Hacker News Reader

Comment by jsf01 | Hacker News Reader

The current Vegas loop is a fun little novelty, like a roller coaster, but it’s so bad as a model for transportation. If we’re already digging the tunnels PLEASE just make a subway. The teslas are so awkward, have to be steered in this small and curved space, require a driver, only fit a few passengers, and on and on. Vegas is in desperate need of higher volume train transportation between the airport, the Strip, and downtown—but I’m afraid this isn’t going to cut it.

View on HN

You’ll never be able to build 68 miles of subway in a place like Los Vegas given the cost structure that exists in America today. The next 1.5 mile phase of the second avenue subway in NYC is going to be $7.7 billion. DC desperately needs more downtown subway capacity to accommodate the new suburban lines it has built. It won’t happen. I think you can count the active subway expansion projects in this country on one hand.

Outside somewhere like New York, a cheap low capacity system that goes more places is better than a high capacity subway that doesn’t go anywhere.

The Second Ave. subway cost explosions are driven by a combination of union grift, overly litigious third parties, onerous design requirements driven by politics rather than engineering, and the requirement of building stations in the middle of New York. The problem in DC is similar but also impeded by the fact that Metro does not want to take responsibility for any new track (see the Purple Line fiasco).

Your point still stands that America's approach to infrastructure (and underground in particular) is costly; however, those costs are not nearly as bad outside of major metropolitan cities. Underground work in the Vegas area is relatively inexpensive in my experience (I have worked on underground projects in all three of these locations within the last decade).

New York and DC are especially bad, but LA still paid a billion dollars a mile for its recent subway expansion. If built as a subway, the 68-mile network would cost at least $50 billion. The gross domestic product of the entire Las Vegas region is just $136 billion. A $50 billion infrastructure project would be like New York City taking on a $750 billion infrastructure project.

You could build a much smaller subway network. But at that point, the value proposition of the network in a sprawling city like Las Vegas evaporates. (Especially in a city as hot as Los Vegas, where long walks from subway stations to destinations will be miserable.) Baltimore is much denser than Las Vegas, and the one underground line has a ridership of just 2 million people annually. Limited subway networks that don’t go anywhere are worthless.

NYC's subway system has nearly 1000 miles of track. The cost of that was spread over years. Similarly, the 68 mile network proposed here will be built over years. Based on Boring Company's own PR, it took them a year to build 1.7 miles of track at the convention center. 68 miles will take them 40 years. Considering the reality of construction schedules and the fact that none of this is permitted or designed and using your $50B number, this will take half a century or more to build. So that amounts to less than $1 billion a year. Does NYC take on $15 billion in infrastructure projects a year? Probably.

I find it entirely plausible that NYC would take on $750 billion in infrastructure projects over half a century. Hell, if NYC were to attempt to build their own subway network again using the Second Ave. number (and conceding that it's inflated), it would take nearly $5 TRILLION dollars. Handicap that number for the inflated cost and we're still looking at NYC having already spent multiples of the amount you are asserting NYC would equivalently have to take on on a project-to-project basis with this.

As to the rest of your post, I agree - this proposed 68 mile tunnel system in a place that lacks the requisite urban density for public transit efficiency will have limited ridership and will be worthless, if it's ever completed. More likely it will be abandoned partway through and will only serve to become a literal obstacle to future tunnels/development.

NYC has not spent $750 billion on subway expansion (adjusted for inflation) even over more than a century. To use the IND 8th Avenue line as a benchmark, the second to last trunk line, the 14 route miles were built for about $4 billion in today’s money. The whole system has 248 route miles, almost all of it built before 1948. Assuming it was built at the same per-mile cost as the 8th Avenue Line, that’s about $70 billion or so in today’s dollars for most of the original network.

The NYC MTA spends $7 billion a year on capital expenditures, not $15 billion. Most of that is maintenance, not expansion. No transit system can spend anything close to 100% of their budget on expansion over that time horizon. So your $1 billion a year is probably more like $2 billion accounting for maintenance. That is more than the entire Las Vegas budget.

It seems like an interesting and worthwhile experiment. It seems almost inevitable that if it works out, the drivers will be removed and they'll switch to tailor-made automated vehicles. It could start to make a lot more sense as a sort of point-to-point automated rail equivalent.

Perhaps a regular metro would suit Las Vega better, but if that's out of reach why not give this a shot? It's Las Vega after all.

I really like this vision for the system. If they can retrofit in some rails and remove the drivers, it suddenly makes a lot of sense. I hope that’s where this ends up!

Since the first one cost about 47 million, it seems reasonable to guess the others will cost some multiple of that. The opportunity cost of several hundred million dollars seems... bad? Just have a subway!

Several hundred million dollars at best gets you a mile of subway in an American city: https://www.vox.com/22534714/rail-roads-infrastructure-costs...

Experimentation with alternatives is desperately needed.

His whole thing for a while was how it was going to be cheaper to dig tunnels by using new techniques etc, that would equally apply to subways - but a feel like it was probably bullshit.

The new technology is smaller drills that make smaller tunnels.

They are electric and can start at street level and dig themselves down.

This is ridiculous misinformation. "Drills" come in every size already, from the 12-inch range all the way up to 60 feet in diameter. All modern "drills" that erect a liner behind them are electric - air quality and safety issues (along with the massive energy requirements) mandate the use of an umbilical from the surface and prevent the use of diesel or gas in the tunnel for these machines. None of Boring Company's "drills" are able to start at street level and dig themselves down, they require a ramp or shaft just like every other TBM. This is physics, the Boring Company has made no technological advances. They literally bought their TBM from someone else and put it in the ground with a new paint job.

Here are pictures of the drills launching at ground level and digging themselves down [1], and a third party source discussing it as an innovation in boring [2]

https://www.boringcompany.com/prufrock

https://www.construction-europe.com/news/6-ways-tunnel-borin... -evolving/8026001.article

This would be called a launch ramp in the industry. It is an adaptation of HDD approaches to larger diameter machines. It has been going on for decades. It is only viable where space and the final geometric constraints of the project permit. Additionally, it requires very specific geologic conditions or the construction of a structural headwall in order to avoid a sinkhole.

For example, what Boring Company is doing has already been done in the extreme limit: Herrenknecht sells a vertical shaft boring machine that is essentially a TBM turned 90 degrees. https://www.herrenknecht.com/en/products/productdetail/verti...

Average subways in the US is. Billion per mile.

This one is reported as no cost to the city.

Subways are incredibly expensive to build. The average subway cost in the US is approximately 1 billion per mile.

Public transportation systems all over the country currently have huge budget gaps because ridership keeps going down.

Meanwhile, this provides 68km of low throughput transport to the city, for the public price of zero.

This is really making good the enemy of the perfect. Especially When the general public doesn't agree on the perfect, or want to pay for it.

>Public transportation systems all over the country currently have huge budget gaps because ridership keeps going down.

The pandemic basically destroyed ridership in a lot of places. Boston--both the subway and the commuter rail--have crept back up to "only" being about 30% below pre-pandemic levels (and has probably been impacted by some disruptive long-deferred maintenance as well). But one of the consequences is reduced schedules which makes public transit relatively less attractive.

> 1 billion per mile

That seems incredibly high. My city in France just finished building a new subway line and the cost was ~100 million € per km (170 million $ per mile)

Not sure why there's an order of magnitude difference?

Now consider 7.7 billion for 1.5 miles:

https://gothamist.com/news/cost-of-second-ave-subway-extensi...

Somewhat related: New York metro unveils $30M staircase

https://nypost.com/2022/05/16/mta-unveils-stunning-30m-stair...

only slightly less egregious, BART completes $41 million pedestrian bridge:

https://www.bart.gov/news/articles/2022/news20220203

It is incredibly high. The US is uniquely bad at building infrastructure at a cost that makes sense.

> Public transportation systems all over the country currently have huge budget gaps because ridership keeps going down.

COVID and remote work did this. Might be less of a concern for a captive tourist audience. A line to the airport and 4-5 big intersections or attractions could be pretty effective.

One issue is that Boring Company is digging tunnels that are 12 ft in diameter. They can be used for subways [1] but it's quite narrow by modern standards.

--- [1] https://en.wikipedia.org/wiki/London_Underground#Deep-level_...

Thats the width of most tunnels in the Tube. I always thought it made sense to have the tunnels be smaller

There is also the Glasgow Subway [0] which uses 11 ft diameter tunnels

[0] https://en.wikipedia.org/wiki/Glasgow_Subway

What’s your thinking there? The bigger tunnel boring machines needed and the extra dirt excavated are cheap in the bigger picture. The trend is clear within London alone and globally, towards larger tunnels.

How are they cheap in the big picture? Average cost is a billion per mile in the US and the latest numbers out of New York are more than 5 billion per mile.

Vegas hyperloop was <30 million per mile, so somewhere between 3% and 0.6% the cost per mile.

Reportedly, in this expansion, private companies will carry the entire cost.

Most of that high cost isn't due to the use of specific tunnel boring machines. In Europe they pay 1/6th of the US cost and still dig much wider tunnels with multiple tracks that support higher-capacity trains and safety features that the Vegas tunnels lack.

I agree that there are a lot of reasons for the cost differential between the US and Europe, and many reports and papers on the topic. From what I've read, tunnel costs are climbing quickly in Europe.

At the end of the day The boring company seems to be able to avoid these high costs using a combination of their technology, simplified goal, and business approach while nobody else has been able to bring costs down for a large tunnels in the US.

It would be great if someone could figure out how to bring large Tunnel cost down in the US, but until someone does, smaller tunnels costing a few percent of the cost have a competitive advantage.

My understanding is that the problems driving large tunnel construction costs in the US are driven by a shortage of labor expertise, Labor bloat compared to other countries, and institutional inability to manage large projects and avoid scope change.

>> My understanding is that the problems driving large tunnel construction costs in the US are driven by a shortage of labor expertise, Labor bloat compared to other countries, and institutional inability to manage large projects and avoid scope change.

This understanding is not correct. America's tunneling cost problem is driven by labor costs, unions, politics, and third party litigation. The institutional inability to manage large projects is a result of the latter three of these: the people in charge when the contract is signed are often not the people in charge when the job is half completed, and that change in leadership leads to scope changes that should (justifiably) be paid for.

I dont see a meaningful distinction between the causes you claim and what I said.

I agree third party litigation warrants highlighting, but I consider that as part of institutional inability to manage large projects.

Likelwise, I see high labor costs and unions as equavalent to lack of skilled labor and bloated labor costs.

At the end of the day, we are saying the same thing. Cost of workers is higher, the US uses far more workers, and the US legal and political systems are not condusive to efficient execution.

Your articulation is not unique to large diameter tunnels, the generic costs you point to are endemic to any diameter tunnel in America and also to any infrastructure project in America that takes place in an urban environment. To say that these things are a "large diameter tunnel" problem is to miss the unique distinctions between TUNNELING costs and INFRASTRUCTURE costs in America. At a high level we agree, but that's not useful for finding a solution to the specific problem of American tunneling costs.

For example, the labor shortage impacts all construction in America but tunnels require even more specialized laborers. Unions drive up the cost of all construction in America, but specific unions (like the Sandhogs) have specific provisions in their contracts that drive up the cost even more. Scope changes impact all construction, but scope changes underground cost orders of magnitude more than surface scope changes (add an offramp above ground versus add an offramp underground). Third party litigation impacts all construction, but the nature of tunneling increases the risk of third party litigation because settlement/sinkholes can occur vast distances away from the work.

Again, I think I agree with you on the nature of the problems.

Where I disagree is that the boring company is approach minimizes the magnitude and impact of these challenges.

For example, if the project scope isn't dictated by politicians, it is less susceptible to political whims and scope changes.

If the project scope is smaller, the we all challenges will be fewer. If the tunnels are smaller, service requirements are lower.

It would be interesting to know the boring company uses union labor or not.

This is 68 miles of tunnel. This is an enormous length of tunnel, its scope is anything but small.

I'm not sure what you mean by service requirements for a tunnel, but I assume you mean maintenance. Maintenance requirements for tunnels are only tangentially related to diameter once the final lining is in place, it is mostly driven by length. Realistically there is only so much maintenance that can happen for a tunnel or for any underground structure from the structural perspective since you can't look outside of the tunnel from the inside. So you overdesign. Maintenance requirements thus amount to visual inspections of the inside and maintenance of appurtenant mechanical/electrical systems. These mechanical/electrical systems become more reliable and more efficient with relation to tunnel diameter, and so your maintenance requirements actually decrease on an aggregate basis (i.e. - a tunnel that can support 2 lanes of traffic is less costly to maintain than 2 tunnels that support 1 lane of traffic).

And the idea that anything ran by Elon Musk is not susceptible to whims and scope changes is laughable.

Digging a wider tunnel requires a bigger tbm, which adds little to the whole cost. Hs2 put out a paper, though I don’t think it compares sizes. https://assets.publishing.service.gov.uk/government/uploads/... The NYC example is terrible, because the new part of the tunnel is only short and the whole project costs a lot does not mean each mile of the tunnel costs that much.

I was first asking why they thought small tunnels are better, is that purely on financial grounds? Low cost, low benefit.

Why is it not appropriate to compare a whole project cost the whole project cost per mile? The New York extension was 7.7 billion for 1.5 MI and the Vegas Loop was 47 million for 1.7 MI. Now you have a project cost per mile and you can evaluate benefit from per mile. Biggest Loop was unquestionably much lower total benefit. Then you can compare benefit per dollar. At 1 200th the price does it provide more or less than 0.5% the value.

I guess my thinking was that it works pretty well in London to use smaller tunnels, and I don't immediately understand why larger tunnels are needed. If you're familiar with why tunnels have been trending larger, I'd love to know.

In animations of the intended final design, there will also be larger vehicles which can carry more people at once and will of course be automated. I don't think it will be materially different from a subway (train) in that regard where volumes are high. It will just be more flexible.

The concept animations of the convention center strip were similarly far from what was delivered. Real "skyscraper covered in parks" energy.

For the love of god, why does there need to be yet another lane? It will emphatically not solve the problem. Driving teslas through a tunnel is not flexible, it's a recipe for induced demand and gridlock.

Build a subway! High volumes of people get moved very efficiently. Moving small numbers of people in personal vehicles is a terrible idea. Flexibility is a bad feature here.

Subways are not particularly efficient from a cost perspective. One of the main reasons for the Boring company’s existence is the extremely high costs associated with subway construction. They are trying to bring the costs down by one or two orders of magnitude. This is material.

Yeah, I think the original vision for the Loop tunnels is a lot more practical and compelling than what currently exists in Vegas.

The primary advantages over a subway are:

1. They're point-to-point, requiring no stops between origin and destination

2. They integrate with the existing road system, allowing point-to-point transportation even between destinations where the tunnels don't run

3. They're compatible with existing electric cars, allowing owners to take their own private vehicles directly to their destination without needing to transfer between cars or share a space with strangers

4. Small tunnel diameter reduces the cost vs larger subway tunnels

Of those points, only #4 got implemented in the Vegas loop and even that is debatable. Plus a lot of other important features like automated driving and dedicated mass transit vehicles got left out, which greatly reduces the ridership capacity of the tunnels compared to what they theoretically could support.

How is Vegas in desperate need of transportation between the airport and the strip? Of all the airports I’ve flown to Vegas was the closest to the downtown core and had the least traffic.

You still have to use a taxi/rideshare service, or rent a car. Compared to other cities, those methods are pretty cheap and accessible in Vegas... but it's still expensive and inconvenient compared to a real railway transit system.

Pretty much. Although I'm not sure how useful it would be to have a public transit system from the airport even in the best case. Given how huge the casino complexes are, you'd pretty much need a stop at every casino (and you still might have a long walk to the front desk as is the case with the little loop train that already exists).

It's not really more expensive. In Denver it's $10.50 per person to take the train from the airport to downtown. At the time of this post an Uber from LAS to the Bellagio is $11. If you have two people then the Uber is about half the cost per person of the train in Denver.

Taxi/rideshare also takes you right to your hotel instead of a station.

I would imagine that you could cut out 80% of the lanes of traffic on the strip if there was a dedicated train from the airport to the strip continuing to other points of interest for tourists. Maybe you could even make the strip pedestrian-only.

The strip is basically a walking area like a theme park, except that it’s cut down the middle by an eight lane road and supplemented with overpasses. Cars get priority over people, who ironically mostly don’t have a car while they’re on vacation in Vegas.

This is an avenue that’s already served by frequent double decker buses. The ridership volume is there at least for a light rail tram system.

You might imagine that but I'm very skeptical that 80% or anything like that of the traffic on the strip is people going back and forth to the airport.

1/3 of urban core traffic is just people cruising around looking for free parking. Then consider how much is unlicensed taxi services like Uber. A lot of urban core traffic is self-inflicted through lack of public transit.

Vegas doesn't really have an urban core as such and AFAIK, there really isn't free street parking near the Strip (and also AFAIK all the casinos now charge).

Transit from the airport would stop at each casino. That would be the bulk of traffic (inter -casino traffic)

There are a lot of casinos. You could have a shuttle that serves some subset of hotels. Maybe that even exists today. (There are actually a couple of limited trains today along the Strip.) But people by and large don't love that sort of thing because shuttle buses with a dozen or so stops can take a long time. Rail wouldn't be a lot faster.

Sure, but it’s rather convenient that the route from the airport also takes you up and down the main drag.

Let’s say I’m staying at circus circus and I’m seeing a show at the MGM grand.

It’s an hour walk, 24 minute bus ride plus the wait for the bus (15 minutes divided by 2 on average), or a 12 minute drive.

The Vegas monorail exists but it isn’t anywhere near the actual street where the pedestrians are, and the station locations are so far from where you want to be that you might as well walk.

Therefore, if I’m a typical tourist I’m taking an Uber or a taxi for this trip.

That’s why the strip has so many lanes of traffic. I’m incentivized to get in a car and I haven’t even left the “theme park” boundaries.

Agreed, add a monorail/light rail along side the pedestrian only lanes running down the strip and you have a winner. Always suspected casinos don't want you to roam. Also with F1 now it will never happen.

Your assumption is that only small EVs will go through these tunnels in the future. I think that assumption is flawed.

Tesla announced it wants to make high-passenger-count commercial EVs (akin to vans and buses):

https://www.tesla.com/ns_videos/Tesla-Master-Plan-Part-3.pdf

EVs that carry 8, 12, or more passengers (perhaps by chaining the EVs) would significantly increase the efficiency and throughput of these tunnels.

And then there's a big question mark on self-driving, which, if achieved, would significantly improve the economics.

12 people would be about 5% of the capacity of a single NYC subway car: https://en.wikipedia.org/wiki/R160_(New_York_City_Subway_car...

Are you suggesting that Tesla is planning to develop underground trains?

That would make the whole thing even dumber than I firstly thought.

Or Elmo could make an electric tram built just for these tunnels. Could use Tesla drivetrain to reduce design costs.

Let's hope there's never a fire.

Reply on news.ycombinator.com