Balancing Civ 7

[rocksinmypath]

This post is partly inspired by Potato McWhiskey's recent video where he brings up various balancing problems, including production being overpowered, as some of the big problems he has with Civ 7. I wanted to add to the discussion by talking about what I currently see as big issues with balancing in Civ 7 and how I think these issues can be addressed. Before I start, I want to lay out a couple things. First of all, I'm currently approaching 300 hours in play time, and almost all of it has been spent on the Age of Antiquity. I haven't touched Modern Age yet, and I've only dabbled in Exploration to the point where I can already see some big issues with it (e.g. snowballing, city states getting wiped out within the first few turns, cultural legacy path). I'll only focus on Antiquity as I don't have enough experience in the other ages. Secondly, I'll mostly be talking about core mechanics of the game that apply to all leaders and civs. I'm not particularly concerned about unique abilities and mementos breaking a game that's otherwise very well balanced.

Note: I was going to discuss all of the balance problems that I currently see in Civ 7, but I realized that would make this post too long, so I'm just going to talk about balancing production in this post.

Production is King

Currently, there's a consensus that the strongest opening is to look for a production-heavy start (if you can afford to), and then research Pottery, build a Brickyard, and then research Masonry to maximize production on Mines, Clay Pits and Quarries in your capital. Alternatively, you could open with Animal Husbandry and a Saw Pit. Production resources such as Cotton, Hides, Sheep and Gypsum are recognized to be among the strongest opening resources. Production is quite obviously very strong that it feels like you're losing out if you deviate from the Brickyard or Saw Pit opener or spawn in a location that doesn't have these production resources.

At a fundamental level, the problem with production is simple. By following the "meta", you can very easily get yourself a significant amount of production, and there are so many ways you can use the early production without any downside.

When you start the game, your capital starts with 5 production per turn. If you're lucky enough to start next to a Hides tile, in a couple turns, your production balloons up to 9 per turn. The resource itself yields 3 production, and it adds 1 production to the tile it sits on. Just by sheer luck, you add 80% to your production base at the start of the game. One easy way to nerf production, or specifically production-focused starts, is to increase the production base provided by the Palace to, say, 10. That should effectively halve the impact the first few tiles have on how your game will go. This can be accompanied by nerfing production resources, or maybe even resources in general. I'm puzzled by the decision to make resources provide yields to tiles like they did in Civ 6, even though the resources can now be slotted into settlements to provide bonuses. They really don't need to, and in the case of Hides that we're looking at, removing that 1 production should have a fairly significant impact on reducing early production gain. In addition, production resources can be realigned to be less "front-loaded". For instance, instead of providing flat +3 production, Hides can provide +1 production and +10% toward unit production. With these adjustments, a single Hides tile goes from boosting the base production by 80% to just 10-20%.

Another very easy way of getting lots of production early game is to open with either a Brickyard or a Saw Pit. I'd argue these buildings are currently better than a lot of the wonders you can build in Antiquity. Typically, you can easily time your first Brickyard or Saw Pit such that as soon as you're done building it, you hit population 5 and can start working on your first settler. Of the five citizens, one lives in the Palace and another in the warehouse building, meaning there are three citizens working rural tiles. It's not difficult to have all three of them working improvements that synergize with the warehouse building of your choice. Your first Brickyard, for instance, by the time you're creating a settler, could easily be generating 4 production per turn (1 base, 3 from tiles), and that Brickyard only cost you 55 production to build, which means you're breaking even on it in 14 turns. That's an insane return on investment. For context, let's compare that to the Carthaginian Cothon. It costs 120 production. The maximum production yield you can get on a Cothon is 8 (before Masonry II and excluding wonder adjacencies), and that requires 6 adjacent Coast or Navigable River tiles, which is almost impossible because the building also needs to be adjacent to land. Ignoring the maintenance of 2 happiness and 2 gold per turn, it would take you 15 turns to break even on that unicorn Cothon. Let's also not forget that the Cothon unlocks much later and scales worse than the warehouse buildings, at least until you can afford to get a lot of specialists.

So, how do we nerf the Brickyard and the Saw Pit? Two obvious solutions are increasing their cost and reducing their yields. The first solution is no good for reasons we'll discuss later. There are different ways of reducing the yield of warehouse buildings. The naive method involves just reducing the yield that are added to the buildings and the yield these buildings add to relevant improvements. Unfortunately, there's not a lot of room for adjustment because these buildings have building yield of 1 production and improvement yield of 1 production. Removing the building yield will not make a huge difference as the difference in 1 production makes the Brickyard from our previous scenario break even in 19 turns now, which is still too good. It's difficult to reduce the improvement yield because then we'll have to deal with fractions, which are ugly to look at. Instead of just naively reducing yields, I propose the concept of warehouse capacity. Instead of allowing warehouse buildings to boost any number of relevant tiles, we put a cap of, say, two improvements. That puts a cap on how effective the Brickyard and the Saw Pit can be. We can add ways of increasing warehouse capacity through unlocking certain techs and civics or by having specialists on warehouses increase capacity. This would make Currency more desirable to unlock faster and would indirectly boost science because now, in order to get the most out of these buildings, you can't just afford to stay at the bottom half of the tech tree.

Making production more difficult to obtain isn't the only way to make production-centred strategies less powerful. We can also make it easier to incur production surplus in the early game and make it really painful to have production surplus. Currently, neglecting early-game science and culture is not a big deal because, even if you focus all your attention on maximizing production, it's difficult to run into a situation where you're meaningfully bottlenecked by those yields. One of the causes for this is that things cost too much production to obtain in the early game. Currently, 2nd-tier buildings like Altar and Library cost 90 production. These buildings require 18 turns to build, assuming our capital doesn't add to the 5 production it gets from the Palace. If we increased the capital's production by 5, we'd save 9 turns on constructing one of these buildings. That's a significant difference. If we add another 5 production, we shave off another 3 turns, which is still significant. The first time an additional 5 production doesn't result in at least 1 turn of difference is when we go from 20 production (4.5 turns) to 25 production (3.6 turns). (Yes, I understand that excess production overflows to the next construction and that means this difference still matters. I'll address that later.) This means that you can double, triple and even quadruple your capital's production and still reasonably argue that you should continue chasing more production. Let's look at what happens in an alternative scenario where we've made all the changes that I proposed above, including setting the base production to 10 and nerfing production resources and warehouse buildings. This means that we'll start with a base production of 10 and increment by 3, rather than 5 production, since each unit of production is harder to obtain. We'll also double the cost of buildings, to maintain how long it takes to construct them given just the base production. Let's say how many increments of production we need before any further increment would be insignificant:

10p -> 18t
13p -> 13.8t
16p -> 11.3t
19p -> 9.5t
22p -> 8.2t
25p -> 7.2t

It takes five increments of production to get to a point of production saturation, while it only took four increments to get there in the original scenario. This means that players will still chase production, probably even harder than before, even though we made production harder to obtain. If we were to actually keep the cost of the 2nd-tier buildings at 90 production, despite the doubling of the base production, we'd only need three increments (vs. four in the first scenario) before reaching the point of production saturation, making it less optimal to chase production.

In a similar vein, we can look to fix the production cost of civilian units like the Settler and the Merchant constant with respect to how many of these units you've already created. This will not only further weaken production, but it will also make other deterrents to mass-producing these units more prominent. The settlement limit mechanism provides a significant penalty to over-settling. If we nerf resources as I suggested earlier, aggressive settling makes even less sense. By keeping the cost of Settlers constant, we can expect the settlement limit to become more even more dominant, and consequently, yields that allow you to increase the limit (culture and science) or circumvent the problems that occur when you exceed it (happiness) to rise in value relative to production. Merchants, on the other hand, are typically bottlenecked by resource capacity and trade route range. Allowing these deterrents to prevail means that gold generation and town specialization, specifically the Trading Outpost specialization, become more important. (I won't be talking about the imbalance between cities and towns in this post.) The easiest way to improve resource capacity is to convert towns to cities. A city gets one extra resource slot, can acquire buildings that provide additional slots and are allowed to accept city resources. As long as there are effective deterrents to mass-producing civilian units, letting their costs rise with the number of copies you've already created only serves to ruin the game's balance.

Reducing the cost of items that can be acquired with production, with the aim of making players reach production saturation early, can be an effective way of making production maximization less rewarding. We can take that a step further by making overproduction really hurt. I can think of two ways to achieve this. One is to make surplus production decay over time. Whether it's production that's overflown from the previous project or production that the player has been deliberately not using by force-ending turns, we can decay production not used this turn by a factor of let's say 50%. This should make both production saturation and force-ending much less desirable than they are now since both these actions essentially result in the player throwing away production. Another way to inflict pain on production chasers is to introduce maintenance to early-game military units like the Warrior and the Slinger. As it stands, there's no downside to producing these units and you can sink lots of production into them because you can produce them repeatedly. The only reason you might not be doing this when you have more production lying around than you know what to do with is that you can force-end turns to save production for more impactful things that will unlock in the future. If we just ignore the force-end exploit, there's not only no downside to mass-producing these units, there's quite a lot of upside to doing it, as a horde of early units can help you run over independent powers or other players. Mass-producing Scouts doesn't feel quite the same because there is a rapidly diminishing return on Scouts. If the maintenance of even just 1 gold per turn for every Warrior or Slinger feels too harsh, we can choose to make the maintenance apply only under certain conditions. Perhaps you should only pay maintenance after your 4th unit. Maybe they shouldn't incur maintenance when they're in your territory. This would make it so that there's still no downside to producing these units, but you also don't get much out of them unless you're having to defend against an invasion.

 

[Yzman]

Interesting stuff. How about having food be able to be used for settlers similar to other games? Would that help some of the difference?

 

[rocksinmypath]

Interesting stuff. How about having food be able to be used for settlers similar to other games? Would that help some of the difference?

Yes, because it would remove one way of spending production, making it less valuable, and add one way of spending food, making it more valuable. I don't think this is particularly feasible, though, because this change would essentially require two production queues, and the level of UI complexity that would result in I feel is uncharacteristic of the game. Having said that, Civ 6 incorporated food into settler production by making a city lose population when it finished a settler. This would be a way to involve food without complicating the UI, but with the way population works now, it would be a little weird to impose population reduction as a cost.

If what you're trying to do is to buff food it would be much more complicated. I made no mention of food in my original post because I think food has to be balanced almost completely separately from other yields. The reason food is so weak right now is because of what I call the "growth function", which determines how much food is required to grow a settlement's population by 1. In Antiquity, even if your city was generating 50 food per turn, it would take it 20 turns to grow from 9 citizens to 10. This makes it so that investment into food infrastructure just doesn't make sense in most cases. I was actually going to make my post about how bad the Granary is by showing that, in a situation where you'd think it'd be a good investment, you would need around 60 turns to break even on it. One obvious solution to the food problem is to "flatten" the growth function, but I suspect the food function is one of those mechanisms, like the age reset, that's designed to make early snowballing more difficult. Trying to significantly alter this function could just destroy the game's balance.

The way I would approach balancing food is to change how food buildings work. The reason it doesn't make sense to invest in a Granary is that, if you make any attempt to lean into it and try to get the most out of it by, for example, building more Farms, the game severely punishes you for it. We can lessen the punishment by discounting citizens working Farms when calculating the population of a settlement with a Granary for the purpose of calculating the growth requirement. For instance, if a settlement with a Granary had six citizens and two of them worked Farms, then the settlement's effective population would be five, if we counted each farmer as half a citizen. As I suggested for other warehouse buildings, there should be a capacity limit on the Granary as well, or in other words, there should be a cap on how many citizen discounts the Granary gets you. I would apply similar changes to other food buildings like the Garden and the Bath, where building one of them would result in a decrease of 1 to a city's effective population.

 

[stealth_nsk]

It would be interesting to see some production vs. gold comparison. You could buy a lot of things with gold and do it instantly, you also get gold from towns.

But food and growth probably need some careful balance touches. Currently I focus on maritime food only.

 

[Taefin]

I’m not sure I agree with the production concerns, although do think food is a little weak at the moment, due to the growth function as you cite

The production warehouses are needed to bring tile production up to usable levels, and act as a little more than a small speed bump to doing so, but they also are difficult to optimize while also chasing resources that will usually all use different improvements. If anything, your idea of boosting initial palace to 10 production could help make production less essential in early game. Past early game, you will have already achieved as many production improvements in a city as feasible before specialists start to compete with additional improvements.

Regarding food and growth, my intuition is that a granary has an effect akin to having an extra population in a city than you would without it (since the food it produces x the turns you have had it resembles the scaling cost of growth, at least early on). Thus a granary could be through of as being worth an extra production improvements or the value of an extra specialist. Admittedly it’s probably best to either put all your food improvements in towns, where the farmers won’t count against the growth of the food they produce, or to view farms as a reserve of pops in a city that are to be converted into specialists or production improvements after using them to get a few extra pops (since the food they produce will at best give you 1 extra pop above what you would have otherwise, which has no benefit until you use that pop).

The question then seems to be, do you need to have produced the several extra things you used that extra early production on prior to when a city with farms converts its farms into production improvements and/or specialists, and now has higher outputs owing to its higher population (maybe for 100 turns until the growth curve gets steep enough that the extra food gained this way becomes less than an extra pop). Given that, as you said, the settlement limit prevent as much value from producing more settlers and you can expand to settlement limit rather quickly without maximizing production.

 

[rocksinmypath]

I’m not sure I agree with the production concerns, although do think food is a little weak at the moment, due to the growth function as you cite

The production warehouses are needed to bring tile production up to usable levels, and act as a little more than a small speed bump to doing so, but they also are difficult to optimize while also chasing resources that will usually all use different improvements. If anything, your idea of boosting initial palace to 10 production could help make production less essential in early game. Past early game, you will have already achieved as many production improvements in a city as feasible before specialists start to compete with additional improvements.

Regarding food and growth, my intuition is that a granary has an effect akin to having an extra population in a city than you would without it (since the food it produces x the turns you have had it resembles the scaling cost of growth, at least early on). Thus a granary could be through of as being worth an extra production improvements or the value of an extra specialist. Admittedly it’s probably best to either put all your food improvements in towns, where the farmers won’t count against the growth of the food they produce, or to view farms as a reserve of pops in a city that are to be converted into specialists or production improvements after using them to get a few extra pops (since the food they produce will at best give you 1 extra pop above what you would have otherwise, which has no benefit until you use that pop).

The question then seems to be, do you need to have produced the several extra things you used that extra early production on prior to when a city with farms converts its farms into production improvements and/or specialists, and now has higher outputs owing to its higher population (maybe for 100 turns until the growth curve gets steep enough that the extra food gained this way becomes less than an extra pop). Given that, as you said, the settlement limit prevent as much value from producing more settlers and you can expand to settlement limit rather quickly without maximizing production.

The point is to make production less essential, especially in the early game, so that you're forced to chase science and culture instead in order to unlock more things you can build with production. If you can't build things significantly more quickly with the extra production you can get from something like a Brickyard, either because the extra production represents only a small fractional increase or because things you can build are so cheap that they don't take many turns to build either way, that means the investment you had to make to acquire that extra production wasn't worthwhile.

The production warehouse buildings may be difficult to optimize if your goal is to make close to every single improvement your capital will ever have benefit from the buildings, but you don't actually need that many relevant improvements to make these buildings spectacular investments. If you had just one Mine, you'd break even on your Brickyard in 28 turns. If we assumed that 1 science = 1 production and that its maintenance of 2 gold and 2 happiness was equivalent to 1 production, a 2-adjacency Library would just about be competitive with the Brickyard (90 production / (2 base yield + 2 adjacency - 1 maintenance) = 30 turns). If you had two mines, you'd break even in 19 turns. If you assumed 1 gold = 1 production, which it clearly isn't, you'd need to work 9 desert tiles to break even on Petra in 19 turns (375 production / (2 base yield + 2 * 9 citizens) = 18.75 turns).

My proposal would make it so that, even if your capital had an abundance of Mines, you'd only be able to get a small amount of production from the Brickyard, until you made further investments in the form of unlocking new techs and adding specialists, which would allow the Brickyard to yield more production. Let's look at a specific implementation of this idea:

- The proposed Brickyard has no base yield and has a capacity of 2, which means it will boost up to 2 Mines in the city.
- Adding a specialist to the Brickyard adds 1 to its capacity. Assuming there are already at least four Mines in the city, this means each specialist you can assign to it provides 2 science, 2 culture and 1 production.
- Optionally, make it possible to increase Brickyard capacity by unlocking certain techs and civics. This could replace the current bonus of +1 production to all production buildings unlocked at Masonry II.

The proposed Brickyard requires an additional investment of two citizens, as opposed to one, to break even in 28 turns, and, disregarding the third point above, you'll probably almost never be able to beat this because of how late specialists become available. This should bring the Brickyard in line with other buildings it has to compete with in the early game.

 

[Taefin]

If you had just one Mine, you'd break even on your Brickyard in 28 turns.

I don’t think that this is your strongest point, and so please don’t read this as picking on the easiest target in this argument, but to illustrate a dynamic that I think your analysis is disregarding:

If you break even in 28 turns (or X turns with more mines) that means that you have been operating with less production going toward other things for X turns, and it will take X more turns before you are able to build something of equal cost that you would not have been able to otherwise. The brickyard is an important building if you have mines/quarries, and I’ll build it eventually anytime I have some and the space, and immediately if I can slightly delay what I would build next, but anything you bought before the brickyard (such as a settler, although I recognize your argument that for your capital you can get a brickyard out before hitting pop 5) will also have ROI for every turn earlier it is built.

I think the brickyard works well as is. Its price is not punishing prior to getting its own bonus, and it means that mines are not instantly powerful until you spend 5-6 turns building it, and it competes with other things, like another two units that can do more scouting or disbanding an IP, which can produce better yields than are lost from a 6-10 turn delay of the brickyard.

I’m not a min-max expert, but the omission of those factors leads me to doubt this argument. From another view, I don’t think the fact that a city with mines will always build a brickyard takes away from the interesting choice of whether to build it first.

 

[rocksinmypath]

I don’t think that this is your strongest point, and so please don’t read this as picking on the easiest target in this argument, but to illustrate a dynamic that I think your analysis is disregarding:

If you break even in 28 turns (or X turns with more mines) that means that you have been operating with less production going toward other things for X turns, and it will take X more turns before you are able to build something of equal cost that you would not have been able to otherwise. The brickyard is an important building if you have mines/quarries, and I’ll build it eventually anytime I have some and the space, and immediately if I can slightly delay what I would build next, but anything you bought before the brickyard (such as a settler, although I recognize your argument that for your capital you can get a brickyard out before hitting pop 5) will also have ROI for every turn earlier it is built.

I think the brickyard works well as is. Its price is not punishing prior to getting its own bonus, and it means that mines are not instantly powerful until you spend 5-6 turns building it, and it competes with other things, like another two units that can do more scouting or disbanding an IP, which can produce better yields than are lost from a 6-10 turn delay of the brickyard.

I’m not a min-max expert, but the omission of those factors leads me to doubt this argument. From another view, I don’t think the fact that a city with mines will always build a brickyard takes away from the interesting choice of whether to build it first.

The Brickyard doesn't exactly compete with early units, though. Pottery takes 9 turns to research on Standard speed. Your second Scout takes at most 6 turns to train. If you're lucky with discoveries, you can time Pottery with the completion of the second Scout. In most situations, though, you're probably past the halfway point in training your third Scout before you can start working on the Brickyard. The Brickyard usually only competes head-on with your fourth Scout, but do you usually need that many Scouts? Also, I don't see how you would often end up in a situation where you have to disperse an IP before you have your first Commander. IPs often provide significant long-term value if you decide to develop them into city states. If you do decide to disperse them, you're potentially losing a lot of XP by doing it without a Commander. Besides, in my experience, without any combat bonuses, it's basically impossible to defeat the IP Warrior that defends its base with just a single Warrior of my own. With two Warriors, I have to hope that it doesn't start spawning more units before I can disperse it. I'd say three is the minimum number of Tier 1 units you need to be able to safely disperse an IP. If you're going to commit to the early IP dispersion strategy, you have to sink 90 production into military units before doing anything else like training a Settler. Compare that to going Brickyard first, followed by a Settler. Let's assume we were able to reach 10 production in the capital up to this point with the Brickyard potentially adding 2 more. That means our first Settler's out 10 turns after we started building the Brickyard. In the other world where we start with three Warriors instead, our third Warrior completes after 9 turns. The Settler that follows the Warriors will lag the Settler in the first world by 4 turns. You might be able to time your first IP dispersion such that it significantly accelerates the training of your second Settler. Again, though, that comes with the opportunity cost of lost XP, long-term benefit of befriending IPs (exacerbated if you're picking on a friendly faction), and the outcome also depends on the specific type of IP you disperse. And accelerating the second Settler by a few turns is the most value you can expect to get out of this situation. At this point, you're mostly likely bottlenecked on production, so dispersing a military or gold IP should provide the most value. Dispersion of culture and especially science IPs is suppressed in value because, in this specific scenario, you don't have enough production to be able to quickly build all of the buildings you get from unlocking new techs and civics. If, for instance, we were to use the science we got from dispersing a science IP to research Writing faster, we would still need production to be able to build a Library to benefit from the tech. If we unlock Irrigation instead, that allows us to settle our third town more quickly without going over the settlement limit, but we need 210 production to get three Settlers. The only tech at this point that can get us immediate value without any production investment is Masonry, as Masonry provides +1 production to Mines and Quarries, but in the specific scenario that you chose, we only have one of these improvements, so this isn't all that useful either. You could argue that you don't have to stop dispersing IPs after the first one, but your units need time to recover health, the next IP might be located quite far away from where your units are, and you still incur a lot of opportunity cost for subsequent dispersions.

My point is, your alternative strategy to the Brickyard opener can work in some situations, but when you consider all the opportunity cost you incur and the amount of luck you need to get sufficient value from it, isn't it easier to imagine a situation where instead of just one Mine you can work with your first few citizens, you can have two or maybe even three? Alternatively, if there aren't many such tiles, you can open with a Saw Pit and easily get 3-4 production from it from the get-go. Otherwise, it would mean that you basically just have Farms, Plantations, Pastures and Fishing Boats, and that can lead to a production-rich start that doesn't require either production warehouse building as long as you have plenty of Cotton and Sheep in your capital. But there's no reason to assume that these starts are more common than Gypsum-, Kaolin-, Iron-, Marble-, Jade-, Salt-, Ivory-, or Hides-rich starts.

 

[Taefin]

Very interesting.

Disclaimers, the “one-mine” was only a mention to an earlier part of the conversation, I generally agree production is the way to go in the capital if possible.

I think you’re right that I forgot how long discipline takes to get online without some RNG on goodie huts.

As for dispersing, now that the AI competes for city states (at least with AI mod) I only get a chance to suz two usually, and so non-dispersed ones are often wasted.

That said, the early game has been surprising peaceful lately, even from IP, with the exception of needing to defend town. When early units aren’t needed for defense, it seems like the relevant warehouses should be quick to get online and begin rewarding the investment. If threatened with early pressure from an IP (or an AI, is that even possible) I suppose it could backfire, but it hasn’t happened to me in the last several games.