Friday evening, the biggest rocket ever built flew again, and this time the upper stage came home in one piece. In the ocean, sideways, floating belly up. Still counts.
What happened
Starship lifted off from Starbase, Texas at 6:51 p.m. EDT on July 24, only the second flight of the bigger V3 design [2]. On its suborbital path the ship deployed 20 next-generation Starlink V3 satellites, the large flat ones with laser links, and the last two even switched on lights and looked back at the ship as they drifted off [2]. They then fell back and burned up, by design [2].
Then the good part. The ship relit its engines, came back through the atmosphere, flipped, did a landing burn and made what SpaceX's Dan Huot called its "softest splashdown" ever, before tipping over to float belly up in the Indian Ocean [2]. In between, the ship relit one of its Raptor engines in space, a capability future ships need to get into and out of orbit [2]. "This is a dream scenario for this team that's trying to get this heat shield data," Huot said [2].
The booster had a harder day. Space.com reports only five engines appeared to fire during its landing burn when 13 were needed, and it hit the Gulf harder than planned [2].
Why "intact" is the headline
Reusability is the entire Starship business case. A rocket you throw away is an expensive firework. A rocket you land, inspect and fly again is an airline. Starships usually explode after they tip over, as Space.com dryly notes [2]. Getting one back whole means engineers can look at the actual heat shield instead of guessing from telemetry. That's how you get from "it survived" to "we can fly it next week."
Why the Starlinks matter to normal people
Those 20 satellites weren't headed for service; the flight path was suborbital on purpose [2]. Starship is designed to carry 60 of them at a time [2]. But proving the deployment system works for the big V3 satellites is step one toward putting a lot more internet capacity in orbit per launch. If you're a satellite internet customer, this is the plumbing behind future speed and capacity.
The fair caveat
One great flight is not a program. The previous V3 flight in May had serious engine issues [2], and the booster landing on this one was rough. Space is hard. Twice as hard when you're also trying to make it reusable.
Bottom line
Flight 13 was the most encouraging Starship test yet: payloads out, engines relit, ship home in one piece. The booster still owes everyone a better landing. If you like watching engineering happen in public, this program remains the best show on Earth, and just above it.