What is the best shape of a city? Modelling effect of urban form on distance | Hacker News Reader

What is the best shape of a city? Modelling effect of urban form on distance

journals.sagepub.com

42 pointsrustoo28 comments

This is not quite a spherical cow and there is something to be learned here, but really, I'd like to see this with

- non-square-grid cities with highways, public transport, …

- realistic synthetic populations doing non-random destination selection (nobody randomly goes to a supermarket at the other end of a city, after all)

- actual traffic simulation with jams (or at least using real traffic data)

and then evaluate both the energy use and travel times. That being said, doing this kind of thing, e.g. with ActivitySim, even for one real city, is quite a bit of work.

Google Maps has all that data.

You can look at small parts of it there. And sometimes you can't even buy what you're looking at, GMaps API won't hand out public transport trips in some countries. And where you can, it's fairly expensive. (No idea about traffic or people movement data.)

Restricted access to the data makes it hard for you to do it.

But that’s a you problem.

Not only that, it also means that some other researcher can't do it either, and I can't get free access to the results.

In any case, do you have some helpful comments too, maybe?

The reality is a lot depends on your economy. A manufacturing and port city will necessarily be very different urban form to a white collar office job city or a beach/lifestyle city full of remote workers. An egalitarian city will be very different from a socially segregated city. Etc.

Cities are also never perfectly efficient because they are always evolving in function and use

Tangential, but Barcelona has superblocks which seem to work fine:

https://citychangers.org/barcelona-superblocks/

Also the blocks are octagons which improve traffic flow: https://www.reddit.com/r/CityPorn/s/iCcOGMqoNH

Here's a preprint from 2025: https://arxiv.org/abs/2507.00100v1

And an explainer/simulator here: https://vis.csh.ac.at/cities-morphology/#/

Surprisingly it’s not a line

I think an unplanned city or town that isnt a grid is the best. Grid lines become boring very quickly because they are repetitive without any interesting parks or places that are different to stop and look at. If everything is the same its probably has the same busy roads with little or no trees or shade for a hot day.

This was an interesting tangential read. Fascinating to read about the first Sim City games and the assumptions/calculations it made in scoring. https://news.ycombinator.com/item?id=49476193

Isnt it that Singapore is "the greenest city" in the world? And since the started late to urbanize, the design of the city & districts was led by a "scientific approach", regarding air circulation etc., IIRC?

Hexagons are bestagons

(apparently a city based on a hexagonal grid would be more efficient)

(has played sim city thousands of times since 1993) "I got this."

The Line begs to differ!!

https://en.wikipedia.org/wiki/The_Line,_Saudi_Arabia

+1 on Barcelona, it feels kinda boring, but boring is good if the goal is living well

Excellent. I am looking for better strategies for Cities Skylines

Cities Skylines is not a game that rewards strategy, because of how it works it is primarily about defending against and exploiting the simulation's deficiencies

It's definitely a drag battling cars most of the time. The path optimisation that forces them to all choose 1 lane when 3 lanes are available and go to the same direction is so annoying.

I would like a better alternative but eh, it's just spare time fun thing

Just look at Magnasanti for the most efficient sim city

Without getting behind the paywall, the abstract says that how far you have to travel determines how far you have to travel??

Every 60 seconds a minute passes

Well stop clapping you stupid git

Iykyk

I was going to dismiss this and then I read the whole abstract and that is, pretty much what it says. Oh but I guess it also says they came up with a model for how far apart things are determining how far you have to travel to get there

I think the best shape for a city is a Mobius strip: go anywhere without turns, so no traffic lights needed.

If you're thinking in terms of optimizing car traffic there's a good chance this city will suck

I forget that actual reference.

But there was an article or map or something talking about the placement of towns, I think, back east (or the south, or Virginia).

Simply, the towns were placed within a days travel of each other. And the communities they served were, effectively, about a 1/3rd days travel in radius. That's how far folks would go to get into town, and being a day apart facilitated long range travel instead of camping in the wilderness.

The towns were basically small towns designed to serve those local communities. Cities formed on the intersections of such roads to support more trade and traffic.

I don't recall if it talked about towns along rivers where transit might be faster.

But it was an interesting article with a map of actual towns.

Reply on news.ycombinator.com