Beyond aggressively optimistic timelines, I find it difficult to disagree with the premise. The aggressively optimistic timelines is also what makes it feasible to even attempt these things, where e.g. the amount of iteration required for Starship would have broken most other companies.
> In the long term, space-based AI is obviously the only way to scale.
In the long term - all mass and energy available is outside of Earth - what is here is not even a rounding error. If you wish to continue scaling compute it then becomes a question of time before you'd want to go off planet. Personally I'm quite keen to see near term space based compute explored, as it could end up becoming a much better trade-off than allocating ever more ground to power and operate terrestrial compute which directly conflict with the biosphere.
SpaceX started the Starlink design phase in 2015 - started launching Starlink satellites in 2019 - and they now have the most dominant satellite constellation ever deployed by a large factor. They have their own launch systems, launch sites, satellite bus, communication stack - both in-house designed and built.
What is really going to be that difficult with space-based compute? Radiation hardening and cooling? These are clear engineering challenges that can be simulated, tested with earth analogs, and then rapidly iterated across design generations. There's napkin math all over the internet on this, but it really seems like small challenges compared to the other engineering SpaceX have already sorted.
Beyond radiation / cooling / servicing - it seems like the biggest hurdle is to crack the scaling of designing / scaling the necessary amount of compute they will need to scale space based compute according to the laid out plans.
In case anyone is wondering how Tesla’s stock price remain wildly detached from its business reality, keep these four words in mind. If you can convince people that anything about you and your business has to be evaluated on a literally astronomical timescale, you can justify any valuation you desire, because your believers will give you infinite time to realize their investment returns. It has nothing to do with business. They are selling you a vision — which can also come in a pill form, labeled "salvia" and sold at gas stations.
I still see people say the cybertruck is built for mars environments, conveniently ignoring the vast technological and economical barriers stopping us from driving commercially produced vehicles on mars. This space data center thing is the same deal. It doesn't matter how long it will take to solve the technical issues with cooling, radiation, maintenance. It doesn't matter if it will make economical sense or not. It doesn't matter if spacex will be the one to actually do it. You just have to believe, and give them some time — a lot of time, so much time that a monkey can type out Hamlet and type it out again backwards.
See also the buffoonery coming out of Bay Area "effective altruist" and "longtermism" communities.
I fully agree on the reality distortions and valuation chaos surrounding Tesla. This does also follow from the company being very volatile and chaotic, which becomes harder to price. How do you accurately price in e.g. Optimus - it seems really hard to tell at this point - which I guess is also one of the motivators for these strategies.
However, in this particular trajectory, SpaceX did build the rockets and did build Starlink which is now the best global-scale wireless communication network for many use-cases. Stretching this trajectory to scale up the technology to facilitate in-space computing is vastly more grounded than Shakespearean monkeys.
If it were only retail investors, your assumptions could make sense.
However plenty of the share ownership is institutional investors. Most of them care a bit more about fundamentals. (I'm ignoring passive investors just using indexes).
Chipping in here. There's a lot of speculation on this subject matter, all of which entirely wrong.
Technical concerns aside, the main risk is financial. Success is based on the premise that we need this enough that the costs are justified but the costs are going to be much higher. That is totally unproven on any financial modelling scenario I've seen. In fact there's likely no actual ROI on what has been spent so far and no qualification of demand. With geopolitical problems on the table, no one is going to fund this.
The idea is completely dead before the first node leaves the planet.
Not only that! First mover advantage is mostly bunk. Even if none of what you say was true, valuing this on a horizon far enough out to solve all the technical problems gives a lot of space for competitors to emerge. So even if the idea was both technically and financially workable, there is no guarantee that an investment in SpaceX specifically would be the right move over the same evaluation term as the tech will take to play out. It only makes sense from a speculative angle because you know that if they present any more concrete excuse to believe that the tech works, the price will temporarily go through the roof.
As problems go, radiation and cooling seem to have relatively low dimensionality compared to the other problems. It seems to be mostly a question of optimizing within the dimensions of dissipation / structure / deployment / service / cost / weight. When all is said and done, the cooling solution will end up being a module that can deal with some power dissipation, cost X amount, weight Y amount, have structural interface Z. This seems like something a relatively low number of engineers can iterate on largely isolated from other concerns. SpaceX does have 5000+ of them.
Comparing this to scaling the production of compute where they try to work outside the bounds of ASML (~40k employees) and TSMC (~80k+ employees), and where there is a huge number of degrees of freedom in many, many layers of the stack that have complicated interactions.
With radiation and cooling, SpaceX also has plenty of experience with both already given that they've had to solve this on existing satellites. Overall, Terafab just seems like a far harder challenge, and where I'd be more wary on timelines.
Just go to Google Maps. Just do it. Type in Arizona, or new mexico, texas whatever. Do you see all this brownish stuff? Yeah thats just empty land with A LOT of solar.
So in the long term, what do you think is cheaper and easier to maintain, upgrade, handle etc.?
A Space operation on which you need to send compute hardware constantly upwoards or a fiber connection to some more 'remote/dessert' like area which has a lot of energy available?
Starlink is not a game changer at all. It has 8-10 Million customers, from which plenty of peopple just use it for holidays, or upping there already existing internet line or because its faster to deploy than a cable.
Our planet is already very well connected. Putting lines in the ground is necessary anyway because you still need energy / powerlines.
Of course this can be done, thats NOT the question. The only question is, if its worth it and its not.
Sending some servers up in space is margins more expensive than sending some servers on trucks (you need anyway) to another earth location.
> Just go to Google Maps. Just do it. Type in Arizona, or new mexico, texas whatever. Do you see all this brownish stuff? Yeah thats just empty land with A LOT of solar.
'Brownish stuff', known more generally as natural ecosystems.
> So in the long term, what do you think is cheaper and easier to maintain, upgrade, handle etc.?
How long a term does your imagination stretch to? Are you really arguing that once provisioning, cooling, automated scaling in space, and off-planet mining are all solved problems, that shitting on our planet will still be the cheapest most maintainable option?
Compute in space is doable, we already send plenty of computers up there, technologically, it is not even a challenge. It just doesn't make sense economically, even with Starship, it is making things harder for no good reason.
Starlink is different, it makes sense. Covering the entire Earth, including the oceans with cell towers for global internet connectivity is harder than having a satellite constellation. The opposite situation from datacenters.
I still don't get why so many people who watch/read scifi conclude "yes this is a legitimate and realistic vision of the future, and not Wizards and Dragons but with Technology"
Going to mars or staying on the moon will be a Darwin Award-level adventure.
Honestly, I don't see (sea?) it. Every advantage of space are found in oceans/seas, especially if we use dead zones where aquatic life is already dead. The cooling is cheaper, tide+wind+solar is cheaper than space solar (I know someone who worked on a lens to observe the sun, the satellite was launched but due to being cheap on the solar panels, the sunlight and radiation chipped away the coating that found itself attracted to the most massive object in the area, the lens). Anti-corrosion is cheaper than light radiation protection, and servicing is way easier and cheaper.
But if solar panel is significantly cheaper and latency doesn't matter you can have servers in any part of the world. Even if they are not up 50% of the time due to limited battery it would still be cheaper.
biosphere interference from ground infrastructure? any idea the ground infrastructure it requires to support space based compute operations? i have a feeling that is comparable if not more impactful
you also shrug off cooling. this is not a solved problem in any way. its not even approachable as of yet. the vast size of the radiators will be hilarious regardless.
you ignore power generation. solar is not an option. so we also need nuclear reactors for these orbital data centers. thats cool spacex can just branch out into nuclear too! love the idea of unmanned nuclear orbiting behemoths.
speaking of orbital.. what is their orbit? do they go out to Lagrange points? hilariously far? or do they stay close? hilariously fuel intensive to stay out of the atmosphere for such massive structures?
but hey, maybe we distribute spaceX-AI gpu's across starlinks. a couple solar panels and a tesla battery per gpu. all launched there by spacex
'all mass and energy available is outside of earth'
Yeah, and out of range for compute data connections too.
I don't agree with the feasibility or ANY sort of practicality to this whatsoever. Im all for going for it, but I wish everyone could just admit that we're doing it because it's cool, not because it's useful. I get why Elon wont say that, but not us.
Your feelings are obviously your own, but a Starlink terminal isn't that big and can transfer quite a lot of tokens.
Every single satellite has sufficient cooling for its power production, otherwise they would be frying. Waste heat from a GPU is not materially different from waste heat from an amplifier. That's not cooling entire racks, but I don't think anybody talks about putting entire racks in space anymore.
I'm very much pro nuclear, but a solar cell in a sun synchronous orbit is pretty great too and eliminates most battery requirements
I very much doubt the economics of this makes sense, but I don't think a lot of your criticism is valid.
There are literally enormous problems powering AI data centers on the Earth right now. No, we don't have the power on Earth.
In terms of launch cost, Starship makes launch cost negligible. Some estimates are that it will cost less to launch a tonne to orbit, than to ship across the US by train.
Even if this figure is slightly low, that has nothing compared to the cost of real estate, construction costs, all of the building codes required to build a data center on Earth. These things all still apply underground, and underground is going to require additional shoring and structural engineering, to ensure that the structure is not crushed, damaged, and so forth.
All of this may be true but the scale that Musk is talking about would require an immense amount of solar panels -- and if he has the means to produce so many solar panels why not use them to solve our climate and energy crisis on Earth?
Seems more like a grift to me, after the car grift and the Mars grift didn't pan out.
I’m not saying the math checks out, but the argument is that you get full sun with no atmospheric losses 24/7, so you produce way more energy per panel, and you don’t need batteries, because the power production is consistent and predictable.
>These are clear engineering challenges that can be simulated, tested with earth analogs, and then rapidly iterated across design generations.
They can. But in Elons case, its going to be his style of sending failure after failure up in the space, getting something working part time, lying about it and exaggerating how good it is, and then making fun of others for not using his inferior product.
> Beyond aggressively optimistic timelines, I find it difficult to disagree with the premise. The aggressively optimistic timelines is also what makes it feasible to even attempt these things, where e.g. the amount of iteration required for Starship would have broken most other companies.
Instead of wasting huge amounts of land to farming, restaurants and transportation of food it would be so much better if everyone just had a Star-Trek style food replicator in their house.
None of the tech exists but fuck it. Why bother with realities of life?
I am raising 200 Trillion Dollars for AI Space FoodX. Who is in?
Yeah, I agree. A massive radar network, passive or active is the most likely possibility I have come across. You'd need a LOT of compute at each node to get the most out of the network. I found this video[1] to be a pretty convincing analysis of the absolute max capability you could expect, and it would indeed be impressive.
Local, state and international governments who wanted to crack down on AI could just arrest and execute the owners. None of whom plan on living in space anytime soon.
I assume because the Mars goal is as good as dead with what they're finding out about the complexities of building Starship that they can barely get it back down to this planet, never mind back from a second one.
This "space datacenters is more important than colonizing the universe" thing is just to deflect from what would be an inevitable failure because if they do this pivot, they can push out the timeline for that further than the original 2026 on Mars goal that they are about to wildly overshoot.
SpaceX perfected Falcon 9 reuse, they perfected Dragon, they perfected Starlink. Are you seriously going to bet they can't improve on the Space Shuttle? Which is what Starship/Super Heavy is, Space Shuttle idea implemented correctly.
"what they're finding out about the complexities of building Starship that they can barely get it back down to this planet, never mind back from a second one."
I would argue that complexities of building Starship are already a solved problem. Boca Chica built a lot more test units than there were (test or production) Apollos and the "factory for rockets, churning them out in regular intervals" part seems to be mastered. They even made three iterations of Raptor, and the third one looks really promising so far.
What is far from perfected is the heat shield and I agree that it is a critical problem.
"it, they can push out the timeline for that further than the original 2026 on Mars goal that they are about to wildly overshoot"
True, but this seems to be ubiquitous in space industry. I am old enough to remember talking about the US going back to the Moon in the 1990s. But the goal, declared by presidents (who have a lot more power at their hands to fulfill it) kept being pushed back and back, always into the next decade, then the next...
If you tolerated it from the government, you should probably tolerate the same from Musk, for the sake of consistency.
The more straightforward explanation is that it's a story that Elon (probably correctly) thinks will sound good to wall-street and enable him to take a ton of the publics money when SpaceX IPOs and gets added to the S&P for himself.
In other words good old fashioned plausibly deniable securities fraud.
As an investment narrative it was ideal to justify rolling Elon's unprofitable second tier AI company and the debt-ridden mess he made of Twitter into his highly successful space company, ensuring the investors in those get paid off by the SpaceX IPO...
Check out this video that goes into a very deep technical explanation about how the satellites can be used as a Synthetic Aperature Radar to build a realtime representation of the entire globe at meters of resolution: https://youtube.com/watch?v=jbp3kdJZ1_A
This really isn't true. Infrastructure build outs, space mining, the power generators and datacenters needed by the world's current best funded and most energetic sector all depend on more launches and larger cargo holds.
This is correct. The only problem that "data centers in space" solves is the problem of trying to scale a rocket company where the potential demand for rocket launches is simply not that big.
I don't know why space marines aren't a thing yet. The USA could put a rapid reaction force of Tier 1 Special Forces onto a space station and deploy them through atmospheric re-entry anywhere on Earth within 30 minutes.
I can only assume "too easy to track" is part of the logic.
Ditto for kinetic strikes. That was super hyped up.
The whole kinetic strike concept is 100% complete idiocy.
There is zero merit and zero gain from lobbing pole sized object at terrestrial targets, and I blame people having negative understanding of orbital dynamics for the whole concept getting popular in the first place.
Problems are:
1) You pay every single Joule of impact energy (and more!) in rocket fuel for getting the thing up there in the first place, which is an abysmal deal.
2) You can't actually "drop" anything from orbit once its there, you have to accelerate it while being trivially observable (and trackable) from earth by 30 year old radar technology.
3) You could literally do the same thing by launching purely kinetic ballistic missiles at targets. Non one ever does that for a reason-- its difficult, expensive and ineffective at the same time. Basically the only benefit is demonstrating that you could have delivered an actual nuclear payload in the same way.
The cost would be insane. And it wouldn’t be near 30min, you’d need lots of teams to reach this, driving the cost further up. Need to rotate them on a regular basis. And soldier without gravity for months at a time are definitely not fit for combat.
Kinetic strikes sure. It seems like space marines would be incredibly easy to shoot down. They would be on a ballistic re-entry and must slow down without extreme g-forces before they reach the ground.
> potential supply of rockets outstrips demand by a lot.
IDK I think plenty of people will want to go to space or even cut 24 hour flights across the world to 90 minutes.
As for experience - it's going to be pricy, but look how many multi-million dollar yachts are out there, parked, doing nothing. People do have money for such experiences.
I think for travel around Earth, supersonic passenger aircraft are more feasible than rockets. Even if we consider sonic booms, a lot of routes where rockets would be desirable are across uninhabited oceans.
Even as just an investor sell, its pretty smart. Basically nothing changes for SpaceX, they just keep trying to improve launch throughput. If that payload does end up being data centers, great, he's right. If it doesn't, oh well, he still has a hugely successful space program.
Not that I think we'll end up increasing our total launch payload throughput by over 3000x within 3 years like he suggests.
He is talking about distributed AI, with their own AI chip, ( may be they can work at higher temperatures allow it to slowly cool to space ? ) not space station size server farm. By that, energy requirements will also be reduce, my biggest concern is, if every one starts doing it, in no time, millions of satellites will be in the space
He's trying to position his commercial space launch business in front of the apparently unlimited firehouse of Ai capital. "IN SPACE" is worse in every way as a compute environment.
"The reasonable man adapts himself to the world; the unreasonable man persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man"
Right but it's famously difficult to cool things in space since you have basically zero convective or conductive heat transfer, so I don't think that makes a lot of sense.
Honestly, I think he's spot on, and I normally am not fond of Elon's public behavior. I mentioned in another thread that they're getting around having to ask permission to build datacenters by doing it in space. The entire thing is to avoid NIMBY stuff I'd bet.
It really depends on scale. There will be enough terrestrial vetoes that if what we build is 10-1000x what people are already halting through legal challenges…
I doubt it. Like, I hate to have to be the bearer of bad news, and maybe it’s my weird arctic anarchist soul, but, the old world order, the need for these companies to follow rules at least in spirit? That’s dead now. There are no laws but the laws of physics and the the laws others force your organization to follow.
I recognize that that is distressing to people, hell, it’s been obvious to me since I was at OWS in my 20s. But we are in a new world now and the old rules don’t apply. A company that has the backing of the government to launch their spacecraft will simply do it. You think Texas is going to stop them? Or Florida? Or even California? Of course not.
A lot changes in a world where you can plan things out with AI. A lot changes in a world with abundance. If we play our cards right we could have the culture, but that means letting go of the conservative yearning to put things back to how they were. The old world is 10 light years away now, it wasn’t as great as we remember it and it ain’t coming back.
And if I had to choose, I’d much rather have datacenters in orbit than one burning hydrocarbons loudly 2 blocks from my kids’ school.
It would be orders of magnitude cheaper to buy up islands and space in countries that don't care, and then find ways to connect them to the required infrastructure, than it would be to build them in space.
Hell, it would be cheaper to figure out how to build them on the ocean.
I think Prospera and its kerfuffles with Honduras put a definitive end to the "islands" idea.
Governments can change and the next one may be very unfriendly. "Rich gringos/infidels/colonizers are abusing our land sold for sordid money" is a very efficient populist call almost everywhere on the planet.
Admittedly it is not my field, but back of the envelope calculations in a sun synchronous orbit with the radiators pointed towards deep space seem pretty plausible with about 1.3 to 1.7 ratio of solar area to radiator area.
Like, it's not "great" but if you're not flying around the sun every 72 minutes or whatever and you can keep your panels sun on and radiate into deep space, the numbers aren't bananas.
You can build a completely self-powered (and water-free) datacenter in the middle of nowhere for far cheaper than the satellite version. The NIMBY factor isn't so powerful as to keep datacenters off entire continents. Going to space for that is very stupid.
Building datacenters in a medium where the main waste product (heat) is incredibly difficult to get rid of, there is zero opportunity for maintainance, and the fuel to get to site costs more than the site does. Makes perfect sense, spot on!
Does the fuel cost that much? Just doing some back of the napkin doesn't seem to bear that out. Looks like the fuel load is about $2M, and gets you 100 tons to orbit. I think an inference-optimized NVL72 GB300 rack costs around 3x that, >$6M. That thing eats about 150kw, call it 10 pallets of 30 500W solar panels. Each pallet's about a ton, and costs about $10k. Let's be conservative and say the radiator's about the same weight. In reality, they're not going to be using commercial panels with heavy glass facing designed to resist hail, so should be better than this.
But anyway, conservatively, about 20 tons each, it seems like you could fit at least 5 of these per starship, assuming it's weight and not volume limited. Doesn't seem like fuel's a prohibitive portion of the cost here. But if they can't get it to their no-refurb-between-launches target, then that might be a significant part of the cost.
This person made self driving cars work years after they’d been written off, made reusable rockets and has people with locked-in syndrome speaking to their families. Why do you think he wouldn’t be sane?
in the very broad shoulders of long term, he's probably right.. its why the concept of a dysonsphere is around.
you can get uninterrupted 24/7 free energy.
but yeah, the tech is a long way away.
*Edit: lol
My estimate is that within 2 to 3 years, the lowest cost way to generate AI compute will be in space.
Dyson spheres (or the more plausible sounding Dyson swarms) are not an actual physically possible thing, they're just a nice sounding sci-fi trope, like teleporters or replicators.
Freeman Dyson invented the concept as a joke against SETI, especially designing it to sound quasi-plausible.
In reality, there is no way to create a stable structure of this size, it would be like trying to balance a building on the top of a pinhead - except the pinhead is a chaotic, unpredictable star. And the amount of energy required to displace multiple planets worth of mass, manufacture some amount of it into complex satellites, and then displace this amount again to a "stable" Solar orbit simply doesn't exist in the Solar system, on any plausible time scale (it would take many thousands if not millions of years worth of solar power to do so).
I agree - long term I can see highly distributed compute ( like tons of small satellites ) becoming a cool space thing. And eventually a ringworld like thing or dysonsphere
Pretty normal for Elon: big promises, generate interest and funding, then fail to deliver. But by that time, he’s got his trillion-dollar paycheck and is working on his next scheme.
US history is more complicated than that, and aside from those four years of hot war, more ambiguous.
Henry Ford was a big Nazi sympathizer, and the Apollo program was led by an actual card-carrying Nazi engineer with a history of overseeing slave labor in a concentration camp.
Which is not meant to defend Nazis, just correct the myth that the US was once somehow morally pure in this regard.
I am sick of living in this world where the richest scam artist can get richer and richer and richer with lies and lies and lies and empty promises and there is no SEC, no anything to stop him.
What makes Elon complicated is that he is not just a scam artist. He has an eye for talented people that do good engineering while working for him, in spite of his personal flaws.
For all the lies, bad behavior, and broken promises, SpaceX's achievements and reliability record is still incredible, X/Twitter hasn't crashed and burned after all the layoffs and drama, and Tesla (until recently due to his meddling) had a lock on the leading the car industry's direction & doing a lot to drive practical electrification globally.
> What makes Elon complicated is that he is not just a scam artist. He has an eye for talented people that do good engineering while working for him, in spite of his personal flaws.
Solar panels in space are 5 time more expensive to build than on earth (not talking into account launching them to space), while being 5 to 10 time more efficient. They also degrade 5 to 10 time faster, not accounting for solar flares. Deorbiting solar panels (and satellites) is also a huge environmental issue, as I dislike heavy metal in my food (and you should too). It isn't a real issue yet because we didn't send enough up there for the quantities to be an issue, but idiots seems persuaded we should increase the quantity of heavy metal sent in orbit without fixing this issue first.
Even assuming "that's it", why not just install it in e.g. Morocco instead? It's not like space is any easier to access than the Sahara, and saving a few dozen ms of network latency isn't particularly valuable when your TTFT is measured in tenths of a second. Sure, sun synchronous orbits are a thing, but you also need more expensive panels and the comparative efficiency will decline over time vs land-based hardware as your chips fail (wasting that part of the resource budget) and the land hardware gets upgraded.
The number of political actors that can stop you from building in Morocco (or confiscate/damage your invested capital once you deploy it) are numerous. The number that can do so in space? Maybe a half dozen. We’re already seeing states and municipalities in the US moving to ban data centers and the energy infrastructure needed to power them. Building in space faces no such procedural roadblocks.
The economics still seem like an open question, but if the demand for compute is high enough, space based data centers might be the only option
If AGI were to happen, or if AI became a trillions-of-dollars-generating industry, you wouldn't want to have your data-centers which might be the most valuable thing on Earth be located in a foreign country. All this investment in infrastructure is not purely based on where the industry is now, but predicated by where those who are bullish about it think it will be in 5-10 years.
I think Elons version is totally crazy but the idea of edge computer (maybe for latency or something) on each satellite above your head could make sense. It could even integrate well with larger terrestrial datacenters (like your example of Morocco) depending on use case
Elon explained the logic at length in an interview: Cheaper != Available.
The availability of power is the constraint almost everywhere, no matter how much money you throw at it.
Gas turbine production has a many-year backlog. Everybody that can make the single-crystal superalloy turbine blades is fully booked for most of a decade and can't expand capacity for years (at least).
Meanwhile, putting a slightly larger solar panel onto a satellite is a trivial engineering excercise and has no blockers in 2026.
Disclaimer: Personally, I suspect all this AI-in-space "talk" from Elon is just cheap marketing to boost the IPO of xAI.
You've got to give him credit though. His caustic managerial style seems to have borne fruit despite his lack of engineering or technical skills. He has been supremely effective at defining a vision(however delusional) and attracting funding.
Will we get to Mars soon? Hell no. But we may end up with a world-leading launch provider based in the US and that's a clear win for the country.
> despite his lack of engineering or technical skills
At least he has B.Sc. in physics and got admitted into Stanford.
I think what Elon says is better explained not as a promise what would happen, but rather as a goal which they're going to aspire to. It kinda supports the idea "we're in business of converting impossible into late". If Elon will start offering more "realistic" schedules, the pace of SpaceX will slow down, perhaps considerably. So, yes, it's "Elon time", which historically isn't particularly precise, but still useful.
You absolutely do not, under any circumstances, have to give him credit.
Chronic over-promise, underdelivery.
Where was the nearly 3T of fraud he said he'd uncover in the US government, again? Was that a clear win for the country?
But hey at least he's effective at getting people to give him money, I guess, which is an indistinguishable "skill" from that of someone who is able to convince people to buy an online course on how to make money online.
He just does it at a bigger scale so people are quick to suck him off. How we are still falling into the "money = smart/competent" trap in <<current year>> is beyond me.
People fuck up. I fucked up three things yesterday. Fortunately I am not as known as Musk, so no one tears into me at Hacker News and my fuckups remain hidden.
Nevertheless...
"underdelivery."
Both Falcon and Starlink are quite major improvements over previous status quo. It is not just a question of having a nice WiFi during your flight. If you are interested in some very practical consequences, look at the Russo-Ukrainian war and the role Starlink plays there.
Don’t buy into the 2010’s Tony stark persona. His momentum is clearly slowing because he can’t put his politics and rather fucked social values behind business sense.
I have immense appreciation for what SpaceX has done for humanity. I’m not being dramatic. Reusable rockets alone is an incredible achievement. But he’s lost the plot. He needs to drop his right wing bullshit and stardom chasing if he wants to be taken seriously again. The dude won’t even acknowledge his own kid because of his politics. I will never trust someone who makes that decision, personally. His judgment is beyond clouded.
The Elon bros will be mad but whatever. One day he’ll maybe remember why folks liked him. Hitching his wagon to Trump was a dumb move.
Not getting invited to the EV summit would have pissed me off if I were in his place. The Trump thing; it sounds like the government was going to go after him for various violations, and hitching his wagon to Trump was his way of getting out of that jam.
Beyond aggressively optimistic timelines, I find it difficult to disagree with the premise. The aggressively optimistic timelines is also what makes it feasible to even attempt these things, where e.g. the amount of iteration required for Starship would have broken most other companies.
> In the long term, space-based AI is obviously the only way to scale.
In the long term - all mass and energy available is outside of Earth - what is here is not even a rounding error. If you wish to continue scaling compute it then becomes a question of time before you'd want to go off planet. Personally I'm quite keen to see near term space based compute explored, as it could end up becoming a much better trade-off than allocating ever more ground to power and operate terrestrial compute which directly conflict with the biosphere.
SpaceX started the Starlink design phase in 2015 - started launching Starlink satellites in 2019 - and they now have the most dominant satellite constellation ever deployed by a large factor. They have their own launch systems, launch sites, satellite bus, communication stack - both in-house designed and built.
What is really going to be that difficult with space-based compute? Radiation hardening and cooling? These are clear engineering challenges that can be simulated, tested with earth analogs, and then rapidly iterated across design generations. There's napkin math all over the internet on this, but it really seems like small challenges compared to the other engineering SpaceX have already sorted.
Beyond radiation / cooling / servicing - it seems like the biggest hurdle is to crack the scaling of designing / scaling the necessary amount of compute they will need to scale space based compute according to the laid out plans.
> In the long term
In case anyone is wondering how Tesla’s stock price remain wildly detached from its business reality, keep these four words in mind. If you can convince people that anything about you and your business has to be evaluated on a literally astronomical timescale, you can justify any valuation you desire, because your believers will give you infinite time to realize their investment returns. It has nothing to do with business. They are selling you a vision — which can also come in a pill form, labeled "salvia" and sold at gas stations.
I still see people say the cybertruck is built for mars environments, conveniently ignoring the vast technological and economical barriers stopping us from driving commercially produced vehicles on mars. This space data center thing is the same deal. It doesn't matter how long it will take to solve the technical issues with cooling, radiation, maintenance. It doesn't matter if it will make economical sense or not. It doesn't matter if spacex will be the one to actually do it. You just have to believe, and give them some time — a lot of time, so much time that a monkey can type out Hamlet and type it out again backwards.
See also the buffoonery coming out of Bay Area "effective altruist" and "longtermism" communities.
I fully agree on the reality distortions and valuation chaos surrounding Tesla. This does also follow from the company being very volatile and chaotic, which becomes harder to price. How do you accurately price in e.g. Optimus - it seems really hard to tell at this point - which I guess is also one of the motivators for these strategies.
However, in this particular trajectory, SpaceX did build the rockets and did build Starlink which is now the best global-scale wireless communication network for many use-cases. Stretching this trajectory to scale up the technology to facilitate in-space computing is vastly more grounded than Shakespearean monkeys.
All share price fundamentals are based on the long term. Short term trading is why some shares are very briefly high or low.
5 replies →
> believers
If it were only retail investors, your assumptions could make sense.
However plenty of the share ownership is institutional investors. Most of them care a bit more about fundamentals. (I'm ignoring passive investors just using indexes).
3 replies →
I love how in today’s quarter-based world a decade is now a “literally astronomical timescale”.
Chipping in here. There's a lot of speculation on this subject matter, all of which entirely wrong.
Technical concerns aside, the main risk is financial. Success is based on the premise that we need this enough that the costs are justified but the costs are going to be much higher. That is totally unproven on any financial modelling scenario I've seen. In fact there's likely no actual ROI on what has been spent so far and no qualification of demand. With geopolitical problems on the table, no one is going to fund this.
The idea is completely dead before the first node leaves the planet.
Not only that! First mover advantage is mostly bunk. Even if none of what you say was true, valuing this on a horizon far enough out to solve all the technical problems gives a lot of space for competitors to emerge. So even if the idea was both technically and financially workable, there is no guarantee that an investment in SpaceX specifically would be the right move over the same evaluation term as the tech will take to play out. It only makes sense from a speculative angle because you know that if they present any more concrete excuse to believe that the tech works, the price will temporarily go through the roof.
Yes, beyond the three things that are the hard parts it's easy.
As problems go, radiation and cooling seem to have relatively low dimensionality compared to the other problems. It seems to be mostly a question of optimizing within the dimensions of dissipation / structure / deployment / service / cost / weight. When all is said and done, the cooling solution will end up being a module that can deal with some power dissipation, cost X amount, weight Y amount, have structural interface Z. This seems like something a relatively low number of engineers can iterate on largely isolated from other concerns. SpaceX does have 5000+ of them.
Comparing this to scaling the production of compute where they try to work outside the bounds of ASML (~40k employees) and TSMC (~80k+ employees), and where there is a huge number of degrees of freedom in many, many layers of the stack that have complicated interactions.
With radiation and cooling, SpaceX also has plenty of experience with both already given that they've had to solve this on existing satellites. Overall, Terafab just seems like a far harder challenge, and where I'd be more wary on timelines.
10 replies →
Just go to Google Maps. Just do it. Type in Arizona, or new mexico, texas whatever. Do you see all this brownish stuff? Yeah thats just empty land with A LOT of solar.
So in the long term, what do you think is cheaper and easier to maintain, upgrade, handle etc.?
A Space operation on which you need to send compute hardware constantly upwoards or a fiber connection to some more 'remote/dessert' like area which has a lot of energy available?
Starlink is not a game changer at all. It has 8-10 Million customers, from which plenty of peopple just use it for holidays, or upping there already existing internet line or because its faster to deploy than a cable.
Our planet is already very well connected. Putting lines in the ground is necessary anyway because you still need energy / powerlines.
Of course this can be done, thats NOT the question. The only question is, if its worth it and its not.
Sending some servers up in space is margins more expensive than sending some servers on trucks (you need anyway) to another earth location.
> Just go to Google Maps. Just do it. Type in Arizona, or new mexico, texas whatever. Do you see all this brownish stuff? Yeah thats just empty land with A LOT of solar.
'Brownish stuff', known more generally as natural ecosystems.
> So in the long term, what do you think is cheaper and easier to maintain, upgrade, handle etc.?
How long a term does your imagination stretch to? Are you really arguing that once provisioning, cooling, automated scaling in space, and off-planet mining are all solved problems, that shitting on our planet will still be the cheapest most maintainable option?
2 replies →
Compute in space is doable, we already send plenty of computers up there, technologically, it is not even a challenge. It just doesn't make sense economically, even with Starship, it is making things harder for no good reason.
Starlink is different, it makes sense. Covering the entire Earth, including the oceans with cell towers for global internet connectivity is harder than having a satellite constellation. The opposite situation from datacenters.
It’s trading political difficulty for engineering difficulty.
There are now quite a few politicians running on a platform of banning data centre construction projects.
1 reply →
how does the economics compare at scale building a persistent 24/7 supply of 1,000 gw isnt easy or cheap
1 reply →
It’s like saying within 2-3 years the sun will go out.
Almost correct, yes.
> In the long term - all mass and energy available is outside of Earth - what is here is not even a rounding error.
Define “long term”. Nuclear energy is practically unlimited, plus fusion (if it ever works).
I still don't get why so many people who watch/read scifi conclude "yes this is a legitimate and realistic vision of the future, and not Wizards and Dragons but with Technology"
Going to mars or staying on the moon will be a Darwin Award-level adventure.
> all mass and energy available is outside of Earth
Manufacturing capabilities are quite lacking, though, in the short and medium terms, so this doesn't seem all that relevant.
Maybe a self-contained, modular solar panel / radiator / compute unit could be built, but it will be manufactured on Earth. (Where the fabs are.)
And it still seems easier to put solar panels and batteries near the data centers that SpaceX is already building on Earth.
> What is really going to be that difficult with space-based compute?
Stopping some random rogue nation blowing it up.
Honestly, I don't see (sea?) it. Every advantage of space are found in oceans/seas, especially if we use dead zones where aquatic life is already dead. The cooling is cheaper, tide+wind+solar is cheaper than space solar (I know someone who worked on a lens to observe the sun, the satellite was launched but due to being cheap on the solar panels, the sunlight and radiation chipped away the coating that found itself attracted to the most massive object in the area, the lens). Anti-corrosion is cheaper than light radiation protection, and servicing is way easier and cheaper.
In the long term, the biggest problem is that space data centers are very hard to defend against missiles.
But if solar panel is significantly cheaper and latency doesn't matter you can have servers in any part of the world. Even if they are not up 50% of the time due to limited battery it would still be cheaper.
whose wasting 50% of their silicon production capacity to save on power. doesnt make alot of sense even with money silicon is hard to buy at scale
1 reply →
biosphere interference from ground infrastructure? any idea the ground infrastructure it requires to support space based compute operations? i have a feeling that is comparable if not more impactful
you also shrug off cooling. this is not a solved problem in any way. its not even approachable as of yet. the vast size of the radiators will be hilarious regardless.
you ignore power generation. solar is not an option. so we also need nuclear reactors for these orbital data centers. thats cool spacex can just branch out into nuclear too! love the idea of unmanned nuclear orbiting behemoths.
speaking of orbital.. what is their orbit? do they go out to Lagrange points? hilariously far? or do they stay close? hilariously fuel intensive to stay out of the atmosphere for such massive structures?
but hey, maybe we distribute spaceX-AI gpu's across starlinks. a couple solar panels and a tesla battery per gpu. all launched there by spacex
'all mass and energy available is outside of earth' Yeah, and out of range for compute data connections too.
I don't agree with the feasibility or ANY sort of practicality to this whatsoever. Im all for going for it, but I wish everyone could just admit that we're doing it because it's cool, not because it's useful. I get why Elon wont say that, but not us.
Your feelings are obviously your own, but a Starlink terminal isn't that big and can transfer quite a lot of tokens.
Every single satellite has sufficient cooling for its power production, otherwise they would be frying. Waste heat from a GPU is not materially different from waste heat from an amplifier. That's not cooling entire racks, but I don't think anybody talks about putting entire racks in space anymore.
I'm very much pro nuclear, but a solar cell in a sun synchronous orbit is pretty great too and eliminates most battery requirements
I very much doubt the economics of this makes sense, but I don't think a lot of your criticism is valid.
6 replies →
You can put computers underground. Cheaper than launching into space.
Why put them in space? Power? We have that on earth.
There are literally enormous problems powering AI data centers on the Earth right now. No, we don't have the power on Earth.
In terms of launch cost, Starship makes launch cost negligible. Some estimates are that it will cost less to launch a tonne to orbit, than to ship across the US by train.
Even if this figure is slightly low, that has nothing compared to the cost of real estate, construction costs, all of the building codes required to build a data center on Earth. These things all still apply underground, and underground is going to require additional shoring and structural engineering, to ensure that the structure is not crushed, damaged, and so forth.
6 replies →
All of this may be true but the scale that Musk is talking about would require an immense amount of solar panels -- and if he has the means to produce so many solar panels why not use them to solve our climate and energy crisis on Earth?
Seems more like a grift to me, after the car grift and the Mars grift didn't pan out.
I’m not saying the math checks out, but the argument is that you get full sun with no atmospheric losses 24/7, so you produce way more energy per panel, and you don’t need batteries, because the power production is consistent and predictable.
3 replies →
>These are clear engineering challenges that can be simulated, tested with earth analogs, and then rapidly iterated across design generations.
They can. But in Elons case, its going to be his style of sending failure after failure up in the space, getting something working part time, lying about it and exaggerating how good it is, and then making fun of others for not using his inferior product.
Pretty much like everything else he has done.
> Beyond aggressively optimistic timelines, I find it difficult to disagree with the premise. The aggressively optimistic timelines is also what makes it feasible to even attempt these things, where e.g. the amount of iteration required for Starship would have broken most other companies.
Instead of wasting huge amounts of land to farming, restaurants and transportation of food it would be so much better if everyone just had a Star-Trek style food replicator in their house.
None of the tech exists but fuck it. Why bother with realities of life?
I am raising 200 Trillion Dollars for AI Space FoodX. Who is in?
"aggressively optimistic timelines is also what makes it feasible to even attempt these things"
yawn, people keep making this excuse on behalf of the South African investor with poor technical expertise.
> Yes, Elon is very sane.
tbf, a 'sane' person wouldn't have started a rocket company and an ev company, at the same time, in a recession.
He has never been sane. and that has made all the difference.
He didn’t found Tesla, he bought it then called himself founder.
That's pointless nitpicking, when he bought it it was so early you can easily say he made everything of it.
2 replies →
Tesla and SpaceX were both founded outside of recessions about 1 year apart, between the early 2000's recession and the great recession.
You are saying he got lucky? That doesn’t necessarily mean he will continue to be lucky.
They mean, to attempt things that seem very difficult to normies like you and me, and be successful at it, one needs to be insane.
2 replies →
I'm basically assuming that "space-based data centers" are some Glomar Explorer-style cover for something else.
Yeah, I agree. A massive radar network, passive or active is the most likely possibility I have come across. You'd need a LOT of compute at each node to get the most out of the network. I found this video[1] to be a pretty convincing analysis of the absolute max capability you could expect, and it would indeed be impressive.
[1] https://www.youtube.com/watch?v=jbp3kdJZ1_A
Well, that likely already exists as Starshield - not to mention all the pubmic SAR sats everyone has by this point.
It's putting AI processing out of the reach of hostile local, state, and international governments. Does it need to be a cover?
> It's putting AI processing out of the reach of hostile local, state, and international governments
It isn't... the hostile local government can seize the ssh keys you use to control it and take it over just fine.
The hostile international non-local super power just gained a new ability to jam communications or destroy it with a bit of deniability too.
3 replies →
Local, state and international governments who wanted to crack down on AI could just arrest and execute the owners. None of whom plan on living in space anytime soon.
>hostile ... international governments
you mean other than China, russia, NK and Iran?
9 replies →
A cover is going to have a plausible enough sounding justification that you’ll believe and defend
I assume because the Mars goal is as good as dead with what they're finding out about the complexities of building Starship that they can barely get it back down to this planet, never mind back from a second one.
This "space datacenters is more important than colonizing the universe" thing is just to deflect from what would be an inevitable failure because if they do this pivot, they can push out the timeline for that further than the original 2026 on Mars goal that they are about to wildly overshoot.
SpaceX perfected Falcon 9 reuse, they perfected Dragon, they perfected Starlink. Are you seriously going to bet they can't improve on the Space Shuttle? Which is what Starship/Super Heavy is, Space Shuttle idea implemented correctly.
"what they're finding out about the complexities of building Starship that they can barely get it back down to this planet, never mind back from a second one."
I would argue that complexities of building Starship are already a solved problem. Boca Chica built a lot more test units than there were (test or production) Apollos and the "factory for rockets, churning them out in regular intervals" part seems to be mastered. They even made three iterations of Raptor, and the third one looks really promising so far.
What is far from perfected is the heat shield and I agree that it is a critical problem.
"it, they can push out the timeline for that further than the original 2026 on Mars goal that they are about to wildly overshoot"
True, but this seems to be ubiquitous in space industry. I am old enough to remember talking about the US going back to the Moon in the 1990s. But the goal, declared by presidents (who have a lot more power at their hands to fulfill it) kept being pushed back and back, always into the next decade, then the next...
If you tolerated it from the government, you should probably tolerate the same from Musk, for the sake of consistency.
> they can barely get it back down to this planet
Being the first rocket in history where both parts reached the ground ready to land is a pretty good start.
And if Starship can't land then any space datacenters are just as or even more unlikely, so that explaition makes no sense what so ever.
The math works out if you project certain macro trends out a sufficient amount of time.
I think if fusion is real, it might not be so advantageous until space mining is a thing.
Maybe coverage is directed outward from Earth ? It could be quite an upgrade to the "UFO" TV series SID (Space Intruder Detector).
The more straightforward explanation is that it's a story that Elon (probably correctly) thinks will sound good to wall-street and enable him to take a ton of the publics money when SpaceX IPOs and gets added to the S&P for himself.
In other words good old fashioned plausibly deniable securities fraud.
As an investment narrative it was ideal to justify rolling Elon's unprofitable second tier AI company and the debt-ridden mess he made of Twitter into his highly successful space company, ensuring the investors in those get paid off by the SpaceX IPO...
They'll put up thousands more starlinks and track every mobile device on the planet simultaneously, might as well have a homing beacon in your pocket.
Check out this video that goes into a very deep technical explanation about how the satellites can be used as a Synthetic Aperature Radar to build a realtime representation of the entire globe at meters of resolution: https://youtube.com/watch?v=jbp3kdJZ1_A
1 reply →
One of the underrated topics about space right now is the potential supply of rockets outstrips demand by a lot.
We're simply out things we can profitably send to space so SpaceX and others are trying to come up with ideas to induce demand.
My understanding is that Starlink mostly grew out of the same need to justify scaling up rocket production.
Before LEO internet constellations, even the leading nations had just ~20-25 launches per year each, and a good chunk of those were for ISS services.
Other than the occasional GNSS, weather, scientific, broadcast and surveillance satellite, there's not all that much worth sending into space.
This really isn't true. Infrastructure build outs, space mining, the power generators and datacenters needed by the world's current best funded and most energetic sector all depend on more launches and larger cargo holds.
I, and I'm not alone, would pay a giant pile of money to go into space for a holiday.
3 replies →
This is correct. The only problem that "data centers in space" solves is the problem of trying to scale a rocket company where the potential demand for rocket launches is simply not that big.
I don't know why space marines aren't a thing yet. The USA could put a rapid reaction force of Tier 1 Special Forces onto a space station and deploy them through atmospheric re-entry anywhere on Earth within 30 minutes.
I can only assume "too easy to track" is part of the logic.
Ditto for kinetic strikes. That was super hyped up.
The whole kinetic strike concept is 100% complete idiocy.
There is zero merit and zero gain from lobbing pole sized object at terrestrial targets, and I blame people having negative understanding of orbital dynamics for the whole concept getting popular in the first place.
Problems are:
1) You pay every single Joule of impact energy (and more!) in rocket fuel for getting the thing up there in the first place, which is an abysmal deal.
2) You can't actually "drop" anything from orbit once its there, you have to accelerate it while being trivially observable (and trackable) from earth by 30 year old radar technology.
3) You could literally do the same thing by launching purely kinetic ballistic missiles at targets. Non one ever does that for a reason-- its difficult, expensive and ineffective at the same time. Basically the only benefit is demonstrating that you could have delivered an actual nuclear payload in the same way.
The cost would be insane. And it wouldn’t be near 30min, you’d need lots of teams to reach this, driving the cost further up. Need to rotate them on a regular basis. And soldier without gravity for months at a time are definitely not fit for combat.
2 replies →
> Ditto for kinetic strikes. That was super hyped up.
Dropping steel rods from orbit didn't seem so crazy. But I've never seen a detailed evaluation of the idea.
Kinetic strikes sure. It seems like space marines would be incredibly easy to shoot down. They would be on a ballistic re-entry and must slow down without extreme g-forces before they reach the ground.
> potential supply of rockets outstrips demand by a lot.
IDK I think plenty of people will want to go to space or even cut 24 hour flights across the world to 90 minutes.
As for experience - it's going to be pricy, but look how many multi-million dollar yachts are out there, parked, doing nothing. People do have money for such experiences.
I think for travel around Earth, supersonic passenger aircraft are more feasible than rockets. Even if we consider sonic booms, a lot of routes where rockets would be desirable are across uninhabited oceans.
2 replies →
Even as just an investor sell, its pretty smart. Basically nothing changes for SpaceX, they just keep trying to improve launch throughput. If that payload does end up being data centers, great, he's right. If it doesn't, oh well, he still has a hugely successful space program.
Not that I think we'll end up increasing our total launch payload throughput by over 3000x within 3 years like he suggests.
He is talking about distributed AI, with their own AI chip, ( may be they can work at higher temperatures allow it to slowly cool to space ? ) not space station size server farm. By that, energy requirements will also be reduce, my biggest concern is, if every one starts doing it, in no time, millions of satellites will be in the space
He's trying to position his commercial space launch business in front of the apparently unlimited firehouse of Ai capital. "IN SPACE" is worse in every way as a compute environment.
"The reasonable man adapts himself to the world; the unreasonable man persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man"
-George Bernard Shaw
He is talking about energy costs.
Right but it's famously difficult to cool things in space since you have basically zero convective or conductive heat transfer, so I don't think that makes a lot of sense.
Grifter gonna grift
Still waiting on these 2014 fully self driving cars, back when Uber promised to buy every single model S they could produce.
Now he's late on his mars promises so he's pushing some new bullshit timeline.
Honestly, I think he's spot on, and I normally am not fond of Elon's public behavior. I mentioned in another thread that they're getting around having to ask permission to build datacenters by doing it in space. The entire thing is to avoid NIMBY stuff I'd bet.
It really depends on scale. There will be enough terrestrial vetoes that if what we build is 10-1000x what people are already halting through legal challenges…
I doubt it. Like, I hate to have to be the bearer of bad news, and maybe it’s my weird arctic anarchist soul, but, the old world order, the need for these companies to follow rules at least in spirit? That’s dead now. There are no laws but the laws of physics and the the laws others force your organization to follow.
I recognize that that is distressing to people, hell, it’s been obvious to me since I was at OWS in my 20s. But we are in a new world now and the old rules don’t apply. A company that has the backing of the government to launch their spacecraft will simply do it. You think Texas is going to stop them? Or Florida? Or even California? Of course not.
A lot changes in a world where you can plan things out with AI. A lot changes in a world with abundance. If we play our cards right we could have the culture, but that means letting go of the conservative yearning to put things back to how they were. The old world is 10 light years away now, it wasn’t as great as we remember it and it ain’t coming back.
And if I had to choose, I’d much rather have datacenters in orbit than one burning hydrocarbons loudly 2 blocks from my kids’ school.
8 replies →
It would be orders of magnitude cheaper to buy up islands and space in countries that don't care, and then find ways to connect them to the required infrastructure, than it would be to build them in space.
Hell, it would be cheaper to figure out how to build them on the ocean.
I think Prospera and its kerfuffles with Honduras put a definitive end to the "islands" idea.
Governments can change and the next one may be very unfriendly. "Rich gringos/infidels/colonizers are abusing our land sold for sordid money" is a very efficient populist call almost everywhere on the planet.
How do you cool them?
https://www.chaotropy.com/why-jeff-bezos-is-probably-wrong-p...
Admittedly it is not my field, but back of the envelope calculations in a sun synchronous orbit with the radiators pointed towards deep space seem pretty plausible with about 1.3 to 1.7 ratio of solar area to radiator area.
Like, it's not "great" but if you're not flying around the sun every 72 minutes or whatever and you can keep your panels sun on and radiate into deep space, the numbers aren't bananas.
4 replies →
https://youtu.be/FlQYU3m1e80
"Is It Really Impossible To Cool A Datacenter In Space?" - Scott Manley
tl;dr -> not impossible.
1 reply →
You can build a completely self-powered (and water-free) datacenter in the middle of nowhere for far cheaper than the satellite version. The NIMBY factor isn't so powerful as to keep datacenters off entire continents. Going to space for that is very stupid.
You still have to ask someone permission on the ground. It’s not about the money.
1 reply →
now scale it ti 1 terrawatt
Building datacenters in a medium where the main waste product (heat) is incredibly difficult to get rid of, there is zero opportunity for maintainance, and the fuel to get to site costs more than the site does. Makes perfect sense, spot on!
Does the fuel cost that much? Just doing some back of the napkin doesn't seem to bear that out. Looks like the fuel load is about $2M, and gets you 100 tons to orbit. I think an inference-optimized NVL72 GB300 rack costs around 3x that, >$6M. That thing eats about 150kw, call it 10 pallets of 30 500W solar panels. Each pallet's about a ton, and costs about $10k. Let's be conservative and say the radiator's about the same weight. In reality, they're not going to be using commercial panels with heavy glass facing designed to resist hail, so should be better than this.
But anyway, conservatively, about 20 tons each, it seems like you could fit at least 5 of these per starship, assuming it's weight and not volume limited. Doesn't seem like fuel's a prohibitive portion of the cost here. But if they can't get it to their no-refurb-between-launches target, then that might be a significant part of the cost.
The model 3 was Elons last great idea (if it was even his). Since then, he has been wrong pretty much about everything.
Its to the point where anything he says is guaranteed to be wrong just on the merit that its coming out of his mouth.
Starlink?
1 reply →
This person made self driving cars work years after they’d been written off, made reusable rockets and has people with locked-in syndrome speaking to their families. Why do you think he wouldn’t be sane?
We had all those things without Elon Musk and the alternatives do it better.
long term is doing a lot of heavy lifting here...
in the very broad shoulders of long term, he's probably right.. its why the concept of a dysonsphere is around. you can get uninterrupted 24/7 free energy.
but yeah, the tech is a long way away.
*Edit: lol My estimate is that within 2 to 3 years, the lowest cost way to generate AI compute will be in space.
i think 2-3 years is a very unlikely outcome.
Dyson spheres (or the more plausible sounding Dyson swarms) are not an actual physically possible thing, they're just a nice sounding sci-fi trope, like teleporters or replicators.
Freeman Dyson invented the concept as a joke against SETI, especially designing it to sound quasi-plausible.
In reality, there is no way to create a stable structure of this size, it would be like trying to balance a building on the top of a pinhead - except the pinhead is a chaotic, unpredictable star. And the amount of energy required to displace multiple planets worth of mass, manufacture some amount of it into complex satellites, and then displace this amount again to a "stable" Solar orbit simply doesn't exist in the Solar system, on any plausible time scale (it would take many thousands if not millions of years worth of solar power to do so).
His timelines always assume absolutely everything goes right, and there are no legal or regulatory problems.
I agree - long term I can see highly distributed compute ( like tons of small satellites ) becoming a cool space thing. And eventually a ringworld like thing or dysonsphere
But 2 to 3 years?! Seems crazy
Pretty normal for Elon: big promises, generate interest and funding, then fail to deliver. But by that time, he’s got his trillion-dollar paycheck and is working on his next scheme.
We used to eliminate Nazis, not invest in them.
"We used to eliminate Nazis, not invest in them."
US history is more complicated than that, and aside from those four years of hot war, more ambiguous.
Henry Ford was a big Nazi sympathizer, and the Apollo program was led by an actual card-carrying Nazi engineer with a history of overseeing slave labor in a concentration camp.
Which is not meant to defend Nazis, just correct the myth that the US was once somehow morally pure in this regard.
elon musk isnt a nazi
I am sick of living in this world where the richest scam artist can get richer and richer and richer with lies and lies and lies and empty promises and there is no SEC, no anything to stop him.
What makes Elon complicated is that he is not just a scam artist. He has an eye for talented people that do good engineering while working for him, in spite of his personal flaws.
For all the lies, bad behavior, and broken promises, SpaceX's achievements and reliability record is still incredible, X/Twitter hasn't crashed and burned after all the layoffs and drama, and Tesla (until recently due to his meddling) had a lock on the leading the car industry's direction & doing a lot to drive practical electrification globally.
> What makes Elon complicated is that he is not just a scam artist. He has an eye for talented people that do good engineering while working for him, in spite of his personal flaws.
Elon must have read Rich Dad, Poor Dad.
Elon musk is not a scam artist.
[dead]
[flagged]
Solar panels in space are 5 time more expensive to build than on earth (not talking into account launching them to space), while being 5 to 10 time more efficient. They also degrade 5 to 10 time faster, not accounting for solar flares. Deorbiting solar panels (and satellites) is also a huge environmental issue, as I dislike heavy metal in my food (and you should too). It isn't a real issue yet because we didn't send enough up there for the quantities to be an issue, but idiots seems persuaded we should increase the quantity of heavy metal sent in orbit without fixing this issue first.
Even assuming "that's it", why not just install it in e.g. Morocco instead? It's not like space is any easier to access than the Sahara, and saving a few dozen ms of network latency isn't particularly valuable when your TTFT is measured in tenths of a second. Sure, sun synchronous orbits are a thing, but you also need more expensive panels and the comparative efficiency will decline over time vs land-based hardware as your chips fail (wasting that part of the resource budget) and the land hardware gets upgraded.
>why not just install it in e.g. Morocco instead
The number of political actors that can stop you from building in Morocco (or confiscate/damage your invested capital once you deploy it) are numerous. The number that can do so in space? Maybe a half dozen. We’re already seeing states and municipalities in the US moving to ban data centers and the energy infrastructure needed to power them. Building in space faces no such procedural roadblocks.
The economics still seem like an open question, but if the demand for compute is high enough, space based data centers might be the only option
1 reply →
If AGI were to happen, or if AI became a trillions-of-dollars-generating industry, you wouldn't want to have your data-centers which might be the most valuable thing on Earth be located in a foreign country. All this investment in infrastructure is not purely based on where the industry is now, but predicated by where those who are bullish about it think it will be in 5-10 years.
1 reply →
You need permission to do that in Morocco
I think Elons version is totally crazy but the idea of edge computer (maybe for latency or something) on each satellite above your head could make sense. It could even integrate well with larger terrestrial datacenters (like your example of Morocco) depending on use case
> "We'll need thousands of them!
> Yes, they know.
> Starlink is already planned for a scale of tens of thousands of satellites.
Meanwhile Google installed that many TPUs yesterday afternoon. The idea is still stupid.
So it'll be more like Hertzner[1] in space. Each node just doing its own thing.
Not sure about the cost perspective but, at least that makes more sense than a giant brick floating around.
[1]: https://lafibre.info/hetzner/over-200-000-servers-in-one-pla...
issue is land based will still be cheaper. there are lot of cool things we can do in space, i’m not convinced putting data center is one of them.
Elon explained the logic at length in an interview: Cheaper != Available.
The availability of power is the constraint almost everywhere, no matter how much money you throw at it.
Gas turbine production has a many-year backlog. Everybody that can make the single-crystal superalloy turbine blades is fully booked for most of a decade and can't expand capacity for years (at least).
Meanwhile, putting a slightly larger solar panel onto a satellite is a trivial engineering excercise and has no blockers in 2026.
Disclaimer: Personally, I suspect all this AI-in-space "talk" from Elon is just cheap marketing to boost the IPO of xAI.
10 replies →
You've got to give him credit though. His caustic managerial style seems to have borne fruit despite his lack of engineering or technical skills. He has been supremely effective at defining a vision(however delusional) and attracting funding.
Will we get to Mars soon? Hell no. But we may end up with a world-leading launch provider based in the US and that's a clear win for the country.
> despite his lack of engineering or technical skills
At least he has B.Sc. in physics and got admitted into Stanford.
I think what Elon says is better explained not as a promise what would happen, but rather as a goal which they're going to aspire to. It kinda supports the idea "we're in business of converting impossible into late". If Elon will start offering more "realistic" schedules, the pace of SpaceX will slow down, perhaps considerably. So, yes, it's "Elon time", which historically isn't particularly precise, but still useful.
I have a physics degree. It is not at all the same thing as an engineering degree.
1 reply →
The physics degree is a deception and perhaps fabricated: https://news.ycombinator.com/item?id=35594076 (and it was a BA not BSc though that matters less)
And the Stanford admittance was for materials science, not physics as he lies about
3 replies →
> Will we get to Mars soon? Hell no.
How much did he bring in that timeline?
You absolutely do not, under any circumstances, have to give him credit.
Chronic over-promise, underdelivery.
Where was the nearly 3T of fraud he said he'd uncover in the US government, again? Was that a clear win for the country?
But hey at least he's effective at getting people to give him money, I guess, which is an indistinguishable "skill" from that of someone who is able to convince people to buy an online course on how to make money online.
He just does it at a bigger scale so people are quick to suck him off. How we are still falling into the "money = smart/competent" trap in <<current year>> is beyond me.
People fuck up. I fucked up three things yesterday. Fortunately I am not as known as Musk, so no one tears into me at Hacker News and my fuckups remain hidden.
Nevertheless...
"underdelivery."
Both Falcon and Starlink are quite major improvements over previous status quo. It is not just a question of having a nice WiFi during your flight. If you are interested in some very practical consequences, look at the Russo-Ukrainian war and the role Starlink plays there.
Don’t buy into the 2010’s Tony stark persona. His momentum is clearly slowing because he can’t put his politics and rather fucked social values behind business sense.
I have immense appreciation for what SpaceX has done for humanity. I’m not being dramatic. Reusable rockets alone is an incredible achievement. But he’s lost the plot. He needs to drop his right wing bullshit and stardom chasing if he wants to be taken seriously again. The dude won’t even acknowledge his own kid because of his politics. I will never trust someone who makes that decision, personally. His judgment is beyond clouded.
The Elon bros will be mad but whatever. One day he’ll maybe remember why folks liked him. Hitching his wagon to Trump was a dumb move.
I think it’s tough to stay grounded when you’re as rich as he is. (To be clear, my intent is to explain and not excuse the path he’s taken.)
Not getting invited to the EV summit would have pissed me off if I were in his place. The Trump thing; it sounds like the government was going to go after him for various violations, and hitching his wagon to Trump was his way of getting out of that jam.
[flagged]
2 replies →