[shawabawa3]

> Except it is really, really expensive.

> Mr Musk says the cost of building the route would be in the region of $6bn (£4.1bn), an estimate most agree is extremely conservative.

I'm pretty sure that should say "extremely optimistic". Musk's estimate of $6bn is over 10x cheaper than the high speed train line cost for the same route

[Chathamization]

If I remember right, part of the reason why Musk estimated that the hyperloop would be so cheap was that he assumed it could be built on highway meridians for free, and the tracks were to end end far outside the city centers. It also had a fraction of passenger capacity and number of stations as high speed rail.

It seems likely that for the same amount of service as high speed rail, the hyperloop would be a lot more expensive.

[misnome]

Don't forget that the cost estimate included drilling holes in mountains for magically minature amounts of money than current technology provides.

[lallysingh]

Maybe that's why he's not building it himself?

[freyir]

>> an estimate most agree is extremely conservative.

> I'm pretty sure that should say "extremely optimistic".

A conservative cost estimate is essentially an optimistic cost estimate. A stronger statement might be that it is unrealistic.

[puranjay]

Last I checked, wasn't the project over budget and hitting close to $100Bn in costs?

Honestly, even $50Bn sounds like chump change for a project as ambitious as the Hyperloop. Nearly all large scale public infrastructure projects end up costing billions of dollars over estimated costs anyway.

[tptacek]

If you're saying that the HSR project is over budget, you're right, but you're not helping the Hyperloop argument. The point is that if the HSR people can't get something done for $X, it's unlikely that Musk can do something even more ambitious for $X/10.

[jcranmer]

IIRC, one of the big sources of cheap money is expecting that easements for giant elevated tubes would be available for much less than the land value of the required land underneath.

[tptacek]

The analysis I read, back when the Hyperloop proposal was released, took this into account and still found Musks's cost reductions versus HSR were so profound as to suggest major changes for all of civil engineering.

[elchief]

Like how he's lowering the price of rocketry by 90%?

His last bid was 40% less than United Launch

[ubernostrum]

SpaceX received uncounted billions' of dollars worth of R&D for free thanks to the existence of government-funded space programs, which already solved many of the difficult fundamental problems in the field of rocketry so SpaceX wouldn't have to. And while it's neat to watch the stuff they do, there's very little SpaceX is doing that's revolutionary as opposed to evolutionary. So of course now SpaceX can come in and reduce the costs by building on the existing base of work in the field.

Hyperloop doesn't have anywhere near that amount of free R&D or established, easy-to-increment-to-get-there technologies to draw on.

[Inthenameofmine]

I wouldn't be so sure on that. The R&D transfer and ready human capital was definitely profound in SpaceX's case, but if it was that incremental, why didn't NASA or anybody else do it before?

I think almost the sae can be said for Hyperloop. There's no fundamentally new technology that needs to be engineered for it to work. Maglevs are common in many countries, evacuated tubs aren't complicated.

All that said, the US seems to be the wrong place to start Hyperloop. Regulation and land property will most likely be the biggest hurdle, not technology, especially in California. Hyperloop seems to me like a perfect project in China though, in regard to property rights, regulation, city planning, human capital in the high speed train sector, as well as crazy supply side economics.

[ubernostrum]

why didn't NASA or anybody else do it before?

Why haven't we gone to the moon since the 1970s? Because there wasn't the political will to keep funding the programs that went there. Why was the Space Shuttle such an awkward and ultimately non-viable way to go to space? Because there wasn't the political will to fund genuine reusable launch technology. SpaceX's big advantages are in having the hard up-front R&D already done for it by someone else, and in not having to rely solely on political decisions for the existence of its market; even if there are no more nice-PR resupply missions to the ISS, there are still going to be private actors who want to launch satellites.

[tptacek]

You're moving the goalposts. The argument isn't that evacuated tubes are "complicated". The issue is that the cost estimates Musk provided appear to assume secret Musk space alien technology.

[tptacek]

Why doesn't my laptop battery last 10X longer? Why doesn't my favorite IPA has 1/10th the calories? Can't Elon Musk fix those things first?

[Dylan16807]

That's pretty unfair. He's not making a rocket that's magically 10x lighter, he's making it cheaper. And he is helping make batteries cheaper.

[bobcostas55]

Isn't that exactly what SpaceX did?

[krschultz]

SpaceX entered a market with few competitors, none of which are defacto government agencies. The civil engineering market is much more competitive. Expecting the same level of cost reduction just by having better operational efficiency is naive.

Thus the argument needs to be that Hyperloop is intrinsically cheaper, whether built by Musk's company or any random civil engineering company. I don't see any reason why that is the case, Hyperloop looks a lot more complicated than HSR to me.

[mulmen]

Musk himself didn't expect SpaceX to succeed. It was one launch away from complete failure. You're essentially making the argument "If NASA can put a man on the moon why do I still have to buy toothpaste?"

SpaceX has and continues to do amazing things, that doesn't mean suddenly everything we want is feasible.

[thrill]

They are two different technologies. Budget estimates will involve different tradeoffs with different risk assumptions.

[tptacek]

Civil engineers have looked at the Hyperloop proposal and declared themselves unable to determine how Musk proposes to reduce the costs of long runs of overpass, or of tunneling.

A simpler way to make the same argument:

If Musk can actually build a Hyperloop across California at his proposed cost, he's made revolutionary improvements to civil engineering that will be far more impactful than the Hyperloop itself; the same techniques should revolutionize, well, much of conventional civil engineering! Why isn't that happening?

[fit2rule]

Rockets = big long tubes. Big long tubes = something robots can make easily?

[Avshalom]

The tube isn't the problem. Among various problems include: the building access roads to every pylon base, conducting site surveys and environmental impact studies for each site then figuring out what to do when those studies and surveys declare a patch of dirt to be unacceptable. The original route was also going to need something on the order of kilometers of tunnel through mountains.

[megakwood]

> The point is that if the HSR people can't get something done for $X, it's unlikely that Musk can do something even more ambitious for $X/10

IDK, indeed $X/10 is ambitious but he does have a decent track record with this already. SpaceX is already $X/3 or $X/4 and is on the path to $X/10 with reusable rockets, and it's also more ambitious (rocket reuse, highest payload in the world by 2x, new capsule has powered rocket landing on any surface in the solar system).

In my mind he's earned some benefit of the doubt. Of course, there's still a ton to prove, and the idea sounds looney. You'd also have said that about SpaceX 10 years ago, though.

[tptacek]

Can he make my car get 10X the mileage first? Or my wireless have 10X the range? Argumentum ad Musk doesn't seem like a particularly powerful way to make a point.

[unfamiliar]

It's better to spend $100Bn on something that is tried and tested, than spending $50Bn on a risky proposition that could end up simply not working for whatever reason.

[kumarski]

The golden gate bridge costed about 1.5 Billion dollars when it was built.

$6 Billion is something that I think many civil engineers would consider to be a dubious figure, even as a first estimate. I'm sure Musk is cognizant of this.

[reacweb]

roads are also expensive and trucks (combined with weather) destroy them very fast. Hyperloop cylinder should be far cheaper than trucks in the long run. I think hyperloop would be useful for amazon to reduce delivery delays.