/sci/ - Science & Math

>>12226674
He might have came to the conclusion that solving immortality isn't feasible within his lifetime, and instead wanted to focus on improving things that he can do now.

>>12111593
Reusability and mass production. Falcon 9's core and 2nd stage probably cost about as much if not more to manufacture as an Atlas 5 core, but SpaceX is able to amortize that cost across multiple launches thanks to the landings and recovery. This also goes for the engines, where the sheer, insane volume of Merlin engines needed for Falcon and in particular Falcon Heavy mean that by rocket standards they're practically mass-produced, which again keeps costs low. This isn't unique to SpaceX, as the Soviet Union had similar economies of scale with the NK-33s on the N-1 and Roscosmos today also has similar economies of scale with the Soyuz family which is why they're also dirt cheap to launch. That's the engineering philosophy side of things.

Now, the real SpaceX special sauce is in booster recovery via powered landing, which costs relatively little in terms of materials and added mechanical complexity, but is incredibly dependent on the quality of SpaceX's avionics and flight control software. SpaceX may be similar to ULA, Boeing, JAXA, or Arianespace in terms of their mechanical engineering capabilities, but they absolutely blow everybody else out of the water when it comes to the quality of their programmers and the quality of their software, as their computer science talent is every bit as good as the top CS grads hired by the FAANG companies. As Starliner proved, software is everything, and the programmers at ULA and Boeing don't even come close to the caliber of the Silicon Valley-tier folks working at SpaceX. Things are even worse at JAXA and Arianespace, and you don't even want to think about the quality of the electrical and software engineering staff that the Russians, the Chinese, and the Indians have to make do with. At least they're better off than Iran or North Korea, lol.