A core set of balance changes

[Ahriman]

So we have the ability to do so? This opens up options for us to play?

More options aren't always better, and I disagree that settling new cities should be a good substitute for protecting existing cities late-game.

The rest of this discussion is getting tedious, I'm sick of it.

* * *

Once a nation loses their first wave... its over

Yes, this is true. Its partly because the AI doesn't make a rearguard defensive position, and partly because of the very slow construction speed of units, such that its really all about the army you have before the war starts, new unit construction is low.

Puppet states remain in anarchy unless they have a garrison.

That is a very interesting idea. That would be a very nice way of both nerfing puppets and slowing down snowball mass conquest. [Maybe a unit would also have to be present each turn to raze a city?]

My one concern would be, sometimes a unit is in a city without actually being garrisoned. So it might work better if the requirement was just to have a unit garrisoned or present in the city.

And obviously AI changes would be required, but not incredibly difficult ones.

But I like it, nice one.

 

[Celevin]

I like the idea that puppets need a garrison (Or as Ahriman said, just "present"). It's very neat, and has that killing two birds with one stone feeling.

I'm going to suggest an alteration, though. I'm still under the mindset that puppets should be completely removed. I think we should make it so all *Annexed* cities require a garrison, then have the courthouse remove this limitation (and cost 0 maintenance - courthouses costing maintenance makes me go raze-happy constantly).

I think that's very elegant. Awesome suggestion, Arkangelus.


You should update your original post, Ahriman, to reflect some of this discussion. Here's some things I'm going to suggest you take into account, looking back:
- I don't know if defenses for cities need to be buffed, as they are hardly used anyway. Walls brings archers to 1 damage, and makes warriors/spearmen fall dead. What I think the problem is is since there's a maintenance cost, nobody wants to build them. Take away the maintenance cost, make the AI want to build them more, and we should see a better game. Opportunity cost is a huge enough penalty to deter players from spamming castles. I should, if I have the gold and am being attacked, be able to buy walls and not think about "do I want to suffer 1 maintenance for the rest of the game just for the slightly better chance of saving my city?".
- Don't change "super hill farms", instead just buff mines to 2 production. With the change to maritimes, I think players are already suffering enough in food.

I can't really comment more on changes without seeing a full new list. Too many things are influencing too many things that you might or might not agree on adding.

 

[Arkangelus]

Also agreed on removing courthouse maintanence, it drives me mad that enemy capitals have a 5g/turn penalty attached to them. The cost in hammers and reduced happiness while it builds is penalty enough.

I'd also possibly look at addressing razing a bit, it should have more consequences. Every 2 pop giving an enemy a resistance unit, that kind of thing?

On the walls issue, the upcoming patch is introducing the ability to sell buildings. This will offer humans an advantage, I fear... I can't see the AI buying walls and a castle, then selling them once the city is safe.

 

[Ahriman]

First post updated with some suggestions.

I dislike removing puppets completely (they have potential to be good), and I don't think that an annexed city should be quite as good as a built city.

I think +2 hammers base increase for mines might be too much, I'd prefer to boost them around the same time as Steam Power.

 

[KrikkitTwo]

The problem with conquered cities is the three options for getting a new city give roughly the same benefits

1. Annex
2. Puppet
3. Raze+rebuild

Currently the costs are
#1=must build courthouse while going through period of unhappiness, then 5 gpt maintenance (short+long term)
#2=no control, no units/Wonders.. benefit of no social policies (long term)
#3=must regrow population, buildings, territory (short term)

So currently #3 is the best long term (except for social policy benefit)... and since territory can be bought, and food is easy to come by..and the first few populations come easy.. and conquered cities lose a lot of buildings, It is not as much of a penalty.

#2 is the next best Most of the time, primarily due to the no social policies

If puppets gave a significantly different Type of benefit than Annexing or Raze+rebuild, it would be a worthwhile interesting choice.
That's why I think puppets should give Territory only... that would make them an interesting choice. If you have the Happiness, you can Annex and get production (or Raze and rebuild)... if you don't, you stick with Puppets and get the Territory

Now I'm also of the opinion that Raze + rebuild should almost always be worse than Annexing overall.... currently, both have a short-term cost, but Annexing also has a Long term cost. I'd want to remove the Long term cost of Annexing, to make it prefered... as opposed to Raze/resettle.

That way

Annex->when you want the production, but you need enough happy

Puppet->when you only want the Territory, or just don't have the happy

Raze->if you can't hold it (or its small and in a bad position)


If players could Puppet cities without having to worry about unhappiness, that might reduce a lot of the "I want to raze X city" concerns... Puppeting would in almost every way be better than razing. (except they can take the razed city back and you can't settle there)

It might be interesting to allow players to make Puppets of their own cities as well in this case. (a Great way to manage Hapiness... just "shut down" the city)

Also 'not controlling' a puppet that gave you no benefits and had no costs would be very reasonable. (especially if they didn't build resource consuming buildings, and you could sell their worthless buildings after you had a courthouse).



Having a Output penalty for puppets Would make them different, but not in the Type of benefit. In this plan, a warmonger who conquered multiple Puppets would not be any more "productive" in terms of output than a peaceful empire.

 

[Ahriman]

So currently #3 is the best long term (except for social policy benefit)

Razing loses all the population, all buildings constructed (I'd like to reduce the number of buildings that get destroyed and pop killed from city capture) and pisses off every AI you're in contact with.
I need to understand the diplomacy engine more before I can understand how significant that last part is though.

and the first few populations come easy

This is in part because of Maritime city states. Fixing the MCSs will mean that small cities grow a lot slower.

That's why I think puppets should give Territory only... that would make them an interesting choice.

No, it really wouldn't. It would be lame to have pop that was yours that didn't actually give any benefits. If it gave you nothing but still cost you happiness, it would be underpowered. If it *didn't* cost you happiness, then there is nothing that holds you back from conquering the entire world and puppeting it all.

Raze and rebuild should never be a good option.
Puppet should be a short-term thing that gives only moderate benefits from the city.
Annex and courthouse should be best for long-term growth, unless you're going for a culture win.
The nice thing about the garrison idea is that it works well with the intention for puppets; you puppet while you're at war and can't take a big happiness hit, but you have to slow your conquest by leaving units behind.

In this plan, a warmonger who conquered multiple Puppets would not be any more "productive" in terms of output than a peaceful empire.

And I don't think that is a good design goal.

 

[KrikkitTwo]

No, it really wouldn't. It would be lame to have pop that was yours that didn't actually give any benefits. If it gave you nothing but still cost you happiness, it would be underpowered. If it *didn't* cost you happiness, then there is nothing that holds you back from conquering the entire world and puppeting it all.

Nothing will ever Stop you from conquering the whole world.... Nothing... things may slow it down. (Requiring puppets to maintain a Garrison might be useful for that)

The problem is that puppets encourage you to conquer the whole world, they overall speed the process up, because your outputs benefit from conquest.

You can always raze if you want to just stop your enemy from having 'outputs', so 'no cost' puppets don't stop that, indeed they compete with that, and make razing generally the worst option (because Puppeting also provides a no cost factor)

However, If a 15 city empire made of 10 puppets has the same costs/benefits as a 5 city empire, then Conquest isn't slowed down, but it isn't sped up either.

Because the 15 city empire has more territory to defend, but can't support more/better troops than the 5 city empire. (except by means of Resources)

Raze and rebuild should never be a good option.

I agree with that... however, it is currently the best Long-Term option.
Hence Courthouses should be 0 maintenance... to make Raze+Rebuild no better than Annex in the long term.

Puppet should be a short-term thing that gives only moderate benefits from the city.

Why? Why not just make an Annexed city without a Courthouse give moderate benefits..if Puppeting will always lead to Annexing, why not eliminate Annexing and allow you to order a "low output" Puppet to build a Courthouse... after which if becomes normal.

Annex and courthouse should be best for long-term growth, unless you're going for a culture win.

I agree... hence Courthouses should be 0 maintenance.

The nice thing about the garrison idea is that it works well with the intention for puppets; you puppet while you're at war and can't take a big happiness hit, but you have to slow your conquest by leaving units behind.

Well I think garrisons for puppets would be a good idea. (as would garrisons to get through resistance and razing.)

And I don't think that is a good design goal.

You want conquest to be slowed, this would do it.... conquest is giving you no benefits... unless you have the 'social capital' (Happiness in this case) to benefit from it.

Given that conquest weakens your competitors, profiting from it long term should require some additional work.... if you Have to have exess Happiness and wait to build a Courthouse for any 'output' benefits from conquest, it will be slowed.

In this way you could also Ban Annexing if Happiness was -10, and stop the "Ignore unhappiness" exploit.


Fo a minimal change to Puppets, I'd say stop them from producing ANY culture for Social Policies (but still for tiles), have them produce gold/science/GPP at 30-50% of normal, and require the Garrison (without the Garrison they count as resisting... giving 0 % of everything).

Much preferable would be Gold/Science/GPP at 0%, as well as Building maintenance, Happiness and Unhappiness... and require the Garrison. (without the Garrison they count as resisting... and don't even give you the territory).

 

[Ahriman]

Nothing will ever Stop you from conquering the whole world.... Nothing... things may slow it down. (Requiring puppets to maintain a Garrison might be useful for that)

If puppets require happiness to support, then that significantly slows conquest. If puppets don't eat happiness, then there is no gross cost to a puppet (let alone a net cost), and so if you are at war and you are winning, there is no reason not to continue prosecuting the war and capture and puppet every city they have.

When puppets require happiness, you can't do this without tanking your economy.

If puppets give you nothing and require happiness, then they are pathetically underpowered.

Ergo: puppets should create unhappiness, but should also give you something in compensation.

QED.

You can always raze if you want to just stop your enemy from having 'outputs', so 'no cost' puppets don't stop that

They do if razing multiple cities pisses off all the other AIs enough, and shuts down your ability to trade with them.

I agree with that... however, it is currently the best Long-Term option.

You can't balance by looking only at the hypothetical "long term". The short-term costs (losing the population, which will be harder to get back with weaker Maritime CS, and losing the buildings) are non-trivial. A pop 6 city is going to give you a lot of trade route gold in the time it takes a newly settled city to regrow.
A better solution is to penalize raze and resettle more by keeping more buildings and pop of a conquered city.

Why? Why not just make an Annexed city without a Courthouse give moderate benefits..if Puppeting will always lead to Annexing, why not eliminate Annexing and allow you to order a "low output" Puppet to build a Courthouse... after which if becomes normal.

?
The cost of not having a courthouse is a happiness one. There is a difference between a happiness cost and a productivity cost.

ou want conquest to be slowed, this would do it....

No it wouldn't. Adopting your proposal (where puppets have no gross costs) would lead to endless conquest with no downside.

 

[Celevin]

I think +2 hammers base increase for mines might be too much, I'd prefer to boost them around the same time as Steam Power.

Well I meant +1 (2 in total for a mine), but yeah, a tech change is better so it coincides with lumber mills.


I think the entire puppets/annexing thing is a lot of personal opinion on how they operate, so Ahriman can implement something that he thinks is best. After he gets something concrete down, we can drill him on it. This is his baby, after all. I want to move to happiness scaling.

I argued before that culture didn't scale well with empire size, or world size. This is because there's a static coefficient with it. Well the same problem lies with happiness as well: due to luxury resources, there's a large chunk of static happiness. This chunk makes it hard to balance large/small empires, and obviously hurts larger world sizes. To fix this problem, I propose the following:

1) Let civs benefit from having multiples of the same luxury. So 2 furs will grant 2*5=10 happiness. A lot of work will have to be put into this, as there needs to be UI changes and trade changes.

This will make world sizes scale perfectly with happiness. The problem with it is obvious, however: big civs benefit a lot more from this change than small civs. I also propose the following which is designed to slow down rapid expansion, and bring back more of a "you can support more cities as time goes on" feeling that we had from Civ4:

2) Make the unhappiness per city increase as your number of cities increases. So the first city settled might give 2 unhappiness, then the next city settled will give more than 2 unhappiness, the next more than that, and so on.


I believe the second change will give smaller empires that ability to shine. It will cause bigger empires to really consider if they can support another city, based on if there's luxuries, and how much unhappiness it will bring. Right now it's a no brainer to "settle settle settle" if you have the cash and can take the policy hit.

 

[Ahriman]

After he gets something concrete down

Guys, don't expect anything. I don't have the technical expertise to implement anything like the full set of changes here, even once the sdk is available.

1) Let civs benefit from having multiples of the same luxury.

I would strongly, strongly oppose this. This would be game-breaking.

The whole point of luxuries is that to get lots of them, you have to trade. This is why its important to have good relations with civs, to honor your deals, and so forth. If my 2 furs are as good as 1 furs and another luxury I don't have, why should I ever trade?
It would also destroy the incentives for exploration, particularly discovering other continents. There is supposed to be a big reward for contacting other civs, because now you get a whole set of new/extra luxuries you can trade for.

The much better solution is just to have more luxuries in the game, and then cut some of them out as map sizes decline, so they don't all appear on smaller maps.

Make the unhappiness per city increase as your number of cities increases

This might work, but it would be incredibly hard to balance, and would be non-transparent.
Look at how frustrating the unit maintenance formula is. 2 per city is simple, and it allows you to easily evaluate things like the Indian UA, the Forbidden Palace and Planned Economy, and the marginal happiness impact of making a new city.

 

[Elenhil]

Concerning puppets - great minds think more or less alike, it seems! There's been a similar trend in RFC V suggestions thread. Though that, being a proposed part of RFC gameplay, is perhaps rather too much for a simple Vanilla mod.

 

[Celevin]

Guys, don't expect anything. I don't have the technical expertise to implement anything like the full set of changes here, even once the sdk is available.

All I meant was something written up, not implemented.

The much better solution is just to have more luxuries in the game, and then cut some of them out as map sizes decline, so they don't all appear on smaller maps

This would also work.

I don't see anything inherently wrong in my idea - Yes it would take away uniqueness, but you could still trade for them. You just need to judge if happiness or gold is better for your civ at the moment.

This might work, but it would be incredibly hard to balance, and would be non-transparent.
Look at how frustrating the unit maintenance formula is. 2 per city is simple, and it allows you to easily evaluate things like the Indian UA, the Forbidden Palace and Planned Economy, and the marginal happiness impact of making a new city.

The frustrating part of unit maintenance is all of a sudden out of a sudden out of nowhere units are costing 8 gold each, only every second unit deleted will bring down that cost, and all units cost the same. Also, your unit count can rise/drop easily from turn to turn, much more than your number of cities. Finally, unlike unit maintenance, empire maintenance wouldn't change suddenly the later in the game.

Gold maintenance in Civ4 increased like the way I'm suggesting we do unhappiness in Civ5. I predicted ICS before the game was released. This right here is the #1 reason that people are expanding like maniacs, and are being rewarded for it. Science is *the* currency in Civ, and right now big empires can have much more science due to more happiness. We need to bring down the happiness, and all of the major problems will follow.

 

[Roxlimn]

You could significantly bring down the happiness available for many-city Civs just by requiring cities to be population 10 or 14 before they can make Colosseums.

 

[Elenhil]

There are several ways to address the happiness problem. One is to make distance count. Make the maximum city distance from your capital a factor lowering empire-wide happiness, and you won't get ICS. Logical, too, if you imagine logistical difficulties transporting luxury goods throughout a vast continent. Imagine a country like Russia with a hell of a spread. I tell you, it will take a lot more than a new traded happiness resource to make a change for the whole beast!

 

[Ahriman]

I don't see anything inherently wrong in my idea

The problem is it removes the need to have/find multiple other civs, and to keep on good terms with them. There are lots of behaviors in the game (lying to the AI about declaring war, razing cities, etc.) that are controlled through reputation, and breaking reputation makes other civs more hostile, where they trade with you on worse terms. The only reason for this behavioral stick to actually matter is if being able to trade is really important.

Its a deliberate design that a second or more luxury is useless unless there is another civ to trade it to, and a third is useless unless there are two civs to trade to, etc.
If you can use multiple copies of luxuries yourself, then the strategic loss from losing trade is trivial. Also, there is no longer any need to trade my luxuries for your luxuries.

Also, map scripts seem to cluster some resources more than others (eg incense will often show up in a big patch, whereas marble is much more sparse).

Gold maintenance in Civ4 increased like the way I'm suggesting we do unhappiness in Civ5.

I'm not ruling it out. I'm just wondering; would a sliding scale be better or worse than just shifting unhappiness per city to 3?
I guess the nice thing about moving to a formula is that its much easier to adjust for map size.
But the difficulty is that the system is not designed to accommodate fractional unhappiness well.
So any function is likely to be a step function rather than a clean one.

by requiring cities to be population 10 or 14 before they can make Colosseums

Ugh.... no. Way too artificial, and limiting.

One is to make distance count

Dropping distance was one of the best things about Civ5. It finally allows scattered Empires, and makes colonization less painful. Distance is handled with road maintenance cost, that is sufficient.
Also, distance measures would tend to encourage tightly packed city spamming.

 

[Roxlimn]

Ahriman:

In previous Civ-like games, you would only need buildings like Aqueducts and Hab Complexes and Hospitals to increase a hard population cap. This is a similar thing. Why would a city with 1 population benefit the empire from building a Colosseum? There aren't enough people in it who can actually use the thing!

 

[Moon Pine]

Option A
Make all horses produce only 2 copies, they are far too common. All other resources provide either 2 or 4.

I got 147 copies of horses in last game, but I'm sure it's caused by some bug

 

[Celevin]

I hated distance maintenance. It is one of the things that causes Civ4 empires to look identical. If we solve the entire ICS problem, I'm sure we'll see a lot more neatly spread out empires in Civ5 trying to take advantage of good land.

I'm not ruling it out. I'm just wondering; would a sliding scale be better or worse than just shifting unhappiness per city to 3?
I guess the nice thing about moving to a formula is that its much easier to adjust for map size.
But the difficulty is that the system is not designed to accommodate fractional unhappiness well.
So any function is likely to be a step function rather than a clean one.

I'd much rather take the plunge and move away from integers rather than trying to work with them and finding out it doesn't. I don't see anything wrong with fractional happiness, in fact with the "-20% unhappiness in each city" policy, don't we already see evidence for it, but it's just rounded off for the UI?

The problem with a linear trend is, happiness wise, when will building another city ever become a bad idea? If it's not a bad idea for the second given how many happiness buildings you can make, it won't be a bad idea for the third, fourth, and so on. You're either going to make it so it takes way too long to expand your territory, or way too easy. There's no inbetween with it being linear. A step function on the other hand might be awkward due to the very low numbers for happiness count. Players would know the exact point that they switched, and would hover cities around that point, which would feel artificial. The only way to use a step count would be to actually multiply all happiness / unhappiness values by 100 and deal with it from there, ie the equivalent of using 2 decimal places.

Unhappiness is the new maintenance. In Civ4, we measured "building this city will give me +x gold in trade routes, +y gold in other gold bonuses like tiles in a few turns. It will cost z maintenance". We then measured x+y VS z to see if we should build that new city. We aren't doing anything like this now. What we're doing now is we build that new city, either have the policies/wonders in place, or build a colloseum immediately, and it's a good idea. The only thing stopping explosive growth is opportunity cost for both happiness, and culture.

 

[Elenhil]

Dropping distance was one of the best things about Civ5. It finally allows scattered Empires, and makes colonization less painful. Distance is handled with road maintenance cost, that is sufficient.
Also, distance measures would tend to encourage tightly packed city spamming.

Why are scattered empires, in your view, a good thing? I was of opinion that RL scattered empires are quite vulnerable. Physical separation makes political separation and independence easier. Sadly, there is none in CiV, but why should an artificially separated country like pre-Bangladeshi Pakistan be perfectly OK in CiV? I gather that the only downside in CiV to settling on the other side of the globe is security. But I don't hear of AI taking advantage of it and regularly biting off distant cities to leverage this possible downside for the player.

And why should colonization be less painful? Shouldn't it be more gainful rather than more simple to settle half the world away?

The thing is, being able to settle half the world away to grab a happiness resource is exactly the problem you're trying to solve. Currently even the puniest of civilizations can be a globe-spanning Empire (and reap the benefits of luxury resources in the New World as if it were their back garden). Roads do nothing to prevent it.

 

[Ahriman]

In previous Civ-like games, you would only need buildings like Aqueducts and Hab Complexes and Hospitals to increase a hard population cap. This is a similar thing

There is broad agreement that moving to the Civ4 system was a vast improvement over the artifical hardcaps of aquducts and hospitals required to grow over size X.

And now, happiness is an empire-based mechanic. It makes no sense to impose arbitrary city-based limits on it.

This would be like preventing a small city from building a courthouse in Civ4. Not a good plan.
Its a much better design to use sticks and carrots, rather than highly prescriptive hardcaps which limit the player.

The problem with a linear trend is, happiness wise, when will building another city ever become a bad idea?

Because the fixed happiness bonus from your luxuries, many social policies (protectionism, cultural diplomacy, legalism), difficulty level and wonders is being spread over more and more cities, so the average happiness per city is declining.

Still, its hard to get a formula that doesn't feel arbitrary and forcing (Thou Shalt Have This Many Cities).
Even supposing we could deal with fractional happiness (which is messy), what kind of formula would you propose?
Maybe something like marginal unhappiness from the Nth city = 1.5 + N/K
Where K depends on map size.

Why are scattered empires, in your view, a good thing?

Real life empires were often quite spread out, particularly using sea routes rather than land trade.
Greek and Roman empires went around the Mediterranean coast, they didn't just form a bubble around the capital. Medieval Spain/Hapsburgs controlled scattered territories throughout Europe.
Spain, England, France, Portugal and the Netherlands established major profitable colonies in the New World and scattered all over the globe.

The English ability and the Commerce tree are designed to be able to encourage a scattered playstyle, not just a circular land-based empire.

So: the point is, scattered empires are both realistic and fun.

Also, note that we can't move capitals anymore, so distance from Palace would be very broken.

I was of opinion that RL scattered empires are quite vulnerable.

This is still true even without distance maintenance costs, since it takes longer for your army to get to wherever it needs to be to respond to an invasion.

I gather that the only downside in CiV to settling on the other side of the globe is security.

And road costs, if you don't have harbors and a clear sea route.

And why should colonization be less painful?

Because distant colonization in Civ has always been near useless, because of the punitive distance modifiers.

The thing is, being able to settle half the world away to grab a happiness resource is exactly the problem you're trying to solve.

No its not. I have no problem with being able to do doing this. The problem isn't in building a distant city in a good place, its in being able to build cities everywhere.

Currently even the puniest of civilizations can be a globe-spanning Empire

You mean, like Portugal and the Netherlands?
I don't see the problem. If they're puny, then their possessions will be scattered and hard to defend, and it will be easy to pick them off piecemeal, particularly if they don't have a decent navy.