3 ms·
SpaceX is still struggling with launching frequently and reliably. I'm hopeful that they will solve those problems, but until they do, other launch providers wi
by mikeash 10y ago
SpaceX is still struggling with launching frequently and reliably. I'm hopeful that they will solve those problems, but until they do, other launch providers will have a leg up for customers who really need their payload not to be destroyed and really need it to be launched on time. Keep in mind that some of these satellites being launched cost hundreds of millions of dollars, so paying a few tens of millions extra for a more expensive launch provider isn't necessarily a big deal.
- jacquesm 10y agoThat's true, but overall their failure rate is really not all that bad. Only the Russians are doing better and they're doing better because they are not innovating at all. I could 3 out of 33 as failures, one of those is a partial failure. The only area where SpaceX really can't compete yet is manned flight, everything else looks quite rosy.
- mikeash 10y ago10% is pretty bad. ULA has launched over a hundred times with no failures. The much-maligned Shuttle had two failures in 135 flights (and, strictly speaking, only one of those failures would have prevented a payload from getting to orbit). I'm pretty sure the Russians are the only ones doing worse right now.
- jacquesm 10y agoBut you'd have to compare their first thirty launches with these to compare. Anyway, notice has been served today I think, now let's see what it leads to. For comparison: https://en.wikipedia.org/wiki/List_of_Atlas_launches https://en.wikipedia.org/wiki/List_of_Atlas_launches https://en.wikipedia.org/wiki/List_of_Thor_and_Delta_launches https://en.wikipedia.org/wiki/List_of_Thor_and_Delta_launche... https://en.wikipedia.org/wiki/List_of_Falcon_9_and_Falcon_Heavy_launches https://en.wikipedia.org/wiki/List_of_Falcon_9_and_Falcon_He... Of course the beginnings of what is now ULA were much more innovative than what SpaceX is doing because we simply had not launched any rockets of that size at all and so there were many more failures. It probably won't be the last SpaceX failure either because they plan on making fairly big changes soon and those will come with new failure modes, the best way to get to very high degrees of reliability is to stop making changes other than to fix problems but that does not square with SpaceX's way of doing business.
- mikeash 10y agoWell no, you don't have to compare to the first thirty launches with those when we're talking about competition for launch services. Somebody choosing a launch provider for their satellite today doesn't care what the reliability of the Atlas was in the 1960s, they only care what it is today. If I was arguing that "SpaceX sucks and they deserve to fail because they're so unreliable" then you'd have a point, but I'm not.
- jacquesm 10y agoI think I covered that with: "Of course the beginnings of what is now ULA were much more innovative than what SpaceX is doing because we simply had not launched any rockets of that size at all and so there were many more failures." So of course the situation is not 1:1 comparable. But given that it is a new platform some failures were to be expected, it doesn't take a lot in rocketry to have a failure. How many launch failures would you consider to be acceptable over the course of the first 30 launches for a new platform?
- mikeash 10y agoYou're confusing my comment with some sort of value judgment. I'm not saying it's good or bad. I'm just saying that some customers care more about reliability than cost, and those customers will provide a steady stream of revenue for ULA and friends until SpaceX improves in this area. I have no opinion about what is "acceptable" or not, but my opinion doesn't matter. What matters is the opinion of the customer. There are enough customers out there which find SpaceX's ~10% failure rate to be too high to keep ULA around for now even at a hefty premium.
- jacquesm 10y agoYes, that makes good sense. So then the race is on in a way: will SpaceX be able to get their next platform stabilized before the competition catches up to being able to re-use their first stage? If they can there might be some actual competition which should drive down prices even further.
- rrmm 10y agoEven worse than the expense of the payload is the time: some payloads take several years to build and some literally can't be replaced (e.g. Webb telescope, though it could be replaced it probably wouldn't get funded).