Factorio Forums

So, lots of mentions of artillery come up in the search, none I could see have any real in-depth plan except that such a thing should exist. So, here's my plan for implementing it. My basic philosophy is that it should require a lot of player involvement, it shoudn't just be another automatic damage dispenser with a long range, and should require careful planning to use. Factorio artillery should ideally be a bit more like an artillery game like Scorched Earth or Worms.

First, the research to unlock it requires turrets, steel and explosives and is a red-and-green research topic. The basic artillery piece requires steel, gear wheels and electronics, and is a big unit like an assembly machine, 2X2 or 3X3. The basic shell unlocks along with the artillery research, it is made of steel and explosives. Giving a howitzer a fire mission is like choosing the recipe for an assembly machine, you tell it what type of shot to use, a direction, a range, and how many times to repeat the same shot. I propose adding a tool to the map screen to allow the player to measure range and angles between two points to make this process a little easier. When the howitzer is ready to fire, it opens the breach and will accept the ingredients from the player or an inserter. It ONLY accepts the ingredients it needs for its current shot, no preloading of stacks.

Notice I said ingredients, plural. The howitzer accepts a shell and between one and four additional explosives as the propellant charge, this is in line with real-world gun artillery which keeps the propellant and shells separate, unlike tank cannons and mortars where it's all integrated. The number of propellant charges it needs are determined by the range interval, so shots closer to the base are more efficient as well as more accurate. The shell has a realistic calculated time in the air, and its direction and muzzle velocity are subject to a bit of random noise. Therefore, a longer shot will scatter farther off the target and will have a longer delay from shot to hit. Later game research could implement guided shells that will steer themselves a bit to try and hit a location like a beacon or targeting laser, but will have a limit to how far they can deflect off where the gun puts them. I'd also implement some specialty shells, like gas, smoke, illumination flares and incendiary as later research. The gun has a heat level, and the bigger the charge the longer it needs to cool down before it can fire again. This cooldown can be reduced by plumbing a source of cold water into the howitzer directly, thus making firebases with access to water strategic locations. The firing of the guns produces a lot of pollution, so biters will try and take out artillery as a high priority and fortified firebases will be needed. Keeping up a long-term barrage will mean moving a lot of shells and explosives to the front with a long train line that trails out behind you, not the simple power poles used by turret-creeping.

Of course, implementing it in this way requires the player to be at the firebase, not out on the field spotting, so I'd follow up with a higher-level research option for a field radio kit as an armor module. It can be used to issue orders to howitzers (and possibly other things like trains, but let's keep it focused) from a forward area, and is small and low-power enough to be slotted into a basic modular armor with a minimal opportunity cost (maybe a 2X2 item?). It's still not as simple as clicking the ground and firing off a barrage, but it should be about that easy to do with practice, so long as you take the time to figure out the rough distance and angle from your firebase ahead of time. You can of course open the interface window while continuing to run away/drive, so you can make some panicked requests if need be, but using artillery in general should be about careful plans and not improvising.

I've left out a lot of details about how effective shells would be and what kind of splash damage it should do as the primary variables for balancing. My preference is for shells to not do a lot of damage on their own, and to require sustained barrages to clear a nest. It's also possible to make it more complicated. You could replace guessing the range with figuring out the propellant charge and the barrel elevation and trying to make range that way, which would require real-world math skills and practice to pull off on the fly, or make the howitzers directional, only able to fire in an arc in front of them and you need to place them in a certain direction like a train track. Any additional comments or refinements?

This seems really well thought out. In my opinion though, the difficulty should not be getting shells powerful enough to be effective against targets, but being able to aim at those targets. One idea I had was that you could have to have multiple stations around the map for triangulation or biangulation. Rather that clicking somewhere on the map, I think you should have to set how far away from any three stations the shot would go, for triangulation, or in what direction for biangulation. Both of those methods provide a single point at which the artillery can fire at. Alternatively, you could have multiple tiers of artillery and targeting devices, with the lowest tier making you guess distance and angle, and the highest tier being pinpoint accurate and capable of destroying targets from extremely far distances. A continuation of that Idea could be specialized artillery. Maybe having one that was very accurate and long range, but very low damage, or one which fired nuclear shells, but was horribly inaccurate. I think a good way to do this would be to have a system similar to the power armor (although I must confess I've never gotten around to unlocking it) where you can customize how the artillery behaves, and then produce it. you wouldn't be able to change it after the fact though, to prevent people from simply changing one emplacement to suit all their needs. As for targeting, maybe triangulation would be the most accurate, biangulation would have long range, but that range would be dependent on the spacing of what I imagine would be range finding towers. Once you got to endgame, maybe you could put satellites into orbit that would let you aim precisely at super long distances, or just forgo the artillery all together and have a space laser. I think a space laser would fit very well into the world of Factorio.

Honestly, this implementation of Artillery sounds like an RTS style superweapon.

It could be an interesting part of the dynamic in the conflict with Biters once there is a dynamic. And setting up and defending a thing like this from swarms so you can soften up an otherwise unassailable nest could be pretty fun.

While I do like having accuracy inversely proportional to range, I'm not sure I agree with the manual aiming; That's interesting in an artillery game where it's the whole focus but in an RTS alike that's a bit more micromanagement than I'd like.

So yeah, there's a lot of work to be done before it could happen but this could be pretty interesting.

While artillery sounds cool and setting appropriate, I can't help but think that any feature that allows you to sit in your base while destroying biter nests is going to do more harm than good, gameplay-wise. Making it expensive, or difficult to use, or it attracts worms to burrow into your base... the engagements are still entirely on your terms, just more tedious. What challenge is left, late-game?

Necromancing this as I want to ask: is there a mod that removes the pinpoint accuracy of artillery, and makes it more realistic ?

It's too good as it exists. And I can't find one.