yeah, that’s looking better it still pushes through enemies. ok they should not make way, but gone around?
Local avoidance is still local unfortunately. They will not be able to plan paths around enemies.
how about giving enemy units an obstacle layer or treat them like colliders / map obstacles?
It is possible, but usually it’s very unstable because they are able to fully block some passages that shouldn’t be blocked due to the grid’s coarseness.
There is a simpler kind of local wall hugging behavior in this version. You can see it if a large group of units are attacking (there is a large group of units in a disabled gameobject in the scene that you can use). That might work. However it has some downsides such as that it will not be able to detect that there is a wall in the way, so it may not find a good path around the enemies.
I still didnt managed to run your 3D example in my 2D project. Is there hope to see a RTS2D example in the future? If you were to release a 2D RTS, what would you improve to act like every 2D RTS (getnearest, dont push) like: https://youtu.be/gXAp2hk7mdk?t=12m48s
Nice update! Do you plan to bring your rts example to next release?
Maybe, we will see how it goes.
Would love it! A shame it doesnt work anymore with the new version:
Assets/AstarPathfindingProject/ExampleScenes/Example18_RTS/SC2Avoidance.cs(65,26): error CS1061: Type
Pathfinding.RVO.RVOQuadtree' does not contain a definition forFindAllTouchingCircle’ and no extension method
FindAllTouchingCircle' of typePathfinding.RVO.RVOQuadtree’ could be found. Are you missing an assembly reference?
What is the reason that loading the scene takes sooo long?
Yeah there are some other changes that the branch had to do. So those scripts do not work out of the box with the latest version.
It takes a lot of time because it is recalculating a recast graph. You can create a cache to make it load a lot quicker. See https://www.arongranberg.com/astar/docs/save-load-graphs.php
would you consider to implement an Example17_RTS? I loved your Beta tests
That’s still the plan yeah. So much to do, so little time