Friday Facts #166 - Combat Revisit

[Grimakar]

I would not nerf turret creep. Esp. as long as it is the only way to get rid of the biters. And making a cooldown or delay before they can attack would just make attacking the nests more die-hard, what it already is. But the more research the more twinkle-toed and fun the fights should become, esp. as this is a game about automation and not dark souls or whatever.

[MeduSalem]

I wonder how the infinite research will influence combat.

Because even if all weapons are somewhat balanced out and viable... the Biters won't scale anymore after a certain point because they reached maximum evolution... but the player can go well beyond that with additional research levels.

At some point the combat will be as trivial as it is currently.

Honestly I think that the Biters need a fundamental rework in diversity (different combinations of resistances, etc) to counter the power creep of the player. In higher biter evolutions the diversity of Biters should grow... forcing the player to become more strategic instead of just pushing damage output.

[Grimakar]

I wonder how the infinite research will influence combat.

Because even if all weapons are somewhat balanced out and viable... the Biters won't scale anymore after a certain point because they reached maximum evolution... but the player can go well beyond that with additional research levels.

At some point the combat will be as trivial as it is currently. Even more so if there are multiple viable weapons.

If you spend myriads of ressources to pimp you character, you can except to be a god against those biters, esp in a game about automization.

[MeduSalem]

If you spend myriads of ressources to pimp you character, you can except to be a god against those biters, esp in a game about automization.

Yeah, which is a reason why they should somehow scale with the player to keep up with the progress and to keep it interesting. But that is easier said than done.

[Xerophyte]

I don't mind the biters as an occasional thing to deal with, but I do think that as the game progresses from crafting/survival to logistical challenges and factories the solutions you use to deal with biters should progress likewise. Clearing a biter field should also be similar to setting up a mining outpost in that once it's cleared it stays cleared. The methods you use to deal with the biters should progress towards more automation. You should also be able to eventually eclipse the biters as a force, as that's a natural final act of the whole "become the master of this hostile domain" vibe the game goes for.

As such, specific suggestions:
1) Turn off biter base respawning, or create a building that turns off base respawning in an area. Some blue science tech that uses oil and sulfur to diffuse a biter annoyance cloud, for instance, but attracts more attacks than pollution. Alternatively, just make pollution prevent new base growth and give us heavy pollution belchers.
2) Make the fixed defenses less requiring of your attention once set up, similar to miners. You could have some sort of early-mid game coal-fired military logistic bots that run on fixed tracks and can only rearm and repair early defenses, or add some sort of rearm and repair turret.
3) Late game there should be offensive options that revolve more around the factory building aspect of the game. Giant TA-style Big Bertha cannons or fixed rocket launchers that can fire at bases in radar range, but require decent amounts of resources to fuel and arm. Possibly add an artillery train car for something more mobile. Small yield nuclear artillery shells seem appropriately Factoriolike to me, once nuclear power is in.
4) Make late game biter attacks more wave-like and predictable. Have them build up forces for a bit, then coordinate to attack at a time that the player can be warned about well in advance. This may not be viable for performance reasons, but just knowing when they're coming feels like it would go a long way towards limiting the problem of your train of thought being constantly interrupted by attacks at the other end of the base.

[R3vo]

Many folks here are only talking about the weapons. Combat Revesit for me means to also improve the situational awareness of what's happening around the base.
Right now we have that small icon at the bottom of our screen which either says that turrets are in combat, or objects are destroyed. Those are sometimes mixed up with messages about contruction. I see quite some potential for improvement here, e.g., how many dmg was recieved, how many biters killed etc.

Some other ideas:

- Give us players a way to reduce biter evolution slowly -> there seems to be now way to prevent behemots from being created in a long play
- Add the evo gui mod to the game, so new players get an idea of what's going on and what's causing the biters to evolve
- A building which attracts biters, so we can distract them from belts and train tracks
- A way to automatically defend an area. Why can't we put destroyer drones into drone hubs?
- Remove that stupid explosion sound from combat drones, or, remove the lifetime at all, but reduce their HP so they are easier destroyed by biters
- The tank should be more powerful than any of the personal defense weapons, and it should be upgradeable, we have the means of research, yet there isn't a way to improve armour, speed or research shields.
- The tank should be moved from "logistics" to "combat"
- Make mines more effective, more damaged, replaceable by construction robots so we are able to maintain a minefield automatically -> automation
- Spawn time of spiters per spawner should be reduced. They spawn almost as fast as one can kill them
- Make biter attacks less random, let them communicate and search for common targets. Otherwise we will always need to wall of our whole factory or place turrets everywhere
- Give trains a way to defend, maybe a wagon with a huge gun on top?
- Trains get stuck if there are alot of biter on a track (behemots), even at full speed. That should not happen.
- A research to increase armour and hp of walls!


That's all for now, I hope my feedback will be helpful. Have a nice sunday!

[Rhamphoryncus]

- Trains get stuck if there are alot of biter on a track (behemots), even at full speed. That should not happen.

Spinning blades o' doom we can place on the front?

[Andrzejef]

- Trains get stuck if there are alot of biter on a track (behemots), even at full speed. That should not happen.

Spinning blades o' doom we can place on the front?

GIMME!!!!!!!1
Actually, that's a good idea for train equipment grid, like some dedicated mod

Also, this:

Honestly, I really wish there was more combat automation. I want giant armies of robots crashing against swarms of Biters. I want a war machine. Going out and doing it myself seems counter to the whole point of the game. My two cents.

And also this:

A propos dune . I would love to see some sort of uber enemy like sandworms, that don't attack you unless you come too close, won't attack your base unless provoked, but are practically impossible to destroy. Maybe having to research a 'repellent' device to protect new outposts.

That is an awesome idea. And they only can come out of "normal" earth. So if you build concrete etc you are safe from attacks on that ground from sandworms.

EDIT:
Also, I agree with alingis' tiered combat idea.
And I agree with Optera's points in here and here but I think turret creep can be successfully reduced not by "forced standby" of a turret, as you noticed, but - as I proposed - making some period of time necessary to even place the turret (or any other object IMO) - that way you can't place turrets the instant you know you have troubles.
Such set-up time would be very marginal - if any - "issue" when in calm confines of your factory, but will render turrets as not really desired solution when going in hot into "critical biter situation".

[tehroach]

I am very glad that you have decided to pay some much-needed attention to the combat in Factorio
While I agree that the default Factorio Combat can be lack-luster at times, it is definitely fun at times and given the right modifications can remain fun throughout the game.

Combat in Factorio (for me) is the twitch/reaction aspect of Factorio gameplay.
However, on Factorio's default settings the cognitive flow as described by Mihaly Csikszentmihalyi is definitely out of wack at points throughout the game in relation to tech level and biter evolution.
While a rebalance of the weaponry will help at several points along the flow and address the tech side, it will never solve the problem in its entirety.
IMO, the elephant in the room on this one is represented by the "global evolution level" and its time driven aspects.

On default settings an expert player will always be able to keep ahead of the curve and combat will unfortunately generally flow through the too easy and boring region, but at the same time a novice player will always be behind and combat will generally lie in the difficult to frustrating region.
This being said I believe that the Factorio player base would almost completely fill every point in the spectrum between novice and expert, hence I can easily see a weapon only rebalance approach, only helping the players that fall into the upper regions of this spectrum and yet may still fail the ones at the top.

I really like Factorio to the point it is basically the only game I play these days and I really enjoy the combat.
However, I believe that one of Factorio's strongest points is its ease of modification as currently, I am playing with some 80+ MODS

One of these MODs is my attempt to address the evolutionary problem.
While this MOD succeeds at making the game easier it ultimately fails to address the problem, as it gives too much power to the player and relies solely on the player to "dress for the occasion" and then to resist the urge to send the biters back to the stone age once they have achieved the upper hand.

Hence I would suggest that a great place to start on improving Factorio Combat, would be to open up the combat related sections of code to the LUA interface with more trigger events etc.
ie an on_biter_spawned event etc. and an easy/fast(UPS) way to index and select one/any/all biters, would really aid my efforts and I imagine others as well
IMO the AI could be taken advantage of in this way also if given something like the recipes/items/entities LUA interface to allow further customization of each biter type, ie the same looking biters could be configured to behave differently to each other, this would add variety, which would aid in the elimination of the "same same" problem many people in this thread have described as the eventual grind of Factorio combat.

Honestly I think that the Biters need a fundamental rework in diversity (different combinations of resistances, etc) to counter the power creep of the player. In higher biter evolutions the diversity of Biters should grow... forcing the player to become more strategic instead of just pushing damage output.

ATM the biters while sometimes fun to kill have no life, their homes are just some random groupings of a few different types of "buildings", the biters themselves are lifeless, they spawn, they run they die (sometimes there is enough of them to get a few bites in), generally they are just sitting around waiting for the player to stray too close or to be triggered by a pollution cloud.

If the biters themselves had a purpose other than to just annoy/kill/destroy the player and their hard work. If they had structure and purpose to their bases and a reason to exist and expand. Maybe if the biters behaved like ants, had a queen to protect, a life-cycle and resources to gather, and love for their kin, they too would be more widely loved

ideas

• Queen - All ant colonies have one, IMO biter bases need them too
  instead of biter bases just being groups of randomly placed biter "buildings" what if there was structure to their layout and the biter bases in an area were part of a larger biter colony, controlled by a queen and inter-connected, maybe some sort of tunnel, maybe biters just travel between them on land.
  The Queen would govern the biter evolution level for her colony and lay lots of eggs.
• Life-Cycle - All biters start as eggs, each colony has a limited population that can only be increased by the queen laying eggs, to lay eggs the queen needs to be feed resources.
  (I imagine these eggs as type of biter building which represents many (tens maybe a hundred) individual eggs)
  After some time the eggs hatch into larva (building upgrades), some more time and the building upgrades again to a Hive
  
  In the egg and larval stages, these buildings would be somewhat immune to small arms and piercing-munitions fire, regularly damaged by explosives and highly vulnerable to fire, because of their malleable and slimy state, while in the hive state fire is almost useless, piercing-munitions do regular damage and explosives are highly effective due to the hardened shell and larger size of the occupants.
  
  Over time the hives will release biters as needed by the colony.
  These fresh biters will be soft and gooey ready to take on any role that the colony requires ie worms, buildings, spitters etc. (but vulnerable to small arms and fire until they form their hardened carapace).
• Gathering resources - ATM the resources that exist on Navius are for the exclusive use of the player, but what if the biters also used them, this would mean that the resources outside of player controlled areas would also be subjected to depletion, giving the player reason to go get them ASAP but also a reason for why the biters constantly keep coming back.
  
  While I strongly agree that it is a good idea to remove Alien Artifacts from the recipes of game finishing items, it is a grave mistake to remove them completely as removing them will also remove the only combat reward currently in Factorio, I believe that the higher tiers of weapons, ammo and armor should still require them and if picking them up is such a chore then maybe just alleviate the chore part, ie holding down the 'f' key would gradually increase the pickup radius or something along those lines.
• Pollution as a Biter resource - Biters would simply be attracted to pollution because it aggravated them, but because they would desire it as a resource. This pollution resource would be used as food/building material like other resources, but with the added side-effect of evolving the queen and thereby increasing the base evolution factor of the colony.
  The pollution mechanics would need to be modified not to simply disappear after a period of time but be perpetual and permeate across the entire map, only "disappearing" once it reaches the edge of the known map, where it would be used to calculate the base evolution level of future discovered queens.
  
  Biters themselves would absorb pollution slowly from the air and after a certain amount of exposure would shed their carapace and evolve to the next level.
  and some biter-bases would exist for the sole purpose of rapidly extracting pollution from the air to be carried back to the colony.
  Trees would absorb pollution until they die like they do now, but dead trees would then become a sort-after resource for the biters
• Love for their kin - ATM biters will crawl over mountains of their dead to complete their one desire to kill the player, this in itself makes combat very mundane and repetitive. What if biters did what ants did, when they find a dead or dying ant they don't just ignore it, they tend to it and carry it back to their nest. If defending their home was more important to the biters, than simply chasing the player till death do us part.
  
  This would help alleviate some of the problems related to "turret creep" where you simply place a line of turrets, then aggro the biters to chase you to their death and swoop it to kill the base before enough biters have spawned, rinse and repeat.

Another aspect of Combat that needs to be addressed is the UI, I think vanatteveldt explains it well

One other thing about combat is the interface.
The double function of the Q key is very annoying, as when you switch weapons you no longer have your grenade/capsule in hand. Also, it is extremely annoying the scroll through multiple weapons in a tense fight.

What I think is needed is:
- multiple ways to switch weapons, e.g. using 1-4 or ZXCV and scroll wheel
- a very clear indicator of which weapon is selected, e.g. different reticle, color, whatever
- a way to have a primary and secondary weapon selected, e.g. firing grenades/rockets with the right mouse button
- maybe a 'combat mode', where logibots stop working and left mouse click shoots the primary weapon (e.g. press a button to draw your weapons, and you can't build stuff with weapons drawn since you don't have your hands free)
- stop having to launch combat drones manually, it's annoying and just distracts. For distractors, it makes sense as these are more of an AoE weapon / decoy, but follower bots should just launch themselves somehow.

An earlier poster suggested using roboports and a stack of drones (that will presumably deploy automatically when enemies are near), I think that makes perfect sense. Allowing combat drones in normal (ground-based) roboports would also be a fun addition, I think.

I definitely agree, giving the player a choice to enter combat mode that is easier to control/micro combat and gives more feedback, would be beneficial to the combat aspect of Factorio, it would also somewhat nerf the "Turret Creep" problem, as the player would have to take the gamble of exiting combat mode to enter build mode to build the turret and not be able to attack while doing so.
(As well as eliminating the rare, but annoying, clumsy mistake, relating to the proximity of the 'c' to 'd' key, shotguns and wooden chests full of items I am about to use)

ideas

As the one aspect of Factorio which doesn't allow the player to fully atomize is combat, I remember an idea of a "Control Center" being discussed a while back, I think around the 0.11.x to 0.12.x era, where the player could ultimately reside and completely build their factory from there.

While I sat mainly on the fence about the Control Center in regards to building the actual factory, I could see an idea similar being a good part-solution for some of the combat aspects in Factorio. Robots like the ones in the Robot Army MOD could be autonomously built, but roll-off the factory in their entity form (ie not be mine-able or appear as an item to prevent them being used as a better form of "turret creep") the player could then jump into the Control Center and move the robots around RTS style anywhere on the map that they had radar coverage or if they had a "mobile control center" equipped in their armor could order the robots while in "Combat Mode" anywhere they had line of sight.

Thankyou for taking the time to read this long winded post

[theBeave]

General Ideas for combat engagement:

The current premise is that biters expand in all directions, including TOWARDS pollution. What if we change, and biters behave more like trees? Effectively moving AWAY from pollution?!

Revamp biters and nests to the following (much of this already generally exists):
- Biter nests behave like trees, that is to say that they absorb pollution and eventually die from absorbing too much.
- Every so many ticks (ie 300 ticks), the nest spawns biter (or biters) on par with the amount of pollution absorbed in the last set of ticks and commensurate with its tier- lots of pollution, lots of biters spawned.
- If nest is destroyed or absorbs critical amount of pollution, present group of biters aggro and run towards base or player, per normal.
- Total pollution absorbed is recorded, once a set amount is reached, the nest dies and is replaced by either a higher tier nest or multiple of the same tier -- but, AWAY from the pollution.
- Evolution should be accounted locally, at each base or each chunk if necessary. Global evolution factor seems a bit off. Problems should arise where caused, not universally.

With this setup, pollution both causes biter evolution/response and biter withdrawal. Eventually, this should come to a balancing point if the base produces a flat rate of pollution. Then eventually the cloud will push the biters back far enough that they are beyond the cloud, but very strong. This has the net effect of creating a meat prison for the player. It will take some effort to break out, but he is not really bothered inside his cell.

Nests would have several gradual tiers, as they have the opportunity to replace themselves from pollution destruction; growing from small biters > spiters > big variants > behemoths. Ideally, nests would somehow 'transition' between the pollution destroyed instance and the new higher tier variant further back, just in case the nest broke down right in front of the player and popped into place a bit further back.

This way, combat is a direct response of irritation - no tickle, no fight.

--

I like the idea of special nests/bases on top of ores. Perhaps instead of normal biters dropping loot all the time, the idea of special alien ore protected by unique nests/biters that do not react the same way to pollution. This would incentivize interaction at the option of the player. Pollution could even somehow heal and strengthen the nest/biters in place. So, the challenge grows with time. But these biters would not necessarily aggro and attack from long distances - more static defenses.

[factoriouzr]

I am soooooooo glad this is finalllllyy being addressed.

Here are the more important things that need to be improved when it comes to combat:

1) More varieties/types of biters with unique abilities, eg different attacks, perhaps flight but avoid cheap biter types that can just bypass defences easily such as burrowing biters.
2) better biter AI, eg. biters that explore in a fixed area around their base or migrate from base to base, or try to find a slightly better way to attack, and in general just add more interesting behaviours to biters and add personalities. Perhaps different biters behave differently
3) biters should only build new bases by sending a colonization biter that literally walks to where it wants to start the new base. This way the player can prevent spawns and the biter nests don't just appear in the middle of your base if you don't have buildings nearby. This would fix this issue
4) add lots of customization options on game creation and allow tweaking As many of them as possible after game creation for biter difficulty. Eg. base size, base strength, number of biters in waves, how the waves increase over time, how quickly the biters evolve. Add toggles for each different biter type (more useful when more biter types are added), how quickly concurrent biter wave attacks increase. Add an option for almost constant attacks, etc.
5) add more types of guns
6) add more types of vehicles
7) add more types of items that can be used in vehicle equpment grids and player equipment grids
8) add more unique weapon types like guns, capsules, lighting defence, etc. Ie. more unique categories of weapons
9) better balance between weapons, vehicles and biters

[aristocrat]

I am not a turret creeper myself, but I do think that nerfing turret creeping seems kind of like a (somewhat) bad approach. I think the better approach is to buff some of the lackluster endgame-weapons (like the news post mentioned, explosive rockets), because right now (in vanilla) there is pretty much only two ways to get rid of biter bases in the endgame: turret creeping and destroyer swarms, the former doubling as defensive structures and the latter is consumed upon use. I don't think that it is neccesary to remove a mechanic like turret creeping in this sort of game. Those that despise turret creeping would probably not participate in it right?

IMO: Don't be too quick to nerf/limit turrets, make other options more attractive.

EDIT: grammar

[LucidMoses]

In re balancing combat. Some extra hard areas that don't move or expand (queen nest?) would be a good idea. They would be an obstetrical for building a factory until your powerful enough.

[vanatteveldt]

In re balancing combat. Some extra hard areas that don't move or expand (queen nest?) would be a good idea. They would be an obstetrical for building a factory until your powerful enough.

It could also alleviate the boredom of keeping biters at bay: destroy the queen and the area around her will stop respawning until a new queen spawns.

[Grimakar]

I definitely agree, giving the player a choice to enter combat mode that is easier to control/micro combat and gives more feedback, would be beneficial to the combat aspect of Factorio, it would also somewhat nerf the "Turret Creep" problem, as the player would have to take the gamble of exiting combat mode to enter build mode to build the turret and not be able to attack while doing so.

Somehow the switching of the toolbelt is the switching between combat and build mode, as far as you are using it as I do. And while I am turret creeping, I am only building and doing nothing else, so for me that wouldn't nerf anything.

[factoriouzr]

I am not a turret creeper myself, but I do think that nerfing turret creeping seems kind of like a (somewhat) bad approach. I think the better approach is to buff some of the lackluster endgame-weapons (like the news post mentioned, explosive rockets), because right now (in vanilla) there is pretty much only two ways to get rid of biter bases in the endgame: turret creeping and destroyer swarms, the former doubling as defensive structures and the latter is consumed upon use. I don't think that it is neccesary to remove a mechanic like turret creeping in this sort of game. Those that despise turret creeping would probably not participate in it right?

IMO: Don't be too quick to nerf/limit turrets, make other options more attractive.

EDIT: grammar

I agree

[tehroach]

I definitely agree, giving the player a choice to enter combat mode that is easier to control/micro combat and gives more feedback, would be beneficial to the combat aspect of Factorio, it would also somewhat nerf the "Turret Creep" problem, as the player would have to take the gamble of exiting combat mode to enter build mode to build the turret and not be able to attack while doing so.

Somehow the switching of the toolbelt is the switching between combat and build mode, as far as you are using it as I do. And while I am turret creeping, I am only building and doing nothing else, so for me that wouldn't nerf anything.

I suppose it is like everything else in Factorio, everybody does the same thing; differently.
lol, if that makes any sense to you

So I suppose my "Turret Creep" is also different from your "Turret Creep" for me "Turret Creep" is running in, to a point just outside worm range smashing down a 5h1t load of turrets, then filling the ones not currently being attacked with ammo, then running away guns ablazin. lol

Maybe I don't know how to "Turret Creep"

But you picked up on only a side point to the point I was actually trying to portray.

My main reason that I think a combat mode would really benefit Factorio would also coincide with the other changes I suggested to the biters themselves.
and the fact that I think it would be far better to have separate keys/buttons for each of my weapons rather than have to fuss with selecting them, then firing. I think it would be better to just push one key/button to activate and engage the desired weapon.

[mikiqex]

I start my games with Enemy Bases set to Very High freq. + Very big size, with very small starting area. I love the feeling of imminent danger and I feel cozy in my "castle". I'm big turret creeper and I don't like it, but I can't resist - if the game allows, it's no cheating. In the late game (20 hrs+) I go with big poles and 8-12 laser turrets, it's no competition and feels more like a job, than fun. I'm thinking that laser turrets deals way too much damage, in reality (like ATHENA by Lockheed Martin) it is a single beam that does damage gradually over time and long distance. Also, researching more damage should affect electricity consumption of the turret.

I quite like the "queen" idea, in that case pollution might affect the queen only, which will lay mutated eggs as a result. Queen can be like a "boss", with tough fight, but some nice reward - like alien artifacts (you wrote something about re-imagine their use, this way they may be quite precious). So bases will be like villages, with queens in the center, queen under attack may call for help nearby creeps from other villages.

More evolved creeps may get some more intelligence or just viciousness - attacking poles, trying to go around turrets or something like that - like humans have in their DNA that spiders are dangerous, they may know that turrets are bad (M'kay?)

[Grimakar]

... like humans have in their DNA that spiders are dangerous, they may know that turrets are bad (M'kay?)

Ehm, no?

[TheTom]

Combat at one points needs to be significantly harder. Seriously.

I am walling off my ressource based with one wall. No laser because I want the armor challenge. It rarely looses a turret. Permanent installations get a double wall. The first is NEVER breached.

Even the ammo consumption is not really hard. The result is a nuisance.

At some point biters should turn into a threat that requires logistics. I envision something like the scenes from Starship Troppers where a SEA of biters tries to kill a wall. Not a trickle. Send 100, 200, 500 biters in one run. Something where turrets run out, where I need to fall back to the 2nd wall an rebuild. Not something where I spawn a template over and over and it is never really challenged. I never had a single outpost overrun.