Sign in to follow this  
Scalytank

Pathfinding, and pixelmovement

Recommended Posts

Hello everyone!

I am Scalytank, a Hungarian developer from Serbia. English is not my native language, but i will try my best ^^
I currently develop a game with ika (uses python as script language) I have a working tile based engine, and a pathfinder. However, i want to switch my game, to pixel based movement, And pathfinding.

Now i have this issue.. My game will utilize an action battle, with usually 3-4 enemies at once. I need some theory about useing A* node based pathfinding, with pixelmovement.

My game will consist of 40*40 pixel tiles. Enemies will have 32*32 size, as boundary boxes ,and no bigger. (at least there will be a 100*100 big object, but that wont utilize pathfinding.)

I dont know really where to start. I have this logic, that the enemy will check for the tile where the player is, and that tile will be the middle of the players boundary box. Then he will follow a route to the player, and when he hits a wall, or something, he moves to the center of his tile, and continue his route. (So for example, he is off centered from his tile, because the movement is not gridlocked, and this way ,he can reach other tiles obstruction) And the pathfinder will mark every tile as obstructed, if any of the objects 4 boundary points are on that tile.

If there are someone who made this before, I would appreciate any help given. Thanks for your time ^^

Share this post


Link to post
Share on other sites
Pixel pathfinding is slow, just break everything up into blocks. 4X4 sounds good, then use a grid pathfinding algorithm. The movements can be smothed using a curve fitting algorithm.
[url="http://www.policyalmanac.org/games/aStarTutorial.htm"]Link to translated tutorial[/url] Edited by kd7tck

Share this post


Link to post
Share on other sites
[quote name='kd7tck' timestamp='1348797904' post='4984561']
Pixel pathfinding is slow, just break everything up into blocks. 4X4 sounds good, then use a grid pathfinding algorithm. The movements can be smothed using a curve fitting algorithm.
[url="http://www.policyalmanac.org/games/aStarTutorial.htm"]Link to translated tutorial[/url]
[/quote]
I'd also like to toss in my 0.02, I also agree that using a 1x1 grid for A* is going to be slow(although depending on the size of the level, it might be possible to do it without much performance impact).
Considering your agent's are most likely going to be larger than a single grid unit, i'd recommend giving [url="http://harablog.wordpress.com/2009/01/29/clearance-based-pathfinding/"]this[/url] a read, [b]after[/b] you've given kd7tck's link a read over. Edited by slicer4ever

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

Sign in to follow this