Air Stacking and Environmental Effects Questions Acknowledged :)

I can't possibly imagine how air stuff should behave to completely prevent stacking

Now that there is so much AOE anti-air, the problem of air stacking is reversed. It's so easy to lose your entire airforce to a single peregrine/flak turret that you should avoid stacking as much as possible. The problem is there's no easy way to tell your air units to spread out and they all converge anyways when you tell them to all bomb the same thing.

I think when you give bombers an area attack command, they should just spew bombs into the circular target area instead of trying to pick out individual units. You'd have to rework the T1 bomber to drop more bombs for less dmg or something, but this would be more like the WW2 bomber role that they seem to be trying to emulate. That way they could spread over the area instead of all trying to hit the same thing like they do now.

We thought about it. It's a non issue. There's nothing wrong with it. Trying to 'fix' air stacking is a near impossible task and 'fixing' it would make air nearly impossible to balance effectively. Just think about how it would play out if they couldn't stack. Think about how difficult it would make path finding, about how much air units would have to jockey for position and slow down to hit the right spot and make them even more vulnerable. It's one of those things that sounds good on paper but when implemented, it's worse than 'stacking' itself.

It would probably be a great idea to implement.

And then, a few hours later, someone would coin the term "land 2.0".

It's nice to see this coming from you, in the past when stacking/clipping has come up I've tried to remind people that it's not a simple thing, it's not like there is a toggle somewhere deep in every RTS engine that flips between "Stacking" and "Non-Stacking" and it's only been a case of the Devs forgetting about it.

Air Pathfinding, especially when it's simulated to the degree seen in SupCom/PA, is a totally different beast from Land/Naval and has all kinds of foilables not seen elsewhere. While Clipping/Stacking might have issues, it doesn't mean that the alternative is problem free as you've described.

Mike

Thank you for the reply Scathis.
I still disagree as I dislike seeing X bombers overlap and kill something instantly. It was an issue in SupCom even through shields and is replicated with PA's bombers. Maybe there's another solution, like forced formations, which might more easily fix this.

I respect your opinion and I'm sorry for pissing you off. I know you guys will make it work whatever happens. If you think it'd be a nightmare to implement then I believe you.

Oh you didn't piss me off at all.
It's all good.

It's one of the reasons Flak is so powerful too.

Mmmmkay good ^^

We spoke about Flak's wiping out of Air blobs in that balancing thread too among other things like altitude stacking and preventing aircraft slowing down. It might make for an interesting read if you haven't already and really aren't pissed off

Kay I'm done, fo' reals this time.

Goodbye.

Bye.

Love you. O-O

I think air stacking means that volume is fully modeled.

It is something I would like to see. More in regards to friendly projectiles flying through buildings/units...

In what way would it be easier to snipe a Commander if Air units couldn't drop bombs through eachother?
I agree formations helped a lot with area commands, but we still have the issue where Air units overlap on a run of a single target and can kill it almost instantly. If Air Units didn't overlap you would have time to react before your Commander died. At the moment, unless you've surrounded your base with Flak by default, you will not have time to move or build defences.

If air is stacked it also means that they sustain more damage from the peregrine's AoE

I'm not entirely sure what your solution to the problem is so it's hard to argue against something I dont know. You need over 40 bombers to hit kill a moving commander. How long do you think these 40 should take to drop their payload?

I just had a huge Latin history flashback - you just pulled a classic Cato the Elder!
Hearing that stacking is here to stay is quite saddening, though. Nothing bad about sticking to tried-and-tested in general but stacking is such a Pandora's box I would have very much liked it tackled in a bolder fashion.
And out of curiosity: What makes pathfinding with collisions so much more demanding for air than ground units? Due to the lack of obstacles in the air layer you don't even need a graph search algorithm - steering in the general direction is all you need to get from A to B. And collisions could be taken care of by your flowfield.
Oh, and thumbs up for the Uber responsiveness!

Flowfields are a method of pathfinding, not collision detection.

Which I never spoke of - I said collision prevention. You don't have to check for a collision if you always stay outside a unit's bounding sphere. Which should be possible by means of units exerting influence on the local flowfield so others stay clear.

??? They did talk about environmental effects.

So I looked into it more, and you're right, it can be used in that way. Not sure if that's how Uber are doing though. I recall there was a livestream quite a while ago where they discussed the flow fields, but I can't find a link to it to confirm.

However, there are still issues with this:

• I'm skeptical of the ability of flow fields to produce good visual movement behaviour for air units, especially given the next point.
• RAM usage would be a major concern, and possibly prohibitive of a 3D field.
• This doesn't ultimately solve how the air units all attack a target without overlapping; it only solves the move case.

Since the air layer is just another flat surface that units are moving on there should be no need for 3D representation.
You are right about that attack vector thing - that would require a whole new approach instead of the current 'move over target, drop load'-mechanic. Which means work. But in the end you are always exchanging one bag of problems for another. Me being a selfish person I opt for the upfront increase in work to get a new and improved system in place instead of flinging band-aids around like no one's business. Looking at you, armor-system!

As long as it takes for each of them to fly over and hit the Commander without overlapping or dropping bombs through each other. Basically, altitude-stacked bombers which circle the target and come out in a line one bomber thick to bomb it, before rejoining the circle.

When? I posted this three hours into the stream. We got a reply in Twitch chat from benologus, and Jon talked about LOS being a pain (to which I provided an easier alternative in my OP).