I play on Prince. I've never been in negative happiness past early game. Ever. I average around 100 happiness late game. The ability to grab wonders easier and invest in a much smaller army in lower difficulties (I always have the lowest military score when I'm not warmongering) already makes happiness management not as difficult as higher difficulties.
December Balance Beta - December 3rd (12/3)
- Thread starter Gazebo
- Start date
- Status
Then play india in higher difficulty or China with lots of cities. I played India with 7 cities, my capital was reaching near 40 Citizens, my other cities were around 12 to 20. In every citiy, my capital included, I have got the warning, that further growth would increase my unhappiness by at least 8. So my Civ stagnated popwise and I quitted the game after several rounds, because nothing changed, whatever I have build.
[/QUOTE]in Endless Space you have Approval penalty from Overexpansioning.
On paper this looks right, but you need to look at opportunity costs closely.
First of all, there is a lot less positive happiness in VP (to offset the overall lower unhappiness).
When considering growing in VP, for a city that is slightly above average in growth, you can easily pick up a lot more than 1 unhappiness. So if you are comparing 29 pop cities to 30 or 31 pop cities, the marginal difference can actually be quite a bit bigger than in vanilla, its easily 5 or more unhappiness for those, which would require 2 luxuries to offset (compare to vanilla, where you need half of a luxury).
Not saying I agree with Bite, but for once, his math isn't completely off. I'm really starting to agree that Fealty is bad, and its partially because food just isn't too valuable. I personally think the problem is growing is too easy even without food tiles, so by medieval era I'm not investing heavily in food.
BiteInTheMark
Deity
Yes, but it's one value, one mechanic, and you are able to predict relative easily the outcome of actions to it. Like it was in vanilla civ 5. I don't want to say we need such a easy mechanic. But unless you stop all your growth in cities, and even then, It's impossible to tell how much happiness you will have in 10 turns.
To Gazebo. It isn't subjective, yields per pop decrease with more pop. That's simple math and easy to understand for everyone. And a self regulating punishment which is fine.
But a population modifier and also the tech median modifier are both hidden in the system. It's integrated and interacting with each other (else the tech median penalty wouldn't increase with more pop). And even the need reduction buildings have numbers (-75%...), the resulting impact on the modifier is also weird (a 75% reduction modifier decreases the threshold modifier by around 22% in a city with around 35 pop, while an increase of a tech lead from 3 to 4 increase the modifier by around 16%). But for a normal gamer it's impossible to know or predict that.
And the population modifier Is a unnecessary additional punishment.
You already paying more and more food to get an additional citizen, making the investment more and more less efficient.
Your yields per pop will definitely decrease more and more with every pop, causing already unhappiness. Those both mechanics are already there and you see the source and the result. Why do we need additional, artificial modifiers which can't be predicted by normal human players?
I can't follow all detail of the discussion, but to be honest i'm not a fan of current happiness system too. Gameplay-wise i think it was better without tech median and even without distress.
I mean i understand how it works, i know how to use it, but i just do not see how does that improve game experience
I mean i understand how it works, i know how to use it, but i just do not see how does that improve game experience
Last edited:
Enrico Swagolo
Deity
- Joined
- Jun 10, 2013
- Messages
- 2,387
I have never seen the situation be as bad as in my last game. Sure, I had 30+ cities but they had relatively good infrastructure, and if I let them grow in industrial era, -100+ unhappiness would be assured. Some even had most of the relevant buildings, capital had a ton of wonders but that wasn't enough. I remember this mod being very hard in the happiness management for very many versions, but never have I seen something of this sort before. To lower the hit I need buildings, and to get access to the buildings I need techs, and they're not built instantly so I need production and many turns, if I tech to the tech the needs increase so I need more buildings and before the building's up in an average city, I will have another tech or two so while I might've partially helped with boredom, the illiteracy/distress/poverty are in an even worse shape so I cannot grow anyway because I need their buildings. This means once you're in the unhappy spiral, you effectively stop growing, do so very surgically or you die.
This is not a system I like. I simply changed my happiness values, but I doubt everyone can do this.
This is not a system I like. I simply changed my happiness values, but I doubt everyone can do this.
Happiness is strange in this patch. No matter what I do I cant get into green anymore past industrial. Im not sure what Im gonna do if Im not picking imperialism and taking cities non stop to fight the permanent negative GPT
So we have some people saying they never get into negatives, and others saying that its so much unhappy they can't fight it. It either sounds like the model is too sensitive, and can "fall into extreme values" too easily, or maybe there is some bug that is kicking in with certain playstyles to explain the swings.
Minh Le
King
- Joined
- Aug 20, 2018
- Messages
- 817
Ive never had this problem before. It could be on my end because I applied the hot fix in the middle of my game and things starting to feel strange. Happiness go up and down everyturn. And past Industrial Im staying at -70 to -100 no matter what I do (its 18 unhappy citizens in my 21 pop cities).
I don’t even know where to begin. There’s about three years of design discussions that you weren’t a part of, and you step in and think you have the simple solutions? None of the things you said are original, they’ve all been discussed and tested before. I say this not to invalidate you, but to inform: the ‘solutions’ you list aren’t viable.
In short: it is fairly trivial for the average player to get above the median because, hey, it’s the median. So if you don’t have mechanics to limit players that are above the median, the entire system is a win more mechanic. We’ve been there.
The happiness system is not going to be scrapped. It’s not only not on the table, it’s utterly ridiculous. So either pivot and join in working with the system as is, or stop. It’s that simple.
Now, all that said, I spent some time last night adjusting the UI to give players a breakdown of where the city’s need modifiers are coming from. Stuff like tech median deviation is now exposed to lua. Should help players see why their values are ‘the way they are’.
G
My statement that the hotfix for the other version was savegame compatible had mixed results for players. Not sure why.
G
The values will be coming down next beta, and the median will be going to 50% (is 40%). I’m also bringing the AI’s happiness bonuses down a bit so that the whole system is a little more self regulating.
Lastly, I’m also making it so that ‘free’ population from huts or policies don’t reset the static medians, so it doesn’t hurt as much immediately to grow.
G
I'll agree with this. Regardless of my concerns about the system, its best to keep working with the fundamental concept, instead of just tossing it out wholesale at this point.
That said....a bit of a more radical thought: Right now we have Global penalties filtering through local cities (the needs met in each city). What if unhappiness was purely global?
Basically I have a global need for food, for science, for gold, etc. This could still be a factor of average pop, tech level, whatever. And just like now, if I'm not meeting those needs, it generates unhappiness. But to meet those needs, I can do it however I want.
If I need 100 food I can:
1) Provide 80 from the capital / 5 from 4 other cities
2) 100 food from the capital
3) etc.
So the player gains a lot of flexibility in how to address the system. Tall players bust out their capital, wide players have to get decent infrastructure in a lot of places. Whatever you do doesn't matter. You still have to develop, but you have a lot more freedom in how you do it. It also means none of my actions are "wasted". If I have spent time in City X to develop it, it always contributes...even a little... to my global pools. Whereas right now the system is very binary. Either I reduced unhappiness by X....or I didn't change a thing.
Then, instead of needs reducing buildings, maybe the buildings could multiply the yields effect from that city that is provided to the Global pool. So the university says "+10% science from this city added to the Global Literacy Pool" (the text is terrible it needs work). But that idea is that those buildings could still work locally, and in a way that is very easy for a player to understand.
"Ok I have 50 food in my capital. I need 20 more food to meet distress. So if I build building X I get 50% more food for distress from the capital....oh that will work lets do it"
Now ultimately you may need less needs buildings, and I actually think that is perfect fine. The mod has somewhat transcended that to me. When the happiness system was originally designed, a lot of buildings were still uninteresting and the system was a way to encourage players to build those buildings. I don't think that is nearly as needed any more, I want to build those buildings. So I don't think the happiness system is needed for that anymore, or at least not to the same degree. You could keep some buildings in that bucket, but the number could be reduced if it helped balance the system.
Last edited:
In my honest opinion, you should never say savegame compatible since last time Firaxis did that with their updates to Civ 5 people were riling up and complaining that their save games weren't compatible.
Of course we explored this in the GitHub why, but ehh..
I don't think it's working: I obtain 375 GAP no matter the amount of tourism or the fact I have a themed wonder.
Should it work as designed anyway, what do you mean with 'scaling off tourism/gap', that any point of each in the past 5 turns would be added straight to the base value, and then the final value would be buffed by the theming %?
tu_79
Deity
Happiness is right now the only mechanic that breaks the game. An easy solution is letting that be like a few years back, when we had so much happiness that its only purpose was to provide golden age points late game. But this is not interesting.
Ideally, we could have a system based on the current one which keeps sane values, not too big or too small, that varies smoothly, and in an understandable way. As I said a few months back, partially the problem is trying to fit a linear happiness model into a non-linear growth model. You can make it work for the early game at the cost of being irrelevant in the late game, or make it matter for the whole game at the cost of happiness being too harsh sometimes.
Values can be normalized (artificially made into values in a fixed range), but that won't solve the disparity in the steepness of sources of happiness and sources of unhappiness. The way luxuries are enhanced now follow up better that growth logic, and I had hopes for this to balance the unhappiness from needs. But there's an inherent problem that scales with number of cities. Sometimes, something happens in the world that makes all your cities have some happiness deficit (you got too many techs too fast, or a lackluster civ got destroyed). If you have 5 cities, that could be 5-15 points of difference; the same event having over 20 cities means changes of 30-60 happiness in very few turns. Even with all the damp mechanics in place, the big swings happen for very large empires. It's ok to have some penalties to overexpanding civs, but no one likes such big drops in happiness.
Stalker0 idea for global unhappiness is not that crazy as it seems. That would remove the number of cities from the equation. But I fear that system would be too alien and that'll take too long to balance again.
What could be done is to consider number of cities in the global net happiness. Something like:
Net happiness = (Global happiness - Global unhappiness) * (1 + Map size scaler) / Number of cities
This way, a happiness wave of 10 points in an empire with 5 cities will be felt the same as a wave of 40 in an empire of 20 cities.
Having lots of cities should be naturally worse, since it's harder to get them developed (except for Rome, maybe?)
Ideally, we could have a system based on the current one which keeps sane values, not too big or too small, that varies smoothly, and in an understandable way. As I said a few months back, partially the problem is trying to fit a linear happiness model into a non-linear growth model. You can make it work for the early game at the cost of being irrelevant in the late game, or make it matter for the whole game at the cost of happiness being too harsh sometimes.
Values can be normalized (artificially made into values in a fixed range), but that won't solve the disparity in the steepness of sources of happiness and sources of unhappiness. The way luxuries are enhanced now follow up better that growth logic, and I had hopes for this to balance the unhappiness from needs. But there's an inherent problem that scales with number of cities. Sometimes, something happens in the world that makes all your cities have some happiness deficit (you got too many techs too fast, or a lackluster civ got destroyed). If you have 5 cities, that could be 5-15 points of difference; the same event having over 20 cities means changes of 30-60 happiness in very few turns. Even with all the damp mechanics in place, the big swings happen for very large empires. It's ok to have some penalties to overexpanding civs, but no one likes such big drops in happiness.
Stalker0 idea for global unhappiness is not that crazy as it seems. That would remove the number of cities from the equation. But I fear that system would be too alien and that'll take too long to balance again.
What could be done is to consider number of cities in the global net happiness. Something like:
Net happiness = (Global happiness - Global unhappiness) * (1 + Map size scaler) / Number of cities
This way, a happiness wave of 10 points in an empire with 5 cities will be felt the same as a wave of 40 in an empire of 20 cities.
Having lots of cities should be naturally worse, since it's harder to get them developed (except for Rome, maybe?)
Last edited:
Enrico Swagolo
Deity
- Joined
- Jun 10, 2013
- Messages
- 2,387
Artist burst seems to be quite crappy right now by the way. I've hovered over it after making many Great Works, stealing some from Egypt and others and having combinations of most wonders, and it says it'll give me less than a thousand on epic? I'd rather just get another work. Even if it won't theme, at least I can put it in a city to help lower boredom and/or generate me another policy faster.
BiteInTheMark
Deity
Gazebo, I dunno why you always think I want to go from one extreme to another. I never said I want to scrap the whole happiness system. But instead get rid of the unnecessary double punishments which are hidden, unpredictable and overloaded.
Look how much time you already invested to make the system atleast a bit more transparent and solve happiness swings. How much time you have invested for the save the median mechanic, the prediction of unhappiness generated by more population and so on. Mechanics which are only needed cause of intransparent and artificial modifier calculations.
If you really need a counterweight to the increasing yields and autonomy of cities in lategame, we may could go back to the old more tech = more unhappiness system. It's something the player can be calculate with. Get a better control of it and all the additional helps youve made recently could be removed.
Stalker0 made also a very good attempt, which isn't a completely overhaul. If I see Iam behind in science by 10 points per city with 5 cities, I know I have to gain 50 science in total to decrease my unhappiness by illiteracy. An information how much I need to atleast decrease the unhappiness by one would also help to know better what to do.
Look how much time you already invested to make the system atleast a bit more transparent and solve happiness swings. How much time you have invested for the save the median mechanic, the prediction of unhappiness generated by more population and so on. Mechanics which are only needed cause of intransparent and artificial modifier calculations.
If you really need a counterweight to the increasing yields and autonomy of cities in lategame, we may could go back to the old more tech = more unhappiness system. It's something the player can be calculate with. Get a better control of it and all the additional helps youve made recently could be removed.
Stalker0 made also a very good attempt, which isn't a completely overhaul. If I see Iam behind in science by 10 points per city with 5 cities, I know I have to gain 50 science in total to decrease my unhappiness by illiteracy. An information how much I need to atleast decrease the unhappiness by one would also help to know better what to do.
- Status
