The problem with production costs in Civilization VI

Leyrann

Deity
Joined
Jan 11, 2015
Messages
4,926
Location
Netherlands
Pretty much since the release of this game - over five years ago now - the production costs have frustrated me to no end in various ways. Back then, I already tried various heavy-handed solutions, like slashing them all in half, and more recently I've taught myself basic modding skills simply to try and implement more sophisticated solutions.

However, the fundamental issue with production costs in this game actually lies deeper, and in a way that I do not think can be solved with simple modding - you either need to do a total overhaul, or hope that Civilization VII will be better. I will actually divide this post into two parts, first giving my thoughts about buildings and wonders and then moving on to the main course, districts.

Also, because I'm me, this post got far longer than intended. I'm sure the regulars from a few years ago won't be surprised.

Buildings and Wonders
My issue with buildings and wonders is quite simple. Have you ever wondered why a Workshop is so expensive, or why a Barracks and a Stable are not priced the same despite being functionally identical, save for which units they buff up?

This is because all building and wonder production costs in the game rely on one thing only: Where are they found in the tech tree? This is why a fancy Market (also known as the Colossus) is more expensive than the amazing Pyramids, and why a Workshop takes a whopping 195 production (a non-GS workshop takes almost 100 turns to pay itself off!). There is no consideration for the actual power level of any buildings or wonders, their use cases, or anything else.

I don't usually criticize developers, but in this case, it is simply lazy design. There are no excuses. It's literally five minutes work to go through the xml-file and adjust the production costs to better reflect the value of the buildings and wonders.

There is also the separate issue that late-game buildings, wonders and indeed also units are, in my opinion, overpriced, often taking as long as or even longer than their early game equivalents to complete, even in cities that were settled early and have had a very long time to develop. Gathering Storm somewhat alleviated this issue with buildings (some of which lost more than a quarter of their production cost!), but not quite enough in my opinion. It also did not address units or wonders, even though they suffer from the same problem.

Districts - The Current Situation
And then we get to the big one. Districts.

First of all, I want to mention that I love districts. The ability to make cities sprawl and grow on the map is something I absolutely adore, and I even remember how in Civ V, before VI had ever been announced, I already used the buildings from Great People to create fake city sprawl.

However, the scaling production cost has been something that I've always hated. When I realized that this issue is intimately related to something else I've always disliked about the implementation of districts, I decided to make this post.

But first, let's consider the current situation. Early game, I have no complaints. You build a district in roughly ten turns, maybe you get some adjacency yields (which could use a bit more flavor/accessibility in some cases, but that's a separate topic), and then you get to add buildings to it. All good.

Then you get to the midgame and... wait a second, do districts take fifteen turns now? Why are they getting more expensive in number of turns even though my cities are increasing their production output? And in this new city I just settled in the late medieval era, the projected time for the first district is sixty turns? How long is it going to take to establish this city?

Usually, at least if you've stopped expanding and only have cities that are already developed, this gets somewhat better in the lategame. However the issue is still present to a degree, and many cities will not be able to dip below that early game standard of 10 turns.

This problem was not present in previous games, and it's incredibly annoying.

Districts - Why were they designed like this?
But wait, wasn't it there previously? Let's load up a save game from Civ IV or Civ V, say from the Industrial or Modern Era, and settle a new city. Now, in our big cities, we're currently building Factories, Hospitals (notable omission from Civ VI btw!), Power Plants and more such buildings; they take a while to build, and indeed have a corresponding high production cost. We check our new city, take a look at those buildings... and indeed, they have a similarly long estimated time to build. Sixty, even a hundred terms.

But in that new city, why would we build them? We're building Markets, Forges, Libraries. Cheap buildings, with early-game production costs that are quite feasible for a developing city, settled late. And as it turns out, by focusing on production buildings, it doesn't actually take all that long to reach a point where a Factory, and then a Hospital or Power Plant, can be built in a reasonable time frame.

So where does the difference come from? Well, if you think about it, those cheap buildings are still present in Civ VI. A Library in the lategame is still 90 production. However, they are locked behind the cost of an incredibly expensive district that might be as much as 400 production.

So do we make the cost of a district a flat, say, 80 production? But no, now our developed cities 1- or 2-turn their new districts, making it absolutely trivial to expand.

Let's put it in clear terms. New cities need cheap builds in order to get started, existing cities need expensive builds to not make continued development trivial. And the separation of cheap and expensive builds is one that depends on game time. In the early game, you unlock cheap builds for your cities to develop, and in the late game you unlock the expensive builds.

We've got that working for buildings. But what about districts? Holy Site? Ancient Era. Campus? Ancient Era. Commercial Hub? Classical Era. Theater Square, or indeed several other districts? Classical Era. Except for the Aerodome, the only specialty district that does not unlock in the first two eras of the game is the Industrial Zone, which is located on an easily rushed early Medieval Era tech.

Based on the tech that unlocks them, all districts should be cheap. But unlike in older instalments in the series, districts are a fundamental part of city development throughout the entire game, and by giving them a flat cost, you trivialize these existing cities. Not only that, this particular issue already exists in the game. If you build a Commercial Hub in the Modern Era in your capital, you then 2-turn the Market. Firaxis has, for Civ VI, chosen to make district costs scale with game era, but leave building costs flat.

Districts - The Solution
I could probably write another post this length just talking about various solutions, but it's gone on long enough so I'll go straight to my own preferred solution: I want to see districts that have a flat cost and unlock throughout the game. Take some of the basic buildings out of districts and into the city center, allow all cities to do some general development, and let them specialize with districts. A Library now gets built in the city center, but you need to allocate land to science by building a Campus (unlocked in the Medieval Era) before you can build a University, Observatory or Research Lab. Industrial Zones now unlock at Industrialization. Commercial Hubs unlock in late Medieval or early Renaissance. Holy Sites still unlock very early. And so on.

By making districts a method of specialization, rather than a requirement to build anything but a monument, granary or water mill, you again allow cities to develop with cheap buildings. At the same time, you get to (partially or fully) disconnect scaling costs from districts. Holy Sites will remain cheap, requiring little infrastructure, and Industrial Zones with all their support will be expensive. But neither production cost needs to scale with the game, and that removes a major source of frustration. And, as a bonus, the kind of exploit-y strategy of locking district costs by placing them early.
 
You're right, buildings should have production cost based on tier, and districts taking eternal to make in new cities (without chop "cheating") is not a fun way to address the snowball issue. Personally I think turning districts into a generic city expansion to improve pop and building space and city defense is an option.
 
Yeah, this is definitely one of my pet peeves of the game as well.

Personally, I think allocating a district should not cost any production at all - but on the other hand, the district itself should not give you any yields either, those should always come from the buildings, which would of course cost production. I guess you can say that's a question of semantics, however, as you can consider the district itself the "tier 0" building, but then as you correctly point out, it's deeply illogical that district cost scales much faster with game than building costs.

I guess another way to frame your idea of more buildings going into the city center would be to rework slightly the concept of a "district". What if a "district" is not something you build in its own, but something that manifests itself once you start grouping similar buildings? I.e. you can build a shrine or a library in all your cities, but once you build a temple to join your shrine, you have created a "holy district", any once you build a university to join your library, you have created a "campus".

Of course that thought spawns another - which I have not really thought through - way to rethink the city planning: What if we rethink the idea of buildings being tied to a district, but instead make it so that each city has a number of land plots available for buildings - starting with one (the city center), and then more being unlockable for a gold cost once city grows. Each of these plots would hold three buildings, but then you decide for yourself what buildings you put in each of these land plots? Say, maybe in city A, you want to put a granary, a watermill and a sewer in your city center, while maybe in city B, you want to put a library, a bank and a power plant in your city center? However, if you group buildings of similar type in the same land plot, these will create a synergy effect - with the downside that building only buildings of one type in your city will make people go unhappy because they need buildings of the other type. Now I'm not sure I would myself recommend such a system, because it probably goes too much into sim-city territory, but I wanted to put out the thought now that I had it.

On a different note, you mention modding, and you may or may not be aware that you can solve much of the district scaling problems with modding (unlike the building costs, that are hardcoded, like you say). You can make districts flat cost (which I tried, and found imbalanced, because adjacency yields are so OP), but you can also have softer cost progressions. Notice that there are two different progression schemes, there's the regular one the specialty districts has, but then there is an "aqueduct" cost scheme which works in some other way. Personally I find that using the aqueduct scheme with a scaling rate of 400 gives a fairly good balance. I have also changed workshops to have AOE like the factories, as have I for arenas.
 
Yeah, I would agree with a lot of the above. Although I will say, it doesn't make sense to at one time complain that costs scale based on what era you research the tech for a building or wonder, and then in the second case propose your solution to do exactly that. But that's only a very minor complaint.

But yeah, I do hope that in civ 7, while I do like having districts available, something different needs to happen with them. I don't mind the notion of them simply being free (or a very cheap) to place, but gain virtually nothing inherrent. Essentially they're just like a city council declaring a zoning issue in a location. Or if they came throughout the eras as noted above and had a fixed cost based on that, I really wouldn't mind that either. That would also arguably simulate real life more, where stuff like libraries or markets really tend to be local or randomly spread around, but more developed buildings like universities and research institutes do sometimes cluster.
 
Yep. Production in Civ6 could need some rework. And I agree, many things don't make any sense at all.

I think the way it's implemented in Civ6 is mostly for Balance, but that backfires if you don't design everything thoroughly. Like I think Disctrict production cost scaling is made so that players can't just spam districts, which is, yes, trivial for expansion, but that also impedes Tall play, and given civ6 is heavy expansion rewarding, that doesn't do any good for Tall Play.

It's hard to come up with Ideas on how to solve the production Issue without testing the Idea in the game, multiple times, with different gameplay styles/approaches, and see if it works in practice. One possible Solution that I have, which I also want to test with 4XP, is:

1) Change how Adjancencies work, by reducing the Adjacency Bonuses Districts get + applying Adjacency Bonuses to the Buildings in the District instead,
2) Make District Cost slowly scale with Game Era,
3) Increase the Cost of a specific specialty District with each Copy built,
4) 20-30% production reduction for Districts/Buildings built in foreign Continents (Colonies) (only during the first 50 Turns since the City was settled)
5) Balance all Buildings' Cost based on the Effects/Bonuses they give and not where they are in the Tree. Tier3 Buildings' Cost increase with each copy built.
6) Apply a City based Cost increase modifier based on number of Districts/Buildings built in the City (excluding the ones built in the CC): if City has no Buildings but 2 Spec Districts, then it will get a 10% Production penalty (5% each) when building a District. If the City has 4 non-CC Buildings, then it will get an 8% Production penalty (2% each) when building a Building. (Note: the penalty only applies while the City is building the District/Building)
7) Heavilly readuce the Cost of Military Units + limit the Number of mili Units you can have at a Time, and base it on Military Infrastructure you have built and Num of Citizens in your Cities (+ other things that can affect this, such as Policies, Governments and Wonders),
8) remove the ability to chop build Units, Districts and Projects (how does shopping exactly help my scientists with launching an exoplanet expidition? :think:). Keep it only possible for Buildings (including Wonders).
9) Add more useful City Projects to the Game, especially in late-game, so Cities don't need to always be working on Buildings/Districts/Units. Or even better: have a slider that directs the production towards working Districts with specialists to increase their output.

Ideally, the Cost could be measured in Turns instead of Production, but that would make production a practically useless Resource and take away from the fun of the Gameplay around it.
 
On a different note, you mention modding, and you may or may not be aware that you can solve much of the district scaling problems with modding (unlike the building costs, that are hardcoded, like you say). You can make districts flat cost (which I tried, and found imbalanced, because adjacency yields are so OP), but you can also have softer cost progressions. Notice that there are two different progression schemes, there's the regular one the specialty districts has, but then there is an "aqueduct" cost scheme which works in some other way. Personally I find that using the aqueduct scheme with a scaling rate of 400 gives a fairly good balance. I have also changed workshops to have AOE like the factories, as have I for arenas.

I actually tried making that very change, but it just caused my mod to break. At first, it just didn't read the file, later on it didn't even want to load my mod anymore. Could you share how you built your mod? Perhaps best to do it here: https://forums.civfanatics.com/threads/quick-modding-questions-thread.637803/page-69#post-16246530

Yeah, I would agree with a lot of the above. Although I will say, it doesn't make sense to at one time complain that costs scale based on what era you research the tech for a building or wonder, and then in the second case propose your solution to do exactly that. But that's only a very minor complaint.

I don't think production costs shouldn't scale. I think they should scale sensibly. Not just a lazy "it's in the n'th column of the tech tree, therefore the cost is x".
 
... (how does shopping exactly help my scientists with launching an exoplanet expidition? :think:)...
Gotta have that space fashion! :mischief:

I actually tried making that very change, but it just caused my mod to break. At first, it just didn't read the file, later on it didn't even want to load my mod anymore. Could you share how you built your mod?
This is the code I use. And then repeat for all valid districts. Notice I thought I had parameter 400, but seems like I changed it to 600 later, which is slightly higher cost I think.
Spoiler :

Code:
       <Update>
           <Where DistrictType="DISTRICT_HOLY_SITE"/>
           <Set Cost="50"/>
       </Update>
       <Update>
           <Where DistrictType="DISTRICT_HOLY_SITE"/>
           <Set CostProgressionModel="COST_PROGRESSION_GAME_PROGRESS"/>
       </Update>
       <Update>
           <Where DistrictType="DISTRICT_HOLY_SITE"/>
           <Set CostProgressionParam1="600"/>
       </Update>

Send me a PM if you want the full code to save you the trouble of writing them all, if you want to try it.
 
6) Apply a City based Cost increase modifier based on number of Districts/Buildings built in the City (excluding the ones built in the CC): if City has no Buildings but 2 Spec Districts, then it will get a 10% Production penalty (5% each) when building a District. If the City has 4 non-CC Buildings, then it will get an 8% Production penalty (2% each) when building a Building. (Note: the penalty only applies while the City is building the District/Building)

I don't think that's a good idea, actually. It makes the player feel like they get punished for those previous buildings. Even just changing it to increasing the cost (rather than reducing the production of the city) will already alleviate that. That said, as shown by Civ 5's science and culture costs, it can still be frustrating to play with.

I like most of your other suggestions though.

This is the code I use. And then repeat for all valid districts. Notice I thought I had parameter 400, but seems like I changed it to 600 later, which is slightly higher cost I think.
Spoiler :

Code:
       <Update>
           <Where DistrictType="DISTRICT_HOLY_SITE"/>
           <Set Cost="50"/>
       </Update>
       <Update>
           <Where DistrictType="DISTRICT_HOLY_SITE"/>
           <Set CostProgressionModel="COST_PROGRESSION_GAME_PROGRESS"/>
       </Update>
       <Update>
           <Where DistrictType="DISTRICT_HOLY_SITE"/>
           <Set CostProgressionParam1="600"/>
       </Update>

Send me a PM if you want the full code to save you the trouble of writing them all, if you want to try it.

I'll give that a try, then. I had every adjustment in a single update, that might be the issue.

Finishing my current game first though.
 
I don't think that's a good idea, actually. It makes the player feel like they get punished for those previous buildings. Even just changing it to increasing the cost (rather than reducing the production of the city) will already alleviate that. That said, as shown by Civ 5's science and culture costs, it can still be frustrating to play with.
Fair enough. I want to improve specialization in the Game (using Governors), so that's actually part of that. But I definitely need to balance everything right and make sure the Penalties don't feel like punishment (or too punishing).
 
It sounds like that but it isn't, or at least it shouldn't be. The Penalties should be just tolerable enough so you can't just build all Spec districts in all of your Cities. That's already the case in early/mid game, but only if you're going wide and not a big issue in the late-game. It's meant as an opportunity cost where you have to decide what Spec Districts the City can specialize on, but if you wanna hav'em all in one city then you gotta invest in that. But overall, in my concept, Cities with few spec Districts won't be a big Issue, but the contrary, with bonuses the less spec districts are in the city, so specialized Cities will be quite powerful, and the Penalties (which aren't that big) are needed to balance that.
But that all is just a Concept that is yet to be tested, we can't know for sure if something would work good/bad if we haven't tested and balanced it.
 
It sounds like that but it isn't, or at least it shouldn't be. The Penalties should be just tolerable enough so you can't just build all Spec districts in all of your Cities. That's already the case in early/mid game, but only if you're going wide and not a big issue in the late-game. It's meant as an opportunity cost where you have to decide what Spec Districts the City can specialize on, but if you wanna hav'em all in one city then you gotta invest in that. But overall, in my concept, Cities with few spec Districts won't be a big Issue, but the contrary, with bonuses the less spec districts are in the city, so specialized Cities will be quite powerful, and the Penalties (which aren't that big) are needed to balance that.
But that all is just a Concept that is yet to be tested, we can't know for sure if something would work good/bad if we haven't tested and balanced it.

Except that bottleneck is working fine. You have to grow your cities to extremely unusual sizes in order to build all Spec districts - and I'm talking figuratively there. The average game, you get to build an economy district (Harbor/Commercial Hub), a victory district (Encampment, Holy Site, Campus, Theater Square), a supportive district (Holy Site, Entertainment Complex, Industrial Zone, Government Plaza, Diplomatic Quarter) and one free district in any city.

A developed lategame city (any city settled before the end of the Medieval Era, practically) is not gated by production, but by population (in terms of building districts). It's the developing cities that are having issues here, and in those cities, building everything isn't a desire in the first place; you're not going to reach the point where that is even a consideration for a very long time.
 
You're right, buildings should have production cost based on tier, and districts taking eternal to make in new cities (without chop "cheating") is not a fun way to address the snowball issue. Personally I think turning districts into a generic city expansion to improve pop and building space and city defense is an option.

I very much agree with this. To me it's nuts that a city settled in a river valley can have all of it's districts on the river but then have a random campus 4 tiles away with farms, mines, or whatever in-between. That kind of city planning is just straight up bad from a real-world perspective and more egriously, makes it way to easy for one city to be "everything". Districts should have to be connected by another 2 districts or the city center (Regardless of what city it is). That way, you keep urban areas more condensed. And with that, I think adjacency bonuses with adjacent districts should be expanded to keep this more equal. By doing that, you create more of an interesting choice-do you make a city in a better growth/defense location while forfeiting the ability of building your district more quickly, or, settle that unideal area fast to get a district up and going even quicker.

And to that end, this is also very affected by production. In my eyes, building a district out in timbuktu should have a MASSIVE logistical cost in comparison to building one adjacent to the city center...and as a result, there should be lowered production costs for all districts around the city center. There would naturally be an increasing cost with each additional district, but it should scale in some way to where that district is placed (And in a new city with just 6 tiles, it would be very low. That would allow late game colonies to get an IZ or CH down quickly and be able to build themselves up in just a few turns compared to the hundreds of turns it takes earlier on...but like OP, techs should also make these start up costs reduced as well.

I just agree that, at least for Civ VII, the district systems should be retained, yet streamlined and made more natural to how a city grows. And cities change too-later game techs should have us be able to move districts around, or remove/modify strategic/luxury resources. With just those two things, along with the better way of spacing out/reducing production costs of districts I talked about above, imagine how well this feature could be implemented.
 
Except that bottleneck is working fine. You have to grow your cities to extremely unusual sizes in order to build all Spec districts - and I'm talking figuratively there. The average game, you get to build an economy district (Harbor/Commercial Hub), a victory district (Encampment, Holy Site, Campus, Theater Square), a supportive district (Holy Site, Entertainment Complex, Industrial Zone, Government Plaza, Diplomatic Quarter) and one free district in any city.

A developed lategame city (any city settled before the end of the Medieval Era, practically) is not gated by production, but by population (in terms of building districts). It's the developing cities that are having issues here, and in those cities, building everything isn't a desire in the first place; you're not going to reach the point where that is even a consideration for a very long time.
Which is true for how the current, unmodded, district system works. I stated in a Post above, that I'm also going to rework how Adjancecies work (and many other things, such as Specialists), which are the main reason why you build 3-4 Spec Districts in a City (because of the instant adjacency bonus). With my changes, you won't benefit much from that alone, since most of the Adjancency Bonuses will come from Buildings.

Well, I think I should have explained my Ideas with Districts in a better way, but there are too many changes and I just wanted to share some suggestions on how to improve them.

And to be clear, I'm not reworking Districts just for the sake of it. Apart from its problems mentioned here, it also restricts the Player to specific (optimal) Playstyles. The rework changes many of that, so the gameplay will be different than in the normal vanilla Game (like adjacencies being more flexible, tall isn't just more buildings, but also more specialists...etc). Maybe some Issues will still persist and/or new ones will show up. I just hope to find a balance for going for any playstyle you desire.
 
In my eyes, building a district out in timbuktu should have a MASSIVE logistical cost in comparison to building one adjacent to the city center...and as a result, there should be lowered production costs for all districts around the city center.
Nicely put and very intuitive argument! This could also create a new game mechanic that can be further modified by certain "logistics" oriented improvements e.g. good old Mass Transit from last iterations of Civ. Certain civs themselves may be more "efficient" at cheaply exploiting more distant tiles.

The ONLY flaw here is that conscious choices will now have to be made to avoid having key work tiles in that cheap inner ring because placing districts renders a tile unworkable (though mitigated by virtue of still being able to collect the resource underneath a district).
 
I read this complaint a lot over the years and I still disagree. It feels to me like people simply don't want to invest in production and then wonder why stuff takes forever. Districts become more expensive yes, but it's not like the production of mines and lumber mills don't go up. And then you have domestic traderoutes which get better and better the more districts the target city has, and another boost with communism. And for international traderoutes you have wisselbanken en democracy. And in the late game you should have plenty of gold so after building that district you can just buy the buildings and be done with it. And if you're really in a hurry you can use Reyna or Moksha to buy districts.
 
I read this complaint a lot over the years and I still disagree. It feels to me like people simply don't want to invest in production and then wonder why stuff takes forever. Districts become more expensive yes, but it's not like the production of mines and lumber mills don't go up. And then you have domestic traderoutes which get better and better the more districts the target city has, and another boost with communism. And for international traderoutes you have wisselbanken en democracy. And in the late game you should have plenty of gold so after building that district you can just buy the buildings and be done with it. And if you're really in a hurry you can use Reyna or Moksha to buy districts.

Let me quote from my own work (aka the post you're disagreeing with):

"New cities need cheap builds in order to get started, existing cities need expensive builds to not make continued development trivial."

You're arguing the latter is working fine. I'm arguing the former is not working fine.
 
Let me quote from my own work (aka the post you're disagreeing with):

"New cities need cheap builds in order to get started, existing cities need expensive builds to not make continued development trivial."

You're arguing the latter is working fine. I'm arguing the former is not working fine.

They kind of come down to the same issue - however production costs are set, it's almost impossible to balance both having cities able to cheaply enough get going, while at the same time making sure that stuff isn't too cheap that it spirals out of control.

Plus, making sure to balance the ever-reaching wide vs tall balance. Like, you could make it so that the first district you build in a city is like 25% of the normal district cost, and suddenly those exploration era cities will actually be able to get that first district. But if you do that, then suddenly you're going to find it's cheaper to build a settler and a first district than to keep growing your megapolis, and now you're back to planning out an ICS strategy.

I'm sure there's a balance that works better than the current system. I'd personally like it if instead of district costs simply scaling 10x from the start to the end of the tech tree, maybe if they scaled like 4X based on techs, but also had an internal multiplier for how many districts there are in the city. Or maybe if they made the current discount for under-built districts scale a little more, so it wasn't simply "cheap or not" but scaled more based on how many there were. So if you settle a late city and you're going to build the first holy city in your empire, it would be almost as cheap as if you built that as the first district in your empire. But if you want to build your 14th campus, well, you better make sure the city can handle it.
 
Top Bottom