Why use tiles at all (square or hex)?

[Friar Bellows]

I'm wondering why tiles are used at all in computer games. Computers can handle 2 or 3 axes of real numbers, so why use tiles at all, whether hex or square or octogon or Penrose ?

I suppose designing the AI is easier in a tile-based system. You can navigate the AI units a lot easier, find shortest paths a lot easier, that sort of thing. Maintaining AI front lines might be easier. But I don't really know, I'm just guessing. I'd like to know the advantages of using a tile-based system.

[Victor Reijkersz]

Dear friar,

It is a bit of neccesity in abstract games. Since where exactly on a x,y access would you position an Army Corps? I guess it could be done using some fuzzy logics and moddeling frontlines, but a real x-y axes system is much more suited for tactical wargames. And they do use them. I think Close Combat or Combat Mission are fine examples.

[Moon]

It can be done, obviously, as Airborne Assault-then-Highway to the Reich have shown. But - like so many things in (good) game design - it's always a question of pain vs. gain: just because something is possible, doesn't mean that it will make a given game better.

Will a strategic game on an Army scale like SC benefit much from such a system? If you think about it long and hard, you'll realise that there are only few benefits. At the same time, there is a disproportionate increase in coding effort - AND the game will get inherently more complex to the player, as well.

An operational level game like the one mentioned above (or obviously a tactical game like Combat Mission) profit much more from a system like this. As a rule of thumb, given a certain level of abstractness, a game can sometimes work "better" if the game systems it uses remain on the same level of abstractness.

Martin

[pzgndr]

Combat Mission and TacOps don't use tiles and they play fine. They also have simultaneous turns, which also works fine. An argument could be made to try something like that for a strategic game. As you point out, the path-finding algorithms and AI routines may be better with some sort of tile grid for large strategic games having resources to manage, different countries with diplomatic options, etc. 1000 data points are easier to manage than 100,000, unless you have your own Cray supercomputer at home. It's an interesting idea, and as computers get faster and AIs become better then I expect we'll see more development along these lines.

Besides what the computer can handle, what about the guy at the keyboard? Conceptually we can all imagine the "ultimate" WWII game with a global map having 1 km resolution, battalions which accurately account for every man, vehicle and weapon, and simultaneous game turns lasting maybe an hour. But who could possibly play such a thing? Players can completely finish a game of SC in about 20 hours or so, which is very reasonable for replayability and competition. Turn-based and tile/hex-based systems allow game designers to abstract things so players can adequately manage what they've got and still play a decent game, either TCP/IP or PBEM. So that's a perspective to keep in mind.

[Friar Bellows]

Don't Combat Mission and Close Combat use an underlying tile system? Granted, it's invisible to the user, but isn't it still tile-based? I have no idea, except some hazy memory of it. Would like to learn more about the system used by Highway-to-the-Reich. Thanks for your responses.

[Moon]

Friar, Combat Mission uses a tile system for the editor only, but the combat resolution is point-to-point. In other words - the terrain is built based on tiles or "blocks" of terrain, but after the game starts, lines of sight, targetting, exposure and so on are "real" in the 3D environment.

Martin

[Friar Bellows]

Thanks Martin.