Yes, but if you were that unit, would you get stuck in a corner, or would you persist in trying to find your destination? I suppose if you make it all the way to the target, you've accomplished pathfinding of some kind, by definition, sure. But an algorithm to avoid getting stuck doesn't have to be "pathfinding" as in an expensive algorithm to find a complete path right away.
replyWell, depending on how you implement it, there might not be any way of getting unstuck, imagine a wall that is 3/4 circle for example, with a small opening, could it escape with that sort of naive algorithm? Players tend to be kind of sensitive to "stupid unit movements", and not having pathfinding is a great way of triggering that feeling in every play session.
replyThere certainly are naive ways, although whatever I suggest here will be poorly thought out and will lead to the next problem. But OK, I suggest a scent trail. Each unit paints where it's been and prefers not to go that way again. Then of course it could paint itself into a corner, so it needs a local-maximum-busting strategy too, which probably means marching off in an arbitrary direction to see if it gets unstuck.
replyThis would be behaving like a caged animal, and in a game, that's good, and better than a smarter algorithm. You don't want them to be idiots, but you don't want them to be magic, either.
The old "follow the left-hand wall" maze-solving strategy is another naive way to get out of a trap. It's not fun gameplay, but it's naive and it exists, so better naive strategies do too.
That’s pathfinding. I thought you were opposed to pathfinding?
replyWell, ish. My theory is that you can have units make individual local choices that add up to somewhat inept pathfinding, and it works out cheaper than maintaining live-updated whole-map pathfinding information.
replyNot only is that pathfinding, it sounds like an especially complex pathfinding algorithm that would be pretty challenging to implement efficiently.
replyIn what way?
replyWell, if each unit leaves its own scent trail, that’s a lot of per-unit state, and little to nothing that you can pre-compute for the entire map. You could have all units trails on a global “scent layer” that all units read from, but then you’re basically just building up a graph of common paths that could have been precomputed for the entire map.
replyIt also doesn’t at all address inter-unit collisions, which is a big topic in RTS pathfinding.
When you get stuck, try random directions. That's what my robot vacuum does.
replyNot sure if you've played any RTS lately, but I'm fairly confident people would drop the game relatively fast if they started seeing their units taking off in random directions because they cannot find their way around a wall. But we could also be playing very different games :)
reply