Monday, January 06, 2014
Liebestraum No. 3
Idea of the day: as long as the different kinds of terrain (e.g. plains, mountains, desert) are very well distinguishable from each other I don't see why a continuous portion of terrain type can't be very detailed and varied.
Saturday, January 04, 2014
La Balanga
I'm still trying to decide the overall appearance of the game map. This is going to influence the techincal sides obviously but it will have other subtle consequences.
Usually maps in tbs games just suck, they almost universally use tiles of some form. Tiles are repetitive but have the big advantage of being very readable. The aspect of the tile immediately tells the player the kind of terrain he's going to send his unit to.
Also, tiles mean that drawing the map will be very easy, so modders and scenario makers will have no difficulties in making new contents.
A game which does not let users to make new contents on their own, especially a strategy game, would have a very short life, new maps and scenarios are simply one of the most important building blocks of such games.
The conclusion is that it must be easy to design new maps, with or without tiles. Tiles mean that there must be a scenario maker. No tiles means that an external program must be used to design the map. I tend to rule out vector maps as vectors are much more complex to handle than raster images. However raster images can't be as easily referenced as vectors.
Then there's the issue of how beautiful should be the map: a rich map, full of nuances and details can't be easily read. Just imagine the player in front of a photo-realistic aerial image. Can he tell if something is a farmland, or grassland, if something is a town or rugged terrain? he could move the mouse over the feature and gather all the informations he wants (provided that the problems of georeferencing all the terrain features are resolved), but it would be impossible make calculations just by looking at the map. So the map should have less but more recognizable features. In practice this would mean that the game would have the visual poverty of tiles with the additional difficulty of making new scenarios.
Usually maps in tbs games just suck, they almost universally use tiles of some form. Tiles are repetitive but have the big advantage of being very readable. The aspect of the tile immediately tells the player the kind of terrain he's going to send his unit to.
Also, tiles mean that drawing the map will be very easy, so modders and scenario makers will have no difficulties in making new contents.
A game which does not let users to make new contents on their own, especially a strategy game, would have a very short life, new maps and scenarios are simply one of the most important building blocks of such games.
The conclusion is that it must be easy to design new maps, with or without tiles. Tiles mean that there must be a scenario maker. No tiles means that an external program must be used to design the map. I tend to rule out vector maps as vectors are much more complex to handle than raster images. However raster images can't be as easily referenced as vectors.
Then there's the issue of how beautiful should be the map: a rich map, full of nuances and details can't be easily read. Just imagine the player in front of a photo-realistic aerial image. Can he tell if something is a farmland, or grassland, if something is a town or rugged terrain? he could move the mouse over the feature and gather all the informations he wants (provided that the problems of georeferencing all the terrain features are resolved), but it would be impossible make calculations just by looking at the map. So the map should have less but more recognizable features. In practice this would mean that the game would have the visual poverty of tiles with the additional difficulty of making new scenarios.
Wednesday, January 01, 2014
La ricostruzione del Mocambo
I doubt I can realize the project with quest3d. The main problem is the creation of a tile-based map, I've seen that unity has some plugins that can create it almost automatically, while using quest3d would be nearly impossible, or at least it would not be scalable.
I have to make more tests, but the advantage of using unity is that it has an enormous community plus a ton of plugins, it would be great if I could find one that lets you read shapefiles, this way I'd not have to rely on tiles but I'd directly read fully georeferenced tarrain data.
Quest3d also is a moribund program, I don't think it's worth to use it anymore.
I have to make more tests, but the advantage of using unity is that it has an enormous community plus a ton of plugins, it would be great if I could find one that lets you read shapefiles, this way I'd not have to rely on tiles but I'd directly read fully georeferenced tarrain data.
Quest3d also is a moribund program, I don't think it's worth to use it anymore.
