SpaceX is set to fly its biggest rocket again today — for the 13th time, on purpose

Space 3 min 80 sources

Starship Flight 13 aims for orbit this evening carrying real satellites, while China and Japan race to copy the reusable-rocket trick SpaceX pioneered.

Key takeaways

SpaceX aims to launch its giant Starship rocket for the 13th time today, carrying real satellites for the first time; several past flights ended in explosions, which is by design — each flight is built to learn.
Reusing rockets, once treated as fantasy, is now spreading fast: China caught an orbital booster in a net at sea, Japan flew its first reusable test rocket, and SpaceX reused a booster for the 600th time.
After ten years of watching, astronomers directly photographed the faintest planet ever seen from Earth, near the star Beta Pictoris — a result of patience, not power.

The rocket that keeps flying to break

SpaceX is scheduled to launch Starship — the largest and most powerful rocket ever built — this evening, in a test window that opens at 5:45 p.m. CDT (22:45 GMT) from Starbase, Texas [3][27]. It is the 13th full-scale test flight of the vehicle since 2023, and the second to use the newest design, Starship V3 [19][27]. The Super Heavy booster underneath it is powered by 33 Raptor engines, all of which SpaceX fired together in a ground test days ago [5][19].

What makes this flight different: for the first time, Starship is carrying real cargo. Twenty working Starlink V3 internet satellites sit in its bay, to be pushed out one at a time through a side hatch by a system of pulleys and cables that engineers call the “Pez dispenser” [19][9]. Earlier flights carried only dummy weights that mimicked the satellites’ mass [19]. SpaceX has published what went wrong on the previous flight, in May, and says it has fixed it [3].

That is the pattern worth noticing. Starship is not a finished machine being shown off; it is a machine being found out. Each flight is built to learn something, and several past flights have ended in explosions. The company can afford this because it builds its own hardware fast and flies over empty ground and ocean. When it works, the payoff is large: Starship is designed to lift more than 100 metric tons to low Earth orbit — the zone a few hundred kilometres up where most satellites live — enough to reshape what NASA, the military, and scientists can put in space [9].

The reusable-rocket trick is spreading

SpaceX’s other big idea — landing a rocket’s first stage to fly it again, instead of throwing it away — is now being copied around the world.

On July 10, China launched its new Long March 10B rocket from Hainan island and recovered the first stage 11 minutes later, catching it in a net on a ship at sea [6]. State media confirmed the controlled, powered descent; China’s main space corporation says it aims to reuse that stage before the end of 2026 [6][30]. It makes China the first nation besides the United States to recover an orbital-class booster — a club that until now held only the American companies SpaceX and Blue Origin [6].

Japan’s space agency, meanwhile, flew its own experimental reusable rocket for the first time [41], and the American firm Rocket Lab test-fired the Archimedes engine for Neutron, the larger reusable rocket it is building to compete with SpaceX’s workhorse Falcon 9 [2]. SpaceX itself passed a quiet milestone this week: it flew a reused booster for the 600th time [8]. A decade ago, reusing a rocket was treated as a fantasy. It is becoming the industry’s baseline.

Big money, small planets

The commercial side is running hot. Blue Origin, Jeff Bezos’s rocket company, is reportedly seeking $10 billion in new funding at a $130 billion valuation [18] — a sign of how much investors now bet on space paying off. On the ground, a NASA astronaut, Anil Menon, rode a Russian Soyuz to the International Space Station on July 14 alongside two cosmonauts [7], and NASA’s Artemis II Moon flight drew 149 million views of its coverage as the agency preps the follow-on Artemis III crewed landing for 2027 [10][44].

The prettiest result came from patience, not money. After ten years of looking, astronomers photographed the faintest planet ever seen directly from Earth — a world around the young star Beta Pictoris, roughly 100 times dimmer than its already-famous neighbour [15][21]. Spotting a planet next to a star is like finding a firefly next to a lighthouse; it took a decade of watching and a pair of the world’s best telescopes to pull this one out of the glare [15]. No spectacle, no explosion — just ten years of care, and then: found you.

02 · Lesson · why it matters

A test that can't fail teaches you nothing

A test only teaches you when it's allowed to fail — so how fast you learn depends on how cheaply you can afford to be wrong.

A rocket built to break

Tonight SpaceX plans to fly the largest rocket ever built, for the thirteenth time. Several of the previous twelve ended in fire. That is not the embarrassing part of the story — it is the plan. Starship isn’t a finished machine being shown off. It is a machine being found out, one flight at a time.

That cuts against how most of us think about a test. We treat a test as something you pass. You study, you prepare, and a good result means you got it right. A launch that blows up looks like the opposite of success. But watch what SpaceX actually does with each failure, and a different idea comes into view — one that reaches far past rockets.

The surprise is the whole point

The value of a test is the information it hands you. And information only lives in the surprises. A flight that goes exactly as expected confirms what you already believed; you end the day knowing nothing you didn’t know at breakfast. A flight that fails tells you something new and specific: this weld, this valve, this assumption was wrong. The explosion is expensive, but it is also the lesson arriving.

So SpaceX flies partly to find the next thing that breaks. Each failure isolates one flaw, and that flaw gets fed into the next design. Fly, break, read exactly what broke, fix it, fly again.

If that loop sounds familiar, it should. It is how anyone gets good at anything hard. The cook who has burned the dish, the player who has lost the match, the founder whose first idea flopped — each learned from the attempt that went wrong, not the one that went smoothly. The tight loop between trying and correcting is the engine of getting better. You are inside it every time you learn something the hard way.

Who gets to fail cheaply

But “just try it and find out” is easy advice when a lost rocket costs you a week. It is ruinous advice when a failure carries a crew, or a decade of budget, in the nose cone.

NASA’s Moon programme cannot iterate the way Starship does. Not because its engineers have less nerve — because its failures kill people and burn billions, so it has to try to get everything right the first time. SpaceX can court failure only because it built a structure that makes failure cheap: its own factory turning out rockets fast, an empty test range over open water, the whole machine in-house. Break one, build another next month.

This is the part that hides in plain sight. The “move fast and break things” spirit poses as universal wisdom, a lesson anyone can apply. It is really a luxury of a particular cost structure. The freedom to fail is handed out unevenly — and the person urging you to take bigger risks is often someone whose own failures are cheap. Before you copy a bold strategy, it is worth asking what it costs that person to be wrong, and whether it would cost you the same.

The map of ground you’ve already crossed

And here is the catch that binds even the people running the loop, the ones who seem to have it figured out.

A failed flight reveals exactly one thing: what broke this time. It says nothing about the flaws still waiting further down. Thirteen flights in, SpaceX knows precisely what went wrong on twelve of them. It does not know what will go wrong on the fourteenth. Learning by failure gives you a map of the ground you have already crossed — never the ground ahead.

We carry the same blind spot into everything we learn by living it. Experience is a record of the mistakes you have already made, not a warning about the ones still in front of you. The method that teaches you fastest is also the one method that can never tell you what it hasn’t taught you yet.

A rocket that has flown clean three times isn’t flawless. It just hasn’t hit the next thing. The engineers know this, which is why they keep testing long after it looks safe. We are in the same position about almost everything we think we’ve figured out — a little further along a road whose end we cannot see, mistaking the quiet stretch we’re on for the whole of it. Seeing that clearly doesn’t make the next test less worth running. It just makes it worth holding our certainty a little more loosely.

03 · Lab · your turn

Test to Break

Rehearse how the cost of failure decides whether to learn by flying and crashing or by testing first — and feel that "proven" never means flawless.

04 · Hope · carry this

Every rocket that broke on the way here taught the next one how to fly. Progress has almost always looked like a heap of failures — right up until enough of them add up to something that finally works.