Seems to me that making it hyper-converged, with all storage make it make sense. Ultra dense compute alone isn't ideal. Oxide rack has the features of blade but is hyperconverged with everything. Plus the better integrated software.
Agree. But I think that Shuttle didn't do intermediate tests of these things is part of the reason it never lived up to its potential. During development they lost fact of what they tried to achieve in the first place.
They sacrificed what worked for potential, but tried to take far to big of a step.
I would argue, if you design something that has so many potential pitfalls and so many operational constraints, and so many drivers that make it incredibly expensive and slow, it is understandable why they started overlooking red flags. They would have barley ever lunched at all if they had not overlooked red flags.
This not actually a good reason. First of all, of course you can't get anything better if you never ask for anything better or consider alternatives or put any money into anything better.
If you never invested in anything else, then the M4 Sherman tank would still be the best tank. And then you could say 'we can't get anything better' and continue to use it while refusing to ever even put 1$ into developing anything else.
And actually there are plenty of ways, even without Starship to do these things differently.
I remember when SLS fans in 2016 told me that SLS is real and Falcon Heavy is fake. Even when Falcon Heavy and now New Gleen can do most things SLS can.
One NASA Administer was almost fired for exploring if Orion could launch on Falcon Heavy.
The thing is NASA is not looking for alternatives, and it doesn't matter if Starship is fully proven and operational. People will still say its not the same, because Starship will need refuel and isn't direct.
You can always find a reason to justify one solution if you only ever consider that solution and refuse to even look at any other possibility.
> By your own data, Apollo had 65% more money than SLS/Orion.
I would say by my own data Apollo has to do 800% more work given the point on where it started.
> My point is, Apollo had a clear objective: put people on the moon. When that was achieved, they shut it down.
This isn't true. The Apollo program had many more objectives and was continuing and was about to do many more things. The politics around it just changed.
If it was a single goal, then they could have stopped after Apollo 11.
And of course after moon landings stopped, Skylab and other post-Apollo programs continued. Much of the Apollo hardware continued to operate for quite a while longer.
> SLS objective is: do something NASA-like with astronauts, using current suppliers as much as possible, and better/larger than Apollo. Oh, we are going to ask you to change plans all the time.
Yeah but those changes in plan don't actually change the hardware of the rocket itself. Its always the exact same rocket. It just gets new mission that are designed to work for it, not the other way around. So you can't really say SLS/Orion was delayed because of chaining requirements.
> So it's not about risk averse culture, or the decline of western civilisation, or something like that.
The very fact that the requirements are fuzzy and the political process is a shit-show that gets nothing done and provides bad intensive is exactly the civilization level failure. Just as much as when you try to land on the moon or build a high speed rail line.
> BTW, thanks for the hard numbers, it's a nice analysis.
NASA and government is pretty good on this and lots of people have done work on this, specially Casey at the Planetary Society. So I do not deserve all that much credit.
And for SLS I have been following the program for 10+ years and have been arguing since 2015 that the only rational plan is canceling it. But the congressional alliance behind it is just incredible.
The “% of the federal budget” comparison is mostly a rhetorical trick. It can matter politically, sure, but it’s a terrible way to compare programs across time. Apollo happened before a bunch of Great Society-era spending and later expansions in the federal budget. Comparing shares across radically different eras is basically apples-to-elephants.
I spent some time trying to get solid numbers because I was actually interested in this.
Inflation-adjusted averages:
Apollo-era NASA average (FY1961–FY1972): ~$44.2B/year (2024 dollars)
NASA average over the last ~20 years: ~$25B/year (2024 dollars)
So over FY1961–FY1972 (12 years), that’s roughly $44.2B × 12 ≈ $530B in today’s money for all of NASA.
And what did that buy?
A NASA that was basically inventing the modern space industry:
- building launch sites (LC-39 etc.)
- building huge test facilities and stands
- building control centers / mission operations
- building manufacturing capability at scale
- building/expanding NASA centers
- building DSN and deep-space comms infrastructure
- massive amounts of fundamental research and basic engineering research
- building multiple human spacecraft programs (Mercury → Gemini → Apollo)
- developing major new engines (F-1, J-2, and a bunch of others)
- building multiple rockets and variants
and flying tons of missions, including 6 Moon landings
But of course, NASA wasn’t only Apollo. Even though Apollo dominated, NASA also did a bunch of major non-lunar work: Mariner, Orbiting Solar Observatory, Echo / Telstar / Relay / Syncom, X-15, and the beginnings of Skylab, etc.
That article’s Apollo-only number is around $257B (in 2020 dollars) depending on what you include. I used 2024 $ above for budget. But its close.
Now compare to Constellation and its children (Orion + SLS).
A fair estimate for the cost to get to where we are today is around ~$90B (not counting suits or the SpaceX / Blue Origin landers). And what did we get for that? So far, a few test very incomplete flights.
Artemis/SLS is not doing Apollo-style clean-sheet propulsion development. It mostly reuses Shuttle-era propulsion (RS-25 + solids) with restarts/updates, rather than developing new engines like Apollo had to.
Looking forward gets fuzzier, but current projections suggest roughly:
~$20–25B more before the first crewed lunar landing (assuming the schedule doesn’t slip again)
then for five more landings, under optimistic “one per year” cadence assumptions, maybe another ~$30B or so
So you end up around ~$150B total if everything goes right from here. And note: this assumes huge savings because SpaceX and Blue Origin are spending lots of their own money rather than NASA building its own lander in the Apollo style.
So very roughly:
~$150B (Constellation → SLS/Orion → first 6 landings, optimistic)
vs
~$250B-ish (Apollo-only, depending on inclusion choices and dollar basis)
And my basic point still stands: Apollo had to build the ecosystem, the infrastructure, and the foundational research base from scratch. A gigantic amount of that 1960s infrastructure is still in use today, and 60+ years of engineering and technology progress should matter. That alone should be worth well over $100B in “things you don’t need to reinvent.”
In pure execution terms, it’s hard to argue Apollo wasn’t on a totally different level.
I think you are just talking out of your ass. Please provide some sources.
The milestones and options are all defined in the original contract and each milestone is assigned some monetary value. There were a set of extension option that add milestones for a second lander that NASA choice to pick up. All this was specified in the original contract.
Starship won one small additional contract that I know of, that was about liquid transfer in Orbit, but that just one of 20+ minor contract about space operations.
> no matter what happens to the mission (it'll probably fail), Elon and his buddies will still get to scrape a couple hundred million for themselves
If it fails SpaceX will not get the money covered in the milestones. So if it fails it will 100% be SpaceX that pays. Why are you making stuff up?
Also, SpaceX has been the most successful NASA contractor in the last 20 years and its not even remotely close, so your confident that it fails is just bias.
NASA own estimation is that the SpaceX lunar program will cost more then double what they are paying SpaceX. SpaceX is giving the government an absolutely insanely good ideal and building a lunar lander for like 1/10 of the cost the lunar lander estimates were during constellation. SpaceX will be LOSING MONEY ON THIS DEAL.
Same goes for BlueOrigin, they can only bid because its Bezos hobby project, they will not make money from the lunar lander anytime soon.
All the contracts are public, if there are contracts that SpaceX got for Starship beyond the original lander contract and the minor demonstrator contract I mention above, please link them.
> Funnily enough, the person who decided to grant SpaceX this contract, Kathy Lueders, did so and then immediately decided to quit NASA and work for SpaceX. Nothing to see there.
Kathy Lueders has fantastic reputation with everybody in the know and has worker for NASA for 20 years. Its also wrong to say that it was just her, there is a whole team doing the evaluation with lots of experts involved.And after her the report had to be approved by a whole bunch more people.
If you have any actual evidence of wrong-doing, please come forward.
Funny enough Boeing did actually get caught cheating, a NASA executive was actually fired because he was giving Boeing details about the contract and giving them chances to re-submit the bid.
BlueOrigin's 'National Team' and SpaceX received the same technical evaluation level while SpaceX was only like 50% of the price.
There is really no question that it was the only sensible selection. Anybody would have done the same selection. In fact had Kathy Lueders not selected SpaceX she might have been fired for bias and wrong doing.
It was then followed up by lobbying on the side of BlueOrigin where they got their senator from Washington involved and forced an increase in budget so they could also participate, but they were forced to lower the price by a huge amount as well.
So the result is NASA gets two lander program for about 1/4 or cheaper then was expected 20 years ago during Constellation.
If that's not an amazing deal I don't know what is.
> How could SLS, a rocket that literally worked the first time, be worse than Starship, a rocket that does not work?
Its so crystal clear that you don't know anything about rockets and you're only goal is to perpetuate anti-SpaceX hate.
SLS 'first' flight is literally 95% things that already existed, doing something very conventional. It is literally using engines that were built in the 1970s. And its using solid state boosters form the same factory as those of Shuttle. Its literally just a bunch of old parts in slightly different configuration.
And its already cost 50 billion $ in development without even having to design anything new. The launch cost are absurdly high, so high that NASA can barley fund a SLS any anything else at the time. Notice how during Constellation they never even started to build a Lander, they never had the budget for it.
Starship on the other hand is a completely new architecture, with brand new engine, brand new infrastructure, brand new manufacturing sites and so on. And its trying to do much more then SLS. Its trying to be reusable and support distributed launch, and be a lander.
If all NASA wants is a simply rocket that can launch stuff, then SLS shouldn't be compared with Starship at all. SLS is more like Falcon Heavy or New Glenn, just 10-50x more expensive. Notice how Falcon Heavy also worked the first time it flew, because it was just parts of existing rockets in a new configuration. Its almost like its easier if you build with components that have flight heritage. Crazy how that works.
If NASA wanted just a simple big rocket they could have gotten there much cheaper then SLS. So the whole SLS vs Starship comparison doesn't even make any fucking sense in the first place. The goal of Starship was never to be SLS. Falcon Heavy is already 90% of SLS and if NASA had wanted to, they could have paid SpaceX to boost its performance a bit (something that had been studied 15 years ago already). And now between Vulcan, Falcon, New Glenn there are plenty of options if all you want is launch.
Honestly what kind of idiotic engineering evaluation is it to say 'X worked first time' so its forever better then anything else that didn't work the first time. That's not how we evaluate engineering projects ever even if you were comparing the right things in the first place. This argument just proves that you are not seriously trying to engage with the issues of Artemis program.
I would agree. The heatshield is tricky. But they have shown they can survive without parts of the heatshield. But its a problem for rapid re-usability.
That said, I think Starship architecture can be useful even if this issue is not fully solved.
Starship can be much, much cheaper then SLS even if they throw away the upper stage.
This is just nonsense. First of all, the companies in the 1960s were all there for profit and all made profit. And the politicians in power back then also tried to get contracts to companies in their districts. Why do you think the NASA control center is in Houston?
SpaceX is on a fixed price contract. Dragging the project along is literally costing them money.
By literally any analysis you can do, you will see that in the last 15 years SpaceX was by far and away (its not close) the best contractor to NASA in terms of delivering what NASA wanted.
In fact, by far and away the project that have done the worst, are the project NASA does in the old style where they remain the main designers and operate and only work with private companies as builders. That's exactly why SLS is such a shit-show.
This complete nonsense. The Apollo team was much, much, much, much larger then any V2 team. And mostly Americans.
And their testing procedures were actually very high quality. You don't just accidentally land on the moon and return. That doesn't just happen 'somehow'. That is truly idiotic level analysis.
> Each F1 rocket engine was hand tuned by drilling holes into the "plate" so it would not cause the combustion mixture to vibrate the engine into smithereens.
And that this works was established with lots of both experimental testing and lots of theoretical work.
So much so that F1 is one of the most reliable engines ever used in space flight history.
> Such an approach would never be tolerated today by NASA.
Except of course that the RS-25 engine used by NASA today is known to be less safe then F-1. Having had more failures and generally causing more minor operational problems.
It seems you have absolutely no clue what you are talking about. In fact, NASA own analysis before they were forced to pick SLS by congress indicated that a updated version of F-1 (still relying on the analysis of those people in the 1960s) would be a much better rocket.
Apollo worked because engineers from the top to the bottom made smart engineering focused decisions taking responsibility for their part of the stack, and close working together in teams on a shared goal while having very solid testing procedures for everything.
How many of those traffic accidents could have been prevented if traffic engineering was a serious engineering profession and road deaths were not simply accept as a 'fact of live'.
How many lives would have been saved if a bridge for trains instead of cars were designed?
The budget is actually not that much worse. If you adjust for inflation.
On avg NASA budget was about the same as now. But now we do more things now. But between Constellation/SLS and Orion this new Shuttle based architecture has as much money as Apollo while having done almost non of it. Before it is where Apollo ended up, it will cost much, much more then Apollo.
But even if what you said was true, a gigantic amount of infrastructure that was paid for in the 1960s is still in use today. A huge amount of fundamental research that was required is already done. That alone should make it much cheaper.
Same goes for development, Artemis is not developing any new engines, while Apollo had to develop many new engines.
> The big budget is being allocated to brutes killing people in the streets and a decadent ballroom for the emperor.
Except of course that Korea/Vietnam were much more expensive then what were are doing now.
No it doesn't. Because literally anybody who knows anything about NASA and follows the Space industry in detail has known about most of the issues since 2015 or even in 2011 when this whole new Post-Constellation shit-show started. And many of the problems have been talked about since the day NASA created Artemis. Destin is just more famous then many of the people in nerd forums.
Destin analysis is ok and he makes a number of good points, but it very pro-Alabama (Mafia) inside NASA and contractors since he very clearly is influence by the strong Albama presence and those are the parts of the industry he interacts with.
So Destin misses a huge amount of the relevant puzzle pieces, or he simply doesn't talk about them.
He also simple makes a few assumptions that are fundamentally wrong, namely the different targets of the program. The goal was never to repeat Apollo and landing a few people a few times is totally different from the original goals of Artemis.
I think this claim of being 'way too complex' is a bit over the top.
Sure it was complex for the electronics and some other aspect in the Soviet Union, but not by that much.
N1 actually flew and it mostly failed when engine outs and vibration started to cause other issues with piping and so on.
I think those are solvable problems. With engine reliability going up, whole system reliability would go up to. The piping issues and electronics issues were fixable in time.
Russia was on the right track. They had the right kinds of engines they needed. An engine that could also be used on smaller vehicles to have a shared family. Engines that could be restarted and tested.
They arguably should have started with a smaller rocket with those engines and only gone to N1 when they were reliable.
Wrong. Both were lost because of a fundamentally BAD ARCHITECTURE. And that architecture was bad because the NASA engineers who designed it, had never designed anything like it before and were never able to test or evaluate any of their assumptions.
Columbia would not have been lost if the Shuttle was top stacked, instead of side stacked.
Challenger would not have been lost if not for the use of solid rockets to launch humans.
Both of these design decisions were done to reduce development effort.
False. SpaceX development of Starship is much cheaper then SLS despite using more test vehicles. The claim that building hardware rich is more expensive is not really shown in the data.
NASA has done some analysis on early SpaceX and shown that their methods produced a 10x improvement in cost. And that was with the method NASA uses that often turn out to be wrong.
True but we know for a fact that it doesn't consume 4-5 billion $ a year for the last 15 years like SLS/Orion because SpaceX couldn't afford that. If you actually do some basic math and look at SpaceX revenue and so on, you can make some pretty decent guesses. And SpaceX is analyzed in detail by lots of people.
Starship is irrelevant. SLS was dumb already in 2011 when Starship doesn't exist. Its a dumb system and was never the right system. NASA own analysis showed that.
People who defend SLS on the bases that Starship isn't good don't get it. It doesn't matter if Starship exists. SLS should have been canceled even if you assume the state of the rocket industry in 2015.
Anybody with half a brain and 3h time to do analysis on the topic could figure this out.
Difference is SLS has received 2 billion $ a year for 15 years in a row, while SpaceX get that much once and has to actually cover any extra cost themselves. Why do people just totally ignore money when it comes to SLS.
Not to mention that SpaceX got funding in like 2021, and SLS in 2011.
And SLS works, then why can it only launch every couple of years. I mean what good is a rocket that is so hard to produce that the whole politics and everything around it changes between launches. They basically have to teach a whole new group of people about SLS for each launch.
> while SLS did an orbit of the Moon, with real payload satellites.
If you want things launched to the moon, SpaceX, BlueOrigin or ULA could have done that many times every year for the last 15 years just as well.
Starship isn't just another 'look we can launch some stuff to the moon', its much more, and therefore much more difficult.
You are praising SLS for doing the very, very, very minimum that it should have been doing since 2017. And it will do it at most 3 times until 2027.