Spherical Map

[Spiffor]

Well, Civ2 had the option too, it was called a "buoy" map.

I think a spherical map would be much cooler though. I want my Murmansk to be nearest of the American shores !

[Panag]

Originally posted by Spiffor
Well, Civ2 had the option too, it was called a "buoy" map.

I think a spherical map would be much cooler though. I want my Murmansk to be nearest of the American shores !

hi ,

seconded

and subs should be able to go under the ice

have a nice day

[TimeTraveler]

I'm actually thinking of designing a civ-type game with a spherical map that uses hexagons and the occasional pentagon.

I've worked calculations with geodesic spheres like this and found that the number of tiles on any such map would be 10 times the square of the distance between two closer pentagons.
The distance can be calculated by going a distance x, turning 60 degrees, and going a distance y. The distance is (x^2+y^2+x*y)^(1/2).
Some examples of success are how the dodecahedron has 10*(1+0+0)+2=12 tiles, and the soccer ball has 10*(1+1+1)+2=32 tiles.

For anyone wishing to create a custom-sized map, they can choose an angle and distance between pentagons, possibly using the x-y system if I can simplify the explanation, but I'm thinking of keeping any map from going smaller than 912 tiles.

[LDiCesare]

I'm pretty sure we discussed something like that in the Clash forum. I don't know whcich thrad it was, but it's not that old... Spherical coordinates (hexes + some pentagons) can be done without too many problems in terms of modelisation. Rendering is a bit more tricky, and you'd in fact want a flat minimap somewhere.
About tileless maps, this has been suggested too in the Clash forums. Gary Thomas actually thought of polygons rather than regular polygons to make the map. Thus you could bundle many squares (or hexagons, ...) into one big area, f.e. sea or desert, and have exact coordinates for units. Search for polygons in the Clash forum if you're interested.

[Dauphin]

I think the main problems is not how to do it, but how to do it without puts the computer specs required too high.

I'm wondering how much memory is required to store information on a cartesian grid, and how much memory is required for a polar variable-sized irregular polygon grid.

[Trifna]

Well wasn't Populous (many years old) built on a spherical map? It doesn't seem to add that much more information to manage for a computer. Perhaps it would ask a little more graphically? Well then it wouldn't make it a serious problem since the average computer now has a pretty strong video card, and quite some speed. If a civ game gets on the market in one or two years, just imagine how much cpu strength wont be a limit...

[Dauphin]

Too my recollection Populous wasn't a tiled game, it was a continuous surface. This removes a lot of the problems.

Any game could easily be made on a spherical map if you don't need discrete tiling. Of course removing tile from Civ would lead to a loss of many of its central features.

[Trifna]

Wouldn't a football-type tiling be correct?

[Dauphin]

Football tiling varies depending on the number of tiles being used.

[Trifna]

Then it would simply mean that the number of tiles should be dependant on the tilling :P
It's not that much of a handicap.

[Dauphin]

Its not just a change in quantity but it also effects the shape and orientation of each and every tile.

[Vince278]

Spherical maps have been done successfully in other games. The best example I can think of is X-Com. The strategic map is a globe of the earth that you can zoom or rotate on a North-South or East-West axis at will. Something like this would look really nice in a future Civ, or similar, game.

[geeslaka]

Why use only spheres? Why shouldn't the map be wrapped around other thngs?
My favorite idea for the topology of a map is this:

Originally posted by geeslaka
Another topology came to me in a dream last night. Take the formula that is applied to the top and bottom of the rectangle in the cylinder capped with frictionless hemispheres topology. Apply that formula to cases where units go off the left or right side as well and you get a much more complex topology.

 ________________________________
|         a               b      |
|h                               |
|                               c|
|                                |
|                                |
|g                               |
|                               d|
|           f               e    |
 ________________________________

If a unit goes up from point a it moves to point b.
If a unit goes right from point c it moves to point d.
If a unit goes down from point e it moves to point f.
If a unit goes left from point g it moves to point h.

For other topology ideas of mine see this thread. http://apolyton.net/forums/showthrea...threadid=89943

[Vince278]

Originally posted by geeslaka
Why use only spheres? Why shouldn't the map be wrapped around other thngs?

Because we are trying to accurately simulate worlds?

[geeslaka]

Spheres should be just one of many options, not the only one. If people want to play on a map that resembles the real world they can choose a sphere. Besides, until someone develops a way of dynamically figuring the tiles needed for a spheroid or the tile system is dropped entirely, we are stuck with squares. Why shouldn't we do as much with squares as we possibly can?