Corruption

[therother]

Just added a table with the positions of all the settlements. This should help you to calculate what proportion of your gross income will be deducted for corruption.

[avatar]

Does garrison size affect corruption?

[therother]

Does garrison size affect corruption?

Not that I could see.

[sunsmountain]

Corruption is easy to understand, and linear.

Distance to capital and total/gross income are the ONLY causes of corruption.
Corruption is a percentage of total trade, or Gross Income.
* Example: If you have 1000 corruption in a town at distance 40%, you will have 2000 corruption for that same town at distance 80%.

The distance is rounded, so you can calculate the exact number of squares, but the 'rough estimate' the game provides is accurate enough as it is.

Law is the ONLY solution to corruption.
Adding a law bonus reduces corruption by that percentage.
* Example: If a town is suffering 1000 corruption, adding 10% law reduces this to 900 corruption.

Governors can develop virtues (or vices, unfortunately) that increase or decrease corruption, by adding or removing law bonuses. Prerequisites, from what ive experienced so far (not tested): High distance to capital (70%+), high trade.

[Simetrical]

If you have 1000 corruption in a town at distance 40%, you will have 2000 corruption for that same town at distance 80%.

Close, but not quite. The connection is indeed linear, but you didn't account for either the offset ("grace distance") or cap. Corruption at 40 squares is not double the corruption at 20 squares—it's almost three times that, because the grace distance is approximately 15.88 squares. In other words, the proportion is roughly (40–15.88)/(20–15.88), or 5.85678, despite the fact that you only doubled the distance.

Adding a law bonus reduces corruption by that percentage.

This assertion stands in direct opposition to zhuge's earlier statement that "Law temples (ie Athena) reduce corruption by (0.03)([Gross Income])," although I agree that the latter isn't that well worded (presumably it means per point). Yours also has an attractive simplicity to it, and it seems more intuitive that law reduces corruption by a percentage of corruption rather than a percentage of gross income. However, I must admit that I'm biased somewhat to accept what zhuge says over what you say. Could the two of you perhaps supply the methodology you used to establish your respective claims?

-Simetrical

[zhuge]

This assertion stands in direct opposition to zhuge's earlier statement that "Law temples (ie Athena) reduce corruption by (0.03)([Gross Income])," although I agree that the latter isn't that well worded (presumably it means per point).

Thanks for the reference. I admit to being rather sloppy in my wording but (0.03)(Gross Income) for each level of temple is correct, AFAICT, or should I say each point of law (since generals can add to this as well, by having traits like Drillmaster, Superb Administrator, etc)

However I must make 1 small amendment. It appears that the Admin bonus isn't factored into the Gross Income, so Gross Income here really just means Trade + Taxes + Farms + Mines.

Methodology:
Mouse over Corruption icon in Settlement Details scroll and record Corruption values, before and after demolishing Law temples and before and after moving generals with Law traits in and out of the city. Gross Income calculated by totalling all income sources (mousing over respective icons in Settlement Details scroll). Based on saved files from Greek (turn 17) and Julii (turn 34) campaigns.

Results:

		GrossIncome(GI)	Corruption	Law level		Corruption Difference/GI
Nicomedia	1300		299		1 (shrine Athena)	39/1300 = 0.030000
				338		0 (razed shrine)

Lilybaeum	2259		225		2 (temple Athena)	136/2259 = 0.060204
				361		0 (razed temple)

Pergamum	3073		368		2 (inprogress, blinking)185/3073 = 0.060202
				460		1 (shrine Athena)	
				553		0 (razed)

Domus Dulcis D	798		271		1 (shrine Jupiter)	24/798 = 0.030075
				295		0 (razed)

Damascus	1378(withAdmin)	664		3 (governor with law 3)	112/1253 = 0.089385	
		1253		776		0 (no governor)
		governor has Management 5, supplying an extra 125 Admin income and is a Superb Administrator (Law +3)
			 
Thapsus		1919 		575		2 (temple Athena)	115/1919 = 0.059927
				460		0 (razed)

Thapsuswithgov	1856(withAdmin)	382 		3 (templeAthena+trait)	109/1820 = 0.059890
				491		1 (law trait only)
Thapsuswithgov	1820
excluding Admin
		governor has Management 1, supplying an extra 36 Admin income, is a Useless Assessor (-10% Tax), reducing tax from 982 to 883 and also a Drillmaster (Law +1)

Discussion:
Taking Lilybaeum above as an example, if temples were to have a 10% deduction on Corruption then Corruption should drop from 361 (without a temple) to 325 (L2 temple) or by 36 (roughly 10% of 361). As we can see, it actually drops much more to 225.
The reduction in corruption is therefore (0.03)(Gross Income) per point of Law (with each temple level giving 1 point) as can be seen from the data with Gross Income not including the Admin bonus in this case.

Comments are most welcome.

[Lochar]

I didnt see anyone mention it but it seems a garrison reduces or affects corruption as well.

Last night I bribed an out of the way rebel town and took it over. I happened to have the city detail window up and was buying peasants, after about 5 units I noticed my corruption dropped a bit, it went from 239 to about 200.

I always use my detail screen as for some reason even tho you dont actually have a unit made or structure built the detail screen will still reflect future puchases.

As far as how much garrison affects it, I dont know, I am not smart enough to figure that out..:)

[Simetrical]

I expect that the effect was due to the reduction in taxes and trade. Remember, corruption is a percentage of gross income.

-Simetrical

[sunsmountain]

I agree, simmetrical, that was probably not because of the garrison, but due to decreasing income from tax.

In this post i will make some rules of thumb based on what therother found, yet with % distance to capital instead of squares (even though squares to capital is more accurate).

(Corduba)
Distance to capital: 65%
Corruption: 2194
Gross Income: 4571
% corruption: 48,0%
Amount of law required to eliminate corruption: 80%

(Syracuse)
Distance to capital: 20%
Corruption: 320
Gross Income: 2286
% corruption: 14,0%
Amount of law required to eliminate corruption: 25% (actually, 23,3%)

With distance to capital i mean the % shown on settlement details, NOT the exact number of squares, because this is reduced by grace distance for all cities. Say distance B = A + grace. Distance C (double B) is then 2*A + grace. Distance to capital is used for simplicity's sake.

I mean you're not going to count squares every city every game. This post is for those who dont want to. 'Rules of thumb' stand next to 'True Formulaes'.

Close, but not quite. The connection is indeed linear, but you didn't account for either the offset ("grace distance") or cap. Corruption at 40 squares is not double the corruption at 20 squares—it's almost three times that, because the grace distance is approximately 15.88 squares. In other words, the proportion is roughly (40–15.88)/(20–15.88), or 5.85678, despite the fact that you only doubled the distance.

Corruption at 40 squares (15% distance to capital) is therefore 3 times the corruption at 20 squares (5% distance to capital). The point is that you can use distance to capital, which is more readily available than 'number of squares', which you have to calculate using s2 = x2 + y2, a bit labor intensive.

My previous statement:

Adding a law bonus reduces corruption by that percentage.

should be modified by 0.6. Ie 5% law reduces corruption by 3%, not 5%. I agree with the observations made by others, this is a good 1st rule of thumb.

The % corrupion seems to be about 0.73 times the distance to capital (not squares, don't confuse this with the 0.007935 factor therother found, though there is probably a relation).
Amount of law needed to eliminate corruption is the % corruption over gross income divide by 3 times 5. You will therefore never need more than 108.3% law, if that is at all possible.

maximum distance and minimum corruption at 80% allow us to calculate a more accurate factor than 0.73:
corruption at distance 80% = 60% of gross income
[corruption caps at 65%.]
maximum distance to capital = 80%.

The factor is therefore
%corruption = 0.75*%distance to capital.

2nd rule of thumb. This almost fits all the data. Corruption seems to go on beyond 80%, even though that should be the maximum, by a too large amount to be attributed solely to rounding.

Summary:

rule 1 (known):
every 5% law eliminates 3% corruption
corollary:
100% law (probably maximum) may remove 60% corrupion at 80% distance to capital.

rule 2 (roughly):
% corruption = 0.75 * %distance to capital.

[MackBolan]

I know I cannot be the only one this has occured to. In my Brutii campain i have a kingdom that spans from Southern Italy to Alex/Memphis/Thebes and mostly everything in between save for Pontus and Far East factions (parthia, Armenia, etc). Anyways, as I made my way down the coast I took Antioch. Suddenly my poor city had been bombared with Spys from Pontus, Selucids and Egypt. My revenue in Antioch was dramatically low. Lower than Tarsus, Sidon, Damascus and Jerusalem. Im talkin -800 give or take. Now I had moved my capital to Athens when i began moving on Turkey and then down to Africa. So i guess my question would be, does anyone know the Coruption that Spies cause? How much does it cause and how can I get rid of it. I know it HAS to be the Spies because all the neighboring cities are making more than Antioch and a few of them have close populations. A few within a few thousand. However, Antioch was being Spyed on CONSTANTLY and I would keep finding them. They're just little hooded, dagger bearing, fleas.

[sunsmountain]

Tempting though it may be to blame the spies, the buggers cause unrest (and 15% at least for each faction - not sure for each spy), not corruption. Which would mean you would have to lower the tax rate of Antioch, which would mean Antioch wouldn't 'pull its weight' which means negative income in that city.

Contribution of each city to income is weighed according to total income. If cities are performing below par, they show up on the map as having negative income.

Now if you look at those cities that have a distance to capital of 80%, they are usually the ones with Low Taxes and therefore negative income. To counter the situation, move your capital to the center of your empire (check with your finances tab to get the lowest amount of loss due to corruption). So try out a few cities and see which works best.

Ports, roads, traders do the rest. Make sure you dont use big garrisons for big cities, use Games, Races, & lower the growth rate (dont build farms). If its too late for that: Exterminate.

[MackBolan]

Just outta curiosity, lets say you have the whole RTW world, every faction assimilated into the Roman way of life, and every settlement captured (i know someone has done it on like, VH/VH), is the game designed for Rome to be the best capital, it is pretty much in the center of the campaign map? If not Rome, Id say Messana, or better yet Syracuse would be the best Capital. Anyone have any thoughts?