Hacker News Reader

Comment by motionlessveloc | Hacker News Reader

Starship's commercial viability rests almost entirely on whether it can orbit Starlink satellites more cheaply than Falcon. Falcon is around $3000/kg, Starship was promised at $100/kg, I doubt even Elon believes that number but it does mean they have plenty of buffer and cutting capacity from 50T to 100T is not the end of the world.

I agree that the Human Landing System plan for the moon is a shitshow, but it's almost a distraction for SpaceX at this point, and it would not surprise me if the whole program gets axed by whoever comes after Trump.

View on HN

The problem is not that they are not meeting their payload spec, or that they have not demonstrated in orbit refueling.

The problem lies in the relationship of those two milestones: if you can't meet payload, it takes way too many starships to refuel, no matter how well the fuel transfer works.

Every dead project demonstrates progress every week, right up until the day it is canceled.

It won't get axed, the race with China is on.

Its commercial viability (not program success) depends on Starlink launch, sure. But also: that depends on Starlink's viability and value.

Selling sub-fiber Internet access into a globe increasingly covered in fiber is a silly long-term value proposition.

Starlink is just the civil version of Starshield. As long as the US Army needs unrestricted internet access across the globe Starlink will continue to exist.

I didn't doubt that Starlink will continue to exist.

Is it? I think you might think differently if you challenge yourself to calculate the total addressable market for Starlink.

The vast majority of that market is too poor to afford Starlink and as soon as they're not, their top priority will be to invest the capex to replace it with the vastly superior terrestrial modalities.

Their top priority? Compared to what?

Compared to spending large amounts of money on shitty space-internet?

The elephant in the Starlink business model room is the fact that terrestrial wireless keeps getting cheaper and faster. To be a good Starlink customer you need to be prosperous enough to afford it, while being far enough away from an urban center to have little or no chance of being served by terrestrial wireless services.

That's a very special TAM to begin with, and the harsh part is that it shrinks over time.

As a product manager, I can tell you you're making a mistake in the way you're thinking about this. If you actually try calculating it, you'll figure it out.

Makes sense. Product Management seems to attract the economically illiterate.

Starlink has about 13 million subscribers today and makes around 500$/year net on average per subscriber. It's still growing fast and will get faster and more reliable. You will also be able to use from your phone quite soon. It can very realistically get to 100 million subscribers which is going to generate 50 billion/year of profit. They have 0 competition in that space. Even if they make half of that it's still a 1 trillion dollar company right there.

You are saying "sub-fiber" but Starlink will likely be faster for long round trips once enough new gen satellites are deployed.

That comes before the whole datacenters in space conversation.

> You will also be able to use from your phone quite soon

This, I think, is a much bigger deal than a lot of people realize. A global telephone provider that works in the middle of the ocean or in the middle of the dessert or pretty much anywhere on the planet. The simplification to international travel alone would be worth switching.

Apple is currently doing this via Globalstar's satellites for sos messaging.

Yeah I'll totally spend more than 0 seconds of my life and $0 switching telephony providers so that I can have connectivity when I'm ... checks notes ... in the middle of the ocean.

Huge huge untapped market of people just like me.

Yeah you are not the target demographic. I buy e-sims all the time, have 2 plans from two different countries and I am already thinking about buying Starlink mini (waiting for the new device) because I am often in regions with terrible coverage.

So let's call it 50-50?

Amazon Leo has already launched almost 400 satellites, with expected service 2027 so there's one competitor approaching in that space.

>Selling sub-fiber Internet access into a globe increasingly covered in fiber is a silly long-term value proposition.

Wow, that's so amazing estaerum, you've figured out how to get photonic-bandgap (hollow core) fiber to transoceanic airlines and ships, as well as drones, and satellites in orbit!? Would you mind sharing any hints on this great breakthrough you've made?

...or did you just sorta somehow completely ignore the whole "terminals can be mobile" aspect of Starlink, as well as latency of free space optical vs conventional fiber?

The market for airlines, ships, drones, and satellites is - quite obviously - nowhere near large enough to justify Starship lmao

80% of the Starlink subscriber base is fixed points and that will grow.

>The market for airlines, ships, drones, and satellites is - quite obviously - nowhere near large enough to justify Starship lmao

The only thing that's "quite obvious" is that you're not merely incredibly, confidently ignorant, you're also so deeply uncurious you can't even be bothered to take 30 seconds to check your priors. Starlink's present annual revenue is $11.3 billion with $4.4 billion in income on that. The total global civil aviation (commercial airline) fleet size is over forty thousand aircraft [0]. Expand that to general aviation including business jets etc and it grows to something like half a million. Starlink actually publishes their general aviation plan pricing [1], which for CA is going to be one of the top two plans that don't have data limits and thus $12500 or $20000 per aircraft. From what I can find international flights account for something like 20-30% of all flights, and maybe 5-10% are intercontinental. But plans are per aircraft per month, and plenty of airliners use the same plane to fly across a continent and then also fly intercontiental, which would still need the highest plan. So using a 90/10 split that'd be an average of $13500/plane, for a TAM of $6.7 billion per year just for CA and maybe another $1-4 billion for GA (which is a lot more complicated to reason about). And the profit margins there are going to be high.

So JUST aviation, not even touching on maritime or everything else mobile, has the potential at present market size to approach the entire current revenue of Starlink by itself. And they have barely even gotten started with cellular, or doing other stuff like Earth sensing with the sats that bigger, higher power birds makes more feasible.

You claim it's "quite obvious" mobile applications don't justify Starlink. It's not any such thing.

>80% of the Starlink subscriber base is fixed points and that will grow.

You need a lot more than 5x the residential plans at $55/month to approach $10-20k/month. Of course, some of those fixed points are business/enterprise customers on higher priced plans, but those customers are likely to be using Starlink already (as we are) as part of a redundancy plan alongside fiber, so your original argument about spreading fiber is perfectly irrelevant to those users anyway.

----

0: https://statbase.org/datasets/air-rail-and-water-transportat...

1: https://starlink.com/support/article/78b0428b-4546-0b7f-ea45...

Ah, I see you're very confused.

The question of whether "portable Starlink" (the aviation/maritime++ category we're talking about) justifies Starship's existence is whether the marginal increase in payload capacity or efficiency of Starship over Falcon produces a marginal increase in "portable Starlink" profit that's multiples beyond the amortized development cost and the launch costs of Starship.

Given that Starlink's entire enterprise is probably loss-making if it had to pay for its own launch capacity (it currently pays SpaceX Space $0 and is accounted for just in hardware depreciation, so frontloads profit and all the costs are incurred in coming years)... the answer is almost certainly no.

And for what it's worth, the one aircraft I've been on with Starlink Internet was (by a gigantic margin) the worst in-flight wifi experience I've had in 10 years. I'm curious how much of cost-savings it is over the current alternatives. I hope it's a lot because there's a long road between here and whenever I'd willingly buy a seat on a Starlink-connected flight again.

>Ah, I see you're very confused.

Nah, that'd be you, as well as a very insufferable asshole.

>The question of whether "portable Starlink" (the aviation/maritime++ category we're talking about) justifies Starship's existence is whether the marginal increase in payload capacity or efficiency of Starship over Falcon produces a marginal increase in "portable Starlink" profit that's multiples beyond the amortized development cost and the launch costs of Starship.

And it clearly does, as I showed you. It's not even close, they're constrained on capabilities not demand right now. Starship total dev cost has been about $15 billion. Access to even a tiny piece of the global cellular market justifies that.

>Given that Starlink's entire enterprise is probably loss-making if it had to pay for its own launch capacity (it currently pays SpaceX Space $0 and is accounted for just in hardware depreciation, so frontloads profit and all the costs are incurred in coming years)... the answer is almost certainly no.

Speaking of "lmao", now you're resorting to make believe.

>And for what it's worth

Nothing, in your case. I'm also being kind to you and only talking about the fundamental thing it does that fiber can never do, vs your hilarious suggestion that fiber will rollout worldwide to everyone on the planet in or even in the US in the foreseeable future. I don't think the US has even reached 50% fiber coverage yet.

> Starship total dev cost has been about $15 billion. Access to even a tiny piece of the global cellular market justifies that.

Similarly I can justify developing a $15 billion pizza oven because the pizza industry is $250 billion. And are you even reading the words in front of you? The question is the marginal improvement in their ability to serve that unmet demand via Starship versus Falcon. You said literally nothing about this question and still haven't. Do you understand the question?

> now you're resorting to make believe.

Describing SpaceX's accounting practices and Starlink's actual launch costs is... make believe? Very substantive response.

> I'm also being kind to you and...

I didn't claim fiber will roll out to everyone on the planet, or even everyone in the US. I said the Starlink market is clearly shrinking, and Starlink's success will ironically accelerate that shrinkage because people who get access to shitty Internet are far, far more motivated to get good connectivity than people who have no connectivity at all.

And sure, let's assume fiber hasn't reached 50%. Even broadband is a vastly superior experience to Starlink. You can disprove this by showing me any evidence of a substantial number of people who have access to fiber or broadband and opt to use Starlink as their primary ISP. I would guess this set is approximately null, as someone who ordered Starlink for his house and then filed an FCC complaint to get fiber brought out to it instead. I would have to have been literally braindead to use Starlink.

Reply on news.ycombinator.com