4 ms·

Osaka to Fukuoka in Japan is about 2.5 hours by Shinkansen high speed train with 6 stops. Chicago to NYC is about twice that distance, so 5 hours should be obt

Osaka to Fukuoka in Japan is about 2.5 hours by Shinkansen high speed train with 6 stops. Chicago to NYC is about twice that distance, so 5 hours should be obtainable

jcranmer 9y ago
The question is what the situation on the ground looks like. The short distance for NY/Chicago is around 800mi following more or less I-80. The problem with that route is you're missing any viable station between Cleveland and NY (maybe Scranton, but that's sending you in the wrong direction), and there's no real feasible reuse of that route for any other pairings other than NY/Chicago. The "easy" route via the Erie canal (so Chicago/Cleveland/Buffalo/Albany/NY) is 950mi or so, which is simply getting too long to capture traffic. Going via Detroit and Niagara doesn't really add any time, and cutting the trip in half via Toronto would add some time, but you'd get two viable trip corridors out of it, if you could solve border control issues. The most logical route is the Pittsburgh/Philly route, because you get the topography in your favor east of Harrisburg, reuse of corridor (Pittsburgh, Philly, and Harrisburg are all en route, not to mention the wye at Pittsburgh gives you basically Chicago/NYC and Chicago/DC on the same corridor). But this route is still 860mi, which is hard to keep to 5 hours.
- niftich 9y ago
  To expand this with more detail, NYC-Chicago is between 900 and 1000 miles on existing railroad tracks, depending on which way you go: ~910 miles through Philadelphia -> Harrisburg -> Pittsburgh -> Cleveland; ~960 miles through Albany -> Buffalo -> Cleveland; ~1000 miles through Philadelphia -> Baltimore -> DC -> Pittsburgh -> Cleveland. The route that touches DC is the one that passes through the most valuable intermediate stops, but most of that demand is intra-corridor and doesn't concern Chicago. Furthermore, DC is a logical place to serve as a hub to connect the two corridors: one a high-demand DC-NYC, and one a long-distance Chicago-DC. Since to go into Union Station, you have to double-back, this could be a two-segment train where you'd have one-seat travel between Chicago and NYC, but there would be a mandatory stop in DC. This all sounds good, but the segment between DC and Pittsburgh is awful: it's crowded with CSX freight trains, it's either hemmed in by populated areas or follows river valleys or narrow ravines, and there's much work to be done to bring it up to a state that can host High Speed Rail. Between DC and Pittsburgh, one would basically have to build a dedicated line for High Speed passenger rail. The route through Albany and Buffalo avoids the Appalachians but is forced into the busy Hudson Valley. Route improvements here are expensive. East of Cleveland, the Upstate New York cities along the route simply don't generate enough Chicago-bound demand, nor enough connecting opportunities to make routing this way worthwhile. The route of the old Pennsylvanian Railway through Philly, Harrisburg, Altoona, Pittsburgh is indeed the most logical 'direct' route, both in terms of macro and micro; most of the right-of-way is multi-tracked so even with the heavy Norfolk Southern freight traffic, it'd be competitive; and the route touches Cleveland, Pittsburgh, and Philadelphia, all of which can generate local and connecting passengers. But even this route is substandard for High Speed Rail: most of it is in river valleys, where there isn't space for the overly broad, sweeping curves with large radii that are needed for high speed travel. The gist is, between Pittsburgh and the Appalachian Great Valley -- home to Harrisburg and Hagerstown, on the routes going towards Philadelphia and DC respectively -- significant engineering effort (that I wrote about previously [1]) would be needed to cut down on travel time on what would become one of the most profitable segments of the route: between Western PA and the rest of the Northeast Corridor. As for the Midwest, there are no convenient cities west of Cleveland to chain onto the route. Toledo or Fort Wayne are the most logical choices, but Toledo lacks north-south connections currently, while Fort Wayne is out of the way of the direct route and also lacks connections. Chaining them both along with South Bend would be best, but that's a complicated route that adds travel time. [1] https://news.ycombinator.com/item?id=13414587 https://news.ycombinator.com/item?id=13414587
  - jcranmer 9y ago
    On point, but some addendums: Chicago/NEC isn't viable for HSR until 200mph at least, which means dedicated passenger rails, which means a new corridor anyways. The CSX route from DC to Pittsburgh isn't particularly great from this purpose. Of course, one of the issues with the Keystone to Harrisburg is that the wye is just north of Philly's station, so either you have to run the locomotive around in Philly or you bypass the stop entirely on the way to NYC. In terms of engineering, the worst part is going to be the Ridge and Valley section of the Appalachians, roughly Blue Mountain to Bedford if you're following the PA Turnpike. The problem is that you have steep ridges very perpendicular to your direction of travel, which requires either detouring to find a water gap or tunneling through the base. Once you're on the plateau proper, passenger rail can easily handle the grades. As for your final point: that's the big problem with the Midwest: pretty much nothing's colinear. In contrast, a straight line from from DC to Boston literally passes right through NYC [1] and very near Baltimore, Wilmington, Philly, Trenton, and New Haven, with Hartford, Springfield, and Providence a bit further off. [1] When I drew it from the Capitol to Boston City Hall, the path over Manhattan was from roughly the northwestern corner of Lower Manhattan to the southeastern corner of Midtown Manhattan.
- briandear 9y ago
  Rather that go through all that nonsense, why not focus on connecting the airports to downtown more efficiently — such as how Heathrow has done it. 15 min from heathrow to paddington. Flying is the best mode of long distance travel in the United States — you don’t have to pay for thousands of miles of track, instead you just need to focus on connecting to airports with the cities more efficiently. With airplanes, you can route however you want, scaling with demand. But with a train: what happens if Chicago-NY becomes unprofitable? You’ve spent billions connecting cities rather than simply improving airports. Airports can serve all city pairs, a train is a permanent connection of city pairs. It’s interesting how people here are ok with spending billions to run track, build stations, but nobody has brought up the simplest solution: improve airports and the airport UX. That would be cheaper and benefit more people — not just those living in a major city.
  - Johnny555 9y ago
    That's only half the battle since you still need to get past the TSA checkpoint so if the flight is important to you, you'll get to the airport an hour early just in case the line is significantly longer than usual. Over the past 6 months, I've spent nearly 60 minutes in security more than once due to equipment failure and/or understaffing. Trains are a good backup to airports since airplanes can be grounded (or highly delayed) due to weather, security concerns, equipment problems at the airport, etc. Trains are not (generally) subject to the same concerns so make a good backup. I was in Japan recently and my intercity flight to a city 300 miles away was canceled due to an incoming Typhoon. I was able to book a high speed train that got me in there to catch my flight (and depart before the typhoon hit that airport too). When a blizzard hits NYC, all you can do is sit it out and wait for flights to resume and it can sometimes take days to find an available seat.

Osaka to Fukuoka in Japan is about 2.5 hours by Shinkansen high speed train with 6 stops. Chicago to NYC is about twice that distance, so 5 hours should be obtainable

jcranmer 9y ago

The question is what the situation on the ground looks like. The short distance for NY/Chicago is around 800mi following more or less I-80. The problem with that route is you're missing any viable station between Cleveland and NY (maybe Scranton, but that's sending you in the wrong direction), and there's no real feasible reuse of that route for any other pairings other than NY/Chicago. The "easy" route via the Erie canal (so Chicago/Cleveland/Buffalo/Albany/NY) is 950mi or so, which is simply getting too long to capture traffic. Going via Detroit and Niagara doesn't really add any time, and cutting the trip in half via Toronto would add some time, but you'd get two viable trip corridors out of it, if you could solve border control issues. The most logical route is the Pittsburgh/Philly route, because you get the topography in your favor east of Harrisburg, reuse of corridor (Pittsburgh, Philly, and Harrisburg are all en route, not to mention the wye at Pittsburgh gives you basically Chicago/NYC and Chicago/DC on the same corridor). But this route is still 860mi, which is hard to keep to 5 hours.

niftich 9y ago
To expand this with more detail, NYC-Chicago is between 900 and 1000 miles on existing railroad tracks, depending on which way you go: ~910 miles through Philadelphia -> Harrisburg -> Pittsburgh -> Cleveland; ~960 miles through Albany -> Buffalo -> Cleveland; ~1000 miles through Philadelphia -> Baltimore -> DC -> Pittsburgh -> Cleveland. The route that touches DC is the one that passes through the most valuable intermediate stops, but most of that demand is intra-corridor and doesn't concern Chicago. Furthermore, DC is a logical place to serve as a hub to connect the two corridors: one a high-demand DC-NYC, and one a long-distance Chicago-DC. Since to go into Union Station, you have to double-back, this could be a two-segment train where you'd have one-seat travel between Chicago and NYC, but there would be a mandatory stop in DC. This all sounds good, but the segment between DC and Pittsburgh is awful: it's crowded with CSX freight trains, it's either hemmed in by populated areas or follows river valleys or narrow ravines, and there's much work to be done to bring it up to a state that can host High Speed Rail. Between DC and Pittsburgh, one would basically have to build a dedicated line for High Speed passenger rail. The route through Albany and Buffalo avoids the Appalachians but is forced into the busy Hudson Valley. Route improvements here are expensive. East of Cleveland, the Upstate New York cities along the route simply don't generate enough Chicago-bound demand, nor enough connecting opportunities to make routing this way worthwhile. The route of the old Pennsylvanian Railway through Philly, Harrisburg, Altoona, Pittsburgh is indeed the most logical 'direct' route, both in terms of macro and micro; most of the right-of-way is multi-tracked so even with the heavy Norfolk Southern freight traffic, it'd be competitive; and the route touches Cleveland, Pittsburgh, and Philadelphia, all of which can generate local and connecting passengers. But even this route is substandard for High Speed Rail: most of it is in river valleys, where there isn't space for the overly broad, sweeping curves with large radii that are needed for high speed travel. The gist is, between Pittsburgh and the Appalachian Great Valley -- home to Harrisburg and Hagerstown, on the routes going towards Philadelphia and DC respectively -- significant engineering effort (that I wrote about previously [1]) would be needed to cut down on travel time on what would become one of the most profitable segments of the route: between Western PA and the rest of the Northeast Corridor. As for the Midwest, there are no convenient cities west of Cleveland to chain onto the route. Toledo or Fort Wayne are the most logical choices, but Toledo lacks north-south connections currently, while Fort Wayne is out of the way of the direct route and also lacks connections. Chaining them both along with South Bend would be best, but that's a complicated route that adds travel time. [1] https://news.ycombinator.com/item?id=13414587 https://news.ycombinator.com/item?id=13414587
- jcranmer 9y ago
  On point, but some addendums: Chicago/NEC isn't viable for HSR until 200mph at least, which means dedicated passenger rails, which means a new corridor anyways. The CSX route from DC to Pittsburgh isn't particularly great from this purpose. Of course, one of the issues with the Keystone to Harrisburg is that the wye is just north of Philly's station, so either you have to run the locomotive around in Philly or you bypass the stop entirely on the way to NYC. In terms of engineering, the worst part is going to be the Ridge and Valley section of the Appalachians, roughly Blue Mountain to Bedford if you're following the PA Turnpike. The problem is that you have steep ridges very perpendicular to your direction of travel, which requires either detouring to find a water gap or tunneling through the base. Once you're on the plateau proper, passenger rail can easily handle the grades. As for your final point: that's the big problem with the Midwest: pretty much nothing's colinear. In contrast, a straight line from from DC to Boston literally passes right through NYC [1] and very near Baltimore, Wilmington, Philly, Trenton, and New Haven, with Hartford, Springfield, and Providence a bit further off. [1] When I drew it from the Capitol to Boston City Hall, the path over Manhattan was from roughly the northwestern corner of Lower Manhattan to the southeastern corner of Midtown Manhattan.
briandear 9y ago
Rather that go through all that nonsense, why not focus on connecting the airports to downtown more efficiently — such as how Heathrow has done it. 15 min from heathrow to paddington. Flying is the best mode of long distance travel in the United States — you don’t have to pay for thousands of miles of track, instead you just need to focus on connecting to airports with the cities more efficiently. With airplanes, you can route however you want, scaling with demand. But with a train: what happens if Chicago-NY becomes unprofitable? You’ve spent billions connecting cities rather than simply improving airports. Airports can serve all city pairs, a train is a permanent connection of city pairs. It’s interesting how people here are ok with spending billions to run track, build stations, but nobody has brought up the simplest solution: improve airports and the airport UX. That would be cheaper and benefit more people — not just those living in a major city.
- Johnny555 9y ago
  That's only half the battle since you still need to get past the TSA checkpoint so if the flight is important to you, you'll get to the airport an hour early just in case the line is significantly longer than usual. Over the past 6 months, I've spent nearly 60 minutes in security more than once due to equipment failure and/or understaffing. Trains are a good backup to airports since airplanes can be grounded (or highly delayed) due to weather, security concerns, equipment problems at the airport, etc. Trains are not (generally) subject to the same concerns so make a good backup. I was in Japan recently and my intercity flight to a city 300 miles away was canceled due to an incoming Typhoon. I was able to book a high speed train that got me in there to catch my flight (and depart before the typhoon hit that airport too). When a blizzard hits NYC, all you can do is sit it out and wait for flights to resume and it can sometimes take days to find an available seat.