3 ms·

As long as we're putting trains in those tunnels! Cars on freeways or sleds don't carry nearly as many people/hour.

njarboe 9y ago
By trains do you mean buses? or lots of buses. Or how about many buses mixed with cars. You should do a quick Fermi calculation and you will find that with a close packed stream of single passenger vehicles at 200kph, the people/hour is higher that BART at rush hour. Use vehicles where people stand and are packed like sardines, as it is on BART at rush hour, and then you are getting a crazy high people/hour. If you really need to have all new transportation systems be public trains to feel good about building it, then, yes, there will be public trains in these tunnels.
- jakecopp 9y ago
  Buses are good too, though bus rapid transit doesn't have the appeal of trains. I'm not sure if this is a problem inherent to buses or BRT systems having worse service on average giving them a bad image. 200kph close packed cars would be nice, though you have to deal with: - Close packing which definitely needs autonomous driving. I want something we can build in 2018. - Merging at 200kph in tunnels. Those on and off ramps are expensive to dig. - Load balancing between on/off ramps. If everyone wants to get off the same exit at 200kph I don't know where the load goes when you have to transition to 50kph traffic. Are you going to disallow exiting from an off ramp if it's at capacity and would slow down the tunnel? Where do these people get off? - Where do you park all of these cars? That's a hell of a lot of downtown parking! How much does downtown land cost per square metre, multiplied by the area of the sum of cars in the city? Trains solve all these problems, and more importantly are here right now! It's not a technical problem, it's a political one. I don't understand the obsession with all these far off possibilities that might be better than rail. I'm just as big a fan of Elon Musk as the next person but Hyperloop has given a lot of ammunition to opponents of rail. "Hyperloop/The Boring Company is coming so we shouldn't build efficient metros that are already implemented in Japan/China/Europe and reduce reliance on cars and foreign oil."
  - njarboe 9y ago
    "lose packing which definitely needs autonomous driving. I want something we can build in 2018" Yes, automatic driving only in these tunnels. Current tech in a Tesla Car (and some others) would already be able to automatically drive in a tunnel today with very short spacing, no problem. Here in 2018. "Merging at 200kph in tunnels. Those on and off ramps are expensive to dig." Yes. Tunnels are expensive. They are getting more expensive to dig. Thus the Boring company. This is the biggest problem with a tunnel system, but tunneling for trains is not cheaper. "Load balancing between on/off ramps." An automatic car can know if an off ramp is going to be full. Pay extra(congestion pricing, progressives love that) to exit or take one before or after. "Where to park" Automatic electric car goes back home until Miller Time or is rented to other people for the day. Rail is not a point to point system. This is the problem. How are you getting trains to within a 5 minute walking distance of everyone's home in LA and can go direct (no train transfers) to a 5 minute walk to anywhere else they want to go. Not likely. With this system, being a 5 minute drive to the nearest tunnel entrance and then the connected tunnel system puts you out with a 5 minute drive to your destination. This would be possible. The great thing about cities are the people. Being able to have a huge set of people that people can interact with and have long term relationships with. Groups formed by these people. I think it would be cool to have a city where most people can have a piece of land to grow things on and also have the possible interaction and long term relationships with >10 million other humans.
    - jakecopp 9y ago
      "Pay extra(congestion pricing, progressives love that) to exit or take one before or after." It's a physical impossibility, no congestion pricing will make more supply. If it's 8:30am on a weekday, and you have a large suburban area serviced by a tunnel exiting in a much smaller downtown area, how will there ever be enough exits? Have you seen how crowded a European train station is at peak hour? All these cars would have to go somewhere. Some back of the napkin maths in an ideal scenario: 200kph = 55.5556ms tunnel, 1 lane, 3 metre windscreen to next windscreen distance (car length + distance in between). 55.556 / 3 = 18.51 ish cars/second = 66,636 cars per hour. This site [1] says LA's average off peak average speed is 26.8mph. Super rough estimate. This graph [2] says 26.8mph gives you ~1800 cars/hour/lane. So that's 37 downtown lanes per one lane of 200kph Boring Tunnel traffic. 37 lanes are 3.7m wide, so that's 137 metres wide! Sydney's current trains take 28,000 people per hour [3] in the real world, in 2018, driven by humans, with humans exiting stations. Put in 2 - 3 train lines underground, which is say 6 metres wide, chuck a station on top with escalators pumping people in and out, and people can walk through the city. Put a park on top. Sydney Central Station has 27 platforms, 10 underground. That's like 280,000 people per hour on the underground lines by those dodgy estimates, or 5 x 200kph 3 back to back traffic lanes (not tested yet), or 185 lanes of 26mph traffic which is ¾ of a kilometre wide! And it was built in 1906 out of sandstone! There are ~ 5 CBD stations in Sydney. A 925 lane wide road sounds like fun. And its here in 2018 and it costs me AU$1.50 to use. Cars don't belong in cities. Cars belong in the suburbs, which is where they can drive you to your train station. Tokyo's Shinkansen was opened in 1964 and went 220kph. If the train is more than 1 minute off the timetable you get an apology letter for your employer. They took the original trains out of service in 2008. The USA is so behind. As I said, transport isn't a technical problem, it's a political one. The future is here (high speed rail), it just isn't evenly distributed yet. [1]: http://infinitemonkeycorps.net/projects/cityspeed/ http://infinitemonkeycorps.net/projects/cityspeed/ [2]: http://moderntransit.org/solution.html http://moderntransit.org/solution.html [3]: http://www.abc.net.au/news/2014-04-11/barry-ofarrell-sydney-trains-claim-doubtful/5371446 http://www.abc.net.au/news/2014-04-11/barry-ofarrell-sydney-...
dmix 9y ago
> Cars on freeways or sleds don't carry nearly as many people/hour. Almost entirely because of how car ownership is structured now. People most often buy 4dr cars and only use 1-2 seats 90% of the time. A subscription-based self-driving fleet car system would change this dynamic where you only order a 4dr+ car/SUV/van for a ride when needed. Otherwise a small 2dr electric vehicle without a large front motor will likely be the default in the future. On-demand cars will be great for the environment and efficiency of highways... Not to mention there has been some talk about self-driving cars linking up like trains on freeways to conserve energy and better utilize space at very high speeds.
- jakecopp 9y ago
  I am optimistic about self driving cars, but we also lower to floor of possible vehicle occupancy to 0. I'm curious to see if average occupancy goes up or down. Linking up cars sounds great! Maybe we could make the cars electric. And we could create escape hatches to walk between them. We could make them shorter so the car is essentially a seat. And then we could make them 100% subscription based and shared so you never take them away. I think I might just have reinvented trains! Our obsession with automobiles borders on insanity. Yes, the last mile is hard to do with public transit but for high capacity, high cost thoroughfares trains win every time.
  - dmix 9y ago
    > but we also lower to floor of possible vehicle occupancy to 0 That's a great point that I hadn't yet considered. In terms of finite energy resource demand it's not as big of a deal when the average car both smaller and electric. But in terms of traffic and related resource utilization this will ultimately become a far more common phenomenon, even possibly in the top 1-2 most frequent scenarios as driverless cars attempt to reach a new client or drive to the closest automated charging port. I'd be curious to see the traffic/resource impact of such a hypothetical scenario. Although I anticipate some smarter people than me are already calculating these costs. At least I hope so!
    - jakecopp 9y ago

By trains do you mean buses? or lots of buses. Or how about many buses mixed with cars. You should do a quick Fermi calculation and you will find that with a close packed stream of single passenger vehicles at 200kph, the people/hour is higher that BART at rush hour. Use vehicles where people stand and are packed like sardines, as it is on BART at rush hour, and then you are getting a crazy high people/hour. If you really need to have all new transportation systems be public trains to feel good about building it, then, yes, there will be public trains in these tunnels.

jakecopp 9y ago
Buses are good too, though bus rapid transit doesn't have the appeal of trains. I'm not sure if this is a problem inherent to buses or BRT systems having worse service on average giving them a bad image. 200kph close packed cars would be nice, though you have to deal with: - Close packing which definitely needs autonomous driving. I want something we can build in 2018. - Merging at 200kph in tunnels. Those on and off ramps are expensive to dig. - Load balancing between on/off ramps. If everyone wants to get off the same exit at 200kph I don't know where the load goes when you have to transition to 50kph traffic. Are you going to disallow exiting from an off ramp if it's at capacity and would slow down the tunnel? Where do these people get off? - Where do you park all of these cars? That's a hell of a lot of downtown parking! How much does downtown land cost per square metre, multiplied by the area of the sum of cars in the city? Trains solve all these problems, and more importantly are here right now! It's not a technical problem, it's a political one. I don't understand the obsession with all these far off possibilities that might be better than rail. I'm just as big a fan of Elon Musk as the next person but Hyperloop has given a lot of ammunition to opponents of rail. "Hyperloop/The Boring Company is coming so we shouldn't build efficient metros that are already implemented in Japan/China/Europe and reduce reliance on cars and foreign oil."
- njarboe 9y ago
  "lose packing which definitely needs autonomous driving. I want something we can build in 2018" Yes, automatic driving only in these tunnels. Current tech in a Tesla Car (and some others) would already be able to automatically drive in a tunnel today with very short spacing, no problem. Here in 2018. "Merging at 200kph in tunnels. Those on and off ramps are expensive to dig." Yes. Tunnels are expensive. They are getting more expensive to dig. Thus the Boring company. This is the biggest problem with a tunnel system, but tunneling for trains is not cheaper. "Load balancing between on/off ramps." An automatic car can know if an off ramp is going to be full. Pay extra(congestion pricing, progressives love that) to exit or take one before or after. "Where to park" Automatic electric car goes back home until Miller Time or is rented to other people for the day. Rail is not a point to point system. This is the problem. How are you getting trains to within a 5 minute walking distance of everyone's home in LA and can go direct (no train transfers) to a 5 minute walk to anywhere else they want to go. Not likely. With this system, being a 5 minute drive to the nearest tunnel entrance and then the connected tunnel system puts you out with a 5 minute drive to your destination. This would be possible. The great thing about cities are the people. Being able to have a huge set of people that people can interact with and have long term relationships with. Groups formed by these people. I think it would be cool to have a city where most people can have a piece of land to grow things on and also have the possible interaction and long term relationships with >10 million other humans.
  - jakecopp 9y ago
    "Pay extra(congestion pricing, progressives love that) to exit or take one before or after." It's a physical impossibility, no congestion pricing will make more supply. If it's 8:30am on a weekday, and you have a large suburban area serviced by a tunnel exiting in a much smaller downtown area, how will there ever be enough exits? Have you seen how crowded a European train station is at peak hour? All these cars would have to go somewhere. Some back of the napkin maths in an ideal scenario: 200kph = 55.5556ms tunnel, 1 lane, 3 metre windscreen to next windscreen distance (car length + distance in between). 55.556 / 3 = 18.51 ish cars/second = 66,636 cars per hour. This site [1] says LA's average off peak average speed is 26.8mph. Super rough estimate. This graph [2] says 26.8mph gives you ~1800 cars/hour/lane. So that's 37 downtown lanes per one lane of 200kph Boring Tunnel traffic. 37 lanes are 3.7m wide, so that's 137 metres wide! Sydney's current trains take 28,000 people per hour [3] in the real world, in 2018, driven by humans, with humans exiting stations. Put in 2 - 3 train lines underground, which is say 6 metres wide, chuck a station on top with escalators pumping people in and out, and people can walk through the city. Put a park on top. Sydney Central Station has 27 platforms, 10 underground. That's like 280,000 people per hour on the underground lines by those dodgy estimates, or 5 x 200kph 3 back to back traffic lanes (not tested yet), or 185 lanes of 26mph traffic which is ¾ of a kilometre wide! And it was built in 1906 out of sandstone! There are ~ 5 CBD stations in Sydney. A 925 lane wide road sounds like fun. And its here in 2018 and it costs me AU$1.50 to use. Cars don't belong in cities. Cars belong in the suburbs, which is where they can drive you to your train station. Tokyo's Shinkansen was opened in 1964 and went 220kph. If the train is more than 1 minute off the timetable you get an apology letter for your employer. They took the original trains out of service in 2008. The USA is so behind. As I said, transport isn't a technical problem, it's a political one. The future is here (high speed rail), it just isn't evenly distributed yet. [1]: http://infinitemonkeycorps.net/projects/cityspeed/ http://infinitemonkeycorps.net/projects/cityspeed/ [2]: http://moderntransit.org/solution.html http://moderntransit.org/solution.html [3]: http://www.abc.net.au/news/2014-04-11/barry-ofarrell-sydney-trains-claim-doubtful/5371446 http://www.abc.net.au/news/2014-04-11/barry-ofarrell-sydney-...

dmix 9y ago

> Cars on freeways or sleds don't carry nearly as many people/hour. Almost entirely because of how car ownership is structured now. People most often buy 4dr cars and only use 1-2 seats 90% of the time. A subscription-based self-driving fleet car system would change this dynamic where you only order a 4dr+ car/SUV/van for a ride when needed. Otherwise a small 2dr electric vehicle without a large front motor will likely be the default in the future. On-demand cars will be great for the environment and efficiency of highways... Not to mention there has been some talk about self-driving cars linking up like trains on freeways to conserve energy and better utilize space at very high speeds.

jakecopp 9y ago
I am optimistic about self driving cars, but we also lower to floor of possible vehicle occupancy to 0. I'm curious to see if average occupancy goes up or down. Linking up cars sounds great! Maybe we could make the cars electric. And we could create escape hatches to walk between them. We could make them shorter so the car is essentially a seat. And then we could make them 100% subscription based and shared so you never take them away. I think I might just have reinvented trains! Our obsession with automobiles borders on insanity. Yes, the last mile is hard to do with public transit but for high capacity, high cost thoroughfares trains win every time.
- dmix 9y ago
  > but we also lower to floor of possible vehicle occupancy to 0 That's a great point that I hadn't yet considered. In terms of finite energy resource demand it's not as big of a deal when the average car both smaller and electric. But in terms of traffic and related resource utilization this will ultimately become a far more common phenomenon, even possibly in the top 1-2 most frequent scenarios as driverless cars attempt to reach a new client or drive to the closest automated charging port. I'd be curious to see the traffic/resource impact of such a hypothetical scenario. Although I anticipate some smarter people than me are already calculating these costs. At least I hope so!
  - jakecopp 9y ago