Is the culture and science increase per city and empire modifier making wide too hard?

[ElliotS]

Taking this from another thread to make it more visible, and enhancing it:

I'm wondering if the 7% culture/science increase on standard isn't too high. It seems to really favor tradition and taller play. (which we see at least among the AI doing much better)

Between that and the very punishing empire happiness modifier I think peaceful wide is in a really bad spot and warmongering is in the same spot it's always been: Good because the player does war better than the AI.

Players can win as a wide warmonger, but rarely to never do I hear about players winning wide as a science victory or anything else. As a player I tend to feel like unless I'm warmongering cities past 4 to 6 are often a liability, even if I have a ton of room and especially past the first hundred turns.

The bottom line for me is that new cities are bad to get in most cases past your first few. The fact that by taking advantage of really nice spots to settle you often actually slow down your science and policies even with tons of investment, seems like a big problem to me.

Settling a new city isn't easy. It's not risk-free. It shouldn't slow you down more than it helps you.

I think our goals (and the reasons the modifiers are where they are) are as follows:

• Discourage settling of many low-value cities (Infinite city spam was the reason the modifiers were added to vanilla)
• Allow a peaceful tall civ to compete with a wide civ, so as to avoid "whoever has the most cities wins"
• Temper the gains from warmongering, so as to not make warmongering the only game in town.

Here are the problems I see with what we have:

• Building new cities is often a negative action, especially into the game. That seems really counter-intuitive and wrong in a civilization game.
  
• Warmongers can largely bypass this malus with puppets.

I think that the power of warmongering has largely been mistaken with the power of wide, and wide has been over-nerfed in an attempt to hold back warmongers.

I would like to test putting a 33% (give or take) yields malus on courthouses, reducing the culture/science increase to maybe 4%, and reducing empire unhappiness modifier by some amount. (and maybe giving a better way to counteract it globally.) It would buff raze and replace, but that's always been the weakest option in 90% of cases so I don't mind.

I haven't played for a while, so maybe I'm off base, but in the 5ish games I've played this patch this seems to have been very clear to me. It also seems like a lot of threads have proven how detrimental new cities are recently. (I think much of the issues with pioneers and colonists is because of this, and not necessarily the units themselves.)

Finally I'd like to add that in my mind the goal is that settling new cities is an advantage, but not such an overwhelming advantage that Tall cannot win or compete. There are added issues and struggles that come with more cities, and a lot of other advantages to seek to compensate.

So if my recommended tweaks don't get there, the numbers should be adjusted again. I just think they're too high right now.

 

[CrazyG]

I made an excel sheet to look at this for anyone to download. You can tinker with the values to see how a change would affect the culture a city needs to produce.

The Break Even Culture Line would be how much culture is required for a city to break even. This is a bare minimum, realistically you probably want cities to be a few points ahead of this (because your other cities spend resources to get that new city started), and they need to produce science too (science = culture is a pretty good guess).

Here are some sample numbers. Let's say you want your 6th social policy by around turn 100. How much culture does a new city need to break even?

Spoiler :

Cost of 0.07 (current cost in standard)
The city needs 3.9

Cost of 0.1 (current value in small)
The city needs 5.9

Cost of 0.05 (current value in large maps)
The city needs 2.7

To interpret this, let's say you have a city that produces 6 culture on turn 90.
On small, it effectively produces about 0 culture.
On standard, it effectively produces about 2.
On large, it effectively produces more than 3.

6 is pretty high for a non-capital city on turn 90. Typically you only get about 2 from social policies, 2 from a monument, so you need another 2 from a pantheon, unique trait, arena, or terrain. 6 science would be hard to reach too.

 

[CrazyG]

Here is the doc for anyone interested in tinkering.

 

[ElliotS]

I made an excel sheet to look at this for anyone to download. You can tinker with the values to see how a change would affect the culture a city needs to produce.

The Break Even Culture Line would be how much culture is required for a city to break even. This is a bare minimum, realistically you probably want cities to be a few points ahead of this (because your other cities spend resources to get that new city started), and they need to produce science too (science = culture is a pretty good guess).

Here are some sample numbers. Let's say you want your 6th social policy by around turn 100. How much culture does a new city need to break even?

Spoiler :

Cost of 0.07 (current cost in standard)
The city needs 3.9

Cost of 0.1 (current value in small)
The city needs 5.9

Cost of 0.05 (current value in large maps)
The city needs 2.7

To interpret this, let's say you have a city that produces 6 culture on turn 90.
On small, it effectively produces about 0 culture.
On standard, it effectively produces about 2.
On large, it effectively produces more than 3.

6 is pretty high for a non-capital city on turn 90. Typically you only get about 2 from social policies, 2 from a monument, so you need another 2 from a pantheon, unique trait, arena, or terrain. 6 science would be hard to reach too.

Thanks for that work, it's a really good way to visualize and play with the numbers.

Honestly I'm feeling really sure that it makes no sense whatsoever that the numbers change based on map size. It's absolutely insane that building a city on a standard map could lose you culture, where the same city on a bigger map can gain you culture. It also makes user-feedback way more inconsistent.

I would like to change all maps to 4% or 5%.

What do you think?

 

[CrazyG]

I would like to change all maps to 4% or 5%.

What do you think?

The numbers at minimum indicate that small maps are screwy. It's really hard to get a city that actually earns culture or science. I've tried them before when I had a weak computer (so I wanted faster AI turns) but it always felt off.

Your capital + puppets seems like the ideal strategy, with maybe a couple normal settlers to help with faith or military supply. Does anyone play small map sizes regularly at high difficulites? I'd be interested in hearing their experience. From earlier in other threads, the large size players appear very reluctant about people who don't even use their settings changing their settings, which is a completely genuine concern. I imagine small size players would feel the same.

On standard, I'd be happy to try out 5% for at least a patch.

 

[Kim Dong Un]

Taking this from another thread to make it more visible, and enhancing it:

I'm wondering if the 7% culture/science increase on standard isn't too high. It seems to really favor tradition and taller play. (which we see at least among the AI doing much better)

Between that and the very punishing empire happiness modifier I think peaceful wide is in a really bad spot and warmongering is in the same spot it's always been: Good because the player does war better than the AI.

Players can win as a wide warmonger, but rarely to never do I hear about players winning wide as a science victory or anything else. As a player I tend to feel like unless I'm warmongering cities past 4 to 6 are often a liability, even if I have a ton of room and especially past the first hundred turns.

The bottom line for me is that new cities are bad to get in most cases past your first few. The fact that by taking advantage of really nice spots to settle you often actually slow down your science and policies even with tons of investment, seems like a big problem to me.

Settling a new city isn't easy. It's not risk-free. It shouldn't slow you down more than it helps you.

I think our goals (and the reasons the modifiers are where they are) are as follows:

• Discourage settling of many low-value cities (Infinite city spam was the reason the modifiers were added to vanilla)
• Allow a peaceful tall civ to compete with a wide civ, so as to avoid "whoever has the most cities wins"
• Temper the gains from warmongering, so as to not make warmongering the only game in town.

Here are the problems I see with what we have:

• Building new cities is often a negative action, especially into the game. That seems really counter-intuitive and wrong in a civilization game.
  
• Warmongers can largely bypass this malus with puppets.

I think that the power of warmongering has largely been mistaken with the power of wide, and wide has been over-nerfed in an attempt to hold back warmongers.

I would like to test putting a 33% (give or take) yields malus on courthouses, reducing the culture/science increase to maybe 4%, and reducing empire unhappiness modifier by some amount. (and maybe giving a better way to counteract it globally.) It would buff raze and replace, but that's always been the weakest option in 90% of cases so I don't mind.

I haven't played for a while, so maybe I'm off base, but in the 5ish games I've played this patch this seems to have been very clear to me. It also seems like a lot of threads have proven how detrimental new cities are recently. (I think much of the issues with pioneers and colonists is because of this, and not necessarily the units themselves.)

Finally I'd like to add that in my mind the goal is that settling new cities is an advantage, but not such an overwhelming advantage that Tall cannot win or compete. There are added issues and struggles that come with more cities, and a lot of other advantages to seek to compensate.

So if my recommended tweaks don't get there, the numbers should be adjusted again. I just think they're too high right now.

Beautiful post! I think you hit the nail on the head in terms of diagnosing the root of the problem and where my concerns actually stem from. I normally don't like poking Papa Bear, but he's on record saying how he can't feasibly see all the posts throughout the forum, and that's understandable, so if I could highlight any recent post for him to lay eyes on in regards to pin-pointing a big underlying issue, it's this one... @Gazebo

Your capital + puppets seems like the ideal strategy, with maybe a couple normal settlers to help with faith or military supply.

So like I've said for a while now: the infamous 6 city turtle is the best and true way!

Yeah, I'd really prefer if that were not the case...

 

[ElliotS]

From earlier in other threads, the large size players appear very reluctant about people who don't even use their settings changing their settings, which is a completely genuine concern. I imagine small size players would feel the same.

I actually really like playing on huge, though it messes with wonder balance. I'm not speaking from ignorance. Moreover I'm actually suggesting that we make all maps use their settings (or lower by 1%) which should make the suggestion palatable to them. I wouldn't want to raise their cost to standard's 7%, because as mentioned it's too high. That's generally where I've seen push-back from them.

 

[CrazyG]

I think that the power of warmongering has largely been mistaken with the power of wide, and wide has been over-nerfed in an attempt to hold back warmongers.

The current tourism situation is a really bad example of this. 8 city progress puts you at -56% tourism. Great Works or wonders don't really scale with number of cities.

Wide tourism with progress was a legitimate strategy. Civs like Polynesia are specifically built to do it.

 

[Stalker0]

The current tourism situation is a really bad example of this. 8 city progress puts you at -56% tourism. Great Works or wonders don't really scale with number of cities.

Wide tourism with progress was a legitimate strategy. Civs like Polynesia are specifically built to do it.

I will definitely say, while I don't necessarily agree that the Science/Culture mods are off, I do think tourism is very harsh with its scaling. I don't see how wide tourism is really possible at the moment.

 

[Ninal]

Great post @ElliotS ,

I do agree that wide-peacefull play seems to be in a really bad place. I do support reducing the culture/science to 4% on all maps. This would make settling new cities more viable. But i still wonder if progress can be fast enough to safely settle enough cities. If iT can’t the extra long term rewards of the lower culture/science increase might be less meaningfull.

 

[Gidoza]

The current tourism situation is a really bad example of this. 8 city progress puts you at -56% tourism. Great Works or wonders don't really scale with number of cities.

Wide tourism with progress was a legitimate strategy. Civs like Polynesia are specifically built to do it.

What if the 7% Malus was multiplicative? That is to say - the first city knocks you back to 93%, but your second expo knocks you down to (93% * 93%), the third (93% * 93% * 93%), and so on. If you ask me, the additive 7% is the problem because it makes each consecutive city damage the output of all the others disproportionately.

With my suggested change, the 25% bottom for tourism would come at 19 expansion cities (instead of 11), which seems more reasonable.

To clarify this further: when you have 10 expansions, your Tourism is reduced by 70% total, and the 11th expansion moves that to 75%. Inverted, that's 30%, moving to 25%. 30 to 25 is a total of a 16.67 % decrease - which is more than double the additive 7% that is being applied. In other words, each city you expand to not only decreases your tourism (and science and culture relatively), but it decreases it MORE with each city added. The equivalent reduction in my formula would be 30% - 7% = a 2.1% decrease, which is the effective 7%. This makes each new expansion city authentically equivalent.

When applying this idea to Culture or Science, the appropriate way to make it work would be to "effectively" *decrease* one's science or culture value per extra city founded. However since early-game numbers are low and this would create decimal numbers, we instead just display the original science/culture generation, and multiply the science/culture required for the next tech/policy by the inversion. For example:

Current Science = 100
Next tech requires 10,000 science (100 turns)

Add 1 city = Science is now 93 (107.5 turns)
But instead, the display would be:
Science = 100
Next tech requires 10752 science (107.5 turns)

That way each city's effective multiplier is the same for each science/policy as well.


@ElliotS


EDIT: Also with regard to Wide Tourism is simply the issue that it's not possible to fill up all the GW slots, nor do GW people spawn any faster after having filled up the guilds. One pretty much has to conclude that the Progress goal ought to be more of a sciency one because of this unless Progress cities magically started producing GPP out of thin air.

 

[Zanteogo]

One thing to take into account that cannot be easily shown by pure numbers is the bonus wide civs gain from holding extra land.

These include:

-Higher strat and lux resource access (monopolies!)

-Blocking other civ access and movement

-Stronger religious spread due to ease of coverting own cities rather than other civs (considering how religious spread is compound by multiple cities)

These aren't as easy to qualify because its not a direct loss like the expansion penalty is. It's also why we have to be careful when messing with said penalty too much because it will quickly tip the scales into wide being best play.

I would like to note that I personally think that wide and tall should be both realistic options, I think wide should be on average slightly better due to it being more difficult due to holding more risk. Tall means you just don't make more cities and focus on other things instead. Wide means you have to focus on building settlers, which slow down cities, plus protect and build weaker new cities.

The other downside to wide is how military strength is decided. It's divided by number of cities. The AI picks targets partly due to military strength. This balances out once your new cities are up and running and building their own units, but until then you put a bigger target on your back. (In addition to the teritoral dilpo malus wide already making you a target)

 

[DeAnno]

The current tourism situation is a really bad example of this. 8 city progress puts you at -56% tourism. Great Works or wonders don't really scale with number of cities.

The tourism malus is a whole extra issue in that the math of it is implemented in such a way that it doesn't work like science or culture. For example if your extra cities modifier is 50%, your techs and policies are 67% speed, but your tourism is 50%, because Tourism subtracts from the rate (in the numerator) instead of adding to the target (in the denominator.) Despite being similar for a low modifier, these ARE NOT the same thing. Ideally, Tourism should be modified to be like the other two, which would imply a rate penalty of 1-1/(1+modifier) = modifier/(1+modifier). For example if your modifier for tech/culture is at 50%, your tourism multiplier should be 0.5/(1+0.5) = 33% penalty to tourism. This means your tourism would come in at 67% strength, just like science and culture.

 

[Deljade]

Playing on 15 civ huge I will say that the balance seems good, civs are expected to have 7-12 cities depending on the landmasses and the type of civ/ai personality. And if you want wide science victory, just get Order, simple as, the bonuses scale very well with empire size. So as long as huge map city founding balance is not tampered with do what you will, I'd say. But it does seem to me that 5% is a tad too low for standard, where players are naturally expected to have less cities, may be 7 would be a good one to try? Still 10 on smaller maps.

 

[ElliotS]

where players are naturally expected to have less cities

That's not really accurate. Huge maps have more land and more players, and the amount of land per player isn't significantly bigger.

More importantly: Arbitrarily trying to restrict number of cities on standard by making cities bad to build is dumb. 7% means that most cities are culture and science drains on your empire for eras or even the whole game. (Not to mention happiness.) It's dumb that doing something risky like building a city in a coveted spot is often worse than not building it not because of military threat but because the city will actually hurt your culture and science output.

 

[Deljade]

Maybe I've not played Standard size in a while, how many decent (5 tiles apart) cities does each player get? I remember doing 4 and maybe 2 more later on, so 6 but with full 6 tiles distance between them, due to vanilla's dumb restriction on National Wonders.

 

[CrazyG]

But it does seem to me that 5% is a tad too low for standard, where players are naturally expected to have less cities, may be 7 would be a good one to try? Still 10 on smaller maps.

The standard value is already 7.

If 5% is right for large, and it is also right for huge, why is it wrong for standard? The differences between large and standard are comparable to the differences of large and huge. I think ElliotS is challenging if the value should be different on different map sizes at all. It's been that way for a long time, but I can't really come up with a reason why they should be different (other than it's the way it's been).

Then on small maps, why 10%? Does playing on a small map mean that each city (other than the capital) will produce double the culture and science that the same city would on a large map? That's roughly what penalties of 10% vs 5% imply. I don't see why the city on small would have significantly higher science. Am I missing something about map generation?

 

[ElliotS]

The standard value is already 7.

If 5% is right for large, and it is also right for huge, why is it wrong for standard? The differences between large and standard are comparable to the differences of large and huge. I think ElliotS is challenging if the value should be different on different map sizes at all. It's been that way for a long time, but I can't really come up with a reason why they should be different (other than it's the way it's been).

Then on small maps, why 10%? Does playing on a small map mean that each city (other than the capital) will produce double the culture and science that the same city would on a large map? That's roughly what penalties of 10% vs 5% imply. I don't see why the city on small would have significantly higher science. Am I missing something about map generation?

Nope. In fact having less civs means you'll actually get LESS science because of less trade route opportunities, trade partners, cities to spy on and and people having already researched a tech.

But some bizarre logic of "more cities on bigger maps" that seems correct at first glance but doesn't add up under scrutiny means that the penalty has been in place like that since vanilla.

Thing is that in Vanilla the penalties were all small, so it never became an issue. Only truly worthless cities would hurt you regardless.

We've just pushed it to a point where I think it's causing serious balance issues.

 

[Deljade]

The standard value is already 7.

If 5% is right for large, and it is also right for huge, why is it wrong for standard? The differences between large and standard are comparable to the differences of large and huge. I think ElliotS is challenging if the value should be different on different map sizes at all. It's been that way for a long time, but I can't really come up with a reason why they should be different (other than it's the way it's been).

Then on small maps, why 10%? Does playing on a small map mean that each city (other than the capital) will produce double the culture and science that the same city would on a large map? That's roughly what penalties of 10% vs 5% imply. I don't see why the city on small would have significantly higher science. Am I missing something about map generation?

The way I would justify it would be that, simply, there are more tiles per player on bigger maps, therefore, room for a couple more cities. Of course this is dependant on map generation, which is why I think it's important to know how many cities a player is expected to have. But a simple calculation of total tiles gives the following results.

Duel (2 players) - 480
Tiny (4 players) - 504 tiles per player
Small (6 players) - 462
Standard (8 players) - 520
Large (10+2 players) - 554
Huge (12+4 players) - 640

 

[ElliotS]

The way I would justify it would be that, simply, there are more tiles per player on bigger maps, therefore, room for a couple more cities. Of course this is dependant on map generation, which is why I think it's important to know how many cities a player is expected to have. But a simple calculation of total tiles gives the following results.

Tiny - 504 tiles per player
Small - 462
Standard (8 players) - 520
Large (12 players) - 554
Huge (16 players) - 640

The difference between large and huge is much bigger than standard and large, so based on your logic shouldn't large and huge have different values? Also the difference between standard and large listed is 34 tiles, less than a full city's 3 tile radius of 36 tiles. So maybe one extra city expected?

And most importantly none of that matters because your cities don't magically produce a ton more science or culture on small or a ton less on huge. They produce about the same amount. No reason that they should have different penalties.