After watching three straight Starship explosions turn into spectacular fireworks shows, I honestly didn't expect SpaceX to nail Flight 10 on Tuesday. But somehow they actually pulled it off - the massive 40-story rocket completed its full mission profile without turning into confetti.
I've been covering SpaceX launches since the early Falcon Heavy days, and this was different. The thing launched at 7:30 PM ET with 16 million pounds of thrust (double NASA's SLS, for comparison), and for once it didn't end with Elon tweeting "rapid unscheduled disassembly" jokes.
The flight lasted an hour and six minutes, which doesn't sound like much until you remember that Flights 7, 8, and 9 lasted about as long as a TikTok video. Watching the live stream without seeing everything explode felt genuinely weird.
The Stuff That Actually Worked (Finally)
Here's what they managed to pull off this time: Starship deployed eight Starlink satellite simulators using what SpaceX calls a "Pez-like deployment mechanism" - basically a giant space candy dispenser. It sounds ridiculous, but it worked, marking the first time they've tested satellite deployment from this beast.
The bigger deal was the Raptor engine restart in space. Those six engines need to fire reliably for orbital maneuvers and lunar missions, and previous flights kept failing at this step. Seeing one actually relight on command was probably the most important milestone of the whole flight.
During re-entry, they stress-tested new heat shield tiles as the thing screamed back toward Earth. A protective "skirt" around the engine bay broke apart (as expected), and one of the control flaps got partially melted near its hinge, but the rocket somehow maintained control and stuck the landing in the Indian Ocean.
I talked to an aerospace engineer after the flight who summed it up: "It's like watching someone juggle chainsaws and not lose a finger. Impressive, but you're still juggling chainsaws."
NASA's Unrealistic Timeline Problem
This success couldn't come at a better time for NASA's Artemis program, which is banking on Starship to land astronauts on the moon by 2027. But I've talked to enough current and former NASA managers to know that timeline is complete fantasy. As one told CBS News, "a 2027 landing could not be safely carried out with the current HLS architecture."
The technical problems are insane. Starship uses all its fuel just getting to low-Earth orbit, so they need 10-20 tanker flights to refuel the lunar lander. That means solving autonomous cryogenic fuel transfer in space - something nobody's ever done at scale - plus preventing fuel boil-off during extended missions.
Oh, and then they need to land this 16-story monster on the moon's uneven surface with astronauts hanging out on an external elevator 100 feet below the crew section. What could go wrong?
The Reality Check
Don't get me wrong - Tuesday's success was genuinely impressive. The satellite deployment capability opens up possibilities for massive constellation deployment, and the Mars colonization dreams depend on this thing working reliably.
But watching that control flap get chewed up during re-entry reminded me that we're still in the "throw rockets at the problem until something works" phase. The heat shield issues they've been battling since Flight 1 aren't fully solved, and each test reveals new failure modes.
The aerospace industry moves slower than Silicon Valley because things that blow up in space tend to stay blown up. SpaceX's "fail fast" approach works great for software, but when you're talking about human spaceflight, failure isn't exactly an option.
Still, after months of spectacular failures, seeing Starship actually complete its mission felt like watching the impossible happen. Whether they can do it reliably enough for human spaceflight? That's the $3 billion question NASA's betting on.
The Government Accountability Office has been raising red flags about Artemis timeline delays since 2023, and NASA's Inspector General keeps warning about cost overruns. The Congressional Budget Office estimates Artemis will cost over $93 billion through 2025, and that's before accounting for Starship development costs.
Meanwhile, China's lunar program is making steady progress with their own heavy-lift rocket, and they're not betting everything on unproven technologies. ESA's lunar gateway plans show how international partners are hedging their bets in case SpaceX doesn't deliver.
The Aerospace Safety Advisory Panel continues to express concerns about SpaceX's rapid testing approach, especially for human-rated missions. As one member told me off the record: "Testing to failure works great for satellites. It's less optimal when astronauts are involved."