Friday Facts #286 - Pollution cleanup

[dog80]

now that i see the cute biters on the last image blabla i thought about "leaders" that buff friends through an aura like more attackspeed/runningspeed/attackdamage, and have special buildings in the nest that spawn these or so. This will result in more tactical building because some areas get stronger/more frequent attacks. Right now all attacks are more or less the same despite in the amount, but this evens out when you have reached a certain amount of pollution, because it spreads so far. Only when you have a big desert, ofcourse there are much stronger attacks than on a forrest. Idk i havent checked our biter mods so far. I was playing alot of poe recently and this is a common concept that groups get buffed by totems or rare "leaders". Anyways great news and big respect kek

[T-A-R]

.... and have special buildings in the nest that spawn these or so. This will result in more tactical building because some areas get stronger/more frequent attacks. Right now all attacks are more or less the same despite in the amount....

A fleshy beacon that buffs the worms and spawners... fruity land mines.. mechanics which are in game already. Nests with only one type of enemy, but tweaked ones, could also bring simple variaty.

[leadraven]

"emissions_per_tick_per_watt" and "pollution_absorption_per_second ".
Why ticks and seconds within the same context?

And what if the update rate drops below 60 ticks/second?

I can say the same about any in-game mention of seconds. For example, recipe times or mining speed.

[ejg]

I think it's quite unclear what this means:

Spawn values for current evolution factor (0.73)
35%: Medium biter: 20 pollution
64%: Big biter: 80 pollution

I think it's pretty clear:
- Evolution factor is unexplained, I agree. It also doesn't matter. It grows over time (from 0 to 1, but you don't need to know that).
- Percentages are change of spawning. This spawner has a 35% chance of spawning a medium biter, and a 64% change of spawning a big biter.
- Pollution values is how much pollution it will absorb to create those biters:
-- If it makes a medium biter (35% chance), it will absorb 20 pollution to create it.
-- If it makes a big biter (64% chance), it will absorb 80 pollution to create it.

Estimating Ammo consumption would be:
- Your electric miners generate 10 pollution per second. Let's say you have 100 miners on a patch of iron. That generates 1000 pollution per second.
- Therefore, you'll generate 50 medium biters per second, or 12 big biters per second.
- A piercing round does 8 damage by itself, and with bullet damage 3 it gets +60%, so 12.8 damage.
- A medium biter has flat 8 plus 10% damage resistance, so 12.8-8 = 4.8, 4.8 x 0.9 = 4.32 damage per bullet = 43.2 damage per magazine.
- A medium biter has 75 Hp, so it will take 1.74 magazines to kill each one.
- 50 medium biters per second will therefore require 87 Armor Piercing Round Magazines per second to clear the waves at steady-state.

The same math can be done for Big Biters, then use a weighted average between the two requirements to determine your actual bullet use rate.

Cheers!

I think that pollution works in three steps like (not necessarily in this order):

(1) pollution is produced in a single area block
(2) pollution is diffused to neighbouring blocks using some Jacobi numerical derivative approach, effectively smoothing pollution across the map and slowly spreading it (smoothing at the border of pollution spreads the pollution)
(3) pollution is destroyed in every single block by the entities in that particular block, e.g. by grass, sand, trees, waters, and spawners.

So the pollution created somewhere in your factory will not entirely be transformed by spawners to biters. First of all other things absorb pollution as well, and secondly: the pollution has to reach the spawners by diffusion, although convection with wind direction would be a nice addition IMO.

[Nidan]

"emissions_per_tick_per_watt" and "pollution_absorption_per_second ".
Why ticks and seconds within the same context?

And what if the update rate drops below 60 ticks/second?

I can say the same about any in-game mention of seconds. For example, recipe times or mining speed.

A factorio second is 60 ticks, always. Except for FPS/UPS displays, which must use wall clock.

[McDuff]

This is the first time in any playthrough that I've really *felt* the pollution mechanic, and I really like it, it feels much more fleshed out now, and all these abilities to surface the info and see what you're doing and what changes mean are really nice.

My current map is on Railworld settings and it actually feels pretty well balanced (so far) between pushing out and biter aggression.

I think one avenue to explore would be more options for mitigating pollution, perhaps as you get into the more late-game stuff. Either through efficiency or some kind of capture and sink system.

[SHiRKiT]

This was a very interesting read!

[bobucles]

50 medium biters per second will therefore require 87 Armor Piercing Round Magazines per second to clear the waves at steady-state.

Ouch! Each AP magazine takes 14 ore to produce so that means you will need 1218 ore per second to support the pollution output of 100 miners. Since 100 miners have a base production of 50 ore per second, you might run into a bit of a problem. Of course that does not include the extra pollution of electricity, assemblers and other things.

Pollution looks like it's going to be a huge problem to deal with. Will players need to form a non aggression pact with trees? That would be the worst.

[Koub]

This would be true if at least 50 spawners were sitting on top of the miners. Usually, the spawners are significantly further away, while the pollution spreads, and is gradually absorbed into the chunks. Moreover, there is a limit at which spawners do actually spawn, converting stored pollution into actual biters.

Taken from https://wiki.factorio.com/Enemies#Evolution

Besides choosing what kind of biter will be spawned, the evolution factor also influences the spawning interval. This interval (spawning_cooldown in the enemy-spawner definition) is interpolated between 360 (0 evolution) and 150 (1 evolution) game ticks (= 6 to 2.5 seconds).

So at 0.73 evolution (for which the calculation has been done), you'd have one biter spawn from each spawner every 206 ticks, or around 3.5s
You'd need to account for the number of spawners in your pollution cloud, and deduce the maximum wave size and attack rythm.

[bobucles]

I think it helps illustrate the point that you don't want biter nests in range of your pollution at all. For every ounce of pollution that a nest soaks up, the player ends up paying the biter tax many times over in terms of ammo. For a bullet based defense at least, the winning move is to have zero nests in range of any pollution at all.

Of course that story may change with higher tech weapons. Flame throwers and lasers boast a massive increase of lethal efficiency, and generate less pollution at the same time. There may be a point where pollution doesn't matter and it's simpler to pay the biter tax while unleashing a hail of laser firey death.

[mamuto]

That would also be a way to reintroduce the pink orbs which I miss pretty much.

I saw old game play footage and thought that was a neat concept. But I started playing factorio (0.16) way after that was removed.

I understand that would be hard to automate. I was thinking it would be neat if they brought it back where destroyed nests drop orbs, but also had the option of essential “capturing” a nest by killing off all the bugs and then hooking up pipes and pumps to a nest and it would suck the juices out then those would feed into chemical plants and continue making orbs. Once it is hooked up the nest would stop producing spawns, or if the pipes got backed up then they would start spawning again. (Or instead of fluids you could simply place a inserter next to a nest and it would grab orbs, but if the orbs begin to build up then it would start spawning). Also higher level nests (that have been on the map for a while) would produce more fluid/orbs than new nests, but as you drain them over time they would yield less until the nest is destroyed because it has been sucked dry.

That would make for challenging game play since you have to keep attacking the spawns while you get everything hooked up. You would need an option on the turrets to not attack nests.

[Pawz]

This interval (spawning_cooldown in the enemy-spawner definition) is interpolated between 360 (0 evolution) and 150 (1 evolution) game ticks (= 6 to 2.5 seconds).

So at 0.73 evolution (for which the calculation has been done), you'd have one biter spawn from each spawner every 206 ticks, or around 3.5s
You'd need to account for the number of spawners in your pollution cloud, and deduce the maximum wave size and attack rythm.

I think this is exactly what's missing. Forget 'evolution factor' it's meaningless because it's something we have to use to further calculate a useful number.

Saying "Will spawn biter every 3.5s with 64% chance ... " would be instantly clearer.

Another point with regards to difficulty -- is it possible to increase the # of biters spawned? So it looks more like..
64% chance of spawning medium biter
36% chance of spawning 4 small biters
(I'm trying to think of good ways to make biters more 'reactive' to pollution without increasing the spawn interval)

[cappie]

I just mentioned to kovarex that I found it funny how we both worked on the same stuff on the same day.. (although I worked on Grafana that day, not in-game graphs, yet I think I had to deal with a bit more data.. about 27 data points per server, 300 servers in total, for over a million timed entries)... even plotting the dots nearly crashed my browser, so I had to come up with a better way of separating the data, which I did (sorry, can't write an entire FFF here myself)

Man, I love these insights into what makes stuff tick at a game company such as Wube.. Also, the amount of work you guys are pushing through is just awe-inspiring.

[Griggers]

Can anyone tell me if the current pollution values are intended?

I'm experiencing a massive change in difficulty (from hard to almost impossible) on a low tree desert deathworld from early 0.17 to 0.17.12

Unless I'm missing something these values just don't feel right.

Example (not including boilers and natural dissipation):

Before 0.17.12:
An electric mining drill would produce 9 pollution/s and a small biter would cost 200.
This means that a single electric mining drill would end up generating one biter every 22.2 seconds

Since 0.17.12:
The electric mining drill produces 10 pollution/s and a small biter costs 4 (2 in Deathworld)
In this case the drill is generating 5 biters every 2 seconds (Or 5/s in DW)

As a result by not looking at the changes to the machines and only at the biter costs, these values make the game at least 50 times harder

And to make it even worse the pollution generated from creating ammo far outweighs the cost of biter generation.
1 piece of normal ammo costs 4 ore and can kill 3.333 normal biters or 5 spitters.
But creating this ammo generates 80 (from drill) + 25.6 (from furnace) = 105.6 pollution
Which means to kill 3.33 biters you let the game generate at least 26 more biters (52 in DW) and this is without calculating the pollution from the boiler.
So everything added together it is even more brutal that my calculation suggests.

[Koub]

Can anyone tell me if the current pollution values are intended?
I[...]
1 piece of normal ammo costs 4 ore and can kill 3.333 normal biters or spitters.
But creating this ammo generates 20 (from drill) + 25.6 (from furnace) = 45.6 pollution
Which means to kill 3.33 biters you let the game generate at least 11 more biters (22 in DW).

There might be a balancing issue - or not, I can't tell.
However, the direct conversion pollution generated by a mlining drill => Biters created is not an accurate measurement . See viewtopic.php?p=414099#p414099

Tests should be made on a testmap, with real numbers measured in the game to back up, and a thread should be created in balancing subforum if the experimental results indeed seem off balance.

[BlueTemplar]

Yeah,

Can anyone tell me if the current pollution values are intended?

I'm experiencing a massive change in difficulty (from hard to almost impossible) on a low tree desert deathworld from early 0.17 to 0.17.12

Unless I'm missing something these values just don't feel right.

Example (not including boilers and natural dissipation):

Before 0.17.12:
An electric mining drill would produce 9 pollution/s and a small biter would cost 200.
This means that a single electric mining drill would end up generating one biter every 22.2 seconds

Since 0.17.12:
The electric mining drill produces 10 pollution/s and a small biter costs 4 (2 in Deathworld)
In this case the drill is generating 5 biters every 2 seconds (Or 5/s in DW)

As a result by not looking at the changes to the machines and only at the biter costs, these values make the game at least 50 times harder

And to make it even worse the pollution generated from creating ammo far outweighs the cost of biter generation.
1 piece of normal ammo costs 4 ore and can kill 3.333 normal biters or spitters.
But creating this ammo generates 20 (from drill) + 25.6 (from furnace) = 45.6 pollution
Which means to kill 3.33 biters you let the game generate at least 11 more biters (22 in DW).

But also, try comparing to 0.16.51 ?
Also, you're not counting all pollution resulting from mining with that electric drill :
viewtopic.php?f=11&t=67201

[Ojelle]

Also the grond removes pollution for Free. So not all of the ammo pollution is converted in bitters.

[Unknow0059]

We've experienced the problem where if you have infinite time to work on something it ends up taking infinite time to finish.

That's a very quotable statement.

[Unknow0059]

Cliffs are total garbage.
I could build a useless machine to get to 99% evolution.

Jokes aside. No one cares about polution after 10 hours. The late game needs some more interesting and challenging stuff. The mid game got really extensive lately but for late... Let me grab mods cause the game delivers not that much. Just bigger values.

Why do you consider them garbage?

[SuperSandro2000]

Cliffs are total garbage.
I could build a useless machine to get to 99% evolution.

Jokes aside. No one cares about polution after 10 hours. The late game needs some more interesting and challenging stuff. The mid game got really extensive lately but for late... Let me grab mods cause the game delivers not that much. Just bigger values.

Why do you consider them garbage?

What do they add to the game except a temporary annoyance?