These people are so fucking dumb. Let me do some quick math.
Any given week day, at peak hour, any train from Madrid's metro can be carrying around 500 people. Let's consider that some of them may be traveling together, so let's say that being optimistic, they could be divided in 250 robotaxies.
Now, that's one train. At any moment there are between 10 and 20 trains going in that line in that same direction. And as much in the opposite. That makes it between 5k and 10k robotaxies. To cover a single metro line
Madrid has 16 metro lines. That makes it between 80k and 160k robotaxies just they could move the same order of magnitude of people our Metro moves in any given workday. If you want to cover really busy moments, like big sport events, you need twice that amount (the nominal max capacity of metro trains is around 1200 people per train, and believe me, I've seen them so full that you couldn't get in pretty often).
That's just for the subway. Local commuter trains and city buses, combined, are actually carrying more people daily than the subway. So let's double the robotaxies fleet again if they want it to "kill public transport". So you need between 350k and 700k robotaxies just to be able to kill the public transit of a single major city.
That's about 1/6th of every Tesla ever made. Just. For. One. City.
@javi@cygnathreadbare The sentence just doesn't apply in my opinion, autonomous vehicles will come in many form factors, not just "cars", and there will be public and private fleets, it's not about public vs private. Tesla will probably end up licensing their technology to many companies. Autonomous and algorithmically optimized ride hailing will be much cheaper, more efficient, and much more convenient, and will achieve comparable throughputs even for high density routes.
@javi@cygnathreadbare This isn't about musk or Tesla, it's a bigger technological trend, similar to how distribution has changed with things like amazon prime or to how server infrastructure has changed with cloud providers etc..
Yeah but the thing is that there are physical limits autonomous driving can't cross. No matter how good the algo gets, it won't ever be more efficient at moving people than a train. Even than a bus line
@javi@cygnathreadbare trains have high throughput but not as much as you would think mostly because of the space and time needed between trains. They are also very rigid and inflexible. They are nice for some main routes but that's it, the inflexibility of these systems is what will kill them in my opinion. Autonomous fleets will be much more flexible and "fluid" and they will provide very predictable and point to point transport, so very convenient and it will be very cheap.
My local subway trains carry up to 1250 people and the time between trains, in peak hours, is 3 minutes. That's 25k people per hour per station. With >300 stations all over the city.
Let's say a robotaxi can make two trips per hour. So to get the same max potential capacity you need 12k taxis. Per station.
@javi@cygnathreadbare The throughput isn't really a problem since you can have bigger "cars" that will be fully utilized for main routes, it's like a fluid train where each wagon can detach and go to different destinations. At some point they will drive very fast and very close to each other due to superhuman perception and reaction times etc.. that could take quite a bit of time though.
They can't drive fast in a city. 50km/h is the top within a city here, most of the streets are 30. And every time they drop a passenger, they stop all the other "buses" going through the same street. There are physical limits that can't be solved with algorithms
Also, public transit routes are already optimized. They have been under optimization for decades. You can't have a bus stopping at will to drop a passenger in whatever exact point they want: buses have special infrastructure so they can stop without interrupting the entire tragic flow. I would say all of this is a pipe dream, but I strongly suspect it's mostly a con.
@javi also more people commute in one direction than the opposite so there's no way all those robo taxis are going to fit all at the same time in the same place.
But that is precsely the hope and plan of the car industry.
To save theplane, we must drastically reduce the burning of fossil fuels. For electricity generation we can switch to solar, wind, and hydro; and that is already happening. For transportation, we must expand public transportation, improve the biking infrastructure, put housing within walking distance of work until 90% (say) of human and cargo traffic that now uses cars and trucks switches to those modes. >>
@javi >> But then 90% of the people will see no reason to own a car. THAT is the car's industry nightmare.
So the car industry is doing what they have been doing for the past century: fight public transportation, so that each family, no matter how poor, MUST buy at least one private car. >>
@JorgeStolfi
your estimation is overly optimistic. I'm all for public transport but that can only cover popular destinations efficiently. You will still need a car to go to places where there is no public transport available or your travel will take a very long time. @javi
OK, maybe 90% is unrealistic. But ~60% of people live in cities. If most of those people can use metro, train, or fast bus for commuting to work and regular shopping, most of the remaining uses of personal cars can be supplied by taxis and rented cars.
As for people in rural areas, they will probably still want to have personal cars; but how much travel do they actually do?
High-speed trains can cut down airplane use. Shipping is going to be a problem though. >>
I don't have much hope of governments outright forcing most people to switch away from personal cars. At most the gov can nudge them a bit by increasing taxes on cars and gasoline, closing streets to cars, etc; but those measures will be unpopular and will be easily overturned under a democratic gov.
I hope the same result can be achieved by simply offering people better (faster, cheaper, ubiquitous, and comfortable) public transportation alternatives.
@hajovonta@JorgeStolfi@javi Then make public transport available to all places. Busses can go anywhere cars can go, all you need is a few bus routes to actually go there. Slightly increased travel time is definitely worth it compared to having to own, maintain, and fuel a car.
@tyil
Nah. I live in Eastern Europe, and these discussions often sound as "entitled". We have underdeveloped public transport, because to run these efficiently, they need many people to travel together. That's why I wrote it's only feasible to cover popular destinations.
Just to give an example, if I wanted to get to work by public transportation, it takes 80 minutes with two line changes. By car, it's ~30 minutes.
like, I grew up in a 3500 people town. Still, we have a bus every 10 minutes to the nearby towns, both a bus and a train every hour to the provincial capital. Those buses and trains were almost empty most of the time: sure they would get full at peak hours, but taking 11am train? you are going to have a nice trip around spanish northern coast with a train almost to your own. It's possible to have something like that everywhere. But public transport needs to be considered a cost to the state, a right of the citizens that need to be provided for, and not a for-profit business industry.
Add comment