This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.
Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.
We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:
-
Shaming.
-
Attempting to 'build consensus' or enforce ideological conformity.
-
Making sweeping generalizations to vilify a group you dislike.
-
Recruiting for a cause.
-
Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.
In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:
-
Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.
-
Be as precise and charitable as you can. Don't paraphrase unflatteringly.
-
Don't imply that someone said something they did not say, even if you think it follows from what they said.
-
Write like everyone is reading and you want them to be included in the discussion.
On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.
Jump in the discussion.
No email address required.
Notes -
What's your take on it's performance so far, in that regard? It seems to have taken quite a bit of time for it to pick up speed during launch, just with 16 tonnes of the dummy payload. It's hard to imagine it taking off with double that, let alone the 100 tonnes they're targeting.
Very disappointing out of context, then reasonable with a little context, then worrying with a lot of context.
With their current "v1 booster, v2 ship" stack, they claim max payload to Low Earth Orbit of ~35 tons, basically twice what they can get from a Falcon 9 with a drone ship landing. A new stack and test flight of Starship is something like 4 times the cost of a Falcon 9 flight, so we're looking at twice the price per kg, when they were shooting for 20 times less, so they're off by a factor of 40. Demonstrating booster reuse this early means that ongoing costs would be less than new stack costs, but until they can reuse the ship too or increase the payload they're still at maybe half of the price per kg of Falcon 9, roughly what Blue Origin is trying to hit with New Glenn. They'd be way ahead of Falcon in price per m^3, which for older Starlink designs might have been more important (they were limited by the size of the Falcon 9/H fairing much more than by the payload capacity), but these days they're launching "Starlink 2 mini" and "Starlink 2 mini optimized" satellites that squeeze down more densely and actually use the Falcon 9's mass capability. Regardless, billions of dollars of R&D to save a few hundred million a year on launch costs would be very disappointing in LEO.
Past LEO the current design would be a total failure. Their goal of using at most a dozen refueling launches at full-reuse prices to get a hundred tons out of Earth orbit would be awesome. They'd be able to put dozens of people on the moon more cheaply than they currently send four people to the ISS. But with their current performance, requiring three dozen launches at partial-reuse prices to get 35 tons out would mean they'd eat a loss just fulfilling the HLS contract.
So, TL;DR: very disappointing.
When I watched, it looked like it cleared the pad pretty quickly as soon as it started moving at all, but for some reason it wasn't released from the pad for like 5 seconds. Makes me wonder if some sensor reading almost triggered aborting the launch. They did shut down one booster engine about 2/3 of the way through its flight, and didn't relight it for the boostback burn, but just one problem engine wouldn't have been enough problem to possibly cancel a liftoff.
That's because imagination is an inadequate substitute for math. ;-)
The v1+v2 stack is around 5200 tons. By far most of what any rocket does with its fuel is accelerating the rest of its fuel. If they were to add another 16 tons of payload to that stack, it would not have half as much acceleration at takeoff, it would have 99% as much acceleration at takeoff (about .36 instead of .365 g's). Another 68 tons after that gets you down to around 94%. The rocket equation is a harsh mistress, and one of the consequences of it is that, for any rocket with enough delta-V to get to Earth orbit, the payload mass at liftoff is practically a rounding error.
The extra 200 tons of propellant in the v2 ship does make a bit of a difference to their initial acceleration, though. What their current descriptions call a "v2" Starship would have been in between v1 and v2 in their earlier talks; it's basically "v1.5". It's got the stretched ship from what they previously called a "v2" stack, but not the stretched booster with higher-thrust Raptors. From a performance point of view they've upgraded half the ship so far, and in an imbalanced way. They still think they can get 100 tons to LEO out of what they're now calling "v3" (what was "v2" on their prior timelines) with both ship and booster upgraded. Flying newer ships on older boosters isn't a performance thing, it's a "need to test after major changes" thing, and in hindsight they really needed to test after major changes. The higher-thrust Raptors on the next booster version should give it around .41 g's at liftoff even with both stretches. That also means higher acceleration through most of the trajectory (rockets start accelerating faster and faster as each stage's fuel burns away and the same thrust is lifting less weight; it's common to start at around .25 or .3 g's but throttle down before the end to cap the acceleration at 6 g's to go easy on the payload), which means less nasty gravity losses (imagine a rocket "taking off" at 0 g's acceleration - it's still producing 1 g but only fighting gravity with that and getting no velocity) and more efficiency, which means much more payload makes it to orbit. The payload at liftoff is practically a rounding error compared to the weight of fuel and oxidizer, but when Starship hits orbit with only as much propellant left as is necessary to get back down, the payload should be more than a third of the total mass.
We saw this same sort of growth with Falcon 9. The first "v1.0" stack was around 320 tons at takeoff and could put 8 or 9 tons into orbit, but they kept making the engines more powerful and that let them stretch the fuel tanks and densify the fuel, and today the final "v1.2 Full Thrust Block 5 Why-Cant-SpaceX-Name-Versions-Sanely" stack is around 550 tons at takeoff and can put at least 22 tons into orbit. They typically spend some of that growth on things like earlier staging and landing legs and landing fuel, and now they can put 17 tons into orbit while landing the booster again afterwards. For booster landings, Starship doesn't have to spend anything more than it already has. Earlier staging and landing fuel are already in the current accounting, and they replaced "landing legs" with "giant robot arms on the launch tower" and somehow that's repeatedly worked because I guess we live in some kind of sci-fi anime now.
They could finish up their immediately planned upgrades, and if everything works on upper stage reuse too, hit their original goals. With an additional booster upgrade (and an extra 3 engines on the upper stage), they still think they can double their original (2017, after they scaled down from the 2016 trial balloon) payload goals, and that would be amazing but not a priori impossible.
So, TL;DR: reasonable with a little context.
The trouble, in the long term, is that "if everything works on upper stage reuse too". They could give up on upper stage reuse completely, get to spend a few tens of tons on more payload instead of on heat shielding and flaps and landing propellant, and even with the design they're testing next year they'd be good enough for HLS and for Starlink launches and for continuing to price under their competitors' next generation. But, unless they can do upper stage reuse, that still doesn't put a colony on Mars. Their existing "can reenter and then do a soft touchdown afterwards despite some damage" ships would actually be fine for getting to Mars, since that atmospheric entry isn't inherently as bad as Earth's (less free oxygen in the shock plasma, plus significantly lower speed entry) ... but then to bring people home they need the same upper stage to get back to Earth afterward, with no more refurbishment than they can accomplish in situ, with the reentry at Earth now at a significantly higher speed. With their dream architecture, getting to Mars is (relatively!) cheap and getting back afterward is practically free, but if they can't make upper stage reuse bulletproof then getting to Mars is too expensive to do regularly and getting back is impossible (without some other gimmick like carrying a separate reentry capsule). Part of why their flights 7 through 9 were so awful was that such major regressions on the v2 ship were really embarrassing, part of it was that the flight 7 and 8 failures were at the worst possible part of the trajectory, but part of it was just that they've been trying to do more heat shield robustness tests (removing tiles here, experimental tiles there) since January and they didn't actually get a v2 ship to reenter properly and run a test until August.
So, TLDR; worrying.
Thanks for the detailed response!
Yeah, I take it back. When I was watching a livestream, someone commented on it taking it's time (probably it being held, as you say), but what really gave me the impression was a post-launch commentary video, which, looking at the original stream again, must have shown the takeoff in slowmo.
Eh, it would be awesome, but unless some rabbits get pulled out of several hats, Mars feels like a distant dream.
You're welcome, but it's not entirely to my credit. I reserve the right to keep giving you shit about "three years AFTER" from now until the end of time, and I figure if I don't at least try to be as helpful as I am annoying then that's probably not good for my soul.
I actually forgot to mention what I think is a very important detail in the "worrying" category: the fact that they have a shot at reaching their target payload isn't actually because their initial estimates were good, it's because their underestimates of engine performance are making up for their overestimates of structural performance.
On the one hand, the Raptor looks like it'll end up with a thrust-to-weight ratio similar to what they eventually hit with Merlin, despite much higher specific impulse, and it's already better than the best non-SpaceX engines in history, more than twice the next best engine with equal or better specific impulse. The Raptor 3 is incredible, not just in the metaphorical "higher thrust than Blue Origin engines twice its size" sense, but in a literal "the United Launch Alliance CEO thought it must just be an unfinished model until Gwynne Shotwell posted video of it firing" sense.
On the other hand, the mass growth on both Superheavy and Starship has been heinously bad. They need the extra power, because they're something like 50% over their estimates on dry mass. On Superheavy that's not such a big deal so far (every extra 10 tons of dry mass on the booster only means losing around 1 ton of payload), and not likely to be a bigger deal in the future (they're already landing and reflying them), but on the ship that's a huge deal (every extra 10 tons of dry mass on a ship that's doing a powered landing is a tad over 10 tons less payload) and it's still in danger of further growth (soft-landing a ship is 80% of the way to reflying it again, but there's no telling how much extra bloat that last 20% will require).
I'd put maybe 33% odds on them sending an unmanned (save for Optimus androids) one-way ship or two in the 2029 launch window, albeit probably to crash on arrival. 66% in the 2031 window, with slightly-better-than-even odds of survival by that point. Just a baby step compared to a colony, but still, it's been super exciting when we landed a couple 1-ton robots on Mars and I can't wait to see what scientific "rideshares" in a 100-ton payload will look like.
What pushes us to "distant" is that those launch windows only come every couple years. On Earth, when SpaceX flubbed a test super-badly then it took them 7 months to launch the next, and when a test goes well it takes them 1 or 2 months to launch the next, and they're gradually speeding up the cadence overall. For Mars entry, no matter how well or how poorly a test launch is prepared, it's 6 or 7 months before they reach Mars and find out, and it's another 19 or 20 months after that before they can launch a redesign. If they want to test Earth return too, with a conjunction class mission to give refueling time, it's 29 months before the Earth reentry and then another 23 months after that before they can launch a redesign. That's a horrible rate at which to iterate, for a company that's always depended so heavily on iterative testing.
Feel free. Like I said the first time, my bets / opinions on Musk are not based on expertise, and I hope I didn't come off as pretending that they are. In fact, part of my shtick nowadays is proving the superiority of Vibe Analysis over deliberate reasoning, so I suppose my ignorance only works to prove my point here.
Well, if you want to bet, I'll be more than happy to give you 3:1 odds on this one. Even if it's smooth sailing from here on out, I don't know if they'll make a go for it. No one asked him to go to Mars, it's Elon's own personal dream, there's no money in it. OTOH he does have a contract for going to the moon, and his investors might want to see a better return on Starlink, and the things that have to be achieved before he gets beyond LEO make it so that "smooth sailing" is far from guaranteed. I want to see how that orbital refuelling works out, and if it handles boil-off well enough that it doesn't turn out they underestimated the amount of necessary launches by a factor of 2-3x.
Finally, there's the competition and the political risks. If Bezos swipes the moon from under Elon, the investors could very well say they're done here. If the competition can provide a tolerable alternative for Starlink, at least for the Pentagon, and the Dems win the next election, they'll stop at nothing to fuck him over.
Ah, but the catch is that most of the other sides of your bets (myself included) are probably likewise using motivated reasoning, not deliberate reasoning. "Elon Time" has been a thing since at least Falcon Heavy (announced Apr 2011, first launch planned by end 2013, first launch accomplished Feb 2018, 82/32 months = 156% schedule slip). BFR announcement was Sep 2017 with first unmanned Mars launch planned by Nov 2022, so that'd just make the 2031 launch window after the same magnitude slip, and it's a much harder challenge so expecting the same level of slip should probably be a best-case scenario not a median-case.
Would you take $33 of mine to a charity of your choice vs $100 of yours to a charity of mine? (Probably just Givewell or a top pick of theirs) Official judgement based on whether there's a Starship-derived upper stage en route to Mars by July 2029 (if they're running late SpaceX might try some kind of Hail Mary pass after the best of the launch window has passed) but more likely February 2029 if they launch something on time, or I'll call it at the end of 2028 if they clearly have zero plans to launch anything. Yeah, I just figured out that "33%" was motivated reasoning on my part, but if we keep the bet small enough to just rub my face in a loss then I'd be in anyway.
IMHO part of why SpaceX has been a success and e.g. Blue Origin (with more investment and a head-start) hasn't yet is that Musk's employees implicitly asked him to go to Mars. At some point I guess people were willing to work crazy hours at SpaceX for barely-competitive wages because the stock options made up for it, but at least in the beginning the only thing SpaceX offered employees was the promise of being able to just get important things done, not to just eventually co-chair a committee to review the recommendation to change the color of the book of regulations against doing things. The list of "first privately-funded X" (liquid rocket to orbit, spacecraft recovery, ISS, GEO, humans to orbit) and then "first X" (booster landing, ocean booster landing, rocket with a 120-launch success streak) and "most powerful X" (operational rocket, rocket), while Constellation and SLS were turning into dead ends, keeps the dream alive. If SpaceX ever pulls a bait-and-switch on that, and just focus on e.g. Starlink as a cash cow while ignoring Mars, eventually their best people will go elsewhere and they'll rot Boeing-style from the inside out.
At this point SpaceX is the investor, buying back $500M of their own shares last year, and at the rate Starlink is growing (7 million subscribers now, up from 6 million in June and 5 million in Feb) they're not likely to change that soon. They're still letting employees sell their shares to outside investors too, but AFAIK the last time they issued new shares for investment was Jan 2023.
The existing investors could turn on Musk, and I'd expect a shareholder lawsuit if he gets Spruce Goose "the next Starship will be made out of wood!" crazy, but right now he's still reportedly got the majority of voting shares, and "we're mad because the company that's been talking about going to Mars for decades is going to Mars" probably wouldn't even make it past a Delaware court.
I'm hoping the competition can provide a tolerable alternative, but so far the best out there is Kuiper, 100 satellites launched (out of a planned 3236), half via the cancelled Atlas V rocket and the other half via Falcon 9. Even with Falcon launches, Kuiper has an upcoming July 2026 deadline to launch the first half of their constellation, and I don't think they're going to make it. Hopefully Trump is still pissy enough at Musk that his FCC will waive the "may result in Kuiper’s authorization being reduced to the number of satellites in use on the milestone date" consequences.
Plus, the competition isn't even yet proposing an alternative for Starshield. SpaceX had put up several hundred commercial satellites and begun paid service before they even started putting up the military sats.
I guess there's still a lot of time between now and the next administration. New Glenn isn't even planning to launch its first KuiperSat load until "mid-2026", but by 2029 they could really be in business. There's not a lot of time between the next inauguration and the subsequent launch window, though. If SpaceX actually is prepping for a Mars launch in February 2029, I'd be astonished if the Dem's "First 100 Days" list in January 2029 was topped by "1. From Hell's Heart, We Stab At Him."
It's not so much motivated reasoning as trusting your gut, and a big part of making your gut reliable is being able to tell the difference between what you think is true, and what you want to be true. While at this point I do have some ego invested in this, I think being right would be worse for me than being wrong. All I win if I'm right is an ego boost, but I lose one of the biggest social media platforms that single-handedly turned the socio-political tide away from a thousand years of darkness that I was foreseeing. If l'm wrong all I lose is some ego, but gain the ability to go on a moon fly-by cruise, or some crazy shit like this.
Now you may say that your decision is also not motivated, and you're just trusting your gut. That's fine, may the best gut win.
It's not about "Elon Time", it's about "Elon Hype". The problem with the Hyperloop wasn't the timing, it was that the idea was retarded. Same with the Cybertruck. The Boring Company might have been cool, if it actually delivered super-cheap tunnels, but I don't think they really outperformed anyone on the matter of costs. Tesla has a whole bunch of products in the pipeline now that were announced as revolutionary, just as the Cybertruck was, and are likely tu suffer a similar fate. A Robotaxi that needs a worker constantly holding his hand at an emergency shutdown trigger is no Robotaxi. This stuff is going to keep repeating with FSD, Cybercab, Semi, and Optimus.
Now maybe, just maybe, SpaceX still has the mojo, but I'm not counting on it.
Yeah, that's my preferred way of dealing with it as well, for privacy reasons.
That's an interesting take, I suppose it would explain why he keeps making these "Mars update" speeches. OTOH, I don't think there's a lot things that could demotivate you more, as working for someone who keeps promising insane achievements are just around the corner, while being the grunt charged with actually achieving them, and who knows exactly how far away you actually are from it. Ask me how I know.
That kinda makes me think that the buyback was about maintaining control, rather than any sort of investment (and strictly speaking, how could it be otherwise? They've spent money that could have gone on development, in order to buy paper).
Isn't that underperforming relative to what was promised to investors? I think I heard the somewhere it should have been 20 million by now.
We've got a deal, then! See you in 2029 (other geopolitical events, the existence of TheMotte, etc, willing). Want to let me know a "default" charity pick for if and when you win, in case I can't track you down then?
In an estimate from ten years ago, when they were trying to round up a bunch of investment at a $12B valuation? Yeah, they're behind that schedule, but it's hardly a promise. Investors who want a promise 10 years out are currently happy to buy T-bills at 4.25%. Probably some investors gave up during the latest ($400B valuation) buyback, and have had to dry their tears with wads of $50s instead of $100s.
That said, I'm not a SpaceX investor (except indirectly via Alphabet) and don't plan to be; I'm just refuting the idea that they're dependent on continuing investment for continuing R&D funding. Their investment over their whole history appears to be less than their revenue in 2024 alone.
Right now it would either be Genspect or Themis Resource Fund, I think the latter should stay relevant even if, god willing, the whole trans mania finally blows over. If not, some kind of uncucked Free Software org, but I don't have specific recommendations here, since they tend to be subject to corporate and progressive takeovers.
You?
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link