Factorio Forums

I've noticed a few critical issues with combat in Factorio which it seems either the developers are missing these or they don't know how to fix them. I'd like to point out the flaws and give broad, open suggestions about how to avoid them.

First I'll start with some goals of the combat system. Not everyone will agree on what the goals are, but I'm keeping it very general to better illustrate the purpose and goal of my post. Combat should enable the player to defeat biters and biter bases with scaling difficulty, granting the player room to adapt their strategy and spend time/resources to upgrade their technology in order to tackle greater threats both by choice and by game progression.

Aside from balancing the numbers on enemies and weapons, there are some areas in which the way combat is setup simply drains away the ability for the player to respond in that situation. The player is then forced to avoid the situation entirely, which shoehorns gameplay into a handful of viable tactics rather than allowing players to become inventive with their strategies.


1.) Turret Creep
I think turret creep is a viable tactic and should be allowed to remain in the game, albeit with some balancing (nerfing) work done. But the main problem with turret creep is that the way biter bases are set up makes it next to impossible to assault them in the early game using any other fashion than turret creep. For most of the game, turret creep is THE way to assault bases, then suddenly you get the tank and assaulting bases is easy. This issue is rooted in multiple other issues...

2.) Biter Base Waves
Biter bases spawn waves of enemies very quickly. If you are powerful enough to take out the whole wave before the next wave spawns, then it's easy to assault the base (as long as the turret worms aren't an issue). But if you are slightly weaker and need some extra time to take out the wave, you cannot assault the base without plopping down turrets. Also, biters tend to die pretty fast but move fast and deal damage fast. This is great for roaming biters, but a better solution for biter base defense might be to give them slower-moving enemies that have a lot more hit points and won't chase the player very far before returning to the base--but have respawn waves occur far less often. This way a weaker player can gradually weaken the base until they have removed enough of the defenders to assault it. There's still a time component, but the respawn timer might be more on the level of minutes rather than seconds. It could be fun to try to pick apart an enemy base over the course of several minutes.

3.) Turret Worms
Biter base turrets deal incredibly high damage, making them effective as base defense. Problem is, they're a little too effective. Before powered armor, they tend to blast you to bits in mere seconds. So assaulting a base covered in worms is next to impossible in the early game without using the turret creep method. But once you get a tank, you can snipe even the largest worms from outside their firing range. Thus in one fell swoop, the defensive power of the worms goes from overwhelmingly strong to basically zero. A way around this might be to simply lower the worm damage in order to make it viable to use assault options that don't out-range the worms--because as long as the only way to beat the worms is to outrange them, the player is shoehorned into using the few things that have that range (or high-end powered armor). Another option could be to give them sort of a "falloff" range: their effective range could be the same as current, but their maximum range could be extended much further. Tanks could attack them from outside their effective range, thus the turret worms' shots are less likely to hit the tank and also travel slow enough you might just move out of the way. A third option could be to make the largest turret worms more effective at hitting larger targets like the tank--perhaps they have a lower fire rate with a more powerful shot that is best at piercing armor, and the player on foot can simply weave side to side in order to avoid them.

4.) Spawners Have Low HP
Once you've got the base defenses down, it's a trivial task to take out the spawners. You've got to be strong enough to handle everything the base can throw at you within the span of about 5-10 seconds or you can't really assault it. But once you climb over that ledge, suddenly it's super easy. It would be nice if biter and spitter spawners had much more hit points and possibly even a bit of armor, perhaps their armor would even increase as evolution factor increases. That way when you're whittling the base down, you have a time factor in which you've got to struggle to take down at least one or two spawners before the next wave spawns or else you might have to evacuate and go back to creeping up slowly.

All of these suggestions give the player time to move, think, react, and strategize. It gives time for player tactics to work so that situational response is viable, and the player isn't forced to find silly exploits or research top level technology just to assault biter bases. It breaks the player's strength progression down from a few giant leaps into a bunch of smaller steps.




edit: here's a partially-related suggestion which could further smooth out biter assault progression:
Add some kind of hive structure to some bases which occasionally creates new spawners near it. The hive structure could occur in any base that gets particularly large, and could help the base to expand faster while making it less likely for new bases to spawn in its vicinity. So this would make the size of biter bases vary significantly, with some bases becoming enormous, but total base coverage wouldn't change much. The hive structure ought to be very durable but otherwise defenseless. A player may assault the base and take out the spawners but leave the hive structure behind simply because it is too durable, but if they don't come back soon enough the hive will rebuild the base.

A lower power player can assault the weaker bases to clear some room for expanding, but in order to eradicate the largest bases, the player would need considerable technological advancement and/or would have to put a lot of work into the effort. There should also be some reward for taking out the hives, to create a late-game incentive for destroying biter bases other than simply eliminating threats to the base.

You might want to read this viewtopic.php?f=6&t=3394

At the start of the game you dont have much, just your handguns and like 2 defender bots which are a joke. In comparison in later game you get personal shields and like 130? destroyer bots. They must be doing hundreds of times more damage.
Whats there ,if you are playing with any setting that requires you to destroy alien nests early. Right turret creep. You got like everything high end already. Turrets with piercing ammo, laser turrets. Just the personal robots missing that make this way faster. But its the only way.

Unless you play a default game and dont need much resources until late game. But whats a value of a default game, you dont need trrains in it.

But why the discussion, the combat is getting rebalanced with 0.15 and this is all becoming nonsense.

Ragnaman wrote:You might want to read this viewtopic.php?f=6&t=3394

I have read it and I am following it. Why is this relevant here?

aober93 wrote:But why the discussion, the combat is getting rebalanced with 0.15 and this is all becoming nonsense.

It is getting rebalanced, not fixed. Here's a direct quote from Twinsen in FFF #169: "As you might have read in Friday Facts #166, we wanted to do some combat balancing. First, to not bring the hopes of everyone up too much, this did not mean a combat overhaul. It means mostly tweaking numbers to make the game more fun and make some of the the weapons more viable. No new entities or new mechanics."

The purpose of this discussion is to provide ideas for a future combat overhaul.

Game lacks a really expensive, really effective weapon of mass destruction.

We already have a rocket launch pad. Why not put a nuke in it?

Target an area. Boom! Biters be there no more (but pollution aplenty so expect their friends to come roaming back.)

Well, this suggestion could be fit into a single sentence:
Low down the risks.
Which I agree with. Currently everything can kill almost everything else in a fraction of a second and with a combat that short there's not much of a thought to invest in it.

It like those old rts, in CnC and RedAlert, a lifetime of a unit in combat was measured in fractions of a second and you couldn't actually do anything fancy but to send a horde of them and hope for the best. Starcraft did add a slicker interface, so that a person with a reaction of a jet pilot could make most out of that time. But it was Warcraft3 and its nerf guns and nerf swords that allowed mere mortals to apprehend all aspects of the combat.

Factorio currently feels like starcraft, except you're playing as a lone marine. So with you being that fragile the cost of mistake is sky-high thus you're forced to various cheesy tactics in order to progress.

So, solution might be to slow down the process of everything dying. If bigger biters were guaranteed to have a nibble or two on the turret actually thought out defense setups might not be useless. And with player not being accosted with infinite stream of deadly shrimps inside combat range of a spawner, he might actually consider giving it a spin.

Of course, another alternative could be to lower the combat risks by not forcing the player into it, but automated warfare is already discussed elsewhere.

^
This exactly.

I feel that it should be difficult to encroach upon a native base (especially with lower tech), but I also feel that making the player die really quickly isn't the way to do it.

It's nice that the biters and spitters die really quickly, because then a decent defense setup means your base often takes zero damage (especially at lower evolution factors), which is great because then you don't have to constantly run back and forth repairing everything just because you don't have repair bots setup yet.

But attacking a biter colony should take longer. That's why I suggested having a different kind of unit to defend their nests than what they use to attack the player's base.

Yeah the combat scaling definitely isn't there and the interface is needlessly clunky in places. I'm not against turret-pushing as a form of offense, but the fact that turret-pushing is mainly a challenge of interface management rather than one of tactics/strategy is the main problem there.

(Quick background: I've worked on games like AOE2: The Conqueror's, Rise of Nations, and AOE3: Asian Dynasties and I'm currently a designer on another more recent strategy game.)

Having finally beaten the game for the first time, I felt like difficulty didn't scale very interestingly and there'd be a lot of upside for me if these things were improved. Mostly I wanted to feel like I was constantly striking a balance between developing my production and developing my defenses, yet mostly it was just about combating production issues where one resource or another fell short of the demand and I spent my time addressing that non-threatening problem.

Basically the way I see it:

1. Tower Management Improvements. Using turrets in combat is too clunky currently. The solution I'd recommend would be:

a. When you place a Gun Turret up to 25 of your lowest-tier ammo automatically gets placed in it.
b. Gun Turrets begin in a vulnerable 'deploying' state where they can't fire and take double damage. Aliens will switch to attack deploying turrets if they're very close. While your character is nearby, one Gun Turret at a time will be activated every 3 seconds. Otherwise they take 10 seconds on their own.
c. (Optional) Stack-averaging. It doesn't really add much to the game that damaged turrets don't stack. Instead, they should stack but damage should be averaged across the stack. Meaning if I have one 100%-hp turret and pick up a 50%-hp turret I'll have a two stack where both have 75% hp.

2. Enemies. The current spawn system doesn't really create a good attack rhythm.

a. Predictable Attacks. At dusk, two icons appear for a few seconds displaying the state of the planet's two moons. One has a 4-day cycle, the other has a 5-day cycle. When either moon is full, a medium-sized attack comes that night. When both are full at once, a large attack comes. The attacks are tuned to the player's progression. These should be some of the bigger attacks the player faces, creating a predictable event the player needs to plan around a little.
b. Less Predictable Attacks. Additionally, random pollution-based attacks should continue to happen, but they should be visible ahead of time if you have RADAR coverage so you can see them building up and possibly augment your defenses there before the attack launches.
c. New Turrets. New turrets should be vastly superior to older ones. I was able to use Gun Turrets exclusively in my latest playthrough and that's pretty dull.
d. Better Difficulty Ramp. One elegant way to implement this (since attacks are already driven by pollution) would be for all three types of rocket parts (control unit, fuel, low density) to require Assembling Machine 3 and change AM3 to produce dramatically more pollution than lesser AMs (and therefore dramatically ramp the power of the attacks you're hit by).
e. Fix all-or-nothing nature of attacks? This seems in part my own fault due to my own playstyle of making a single strong 'fence' of turrets with very little on the fleshy center of my base, and so, "No, learn to play scrub" is possibly a legit response to this one, but it feels like attacks shouldn't necessarily run rampant through your entire base if they slip through. It might be that any given alien attack has a max amount of HP damage it'll do before the aliens give up.

3. Skill shots
I'd also love an even larger combat redesign to be skill-shot focused. Basically replace the auto-lock-on system with something more like Nuclear Throne (top-down shooter like Gauntlet, Smash TV, Robotron, etc), where your cursor is where you're shooting towards when you hit spacebar and bullet travel times aren't instant (except for laser weapons). Auto-lock combat is going to be really tough to make compelling. A couple weapons would use the old auto-lock-on system (one would be like the M56A2 Smart Gun used in the movie Aliens, which auto-aims at enemy targets), but these would be balanced to only be valuable for low-skill players (essentially if you're bad at aiming these weapons will put out decent DPS for you, but if you're good at aiming you're going to get a lot better performance out of the skill-shot based guns, which would be most of the other weapons.)