Land Units Shooting Air, Realistic Interaction but Rebalanced.

I'm sorry, I went away for the weekend. Internet was on my phone, and firefoxes android browser is next to useless for posting on these forums. But I had a big think and so on.

The point is that in this situation, you don't have dedicated anti-air, and your opponent is taking full advantage of that fact, and punishing *you*.

There are 5 battle-spaces in the game - land, naval, air, orbital and intel. Your opponent is fighting in the land battlespace, and he is also mindful of the air battlespace. You are fighting in the land battlespace and have completely forgotten about the air battlespace.

This mechanic can be construed to reward a player who is just spamming one unit or one unit type. The army was built without anti-air vertebrae in the spine.

I like how this post firmly addresses both points of the argument in a balanced way.

In reality, an M1 Abrams tank can not raise its barrel high enough to engage anything but distant, low flying aircraft. The 50 cal machine gun is effective against low flying aircraft. A helicopter can engage without line of sight. A fixed-wing aircraft can engage from high altitude.

My first issue is that the argument is flawed. The interaction would be unrealistic.

But that's a silly nitpicking point.

So in terms of actual gameplay: the issue is that this game is about macro. If your army only consists of 1 unit, your army has a weakness. Your army is designed to be hard countered. If your army had consisted of 1 or 2 more AA, and 1 or 2 less ground units, this wouldn't be an issue, your army would have had to be countered with a range of units. As it is, you built your army around a weakness, your opponent exploited that weakness. That's my issue with it from a game-play perspective. Your army make up discourages soft countering, in favour of hard counters.


Secondly, for me the "soft-unit interaction" comes from unit interaction as a group, not unit interaction as singular entities. I don't see the t1 units as being "generalists". I see them as being the backbone of the army. Is that purely terminology? No - 1 ant can engage 1 soft target well. 1 spinner can engage 1 aerial target well. You need to throw multiple ants at slammers, you need to throw multiple spinners at bombers and even more at the t2 aircraft. Personally I always throw a couple of AA into my armies, and the AI always has a terrible day when it attacks me with bombers. Likewise, when I get attacked by bombers, my groups without AA are first to be destroyed. My ants form anti-land vertebrae, my spinners anti-air vertebrae, and I have a backbone I can add levellers, shellers, doxen, slammers or stompers to.

An army with 1 leveller and 7 ants takes more time to kill than 2 levellers. Both armies are vulnerable against air. An army with 1 leveller, 6 ants and 1 spinner costs the same as 2 levellers, takes 2 levellers the same amount of time to kill and has defence against bombers.

An army with 13 ants and 1 spinner, vs 14 ants, performs very well - because the extra fire power the second army has is counteracted by the fact that two seconds worth of damage are useless.

Sure, 13 ants still loses. But why would you engage in the first place? They forgot about air units, so just drop bombs on them. Send your army in to mop up.



Considering the other direction of maths (warning - bad probability is bad, but as I recall probability curves this is a near enough guestimate), and using real dps - A leveller does 500 damager per shot, 1 shot per second. It doesn't kill another leveller in 1.25 seconds, but in 2 seconds.

If 1 ant has a 1/50 chance of hitting a target, its average dps over a long period of time will be: 0.84.

In a mob of 50 ants, the average dps would be 42, and the bomber gets killed in 2 seconds. That's going from an infinite amount of time above the enemy to a limited amount, for no other reason than not building the anti-air backbone of your army.

If the probability is 1/1000, the average dps for the mob is 2.1, it takes 38 seconds to kill the bomber. But that's still "unfair" to the other player - after all, you don't have dedicated antiair. You could have built a fighter from scratch and flown it 1358 distance units in that time. That includes a 5 second factory roll off time. It's still just a substitution for your poor army management.

In the previous example given on an earlier page, 50 ants are exterminated by 1 bomber. Those losses would have been entirely preventable with just 1 spinner in the mob, or a fighter flying overhead. Spinner is preferable because it acts as extra health for the blob, whereas the hummingbird does not.


Personally I don't feel that spinners should be able to engage land, in keeping with ants not being to engage air. Your army should consist of both, and to leave one backbone out is folly.



I'm not, you understand, against the idea of a generalist unit. But the anti-land tank and the anti-air tank should be two units distinct from the generalist unit.

Considering Rise of Nations - Rise of Nations had units that were hard counters against others, and then it had soft counters as well. Everything could kill everything. Light infantry kills archers fast. Cavalry kills light infantry and archers, but in sufficient numbers light infantry and archers will overwhelm Cavalry. Considering Earth 2150, chainguns and rockets attacked air units and attacked land units as well. (Mind there was no dedicated anti-air in that game).

So placing the 4th t1 combat unit, the generalist, does, as you rightly identify, make a dedicated anti-land task force more specialised towards their role, and vice versa.

In Earth 2150, chainguns were the supreme anti-air weapon until you had homing rocket ammunition. Then homing rockets were awesome against both air and land. But "fighters" and "bombers" could still engage anti-air units and their defending units (in fact there were some campaign missions I solely used the air force for)

I'm not convinced it would work from a balance perspective. With the way PA treats unit engagements, the emphasis on producing large volumes of units, and the emphasis on low unit survivability, it seems like it would create a no-fly zone for enemy aircraft as effectively than spinners do, unless you lowered the damage of the AA weapon to be much lower than dedicated aa units. It certainly doesn't give you the chance to make hit and run attacks. Because of the way unit focus firing works, a group of 20 - 40 hybrid tanks raiding outlying mexes would require a relatively large investment in ants/doxes to remove.

First I should say that you shouldn't use probabilities when talking so generally about ground hitting air. You cant just say its a 1/50 chance or a 1/1000 chance. It depends on how they are used. If the bombers fly over, drop there bombs and leave, you should have no chance of hitting them. However if the bombers slow down and pull a really tight/slow turn in range of the ground forces then your tanks should have a reasonable chance of hitting them.

Your (and many others) problem is that you simplify things too much. You go to extremes and end up with a binary result.

Your problem is that you think too much about big ideas, and don't think about the details. 1 in 50 was pulled from earlier in the thread, and it's still clearly too high. As a thought experiment, 1 in 1000 is used, and it still gives a value which is disadvantageous to the player with the strong air force. I haven't gone to any extremes yet. I've considered reasonable values. My question for you is "What is a reasonable chance?"
(Is 1 in 1000 reasonable? Perhaps not, but considering that 1 in 50 gave a time equally as fast as 1 spinner acting alone, I believed it was)


PA uses physics based projectiles, correct? It's the intersection of two 3D straight line vectors. There are a finite number of ways that they can fail to intersect, and a finite number of ways that they can intersect. All you did is change the probability from being general (1 in 50) to being conditional probability (1 in 50, if such and such a condition is true).

And what's the condition? If bombers are in range of a target.

Is inaccuracy decreed by a random number God, which is probability at its finest? Or is it decreed by the interaction of two vectors, in which case observant players will always be able to avoid the shot? Or some combination thereof? Either way we are talking about a boolean result - vector intersection is true, or it is false. There are a finite number of successful shots, a finite number of failed shots. Simplifying the situation so it is explainable by a probability model is perfectly valid. It's also the only way that the programmers can get it into the game.

Once again, if air units are slowing down in friendly air space, they have every reason to do so. You lost air superiority, you're losing the air war. Build fighters and spinners.

Not sure if you realise this or not, but if tanks catch bombers/fighters landed, they tear through them. Bombers can't fight back in that situation.

There is more to it then just inaccurate weapons. Its about turret turn/pitch speed. So if an aircraft flew over at high speeds there would be no chance of the tanks shooting it. However if an aircraft slowed down or stopped near the tanks, the tanks would have enough time to turn and aim at them, and then shoot at them.

Your problem is that your skipping a load of balancing features from your estimates.

Did you not consider that the cases where the tank does not fire a projectile to hit a valid target is accounted for in the "did not hit" part of the probability? Accuracy is just a measure of whether or not a target was hit. For all practical purposes, there is no difference between a shot that was never fired, and a shot that was fired and missed it's intended target.

You are missing the point. This has nothing to do with shooting and missing/not shooting at all. Using probability for the whole ground vs. air is flawed. It varies greatly depending on type of unit, its speed and what its doing.

What I am saying is that for fast flyovers and quick attack, tanks won't be able to line up there turrets. They won't shoot.

Now lets look at a slow bomber doing a tight turn at around the max range of the tanks. The tanks don't have to turn there turrets much and they have plenty of time. They get to shoot the bomber because it made the silly choice of slowing down/stopping near the enemy.

I thought more. I took a couple of days just to make sure reply was sensible and good.

Probability gives you an expected dps. That's literally the only reason I used it. To compare a weapon with much much smaller dps than the main weapon. Rather than saying "The tank has an anti-air chain gun which does 0.42 dps, and a laser cannon which does 42dps"

The problem with that system is if the player is paying attention, he'll just make the bombers attack from multiple places, or otherwise not at 12 o'clock. And then it comes back to a binary hit/miss system again.

And you know. Oft repeated reason that you don't have AA so bomber has every right to be there. But also the fact that the bomber default attack behaviour slows down and stops near the enemy, so it kind of guarantees the bomber to being a very expensive throw away unit in every situation, even when the enemy does not have AA...

I'm beginning to come around to your way of thinking though. I reinstalled E2150 and had a play.

Rockets are high damage but miss. Chain guns are low damage, but hit lots. Both weapons are ok, but not great, against armoured targets. Land weapons (such as cannons and grenades) do high damage against armour, but can't target air. It works well and does feel realistic. In that game, chainguns are basically exclusively anti-air specialists, rockets start to replace them later in the game but are more generalist units, cannons and grenades are anti-land.

I feel like an important supplement would be the concept of armour for different units (So that certain weapons did less damage against some units) I guess I am generally supportive of the idea as an all encompassing principal, but I do feel that it's not really the place of a dedicated anti-land unit, such as the Ant, and I do strongly feel that it's clashing with a unit that already exists to the point where players aren't focusing on creating mixed armies of units.

Someone brought up the idea of a tank-hunter, which is kind of where the Ant is already. So the Generalist would need to be separate (and it doesn't make sense to raise the barrel to engage air, it makes more sense to have dumb fire rocket pods or epic chainguns)

The problem I foresee is that in most RTS games I can think of, that have "soft-unit-interaction", you find two units. One has the most dps, the other has the best health to cost ratio. And you just spam both of those together because it's "good enough". Especially with ranged units. The idea of kiting makes many ranged units absolutely supreme in every situation.

Look at PA. You've got perfectly good bot and vehicle combination, and everyone sticks with vehicles because the added mobility that bots have does not make up for their lack of range unless the tank target leading is bugging out.

I guess a generalist "dumbfire rocket" unit would be ok if it took 3-4 shots to kill a T1 bomber/ant and missed the bomber lots, because then it wouldn't outclass either the Spinner or Ant. And looking at T1 Naval I see the hole - but I think that in this case, there should be 3 units (anti-land, anti-air, generalist) in T1. And then T2 can get flak or a multimissile launcher anti-air unit, with the ability to hit multiple units at once or concentrate all on one target.

Despite that, I honestly do feel that the best solution is just to leave it out, and force players to consider that there are multiple battle-spaces that they are fighting in, and equip themselves appropriately.