Multiple turret types/general turret rebalance

Post your ideas and suggestions how to improve the game.
Post Reply
User avatar
Drury
Filter Inserter
Filter Inserter
Posts: 679
Joined: Tue Mar 25, 2014 8:01 pm
Contact:

Multiple turret types/general turret rebalance

Post by Drury » Sun May 11, 2014 11:05 am

This was originally meant to be a reply to another thread about turret creeping, but then it got out of hand... Credit goes to just_dont for inspiration.

There could be 3 types of turrets - gun, energy and fire. Gun turrets would have to be fed ammo magazines by a belt, energy would require electricity, and fire would require a liquid pipeline feeding them with liquid fuel straight out of oil processing.

Gun turrets would be the most powerful type as they'd require electricity (for inserters) as well as ammo which has to be produced and delivered by a belt. For balance, I'd make it impossible to feed gun turrets manually (so it's not possible to turret creep with gun turrets).
Fire turrets would be the mid-powerful type, they would require no electricity at all, only a pipe - however, their fuel would still have to be produced in chemical plants. A good compromise for oil magnates and pyromaniacs. You're going to need less of these as they generally kill more biters at once, which also somewhat keeps their fuel consumption at bay. Oil is precious.
Energy turrets would be the least powerful type, but easy and cheap to run - all you have to do is plug them into a pole. Excellent for distant outposts, but you need a lot of them to get anything done.

Each turret type would include several sub-types - a short-range-yet-powerful one, a mid-range one and somewhat less powerful in terms of DPS (but makes up for it with it's effective range), and a super type with longest range and highest damage, but requires electricity feed regardless of type and doesn't target small and medium biters. Super type is also much larger and difficult to manufacture.

Gun:
Shotgun turret - poor range, requires shotgun shells. Shotgun in a turret form.
MG turret - what we have now. Chews mags, shoots good. Weaker damage output than shotgun, but manages to do more damage overtime due to extended range
Railgun turret - highest rate of fire of all super-type turrets (60rpm), requires special ammo AND direct electricity feed, but two-shots big biters, each shot penetrating everything in it's way, and has 1,5-2x the MG turret's range. Endgame goodness, three of these things supported by MG turrets to clear out the small ones will keep your wall untouched.

Fire:
Flame turret - short-range flame-throwing. Very hot.
Napalm turret - pumps out a long stream of fiery gel that spreads when it hits the ground. Good range, fun to watch. Weaker damage output than flame turret, but manages to do more damage overtime due to extended range.
Plasma turret - as hot as it gets. This thing doesn't only leave burning corpses, it outright evaporates everything it hits, leaving a scorching trail of blue flames. Works similarly to napalm thrower, but has a much longer range and is much more lethal - at the cost of very high plasma fuel and electricity consumption. Good at repelling large amounts of biters at once, but not a constant siege due to it's hunger for costly fuel.

Energy:
Tesla coil - as usual, short range, very shocking.
Laser turret - zaps biters at distance, weaker damage output than tesla coil, but manages to do more damage overtime due to extended range
Lightning cannon - fires a long lightning bolt that hits a big biter directly and then jumps to nearest medium and small ones. Despite this, it's the weakest super-type turret and what's worse, it charges up even slower than your laptop. Good for endgame outpost defense when big biters come into play, but similarly to plasma turret it can't repel a constant flow of biters due to very slow rate of fire. Unlike plasma turret though, it doesn't fare well by itself, and is mostly a means to weaken lots of enemies at once so they're easier to kill by accompanying laser turrets.


I tried to incorporate rockets/explosives as well but somehow they didn't fit in any equations, and besides, I think 9 types is enough to choose from. Keep in mind these were primarily thought up to be practical to use and balanced, not to be realistic or cool to watch, so don't get the wrong idea when you read about tesla turrets and railguns - they're all where they are to fit their purpose.

This idea also includes rebalancing of laser and gun turrets, but since it's more about the new things than rebalancing old things I put it here instead of balancing board.
Image

Coolthulhu
Fast Inserter
Fast Inserter
Posts: 201
Joined: Thu Apr 04, 2013 9:55 am
Contact:

Re: Multiple turret types/general turret rebalance

Post by Coolthulhu » Sun May 11, 2014 7:14 pm

I don't like the grouping of types.
  • You get clear progression within a type, which is a bad thing as it limits choice.
  • As weapon types are associated with damage types, balanced biter hordes would mandate having reasonably developed weapons from all tiers, once again limiting choice
  • Progression in each type is similar, meaning you can't really make decisions like "I'll rush gun tech now, because I want long range for now"
Individual turrets:
Shotgun at lvl 1 doesn't really make sense, as shells come after basic mags. Swapping it with MG would work better. Shotgun turret would need good armor (maybe even comparable to walls) and thus high cost, as currently shotgun doesn't shoot over walls (unless that would change).

Railgun sounds kinda OK, except for the pierce - it would promote strange wall designs in order not to make it waste ammo when it can't pierce, not to mention it's unrealistic. I'd lower the rate of fire to make it more of a sniper turret. Sounds like a nice sink for depleted uranium (and energy) and thus would work nicely with nuclear power plants if they were to be implemented.

Flame and napalm are too similar. I'd merge them. I'd also move the flame effect from ground onto biters: sticky flame instead of flame floor would be easier to balance.

Long range on plasma doesn't make much sense. It sounds like you wanted to fill the role of an artillery turret here (high cost, long range, high damage, hard to shoot for long time).
I'd add a cannon turret to cover this role (except for large energy cost) and change plasma to short range burst area damage.

I don't like the "videogame lighting" part of lighting cannon. So far Factorio does a nice job at realism. Also, once again this looks like a nice place to add an explosive ammo turret here - shrapnel or toxic gas canisters.
Instead, I'd see lightning cannon as a laser turret with huge rate of fire and huge energy cost with damage-per-watt lower than laser, but high target switching speed. It would make it good against large numbers of spread weak or damaged biters, but bad for small groups of big ones

User avatar
Drury
Filter Inserter
Filter Inserter
Posts: 679
Joined: Tue Mar 25, 2014 8:01 pm
Contact:

Re: Multiple turret types/general turret rebalance

Post by Drury » Sun May 11, 2014 7:50 pm

Oh, that's not the way I meant it.

Building gun turrets doesn't mean you can't build lasers elsewhere - from tech tree standpoint you'd research all 3 types of a single tier in a single tech. As I said, it's done this way for practical reasons, coolness points are secondary. It's basically to give you better tools for defending your base with and being more flexible at solving different situations.

Flame is different from napalm because it's effect spreads more - a trait shared by all level 1 turrets (and videogame flamethrowers in general).

Image

Napalm is a constant stream, sorta like garden hose. Except the liquid's on fire. It's effect spreads as the liquid that hits further away makes these fiery puddles, so it has some area denying effects, a specific trait of the fire type (as explained in the OP).

Image

You can't merge them because then you wouldn't have tier one fire turrets, meaning the whole hierarchy would get predated and gear people towards using napalm turrets instead of MG turrets before they get researched.

So far Factorio does such a nice job at being realistic that you're a guy who doesn't need to eat, drink or sleep, yet can carry huge factories and solar farms in his pocket, place down transport belts that somehow magically work a lá perpetuum mobile, and has the ability to craft complex circuitry by hand in a matter of seconds... I'd say on a real scale of realism Factorio isn't exactly up there - thank god, reality doesn't make exactly a fun videogame. Sometimes you have to do things unrealistically for the sake of balance. Now that that's out - lightning gun isn't even unrealistic, haven't you heard that the US military are on to it? They managed to get laser to make an ion path for a lightning bolt to follow, essentially resulting in what I describe here. I still rest my case - realism isn't the order of the day when it comes to making videogames and shouldn't ever be.

Yes, plasma turrets are super turrets, super turrets are artillery. This is once again a shared trait of every super type. Can't put cannon turrets in energy type, gun type has railgun as it's super turret. Plasma also can't be short range burst area damage because that's tesla coil. As I said, tier one is short range.
Image

Coolthulhu
Fast Inserter
Fast Inserter
Posts: 201
Joined: Thu Apr 04, 2013 9:55 am
Contact:

Re: Multiple turret types/general turret rebalance

Post by Coolthulhu » Mon May 12, 2014 1:34 am

Lighting gun: I though about that exact real life example for why your idea of lighting gun is unrealistic. It would ground itself out on first target. Also, I've yet to see forking lightning done well in games. So far it was always just "fireball but blue and fast".

About realism: have you heard of suspension of disbelief? To sum up the concept: things we don't know or abstractions aren't jarring, while redefinitions of real life concepts or logical inconsistencies are.
Dragons and teleportation machines never existed, so by themselves they do not break suspension of disbelief. Coal burning underwater, gold being stronger than steel or vegan cats do, unless a fantasy or sci-fi explanation is provided.
While realism "loses" to gameplay advantages, ideas you provided can be easily made significantly more consistent with real life without making them less interesting. In fact, clever use of real life concepts is often more "fun" than handwaving them with sci-fi.

About the 3 tiers - as I said, the progression of short range->mid range->area artillery is itself a boring idea. The only incentive to mix types would be different resistances and turrets would be mostly interchangable. Instead of the choice being "do I need plasma, some lasers, railgun, cannon or dark matter accelerator?" you're much more likely to end up with "do I need the range to be long?" and "do enemies have resistances?". Something as simple as shuffling the tiers in types (making tesla lvl 3 and laser lvl 1, shotgun lvl 2 and MG lvl 1) would improve things without negative side effects, other than breaking the arbitrary symmetry. Standardization can be good, but here the cost is just too high and gain too low.

Flame and napalm: I know about napalm in real life and flame in games, it's just they wouldn't be different enough as you described them to warrant splitting them. Since area denial isn't a thing without good AI or ranged enemies, it would only get different when biters reach the walls (thus kinda defeating the purpose of napalm having longer range). Other than that, it's basically just range difference.
Also, flame and tesla sound much too similar, with the only real difference being that flame would require clunky pipe system to go through the wall weakening defenses.

Phy
Burner Inserter
Burner Inserter
Posts: 18
Joined: Sat May 10, 2014 4:42 am
Contact:

Re: Multiple turret types/general turret rebalance

Post by Phy » Mon May 12, 2014 2:03 am

The Railgun should be really high damage and really slow firing. The ultimate in anti armor or big enemies, but really bad for swarms. I don't think the T3 turret should just be a straight upgrade from T1/2.

User avatar
Drury
Filter Inserter
Filter Inserter
Posts: 679
Joined: Tue Mar 25, 2014 8:01 pm
Contact:

Re: Multiple turret types/general turret rebalance

Post by Drury » Mon May 12, 2014 6:16 am

Well as I said, T2 turrets have lower DPS than T1, so you want to back them up with T1's (presumably by placing T1's in front of T2's). The range bonus is there so the turrets can start damaging biters before they reach the walls. T3's are meant to be endgame tech.

Different types aren't interchangeable because as I said, they differ in purpose. Laser turret is vastly different from napalm, which is vastly different from MG.

To elaborate - laser turrets are crap but don't need ammo, meaning you can deploy them anywhere within your energy network and they'll work, no need to produce and delver ammo. Napalm is quite the opposite - ammo is difficult to manufacture and deliver but the result is worth it when an occassional pack of biters comes along. MG is the optimal solution as the ammo isn't difficult to manufacture, meaning they can keep on shooting forever, but the turrets need a solid framework behind them, with belts and inserters and whatnot.

Resistances aren't important at the moment since biters have all the same values. You can easily tweak turret damage to get desired DPS on the biters.

Again, flame and tesla are similar because they're T1's - but not quite the same. See differences between fire and energy type.
Image

Kiapha
Inserter
Inserter
Posts: 23
Joined: Fri May 09, 2014 2:15 pm
Contact:

Re: Multiple turret types/general turret rebalance

Post by Kiapha » Mon May 12, 2014 6:46 am

Personally, I think that for the electrical defenses, you should start with something ground-based such as an electrified floor for short-range but high damage. From there, consider going to a tesla-coil or a laser. Both lose damage potential based on range, though laser works at a much longer range. Allow a tesla-coil to arc to targets that are touching for minor additional damage, but no large jumps. Swarms that are at point-blank range would suffer, yet otherwise it would act more as a high damage single-target bolt. Perhaps allow customized energy turrets where you can build a multi-block structure and fill it with modules to change how it acts. This would be a nice way to solve the "I need walls with four-thick laser turrets" that you might run into on the more insane difficulties, while still allowing a similar level of resource consumption, plus it would look better in many ways.

For shotgun turrets, you do not particularly have to set it up to fire over walls, you could attempt to create 'bunker walls' which have less hp, some armor, and allow any ranged attacks a chance to pass over them based on range. Have adjacent sources get a 100% chance. This allows the possibility of ranged enemies in the future being able to fire into your fortifications.

As far as I have seen, railguns have exactly one thing to do with speed: one projectile at ungodly speeds followed by an extremely long recharge time. I would say give it a huge amount of damage and track that. If it kills something, save the leftover damage and keep the shell moving in a straight line. If it hits and kills another thing, repeat with the lower-damage shell. It would fail incredibly against swarms without heavy support, but would rip a corridor through them with ease. Give it priority targeting to larger and heavier armored targets to maximize its effectiveness. This penetration effect would also solve the issue of trying to target something it can actually hit that still fits the targeting criteria -- ignore targets in between, just aim for the big bad nasties and carve a tunnel to them. :P Addiitonally, railguns require insane amounts of electricity. The combined reliance on expensive ammo and a huge power grid should make them suitably endgame, and one of the most powerful weapons.

Perhaps also add in a sniper turret, slow rate of fire, high damage, 3-4x normal range.

Artillery/mortars could be useful as well. Set them as having a very long range and a fairly large minimum range, very slow fire rate, and large splash damage, along with a bit of inaccuracy. This would make them useless against single enemies, but quite useful against swarms.

Perhaps also have some high tech level defenses that are more drone hangars than turrets, build a few combat drones and have them deploy from it, then return to recharge/reload after battle?

just_dont
Long Handed Inserter
Long Handed Inserter
Posts: 57
Joined: Thu Apr 17, 2014 1:24 pm
Contact:

Re: Multiple turret types/general turret rebalance

Post by just_dont » Mon May 12, 2014 10:32 am

As a person who helped sparking this topic to life, I feel compelled to write my thoughts here. There will be two parts in it -- one about "current biters vs. current towers", the other one more about Drury's suggestion in full.

1. A Blast of Steel For An Unhappy Biter.

Regarding to our current opposition, consisting of 3 types of melee units with different armor/hp values but the same role, I feel like having extended tech trees with many turret types is a SEVERE overkill. The enemy is quite plain, so there's no real need to punish it with a myriad of a bug-humiliating ways. Currently, I feel there's only two (intertwined) pressing problems with turrets: a) nobody wants SMG turrets because they are more difficult to set up for full automatic service; b) Laser turrets are much more easy to set up, AND they also more powerful.

So, addressing the current situation is all about fixing those problems. In the most simple way, it's possible by adding just ONE more turret type: "end game" (meaning, high damage), and "bullet type" (meaning, ammo must be fed into it). So, we would have 1 "basic" type -- SMG turrets (weak but available early), 1 "intermediate" -- laser turrets (less weak, available mid-game, very easy to set up), and 1 "hi-end" (available late-game, require ammo feed, but much stronger than lasers).
Some other things have to be adjusted (lasers would likely require a slight damage/speed nerf to fit into this new categorization), but otherwise it'll bring us some nice progression for turrets without big gaps.

2. One Thousand And A Half Ways To Decimate A Bug Horde.

Now, to Drury's suggestion. First off, I want to say that unless we would get MUCH more diverse opposition (shooting biters, flying biters, "carrier" biters, exploding biters, etc), a diverse turret system is not even needed. Secondly, having much more enemy types and turret types -- well, the game will take much more "Tower Defense" type colors. I'm not exactly sure that devs' want Factorio to become more tower defense-ish type of game. But I may be wrong.

On that aspect, I want you to go and look at a small TD game called "Harvest: Massive Encounter" (written by a certain Jeb who nowadays updates Minecraft). The game has only FOUR tower types (and three of those are more "auxiliary"), but still manages to be great. Why is that? Because it takes on a same idea "combining big stuff out of small elements" that Factorio has at its core. In "Harvest", your basic laser towers (and also your ONLY mainstream tower type) can be linked to each other to provide "boosts" to the last tower in that chain. This seemingly small idea leads to LOTS of interesting designs and variations.

So, returning back to Factorio. If Factorio one day will take more of that tower defense-ish approach, and there will be diverse bug hordes against us, I feel like our turrets should then somehow use that "big things out of small things" approach as well, and become modular, combining, or somesuch. And then it would be awesome.

User avatar
Drury
Filter Inserter
Filter Inserter
Posts: 679
Joined: Tue Mar 25, 2014 8:01 pm
Contact:

Re: Multiple turret types/general turret rebalance

Post by Drury » Mon May 12, 2014 12:30 pm

just_dont wrote:As a person who helped sparking this topic to life, I feel compelled to write my thoughts here. There will be two parts in it -- one about "current biters vs. current towers", the other one more about Drury's suggestion in full.

1. A Blast of Steel For An Unhappy Biter.

Regarding to our current opposition, consisting of 3 types of melee units with different armor/hp values but the same role, I feel like having extended tech trees with many turret types is a SEVERE overkill. The enemy is quite plain, so there's no real need to punish it with a myriad of a bug-humiliating ways. Currently, I feel there's only two (intertwined) pressing problems with turrets: a) nobody wants SMG turrets because they are more difficult to set up for full automatic service; b) Laser turrets are much more easy to set up, AND they also more powerful.

So, addressing the current situation is all about fixing those problems. In the most simple way, it's possible by adding just ONE more turret type: "end game" (meaning, high damage), and "bullet type" (meaning, ammo must be fed into it). So, we would have 1 "basic" type -- SMG turrets (weak but available early), 1 "intermediate" -- laser turrets (less weak, available mid-game, very easy to set up), and 1 "hi-end" (available late-game, require ammo feed, but much stronger than lasers).
Some other things have to be adjusted (lasers would likely require a slight damage/speed nerf to fit into this new categorization), but otherwise it'll bring us some nice progression for turrets without big gaps.

2. One Thousand And A Half Ways To Decimate A Bug Horde.

Now, to Drury's suggestion. First off, I want to say that unless we would get MUCH more diverse opposition (shooting biters, flying biters, "carrier" biters, exploding biters, etc), a diverse turret system is not even needed. Secondly, having much more enemy types and turret types -- well, the game will take much more "Tower Defense" type colors. I'm not exactly sure that devs' want Factorio to become more tower defense-ish type of game. But I may be wrong.

On that aspect, I want you to go and look at a small TD game called "Harvest: Massive Encounter" (written by a certain Jeb who nowadays updates Minecraft). The game has only FOUR tower types (and three of those are more "auxiliary"), but still manages to be great. Why is that? Because it takes on a same idea "combining big stuff out of small elements" that Factorio has at its core. In "Harvest", your basic laser towers (and also your ONLY mainstream tower type) can be linked to each other to provide "boosts" to the last tower in that chain. This seemingly small idea leads to LOTS of interesting designs and variations.

So, returning back to Factorio. If Factorio one day will take more of that tower defense-ish approach, and there will be diverse bug hordes against us, I feel like our turrets should then somehow use that "big things out of small things" approach as well, and become modular, combining, or somesuch. And then it would be awesome.
Well now we're getting somewhere.

Actually the basic idea was to make laser turrets less powerful than gun turrets for obvious reasons - gun turrets are more expensive to set up and run. Basically my post was originally meant to be a sort of appreciation station where I nod to everything you said. But then I realized that laser turrets make no sense anymore - you research inferior tech. So I got the idea of putting laser turrets and gun turrets in the same place in the tech tree, where gun turrets would be this complex stationary thing that requires planning and infrastructure and such, but are very good at defending your main base, and cheaper disposable laser turrets which only need electricity poles next to them. So you'd have two types of turrets for two types of environment - an efficient but complex type for defending complex infrastructure, and an easy but weak one for disposable outposts.

And at that point I decided to take a better look at base defenses in this game as a whole and see what could be done better. One thing that popped up instantly - make turret upgrades physical. This resulted in tiers. I always thought it's not a lot of fun simply clicking at gun turret upgrades in tech tree window and having all the turrets magically do more damage without any work needed by the player, without sorta looking back and fiddling with the framework to upgrade it - especially since you have to do it with things like furnaces, belts and assemblies. And to give it that extra kick I made the tiers dependent on each other - a good defense framework should include every tier. At the same time, tech tree progression isn't broken and each tier can do something that the previous tier couldn't - while still being dependent on the lower tier. I used extended range for this, as it means that with multiple turret types you have multilayered defense, getting hotter and hotter closer to the walls. This is why T2's are weaker than T1's. Super ones have a support function, you're going to have normal turrets doing the most work. They are mostly there to weaken the horde before the main slaughter.

That is the core design and the cool things like railguns and tesla coils are there mostly as the most appropriate fancy sci-fi concepts for the core design behind the turret - once again they're where they are to fit their purpose, not to be realistic or cool. Maybe there's a way to implement explosive type, but I can't imagine a situation that isn't already covered by the previous three types (and even 3 with 3 tiers each is a bit too much).

Why all that stuff? To make base defense more interesting than just slapping down a wall of laser turrets and plugging them into poles. As simple as biters are, they deserve to die in a true Factorio fashion - to fall victim to complex, sinister and ingenious designs.
Image

Post Reply

Return to “Ideas and Suggestions”

Who is online

Users browsing this forum: boskid