alt Hacker NewsStarship V3
Comments
He again mentions data center in space.
He has to be the biggest richest idiot on the planet.
It should be a lot cheaper to just buy massive solar (wait, couldn't he just make them himself with his tesla roofs?) and batteries (which Tesla also makes) and put Datacenter in some dessert and put fiber to that place...
But it seems he needs some angle to push all this necessary investment into something?
Are we now in the phase of 'lets play scifi' just because we can't come up with anything else?
Btw. Starlink is already 'cheap', with only 8-10 Million customers and doesn't scale easily. So that will not just be able to keep up with his mars stuff...
Oops, I read too far and come across these bangers in the next post:
> In the long term, space-based AI is obviously the only way to scale.
> My estimate is that within 2 to 3 years, the lowest cost way to generate AI compute will be in space
Yes, Elon is very sane.
I really hope I get to see a permanent settlement on Mars or the moon. I don't care who settles it I just want to see humanity reach for the stars.
These datacenters in space will become space junk. Doesn't seem to be sustainable. With a million of these it's hard to imagine why this would be a good idea, and it starts looking like insanity.
Those Raptor 3 engines are a thing of beautiful simplicity compared to their forebears...
Gotta pump that Grok IPO /s Seriously though, the whole SpaceXAI makes zero sense to me. SpaceX was a wonderful company and there was zero need to pollute it with Twitter and a service that creates sexual images of people without their consent.
The new more powerful engines with built in heat shield are a phenomenal achievement. Hopefully they perform as good as they look!
"My estimate is that within 2 to 3 years, the lowest cost way to generate AI compute will be in space"
Soooo, how much did he put on that outcome on polymarket?
Close ups of the tail fins and the hull exterior have little hex tiles covering the entire tail fin assembly. There's also different sizes of tile. Exciting to see if that will be enough structural reinforcement.
Pay your taxes, Elon
The thing which is seemingly missing from this is their current largest hurdle emerging from the V2 testing. The heat shields keep failing.
I guess the focus is going to be on getting stuff up, rather than back down. Thus the Starlink and data center plays, not human space exploration.
Saw some photo's and the first thing that stands out: the American flag. What's up with that? If you see a product launch in Europe there will be no flag in the photo's (non that I ever saw).
Incredible to get insight into the new things they're trying. Back in the day of the old Space Race this kind of thing was impossible and now an enthusiast can just follow along as incredible feats of engineering are performed. Great stuff!
I imagine at least some of the reason to chase the AI datacenters in space thing is because Starship is "too capable" if it succeeds. It makes available a technology that does not have a short-term utility that people will pay for. Starlink was something that's been useful as telecoms but perhaps that market is saturating. It makes sense to pursue what is currently high-utility but is not being met because of terrestrial constraints.
Well, good luck to him. A lot of smart people are chasing this idea and I can't seem how it could work, but I was honestly surprised that Tesla hit its production goals, and I was honest surprised that SpaceX hit success so fast, and I was honestly surprised by the rise of LLMs, so the truth is there are lots of paradigm shifts I just miss: BEVs, cheap space, AI.
Someone once tweeted something like:
> Less intelligent people perceive more intelligent people as incredibly lucky. They always make inscrutably stupid decisions, unjustified by visible information, and somehow fate rewards them for this.
But also, I'm just hoping that a new era of space exploration will open up in my lifetime. That sounds incredibly cool! And I dare say there are many people like me in the US at least judging by the popular baby names of this era, which have seen spikes in Aurora, Nova, and Luna - and in the one my daughter has: Astra.
Page banned in my country apparently
One more week
> Liftoff will occur at 6:30 p.m. ET on Monday (May 19)
https://www.space.com/space-exploration/launches-spacecraft/...
Definitely some cool photos of Starship V3 - how much of this is new info vs just a press release style announcement? I havent been following the latest rocket news much
Your daily reminder that there is no scenario in which putting data centers in space is easier than putting them in Texas, or Morocco, or literally anywhere else.
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 wonder how you solve the cooling issue as you can only shed heat via radiation.
Spacex may be the most important company on the planet. What greater goal is there than expanding humanity to the stars!
Yay, go Elon!
What SpaceX has accomplished is just phenomenal.
V3... the third major redesign, and the second unplanned one. How many verisons will it take? Will Starship beat Full Self Driving into production, or will the hyperloop steal the show? Will it take longer than the Tesla Semi (9 years and counting) or will it pull a Tesla Roadster and never launch at all? Either way, it's sorely needed to meet Musk's goal of landing on Mars by 2018. Or at least to get to zero new cases by the end of April.
I used to follow Starship so intently, similarly NASA things, but Musk's antics, politicising of everything he touches, the increasing use of NASA as US propaganda, has all really put me off it. It's hard to get excited about these things anymore, which is sad because they're otherwise legitimately exciting.
I was disappointed when this was not the command line prompt library
It's a fascinating design but it's been 14 years since the concept was first announced and it's never really completely worked. If it ever was possible, it's not clear the talent for it still works for SpaceX.
Reading reports of people objecting datacenters build in their states I wonder how Florida residents feel about the Spaceport ? It will certainly be more distruptive than datacenters.
Quick update for the folks passionate about space things (since this thread is full of unrelated comments):
V3 is their first Starship family big upgrade, containing lots of learnings from previous tests, and the big engine upgrades. V3 engines are the first iteration of a production engine, with lots of sensors and auxiliary systems integrated into the engine itself. Besides the improvements in thrust, they've streamlined the production, moved a lot of stuff "inside" the engine (the first iterations looked like something out of the steampunk era), and they've simplified lots of fire/heat protection.
The Booster and Ship also got some major redesigns in the way they're handling fuel, the "thrust puck" (the area where the engines get mounted) and so on. It's also a bit taller, helped by the engine upgrades. TWR has also improved, with estimates at 1.6. This should be visibly faster to clear the tower and "jump" the launch.
They are also adding ~44tons of simlinks (starlink simulators, dumb payloads). So they seem to have improved the margins for orbital payload a lot. New this launch will be a few sats that have comms & cameras on them. Hopefully we'll get to see outside shots of Starship from these things, on orbit. They've filed FCC paperwork for this, and they'll likely use it to inspect the health of the heatshield on orbit.
They've also updated the launch tower, with a flame deflector, and a new deluge system.
This flight will be still suborbital, testing payload deployment, booster return to a fixed point somewhere in the coastal waters, and the ship aiming for somewhere in the Indian Ocean. They've also removed some parts of hte heatshield, to test how it handles that. (on a previous flight the ship still nailed its simulated landing with huge gaps in it, from multiple tiles missing intentionally).
If everything works on this flight, the next one is planned to be orbital.