Oolite Bulletins

submersible wrote:

90 splats , Now with a 'coastal-clumpiness' fudge.
<image snip/>

That's looking even better.

What I was wondering though is can you either extract lists of "large" cities' positions from the data or more likely and preferable is start with a list of cities and their locations and add those to the algorithm (though chosing those locations would be an algorithm in its own right). The reason I ask as it would be great to do overlays that point out the Capital and maybe some major cities.

PhantorGorth wrote:

submersible wrote:

90 splats , Now with a 'coastal-clumpiness' fudge.
<image snip/>

That's looking even better.

What I was wondering though is can you either extract lists of "large" cities' positions from the data or more likely and preferable is start with a list of cities and their locations and add those to the algorithm (though chosing those locations would be an algorithm in its own right). The reason I ask as it would be great to do overlays that point out the Capital and maybe some major cities.

That's looking great!

And yeah, that would be really cool. BTW, the capital doesn't have to be big...

Rorschachhamster wrote:

That's looking great!

And yeah, that would be really cool. BTW, the capital doesn't have to be big...

That's is very true, after all I know this because I live in Edinburgh. My statement was more about marking say the ten most prominent/important cities or rather actually specifying them and then they get added to the map (which doesn't necessarily mean the Capital is the largest).

PhantorGorth wrote:

...start with a list of cities and their locations and add those to the algorithm (though chosing those locations would be an algorithm in its own right).

The Ouroboros strikes again!. The macro that is placing the splats could be made to write out the splat positions to a file (points on a unit-sphere), however determining which should be a capital or major city is probably more of a difficult problem best left to hoomans.

Example from before , but with red dots where the splat was actually placed - many of these fall in the sea.

submersible wrote:

PhantorGorth wrote:

...start with a list of cities and their locations and add those to the algorithm (though chosing those locations would be an algorithm in its own right).

The Ouroboros strikes again!. The macro that is placing the splats could be made to write out the splat positions to a file (points on a unit-sphere), however determining which should be a capital or major city is probably more of a difficult problem best left to hoomans.

Example from before , but with red dots where the splat was actually placed - many of these fall in the sea.

Interesting - the point where the splat falls is not automatically the point with the biggest density?

Oh, and for some planets the possibility for having these in the sea (or above the sea, actually) would make sense, I think...

Rorschachhamster wrote:

Interesting - the point where the splat falls is not automatically the point with the biggest density?

Yes - I am a little curious about that too. In fact - the whole splatting approach might be a red herring. Alas - more experiments needed.

Rorschachhamster wrote:

Oh, and for some planets the possibility for having these in the sea (or above the sea, actually) would make sense, I think...

Certainly - I imagine at least 4 different city density bias behaviors; Coastal Clumping, Shallow Dwellers, Oceans Deep and Mountains High. Defined by some inscrutable means based on inhabitants, sun, solar-flares, earthquakyness ,etc.

It's looking great .