[ploxiln]

I didn't know the SRB components were also re-used!

It's a useful clarification that the Space Shuttle program had a large amount of reuse in the design ... but it still cost over a billion dollars for each reuse, many multiples of e.g. fully expended soyuz. So until SpaceX proved that reuse could be done well, and economically, it wasn't obvious. Now it's super obvious and practically all space programs are frantically trying to catch up.

(The SS could carry fairly large ISS module payloads into orbit, while cheaper rockets/spacecraft couldn't ... but couldn't get to high orbits or beyond ... details details.)

[_0w8t]

There are estimates that the production cost of Soyuz is less than 20 millions USD, https://space.stackexchange.com/questions/48592/what-is-the-...

It is not clear if SpaceX is there per flight. They still need to spent several weeks on refurbishing of the landed booster.

[Faark]

We should keep in mind that Soyuz is the name of both a manned space capsule (like dragon / cst100), but also a rocket family (like falcon).

You link seems to be specifically about Soyuz MS, a capsule. No idea how much SpaceX spends on refurbishing dragon.

[Robotbeat]

Several weeks on-site vs years in a factory on the other side of the country. It’s a clear win. Marginal per launch cost is on the order of $15 million, including the price of a new upper stage. More full cost including GSE and amortization on the booster and fairing more like $25 million.

[zugi]

You are right that no one knew that Space X's approach would work until they demonstrated it. However, it was clear that the shuttle's "reuse" approach was a dismal failure early on - even before it was built, the sheer mass shuttled to and from orbit every time was a giant red flag to anyone who understands the basic rocket equation. That also drastically limited its altitude and usefulness. In a world where practicality matters, it should have been abandoned early on in favor of a return to expendables, until we figured out something that actually worked.

[justapassenger]

No outsider knows if Falcon reuse is economical. SpaceX keeps on raising money (they have a lot of going on, so that’s not surprising) and without insight into financials we don’t known if their launches are subsidized by VC money.

[sdifognoinio]

I think another factor many people neglect is that SpaceX benefits from economies of scale. SpaceX launches very large rockets and launches them very often. Large rockets and a regular cadence are both known to reduce the cost per ton to orbit. SpaceX has the ability to put a whole lot of tonnage in orbit and this immense capacity means they can do it cheaply.

However, this only works if there is enough demand to use all their capacity. At the moment this demand simply does not exist. SpaceX manufactures its own demand. The majority of its capacity goes to launching its own satellites. This lets it keep its cadence, but literally burns the company's own resources to do so.

Whether or not you think Starlink makes any sense in the long run, at the moment it is losing money hand over fist. It serves as an indirect subsidy for SpaceX's other customers.