Maps, theory discussion

[geoschmo]

I’ve been thinking a lot about maps lately and what makes a good one for Se4. In my opinion the bane of good map making is the ancient race trait. This is speaking specifically about maps for two player games such as Koth or the Ratings bash tourney. You try and design a good map that is even and gives both players a decent chance, and then one of them takes the ancient race, finds the choke points, grabs 75% of the map in the first ten turns and then just starves the other guy out of the game. Biding your time and building up an attack fleet, while the other guy struggles to defend himself from attack everywhere at once.

I had hoped that using spiral arm quadrants would make this less serious, but it doesn’t really help. In some cases it makes things worse because the ancient race player can come at the other guy from two directions around the center of the map.

My philosophy is that starting out close to the other guy isn’t so bad, as long as the other guy hasn’t got you boxed into 5 systems also. I stumbled on a new variant of map that I think may solve the problem. I think it gives the non-ancient race guy a decent shot, without totally taking away the advantage of the ancient race. I call it the “C-type Spiral Arm”, or C-map. All it is a spiral arm map that doesn’t connect all the way around the center.

Here’s an example:


They are generated randomly on occasion, but they are easy to manufacture. You simply create a random spiral arm and then remove a couple warp points to break the connection on one side. (Don’t forget to remove both ends of a warp point pair.) Then you set a couple common starting points approximately across the center from each other so that each player has around the same size “arm” behind him.

The ancient race player will still have an advantage of knowing where the other guy is, and on which side the map is broken. He can send everything the shortest distance between them and should still be able to capture the bulk of the map. However, the non-ancient race guy, once he figures out where the enemy is, can defend the choke points on one side and have a good amount of systems that can build up unmolested. This way even though he’s at a disadvantage in number of systems, he still has a shot at holding the other guy off and maybe breaking out if he plays well.

Another kind of map I’ve been playing around with lately is a sphere map. These aren’t so easy to make. You take a grid map, it works better if it’s a full grid, 13x9, and then make new warp points to wrap around side to side and top to bottom. This way the map really has no corners and everybody can get to anybody from any direction. It changes the strategy of the game quite a bit. There are a couple included in the FQM stadard edition if you need to see what I am talking about.

However, I have realized that it’s not really a sphere, even though I am calling it one. A sphere represented on a grid would wrap around side to side, but not top to bottom. To make a truer sphere it would need a “polar” system on top and one on bottom that was connected to all 13 of the systems along it’s edge.

I am trying to imagine what 3D shape my original idea represents. Not sure it is a shape that can be truly represented in three dimensions. Any ideas? Does it make a donut shape? What's that called? A torus? This stuff hurts my head when I think about it too much.

[ January 19, 2004, 14:12: Message edited by: geoschmo ]

[Loser]

Yes, that would be a torus.

If you take a rectangle and connect the top edge to the bottom and the left edge to the right you have a torus.

[dHay]

Wrap around the top and sides makes a donut shape.
Donut is a very good shape for limited choke points.

http://www.wolfpackempire.com/ plays a donut world.

The number of sectors in your small game maps could make it difficult to get the effect of a donut. Not enough distance, so pretty much everything is a couple of hops away. Which might remove a lot of the terrain tactics.

[Cipher7071]

Topographically speaking, the torus is really the same thing as a spiral map, unless, of course, you break it into a C like you did in your sample map. It really does sound like you are looking for the topographical equivalent of a sphere.

edit:

Instead of thinking about it in terms of your common garden variety Earth-globe, try thinking about it in terms of a ploygonal solid with a system at each of the vertices. In other words, there doesn't need to be a clear-cut 'pole' system. Any vertex might be seen as a 'pole.'

First consider a cube, and what you would need to do to represent it in 2-D on your computer screen. Draw it on a piece of paper. The trouble then becomes the number of crossing warp-lines that this makes. A cube can be un-crossed without too much difficulty (try it!), but as the number of systems (vertices) increases the map will become too messy before you get anywhere near the number of systems in a large quadrant.

That's why the compromises are made. Hope this gives you some ideas.

[ January 19, 2004, 15:13: Message edited by: Cipher7071 ]

[geoschmo]

Originally posted by Cipher7071:
Topographically speaking, the torus is really the same thing as a spiral map, unless, of course, you break it into a C like you did in your sample map. It really does sound like you are looking for the topographical equivalent of a sphere.

Actually, the C-map and the sphere map were two totally different trains of thought. I merely put both of them in one post. However, a torus map broken into a C is just a cylinder, right? Regardless of it's 3D shape I mean, since that is ignored for game purposes.

Originally posted by Cipher7071:
edit: Instead of thinking about it in terms of your common garden variety Earth-globe, try thinking about it in terms of a ploygonal solid with systems at each of the vertices. In other words, there doesn't need to be a clear-cut 'pole' system.

Not toally sure what you are getting at here. Instead of having a polar system would you have interconnects between all the systems at the top? Not connecting them top to bottom, but like spokes across the end of a cylinder I mean?

[Cipher7071]

Geo, you caught me in the middle of an edit...go back to my original post.