New build - July 23rd, CBP v.7, CP v. 57

Status
Not open for further replies.
I made a 5% adjustment (and by that I mean I lowered the numbers by 5). I started the same game back up so we will see if I can get out of the unhappy hole a bit easier.

That said, I looked at some numbers to put the current issue in context.

Currently my global average for Literacy is 6.5 per pop. (btw I think is a bug, most of my cities show 6.5, but one shows 7.5 for the global average).

I have a Size 14 city. It has every science building it can have (Library, University, Public School). It has every scientist slot filled, and it has 5 more science from Trade Routes.

It is producing 58 science total, or 4.15 per pop. So I have a deficit of 32.9 science in order to clear my Illiteracy.

To put that in context, in order to crack the illiteracy number at all I would need to get another 19 science (to get to 5.5 science per pop If I understand the system correctly). Only the "hokey" observatory could get me that...if I could build one.


What that tells me is that by this point in the game (Turn 261) most cities can't use yields to really get their unhappiness down. The science buildings can take you from crazy illiteracy down to moderate, but only focused illiteracy knockdown (like the zoo) will let you push past that point.

With that thought, I don't think tweaking the yields is necessarily the answer, though maybe some is appropriate. I think we need to take another pass at the happy buildings for more focused unhappy knockdown.


Also, one thing I was considering. There is actually a sneaky "metagame" with the new happiness system. If the world goes Tall...then wide becomes much more difficult. If most of the world goes wide...the world will generally be happier overall, and Tall's happiness increases even more. I don't know how strong the effect is overall but its present. I don't know if that's good or bad, but its interesting.
 
5% didn't do it I'll have to readjust some more.

Gazebo, what were the numbers before this update? I though the unhappy balance was pretty good at that point, so I want to use it as a reference.
 
Also, Gazebo I wanted to ask if there is something special going on with disorder.

My capital has every defense building available to it right now, all the way to military base. I have a machine gun in my garrison, which is the best ranged unit I have right now.

I have a defense of 4.01 per pop, and the global average is 8.2

I don't understand how a fully defensed out capital fueled by tradition growth can be behind the global "average" by such a degree?
 
Good points. I'll look at disorder and make sure the function is correct. Once the CEP buildings are in it'll be easier for us to decide as a group where happiness/unhappiness reduction should come from. Also, that meta game theory is fascinating. I hadn't considered it, but I like it, as no two games will have the same happiness curve.

G
 
I don't understand how a fully defensed out capital fueled by tradition growth can be behind the global "average" by such a degree?
I'm not too surprised, that is a consequence of relatively static defence and "per citizen" counting.

For example, a newly founded city will have defence 9 or something, but one citizen. As the city grows, the defence per citizen will drop, incurring more unhappiness immediately. It just doesn't increase fast enough and has sharp spikes with each defensive building/era. The larger your city, the more pronounced the effect.

That's exactly why Funak raised the point that city defence is a poor "yield" because until most others it doesn't scale well with city size because of the large constant term (and why I said making city defence more static CEP-style will exacerbate the problem).

Mathematically, there are only two ways around this problem: either compare average city strength on a city level, disregarding the population-based part of city strength or compare average city strength per citizen, disregarding the constant part (something like 9 x era modifier).
Also, that meta game theory is fascinating. I hadn't considered it, but I like it, as no two games will have the same happiness curve.
Interesting! If this proves being inconsistent, we might need to implement a "fudge factor" weighting the averages vs. relative empire sizes, i.e. if you have more cities than the world average, you get a break, if you have less, you have to work harder (this wouldn't change the averages, just what's the "effective" average for each civilization it compares against).
 
The calculations for CEP didn't reduce defence from population very much. Mostly adjusting the tech modifications and slight reduction in population, and then most significantly a reduction in base defence which is modified from there. I'm not sure I would say it smooths it out either (most of the value is then provided by defense buildings and garrisons). That all resulted in a significant reduction by the later stages of the game in city strengths, in exchange for extra hit points from buildings. Most of the "smoothing out" would be for city combat rather than this mechanic.
 
The calculations for CEP didn't reduce defence from population very much. Mostly adjusting the tech modifications and slight reduction in population, and then most significantly a reduction in base defence which is modified from there. I'm not sure I would say it smooths it out either (most of the value is then provided by defense buildings and garrisons). That all resulted in a significant reduction by the later stages of the game in city strengths, in exchange for extra hit points from buildings. Most of the "smoothing out" would be for city combat rather than this mechanic.

In all honesty a city without a defending unit falls in one turn in CEP (whcih imo is fine, cities should be defended and sneakattacks should be effective)

That's exactly why Funak raised the point that city defence is a poor "yield" because until most others it doesn't scale well with city size because of the large constant term (and why I said making city defence more static CEP-style will exacerbate the problem).
Yay someone gave me credit for something.

Mathematically, there are only two ways around this problem: either compare average city strength on a city level, disregarding the population-based part of city strength or compare average city strength per citizen, disregarding the constant part (something like 9 x era modifier).

You could also, especially if the CEP citystrength system is added, just totally remove population out of the equation. For even more flavor you could base the global average value, not on an actual global scale but instead on other cities close to you (If you're the least defended city in the area your citizens are going to panic/your city is going to be the target for raiders). This would give newly founded cities massive disorder unhappiness (which imo is fine, we don't have any walls and there are only 10 people living here after all, also our houses are made out of toothpicks), making it somewhat important to bring a unit for garrison.

Also I'm going to keep voicing that the unhappiness is overtuned. It is physically impossible to have the "wrong" ideology, because any slight preassure is going to throw you down a spiraling curve to -20. Even the AI understands this which is why in my last 4 games all the AIs adopted freedom.
I don't know how the global average is set, but if it is actually the global average(which seems extremely weird considering how high the number goes) it would probably be more reasonable to drop it down from the bottom 50% getting penalties to the bottom 25%.

In all honesty the current system punishes growth extremely much, probably way too much
 
In all honesty the current system punishes growth extremely much, probably way too much

In my emperor game (CP v57 first build / CBP / CSD), I could keep my wide empire (13 cities at the end) happy (sometimes barely) during all the game. And that's with taking 3 capitals and without most of the happiness buildings before the very end of the game. So sometimes at least it may work.

But it may be a consequence of nearly every civ playing wide or being a bit outteched... It was dragging the global average down and helping me.

It's a very unpredictable system because it all depends on things you don't see (what happens in the rest of the world) but I like it. When things go south though, you don't really have any solution, even buildings don't help that much. But those buildings and systems will be reviewed soon if I understood correctly...
 
In my emperor game (CP v57 first build / CBP / CSD), I could keep my wide empire (13 cities at the end) happy (sometimes barely) during all the game. And that's with taking 3 capitals and without most of the happiness buildings before the very end of the game. So sometimes at least it may work.

But it may be a consequence of nearly every civ playing wide or being a bit outteched... It was dragging the global average down and helping me.

It's a very unpredictable system because it all depends on things you don't see (what happens in the rest of the world) but I like it. When things go south though, you don't really have any solution, even buildings don't help that much. But those buildings and systems will be reviewed soon if I understood correctly...

I was talking about the latest build. Also happiness was totally fine for the most part of the game, just the last 3 eras where unhappiness skyrocketed.
 
I was talking about the latest build. Also happiness was totally fine for the most part of the game, just the last 3 eras where unhappiness skyrocketed.

Ok. I'll make a game with the latest build and I'll report back.
 
The points regarding disorder make sense. I can revert back to just measuring defensive buildings and garrisons pretty easily. I think the 'meta' is fine- variability in average per game means that there won't be a standard 'build' for each game, which is nice. The key is giving players the tools to deal with unhappiness directly, which will come with buildings.

Right now there's only two unhappiness-reducing late-game buildings. That, plus ideology pressure pain, does make it difficult to deal with late-game unhappiness. I'm aware of this, but I've been holding my cards until the CEP stuff is in. Tweaking ideology unhappiness down a bit, and perhaps buffing ideology tenet happiness a bit, would probably help as well.

G
 
The points regarding disorder make sense. I can revert back to just measuring defensive buildings and garrisons pretty easily. I think the 'meta' is fine- variability in average per game means that there won't be a standard 'build' for each game, which is nice. The key is giving players the tools to deal with unhappiness directly, which will come with buildings.

Right now there's only two unhappiness-reducing late-game buildings. That, plus ideology pressure pain, does make it difficult to deal with late-game unhappiness. I'm aware of this, but I've been holding my cards until the CEP stuff is in. Tweaking ideology unhappiness down a bit, and perhaps buffing ideology tenet happiness a bit, would probably help as well.

G

Balance is key, in vanilla AIs could be in "revolutionary waves" and still be at over 60 happiness, and I'm pretty sure we don't want that.
 
I think my main issue with this happiness system is that it's just not consistent - I can do the exact same build and cities on the same map with the same civs and be fine one time but not fine the other
 
The key is giving players the tools to deal with unhappiness directly, which will come with buildings.

I agree. We want to ensure we are hitting the spectrum.

If happiness is high in my particular game, then I don't even need any happiness buildings..or if I have them I have a lot of unhappy.

If my happiness is low, then I can still remain decently happy using all the happiness buildings.


If we can handle those extremes then the variance I think will be fine.
 
I think my main issue with this happiness system is that it's just not consistent - I can do the exact same build and cities on the same map with the same civs and be fine one time but not fine the other

I actually like that it's not the same every time, you have to adapt to whichever happiness margin you get.

I just started a new game with the latest build. Now I'm -4 happiness but it's pretty logical : as Assyria, I decided to attack Denmark and took 3 cities as puppets so I've got 7 cities early in the game now and I'm totally overextended, my territory is huge.

So I get isolation unhappiness (5) and some pillaging unhappiness too due to wars. It makes sense and I'm where I would expect to be with my strategy. I can solve some unhappiness by connecting cities and removing pillaged squares.
 
I actually like that it's not the same every time, you have to adapt to whichever happiness margin you get.

I just started a new game with the latest build. Now I'm -4 happiness but it's pretty logical : as Assyria, I decided to attack Denmark and took 3 cities as puppets so I've got 7 cities early in the game now and I'm totally overextended, my territory is huge.

So I get isolation unhappiness (5) and some pillaging unhappiness too due to wars. It makes sense and I'm where I would expect to be with my strategy. I can solve some unhappiness by connecting cities and removing pillaged squares.

The earlygame is fine, it is the lategame that isn't sustainable.
 
As this thread seems to accept general feedback, here are the two things I noticed in my early game with the latest build (emperor / CP v57.1 latest / CBP 7 / CSD) :

- Barbarians were crazy. Now unlike my last game I was on a big empty continent where my closest neighbour was far away, so I can see why barb camps were popping like mad, but barbs were really numerous and aggressive... I basically had to build a whole army to fight them, they were rushing from everywhere. It got me behind in tech. I don't say I don't like it, but that's maybe a bit over the top.

Even crazier, I got a barbarian horde CSD quest and there were like 20 barbs or more around my CS friend. Even with 3 units and a great general, I couldn't kill them all in time and the city was sacked. That was madness.

The problem I have with that is that if you don't pop near enemy civs you're already penalized because you don't get trade routes and luxuries. So if you also get tons more barbarians on top of that due to fog of war, that's maybe a bit unfair compared to civs who start near others. I know it earns culture, but still... Or maybe give even more culture for them in Honor tree to balance that.

"Raging barbarians" wasn't even checked, so I don't even want to know how many of them would have rushed my cities if I had checked this.

That's something I didn't notice in my previous game, probably because my continent was smaller, snakier and more crowded.

- The AI made some idiotic focus choices defending their cities. I came with 2 assyrian towers and a few archers / swordsmen in the same turn. Basically the towers are 90 % of the danger because they're 12 strength +200% vs cities AND they give +50% attack cities to the units near them. But the AI civ decided to focus my chariot archer instead. Never once did they focus my towers, which was clearly a blunder.

Overall, looks like another great game starting, I'm middle of the pack in scoring and I've got plenty of unhappiness to fight. That was tons of fun.
 
The earlygame is fine, it is the lategame that isn't sustainable.

Ok, I'll report when I get there. Unless happiness system changed a lot between latest v57 and first v57 versions though, unhappiness was not unsustainable in my late game the last time I tried. Still, I definitely had more problems managing it in late game than in early game (in late game I at least had to build happiness buildings for most of my 13 cities).
 
Status
Not open for further replies.
Top Bottom