
- baq 5y ago
  Starlink is much bigger than that. It isn’t life support. It’s the ultimate cash cow, profits from which are in at least some part expected to fund some part of a Mars mission. This is why SpaceX isn’t public - your wall street shareholders would suffocate Musk with lawsuits before he would be able to finish saying ‘Martian mission’.
  - ncmncm 5y ago
    Agreed that Starlink will be a huge cash cow, anyway until competition surfaces. $60B ARR? At issue is whether and when their planned launch capacity ever finds other customers. Once all 30k birds are up, will there be other uses for the launchers, besides going to Mars? Do they need more than a dozen or two, for the job?
    - baq 5y ago
      Starlink's orbit is so low, especially the large constellation, that it requires almost constant replenishment of birds due to extremely fast decay (order of 1-2 years of useful service life?) So yeah, if they ever launch the whole 30k, they'll be flying quite often; the first replenishment will likely happen before the constellation is complete if the start slow.
- oh_sigh 5y ago
  Eh...SpaceX has investors knocking down doors trying to throw money at them, even before starlink was announced. They'd be fine without starlink, though I am excited for orbital internet.
  - ncmncm 5y ago
    Investors is one thing, customers another. It is possible customers will appear, but most uses don't seem to need many launches. Even a really big telescope doesn't seem to need more than 50 launches, $2.5B. A moonbase needs fuel depot runs in the tens per year, and the occasional crew exchange, all under $1B/y.
- vkou 5y ago
  Alternative theory: People have a decent understanding of the value of satellites, and it's SpaceX who doesn't. If you lower the dollar cost of going to the dentist by 10x, you're not going to see people going to the dentist 10 times more often.
  - rbanffy 5y ago
    If someone made having my mouth professionally cleaned more convenient, I would change the schedule to every 3 months instead of every six. We can anticipate some businesses would benefit from having low-cost access to microgravity environments - right now you carefully select your experiments, because it's extremely costly, but, if it weren't, you wouldn't need to. While satellite/probe/crewed-craft building has an extensive industrial chain with multiple bottlenecks that would hold it back if launch costs plumeted, there are others that have few and that would be enabled. We'd probably need a lot of barriers so that people don't launch cubesats for $100 in volumes that make LEO dangerous.
    - vkou 5y ago
      And if someone had self-cleaning easier, you wouldn't be going in for cleaning at all. Nobody cares about satellites themselves. What they care about is using them for something useful. Communication and imaging. At some point, both markets will become saturated. I don't know what level of demand there is for them, but it's entirely conceivable that it will be long-term limited by something other than launch costs.

dylan604 5y ago

But isn't it SpaceX plan for getting to Mars to require something like 35 Starship sized lifts to get all of the supplies needed for one trip to Mars? Sounds like builtin needs for launch right there. Wasn't this "I can't get to space on my own so I'll sue everyone that can" Bezo's main critique in one of their attempts to make fun of SpaceX?

rbanffy 5y ago
But who'll be paying for that trip? Does NASA have enough astronauts to crew a Mars base in addition to a Moon base? What private companies would want to do with Mars that can't be done on Earth? I'm all in for astronaut selection to be less rigorous (since I wasn't flying supersonic jets and earning a PhD when I was 18 - seriously, the bar is waaaaay too high) and would be delighted to be able to work that remotely, but I'd expect to be paid to go and have a return ticket already paid for.
- devit 5y ago
  As weird as it may seem, it looks like that there may be people willing to pay positive sums of money to go to Mars even without a return ticket.
  - vkou 5y ago
    Those people should try serving three tours on a submarine first, and a few winters in the Antarctic, after. My guess is, that short list of volunteers will get a lot shorter.
    - greedo 5y ago
      The difference isn't being stuck in a metal tube, it's being able to see space. See stars/constellations etc in person. On a sub you don't get to see anything. And I haven't heard much about the views from Antarctica.
      - vkou 5y ago

TrainedMonkey 5y ago

I think you argument hinges on the fact that demand shows up immediately after the price is cut. That would require there is a backlog of payloads which are just waiting for the price to be low enough. That is obviously not true, I believe lower prices are dramatically accelerating the demand as evidenced by space sector investment (1), but the demand curve is lagging because building space vehicles is hard. (1) I could not find a good chart showing the historic Space Sector investment, but there were multiple articles talking about record amount of the investment: https://spacenews.com/space-industry-in-midst-of-transformation-following-record-private-and-public-investments/ https://spacenews.com/space-industry-in-midst-of-transformat...

spfzero 5y ago
>>building space vehicles is hard ...mostly because reliability must be so high and launching mass costs so much. If you can launch mass cheaply, and accept lower reliability, space vehicles should be much easier to build.
rbanffy 5y ago
> That would require there is a backlog of payloads which are just waiting for the price to be low enough. Not only that, but they'd need to integrate with Starship and, right now, we aren't even sure which way the payload needs to go once the target orbit is achieved (it looks like it's sideways, while every other launcher releases the payload forward).

wumpus 5y ago

Falcon 9 launched every 9 days for the first half of this year.

rbanffy 5y ago
It should be able to launch a lot more than that. Maybe the expendable second stage is the limiting factor here.
- wumpus 5y ago
  It seems unlikely that there were more payloads sitting around waiting to launch. If there were, then second stages (always expended), fairings (imperfectly reused), and weather/pad/ship operations are all possible limiting factors.

cptaj 5y ago

The other issue is that a really aggressive price cut hasn't happened yet because SpaceX doesn't have any competitors that can cut prices to match. Why cut further when you're already the cheapest?

rbanffy 5y ago
Because they need demand to pay for the launches.
- Robotbeat 5y ago
  But demand isn’t hyperelastic, is it? If anything, a quick reduction in price would reduce total revenue (and total profit even more) as demand wouldn’t immediately keep up. It takes time for mass demand to respond.

mlindner 5y ago

> Launch capacity as a whole was never really a big bottleneck and is also driven by demand and supply. But even cutting down cost didn’t really meaningfully increase demand, as Falcon showed. I disagree. The excess payload and cheapness of Falcon 9 created a huge market for smallsats and microsats as secondary payloads. If you look at the _number_ (not mass) of operational payloads since Falcon 9 has come online, the numbers have been shooting up dramatically (and that is excluding Starlink). There's been a 5-fold increase in the number of smaller satellites being launched (excluding Starlink) since the last decade. See (pdf): https://brycetech.com/reports/report-documents/Bryce_Smallsats_2021.pdf https://brycetech.com/reports/report-documents/Bryce_Smallsa...

ksdale 5y ago

I think it's hard to say there isn't demand since it was so ridiculously expensive prior to SpaceX. Getting things into orbit at this price point is quite a new thing. In a recent post on this blog, he talked about how before SpaceX, even basic things headed into orbit had to be redesigned entirely to satisfy grueling mass requirements. I wouldn't be surprised if it takes a few more years for people to realize that they're not subject to those restrictions anymore and to fully understand the import of that. (Indeed, the prior post I mentioned is called "Starship is Still Not Understood.)