Artillery Accuracy

Back before microcomputers were available/affordable, I played a very complex board game from Avalon Hill called Squad Leader, which depicted tactical-level WWII combat, complete with every weapon imaginable. Artillery pieces were common, but one of the most feared was the German "88LL" - an 88mm gun that was lethally accurate at long range (hence the "LL" notation).

It got me to thinking that such a concept could be applied to the equivalent of a Big Bertha. That gun would lob shells into the enemy camp but there was never a way to increase its accuracy. Perhaps PA could include a Big Bertha LL? Either as a separate unit or via weapon upgrades? Or even augmented by a separate targeting unit.

Another thought was HE (high explosive) vs. AP (armor piercing) shells... HE would be building busters and area effect, whereas AP would slice through armor like butter, but more concentrated in its damage radius.

Just food for thought.

Big Birtha was always feared in TA, even though it's accuracy wasn't perfect, it almost always killed it's target within a few shots (partially thanks to it's large AOE damage). I don't see improving it's accuracy as doing anything good. Long range weapons should have at least some kinds of downsides imo.

Long range artillery is very clearly going to be present in some capacity. I would caution the devs to be careful with respect to intelligence gathering, as if intel is too easy to acquire then even artillery that is far weaker than it should be can become game-breakingly effective.

Supreme Commander actually built upon the TA artillery paradigm by splitting long range artillery into two categories- shells, and tactical missiles. I think the missile/shell distinction is an excellent one to maintain, also. Shells are cheap to fire, have amazing range, but poor accuracy. Missiles are costly to fire, and slow to produce, and have less range than a comparably-classed shell weapon, but they have pristine accuracy. Both may come in variants with high splash damage, and longer-range, more precise variants that only hit one target.

On the suggestions thread it has been confirmed there will be no shields, which is fine. In SupCom, shields have become the go-to countermeasure for artillery barrage, but we don't need them necessarily. I would rather PA go the route where hiding the positions of your forces, and relying on movement and limited intel to protect you from artillery.

Missiles, on the other hand, due to their extremely precise and destructive nature, need to have a countermeasure. Missile defense can be reworked significantly compared to how it was done in SupCom, though.

You can hide on the other side of the planet

Unless you use the gravity of the planet to slingshot the shells around the planet and hit people in the back.

The moon visits both sides of the planet, and shells are affected by the same gravity, just in a less stable orbit ;-)

On the intel being too easy to gather point I think the main problem with artillery in supcom was just that: if you built an omni-sensor, you could see most of a reasonably sized map, and shell the bejezzus out of it with artillery. If the omni-sensor wasn't present, the "sheller" would have to build a mini-base closer to the "shell-ee" with radar in it, which would be a prime target for the defender, and prevent turtle-y artillery nonsense =P

Accurate long range artillery without shields would obsolete air units and tactical missiles, so it's not a good choice.

Frankly i'd prefer extreme range atrillery to be removed all together, even if it would mean saying goodbye to the Mavor D=

Seems orbital platforms and aircraft could get the job done a lot more simply yeah

If you are using artillery, your biggest challenge should be acquiring solid intel on where to fire. And this should be an actual, real challenge which requires attention and investment of resources in scouting.

One aspect of this is it should cost something to fire. Not much, but it shouldn't be free. This will make firing blindly into the fog of war something players want to avoid, and require scouting to find things to fire upon. In fact, artillery should function in such a way that you really don't want to shoot at individual targets out in a field, or even small forces. Artillery should be reserved for when the enemy has gathered enough stuff in one place that you are very likely to hit something, and you don't really care what.

The Mavor was actually a really awful piece of artillery. Its cost was RIDICULOUS, as a result of its extreme range and accuracy. Good artillery is cheap, powerful, and inaccurate as hell. This means you can get numerous pieces and use them in concert. If each individual piece is inaccurate, then you can simply get more of them and direct them at the same target area. Acquire actionable firing solution, fire everything into that location, and once that intel is stale or the area is confirmed clear, cease fire and wait for further intel.

Also, I think we should rethink the way players use artillery. Typically in past RTS games, artillery targets units, just like a normal combat unit would. I think this is dreadful for game design. Artillery should target *space*, not a specific target unit. If your artillery is so accurate that targeting a single unit out in a field is a reasonable course of action, then you don't really have artillery anymore, so much as an unusually long range main combat weapon.

Totally agree with you Ledarsi, I tend to only use arty in Supcom for spamming tons of projectiles at big armies or large bases, and ANY being able to do pinpoint accurate damage from within the centre of your base without an easy counter (nukes), is always a bit OP

Supcom 2 did this quite well actually, the arty was very inaccurate and did relatively low splash damage, and was only a real danger in bunches of about 4+

Still, intel gathering I think should be harder overall, in supcom everyone just built an omni as soon as there was something they needed to see, removing a lot of stealth from the game

That is an interesting thought. I look forward to seeing how orbital mechanics end up affecting gameplay like this. It promises to be interesting.

You would need some big arti to get the shells into orbit, just saying. Normal artillery will act more as indirect fire, but firing half around the globe? With some bad luck and fixed shell velocity there might even be the problem, that the shell leaves the orbit, it is very unlikely to have artillery which can actually use the orbital mechanics because it would require perfect controll over the shell velocity, not just the launching angle.

Do I need to quote the tagline?

Incorrect. For any given shell velocity there is a ballistic solution to hit any location on the planet merely by adjusting the firing angle, ignoring air resistance. For the Earth, the minimum speed needed to achieve this is about 8km per second. For a 16km radius planet with a surface gravity of 1g (actually considerably larger than the planet we see in the trailer), this speed is 250 meters per second. If you want 'realism', it would be unrealistic for any artillery piece to not be able to shoot all the way round the world.

Always nice to have a person prove that reality is far cooler than people think.

Me on the other hand, I'm to lazy to use any of my old physics knowledge on something like this. Thank you for using yours.

I like the general idea here, which is they become 'attack ground' only, but I'm concerned it'd introduce a considerable amount of micro in to the game.

So, you're saying if a have shell velocity of 10m/s, and 1g gravity... I can hit anywhere anywhere on a Moon-sized planet... just by varying the angle?

No, sir, that's impossible.


Mathematically, for a given shell velocity and acceleration, there's a locus of points that are possible solutions (ala. that you can hit). Increasing shell velocity increases the size of this locus up until it encompasses the entire planet you're on.

Wait, you even point this out when you say the velocity needed on Earth is 8km/s!

The above post is why you should interpret paragraphs as a whole, not single sentences. Don't people get taught reading comprehension in school these days?

On it's own, or as part of a paragraph, the first underlined sentence is still a complete and utter fallacy.

Swap firing angle and velocity around in that sentence and you'd have a leg to stand on.

Obviously we need to have AI that is capable of making decisions for artillery. On the one hand, it is not really feasible for a large scale game to require player input in order every time any artillery unit is going to fire, or change target. On the other hand, if the AI picking targets is poor, then I would rather not have any AI at all. A single scout out in the middle of nowhere drawing every shot from every long range artillery piece on the continent would be incredibly annoying. Especially since in a large scale game we want to encourage covering lots of land area lightly long before anyone starts building huge megabases.

And then there is the possibility of friendly fire- which should also be a big issue for artillery if you aren't careful.

A simple mechanism to group like artillery which are near one another. And then make the group search the entire area within their range, counting each enemy unit or structure as a weight towards that area. Find the area defined by that artillery's dispersion radius which has the highest target weight, and shoot all the artillery in the group at the center of that circle. Probably should also require a minimum weight as well before they will fire at all, depending on the number of guns and their cost.