First Thoughts on Civ VII

I have more time recorded on Civilization VI than any other game in my library. It's been my go-to game to play with Liz since it launched. Before that, I played nearly endless amounts of Civilization II and III, and plenty of IV, Alpha Centauri, and Beyond Earth as well. This is probably the single series of games I have maintained the most interest in across my whole life.

So, naturally, we pounced on the opportunity to pay some extra money up-front for guaranteed access to the first two Civ VII DLCs as well as a chance to play the game a week ahead of its official launch. Of course, these games are always a bit shaky at the beginning, before patches balance out uneven mechanics and DLC adds depth to places that are shallow in the base game. We went into this knowing that we'd need to keep our patience close at hand.

But patience alone wasn't enough to get us through the game as shipped. Civ VII is direly underbaked, missing critical user interface affordances at every turn―most of which have been established standards for generations of Civilization games!—and in some cases being so opaque as to be nearly unplayable. At the same time, it's incredibly ambitious, making major overhauls to the formula that are aimed at addressing flaws that one might consider inextricable from the 4X genre. The result is something that was bound to be polarizing even if it worked.

Dramatic Changes in Civilization VII

The moment-to-moment mechanics of Civ VII look familiar to anyone familiar with the franchise. You still have cities and units on a game board, which still uses the hexagonal tiles introduced in Civ V. You can build buildings in your cities, explore the map, work your way through parallel trees of technology and civics, and pursue either diplomacy or war with other nations you encounter. But the details of almost all these systems have been changed dramatically.

Improvements Without Builders

One of the earliest things one notices upon starting a game of Civ VII is that there's no equivalent to the Builder or Worker unit from previous games. Instead, tiles are improved as part of a city's growth. Each time a city gains a population, it can improve one tile, also growing its borders to include the adjacent tiles (if they aren't too far from the city center). Gone is the choice between mining or farming a grassy hill; each terrain type has exactly one improvement which is automatically applied. Later in the game, you can unlock "unique improvements" which further enhance improved tiles, but these must be built or purchased like buildings.

This immediately gestures at one of the game's core design goals: to mitigate the micromanagement that blossoms as the game wears on. Civ VI already took a gentler stand on this, moving from immortal Workers to Builders that have limited charges. But doing away with this unit type entirely means that the mid- to late-game is no longer bogged down with the weight of choosing actions^[1] for a small army of builders every turn to allow all your cities to keep up with the latest improvements you've unlocked.

I mention this first not just because it's one of the first noticeable changes but because I wanted to lead with a really great design. Builders were compelling in Civ games not by their own merits, but because strategically choosing how to use the available terrain makes the texture of the map extremely relevant and gives each city a unique character. Tying improvements to growth makes sense thematically, hits all the notes that you need mechanically, and provides a huge micromanagement mitigation in one fell swoop.

That's not to say that it's perfect: there are serious issues with the execution of the UI, as there are with every part of the game. It's difficult to tell what benefits improving a given tile will have especially once you start to have synergistic buildings, and it's nearly impossible to know anything about your improvements when you're not actively growing a city. It's an almost unimaginable oversight considering that the "city view" showing details and population allocation has been standard since Civ I.

This points to a much more pernicious goal that appears to pervade this game: simplifying the information presented to the player. It's true that there's a lot of information in earlier Civ games that can feel incomprehensible or even overwhelming to a newer player. But having access to information is also critical to the functioning of a strategy game. You can't meaningfully strategize without understanding the consequences of your choices, and Civ VII chooses to hide that information at nearly every turn.

Districts and Buildings

Perhaps the biggest innovation in Civ VI was the introduction of districts, special buildings which were placed outside the city center and, once placed, unlocked their own individual trees of buildings for the city. Districts added a tremendous amount of depth to city planning largely because of the concept of adjacency bonuses: a district's native yields depended on which other tiles it abutted, including other districts, so optimizing a city involved carefully planning where districts would go to mutually reinforce one another most effectively.

Civ VII brings this adjacency logic to the level of individual buildings. Everything is now placed outside the city center^[2], with two buildings allowed per hex and a requirement that "urban districts"—those with buildings—be adjacent to one another. Most buildings now have adjacency bonuses, and you can accrue additional benefits from placing buildings of the same era on the same hex and even unlock unique bonuses from particular combinations of buildings.

This is, on its face, really cool. Districts already added a ton of texture to city-building, so doubling down on that concept and expanding it to reward other types of geometry is very appealing. It hits what I've always considered to be a crucial strength of the Civilization franchise: its lenticular complexity^[3], the way it provides mechanical depth for players who go looking for it without overwhelming players who are still learning the basics. A novice player can just put buildings wherever and those buildings will still function, while an advanced player can understand the system well enough to plan their building placement many turns ahead and get rewarded for it. It's one of the things that makes these games so fun to come back to over and over: there's always a bit more you can do a bit more cleverly than before.

But lenticular design does exist to some degree in tension with the goals of avoiding micromanaging and simplifying information. In order for a player to be able to engage with the hidden complexity of the game, they need a way to see and manipulate that complexity. When building a city, Civ VII shows very little information about why a building has the adjacencies it does, or what effect it will have on other buildings' adjacencies. It creates a very smart system for hiding complexity and then makes it nearly impossible for those who want it to find it.

Cities, Towns, and the Settlement Limit

The first settlement in a game of Civ VII is a pretty normal city by Civilization standards. The second is something substantially different. All new settlements begin as towns, which means they have no production and can only purchase buildings and units. Any production they would get from improvements or buildings is instead converted to gold. Once they hit seven population, you can choose to "specialize" them, giving some particular bonus and sending all their excess food to adjacent cities instead of growing them further.

Towns can eventually become cities for a substantial sum of gold, which becomes smaller the higher the population is of the settlement in question. Generally speaking, the game is balanced around the expectation that there will be about one town for each city in order to provide sufficient growth and cash flow for the empire.

There is also a limit on the overall number of settlements (towns or cities) you can maintain at once without incurring substantial unhappiness penalties^[4]. This begins at four settlements in the first age, although civics, technologies, and later eras increase the limit over time. It's unclear to me whether the game balance as it stands actually makes it undesirable in practice to go over this limit—certainly I saw Xerxes having twelve cities in my game without succumbing to outright revolt—but the clear message is that the design intent is for this limit to be fairly painful to flout.

Nothing contributes more to the late-game drag of Civilization games than the size of the empire, so the design logic tracks for trying to pull back on the value of spreading out endlessly. In Civ VI the winning strategy by a wide margin was to create as many cities as possible, and a winning game could easily be determined in the first fifty turns simply by being the first to settle a crucial strategic location and use your loyalty pressure^[5] to box out your competitors from the best part of the continent. The settlement limit of course exists to motivate a smaller empire, but towns have a similar effect in essentially halving the number of cities that actively and repeatedly require the player's attention. Without a build queue, they add much less management complexity to the game than a full city.

Such a substantial limit on a civilization's size creates a serious risk, though: at what point is the design undermining something that's critical to the way players enjoy the game? While undoubtedly a fifty-city empire does involve a tremendous amount of arbitrary and essentially-useless choices about what to build, the second X in 4X is "expand" because expanding is fun. And part of what makes the first X, "explore", fun as well is the idea that you'll be able to engage with the territory you explore—something that's much less true if expansion is so limited.

The town system works well for limiting the complexity of a large empire because it reduces the number of decisions that have to be made without reducing the total occupied territory. I don't understand why, with such a powerful tool in hand, the game also feels the need to impose a vastly more arbitrary numeric limit on top. Let players stretch their legs—it's what they're here for!

Diplomacy and Independent Nations

Diplomacy in Civ VI was a bit of a hodge-podge. Some diplomatic actions with other civilizations were just selected from a menu, while some were traded in an exchange view that fundamentally reduced everything to a highly exploitable^[6] gold value. Engaging with city-states used an entirely different system with a different currency, envoys. Engaging with world congress in later ages used yet another system with yet another currency, diplomatic favor.

Civilization VII unifies all of these under a single currency: influence points, which represent your diplomatic influence throughout the world. All menu options with other civilizations and even independent nations (which replace older Civ's barbarians, and which eventually mature into city-states once a civilization has engaged with them sufficiently) are now mediated by these points. There is no more direct trade screen at all, with collaborative projects (which of course cost influence) now the only way of sharing or transferring resources between civilizations.

This is a much cleaner system, and it makes accumulating influence feel valuable and exciting in a way that none of the individual currencies did in Civ VI. Engaging with independent nations and choosing to either wipe them out as barbarians or help them blossom into city-states is charming, although all city-states of a given type now allow a choice from a small pool of suzerain bonuses which is a shame relative to the unique bonuses of Civ VI^[7]. But on the whole, this is a very positive change, adding a lot of depth to the question of where and when you spend your influence and how much you prioritize it relative to other resources.

Ages and the Great Reset

Now, finally, we get to the biggest of all the changes, the true defining structure of Civ VII: the age system. Unlike in previous games, where the most well-defined an "age" got was Civ VI: Rise and Fall's concept of era score leading to golden or dark ages, Civ VII draws such a sharp line between each age that it's fair to describe each one as being almost its own game-within-a-game. Each age has its own tech and civics tree which does not connect to that of the age before or after. Each age has a handful of unique mechanics which aren't accessible before or after, and which feed into quests called "legacy paths" which provide substantial bonuses for the next age. Each age even sees you choose a new civilization, although your leader remains the same.

And, crucially, each age resets the game board.

This isn't entirely true. You keep all your existing settlements, although only one remains a city and the rest revert to towns at the age transition. You keep your improvements and buildings, although most buildings from previous ages have dramatically reduced yields without reducing their upkeep cost at all^[8]. You'll be granted a few new-era military units in exchange for your old ones, although by no means on a one-for-one basis. But your trade routes are severed, your wars are ended, your exploration is curtailed, and all the research and culture you pursed becomes irrelevant.

Judging by the feedback discord I've been reading, this is by far the most controversial change in the game. At the same time, there are real benefits. The distinct ages allow for substantially different mechanics across them, such as the treasure fleets that define the exploration age. The hard resets do a lot to reduce the amount of micromanaging one does as the scope of the game balloons out of control. And, crucially, the separation limits how possible it is for a player to snowball.

Realistically, in older Civ games, if you placed a bet seventy turns in that the leading player wins the game, you'd be right the substantial majority of the time. Resources beget territory, territory begets resources, and it's hard for a player who's winning today to lose tomorrow. And Civilization games are long. How many hours can you just go through the motions of completing a win you already locked in before it's not enjoyable anymore? At what point is it worth dramatically mitigating the overall effect of the first two ages exchange for giving the third^[9] any relevance at all?

I honestly don't know the answer to these questions yet. Part of the problem, as I mentioned in an earlier footnote, is that going through the motions is fun. Rolling your snowball civilization downhill as it inevitably grows larger and larger and crushes all other snowballs beneath it is an intrinsically joyous experience, but it's also one that's bound up tightly with slog and sameyness. I'm impressed that Civ VII took such a big swing to disentangle those, and I'm a bit heartbroken that it did so with a UI so inadequate that it will undoubtedly cast a pall on the more ambitious and thoughtful aspects of the design.

These questions can't realistically be answered without real experience, and right now there's no way to have that experience without slamming into unrelated unambiguous failings. I believe there is a way to pull back on the micromanagement complexity and the late-game slog of the Civilization formula without throwing the baby out with the bathwater. I believe most of the novel designs in Civ VII are really smart steps in that direction. But as it stands, I can't judge it holistically because it's barely playable at all.

If I Ran the Zoo...

If I had to make the call right now, I'd keep most of the changes in Civ VII: the builder-free improvements, the explicitly-placed buildings, the towns, and the diplomacy points. I'd get rid of the settlement limit in exchange for something gentler, like a penalty to net overall yields that limits but doesn't erase the extra gains from each individual city. I'd reinstate unique suzerain bonuses, correlating with each independent city-state's name, with a way to see what the bonus will be before deciding to befriend them.

And, as I see it today, I would choose to do away with the strict age divisions. It's possible that after a handful of patches, 2K will find a way to thread the needle and make this system scratch the same itch as a campaign with full continuity, but I just don't think that's likely. It's too tightly targeted at dissipating the snowball effect, and the snowball is too much what I love about these games. Which breaks my heart a bit, because game development practicalities make it essentially impossible to remove this: it's too tightly wound around the design of the civilizations, the age-specific mechanics, the tech and civics trees, and on and on. Civ VII without it would essentially have to be Civ VIII.

Which I guess, now, is what I'm waiting for. Liz and I will go back to VII for sure and give it another honest try in a few months. Maybe we'll even play a number of games through. But unless they manage a miracle, even once the UI issues are fixed and the game is clearly what they envisioned it, it's hard to imagine that being the sort of game I'd put another 900 hours into.

---

1. Earlier games mitigated this micromanagement overload by allowing immortal workers to be automated, so that they'd choose their own improvements and gradually make your cities better without direct intervention. This obviously doesn't work nearly as well once you introduce Civ VI-style districts which put a lot of pressure on planning ahead for what a city will look like, but even worse than that it undermines the concept of the strategy game as a whole. If the best option for workers is to not strategize, then what value are they adding to the player's life? ↩︎

2. With the exception of a single building slot available adjacent to the palace. ↩︎

3. I used to be a huge reader of Mark Rosewater when I was younger. Since then, I've soured on a lot of his design philosophy, especially after seeing so many situations where he's done an about-face without really grappling with why his thought process was wrong in the past. But I think this concept in particular is extremely valuable across contexts for understanding how (and why) to make something that's appealing for a range of experiences, investment levels, and needs. ↩︎

4. Specifically, each settlement gets -5 happiness for each settlement you have over the limit. What specific effects low happiness has for a city beyond incrementally reduced yields are opaque because of the game's characteristic inability to explain anything at all. ↩︎

5. Another thing that simply doesn't exist in this game. The only disincentive for settling right in the midst of another civ's territory is the diplomatic cost of pissing them off. If someone else settles near you, you can either go to war or accept it, which cuts off one of my favorite avenues for non-violent competition in older games. ↩︎

6. One of my favorite tactics when playing Civilization VI against AIs was to become the world bank. Unless they disliked you, every civilization would be willing to sell you a thirty-turn annuity of one gold per turn in exchange for 21 gold up front, which for those keeping track at home is a 42.9% return on investment. Doing this again and again while maintaining good relations would put the entire world deep in your debt, giving you functional control of the world GDP and making winning essentially trivial even on the hardest difficulties. The fact that the game was still fun with this laughably explotative mode of play is a testament both to the quality of the core concept and the fact that optimizing strictly for balance isn't necessarily the right call for a 4X game. ↩︎

7. Something that feels cool across many different games of many different genres is the idea of a "blank card": a class of game object that can have any mechanical effect that the designer can dream up (and, for video games, program). It's a well of essentially endless depth—actual card games have been plumbing it non-stop for decades—that can make a relatively simple set of base mechanics feel endlessly complex and novel. Civ VII still retains tons of blank cards (buildings, wonders, policies, leaders, and so on) but city-states represented a particular intersection of being randomly present on the map while also making it possible to work towards a known outcome that produced some very fun play patterns. ↩︎

8. This is, in a sense, a solution to a problem posed by the new adjacency-focused building mechanic. A typical Civ VI game would see late-game cities with building and district counts in the high teens or twenties—far more than could be placed two-per-tile while still maintaining a decent number of improvements. Making older buildings clearly outmoded and encouraging their demolition and replacement helps preserve space without having to limit the number of available buildings. ↩︎

9. Yes, there are only three ages in Civ VII: Antiquity, Exploration, and Modern. The medieval era is the most conspicuously absent, but there are also no distinct classical or industrial eras. This has essentially no mechanical bearing, but it cuts deep at the hearts of long-time fans of a franchise whose core promise is playing out the entire development of a civilization. ↩︎