Ask HN: Why hasn't x86 caught up with Apple M series?
418 points by stephenheron 1d ago 597 comments
iOS 26 Launches Sept 15 – Even GPT-5 Doesn't Know It Exists
2 points by rileygersh 4h ago 5 comments
Ask HN: Best codebases to study to learn software design?
100 points by pixelworm 3d ago 89 comments
Stop squashing your commits. You're squashing your AI too
4 points by jannesblobel 1d ago 9 comments
SpaceX's giant Starship Mars rocket nails critical 10th test flight
269 mpweiher 273 8/27/2025, 1:01:57 PM space.com ↗
After the Shuttle program ended in failure, work on reusable launch systems stopped for decades. A similar thing would happen if Starship fails. Space would remain the province of the military and large governments.
Today it costs ~$3,000 per kilogram to put something in orbit (on a SpaceX Falcon 9). Starship aims to lower that to $10 per kg. That's totally crazy, but even if it could get it down to $300 per kg, that would revolutionize access to space.
Data centers in space, biotech manufacturing, and maybe even asteroid mining and energy generation become practical at those prices. To say nothing of telecommunications, remote sensing, and global navigation--all become much cheaper.
And, of course, that drops the price on all the cool science/exploration goals that we always talk about: massive space telescopes, regular probes to all the planets, and crewed exploration.
We're literally at an inflexion point between two possible futures and we don't know which it's going to be. If I were younger I would absolutely try to work at SpaceX to help tilt the chances.
But as it is, all I can do is root for them.
I really don't agree with this take.
It may appear as if SpaceX is the only game in town, but in reality a lot of this technology is commoditized now, and space is as diversified and vibrant as ever.
The starting point today is very different from the post-Shuttle environment. If Starship fails, it is unlikely to be for pure technology reasons and something else will take its place with perhaps better product-market fit.
SpaceX is a symptom, not the cause.
The true dangers to all of these lofty human enterprises are geopolitics, domestic political destabilization and environmental collapse.
Then everything that was not contributing to MOON ASAP was thrown away & massive spending on space hardware essentially normalized. The end result (Apollo) was impressive and achieved the goal (first on the Moon) - but was also totally unsustainable, resulting in a big crunch and a lot of setbacks and slowdowns.
Another contributing factor has been military involvement - again, wee need that milsat on orbit and we need it now, costs be damned. ICBMs not being exactly reusable does not help. Even the Space Shuttle design being perverted into its partially reusable clunky form can be traced back to military requirements.
I still think Shuttle could have worked, if it had been cheaper to evolve it.
Elon's focus on Starship is to make each iteration as cheap as possible, which means they can radically evolve the design without fear. Think of the (recent) switch to hot-staging or the evolution of Raptor.
If Shuttle had been able to evolve like that, I think they could have gotten the cost down. If you think about it, Starship is basically just Shuttle with the orbiter stacked on top of the booster and with propulsive landing.
I doubt it. It's a gigantic kludge that isn't fixable.
For example, the requirement that it land like an airplane meant it needed wings, landing gear, and a full set of flight control surfaces. None of that is useful apart from the landing, and yet it is necessary to push it all into space and re-enter it.
I once emailed Homer Hickam about it, and he was kind enough to reply and said he'd argued the same thing.
Stacking the orbiter on top of the external tank is a non-starter, IMHO. Obviously you'd have to add engines at the bottom, but now your cost goes up unless you plan on recovering the external tank (and how do you do that?).
And now you need another fuel tank for the orbiter, right? Do you extend the orbiter so it can fit an internal fuel tank? Or do you remove the engines and move them to a separate disposable stage?
Having to inspect each and every tile after every trip because they basically didn't work like initially designed was the primary failure of the Shuttle program. It also wasn't nearly safe enough, primarily due to a shitty management culture that was taking over America (and is still currently in power in nearly every business).
The thermal tile technology was for some reason believed to be dramatically easier to design, engineer, and manage than it ever came to be in reality. I'm not convinced that Starship has "solved" the problems inherent in tile systems.
And if it fails, who will spend billions on a new vehicle?
Stoke Space: They are working on Nova, which is designed for 2nd-stage re-use, and they've got a novel architecture. But they are not well-funded and if Starship fails, it is likely that investor sentiment will shift away from full reusability (you know how investors are). And even if they succeed, their current vehicle can only get 3 tons to orbit. That means each launch must cost less than $1 million to get to the $300/kg target. In contrast, Starship can loft 100 tons, so it can cost up to $30 million per launch and still hit the target.
Blue Origin: They are still working on 1st stage re-use, and even assuming they get that to work next year, they are at least a decade away from testing 2nd stage re-use. And their current designs don't have any of the cost-savings in Starship (like launch-tower catch).
And that's it! There are no other companies seriously working on 2nd-stage reuse.
If Starship fails, there will not be another contender for cheap flights to orbit for decades.
There are a lot of other potential technological problems (dozens of engines, stainless steel construction, the belly flop maneuver, etc). Ultimately if Starship would were to fail for technical reasons, it would only indicate the particulars of Starship's implementation don't work. Starship is not the only (or even in my opinion the best) way to achieve full reusability. And partial reusability, which just a few years ago was considered radical, has already been so firmly proven that just about everyone is doing, or trying to do it. The idea of "don't destroy this extremely expensive vehicle after only a single use" won't die for as long as people can see expenses on their books.
If anything, the alternative approach, making a low cost, mass producible rocket has been abandoned, possibly pre-maturely.
Partial reusability won't get the cost down to the ~$100/kg range. And it definitely won't do that and still loft ~100 tons to orbit.
Falcon 9 can get 15 tons to LEO for $45 million, and that's already the lowest price on the market. To hit $300/kg they would need to build a 2nd-stage, launch, recover (on a drone-ship) and refurbish for $4.5 million. That's just not going to happen.
There are only two companies that are actively building hardware for 2nd-stage reusability: Blue Origin (which doesn't even have a prototype yet) and Stoke (which has a max 3-ton payload). If Starship fails, we are not getting $300/kg orbital costs for 1-2 decades minimum.
I agree that Starship has lots of other potential technological blockers (although fewer each attempt--I never thought tower-catch would work the first time). But no other designs are even close to fulfilling the promise of low-cost orbital launch.
Falcon 9 has massively brought down the cost per orbit, and even with the whole world as a captive market, every university in every country putting up cubesats, they still don't have nearly enough payloads to make the economies of scale kick in. Hence Starlink. The majority of SpaceX payload mass has been Starlink, something nobody was even asking for. 300+ launches.
And the idea to reach the economies of scale for Starship is... Even more Starlink. How much Starlink could we possibly need? When will humanity come up with another use for this glut of payload capacity?
Even with the Artemis deadline looming large, SpaceX are still pushing this Starlink angle for Starship, it's nuts
As for "How much Starlink could we possibly need?" I think the answer is simply "YES". Even when you possibly somehow satisfy all your Internet access customers, you can start adding other services, like mobile phone coms (already in progress) or maybe imaging or hosted payloads.
Even at $300/kg to LEO there are a ton of new applications that suddenly make sense. If we get to $10/kg we will literally colonize the solar system.
Falcon 9 launches are already only ~10% government payloads, ~90% commercial payloads. They're already vastly not military/government launches.
Government still props up this whole market.
I should have said that the space economy would remain small.
The minimum propellant travel times are a major barrier.
Did it really stop for decades? I think SpaceX and Blue Origin were both already working on re-usable launch systems around that time
Remember also that when SpaceX started to develop Falcon booster reuse in 2011, every major aerospace company said that reusable vehicles would never make economic sense. Even after the first Falcon 9 recovery and re-flight, most aerospace companies thought reusability was a dead-end and that belief came from the refurbishment cost that Shuttle had to go through.
I count from 1994 (start of EELV) to 2021, when NASA launched astronauts on a reused booster and Peter Beck famously fulfilled his promise to eat his hat if Rocket Lab ever worked on a re-usable launch vehicle.
---
[1] https://en.wikipedia.org/wiki/National_Security_Space_Launch
I don't think Stoke's first vehicle will be large enough either (beyond proving out the technology). They will eventually need to build a much larger one.
Also, Kessler Syndrome is over-feared, like the China Syndrome from the nuclear age.
At Starlink orbits, space debris renters in years, if not months. And even then, having cheap access to orbit would allow us deploy replacements quickly.
At GEO, the distances are so vast that Kessler Syndrome is much less likely (though not impossible).
In between, there is some danger, but again, cheap access to orbit gives us the possibility of clean up.
Almost all the problems can be solved by cheaper access to space, whereas the benefits cannot be easily recreated on Earth.
Is it guaranteed to work? My point is that we don't know, despite confident assertions one way or the other. If it does work, the benefits are clear, and that's why it's worth trying.
My take: The idea that heavy computing in LEO is better than sending data down to Earth sounds very naive. Starlink is great proof that the bandwidth is plentiful and the latency is good, once the equipment is modern. Definitely cheaper and easier than cooling a datacenter.
No comments yet
Nailing it would be without the things above.
The biggest can't-miss milestone was the flawless engine restart. That gives them the go-ahead to hit orbit on the next flight.
They are built to tolerate that, resulting in much better launch volume and weight utilization (sats are stacked on top of each other & held in place by rods that are then released).
The Starship Starlink release demo was quite tame in comparison to that. ;-)
Is it really unexpected that an extremely hot metal pressure tank will rupture when plunged into water?
Since the ship is designed to be caught by a tower and not be plunged into water at all, it doesn't seem like this would be an issue in normal operations.
https://starship-spacex.fandom.com/wiki/Starhopper
And fornlong duration missions also lubricant evaporation, possibility of vacuum welding & atomic oxygen reactions if you spend long in low Earth orbit. :)
Now that they're demoed pez-dispensing v3 dummy Starlinks, I'd assume they'll start launching real ones within ~1-3 months. At that point as long as they can deliver payload to orbit and catch the booster the program is operational and they'll start switching their own Falcon 9 launches over.
The HLS timeline is definitely dicey, but whether Starship winds up being the blocker remains to be seen. Otherwise they've now succeeded enough to "lean launch" Starship with equal/better capabilities to any other existing orbital rocket, and Starlink can fund indefinite further tests/iterations on the rest of their roadmap features (which no one else has).
I think it's still a bit early for that given that they've only had one successful flight and are still testing lots of new design changes, but I think you're right that the capabilities they've already demonstrated are probably enough to make Starship commercially viable even if literally none of the other revolutionary improvements they're working on pan out.
Falcon 9 doesn't recover the 2nd stage at all, and it's already by far the least expensive rocket out there in terms of cost/kg.
I'm surprised they didn't take less risks just to avoid a narrative of failure.
That's the advantage of being privately owned. "Vibes" (hah) don't matter. Public opinion doesn't matter. What matters is executing on your vision / goals. And they're doing that.
The fact that they're bringing in loads of cash from Starlink surely helps. They haven't had the need to raise money in a while, now.
There's a lot more eyes on them now a days, and Musk is much more well known, so it creates a lot more drama - but they've done the exact same process with everything. They even published a montage of failures [1] on the way to their first successful landing 'back in the day.' It was fiery, but mostly peaceful. They didn't even hit a shark!
[1] - https://www.youtube.com/watch?v=bvim4rsNHkQ
It's privately own, might as well learn as much as possible with each dollar spent.
I was very surprised that that flap still held up during the stress testing on atmospheric re-entry.
They specifically said they're testing lighter fins to see how much they would hold. Let's not invent problems when it's an experiment that was clearly stated.
In SRE, we have chaos engineering so I'm wondering if it's the same concept.
They planned a test that would subject various components to stress levels outside of the normal mission profile. The various specific failures that resulted from that may be within expectations but not necessarily planned.
In engineering you want to know that a design will not just succeed at its rated limit, but have some margin percentage of safety above that. To measure that margin often involves destructive tests.
SpaceX's development methods differ a bit from more traditional rocket development by performing some of these potentially destructive tests with full-scale articles in real flight scenarios as part of an iterative process.
It's a different apporach to say the Apollo program, where they did heavy up-front analysis, at the expense of cost-efficiency, speed, and innovation. They had one-shot for a flight, otherwise that's several $bn up in flames.
Even with the last few mishaps, it's an approach that seems to be working. If you look at Starship and Falcon's journey in comparison to SLS and Blue Origin, they have done so much in such a short timespan.
They'll need a higher bar for Artemis but frankly Starship is not the only critical bottleneck there and it's not SpaceX's main financial driver.
I watched the Martian again the other day and I marveled about how much has changed. With Starship progress, almost none of the plot really makes sense (bespoke vehicles and payloads etc). The first mars expeditions will probably be stocked with a thousand tons of gear, enough to easily last a guy 5 years. And if some dude were stranded on Mars, SpaceX could start lobbing things in his direction within maybe 30 days?
The Martian is a vision for a 2035 mission from 2011. We seem likely to beat that!
If Earth and Mars are on opposite ends of the sun, nobody is going anywhere within 30 days. I do not see how anything will change from the one transfer window per ~2 years for the foreseeable future
> The Martian is a vision for a 2035 mission from 2011. We seem likely to beat that!
What, exactly, is that guy doing for those five years? We don't know how to terraform Mars, and it's questionable what having someone on the surface will add to the knowledge we have of surface composition. And then what? That equipment is still on earth - after it's built.
Oh, and how's he planning to get off Mars?
I would comfortably make a $100 bet that there is no chance that we have sent a manned mission to even orbit Mars by 2035, let alone are "settling" it.
Waiting to be rescued. We're not talking about sending one guy to mars for funsies, we're talking about one person left after an emergency. In the book he gets off mars by going to the launcher staged for the next mission, which again is a case of prepositioning extra hardware before sending someone to the planet.
If you assume a team of 5 people with an intended stay of 6 months, 5 years of supplies is a factor of safety of 2. If you send enough supplies to keep the whole team alive till the next launch window, that would keep a single person alive for about 2 decades (ignoring potential storage lifetimes).
A major activity for the Martian would be exploring the location and prospecting for necessary raw materials, like digging for water.
> The effect of this no-abort condition is to make Mars mission design acutely risk-averse.
"Acutely risk-averse" is not SpaceX.
And being acutely risk-averse also underscores my point. If we are actually acutely risk-averse, we aren't going from "still test-flighting and developing the launch vehicle" to "manned mission" in 9.5 years.
I got tingles when the first booster landed on the drone ship, because I knew access to space had just changed in a fundamental way.
First, the time frames are way off. Development of the Falcon 9 took ~5 years (2005 to 2010). The first reused booster came much later (2017?).
Second, Starship is much more expensive for each launch attempt than Falcon 9 ever was.
Third, Starship is significantly more complicated technology-wise, being methane based. There are reasons to do this but it then requires cooling both propellants (instead of just liquid oxygen and RP-1 ie kerosene with the Falcon 9(.
Fourth, Starship has to compete with somethingg Falcon 9 never did: Falcon 9. Falcon 9 is now the most succcessful and cheapest launch platform in history. It is the reliable workhorse of the industry and relatively cheap to launch. Its reuse is proven.
Fifth, the market for Starship is unproven. We can compare it to other launch systems for heavy payloads, most notably the Falcon Heavy, which I believe has only had ~12 launches in almost a decade (compared to the 100+ Falcon 9 launches every year).
You could argue SpaceX will steer customers to Starship but there'll be other competitors (to the Falcon 9) by then.
Lastly, Starship is still so far from being human-rated. So much of the needed tech (eg refuelling in orbit) hasn't even begun testing yet. I can easily see this taking another decade at least.
They are already reusing boosters, so it might already be cheaper than F9 before booster reuse. Once they start reusing the ships, it will be cheaper than F9 with booster reuse because F9 has to build a new second stage each launch.
> Fifth, the market for Starship is unproven
The market for Starship is proven by SpaceX itself. The Starship can add 20x the Starlink network capacity per launch as F9. There are currently around 100 Starlink launches per year, so the market couldn't be more proven.
The launch cost of a Starship today is high, especially if you include development costs, but Musk's goal is a marginal launch cost of ~$1M. A Falcon 9's launch price is ~$70M; Musk claims a "best case" marginal Falcon 9 launch costs ~$15M.
https://www.spacex.com/launches
It’s a bunch of starlink missions. With some dedicated and rideshare missions.
Frankly, it kind of blows my mind what the US pulled off in the late 60's, early 70's with the technology and materials of the time.
That said, I hadn't fully appreciated the size of Saturn V either until I saw it in person in the museum. Like, I had felt it was big, but it was big.
Also, SpaceX is not building rockets, they are building a rocket factory. If they succeed they will have lowered the cost of putting stuff into space by an order of magnitude. The potential rewards are huge.
As with any manufactured item, high volume and iterative design improves the production process and finished product.
https://en.wikipedia.org/wiki/Saturn_V
raptor engines are designed to be cost efficient, as is the rolled steel? that is used for the fuselage
It's way harder to do it the first time.
Manufacturing the Apollo Guidance Computer (which wasn't in the rocket per-se, but was wired up to it and could fly the rocket in certain scenarios) alone consumed around 40% of the US' entire IC production capacity at the time.
Starship has considerably fewer moving parts. And googling 'evolution of raptor engines' gives you some pretty stark images on how simpler things look, in principle.
I know it seems counterintuitive to everyone who grew up in the era of the space shuttle, but the ship is the cheap part, the giant booster is the expensive part.
The ship has a way longer cycle time so starship unit costs are going to dominate fleet construction cost despite being the cheaper unit so knowing exactly how hard you can run them is very valuable information it's worth gleaning by wasting some units early on.
One interesting point is if they actually go for orbit. It would take just a few more seconds to reach something like 200+km / 100km, a place where they could deploy some v3 Starlinks and gather data from the launch (i.e. vibrations, health, dinging on the door, etc). It would be a test where they get more data that's transferable to the new architecture, and relatively low risk of getting stuck in orbit. (low perigee would mean eventual re-entry anyway, hopefully over the ocean) The sats can probably raise themselves from there.
One option is they can run it again with the data gained from missing tiles etc. and see if there is an improvement.
They could also do a similar flight but with an actual orbital insertion and de-orbit if they are confident in the odds of success of the de-orbit burn.
Landing the ship at the launch site means overflying land and potentially populated areas, so I think they're going to want to demonstrate successful control, re-entry, and landing from orbit a few times before attempting that.
But I agree with you, I'd rather have test flight 11 demonstrate at least another successful reentry with no issues (they had a non-fatal explosion on ship reentry in flight 10) before attempting to catch the booster AND landing the ship.
"Great work by the SpaceX team!!!" SpaceX CEO Elon Musk wrote on X after the flight.
Amazing accomplishment. Always a thrill to watch live.
SpaceX conducted 134 launches in 2024 and is targeting a record-breaking 160-170 orbital launches in 2025.
https://www.spacex.com/launches
No comments yet
If the government 100B to dig a big hole and fill it up again, that money would also "go back into those economies". Does that mean it wouldn't be wasteful?
> Project cost US$6.417 billion (equivalent to $33.6 billion in 2023)
> Cost per launch US$185 million (equivalent to $969 million in 2023)
That a manned Apollo mission would/did cost under a billion dollars (todays money) is surprisingly cheap. A single Artemis launch using the Space Launch System (SLS) costs an eye watering $4 billions.
Different metric:
> [1966] NASA received its largest total budget of $4.5 billion, about 0.5 percent of the gross domestic product (GDP) of the United States at that time.
Using that metric NASA yearly budget would with todays GDP be $150 billion dollars.
That said, it would be interesting to have someone really knowledgeable go over what it is that Artemis has and Saturn V didn't, and then break them down and assign each an approximate proportion of the cost delta.
This represents less than 0.5% of the total U.S. federal budget, though it’s one of the most visible and impactful science agencies
SLS is also a pretty weird design due to reusing Shuttle components for a completely different kind of launcher. This saves development costs (maybe) by using existing stuff that has already been developed, but the unsuitability of those components for this system increases per-launch costs. Once NASA runs out of old Shuttle engines, manufacturing new ones is going to cost $100 million apiece if not more, and each launch needs four of them. It was OK for Shuttle engines to be expensive since they were supposed to be amortized over hundreds of flights (and in practice were actually amortized across at least tens of flights) but now they’re being used in expendable launches. If Starship even comes remotely close to its goals, an entire launch will cost less than a single SLS engine.
There are teams of incredible engineers working there because NASA paved the way.
Presumably what we're trying to get at is, in broad strokes, "is Starship more cost-effective to develop than Saturn V" (and I assume the follow-on for that will be to compare the "NASA approach" vs the "SpaceX approach")
But you raise a good point in that the baseline playing field is completely different. The existing knowledge each program started with, be it in materials science, understanding of rocket combustion, heat shield technology, electronics, simulation ability, you name it, it's completely different. So we can find and pull out whatever numbers, but I don't think it's possible for them to say anything meaningful for comparison on their own.
It depends on how different they are. Saturn V was launched 13 times in total. Starship is already 75% of the way there and hasn't orbited once. Ignoring R&D and just going by launch costs alone, that's USD 4B (2025) to orbit 1 Saturn V, vs USD x to orbit 1 Starship, where x >= 1B.
>It only launched 13 times due to being so expensive as to just not be feasible.
"There aren't many uses for such a gigantic rocket. Let's make an even bigger one and hope it works out!"
Apollo 1 - lost on the launch pad, crew killed. very bad Apollo 13 - major malfunction causing loss of mission but crew saved. very not bad
Starship - 10 launches 5 failures. No crew ever so that pressure is also not comparable.
Are we really claiming Starship has achieved 75% of the results of Apollo? That's absolutely ludicrous
Not one of these triumphant 75% achievement in launch numbers would have had a surviving human. Apollo had 0 practice runs. Starship is nothing but practice runs. To equate the number of launches to something so drastically different is just an exercise in futility that I can only assume you're trolling
Read it as "Starship is already 75% of the way to that cost and hasn't orbited once" (you seem to be in agreement)
I bet it will get to the moon cheaper, too, and the Muskonauts will use less expensive lenses than Hasselblads to take photos.
The reason why it matters is that efficiency matters. It's fine if it takes longer, not so much if it costs way, way more, especially if such a huge rocket has limited applications. And as I understand it the consensus is that Starship (or at least a fully-loaded Starship) will never go to the Moon. Once it's in orbit it takes like twenty refueling launches and space rendezvous to fill it up again so it can make the transfer burn. In other words, it's never happening.
Yes the mission profile is more complex, but that complexity can mostly be settled before the astronauts launch on their mission.
NASA seems to think it is a viable plan which is why they selected SpaceX to execute that part of the mission.
> After a multi-phase design effort, on April 16, 2021, NASA selected SpaceX to develop Starship HLS and deliver it to near-rectilinear halo orbit (NRHO) prior to arrival of the crew for use on the Artemis III mission. The delivery requires that Starship HLS be refueled in Earth orbit before boosting to the NRHO, and this refueling requires a pre-positioned propellant depot in Earth orbit that is filled by multiple (at least 14) tanker flights.
I stand by what I said: not happening. I'll believe it when I see it.
Can you imagine if to make a sightseeing trip to another city you had to stop in the middle of the highway and then make 14 round-trips with a second car to fill your first car back up? I can't imagine why someone would approve this plan, other than corruption.
If the alternative was throwing away and building/buying a new car for every trip? Absolutely.
They said the same about landing a first stage booster - impossible and pointless to attempt. And it just happened for the 400th time yesterday.
We didn't get to the moon with a refuelling station did we? How come we need one now? We're really seeing 15 starship launches per moon trip as reasonable, rather than just building a single trip program?
The mission itself is nonsensical. The problems are stemming from the SLS, I'll find a link to a relevant source.
No. We did it by throwing away ~98% of the vehicle on the way there.
> How come we need one now?
Because building a new gargantuan tower and tossing that majority of it into the ocean/deep space every time we need to go the moon is not sustainable.
> We're really seeing 15 starship launches per moon trip as reasonable, rather than just building a single trip program
Yes. Because again. The alternative (dictated by physics) is that we expend the whole thing.
Making trips to the moon sustainable is pointless and nonsensical.
Edit0: good read https://news.ycombinator.com/item?id=40410404
We can also, you know, not. We could put that money to something here on Earth instead of burning it up.
A single trip launch will always be constrained like this due to the tyranny of the rocket equation.
A modular mission system with multiple launches is the best way to expand capabilities and enable things like landing larger payloads for more advanced or long-term missions.
One of many wacked-out things about the plan.
Return payload constraints are probably from using Orion as the return vehicle. Mass to the surface is much higher than Apollo since that is launched separate from the crew.
[EDIT] Apparently there are multiple plans involving even more spacecraft, because why not I guess? It's as you describe for Artemis III, but then gets way more complicated with Artemis IV, involving more spacecraft for some reason.
NASA has optioned an additional lander from Blue Origin but that would be taking the same role as SpaceX's lander, shuttling from lunar orbit to the surface and back to lunar orbit.
1. Starship is never going to be usable for a Moon mission.
2. There's little scientific value in Moon missions.
3. There's never going to be long-term missions to the Moon.
I maintain all three simultaneously.
From wikipedia: https://en.wikipedia.org/wiki/Falcon_9#Pricing
I'm with you. Not happening. We're more likely to come up with a totally different, simpler plan, and do that instead, before this happens.
The fuel that is landed is used to get back from the lunar surface to lunar orbit, not to return to Earth. That fuel stays with Orion in NRHO.
It was pumped, shipped, refined, and trucked to that point using a complex supply chain, enabling your final trip to happen with one fuel transfer.
And it's totally valid for you to have that opinion. But it's your opinion, not "the consensus."
Taking longer at lower cost is a great trade-off for Starship but wasn't for Saturn V. The main driver for Saturn V was the space race against the Soviet Union. Economic interests played a very small role. It was all about being first and compensating for the Sputnik shock.
Nobody but SpaceX knows how much each Starship test costs but the estimates online range from $50 million to $200 million. Presumably, whatever the actual cost, they're more expensive right now while they're redesigning bits and doing custom, one-off work for each flight but it has a long way to go to beat Saturn V for the full mission.
There is nothing wrong with this question. Zero.
Stop eroding this site's community.
> Mars rocket
Very dubious. If you disregard all the SpaceX marketting talk and just go by what they're building, then it's a rocket meant for launching very large satellite constellations as cheap as possible.
As for Mars, Starship might be ready for that in a few years (a year ago I was saying in a year or two, but I've kicked that back). But where are the Mars customers? Who is developing Mars habitation plans and hardware that will be ready in even twenty years? Commercial demand for large satellite constellations obviously exists. Demand for Mars colonization is nonexistent, it exists only in the hopes and dreams of sci-fi junkies.
I really hope I'm wrong, because I'm a sci-fi junkie!
I think as a culture we've lost the ability to compartmentalize. We should be able to criticize and even despise the head of a company, and at the same time celebrate when the intelligence and hard work of the countless smart and hard-working people at that company push the boundaries of what is possible for humanity.
No comments yet
This is not to condemn or argue with anyone who feels differently. But I think we need to be more visible.
For me what this shows that the most important thing for a CEO to be successful is to have money, a vison (no matter how unrealistic or unnecessary) and a cult personality. Nothing else matters. Also it shows that with enough virtual money (I.e.: massively overblown Tesla stock) you can do just about anything.
It's clear that money isn't the defining factor at least. When BO was founded Bezos was the richest man in the world. It has floundered for so long that Musk was able to build up a cult of personality around SpaceX and parlay that into even more money than Bezos.
Jeff loves measurement and control. So he replaced his experienced aerospace guy with the Alexa guy. Because the Alexa guy works the Amazon way: everything measured and tightly controlled.
Will Starship every carry a large enough payload to justify the launch cost? I'm skeptical. Musk's Mars fixation is nuts.
Thanks to the intelligence and hard work of the countless smart and hard-working people he pushed the boundaries of what is possible for humanity.
Still, I find it hard to accept we should compartmentalize and not think about who those rockets were built for and with what purpose.
On the one hand I am a major space nerd and I see the value of what SpaceX is doing. Especially with it really seeming like no one is anywhere near their level. What kind of scientific advancements will be possible once this thing can be used normally and launches like this become commonplace.
But at the same time it is impossible to ignore the Elon situation. And that also directly relates to Trump as well. We are in this bonkers situation where he helped get a largely anti-science administration in power and yet also runs one of the companies that will help science.
It does raise serious questions about whether or not there will be limitations on what types of science can be done. Will they have some line in the sand and say they won't launch satellites that do "X", like maybe monitor climate change.
I think maybe rooting for them to fail is a bit much, but I am sure as hell hoping that someone else can catch up. But in the mean time I will celebrate these achievements cautiously. Recognizing the amazing work that the engineers at SpaceX have put into this, because they do deserve a lot of credit for that.
What if he’s not an idiot?
What if we should actually be listening to what this guy says and considering it?
What if he has the same ability to see what nobody else can see early on in politics…
As he’s shown across the rest of his career?
Lets evaluate that claim, by first defining what an idiot is, and then looking at his history, all the things he said, and all the things he has done.
Ill skip you the trouble of going through that process - he is very much an idiot.
Don't confuse the ability to throw money at something and make it work through sheer cash burn with actual intelligence.
For example: https://www.reddit.com/media?url=https%3A%2F%2Fi.redd.it%2F1...
This statement alone disqualifies him to talk about anything self driving.
Are you the richest man in the world? If not, could it be that what is good for the person in that position is not good for you or most other people?
The "accomplishments" you're listing are mostly just investments that he managed to hype up very well. I'll give him this, he's an excellent huckster. But listen to his opinions? I wouldn't let him tell me what color an orange was.
No comments yet
Please tell us which culture war topics he’s anti-science on.
- Deletion of economic databases.
- Deletion of public health information and historical weather databases.
- Criticism of vaccine effectivness
- Defunding of health and climate research funds.
- Leaving the largest international research alliances in Climate and Health.
- Defunding of NASA
He called the administration anti-science, which is well grounded claim. But no part of the phrasing implies that "republicans" are anti science or that you are.
As far as Trump (and the administration in general) being anti-science. I really don't think I need to list examples of this.
My point with stating it, is it is not unreasonable to ask the question if we are reliant on a company with someone like Elon owning it is what the company will and will not fly going to be dependent on politics.
Advancing human scientific progress, but at what cost?
If Musk does achieve a second foothold for the humanity, then any and all objections to his methods become irrelevant. So far he does deliver. So we wait for the final result.
Also, if you don't know, we've got a war in europe for like 3.5 years already. I'm seriously curious how many times a space-x total program cost since their start in 2000s has been already sunk into that.
Remember, development of the R-7 into the most reliable expendable booster took about 40 years and that with full backing of a Soviet Union.
To have it go to Mars, you have to have something like 20 refueling launches.
Raptor engines are the equivalent of taking a 2.0 inline 4 engine in a car and making it produce 1000 hp. Demonstrating one restart does nothing. You have to demonstrate reliable restart in a variety of conditions.
Vertical landings using engines to slow is inneficient. Sure, you reduce system complexity in not having wings or parachutes, but you gain it all back with things like engine restarts, and the need for high pressure ratio engines.
>I think as a culture we've lost the ability to compartmentalize
There is a limit to how bad a person can get, after which its probably worth for anything adjacent to him to fail so that he can't do the things he does. If Musk was aligned behind someone like George Bush, he would receive a fraction of the hate he does now. But at this point with Doge and his rallying behind a dictator (only to be hilariously dumped and fucked over with the BBB) is bad enough for all hate towards Space X to be justified. We would be better of as a society if people had more of a moral compass and actually did things like quit companies. Its not like there isn't competition with much more sane people at the wheel for Space X and Tesla.
> Vertical landings using engines to slow is inneficient. Sure, you reduce system complexity in not having wings or parachutes, but you gain it all back with things like engine restarts, and the need for high pressure ratio engines.
Well, the most cost effective rocket ever built uses this, but please do elaborate. What system do you propose that would be better?
I was thinkin more along the lines of unconstitutionally cutting things like USAID thats going to actively result in more unnecessary deaths, personal vendetta against LBGT organizations in cutting those budgets, cutting IRS funding which is going to end up costing more ironically as there isn't enough staffing to pursue all the people using tax avoidance
Not to mention in general supporting someone and donating campaign funds who literally tried to coup the government, contributing to him being elected, and thus being responsible for everything that Trump is doing, from fucking up the economy further to playing wanna be dictator with ICE, all of which have long term repercussions.
In comparison to GWB, Trump is going much worse (especially if you want to compare deficit spending), and Musk has a direct hand in that.
>Well, the most cost effective rocket ever built uses this, but please do elaborate.
Cost effectiveness has nothing to do with engineering. For example, I can start a car company, fund it with my other business venture, and then sell cars for $50 and it would be the most cost effective car.
That's bad IMO, but not helping people isn't remotely the same as starting a war that kills them (different if they were US citizens, but in this case they aren't).
> personal vendetta against LBGT organizations in cutting those budgets
I don't even have anything against these organizations, but I would cut their budgets to zero if I were in charge because funding them is none of the government's business.
> cutting IRS funding which is going to end up costing more ironically as there isn't enough staffing to pursue all the people using tax avoidance
This is dumb, but hardly a moral issue.
> In comparison to GWB, Trump is going much worse (especially if you want to compare deficit spending)
I 100% agree, but so was Biden, and I don't blame everyone who supported him either.
Fundamentally I don't blame Elon (or anyone else) for anything that Trump (or whoever they supported) has done that he didn't have a direct hand in e.g. DOGE. It's the nature of politics that you have to swallow a lot of things you don't like when you choose a side to support.
I'm sure there could be better rocket configurations than Starship - but so far Starship is better than other existing schemas. Starship is really good comparing to others - getting the cheapest kilogram to orbit cost in perspective.
I wonder about this scenario..
...
2019: A talented engineer is pumped and excited to join an innovative space tech company who are killing it, they're vaguely aware that Elon is some kind of celebrity but they dont follow social media stuff. They're super passionate and diligent in their work.
...
2025: Elon backing Trump, doing DOGE stuff, supporting far right politcal parties in Germany, doing overtly nazi stuff. The engineer realizes what a vile and irredeemable sack of shit their boss is and now they feel conflicted, their enthusiasm falls away and they begin phoning it in at work while they browse other jobs.
...
This happens X100 or X1000 to varying degrees amongst all employees and you have some very unreliable rockets.
That kind of testing is understandable to a certain extent. But it doesn't make sense to ditch the rocket in the Indian Ocean once you've run those experiments, instead of catching it, and having all the parts available to study.