right now when a 'target that' happens it doesn't let go until that target is destroyed ... so a unit will run past better targets to get to the one he first targetted, like running along a wall to a door, and then there's a great building to blow up right next to the door .. but he runs back along the wall to the original target.
i'm sure there are standard algorithms for improving this; check for retarget regularly, or better randomly?
i can see a performance hit, but this has to be a standard problem in these types of games.
i'm sure there are standard algorithms for improving this; check for retarget regularly, or better randomly?
i can see a performance hit, but this has to be a standard problem in these types of games.
Last edited: