Population density, number of buildings etc

[Ghostman]

How do you people determine how many buildings will be in a city/town? Do you first decide on the population figure and then estimate the number of houses? Or do you create the map first and then figure out the population? Do you even consider this matter at all?

I'm personally haunted by these kind of questions. When I sit down to create a city map, I want the population density and the city size to be at least somewhat realistic. But the towns I want to map are not like modern ones; they are supposed to more or less resemble historical population centers of varying levels of technology. That makes things difficult to me. On one hand I don't think modern cities would make good examples. On the other hand I don't have enough knowledge about historical cities and their buildings.

Just how big were common houses in the past and how many people were living in one? How much did these things vary between different parts of the world and between different eras? I'm thinking that in general houses were smaller and that people were more densely crammed into them than in modern times. The houses may have been more closely packed too, with narrower streets between.

Having actual examples of preserved historical towns (so that the number and size of buildings can be known) with reliable information on their populations in the past would be invaluable help, but I'm not aware of any such examples.

[ravells]

There is a good explanation of medieval city demographics by S. John Ross which answers some of your questions.

[robb75067]

There is a good explanation of medieval city demographics by S. John Ross which answers some of your questions.

Thanks for the link.

[dayday247]

Anyone know of eco-system stimulates or engines?

[Ilwan]

As does Erin´s low fantasy populator, based on the same as the one mentioned above:

find it here: http://www.welshpiper.com/lowfantasy

[Midgardsormr]

It might be helpful to think about how the people live in the society you want to depict. Habitation patterns are directly related to how families form and how people make their living. A society that values independance and has relatively loose ties to family will probably have more houses than one that values cooperation and tight family bonds. In the first case, I'd expect the number of homes to be equal to around a quarter to one-eighth of the total population. In the latter, expect the number of houses to be somewhere around 1/20th of the population, but the houses will be much larger because you'll have four generations all living under one roof.

Other things to consider are subsistence, division of labor, climate, and terrain. All of these things will have an effect on the size of homes, how they are constructed, and where they are placed. A mountain city will have smaller houses built close together because stable flat land is at a premium. A city on a huge floodplain might have larger houses, each with its own garden plot, because there is plenty of space for the city to sprawl out.

Cities with a strong middle class might have tradesmen living in the same building they work in, leading to cities with a small number of stand-alone houses. Those that base their economy on mass-producing goods or harvesting a particular natural resource will probably have a number of houses placed close to a central shop or resource processing area.

It might be worthwhile to look for an old cultural anthropology textbook in any used bookstores you happen to find yourself in. I started asking many of these questions after I took such a course a couple of years ago.

[Badger]

I generally use a ratio of 15% to population for the # of buildings, and the area of the city equal to 1500 square feet multiplied by population. So a city of 5000 residents would have 750 buildings and cover roughly 7,500,000 sq ft.

These numbers are in no way based on historical references, but my 'average' formula for mapping out a town, and haven't failed me yet. They do however seem to work for larger populations of 1500 people or more...