Never been attempted at this scale in space
On the Texas Gulf coast, humanity's most powerful rocket prepares once more to meet the sky — not merely as a test of engineering, but as a wager on whether the moon can be reached by 2028. SpaceX's twelfth Starship flight carries with it the weight of NASA's Artemis program, a redesigned machine built from the hard lessons of eleven previous attempts, and the unresolved question of whether ambition and iteration can outpace the unforgiving physics of deep space. The flight is a single step in a long staircase, but the staircase leads, if all goes well, to the lunar south pole and the return of human footprints on another world.
- Version 3 Starship arrives with 18 million pounds of thrust and a body rebuilt from failure — but it has never flown, and the clock toward a 2028 moon landing is already running.
- Twice delayed by wind and rain, the rocket finally targets Thursday evening for liftoff from SpaceX's brand-new second launch pad, adding yet another untested variable to an already complex mission.
- NASA is quietly hedging, nurturing Blue Origin's rival lunar lander in parallel — a tacit acknowledgment that Starship's readiness is far from guaranteed.
- The flight will stress-test autonomous satellite deployment, heat shield inspection via onboard cameras, and a re-entry banking maneuver designed to rehearse the return of future operational missions.
- Beyond Thursday, the truly unsolved problems loom: transferring thousands of gallons of cryogenic fuel between spacecraft in orbit, and landing a 170-foot rocket autonomously in the shadowed, uncertain terrain near the moon's south pole.
SpaceX is preparing to launch its upgraded version 3 Starship from a new pad on the Texas coast — the twelfth integrated test flight of the world's most powerful rocket system. Built from lessons absorbed across eleven previous flights, the redesigned vehicle carries 33 methane-fueled Raptor engines generating 18 million pounds of thrust at liftoff, roughly double the output of NASA's Space Launch System. The stakes reach well beyond SpaceX: NASA has committed to using a human-rated Starship to land astronauts on the moon by 2028 under its Artemis program, a deadline that depends on SpaceX proving reliable performance across many more flights to come. As a hedge, NASA is simultaneously developing Blue Origin's alternative lunar lander, and its Artemis III mission will test rendezvous and docking procedures using whichever vehicle is ready.
Already delayed twice by high winds and rain, the rocket is set to lift off Thursday evening. The Super Heavy booster will separate and splash down in the Gulf of Mexico — a conservative choice for a maiden flight of heavily redesigned hardware — while the upper stage loops halfway around the planet before splashing down in the Indian Ocean about an hour after launch. Along the way, the spacecraft will deploy 22 Starlink satellite simulators, execute a banking maneuver during re-entry to rehearse future return flights, and use onboard cameras to photograph its own heat shield, transmitting images back to validate inspection methods.
The harder challenges lie ahead. Before any moon mission can proceed, Starship must be refueled in Earth orbit — a process requiring multiple tanker flights to transfer thousands of gallons of supercold cryogenic propellant between spacecraft, a feat never attempted at this scale. How SpaceX will prevent those propellants from boiling off in the vacuum of space remains publicly unresolved. Once in lunar orbit, astronauts would undock from their Orion capsule and descend autonomously in the 170-foot Starship to land near the moon's shadowed south pole — terrain that is poorly lit and incompletely mapped. NASA requires at least one unpiloted test landing before any crew boards. Thursday's flight will offer the first real evidence of whether version 3's redesigns mark genuine progress toward that distant, demanding destination.
SpaceX is preparing to launch an upgraded version of its Starship rocket on Thursday evening from a new pad on the Texas coast, marking the 12th integrated test flight of the world's most powerful rocket system. The version 3 Starship represents a significant leap forward—equipped with redesigned engines, structural improvements, and safety enhancements built from lessons learned across eleven previous flights that ranged from impressive successes to dramatic failures. The 33 methane-fueled Raptor engines powering the booster are lighter and more powerful than their predecessors, capable of generating 18 million pounds of thrust at liftoff, roughly double the power of NASA's Space Launch System.
The stakes extend far beyond SpaceX's own ambitions. NASA has committed to using a human-rated version of Starship to land astronauts on the moon by 2028 as part of its Artemis program, a deadline that hinges on SpaceX demonstrating reliable performance across dozens of test flights between now and then. The space agency is not putting all its eggs in one basket—it is simultaneously working with Jeff Bezos' Blue Origin to develop an alternative lunar lander, a hedge against the possibility that Starship will not be ready in time. NASA's Artemis III mission, scheduled for late next year, will test the rendezvous and docking procedures needed for a moon landing, using whichever lander or landers are ready to fly.
Thursday's flight will be the first to use SpaceX's second launch facility at Starbase, the company's sprawling test site on the Texas Gulf coast. The mission has already been delayed twice this week due to high winds and rain. When it finally lifts off at 6:30 p.m. Eastern time, the Super Heavy booster will separate from the Starship upper stage after reaching the lower atmosphere, then execute a controlled splashdown in the Gulf of Mexico rather than attempting the more ambitious catch by the launch pad's mechanical arms—a conservative choice for a maiden flight of significantly redesigned hardware. The upper stage will continue climbing to a sub-orbital trajectory, looping halfway around the planet before re-entering the atmosphere and splashing down in the Indian Ocean roughly an hour and five minutes after launch.
The flight plan includes several critical tests of new systems. The Starship upper stage will deploy 22 Starlink satellite simulators, with the final two equipped to scan and photograph the spacecraft's heat shield, transmitting images back to Earth to validate methods for assessing the shield's readiness for future return-to-launch-site operations. The spacecraft will execute a dynamic banking maneuver during re-entry designed to mimic the trajectory that operational missions will fly when returning to Starbase. Fifty onboard cameras will document the entire flight, streaming imagery back via Starlink's satellite network. SpaceX has emphasized that the primary objective is to demonstrate each new component in the actual flight environment for the first time, incorporating years of development and testing into a vehicle designed for full and rapid reuse.
But the path from test flights to operational moon landings remains treacherous. The most daunting challenge is one that has never been attempted at scale: autonomous refueling of the Starship lander in Earth orbit. Because the rocket burns most of its propellant just reaching orbit, the lander must be topped off before heading to the moon—a process that will require multiple tanker flights to transfer thousands of gallons of supercold cryogenic fuel from one spacecraft to another in the vacuum of space. SpaceX has not yet disclosed how it will prevent these extreme-temperature propellants from naturally boiling off into gas that must be vented away, a problem that could consume significant fuel reserves.
Once in lunar orbit, Artemis astronauts will undock from their Orion capsule and descend in the Starship to a fully automated landing near the moon's south pole—a region characterized by long shadows, poor lighting, and uncertain terrain. The 170-foot-tall rocket will touch down autonomously, after which the crew will descend via an external elevator deployed from the spacecraft's side. After their surface mission, they will ride the elevator back up and blast off to rendezvous with the Orion for the return journey to Earth. NASA's contract with SpaceX requires at least one unpiloted test landing before the agency will risk astronauts aboard.
SpaceX's track record with reusable rockets—particularly the Falcon 9 booster, which routinely lands at sea and on shore—gives some observers confidence the company can pull off the feat. Others, noting the unprecedented complexity of in-space refueling and precision lunar landing, remain skeptical that Starship will be ready by 2028. What is certain is that Thursday's flight will provide the first real-world data on whether the version 3 redesigns represent genuine progress or merely incremental adjustments to a system that may not be ready for the demands ahead.
Notable Quotes
The booster's primary test objective will be executing a successful launch, ascent, stage separation, boostback burn and landing burn at an offshore landing point in the Gulf of America.— SpaceX
The flight test's primary goal will be to demonstrate each of these new pieces in the flight environment for the first time, with each element of the Starship architecture featuring significant redesigns to enable full and rapid reuse.— SpaceX
The Hearth Conversation Another angle on the story
Why does this particular test flight matter more than the previous eleven?
Because this is the first flight of a fundamentally redesigned vehicle. SpaceX has taken everything it learned from eleven flights—successes and failures both—and built those lessons into the hardware itself. It's not just tweaks; it's a new generation.
And NASA is betting the entire Artemis moon landing on this working?
Not entirely. NASA is hedging by also funding Blue Origin's lander. But yes, Starship is the primary path. If it doesn't work, the 2028 timeline becomes very difficult.
What's the hardest part of actually landing astronauts on the moon with this thing?
Honestly, it's the refueling. Getting to orbit is hard. But refueling in space—transferring thousands of gallons of fuel from one spacecraft to another while orbiting Earth—that's never been done at this scale. And then you have to do it multiple times before the lander can even head to the moon.
Has anyone ever done anything like that before?
Progress cargo ships refuel the International Space Station routinely, but that's a relatively stable, well-lit environment with ground support. This would be autonomous, in deep space, with cryogenic propellants that boil off into gas if you're not careful. SpaceX hasn't even said how they'll solve the boil-off problem.
So there's a real chance this doesn't work by 2028?
Many outside observers think so. But SpaceX has surprised people before with the Falcon 9. The company has a track record of learning fast and iterating. Thursday's flight will tell us a lot about whether they're on the right path.
What happens if they're not?
Then NASA lands on the moon with Blue Origin's lander instead, or the timeline slips. Either way, it changes the whole calculus of the Artemis program.