DESIGN: Government City Caps

[Maquiladora]

Bah, look at some Civ 3 games where people played without ICS. A relatively small amount of cities, and upgrading them.

Is that a bad or good comment about civ3? Isnt the best strategy to build cities only a couple of tiles apart, constantly in civ3?

Yes, I'd love to have the choice of having a small or medium sized empire with developed cities appealing. But, I'd hate to have that choice pressed down upon me with caps or such stuff! I already hate how Civ 3 forces you to fight an Ancient war, whether you want it or not...

Well the caps will always do that, so whats the alternative to city caps? It would be cripplingly boring to have no caps...

Like Hex's idea, if we get rid of the cap we fill the hole with something equally crippling as unhappiness but not as EASILY curable as moving a slider and you can build more cities......... so it needs to be curable by buildings, maybe even tile imps, whatever.

I have 24 cities under Monarchy 20 city cap, crime is high in the far away cities so i garrison some more troops in every one of those far away cities, it was a difficult solution, but it worked. When those cities grow larger ill have the production to build a courthouse in those cities, or maybe i CANT even build a courthouse until the city is size 7? Interesting...

[The Big Mc]

75% of civ fans dont like future techs.
80-99% dont like stealth units, id agree with that except for the good ones (slaver spies).

they lack imagination

i hated the fact in civ2 that i got to the future age in 1580 when I was flying but then got caught up by the ai which eventually attacked and took several cities which they should not have got if I were able to keep the tech gap up

[hexagonian]

Originally posted by hexagonian
...but rather than increase the happiness penalty for exceeding the cap, we greatly increase the crime rate as you fall below 75, and have a more gradual riot chance. The end result are cities that produce very little, but are not in a riot state.

All of a sudden, crime rate becomes even more important to manage - almost just as important as unhappiness.

The more I think about it, the more I'm convinced that this is the way that I want to go regarding how happiness affects your civ.

Well, I was going through the numbers in govern.txt to see if I can affect the rate of crime as I fall into unhappiness.

The numbers

CrimeCoef x
CrimeOffset x

are the only relevant numbers regarding crime, but both numbers seem to only affect the baseline percentage of crime and not the rate that each happiness point changes the crime rate. Too bad, because I'm looking for control of the rate - and I want it to be more drastic than it currently is. (currently about 5% per happiness point - I'm looking for something along the lines of 15-20% per point)

Anyhow, a look though the rest of the files yielded no numbers that control this rate - so has anyone been able to locate the number in the source code?

...or have I missed something?

[Solver]

Hex, it looks like, under your idea, basically any city that is ready to fall into riot, would be losing most its production via crime anyway. Sounds pretty harsh to me.

In CtP, there are already more factors about each city to manage. Not only happiness, but also overcrowding has to be kept at bay - otherwise, cities get overcrowded, and at this point, things start to be looking bad.

[hexagonian]

Actually, in Cradle (as it is currently set up) your city has a great chance of falling into a riot if it falls to 74, which is a lot harsher than what I want. Basically, I want a margin of several happiness points of where crime becomes a greater factor without falling into a riot state.

What I want is a significant effect - the current Cradle setup is almost an all-or-nothing result.

Current Cradle setup
75 - 25% crime
74 - 30% crime/40% riot chance on highest level
73 - 35% crime/80% riot chance on highest level
72 - constant riot state

My preferred setup
75 - 25% crime
74 - 40% crime
73 - 55% crime
72 - 70% crime/10% riot chance
71 - 85% crime/20% riot chance
70 - 100% crime

In my preferred setup, crime is more important. But at the same time, your cities will still be producing something.

In the current setup, most likely, you lose all production capabilities at 73

[Peter Triggs]

Here's where crime is calculated:

double Happy::CalcCrime(CityData &cd, Player *p)

{ 
    double threshold = p->GetCrimeOffset();

    if (m_happiness > threshold) { //if city happiness is above the crime threshold
        m_crime = 0.0; //no crime
    } else { 
        // otherwise base_crime is the ammount that happiness is below the threshold:
        double base_crime = threshold - m_happiness; 
        double cops = cd.GetImprovementCrimeMod() - 
		(double)(wonderutil_GetDecreaseCrimePercentage(p->GetBuiltWonders()) / 100.0);
            // cops is a modifier for buildings and wonders

		double total_crime = 0.01 * base_crime;//total crime is 1% of base crime

		total_crime += cops * total_crime;// modify total crime by cops
		total_crime *= p->GetCrimeCoef(); // and again by the crime coefficient
        
            // m_crime is the final (percentage) value that we want:

		m_crime = min(1.0, total_crime);// so m_crime < 100 percent
		if(m_crime < 0) //and at least 0 percent
			m_crime = 0;
    }

    m_tracker->SetHappiness(HAPPY_REASON_CRIME, 0); // no longer used (?)

    return m_crime;
}

The easiest thing to do would be to expose that "0.01" as a parameter in Govern.txt as "CrimePerCent": "Percentage of base crime when a city is below the happiness threshold".

OTOH, maybe what you're after is a non-linear equation there; something like "0.001*base_crime*base_crime". So if a city is just a little unhappy, crime isn't significant. But as it gets more unhappy, crime goes up exponentially.

[J Bytheway]

Originally posted by Peter Triggs
OTOH, maybe what you're after is a non-linear equation there; something like "0.001*base_crime*base_crime". So if a city is just a little unhappy, crime isn't significant. But as it gets more unhappy, crime goes up exponentially.

Putting on my mathematician hat, I would like to point out that that would give quadratic growth, not exponential growth .

[hexagonian]

That would be ideal to transfer it to govern.txt. Actually, I want it to be a standard crime hit per hapiness point regardless if you are a little unhappy or a lot unhappy. What I would like to maintain is an importance in keeping your citizens happy as a priority. To me, the vanilla game made it too easy because there wasn't enough of a production/riot penalty between 72-75.

My goal is to make it a significant hit in those areas without the all-or-nothing effect that is now in place in Cradle. So a player and all of his cities could still function in that range, but at greatly-reduced values.

After thinking a little more, I would like to avoid having it work the other way - having a 15% reduction in crime for each happiness point over 75, because that would mean that at 77, all crime would be eliminated.

Perhaps a setting in govern.txt that works the same way as the slider setting. In Cradle, you have a (-3) happiness hit per click for a benefit, but only a (+2) bonus if you click in the opposite direction.

[child of Thor]

Hex i really like the idea of having a costly state before a revolt. A revolt should be the last word/means for an unhappy populace to express it's self - a slide into crime/tax evasion etc would, i would have thought, proceed such an event. So you get a last gasp chance to placate your city populace before they take things into their own hands.
One thing about this, maybe it would be good to make an actual revolt much more costly than it presently is. What i'm thinking is that it seems a little unnatural to keep having a city pop into and out of revolt over the course of say 3-10turns.
If we could make an actual revolt more rare but if it does happen it takes more than moving in a few units for a few turns - or suddenly giving the populace an extra half loaf of bread a day?
So an actual revolt becomes a really big deal - you get the warning and the bufferzone of increased losses to crime(as people try to work around the system) - but if you ignore it then you run the risk of maybe loseing the city for a longtime?

I'm not entirely sure - just kinda poped into my head

[Martin Gühmann]

One thing I have to note is, if people are unhappy they could work a little bit more slowly. At least I got the notion that this is the case in Cradle and it seems that it does matter how happy your citiciens are at least new advances came faster with a more happy population, maybe because of less crime.

-Martin

[Maquiladora]

I would rather combat crime with units and buildings than combat happiness with entertainers. Does anyone else find going around every city, every turn and adding and removing entertainers VERY boring?

[Martin Gühmann]

Originally posted by Maquiladora
I would rather combat crime with units and buildings than combat happiness with entertainers. Does anyone else find going around every city, every turn and adding and removing entertainers VERY boring?

I did this in CTP1 until I got the AI Entity then I tried to get rid of all the entertainers, what a pain if you have cities of size 80 and you have to turn 60 entertainers into workers of course per city. In CTP2 I don't take so much care on them but when a city riots I spend it some entertainers until it is happy and if there are to many cities rioting I do it for every city.

-Martin

[child of Thor]

Originally posted by Maquiladora
I would rather combat crime with units and buildings than combat happiness with entertainers. Does anyone else find going around every city, every turn and adding and removing entertainers VERY boring?

Yes! When i bother to use them! I mostly rely on my empire settings and buildings to keep crime and unhappiness in check - only rarely resorting to the specialists if i can help it, mostly in the later part of the game when any new cities need as much a boost as possible to grow.

[J Bytheway]

I'm hoping that we will be able to implement some simple automation of such things as toggling entertainers - I feelt that tasks such as that should be possible to automate relatively easily, and since we're going to have to do it for the AI players anyway, why not allow the human players the chance to use the same code.

[MrBaggins]

So.. actual implementation of this auto-entertainer setting?

I'd suggest a checkbox "Enable Auto Entertainers" and a "Target Happiness" value (with spinners perhaps.)

You should, for expediencies sake, be able to set this up globally too, with, perhaps a button to an additional dialog, maybe, from Empire Manager. The new dialog would set the Auto Entertainers setting for all the cities.