|
Re: What are you working on?
[Re: HeelX]
#392633
01/25/12 09:51
01/25/12 09:51
|
Joined: Mar 2011
Posts: 3,150 Budapest
sivan
Expert
|
Expert
Joined: Mar 2011
Posts: 3,150
Budapest
|
@HeelX, great job, that procedural sea-wall thingie is a really neat idea. I also had fun with terrain editing, brush based vertex manipulation and setting hull and normals is okay, but I'm too stupid for the SDK (and absolutely have no time to dig into that deeply). I simply use the .dll what Widi used in his terrain deformer, but it works perfectly only in A7 compilation, in A8 the first chunk remains unchanged somehow (and unfortunately Widi couldn't help me)... fortunately in my editor saving/loading a vertex modifier array has its role beside direct storage in hmp. moreover, as a relaxation, I'm experiencing with hierarchical pathfinding taking into account different unit sizes and movement capabilities (atm building up the necessary database):
|
|
|
Re: What are you working on?
[Re: Damocles_]
#392647
01/25/12 13:07
01/25/12 13:07
|
Joined: Mar 2011
Posts: 3,150 Budapest
sivan
Expert
|
Expert
Joined: Mar 2011
Posts: 3,150
Budapest
|
yes, I do something like that, to achieve group movement, briefly: all is done within my MapBuilder editor (without manual settings required): 10x10 tiles form a default area (called as cluster), which are automatically subdivided when they contain some homogeneous sub-areas, like an impassable high slope abyss, or water, or badly walkable hill or swamp. Each area move cost is also calculated, and real passability checked too, and stored. It will be used for the high level pathfinding, thus in case of 500x500 tiles, the area array consists of approx. 50x50 clusters = 2500 elements. it can be searched fast enough by A*, taking into account move cost, and later some additional AI influence too. The tile passability is based on calculated maximum unit size (shown in the picture by different coloured squares) depending on 4 different unit capabilities at the moment, by measuring terrain hight and slope data, plus obstacle data updated periodically, plus moving unit data updated frequently. Each units within a group will do its own, low level pathfinding in small tile arrays, using a faster (but more stupid) solution than A*. Now I'm ready with these data structures, plus the A* array systems, plus I have the 2 pathfinder algorithm to be used (should be altered a bit of course). What is next is group order and unit behaviour system to be implemented... I realized that making pseudo codes is useful and fun, but requires a lot of chocolate
|
|
|
Re: What are you working on?
[Re: HeelX]
#392930
01/29/12 18:41
01/29/12 18:41
|
Joined: Dec 2008
Posts: 1,218 Germany
Rackscha
Serious User
|
Serious User
Joined: Dec 2008
Posts: 1,218
Germany
|
@HeeIX: looks very interesting. Can you explain it a bit further, for example where do you need it explicitly, or what are you going to do with it at the end?
MY Website with news of my projects: (for example my current Muliplayer Bomberman, GenesisPrecompiler for LiteC and TileMaster, an easy to use Tile editor) Sparetime-Development
|
|
|
|