1. We have added a Gift Upgrades feature that allows you to gift an account upgrade to another member, just in time for the holiday season. You can see the gift option when going to the Account Upgrades screen, or on any user profile screen.
    Dismiss Notice

The Pursuit of Unhappiness

Discussion in 'Civ - Ideas & Suggestions' started by CGG1066, Dec 9, 2010.

  1. CGG1066

    CGG1066 Minister of Finance

    Joined:
    Oct 30, 2010
    Messages:
    147
    Location:
    California
    So - people have been wanting rebellions and revolutions, and all that; or complaining about civilization-wide unhappiness. However, I think the largest problem is that the game incentivizes ignoring happiness unless you're near one of two thresholds 0 :c5happy: or 10 :c5angry: - which is really odd because happiness has replace health/food(to some degree)/city maintenance/corruption as the limiting factor to just about all aspects of the game. Since decisions are made on the margin - why would I try to improve happiness if it doesn't help me with any of these thresholds. In thinking about how to solve this problem, I think I've reached a simple solution that - in all cases - gives value to additional happiness, as well as gives some other commentators what they want.

    (As always - numbers are gut feelings of what would work, if/when this idea is ever implemented, they will probably need adjusting.)

    Unhappiness causes unemployment

    For each 1 unit of :c5unhappy: between 1-10 :c5unhappy:, 2% of the population is locked in as an unemployed citizen. The first citizen that becomes unemployed is the least productive citizen (as measured by the sum of :c5food: + :c5production: + :c5gold:) in the entire civilization, the second citizen which becomes unemployed is the second least productive citizen, etc.

    This means that if you have 1 :c5unhappy:, you probably wouldn't even notice - as less than 2% of your production goes away (you only lose the least productive citizens, so - really at lot less than 2%). But once you have 7 or so :c5unhappy:, you're looking at a 10% or so loss of resources produced per turn . . . and that begins to hurt. With each additional :c5unhappy:, the losses also increase, as more productive citizens become unemployed.

    (Yes, a player could reassign a citizen from a more productive tile to the tile the unemployed citizen was working if they want, but the city will still have an unemployed citizen "locked in").

    This makes happiness a bit self correcting, unemployment leads to less food production, which can lead to starvation in marginal cities. (other mechanisms could also fit very well here - like killmeplease's emigration mod).

    Growing discontent

    For each unit of :c5unhappy: 10 or greater, the same 2% rule applies, but your army also gets a -5% combat bonus. Once again - that could be important right away, but the real sting comes when these penalties add up.

    These two ideas bring in the marginal decision making that solve the worse problem (which was my original goal). But if you want rebellions/revolutions - well you know you can count me . . . in!:

    All politics is local

    When 34% of the city's citizens becomes unemployed, partisans appear! (notice how this brings a local aspect to happiness. Since unemployment is based on the relative productivity or citizens, it is possible - and I would guess probable - that unproductive citizens are in the same city, or group of cities. So even though unhappiness levels are civilization wide - extreme unhappiness consequences can be localized).

    When in the course of human events . . .

    When 51% of a city's population becomes unemployed, it goes into disorder :c5razing:. After 5 turns of disorder it breaks away. If it is a puppet or a city annexed from another civ, control reverts back to the previous owner. If it is not, then a new civilization not previously in the game takes control (and declares war on you, naturally).

    Your capital will never revolt, even if 100% of it's citizens are unemployed. (Although your citizens would be starving and you would have lost all your other cities)
     
  2. Camikaze

    Camikaze Administrator Administrator

    Joined:
    Dec 27, 2008
    Messages:
    27,312
    Location:
    Sydney
    I certainly agree that graduating penalties are a great idea (and such an obviously good one that I'm left to wonder why on earth they weren't implemented at all). But I'm not entirely sure about this 'unemployed citizens' suggestion. It seems a bit of a roundabout way to implement an effect for each individual graduation along the scale. The exponential effect implicit within this idea is a good one (although a lot of citizens would be equally productive, so the progression would be largely linear). But why not just impact on productivity in general, rather than have all but 2% of citizens impacted on at all?

    Also, I'm not entirely sure about the combat penalty. I'm not sure it logically follows as a consequence of poor happiness.
     
  3. CGG1066

    CGG1066 Minister of Finance

    Joined:
    Oct 30, 2010
    Messages:
    147
    Location:
    California
    To take on your points out of order:

    Well poor morale, for one. But that aside, the combat penalty is in the game now (-33% with 10 :c5angry:) for good reason - you're likely to get all that unhappiness after/during conquest, so it is a check on military expansion once your army gets the penalty and is no longer competitive (whether or not the -33% makes a difference, well that depends on other factors).

    That's a valid point, though I think the unemployment idea adds a bit of depth to it. First off - I'm not so sure it's round-about, the connection seems straight forward as does the logic behind it - to me at least. Unhappy citizens aren't going to work - so they don't. To me, just throwing in a penalty is more round-about; it's an unnecessary abstraction.

    Regardless of that - the main reason is game-play because it creates options. If a city goes into starvation, disorder, or you really need to that one (or handful) citizen's production, there are several things you can do, even if you cannot raise happiness. You can try to shuffle work assignments around, move specialists to tiles, etc. What makes management fun is when you're hit with moments of "oh - !@#$, what do I do!" and you have options to make a solution, rather than just one path.

    Secondly, using unemployment rather than an incremental penalty also creates a localized effect to unhappiness - not an even uniform empire malaise. This makes the rebellion and revolutions possible without relying on randomness or arbitrary criteria.

    Finally (and least of all), for "flavor" purposes, I think seeing the :c5citizen: icon in red sort of hammers home the point that citizens are suffering, rather than just seeing a number's penalty.
     
  4. killmeplease

    killmeplease Mk Z on Steam

    Joined:
    Nov 22, 2007
    Messages:
    2,794
    Location:
    Samara
    i propose quite large happiness rework:
    1. introduce dark age as the opposite to the golden age.
    2. symmetric levels of happiness/unhappiness:
    2.1. (-1..+1): content
    2.2. (-5..-1): displeased, (+1..+5): pleased
    2.3. (-10..-5): unhappy, (+5..+10): happy
    2.4. (-25..-10): very unhappy, (+10..+25): very happy
    2.5. -25: instant dark age, +25: instant golden age
    3. scale happiness with number of cities. e.g. if you have 2 cities, content level will be at -2..+2 (happiness modifier should be equal to some function f(numCities))

    what should be levels effects?
    i think it should be some smoothly increasing effects.
    like persentage of something, e.g. happiness*X%, with +-25*X% at higher levels. plus some bonuces for hitting certain levels.

    gradual effects:
    confedience: population growth rate += happiness*1%
    corruption: waste of gold = happiness*3% if happiness < 0
    enthusiasm: bonus to production = happiness*2% if happiness > 0

    level effects:
    content: enemy units can not heal in player's lands without a medic .
    pleased: enemy units in city radius are being crippled (partisans).
    happy: citizens help defend cities by providing levy units (from earlier era, or from same era if armory/arsenal is built) (patriotism).
    displeased: enemy units can heal in player's territory.
    unhappy: citizens occasionally cripple garrison/city defences, destroying buildings, kill population
    very unhappy: citizens spawn rebels units / instantly capture cities (if there are no garrison)

    this system was developed for revolutions mod (http://forums.civfanatics.com/showpost.php?p=9788159&postcount=32), and it was intended that those effects will be conditioned by empire stability measure (+city loyality). I think we can use happiness as indicator of empire stability, not an unnecessary additional parameter.

    ps: there are some ideas also on how to implement local (in)stability effects by taking various factors into account such as culture, war success, etc
    i think this system can be reworked to utilize global happiness as a general stability factor.
     
  5. killmeplease

    killmeplease Mk Z on Steam

    Joined:
    Nov 22, 2007
    Messages:
    2,794
    Location:
    Samara
    on unemployed,
    i always thought that one becomes unhappy because of unemployment, not otherwise :D
     
  6. CGG1066

    CGG1066 Minister of Finance

    Joined:
    Oct 30, 2010
    Messages:
    147
    Location:
    California
    First off - your point #3 is right on. There is definitely a bit of a scaling issue because the happiness numbers in civV don't correspond to the overall population number. In way this is good because things like luxury resources have a smaller effect (percentage wise) on large empires than on small ones. I compensated for this by having the penalty be a % rather than a number. The problem with re-scaling the happiness level based on population all the time is it can get confusing as population changes.

    I've thought about a "dark age" or "anarchy bucket" idea (in fact, I originally conceived my post to go into that thread), but - thinking about it - there were a lot of issues that needed to be dealt with. Say you were near a golden age - would you first have to 0 out that counter? If so, it seems subject to abuse (you could draw up a pretty large happiness bank and act with impunity there after). If not - well then theoretically you could be 1 turn away from both a "dark age" and a golden age at the same time.

    Another thing that confuses me - are your "confidence" and "enthusiastic" bonuses supposed to replace "we love the king day" and golden ages respectively? If not, aren't they sort of redundant? The game already rewards excess happiness.
     
  7. killmeplease

    killmeplease Mk Z on Steam

    Joined:
    Nov 22, 2007
    Messages:
    2,794
    Location:
    Samara
    confusing - why?

    see no problem here. you can do such a things with gold, no one complained about it
    ps maybe GA/DA points should decrease with greater [and increasing] speed than they are being accumulated:



    right.. but not quite
    my post is confusing as i have not devised yet how to move stability system to the happiness basis.
    GA/DA are empire-level effects, while corruption etc supposed to be city-level.
    some interference with WLTK tough.
     
  8. CGG1066

    CGG1066 Minister of Finance

    Joined:
    Oct 30, 2010
    Messages:
    147
    Location:
    California
    Yeah - I'm not sure about a stability system - it seems like it could get complicated. that alone is probably worth its own thread.
     
  9. Camikaze

    Camikaze Administrator Administrator

    Joined:
    Dec 27, 2008
    Messages:
    27,312
    Location:
    Sydney
    @killmeplease- could you please explain that graph?

    Hmm, that kinda makes sense, although I would much prefer a morale mechanic independent, or at least semi-independent, of happiness.

    I see what you're saying here. I guess I was thinking it roundabout in that there is an extra step in the process between cause and effect. Unhappiness causes unemployment, which causes a penalty to production/food/gold, rather than simply having unhappiness directly causing such a penalty. I guess it does, as you say later on, make the effect more tangible for the player

    These all seem quite fair points. You're making a good case for it.

    I'm a little confused as to the local effects. So you propose that partisans appear when unemployment hits 34%. This is certainly a local effect. But if you cannot directly control what citizens in your empire become unemployed, it's not a local cause. Both the cause and effect have to be local for it be a viable localised happiness mechanic.
     
  10. killmeplease

    killmeplease Mk Z on Steam

    Joined:
    Nov 22, 2007
    Messages:
    2,794
    Location:
    Samara
    H is current happiness
    acc is an acceleration multiplier for happiness translating to GA/DA points.
    pts is GA/DA points (-x = DA, +x = GA points)
    when happiness crosses the 0 level (was negative, became positive, or otherwise), accumulated DA/GA points start to decrease. On each turn when GA/DA points are decreasing, the speed of decreasing (acc) is rising.

    on the graph, we start with -75 pts (these are Dark Age points), and our happiness = 3.
    On turn 2 we have -72 pts (-75+3*1). Happiness has changed (by whatever reason) to 2, and acc has risen to 2. it will rise while happiness and GA/DA points accumulated have opposite tendencies :)) and DA or :( and GA). otherwise it will drop to 1 and stay there.
    On turn 3 we have -68 pts = -72+2*2
    and so on

    So if your happiness was say -5 for 100 turns and you got to -500 DA points, but then your happiness increased and its value now is +5, you have not to wait for another 100 turns to burn those 500 DA points. it will take much less.

    You can check the certain formulas in the spreadsheet:
     

    Attached Files:

  11. Camikaze

    Camikaze Administrator Administrator

    Joined:
    Dec 27, 2008
    Messages:
    27,312
    Location:
    Sydney
    Hmm, I *think* I get it. Seems awfully complicated for a golden age/dark age system. How intuitive do you think it would be?
     
  12. CGG1066

    CGG1066 Minister of Finance

    Joined:
    Oct 30, 2010
    Messages:
    147
    Location:
    California
    Ok - I see it now (I didn't check the formulas, but I see how an acceleration factor can solve the issue I brought up earlier). I'm still cautious about the bucket idea for the following reasons:

    - As Camikaze points out - it's not really accessible. Sure - players like you and me might run the numbers and know what's going on. But for most players, it will be a mystry on why they are approaching a dark age/golden age as fast as they are, and have no idea when it they're going to get one or not. (Anecdotally, when I did regulatory consulting, some regulatory assessments had a similar adjustment factor to them - and many companies - even their CFOs - just never understood them, even though a lot of money was at stake! :sad: I would not expect any more from players in civV)

    - I don't really like the idea of a "dark age" because the player needs an immediate reward to solving a happiness problem. Otherwise, it's most likely the'll just reload or abandon the game. Sure - this isn't symmetric to the "golden age idea" but a player's response to good things/bad things happening is asymmetric.

    - I'm not sure it's necessary - you can tweak it all you want to address these concerns, but the mechanic I suggest in the OP avoids all the problems.​

    I assume by "viable" you mean in terms of realism? Short of scrapping empire happiness (which I'm against because it gets rid of the perpetual civil disorder problem) or creating some other mechanic (which I'm also against because it will just create future complications) - you're not going to get the causes of rebellions/revolutions to become local. I'm not too concerned about this - and it actually doesn't strike me as unrealistic (for example, the riots at the 1968 Democratic convention had a lot more to do with national politics than what was going on in Chicago).

    If by "viable", you mean something that can be exploited gameplay wise - which I would be more concerned about - then I don't see it.
     
  13. Camikaze

    Camikaze Administrator Administrator

    Joined:
    Dec 27, 2008
    Messages:
    27,312
    Location:
    Sydney
    By 'viable' I mean 'workable within the constraints of realism and gameplay'. I think if the effect is going to be quite localised, then the cause should be localised, for both the sake of realism (in that it doesn't make much sense to have localised effects dependent entirely upon global causes; sure, there are exceptions, as you point out, but if your one city with crappy lands gets the pointy end of the stick just because it has crappy land and the rest of the empire has happiness issues, that doesn't make too much sense) and gameplay (in that it isn't entirely good for gameplay if effects don't directly translate from their specified causes). That is why I would suggest either tweaking the happiness system so as to make it a hybrid of global and local causes and effects, or implementing a separate local happiness feature that deals specifically with local causes and local effects.
     
  14. CGG1066

    CGG1066 Minister of Finance

    Joined:
    Oct 30, 2010
    Messages:
    147
    Location:
    California
    I'll try to break it down a little more, and hopefully this will address your points:

    Realism: There seems to be two components of realism which you're concerned about (a) the logic behind why "crappy lands" become the focus of civilization wide-unhappiness and (b) a possible apparent "exploit" where a city is on the verge of disorder can be brought back in line buy building a theater the other side of the world.

    (a) I think there is a deep logical connection between "crappy lands" and unemployment/unhappiness, which you might not see. Since these citizens produce less, their personal income would be much less than their fellow citizens, and they are not going to be able to afford luxury resources (or tickets for sporting events/theater shows) which is an important element for civilization happiness. If these citizens happen to be geographically clustered, well that's where you get your problem.

    (b) I'm not too concerned with this. It might not be the most realistic thing in the world, but it isn't the most unrealistic thing in Civ either. This is always going to be a problem with civilization-wide happiness; and local happiness had the more serve problem of causing cities in perpetual civil disorder, with no broader punishment.

    Game-Play: I think you're forgetting that losing a city to revolt or having to clear partisans really isn't a "local punishment", but instead meant to punish a civilization that has failed to manage its happiness. The punishment is losing a city. The mechanic I propose is a method of selecting which city the civilization loses. There are other alternative methods out there, but most of them seem arbitrary or random.

    You could - as you suggest create other mechanics that solve this issue; run two happiness mechanics instead of one (one local, one civ wide), etc. - but that strikes me as redundant and would really clutter up the game, all in the name of perfect symmetry. And I think you lose more there than you gain. Remember, if there is perfect symmetry between local/global causes and effects, a player would not be penalized for local unhappiness from military conquest - he could just steam-roll over leaving conquered cities unhappy and unproductive, with no civ-wide problems created.
     
  15. alpaca

    alpaca King of Ungulates

    Joined:
    Aug 3, 2006
    Messages:
    2,322
    You can see this throughout the whole game, everywhere: The discontinous era effects that unlock whole policy branches when you research the first techs, Great Scientists yielding free techs instead of a number of beakers, happiness thresholds instead of gradual effects, and so on. I think this is one of the ways in which the devs tried to simplify the game as it's easier to remember a large and discontinuous effect.

    Unfortunately, discontinous effects almost always distort gameplay - if you introduce a threshold, the optimal play is, if it's negative, to get as close to the threshold if possible without hitting it, and if it's positive, to beat it as quickly as possible and then focus on other things. There is a large amount of scientific research that was conducted on this topic in behavioural psychology, and it seems like this kind of reward will always produce this kind of behaviour in any animal that's highly developed enough to learn. Here's an interesting article on the subject relating to gaming, although I can't vouchsafe for scientific correctness because I haven't read into behavioural psychology myself: http://www.gamasutra.com/view/feature/3085/behavioral_game_design.php?page=1

    Note that having some of these ratio schedules is probably not such a bad idea because the level of activity associated with it can be quite fun (ever noticed how you frantically start building happiness buildings when you think you will get below -10 soon?). But Civ5 has a few too many, in my opinion
     
  16. CGG1066

    CGG1066 Minister of Finance

    Joined:
    Oct 30, 2010
    Messages:
    147
    Location:
    California
    I haven't read the latest behavioral psychology research either, but if one just cracked open their old econ 101 textbook, the result should be obvious. Remember Marginal Cost (MC) = Marginal Benefit (MB)?

    That's an interesting article - though I'm disappointed they just focused on behavior psychology rather than the economic aspects of it. - sigh -

    I agree you can't make smooth marginal changes everywhere - but the activity has to be fun. The spurt of building coliseums at -9 :c5unhappy: or 0 :c5happy: (depending what type of game I'm playing) isn't that fun. The spurt of activity when a peace treaty expires can be quite fun . . .
     
  17. Camikaze

    Camikaze Administrator Administrator

    Joined:
    Dec 27, 2008
    Messages:
    27,312
    Location:
    Sydney
    That does make sense to an extent, but I don't think the correlation is strong enough to support tiles with bad yields always missing out. I mean, if we take this slightly OT, it's entirely realistic for densely populated areas to have a lower quality of life than small cities with what would translate to 'horrible tiles' (and remember we're not just talking about modern times here). Big cities are not going to be punished by this, even though that is where a lot of discontent should realistically lie.

    While this is true, I don't think 'well, it could be worse' is an entirely good justification for a feature in the game. I think if you clearly separate local cause and effect from global cause and effect, then you're going a long way to rectifying this problem.

    It may be that 'crappy lands' are in fact a local cause of unhappiness (and unemployment under an expanded system), but attaching a global cause to a local effect (unemployment in a specific city) confuses the issue.

    Oh I don't particularly mind the partisans & revolt as it relates to penalising a civilization as a whole. My problem in the translation from global cause to local effect is the unemployment on tiles. The partisans which then result from this unemployment are a hybrid of global and local, and have global and local effects themselves, so that's not particularly the issue. What is the issue is unhappiness from capturing a new city resulting solely in a rise in unemployment in one far off and relatively unrelated city.

    Oh, I don't think that would be the case. They'd be two clearly distinct features, both nicely outlined. It's just like if you added in health as a factor. It's only really redundant if it does the same thing. If it's designed to have different effects dependent on different causes, then it makes sense.

    Other balance issues could occur, but these could quite easily be sorted. In the instance you suggest, I would assume there would be both local unhappiness in the conquered cities, and global unhappiness due to some sort of 'war weariness'.

    Couldn't have said it better myself. :)

    I've been quietly (well, loudly within this forum, but I don't think that means much) campaigning for graduating penalties (whether they be exponential or liner) for quite a while, so it's good to see that someone else agrees. :goodjob:
     
  18. CGG1066

    CGG1066 Minister of Finance

    Joined:
    Oct 30, 2010
    Messages:
    147
    Location:
    California
    I see what you're saying, I just don't like the "oh - well let's create another mechanic for that" mindset. There are lots of reasons why citizens become unhappy, most of which cannot be modeled into the game in a fun way. Even if you're able to keep all the mechanics simple, the more of them you have, the more cluttered the games becomes.
     
  19. Camikaze

    Camikaze Administrator Administrator

    Joined:
    Dec 27, 2008
    Messages:
    27,312
    Location:
    Sydney
    Eh, I don't particularly like adding features just for the sake of adding features either (it would tend to make lots of existing features semi-redundant and clutter the game), but I do think that splitting up happiness into local and global is neither cluttering the game or making anything redundant. It would be quite a positive change in its own right.
     
  20. CGG1066

    CGG1066 Minister of Finance

    Joined:
    Oct 30, 2010
    Messages:
    147
    Location:
    California
    I haven't been to the forums at all in a long time (nor have I really played CivV) - but I happened upon the site by near accident today and saw that this thread was discussed on Polycast! So I'd thought I'll post a shameless plug here: http://forums.civfanatics.com/showthread.php?t=415025

    Some points of clarification of some ambiguities:

    - As imagined in the OP, the 2% rule applies on a civilization wide bases, not 2% unemployed per city. This means that you don't only have to watch your overall happiness level, but also how it is distributed. You can have a relatively happy civilization, but if all the unproductive citizens are concentrated in one city (e.g., because it hasn't been improved yet) - you will have a problem in that city.

    (I actually view this as a simple rule that creates complex results. Like in real life, a rebellion can be either based on geography, like the American revolution, or occur nation wide at the same time, like the French revolution. Depending on the distribution of your population, one is more likely than the other. Since unhappiness effects different civs differently, there is higher replayability and no formula for a one-size fits all solution).

    - I probably used the word "production" with a bit of imprecision: I don't mean the loss of just :c5production:, but rather the loss of the sum of :)c5production: + :c5food: + :c5gold:) i.e., all economic products produced by citizens (this makes sense because the loss comes from not allowing a citizen to work a tile). That said, I agree that the game is too light on :c5production:, but the fix for that is really both simple and obvious, and I'm disappointed that 2k hasn't just scaled down :c5production: requirements.

    - I would like avoid comparing the "civil disorder" bit from that of pasts civs. In older versions of the game, there were basically cities that were caught in a trap in perpetual disorder. That is impossible here - disorder only acts as (a) a notification that a city has an unemployment problem and (b) a buffer so that you have time to address the problem. If the problem isn't solved in the set number of turns, it WILL break away.
     

Share This Page