Take the 2-minute tour ×
Game Development Stack Exchange is a question and answer site for professional and independent game developers. It's 100% free, no registration required.

I've searched around and it seems like most people are using tile-based map systems.
I suppose this question is more theoretical than practical (I am not very concerned about memory or performance speed), but I want to know: what other ways can a map be created in a game?
A map being a graphic representation of terrain that can be navigated, has entrances and exits, and boundares (no-go zones).

Besides using text files to store and arrays to load tile data, one idea I had was to store a map entirely as a graphic file and use queries on the pixel colour to determine boundaries (ie, you can only move in a certain direction if the way is bright enough in that direction).

What other creative map systems are out there?

share|improve this question
    
storing it as a graphic file using pixeldata is still tile-based if you ask me, its just a different storing method for the same thing... –  MrJre Aug 22 '12 at 14:57
    
Yea, I suppose you're right, but the tiles aren't grass, water, tree, etc., and a pixel is the smallest tile available, so it gives freedom to have map elements of arbitrary smallness. Also, why the -1? I am trying to brainstorm ideas and see if anyone else is making non-traditional style maps. Should this be in meta? –  sideways8 Aug 23 '12 at 1:46
    
do you consider something like this not a 'tile-based map system', javilop.com/gamedev/… ? –  georgek Aug 23 '12 at 3:06
    
Yea, that one looks interesting. It kind of reminds me of Freddi Fish, though I don't know if it would play similarly. It looks like the graphic background doesn't have any notion of barriers, and collision detection is only needed with regard to the images you drop into the map. Neat. –  sideways8 Aug 23 '12 at 3:55
1  
I'm voting to close as not constructive. From the FAQ: You should only ask practical, answerable questions based on actual problems that you face. Chatty, open-ended questions diminish the usefulness of our site and push other questions off the front page. –  doppelgreener Nov 20 '12 at 12:11

2 Answers 2

Instead of using tilemaps you can use general objects transformations. Like: house01 x:151 y:150 angle:30. Also you can generate your map on the fly.

In storing map as bitmap you use 32bits per pixel for example, and in the simpliest tilemap only 8bits(ASCII char). And using bitmap for map generation it's just a way for storing data. Anyway your engine will interpret data as required, so your idea is right about save/load tilemap.

share|improve this answer
    
Can you expound on 'generate your map on the fly.' ? Would such a map be reproducible for the next time the user visits that region? –  sideways8 Aug 22 '12 at 10:55
    
i you use the same seed for the region, then it will be as the first time. –  PATRY Aug 22 '12 at 11:03
    
Ok, that makes sense. That still seems to me to be something you could only do with a tile or sprite oriented map design. I'm really looking for ways that would break the traditional mold of game maps. The more unconventional, the better. –  sideways8 Aug 22 '12 at 11:21
    
You can generate your map from satellite photo ;) As for seed, you can save your generated map (of course if changes were made) and load when you come back. –  bobenko Aug 23 '12 at 16:58

Well, Theoretically speaking, I believe there are three main ways of creating a map: 1. Graphs (which include the tile system); 2. continuous (bouncing box) system and 3. mixture of both.

  1. The graphs approach consists in telling each identity which are its possible movements. It implies that every identity as a position on a graph, and can only move to certain positions: e.g. in a square tile system, each identity can only move to up/down/left/right (and diagonals in some cases), which is basically to say that each tile has a set of links connecting it to other tiles. Forbidden places are set by not linking that node.

  2. On the other hand, the continuous approach, basically the identity can move freely everywhere, subject to the constraints defined by the bouncing boxes of other identities. For instance, the map can be a set of vertices that define a smooth surface, and when the identity moves trough the surface, it is repositioned on top of it, according to the maps height and identity height.

  3. Finally, probably the most used now-a-days is a mix: a graph system to define your tile system and on each tile, you use a continuous approach. Only the tile where the identity is is loaded (and possible close tiles).

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.