City-builder game
A city-builder or city-building game is an umbrella term for any computer game which allows the player to simulate building and managing a city. They are part of the larger genre of construction and management simulation games, and are particularly closely related to transport simulators like 1994's Transport Tycoon, its open source recreation OpenTTD, or the newer indie game Mashinky. They are also conceptually and historically related to 4X games like Civilization. Usually, city-builders have minimal competitive elements and loosely defined or non-existent win and loss states. A player usually designs a city only to achieve a personal goal, and ends the game when they get bored of it and decide to start a new city. Though there may be indicators of city health, like prosperity or happiness, many city-builder games have no "score" to speak of.
City-builders vary in their setting and mechanics, but a general division can be drawn between games set in a somewhat realistic approximation of a contemporary city and those with more stylized fantasy or historic settings. A secondary division can be drawn between desktop/console games and more casual phone games with simpler mechanics. Charm Cities is intended primarily as a realistic, contemporary desktop game, and this article will therefore focus on that subset of the genre.
Contents
History
Before 1990
City-builders are among the oldest video games: text-based mainframe games with recognizable city-builder elements date from the 1960s. The first graphical game that might be called a city-builder was 1981's Utopia, a strategy game in which two players competed to build a more perfect society. Although the game was competitive and involved some minimal military aspects (limited to funding rebellions or using a PT boat to chase fishing boats away), the focus was on creating a happy and healthy populace. Although largely forgotten today, Utopia is credited with laying the groundwork for later development in at least three genres: real-time strategy, 4X, and city-building.
However, the most influential city-building game to date is, without question, SimCity. Developer Will Wright began work on the game as early as 1985, when he was working on another game but discovered that the level editor was, in fact, more interesting than the game itself. But in a market dominated by action and arcade games, publishers hesitated to distribute a game without so much as a clearly defined goal, much less the ability to win or lose. Finally, in 1989, Brøderbund agreed to distribute it - though they stipulated that Wright include a series of timed scenarios with explicit goals. After a slow start, SimCity found its audience and became a commercial and critical success, launching Wright's career.
SimCity was heavily inspired by Wright's own research into urban planning and system dynamics and it set out the basic urban design principles and gameplay mechanics that just about every contemporary desktop city-builder since has followed. The player sets out zones, which are either residential, industrial, or commercial, which provide locations for "Sims" to build housing and workplaces for their city. They must then provide services, like power and police, while avoiding problems like traffic. The more their needs are met, the happier and more affluent the Sims of the city will be, and the more likely they are to build high-value buildings in the designated zones. The player can set a tax rate, which gives them an income based on the value of their city's buildings. The city may occasionally face disasters, such as flood, fire, or even monster attacks. The game's code was released under the GPL free software license in 2008 (though the name "SimCity" is trademarked), and multiple modernized versions exist under variants of the name Micropolis, including a browser-based HTML5/JavaScript version called micropolisJS. Even 30 years later, the game mechanics will seem extremely familiar to anyone who's played another city-builder.
The 90s
One of the earliest fantasy/historic city-builder games was the original Caesar in 1992. It borrowed heavily from SimCity but changed the setting to ancient Rome and added a great deal of complexity, including the requirement to fund unseen military expeditions. Caesar was also a success and spawned three direct sequels, Caesar II (1995), Caesar II (1998), and Caesar IV (2006), as well as an entire series of similar historical games set in Egypt, China, and Greece. Perhaps the franchise's most enduring legacy is the fact that the phrase "plebs are needed" has popped unbidden into my head once a week for the past twenty years.
Will Wright's company Maxis, meanwhile, had been struggling to find a successful follow-up to SimCity; other efforts, like SimEarth, SimAnt, and SimLife had received positive critical responses but failed to achieve the same kind of breakthrough commercial success. In 1995 they went back to the well with a direct sequel, SimCity 2000, and achieved the success they were looking for: SimCity 2000 sold 4.2 million copies and is now seen as one of the most influential games of the early 90s. The interface and graphics were dramatic improvements on SimCity, with an isometric projection that offered the illusion of a 3D city while still using 2D sprites, maps with terrain and elevation, and underground layers. Gameplay was expanded dramatically, with many additional services and more nuanced budget controls.
Despite the improvements, the basic gameplay of SimCity 2000 was not much different in principle from its predecessor. Players still created residential, industrial, and commercial zones (though they could now choose between "low density" and "high density" zones) and tried to encourage Sims to build in them by providing services that raise land value. There was a loss state (extreme deficit spending would get you a "game over") and something vaguely resembling a win state (if you built 300 Launch Arcologies, the most expensive buildings in the game, they would launch into space), as well as more scenarios, but again, most games lasted for exactly as long as they held the player's interest.
SimCity 2000 was also packed full of charming Easter eggs and features. The game is full of hidden Loch Ness monsters, calling moose, Neil Gaiman essays, Superman, and other treats. One of the more fascinating features - almost entirely unique to SimCity 2000 - was the newspaper, which had articles on a number of topics that ranged from "a power plant is reaching the end of its operational lifespan" to "frog convention". This sense of whimsy and fun set the stage for later Maxis games, including the first city-builder aimed at casual players, SimTown (1995), and their 2000 hit The Sims, one of the best-selling video game franchises of all time.
Another key innovation of SimCity 2000 was a toolkit released in a 1994 expansion called the SimCity Urban Renewal Kit, or SCURK. This tool, which allowed users to modify and create sprites for buildings, can be seen as the birth of city-builder modding culture.
The rest of the 90s saw a series of new historical, science fiction, and fantasy games, like Anno 1602, The Settlers franchise, Outpost, and Afterlife, as well as new releases in the Caesar franchise. It also saw work on the earliest open-source city-builders, notably Lincity, started in 1995 as a clone of the original SimCity for Linux. Interestingly, it features not one but two win conditions: either a space launch, as in SimCity 2000, or a certain level of sustainability in development. Though development on Lincity stopped by 1999, it would be resurrected around 2005 as Lincity-NG, with isometric graphics. Lincity-NG is hosted on GitHub but the community is currently mostly inactive.
Two other late-90s Maxis projects are worth mentioning, though neither was a city-builder and both were pretty bad: SimCopter and Streets of SimCity. The first was a flight simulator in which you piloted a helicopter around a city doing various missions, and is remembered primarily for an unauthorized Easter egg inserted by a programmer working with anti-consumerist activists RTMark in which flash mobs of scantily clad men assembled occasionally appeared to make out with each other. (One of the best sentences on all of Wikipedia: "Their fluorescent nipples were drawn with a special rendering mode usually reserved for fog-piercing runway landing lights, so they could easily be seen from long distances in bad weather.") The second is a vehicle racing game with no scantily-clad men worth mentioning. Both were critical and commercial failures. However, the truly remarkable thing about both is that they could import SimCity 2000 savegame files, allowing you to experience your cities in full 3D for the first time. It would be most of a decade before full 3D city-builders became the norm.
In 1999, Maxis released SimCity 3000, which included several new game mechanics but failed to make the same lasting impact on the genre. It remains the only game in the genre to have medium-density zones in addition to low-density and high-density. It featured the ability to make deals with neighbor cities to buy and sell power, water, and waste management services, as well as deals with private entities to build controversial buildings that lower land value in exchange for cash. Land value in general was modeled in a much more detailed fashion, creating much more defined neighborhoods with a distinct class structure.
The 2000s
The rapid increase in consumer use of broadband around the turn of the millennium - and the surprise success of the massively multiplayer online role-playing game EverQuest - ushered in the idea that every genre could be massively multiplayer. The first city-building MMO, 2000's StarPeace (later Legacy Online), was a bold attempt by a remarkable Cuban-Canadian team of indie developers to create an entire world with thousands of players all competing to build their own cities. However, multiplayer is expensive to host, and city-builder games can be far more resource intensive than role-playing games. Despite decent reviews, StarPeace was an overall failure, mostly because it failed to achieve the customer base required to make an MMO either fun or profitable.
A few more successful games cropped up in the early years of the decade, notably Tropico, a game in which you play as the dictator of a Caribbean banana republic during the Cold War and must manage the various factions among your citizenry as well as maintaining relationships with the USA and USSR. The game spawned a franchise currently up to the sixth game.
2003 brought SimCity 4, the first 3D version of the game, and another success for Maxis. Perhaps unsurprisingly, gameplay expanded on some of the ideas in SimCity 3000, notably by including even more complex relationships with neighbor cities. SimCity 4 also allowed players to import characters from The Sims to live in their cities. It had the franchise's first expansion pack, Rush Hour, which added multiple road and transit options, the ability to query which routes Sims are using, and the ability to drive vehicles through the city, as in Streets of SimCity. SimCity 4 also expanded the modding community even further.
2006 brought three new games. First was Tycoon City: New York, a somewhat unusual entry in the contemporary city-builder genre in that players focuses on maximizing the growth of individual businesses but can't place roads, and are limited to attempting to build Manhattan; reviews were mixed.
That year also saw the release of Dwarf Fortress, an indie game with no graphics at all - the entire world is rendered in top-down view using colored ASCII text characters, as in a classic roguelike game - but a complex procedural world generator and AI system. After almost two decades of total development, the game's creators consider themselves almost halfway to version 1.0. The game is a cult classic, famously challenging (the unofficial motto is "losing is fun!"), but full of unexpected emergent behavior that creates real stories out of the lives of the dwarf characters. Though not a traditional city-builder, it contains some of the same elements.
Finally there was City Life, one of the first games to try and muscle in on SimCity's market dominance among contemporary, realistic desktop city-builders. In addition to the typical gameplay mechanics of SimCity, City Life added in a class system, in which every citizen was a member of one of six classes: Elites, Suits, Radical Chics, Fringe, Blue Collars and Have-Nots. Citizens moved freely between classes as the city developed. The Have-Nots were the poorest and least demanding class, the Elites were the wealthiest and most demanding, and the intermediary classes offered two "tracks" from Have-Not to Elite: high levels of education and retail might cause Have-Nots to evolve to Fringe (i.e. working artists) and then to Radical Chics (i.e. wealthy bohos), while physical safety and industry might cause them to evolve to Blue Collars and then to Suits. Persistent unemployment or lack of services could cause citizens to "devolve" back down the ladder. Literal class war was possible, in that citizens of one class might riot if citizens of another class they disliked lived too close. The game received mixed reviews, with one reviewer saying "the solid city-building engine still has legs - just nowhere to walk."
The SimCity franchise returned the following year with SimCity Societies, a major departure from the series to date, and the first SimCity game to fail to impress critics. It was not, in fact, created by Maxis (who were working on Spore), but by Tilted Mill, who had previously made two of the later games in the Caesar franchise. It removed most of the classic SimCity gameplay elements, such as power and water, zoning, and building evolution, and replaced them with a new mechanic in which players had to choose the "societal values" of their city. The city's appearance and soundscape changed as those values changed: for example, promoting the value of "authority" might cause security cameras to begin appearing in-game. The game did not merely include the ability to add mods, but in fact was designed to be modded. Despite its widely praised ambitions and innovations, Societies failed to land; reviewers felt that it was too easy, oversimplified, and had performance issues.
In 2009, the first free and open-source city-builder to use a 3D engine, OpenCity, was released, and version 0.0.62 is still available, though the project is not under active development. The creator and lead programmer was Duong Khang Nguyen, with 3D graphics by Frédéric Rodrigo. It was downloaded over 190,000 times and was bundled with many Linux distributions. The gameplay is broadly similar to the SimCity franchise, with zones that develop as their needs are met.
Also in 2009, Cities XL, a sequel to City Life, was released. It drastically scaled down the class-based socioeconomics of its predecessor down to "unskilled workers", "skilled workers", "executives", and "elites", and overall was a simpler, more SimCity-like city-builder game, with the major innovation being that it was originally intended as a subscription-based massively multiplayer online game. However, the multiplayer element only lasted for a few months, due to a low subscription rate. Transport options were very limited; the game didn't get buses until a patch two months after its release. However, it did allow for curved roads, something that the SimCity games didn't at the time. Two follow-ups, Cities XL 2011 and Cities XL 2012 failed to address the lingering core problem that the game was originally designed for multiplayer first, with single-player something of an afterthought.
Current Generation
In 2013, Maxis launched what was supposed to be their defining city-builder game, simply called SimCity in an apparent effort to make everyone's life just a little bit more difficult. It was the first major SimCity game in a decade, and was hotly anticipated. However, the decision to add multiplayer once again caused serious problems. In particular, the game required an always-on internet connection even for single-player games - mostly for DRM reasons, not for any gameplay function - and at launch there were nowhere near enough servers allocated to handle the required traffic. Players couldn't connect, games froze, and save files disappeared. The launch failures more or less doomed the game; reviewers, who had been quite positive in pre-release, adjusted their reviews downward because of the nearly unplayable state of the game at release. Furious users also had their say; at one point, the game had 1/5 stars on Amazon, and players filed a petition with the White House asking for more consumer protections for games that used persistent connections for DRM.
The actual gameplay of SimCity is basically unchanged from SimCity 4, apart from the multiplayer features. One small innovation was that existing user-placed buildings could get additional modules added on after construction, like an extra garage for a fire station. Another interesting change was that roads automatically carried power, water, and sewage. The major change is in the interface and in the simulation engine. The interface, loosely inspired by Google Maps, changes based on the user's activity, showing the most relevant view of the world for whatever the user's current action is. The simulation engine, called GlassBox, didn't work top-down from high-level statistics but rather bottom-up: it modeled individual Sims as dynamic entities, each with their own AI, and then saw what happened in the city as Sims and resources moved about - not unlike a scaled up version of Dwarf Fortress. Unfortunately, the Sim AI was rudimentary at best; each Sim simply took the first available job and the first available home. However, this bottom-up simulation concept would become the standard for the current generation of city-builders.
The historical city-builder/survival simulator Banished was released in 2014 by indie developer Luke Hodorowicz, inspired by the Anno 1602 series. It too treated citizens as individual agents, though with a much smaller population. Though the original game was relatively small and self-contained, and received mixed reviews, the game had a major advantage over SimCity: an active mod community. Maxis had decided to ban SimCity mods that changed the gameplay, but Banished thrived due to just that sort of mod.
Meanwhile, the long delay after SimCity 4 and the disastrous launch of the 2013 SimCity left an opening for other contemporary city-builders as well. The creators of Cities XL came back with yet another attempted fix with Cities XXL in 2015. This was functionally the same as Cities XL, with few new features, and was by far the worst reviewed game of the City Life franchise.
But 2015 also saw a brand new entrant in the ring, Cities: Skylines. Its developers, Colossal Order, had previously developed a transport simulation game, Cities in Motion in 2011 and its sequel in 2013; both games were about creating a public transportation network in various European cities. As a result, they had already developed both a fanbase of train wonks and a large library of relevant assets and 3D models. They had thought about developing a full city-builder for some time, but only after the failure of SimCity did their publisher give them the go-ahead to start work.
Cities: Skylines is generally considered the finest city-builder game of the current generation, and has far outsold the 2013 SimCity, with over 6 million units sold by March 2019. It uses a similar strategy to SimCity's GlassBox engine, modeling thousands of individual "cims" (their legally-distinct-from-Sims name for their citizens) as agents, but with a dramatically improved AI. The traffic modeling is good enough that it's been used in actual urban planning. The graphics are also probably the finest in any city-builder to date. The game is designed to be modified - Colossal Order has a full-time programmer working on modding tools - which has led to a thriving fan community. No less than eleven expansions (eight of them major) have been released - far more than for any previous city-builder - offering both a continued revenue stream and a way to keep the fanbase energized.
However, Cities: Skylines gameplay is not significantly different from the 2013 SimCity. For that matter, it's not all that different from the 1989 SimCity, either. The basic mechanics remain unchanged. Some of the DLCs have expanded on traditional city-builder mechanics, by allowing more freedom in designing features like parks and universities, but the general principles of the genre remain the same.
Future Prospects
There is no reason to believe that a new desktop SimCity game is in the works. After the flop of 2013's SimCity, Maxis used the SimCity codebase to create SimCity: BuildIt, a simplified casual mobile city-builder, the following year. While SimCity was a critical and commercial failure, BuildIt is now the most played SimCity game ever, and one of the most played phone simulation games ever. It's a "free-to-play" game, which is to say, a game subsidized by that portion of people who become addicted to it. Given the clear contrast in profitability between the desktop game and the mobile game, it's hard to imagine Maxis taking the risk of trying to develop a new desktop game any time soon, especially given the ongoing popularity of Cities: Skylines. It seems that they have decided to cede the desktop genre to Colossal Order (who don't develop for mobile) for the foreseeable future.
The City Life/Cities XL/Cities XXL franchise has received worse and worse reviews with each new installment, and in any case Cities 3XL is a terrible name. A total reboot is not impossible, but seems unlikely. Cities: Skylines, meanwhile, doesn't feel particularly dated, despite being nearly five years old, and Colossal Order is successfully releasing one or two new DLCs every year. A new release is possible but doesn't seem particularly necessary at this time.
It would be hard for any game to compete with Cities: Skylines on graphics or on how well it executes the core mechanics of city-builder games any time in the near future. It stands to reason, then, that the next innovation in the city-builder genre won't be in graphics or core mechanics, but in something else entirely.
