Corruption due distance vs. corruption due number of cities?

[Auren]

So far in every game I have played I have tried to build as many cities as possible without overlapping each other. So I always end up with cities that have lot of water squares, tundra, desert etc.

I am now startign to wonder if I instead should try concentrate on making only cities on good places, so that preferably each square in city are would make 3+ combined food+resources.

That way I would hopefully have lot less corruption since I imagine number of cities have significant effect to corruption. Also all cities would be good producers and would have ability to grow quite fast.

Problem is that computer would build cities "inside" my territory since cultural boundariers wouldn't be tight enough until lot later. On the other hand those cities would prolly be quite easy to either take with culture and kill to starvation or to conquer and raze.


Has anyone made this kind of test and could tell how much effect number of cities have to corruption and how much distance has?

[MarshalN]

I think building cities in places that aren't so desirable is still good. They give you some form of production/culture, and any of those points are useful. Plus, they might be sites of strategic resources later on (like saltpeter in deserts) so they're not worthless.

I am not sure how much number of cities count in corruption. I know that in Civ2 if you build a lot of cities corruption goes up a bit and unhappiness goes up too, but I haven't played enough Civ3 to notice that yet. It wasn't very big in Civ2 -- mostly distance from the capital, and I'd imagine it to be so in Civ3.

[Auren]

I was thinking about having 3-6 unaccessed squares between my cities so that my culture would eventually cover it and I would get all the strategic resources that appear there.

With strategy I nowdays use there is usualyl no squares at all between my cities.

But if corruption is mostly for distance then this idea is of course useless.

[Fitz]

I plan to study this and try and figure out a formula. My review of the guidebook that was pakaged didn't have one, so I may have to do this from scratch. I never played Civ2, so I don't know if there was a specified formula for corruption, but in Alpha Centauri, there certainly was for corruption and extra unhappiness, from both more cities and distance affects, so I'll see if it matches that in style.

If anyone knows a formula for corruption and unhappiness, be it Civ2, Civ1, SMAC, or somewhere in Civ3 that I've missed, please email it to me at junioros@hotmail.com

After I get Corruption and Unhappiness down, I'll move on to Pollution.

[MarshalN]

I don't have a specific formula, but I remember when you get to 15-20, maybe a bit more, cities in Civ2, on deity, your inital population in the city would be a double unhappy guy. Starting with about 7-10 cities it would just be one unhappy guy........ and with those double unhappy guy cities I just made them for that one commerce and whatever else I can squeeze out of them

[greggbert]

From my informal testing the major factor appears to be distance. You should pack your cities in near your capital and Forbidden palace even if 2 or 3 squares ovelap! You will do much better! Once you get more than 10 squares away from your capital, it seems to be more efficient to spread them out.

-Gregg

[solo]

Map size makes a difference, too. According to the strategy guide, the number of cities that can be built before introducing extra corruption depends on map size:

Tiny map - 8 cities
Small map - 12 cities
Standard map - 16 cities
Large map - 24 cities
Huge map - 32 cities

There may also be a difference due to paying level, as I've found corruption to be a much bigger problem on deity than on lower levels. In a tiny map deity game, I had a city that was totally corrupt only 4 tiles away from my capitol, but this was only after adding extra cities beyond the recommended 8. On a standard map game at a lower level this did not happen even though I had well over 20 cities. Instead it seemed that the distance from the capitol or forbidden palace was determining factor on the amount of corruption.

[TrainWreck20]

Kind of frustrating to build the Forbidden Palace, since common sense would say you want it near the opposite end from you original palace. However, corruption would make it take 300 turns, then.

I now have an army of workers planting and chopping forests, but still takes a while.

[sachmo71]

Originally posted by Fitz
I plan to study this and try and figure out a formula. My review of the guidebook that was pakaged didn't have one, so I may have to do this from scratch. I never played Civ2, so I don't know if there was a specified formula for corruption, but in Alpha Centauri, there certainly was for corruption and extra unhappiness, from both more cities and distance affects, so I'll see if it matches that in style.

If anyone knows a formula for corruption and unhappiness, be it Civ2, Civ1, SMAC, or somewhere in Civ3 that I've missed, please email it to me at junioros@hotmail.com

After I get Corruption and Unhappiness down, I'll move on to Pollution.

Sounds like a lot of work, but I would definately wait until the patch come out. It should be next week, but if not it won't be long after that. I know that they are working on corruption in the patch, and would hate to see all of your hard work invalidated. Just a thought...

[Fitz]

I'm not going to be able to get cracking on it for another week anyway. I suspect that they used a similar method to AC, if only because that saves them rewriting loads of code. In Alpha Centauri, corruption loss was due to distance (and government), and extra unhappy citizens was due to too many cities. The number of cities was determined by map size, government, and dificulty level.

On the easiest level, default government, city limits were (in AC):

Tiny map - 9 cities
Small map - 12 cities
Standard map - 16 cities
Large map - 17 cities
Huge map - 25 cities

and on the Hardest:

Tiny map - 3 cities
Small map - 4 cities
Standard map - 6 cities
Large map - 6 cities
Huge map - 9 cities

If Firaxis used the same formula, the diference between the Civ3 documented number of cities and my first set could be easily explained by one thing: Different numbers of squares for the map size names (ie Tiny isn't the same as it was in AC).

Corruption loss was a much more complicated formula, but dependent on distance, level, and government. Assuming that Despotism is the base government, and Chieftan the base difficulty, it may look like this:

Corruption loss = ((Commerce or Production) * Distance / 32)

Test it if you want. Set up a base 8 squares from the center. Give it 4 commerce. See if it loses more than 1 to corruption. I suspect it will, which implies that the number 32 has been changed.

First I'll confirm that the formula's still hold, then I'll calculate the variation in values. Shouldn't be too hard, if the same basic formula's were used. If not, who knows.

[Aeson]

I was running some tests on Corruption tonight, here is some of what I found.

First of all, the optimal # of cities on a map size is the most important factor. I found that on a tiny map, despotism, the 5th city (other than your capitol) will be completely corrupt until a courthouse is built. This is assuming that the other 4 cities are all closer to the capitol than the 5th. Once a courthouse is built, about 1/3rd of its production can be saved. After the 8th city, all others were completely corrupt even with a courthouse (again, assuming that they are built farther away than the previous 8). Building more cities doesn't seem to have much, if any, effect on previously built cities. The exception is if the previous city is farther away from the capitol than the newly built one. Then they just basically swap corruption.

As far as distance is concerned, diagnals count as 1.5 spaces (diagnals in relation to the city squares, not the screen). A city built with 2 diagnal spaces between it and the capitol will have the same corruption as a city built with 3 spaces to the side. It is the 1st and 2nd diagnal that take the penalty, and the 3rd is treated as a regular space.

When a courthouse is built, it seems to cut corruption by 1/3rd roughly. A courthouse helps a city not connected by roads a little bit more than a city that is connected by roads (about 2% corruption). Roads to the capitol reduce corruption by a small amount, 2-6%, though roads don't seem to make a difference if the city is founded with only 1 space between it and the capitol. Airports give the same bonus that roads do, but not if the roads bonus has already been applied. Of course once you have passed the optimal number of cities by 150%, all other cities will be totally corrupt in despotism.

There doesn't seem to be a difference between Monarchy, Republic, or Democracy in corruption due to distance from the capitol. I only ran tests for a couple of short distances under each government, they all came out exactly the same though. Despotism has a slight penalty for distance, again about 2% in all the tests I ran. Communism really differs though, as distance doesn't matter. Just having your palace in communism, about 1/4th of the shields will be lost. A courthouse will cut the corruption in the capitol by about half.

One very interesting thing I noticed. If in a high commerce city, very one sided tax/research/luxury funding results in extra lost commerce that isn't attributed to corruption. All of my cities had 175 Commerce, but at most could only produce 53 Tax revenue, even with only 35 being lost to corruption. If I increased research and luxuries, the "lost" portion would be added to them, without reducing the Tax revenue. Just something to keep an eye on when you have one of those very high commerce cities.

Anyways, a more in depth look is going to have to wait till after the patch (at least from me). I'll post some of the numbers here..

Deity, Commercial Civ (Indians)
note: terrain was edited to allow for more precise calculations and quicker expansion, desert was 25 food, 25 production, 25 commerce, with the whole map desert.

Despotism
(distance, corruption(out of 175 commerce), road reduction, courthouse reduction)

Palace 4, 0, 2
1 Space 35, 0, 9
2 Spaces 37, 2, 11
3 Spaces 41, 2, 13
4 Spaces 43, 2, 13
5 Spaces 47, 4, 15
6 Spaces 49, 4, 17
7 Spaces 53, 6, 19
8 Spaces 55, 5, 19
9 Spaces 59, 5, 21

All the distances were checked with only 1 city and the capitol. There is a modifier for number of cities, though it doesn't seem to change things too much as long as the optimum number of cities hasn't yet been reached. Corruption to Production (sheilds) and Corruption to Commerce (coins) were in all cases identical.

[Auren]

Thanks for great answers everyone, specially Aeson.

Seems that fewer cities that are better placed is lot better than many cities. I guess I have to start avoiding coastal cities which I used to build a lot. It is much better to have just land squares for cities, altough that means you can make far less cities.

Problem is of course that nowadays I don't build that many cities myself after all. In last game I made maybe 5 cities and conquered 20 made by computer.

Another interestign thing is that corruption is based on map size. So if you think corruption is bit too much playing maps with maximum amount of water can help to reduce it's effect a bit. (Less land -> Less cities -> less corruptiom)