How is pollution actually calculated in-game?

[DrFrankinStein93 42]

Hello Simtropolis.

I hope you can help me resolve my confusion.
I was reading in the Prima guide and the SC4Devotion encyclopaedia that the Coal Power Plant air pollution radius was 16 tiles.


So I put it about 16 tiles away in my city but I found out that the pollution was reaching way further than 16.

So it got me wondering what was going on. I wanted to remove the other possible sources of pollution , so I made a new empty flatland city and plopped a new power plant and...

To my surprise, it seems the radius was not 16 but more like 32. I tried with other power plant and I saw similar result.
I also looked at the water pollution radius and also similar result. The radius seems to be double what indicated.

So what is going on? Do we know how pollution is supposed to be handled in game?

Theory time:
The only explanation I saw was in the Prima guide:

Quote

"A building emits air and water pollution every
month. The pollution is emitted at full strength
over the footprint of the building, then outward
in a fixed radius from the edges of the lot; the
strength of the pollution dissipates linearly as it
reaches the outer edge of the radius."

The explanation can still work out but it needs a little tweak.
It seems the "tiles" used in pollution radius are 2x2 and not the standard tile on the map.
It seems to align to what I'm seem overall.

PS: I'm playing with mods but they are mostly fixes and QOL while trying to keep the rest as vanilla as possible. I use the Air Quality Index mod that changes the dataview UI but even without it, I see the same behavior.

[CorinaMarie 29,771]

18 minutes ago, DrFrankinStein93 said:

To my surprise, it seems the radius was not 16 but more like 32.

Your empirical evidence conclusively shows that it does extend beyond the listed 16 cells.

I would surmise that's a deliberate dissipation effect that Maxis coded in. For example if I pour a bottle of pancake syrup onto a 1 foot by 1 foot tile in the kitchen then that is where the bulk of it will go, yet some will creep out to surrounding tiles too. Just not as much.

[DrFrankinStein93 42]

Funny thing you mention that, because Prima's guide explicitly mention the pollution don't extend further than the radius.
Let me test it out, I'll pop a bunch of coal power plant together if the pollution range goes more than 32, we might be into something.

[CorinaMarie 29,771]

7 minutes ago, DrFrankinStein93 said:

I'll pop a bunch of coal power plant together

Good plan. I'm also curious about two of them separated by 65 cells. Does the pollution then almost, but not quite touch each other? If you do 4 of them at the same distance at the cardinal directions does that one center cell stay clear?

I've always had the impression that air pollution was dangerously additive.

[DrFrankinStein93 42]

So first, 4 coal power plants in a square separated by 64 tiles (measured by the road price):


And then, 3x3 coal power plants all bunch together.

[DrFrankinStein93 42]

And it seems the 32 tiles of reach is from the center of the building and not the sides. The 64 tiles separated picture accentuate this a bit more.
Like the coal power plant is 4x4, so 32 tiles away from the center is 30 tiles from the side of the building. I will try to put them 60 tiles away. The circles of pollution should touch but not merge together.

[CorinaMarie 29,771]

4 minutes ago, DrFrankinStein93 said:

And then, 3x3 coal power plants all bunch together.

So we are getting greater additive intensity within the radius, but not exceeding the 32 cell boundary you found.

[DrFrankinStein93 42]

Here's 2 coal power plant separated by 30 tiles:

So barely touching but not merging together.

[CorinaMarie 29,771]

2 minutes ago, DrFrankinStein93 said:

So barely touching but not merging together.

Adds more support to the radius never growing beyond what you've already discovered.

[CorinaMarie 29,771]

Excuse the goofy UI in this pic. (I'm working on a different project.)

Here's the Maxis vanilla pollution data view:

 

And the real point of my post is to ask if you would like those neutral numbered marker lots? I use them to make a ruler in the game.

[DrFrankinStein93 42]

The marker lot seems perfect for our situation. And Maxis default UI is less sensitive, but we have a similar situation!

BTW, I've checked the exemplar property in iLiveReader and the radius value of the air pollution is really 16. So something happens inside the simulation.

[CorinaMarie 29,771]

27 minutes ago, DrFrankinStein93 said:

The marker lot seems perfect for our situation.

Since I didn't use them very often, I put them in the Reward Menu. (That's a bit annoying, but we can live with it (or change it)).

CoriBoom Neutral Marker Lots (Share Version) v1.0.zip

 

30 minutes ago, DrFrankinStein93 said:

So something happens inside the simulation.

All evidence indicates it's deliberate too.

[Simizen 1,099]

so, couple thoughts on this:

• I figured the color change would be useful, but didn't think about tracking stuff on a level like this, so, this is an incredible use of it, and very glad it's actually helping people out with tracking everything
• I think the extra pollution around the buildings is accurate for existing, but, not accurate in what it's doing
• and never actually saw the regular vs color coded views back to back before, and it's so wild seeing the default one (been running my version since I made it, lol)

But, if you've ever seen how SC2013 and C:S work, you'll notice that the air pollution is elliptical/oblong/teardrop shaped, and basically, the core of the pollution, will fit within the initial radius. But, the rest of the pollution, is supposed to go out differently, based on wind patterns. And, I think we're getting evidence of that based on other features in the game, that haven't really been utilized much. Like, we've got clouds, which are mostly aesthetic, but, we've also got the moisture data view, which seems to follow a certain bit of logic, implying that wind is a factor in there somewhere (as seen by the people who've had mountains in their towns and discovered one side covered in water, and the other's drier than bones). But, also the fact that there's even just a rudimentary weather system that seems to have never been fully developed (seen by the clouds at a far enough zoom level where it looks like it's raining).

I'm thinking this is something that just never got enough work on, cause the core seems to match up to what y'all are finding, that the pollution is getting kicked out within 16 tiles of the center of the lot. But, the fact that there's extra air pollution does seem to imply that other stuff is supposed to occur as well. And, tbh, even with my chart, I've found in a couple cities of mine, that where the roads are super heavily used, you can kind of see the air pollution spread out a bit from those hot spots. So, it's likely the devs realized that it would still extend quite a bit away from the point of origin. Just, it might not have been completed and adjusted for potential wind patterns and such. (Though, I could be completely wrong with the wind pattern effects being in this game, but there does seem to have been interest in adding a lot more to this game during development than what we've been given so far)

[DrFrankinStein93 42]

Thanks @CorinaMarie
I'll add your markers to my ever growing plugin list of fixes and patches.

Out of subject a bit:
 

Spoiler

My plugins folder is fulls of fixes and QOL update to get a better experience while keeping the experience as close to vanilla as possible.
Things like I-HT fix, NAM traffic engine (only the default classic engine update), DLL fixes, etc.
The more you look into it the more bugs you find hahaha.
To anyone reading this,  feel free to DM me your plugins suggestions.

 

[DrFrankinStein93 42]

@Ryuu Tenno As far as having wind pattern and such could be a nightmare awesome, I don't think the code handles anything for that. It would have made the game performance too poor for those old Pentium3 PC 20 years ago haha. But maybe you point in something interesting by saying the base radius is almost full pollution and it lowers fast after that, but it's hard to confirm that. We will have to wait for someone to reverse engineer in-depth systems to confirm. Not by me but probably someone smarter can do it one day.

BTW thank you so much for your mods (Air + Water). Those and the Aura mods are SO MUCH BETTER than the original. I skipped the garbage one because the Aura one replace it.

[CorinaMarie 29,771]

16 minutes ago, Ryuu Tenno said:

So, it's likely the devs realized that it would still extend quite a bit away from the point of origin.

5 minutes ago, DrFrankinStein93 said:

As far as having wind pattern and such could be a nightmare awesome, I don't think the code handles anything for that.

I added 100 kp/h YearlyAmbientWindSpeed(0xA7607D86)to the Weather Tuning Parameters exemplar and it has no affect on the air pollution around the power plant. Neither the existing saved one, nor a newly plopped one.

I believe Ryuu's point is Maxis prolly just added hidden wind dispersion code and the evidence suggest that filters out to double the set radius of the pollution source.

 

[Simizen 1,099]

Yeah, I was kinda realizing after I replied, that the stats are saying 200 units of pollution over 16 tiles, and it honestly doesn't make sense that it would simply stick to just those 16 tiles. That's kind of an awkward bubble, plus,. when the pockets are that small, it's pretty easy to contain them and clean them up, vs the work we have to do now whenever it gets out of hand. So, for sure the dissipation is meant to be in play as well.

The good news is, when I was messing around with the data views to get those colors (which surprisingly match up to that of the US AQI before I touched it), the farthest edges all drop off to zero (first attempt covered the entire city tile in green, and I didn't like that, lol). But, since the radius is double, we can at least run some quick math to see how it's dropping off. So, in theory, it should be say, 100 units about 24 tiles out? But, not sure if the drop off is linear or logarithmic (or reverse or however it works out). Though, from the looks of it, it seems to be a slow drop off from the center to the edge of 16 tiles out, and then a sharper drop off from 16 to 32.

11 minutes ago, DrFrankinStein93 said:

BTW thank you so much for your mods (Air + Water). Those and the Aura mods are SO MUCH BETTER than the original. I skipped the garbage one because the Aura one replace it.

No problem!! Glad you like them! I'd been wanting something different from the Maxis default anyway, and this worked out amazingly well imo.

The part that really interests me in this is the fact that Maxis had apparently implemented the values of the US Weather Service's AQI. That's had me wondering if maybe they were intending to do some more with it, but just stuck with the colors they had at the time for convenience (there's still a ton of carry over concepts and code from SC3K from what I can tell, and this was likely one of them).

3 minutes ago, CorinaMarie said:

I added 100 kp/h YearlyAmbientWindSpeed(0xA7607D86)to the Weather Tuning Parameters exemplar and it has no affect on the air pollution around the power plant. Neither the existing saved one, nor a newly plopped one.

dang, that would've been great to see any effects to it. But thank you for checking that. I didn't even know we had a wind speed option in the game, lol

[DrFrankinStein93 42]

40 minutes ago, Ryuu Tenno said:

But, not sure if the drop off is linear or logarithmic (or reverse or however it works out). Though, from the looks of it, it seems to be a slow drop off from the center to the edge of 16 tiles out, and then a sharper drop off from 16 to 32.

Pretty sure it will be linear. I don't think there is logarithmic slope in this game AFAIK. Just bunches of linear stuck one over the others.

Maybe it will be something like :

Range: Pollution
[0, R]:[100%, 90%]
[R, 2R] : [90%, 0%]

[DrFrankinStein93 42]

2 hours ago, CorinaMarie said:

CoriBoom Neutral Marker Lots (Share Version) v1.0.zip

I decided to merge them together so it's a single dat file. Easier to handle.

CoriBoom_Neutral_Marker_Lot_v1_0.dat

[DrFrankinStein93 42]

Just thought about this after a night of sleep. Is there any way to see the actual pollution values with the cursor or maybe exporting the map pollution values in a file (like a CSV file with coordinate and pollution raw values)? 

[CorinaMarie 29,771]

1 hour ago, DrFrankinStein93 said:

Is there any way to see the actual pollution values with the cursor

Yes. But it requires a building to be present where you hover the Query Tool.

Hold down Shift+Ctrl+Alt along with said query tool and then mouse over stuff:

 

You can also create a Query.txt file if you click while holding those keys. It'll wind up in:

C:\GOG Games\SimCity 4 Deluxe Edition\Apps

It is overwritten each time so if you want to collect them you'd need to do that between clicks. (Prolly not worth the effort since the in-game hover over works.)

Edit: This prolly requires the SimCity 4 Extra Cheats Plugin DLL.

[DrFrankinStein93 42]

Thank you that was exactly what I had in mind.

So with this tool, I was able to confirm multiple things and find interesting discovery.

Confirmation:
1) Pollution radius does seems to work as block of 2x2. Meaning radius is really 2 times bigger than the value would suggest.
2) Pollution decrease linearly to 0 over the radius. Center is 100% and 0% at the border. Diagonal distance are calculated using pythagorean theorem.

New finding:
1) Pollution is not the same over the full surface of the building. If you put a Power plant, run the simulation a bit, pause and delete the building, you will see that the pollution is only at full value on the center of the building.
2) The center of the building is not always the true center. Depending on the position on the map, the center is slightly shifted from the true center. I think it is because the code divide the cells position by 2 to calculate the values of pollution with those and then extrapolate it again on true cell position. So in practice, pollution circle could be slightly off-centered.

Here's a simulation in Excel of a coal power plant air pollution (Building is the cyan square, center is true center, AP: 200 R: 16). The tiny number in each cell represent the air pollution level.
Simulation results shows very similar results to game values (some small rounding error in my simulation).



I'm not sure if those finding is generalized to every pollution of every building since I've mostly checked with power plant (almost all 4x4 cells) but there is a good chance it is true. It would be weird to code something unique just to power plant. I think all pollution follow the same rule.

PS:

15 hours ago, CorinaMarie said:

Yes. But it requires a building to be present where you hover the Query Tool.

Best building I need was a 1x1 producing no pollution, so I used the small plaza to fill each tiles around the power plant.

  Edited February 4, 2025 by DrFrankinStein93  
Fix Typos

[CorinaMarie 29,771]

8 hours ago, DrFrankinStein93 said:

I think it is because the code divide the cells position by 2 to calculate the values of pollution with those and then extrapolate it again on true cell position.

I'll add in it might also be influenced by tract boundaries. The Coal Power plant would be an ideal one to plop fully inside a single tract.

Starting in the upper left hand corner, cells 0,0 thru 3,3 form one tract and these 4 x 4 (16 total cells) repeat throughout the city tile. (All the dataview averages are based on tracts (unless setting them to quantize lots).

To find the location coordinates, use Ctrl+X to get the cheat box then type in TerrainQuery and press enter. Now your Query Tool will show this:

The x, y, and z are in meters and fractions thereof. Note that y is used for elevation. And, ofc, the cell x and cell z are more useful for locating tract boundaries. I imagine you could whip up an excel formula to calculate all tract boundaries within a city tile.

 

Another thing I wonder if it'd be useful is to set the Pollution Dataview to show white for all unaffected cells? It would then look like this:

Ofc we wouldn't want this for everyday game play use, but it might be helpful for your investigations.

That area is just a bunch of Coal Power Plants for the main pollution. I achieved a high score of 5385:

[Null 45 3,064]

46 minutes ago, CorinaMarie said:

To find the location coordinates, use Ctrl+X to get the cheat box then type in TerrainQuery and press enter. Now your Query Tool will show this:

I am working on an update to my Query UI Hooks DLL that adds an option for TerrainQuery to show pollution data:

[CorinaMarie 29,771]

15 hours ago, DrFrankinStein93 said:

1) Pollution radius does seems to work as block of 2x2.

You can even tweak the Dataview to show that. Set Interpolate data to False (0x00) to see:

 

And in highly polluted areas doubling the Maximum scale gives this:

 

I do believe keeping the Interpolate set True is good, but maybe the 0x00000800 might be better for the Maximum scale?:

 

[Simizen 1,099]

quick tip: pressing `alt + prtsc` gives you a screen shot of the program you're in on windows, so you don't get window borders or task bars

also, didn't know you could adjust the gradient mix like that, that's pretty cool

and good to know that all the colors are showing up well too

[CorinaMarie 29,771]

32 minutes ago, Ryuu Tenno said:

quick tip: pressing `alt + prtsc` gives you a screen shot of the program you're in on windows, so you don't get window borders or task bars

Here's the funny part: I do indeed use `alt + prtsc`, but because the window goes down behind the Taskbar, it gets captured too. If I didn't use 'alt', you'd see both screens.

And yes, I could tell the Taskbar to auto-hide, but I find that extra annoying.

On the plus side, for pics which aren't Show Us style, I rather like having the time and date included.

[Kel9509 1,475]

19 minutes ago, CorinaMarie said:

And yes, I could tell the Taskbar to auto-hide, but I find that extra annoying.

On the plus side, for pics which aren't Show Us style, I rather like having the time and date included.

I tend to use Win-Alt-Printscreen, which captures the active window and automatically saves it in the Videos/Screenshots folder.  

[Simizen 1,099]

19 minutes ago, CorinaMarie said:

Here's the funny part: I do indeed use `alt + prtsc`, but because the window goes down behind the Taskbar, it gets captured too. If I didn't use 'alt', you'd see both screens.

that is absolutely weird. I wonder why it does that?

19 minutes ago, CorinaMarie said:

On the plus side, for pics which aren't Show Us style, I rather like having the time and date included.

okay, that's fair; time stamping stuff is good for references and such

[CorinaMarie 29,771]

15 minutes ago, Kel9509 said:

I tend to use Win-Alt-Printscreen,

That's cool and it does work for me:

But now I have a 1.58 MB PNG instead of a 234 KB JPG so, other than to show it above, I certainly wouldn't take advantage of ST's free storage this way.

Nor would I want to have to convert each image to save space both here and there.

I use a program from New Horizons Productions which was on the old LEX. It allows setting the JPG quality via a slider and it puts the image in any folder you specify along with your preferred prefix and then a consecutive number.