SpaceX puts ten Falcon 9 launches' worth of Starlink capacity on Starship's first orbital attempt

Count and capacity come apart on this payload. Twenty-six satellites against a fleet of about 11,000 active ones is roughly a quarter of one percent by number ^[7][4]. SpaceX says the same 26 deliver about ten times what a Falcon 9 launch of V2 Mini satellites does ^[4]. Taken at the company's word, a Falcon 9 Starlink flight adds about 2.6 Tbps, and this one Starship is meant to add 26 ^[1]. Each V3 is designed for about 1 Tbps, with solar arrays that generate twice the power of earlier generations ^[3][9].

SpaceX went public in June in the largest IPO on record and is valued at about $2 trillion ^[10]. As of the second quarter, Connectivity was its largest segment and its only profitable one, and the company is counting on Starship to scale it ^[8]. Elon Musk said on the August earnings call: "It's not out of the question that at some point, Starlink will deliver a majority of the world's internet, at least in countries where we're allowed to operate, which is the vast majority of countries." ^[11] Gwynne Shotwell, the chief operating officer, said this month that Starship will also launch "AI compute satellites" in 2027 ^[12].

The clean version of Monday has the ship firing its insertion burn about 25 minutes after liftoff, finishing deployment near the 1 hour 4 minute mark and using a single Raptor to deorbit almost nine hours in ^[14]. A messier version has the ship doing all of that while the booster repeats Flight 13, when 8 of 13 planned engines relit for the landing burn and it splashed down hard ^[15]. SpaceX has since changed engine filtering hardware and relight software ^[16]. In the least useful version, flight controllers cannot confirm enough redundancy for the deorbit burn, and SpaceX has said the insertion burn happens only after they do ^[13]. The mission description does not say whether the 26 satellites would still be released in that case ^[13].

I think even the clean version settles half of SpaceX's case. The company designed Starship to be fully reusable and to lift more cargo for less cost than any earlier spacecraft ^[21], and Flight 14 tests the cargo half. Reuse on this mission extends to two heat-shield tiles recovered from Ship 40 and flown again on Ship 41 ^[17]. The booster is aiming for an offshore landing point and the ship for the Pacific west of Chile, so neither returns to Starbase ^[18][2][22]. Three of the 26 satellites carry cameras to photograph the heat shield in orbit, and SpaceX will use the images to judge whether a ship is ready for missions that come back to the launch site ^[19].

The counter-case is that cost matters less than it looks. SpaceX already runs about 11,000 active satellites to roughly 650 for Eutelsat OneWeb ^[7], some 17 times as many ^[2], and a Starship that adds ten Falcon 9 flights' worth of capacity at a time widens that lead whatever each launch costs. I'd give that less weight. Five of the 13 landing-burn engines failed to relight on the last flight ^[3], and cheap, repeated lift is the design premise ^[21]. The satellites could still overturn this reading from the other side. If the V3s do not raise their orbits and start serving customers within the few weeks SpaceX expects ^[5], the limit on Starlink capacity is the satellite, and the rocket question comes second.