How to evaluate pollution?

[4xel]

Ok, so I ran the numbers for a typical batch of 3 behemoths biters, 4 big ones and 1 medium, assuming an evolution factor of 1 and using the repatition given in the wiki : https://wiki.factorio.com/Enemies#Evolution

This batch totals 10 575 Health points (https://wiki.factorio.com/Enemies#Biters), and requires 77 000 pollution to generate (https://wiki.factorio.com/Pollution#Native_life).

An unupgraded laser turret uses 800 kJ per shot and deal 20 damages.


(800 / 20) * (10 575 / 77 000) = 5.49 kJ/pp
(pp = pollution point)

What does this number means?

For example, if you let a burner mining drill run for one second, it will generate 10pp, which you will eventually have to pay 54,9kJ of electricity for in laser turrets. By comparison, the same burner drill will burn 300kJ of fuel by itself during the same second, which translates into 150kJ of electricity, if we want to keep things comparable. So that's a 36.6% energy penality.

For electric drill and pumpjack, the most polluting per J building in the game, that number raises to 54.9%, meaning when you think they consume 90kW, they actually cost you 1.38kW total.

For other building, use the table found in the wiki here :
https://wiki.factorio.com/Pollution#Pro ... Absorption

The column "Pollution per KW per game second (60 ticks)" also shows the Pollution per KJ if you change the unit, and multiplying the numbers in that column by 5.49kJ gives you the fraction of the power consumed by the machine that you'll have to pay in laser shots (multiply by 2 for burner mining drills and boilers).

Another example worth of notice is boiler : at 15% penality, it means that if you rely on steam power for a high pollution base (I don't know why you would do that, but hey), your steam plants effectively only produce 85% of what you think they produce.

That "Pollution per KW per game second (60 ticks)" column is what affects the pollution effect of production modules. So they directly increase the share of energy dedicated to laser shots.


Now this 5.49 kJ/pp is very pessimistic. Here are some reasons which makes it vary, most of them making it diminish :

• Evolution factor. At 1, the ratio HP/pp is very high, 0.137, very close to the 0.15 of behemoth. Big biters are under 0.1, and small and medium ones are at 0.75
• Techonlogy. At level 7 Laser Turret Damage technology, turret damages have tripled effectively dividing by three kJ/pp
• Laser turret drain. Not a cost per pp, and even tends to be negligible the more you use turrets, ie the more you pollute, but drain has to be weighted for bases at the frontier between low pollution and high pollution, because the alternative is non draining gun turrets with lower upfront cost, and possibly lower upkeep cost as well.
• Mixed defense. I have used Laser because that's the most common defense, being the simplest to setup and only requiring electricity. Gun turrets are about 10 times more energy efficient at their best but eat ores (and their best is normal ammo and uranium ammo with modules, so mid game piercing round kind of sucks compared to Laser), flame turrets and landmines are hard to evaluate.

So overall, I'd say this number should range from 4kJ/pp to 2kj/pp, decreasing with time, assuming laser defense.


TL;DR : you have to pay for the negative extarnality that represents pollution by killing biters and spitter. The price represent a non trivial fraction of your total energy consumption. From a few percents for most machines, about 35% for miners and pumpjack, ignoring Productivity modules (worsening) and defense technology (reducing pollution cost).

The figure I'd go by is 2kJ to 4kJ of laser turret shot per pollution point.

[bobucles]

Honestly, a 15% tax on 0 upgrade laser turrets vs. max evolution is really not that punishing. There's no way you are getting max biters without a handful of laser upgrades (100%-200% is very attainable) and free damage means less energy being spent. Try comparing it to yellow or red ammo and you'll see that lasers are not expensive weapons at all.

Gun turrets are about 10 times more energy efficient

I suspect your math on this is wrong. Mining, smelting and crafting all require energy and ammo takes a LOT of it. The I'm pretty sure the actual energy cost of ammo is higher than lasers, and that's not counting the precious ore.

[4xel]

Honestly, a 15% tax on 0 upgrade laser turrets vs. max evolution is really not that punishing. There's no way you are getting max biters without a handful of laser upgrades (100%-200% is very attainable) and free damage means less energy being spent.

I'm not saying it's punishing, and Laser Turret is definitely the way to go for high pollution strategy. I'm just saying here are the numbers. One way to use the data is to remark that the energy spent on defense is not neglibible, so it does make a difference to improve laser turret tech rather than just adding more of them.

That being said :

Gun turrets are about 10 times more energy efficient

Can we agree unupgraded Laser turret are 40kJ per damage?
(2.4MW, 3 shots a second, 800kJ and 20 damages per shot ; turret's speed does not change consumption per shot)


Cost of mining one unit of ore in electric mining drill : 90/0.525 = 171.4 kJ
Cost of smelting one unit of ore in stone furnace : 3.5*90 = 315 kJ (electric equivalent, assume steel furnace if it bothers you)

Cost of a crafting second in assembly machine lvl 1/2/3 :
180kJ/200kJ/168kJ


Normal ammo magazine (4 mining, 4 smelting, 1s of crafting) :
cost : 2124kJ/2144kJ/2112kJ
Damage : 10 * 5 = 50
Cost per damage : 42.5kJ/42.9 kJ/42.24kJ

Barely more than laser turret, and that's not counting efficiency modules and better tech. Efficiency modules are a five fold improvement, and tech is a two fold improvement (turret specific upgrades are about the same, but gun turret benefits from ammunition tech additionnally, so to speak since the bonus is actually multiplicative).

Since you're likely no longer using normal ammo by module time, let's compare, for the sake of the argument, unupgraded Laser Turret with Gun Turret normal ammo, gun turret damage 2 and bullet damage 2. These tech are far cheaper than laser alone, which requires no less than 200 military packs. In fact I could easily go up to 3, but let's assume oil processing, battery, advanced electronics... are free because useful otherwise (definitely not true in a marathon deathworld).

That's an improvement by a factor of 1.44, or 29.5 kJ per damage, cheaper than laser, and you get ther 15 to 20k worth of tech ressource earlier.



piercing round magazine (14 mining, 19 smelting, 3s of crafting) :
cost :NA/8979kJ/8883kJ
Damage : 10 * 8 = 80
Cost per damage : NA/112.2 kJ/111kJ

I agree they are terrible. Even factoring the twofold improvement from tier3 tech, it's still worse than an unupgraded laser turret, yet you can hardly stick to normal ammo due to biters starting to have 4/10% damage resistance (or a bit less than half what tech 3 normal bullet do, and more than quarter of piercing round damages). With efficiency modules, it still beats Laser turrets handily though. Like by a factor 3-4.

It's worth noting that the tech for flame turret is much cheaper than the one for laser turret, so if for whatever reason one want to avoid laser turret, it's totally possible to go for a gun+flame combo, with flame doing the big work while waiting for uranium ammo.

Edit, I missed 1s of crafting time, but you get the idea.



uranium round magazine (16 mining, 19 smelting, 14s of crafting) :
cost :NA/11151kJ/11073kJ
Damage : 10 * 24 = 240
Cost per damage : NA/48.0 kJ/46.1kJ

Similar but worse than Laser right?

Let's look at a typical endgame Laser Turret, with Laser Turret damage 7 :
Damage of a shot : 66 (20+46)
energy per damage : 800/ 66 = 12.12 kJ


Let's look at a typical endgame gun turret : Uranium Ammo produced with efficiency modules, Gun Turret damage 7, bullet damage 6
Damage of a magazine : 10 * 24 * 2.9 * 2.2 = 1530
Damage vs any spitter : 1530
damage vs big biters : 1310
Damages vs behemoth biters : 1270

cost of a magazine : 2215kJ, assuming -80% due to efficiency modules (also why I assume stone furnace, since conveniently, electric furnace consumes the same). That price could drop lower with other combination of modules, like speed and efficiency.

Cost per damage spitter : 1.44kJ
Cost per damage big biters: 1.69kJ
Cost per damage behemoth biters: 1.74kJ

Ok, not quite tenfolds, more like 7 to 8 times better. Better even without efficiency modules. It only get worse with infinite research, lasers scales linearly with the logarithm of ressources spent, Gun scales quadratically.


I would still use Laser Turret in most of my games, for ressource efficiency, but Gun Turret is definitely NOT strictly worse than Laser Turret (assuming tech is not an issue). If Laser were more energy efficient than Gun Turret, I would see no use for Gun Turret, but as is, Gun Turret are a viable option for

• low energy playthrough : Setting up no drain guns is easier than no drain lasers+power grid + accu
• ressource rich playthrough : Easier and less time consuming to burn your ressource than to conquer lands to put your solar panels
• land poor playthrough : same as above, though nuclear reactor + laser might be an option late game.

In default settings, energy is free only because space for solar panels is free, because you conquer space anyway to built mining and pumping outpost, ressources are the bottle neck. But there are world settings where ressources are free because you conquer space and big rich ore deposit to build your factory anyway, space is the bottle neck. And there are other world setting where time is the precious ressource, and gun + flame turrets make more sense than laser turrets, at least for a long time.


I suspect you overlooked technology, modules, and the size of magazines (10 rounds) in your calculation, that's the only way you can get the numbers the other way around.

[BlueTemplar]

1.) Gun Turret / Bullet Damage 2 requires no MilSci, I'd rather compare with Gun Turret/Bullet Damage 3 (100 MilSci each).
That's +40% damage each, +96% damage total.

For 9.8 damage per shot for normal ammo (5), and 15.68 damage per shot for AP (8).
Against a big biter (8/10% physical resistance), that's only 1.62 damage that passes trough the armor per shot for normal ammo, 6.912 for AP.

(EDIT : The biggest factor for Gun turrets vs Flame/Laser will be access to oil. You can go without oil up to Gun T/ Bullet damage n=°4 - 200 MilSci, +20% damage each ;
vs a big biter that would give you an additional 2.7 damage per shot for normal and 4.32 for armor piercing ammo.)


2.) You forgot to mention (as I already told), that up to 100% of your pollution will be absorbed by terrain and trees, rather than the Spawners.
Again, this is one of the hardest parts to figure out (along with the damage dealt by the aliens) in the pollution cost.


3.) Flame turrets, compared to laser turrets, not only cost 150 less MilSci, but also outrange medium and big worms - quite important for turret creeping !

fire_turret_block_pushing.png (5.48 MiB) Viewed 5542 times

[4xel]

2.) You forgot to mention (as I already told), that up to 100% of your pollution will be absorbed by terrain and trees, rather than the Spawners.
Again, this is one of the hardest parts to figure out (along with the damage dealt by the aliens) in the pollution cost.

If you opt for a high pollution strategy, you can reason at the margin : any excess pollution will generate biters, proportional to pollution you emit. Some pollution will be absorbed by trees and grass, but it does not matter ; if you intend to pollute more by an amount of X, that's X pollution worth of biters more to kill.

pollution absorbed by terrain and forest is like revenue, land lord annuity. Assuming your pollution cloud reaches the bitter, it is constant for a given amount of land, and even scales down with pollution, due to trees dying in presence of excessive pollution.


If your pollution cloud is not reaching the biters, you're in the low pollution game, and gun turret become much more appealing (much lower upfront cost, irrelevant upkeep cost).

I agree, the transition from one to another is not linear, and what I provided is only the marginal cost of pollution in the high pollution game.

The difference between a highly polluting base and a lowly polluting base is much more than one or two turrets or power generators, and would require comparing two different bases, on several criteria (ressource efficinecy, time to rocket, assymptotic growth...), since I don't think one is strictly better than the other.


If that can bring something to the discussion, the New Hope mission where you have to build a plane has a strong reputation of being super hard. As far as m personal experience goes, the first time I tried the mission, I did not cared about pollution (playing the high polluting game) and got crushed very hard. The second time I went slow (low polluting game) and the task was trivial, I was effectively playing peaceful.

[4xel]

1.) I'd rather compare with Gun Turret/Bullet Damage 3 (100 MilSci each).
[...]
For 9.8 damage per shot for normal ammo (5), and 15.68 damage per shot for AP (8).
Against a big biter (8/10% physical resistance), that's only 1.62 damage that passes trough the armor per shot for normal ammo, 6.912 for AP.

I was being conservative but yeah, you can definitely consider level 3. That would translate to 20 kJ per damage compared to unupgraded laser's 40, plus even lower setup cost.

The biggest factor for Gun turrets vs Flame/Laser will be access to oil.

agreed.

3.) Flame turrets is love,

flame turret is life.

I've ran some quick mental calculation, and an unupgraded flame thrower turret powered by depleted pumpjacks looks like in the same league as fully upgraded uranium gun turret as far as energy goes, and oil is renewable, in factorio at least (and plentyfull, as far as turret's needs go). And that's lower estimate, not counting AoE and DoT. Don't quote me on that though (quick mental calculation).

(those pesky worms have some fire resistance though)

[4xel]

Again, this is one of the hardest parts to figure out

Most of the numbers needed are here :
https://wiki.factorio.com/Pollution#De-polluters

The only two hard thing about figuring out the pollution absorbed by the landscape are

• Knowing the frequency, size and richness of forest (as numbers)
• Making sense of a chunk. The data is here, it's just our intuition which is failing

For the second points, for example, the 49 chunks a radar can constantly cover absorbs 1.323 pollution per seconds, assuming grass and no tree. If that same radar is powered by a coal steam plant, it indirectly generates 4.6 pollution, by the boiler alone.

So forest (and maybe small plants, but I do not see their stats) are the main absorber by a very large margin, though a very big body of water, depending on the setting, can have some effect.

A single chunk full of the greenest forest absorb 16*16*0.06 = 15.36 pollution per seconds. We're free to make guesses as to how green and dense forest are in default settings, or in the settings we're playing at. My guess would be between 5 and 12 pollution per chunk per second, depending on the look of the forrest, with most forest being good evaluated at 7.5, accounting for the sttraggler and the reduced density on the borders.

Assuming 7.5, a forest of 7x7 chunks, like the area covered by a radar, absorbs 367.5 per second (that's a typical forest I'd say, maybe a bit small, but definitely not the smallest). That's a bit less 41 electric miners without modules, which can cover a very decent ore patch, or fill half a blue belt of ore.

[mrvn]

When computing the cost of gun turrets don't forget that mining ores and producing ammo produces pollution. Pollution produces aliens, Aliens need ammo to be killed. So a percentage of the produced ammo is needed to counter that.

[bobucles]

With efficiency modules, it still beats Laser turrets handily though. Like by a factor 3-4.

But how much does that really matter? The utility and expense of an efficiency module is directly comparable to the cost of solar panels for lasers. If you're throwing down a hundred efficiency modules to have "amazingly cheap ammo" you can also throw down a hundred solar panels and have totally free lasers. Granted the eff1 module is still a somewhat stronger choice than solar panels with all things considered. The reduction in pollution tax will help ammo and lasers just the same, but energy savings can be stacked into bullets while lasers have no efficiency options. On the other hand, every stack of ammo not used is another investment in more free laser ammo.

I find that players don't use electric furnaces or efficiency modules very much at the time where ammo cost matters. The typical setup for the main meat of Factorio is steel smelters into Assembler 2 machines while using red ammo. In this standard situation lasers are clearly superior but nothing compares to fire.

[4xel]

When computing the cost of gun turrets don't forget that mining ores and producing ammo produces pollution. Pollution produces aliens, Aliens need ammo to be killed. So a percentage of the produced ammo is needed to counter that.

Excellent observation. It should not set the numbers off by more than a factor of 2 for normal ammo, and maybe like 1.1 for uranium ammo, I will correct my math. By the way, I think one can prove using productivity module on ammunition assembly in a high pollution base is a bad idea (intuitively, the philosophies behind gun turret and productivity module are polar opposite, but it still is nice to see a mathematical proof of it).

I find that players don't use electric furnaces or efficiency modules very much at the time where ammo cost matters. The typical setup for the main meat of Factorio is steel smelters into Assembler 2 machines while using red ammo. In this standard situation lasers are clearly superior but nothing compares to fire.

Why they use the worse kind of turret configuration in the game without any of the strategies that could barely justify it, is beyond me. Laser and fire turret is indeed what they need.
In a low pollution strategy, though (cloud not reaching enemies, which are barely ever spawning) you don't have to use efficiency modules if you don't need them (if pollution cloud do not reach enemmy). And if you want to go low pollution then high pollution, it may make sense to ramp solar energy and not bother with efficiency modules, while using red ammo for a short period of time I guess.

With efficiency modules, it still beats Laser turrets handily though. Like by a factor 3-4.

But how much does that really matter?

I don't know but that can be calculated. I would look at setup cost per DPS and setup cost per (DPS times Range). The setup cost of a laser turret is very hard to compress if you want them to be used at full power (scaling savings can somewhat help for laser turrets with low usage, they can share accu), while the size of your ammo factory is directly related to turret usage, so the lower the better.

As far as my personal opinion/wild guess goes,

Laser Turret is definitely the way to go for high pollution strategy.

For low pollution bases, I expect the fact that you need to deal less damage to matter much more than the fact than ammunition cost little electricity, so to answer your question, I don't think it really matters. But I find it worth noting Gun Turrets are not strictly worse than laser turret. Higher DPS may also mean less damage to your infrastructure, especially from biters, (and so does laser's higher range especially vs spitters).


For a low pollution strategy, I would not bother building laser turrets until very late. The reduction of pollution to such an extent that you're barely attacked does favor gun turrets more than laser turrets :

• Less ennemies means fewer turrets shots, which means technology takes longer to pay for itself. Gun turret first levels is nearly 0 technological cost, compared to laser's.
• Faster setup : you drop a turret and 10 magazine and you can forget about it until you no longer need the turret here.
• No electrical drain. Laser turret electric drain is a non trivial expense, in the few percents of an active turret total usage. Now picture inactive turrets, and factories
• Great wall of China : gun turrets are significantly cheaper to wall big areas. Comboes with the two points above. ammunition significantly increases Gun Turret cost, so you'd better ensure biters actually barely spawn, but you shoul dnot need anything else than a few ammo, while laser turrets need power poles. Also, biters have spread far less by the time you can clear them and wall the place with gun turrets.

So for me, in this particular situation, laser turrets sound like an expansive and unnecessary sidegrade, if not downgrade, of a fully functionning secondary emergency system;

[bobucles]

A base that doesn't need defenses doesn't need laser turrets

Stop the presses, we have a breakthrough!

[4xel]

A base that doesn't need defenses doesn't need laser turrets

Stop the presses, we have a breakthrough!

But it still needs gunt turrets, the risk is too high.

And if you only need to defend against small biters, you don't need laser turrets either.

[4xel]

A base that doesn't need defenses doesn't need laser turrets

Stop the presses, we have a breakthrough!

But it still needs gunt turrets, the risk is too high.

If you only need to defend against small biters, you don't need laser turrets.

If you only defend against migrants, you don't need laser turrets https://www.reddit.com/r/factorio/comme ... r_factory/

[BlueTemplar]

Defending against migrants... that have trouble crossing water... using machine gun turrets... brings up uncomfortable associations !

[4xel]

Migrants who won't even pay for our wall.