Hacker News Reader

Comment by madcaptenor | Hacker News Reader

Right. You don't want a line down the entire West Coast, even though it's tempting to draw. But a line from Vancouver to Portland makes sense, as does a "greater California" system - roughly lines from Los Angeles to SF, Sacramento, Vegas, San Diego. The latter is basically the California HSR system that's under construction, plus the proposed privately built line from LA to Vegas. (Levy also proposes LA to Phoenix; Phoenix is further than Vegas but also bigger, so maybe it makes sense.)

Even in the east there are some gaps. It's obvious that a midwestern network centered on Chicago and a southeastern network centered on Atlanta make sense, but it's a bit more of a stretch to connect those to the northeast.

View on HN

Why not? The great advantage of high speed rail is that with intermediate stops along the route you can service smaller cities which previously, or as you just did, would be considered flyover country. Thus making the value of the system greater than just the end terminuses.

Just looking at the towns between Portland and Sacramento Salem, Eugene and Medford exist. Neither would themselves ever be valuable enough for HSR, but as part of a larger system they definitely would bring value.

Especially since you would get 3.5 hour trains Bay Area <-> Seattle and 4.5 hour trains to Vancouver.

That is right at the limit of when flying starts to make more sense from a time perspective.

Edit: Here's a good video on the concept. From a comment (by the author) he says that France and Spain has many lines in the range of 4 given his scoring. It just seems miniscule compared to the enormous potential of DC <-> Boston corridor.

"U.S. High Speed Rail: What's Next? Analyzing Extensions and Expansions, and What Makes Sense"

https://youtu.be/zxiGY8p2rCo

I'm sure they bring some value but my understanding is they don't bring enough value. But I could be wrong!

Here's a good video on one method of calculating what the sum of the smaller individual parts would be. This is for the north east corridor but the same thinking applies to any rail project.

It seems like about a million people live there, not nearly enough individually, and trains would likely alternate at which locations they stop to bring the total travel time between the larger areas down.

Might be hindered by the mountainous terrain though making the cost prohibitive.

"U.S. High Speed Rail: What's Next? Analyzing Extensions and Expansions, and What Makes Sense"

https://youtu.be/zxiGY8p2rCo

Thanks for the link! Will watch later.

Because not enough people live along the way. Sure you an build track and run trains, but 5 hours on a train is about the time where flying is enough faster that people will fly instead of taking the train. Less than 5 hours train competes well (stations are closer to you, and no long security lines), but after that airplanes are enough faster that few people would use a train. That means only a small number of people will ride the train for those middle stations.

Sure if you are building a track you can put in stations in towns that don't generation much traffic, but you still need traffic from somewhere and it won't come.

Well with Maglev, Portland to San Francisco could make sense via Sacramento. You wouldn’t want to do it on the coast because of the mountains, and there’s at least Redding and Ashland and a couple of other places in-between.

At 500 km/h (310 mph) you could feasibly do Sacramento to Portland in under two hours. The less straightforward question is how much of the metro area do you serve around those two cities, and do you connect that line directly to San Francisco or do you run a separate line to Sacramento via the Delta? Do you go a sort if L-shape around Stockton first? The politics of this could push travel time up, but at 500 km/h you can cover a lot of ground, much of it fairly empty.

So a hypothetical Best Coast system would connect Vancouver, BC to Portland, Portland to Sacramento, Sacramento to Reno, Reno to Las Vegas, Sacramento to LA, San Francisco to Sacramento via Stockton, San Francisco to LA, LA to Tijuana via San Diego, LA to Las Vegas, LA to El Paso via Phoenix, Phoenix to St. George, Las Vegas to Salt Lake City via St. George, San Diego to El Paso via Tucson & Mexicali and now you’re in Texas where options include El Paso to New Orleans via either Austin & Houston or San Antonio & Houston, Brownsville to McAllen, Brownsville to Houston via Corpus Christi, Corpus Christi to San Antonio, Houston to Dallas, Dallas to Oklahoma City and I’m probably missing some, but you have the workings of a Gulf Coast constellation anchored by Texas on one end and Florida on the other.

Thing is, I’ve worked this all out on paper too, including a Northeast, Southeast and Midwest map that looked much like one that someone linked to up the thread. Problem is, our choice of infrastructure is downstream from our cultural preferences which in turn are shaped by the infrastructure our ancestors built in the decades prior.

El Paso is about 725 miles away from San Diego according to Siri, so with the best trains in the world you could be inside the Texas rail constellation I briefly outlined above in about 2-3 hours which in turn could serve as the basis for a Gulf Coast constellation connecting Florida and the Southeast connecting to the Midwest and the Northeast and then to Canada.

It’s not built though because people just fly instead. We worked out how to get cheap air travel long before we figured out super-fast transcontinental rail travel that probably doesn’t make sense coast to coast but if it already existed, probably would make sense going coast to middle and middle to middle and would just happen to connect the coasts. Or you could just fly, which is what we do, and since that already exists and is even faster than rail travel, metro areas can figure out how their own inter-urban rail systems that are slow and local where it makes sense to build them and just try to make sure the airport is connected too. I like trains, but not so much that I’m willing to toss hundreds of billions of public money into some kind of “national rail system” whatever form it took for whatever prestige it might bring. Jumbo jets are cool too.

Wind resistance at ground level means that an airplane is much more energy efficient at those distances and speed. As such those distances are unlike to work just because an airplane is so much cheaper even if because of speed it is time competitive.

And for short distances we are getting very close to where electric planes can show up.

There really isn't much between Portland area and San Francisco area, and there isn't enough potential economic activity that an HSR would induce (it is too mountainous, which also means building HSR would be more expensive).

The problem is that TSA isn't as bad as it was right after 9/11, and baggage tracking is much better on all airlines. You no longer need to arrive 2 hours before the scheduled departure and spend an hour collecting your bags.

With TSA PreCheck, I can reliably go from curb to gate in less than 15 minutes. If you're not someone who feels the need to be the first one on the plane, that means you can arrive at the airport 30 minutes before the scheduled departure time.

So, as a practical matter, that means a SJC->SEA flight is at least an hour shorter than the hypothetical train.

I'm also very skeptical that a train could reach Seattle in 3.5 hours from the Bay Area. That would require an average speed of over 200 mph on the great-circle path.

California High-Speed Rail only promises an average speed of 150 mph between LA and San Francisco (w/ a world-class top speed of 220 mph). Additionally, geography dictates a more circuitous route. CAHSR route-miles between LA and the SFBA (similar terrain) are 25% greater than the straight-line distance.

Realistically, the train would take almost 6 hours, and a plane would be less than half.

Yeah, I haven't flown for a while and I gather there's still a certain level of travel chaos. But pre-pandemic, I'd get to the airport early because it's more relaxing for me and my limo company doesn't like to cut things close. But with TSA Pre, I was rarely more than 15 minutes through security and often much faster. Backups happen and I'd rather build in slack for them. But in my experience, at the US airports I fly through, the "security theater" is rarely onerous.

South of Eugene and north of Redding, the land is mountainous and would be extremely difficult to build a straight enough line to serve as HSR. The existing Amtrak Cascades service goes between Eugene and Vancouver BC.

I agree, a high speed line between Eugene and Redding seems like an overkill. However a traditional electrified railway with a stop in Medford would be pretty sweat.

With the planned California High Speed rail going between Sacramento and Los Angeles, and the proposed Cascadia high speed rail going all the way to Eugene, this traditional link would enable a sleeper train between Seattle and Los Angeles in something like 10-13 hours. That is way better then today’s Coastal Starlight which makes the trip in 36 hours.

> Especially since you would get 3.5 hour trains Bay Area <-> Seattle and 4.5 hour trains to Vancouver.

Hiroshima to Tokyo is just over 3.5 hours on the Shinkansen, and it’s a 500 mile trip. SF TO Seattle is roughly 800 miles… You’re making a very optimistic projection.

> Just looking at the towns between Portland and Sacramento Salem, Eugene and Medford exist.

You’re barely cracking 500k people and covering the most difficult terrain on the entire corridor.

Reply on news.ycombinator.com