I LOVE Cities Skylines. I know it seems like I spend more time complaining about games than I spend enjoying them – and that’s probably true in a lot of cases – but Cities Skylines is one of those games that gets more interesting every time I come back to it.
For the last few weeks I’ve been on a huge Skylines bender. I’m sure you’re familiar with this sort of thing. You start skipping sleep. You stop talking to people. You end up with a game on one monitor and its wiki open on the other. Factorio and Dwarf Fortress players will definitely know what I’m talking about.
It got really bad this time. I’d download a dozen mods and buildings to add to the game, play for a few hours, then slam down some junk food while reading the Skylines subReddit, then watch some YouTube videos about the game, then download more mods and start the whole thing over again.
I’m mostly recovered at this point. I think. My rehab counselor said I’m making good progress and I’m allowed supervised access to a computer now. My family is adjusting to having me interact with them again. So I’ve decided to pretend like I was actually working this whole time by making a video about the city-building genre and why I love Skylines so much.
In my work, I’m usually whining about how much better things were in the Good Old Days of gaming, but not this time. This is it. Cities Skylines is the zenith of the city-building genre. City simulation has never been this good. In fact, thanks to DLC and mods, this game is better now than it was when it came out in 2015. And it was already a great game back then.
This isn’t a simple clone of the classic city-building games with a new graphical paint job, this is a deeper and more interesting simulation. I tried going back to old Sim City classics after playing Skylines, and the older games felt shallow and repetitive in comparison.
I want to tell you why I think this game is so good, but before I do that we need to go on…
A Meandering Digression About Non-Games
You know the dumb, annoying argument people have over whether or not something is a game? That debate is really old and worn out at this point, but back in the good old days it wasn’t so polarizing.
Today, calling something a game or not is practically a political statement. Depending on which way you go, people will accuse you of pushing an agenda or of being a gatekeeping asshole. Everyone is either territorial or sanctimonious about it these days, but back before the rise of the Twitter hate machine, this debate was just a curiosity. Asking if random title X was really a game was like asking if the Doom Marine had a mustache. It was fine if you disagreed, because there was no real answer and it didn’t matter anyway.
In the spirit of those happier days, I want to talk about whether or not a city simulator is a real game.
Will Wright – designer of the original Sim City way back in 1989 – actually called his creation a “software toy”. He felt like Sim City didn’t really fit with all the other things we were calling games. He actually had trouble finding a publisher because the Sim City wasn’t anything like the successful titles of the day.
There was no win state except for whatever goals you might have chosen for yourself. There was no definitive lose state in the form of a Game Over screen. While there were challenges to overcome, the system wasn’t deliberately adversarial. Sim City wasn’t like Mario Brothers where your goal was to beat a final boss. It wasn’t like Asteroids where your goal was to survive for as long as possible. It was just a system based on rules, and you were free to do whatever you wanted within that system. Its closest analogue in the real world would be something like constructing a model train, and most people wouldn’t call that a game.
SimCity was about experimentation and creative expression. Legos aren’t a game. An ant farm isn’t a game. A model train set isn’t a game. They’re fun, but they’re not games.
So SimCity isn’t a game, right?
I don’t think it’s fair to say the game is completely devoid of win / loss conditions. If you’re deeply in debt and no longer have the money to sustain your crumbling city, then that kinda feels like it’s game over, even if the game doesn’t say so explicitly. At the same time, the game shows you how much money you have and how many people live inside your city. That feels an awful lot like your “score” in a traditional game. The game never says so, but almost every new player assumes that their goal should be to push both of those numbers as high as possible. That sounds a lot like an implicit goal.
Moreover, the systems of a simulated city don’t really allow for freeform building like a pile of Legos. You can’t just build anything you want. You need to plan out zoning, minimize pollution and waste, balance your budget, and most importantly you need to keep traffic flowing. That’s a lot to keep track of, and it will take you a lot of hours before you know how to construct things optimally. That sounds a lot like a strategy game, with a splash of puzzle-solving. Optimizing strategy and solving puzzles definitely sound like game-like activities to me.
And finally, the old Sim City games had the option to allow for random natural disasters. If they’re enabled, then the game becomes deliberately, explicitly adversarial. Your goal isn’t to just play with a toy train set, your goal is to design a city that can withstand the setbacks the game creates for you.
So is Will Wright wrong? Is Sim City a game or not?
The good news is: Yes. Or no. Actually, it literally doesn’t matter.
The thing is, you can float seamlessly between these two extremes. As a new player, I treated it 100% like a game. Everything was about making the numbers go up. More money. Taller buildings. More people. More prestige buildings. More roads with more lanes. This leads to a very brutalist style of gameplay. You lay down a grid of your best available roads and build as dense as you can, making this ugly Borg Cube for people to live in.
And that works. That’s a totally valid way to play. But over time, I got a little tired of building Megacity One over and over again. I was winning according to the game mechanics, but I started thinking about making the city more beautiful. Not just beautiful as rated by the AI to maximize the citizen happiness number, but actually more beautiful. Well, maybe it’s not fair to call America’s carpet of suburban sprawl and strip malls beautiful per se, but there is a strange satisfaction that comes from creating my own version of it.
And from this point I started looking for techniques for making the city look more real. It was less about filling in the map and making numbers go up, and instead it was more about literal worldbuilding where I designed communities and connections based on purpose and tried to grow a city organically based on the terrain and available resources.
Like most people, if I like a game then I start off being initially obsessed with it, but over time my interest will fade. I might come back to it a few years later, or I might not, but nothing will ever top that big push at the beginning when I’m getting familiar with the game and its systems. With Cities Skylines, this was kinda backwards. I liked it at first, but every 18 months or so I come back to the game and discover I like it even more, or in a new way.
This time around, I leaned hard into the creative side of the game. I installed the infinite money cheat and a ton of community-made buildings from the Steam Workshop. The game was less game-y than ever before. It started off as a strategy game, and gradually became a very large and complicated train set.
The point is that I played Skylines both as a game and as a creative building tool. There wasn’t a clear line between these two experiences, but instead a gradual shift away from mechanical challenge and towards creative building.
The point is that the “Is this a game?” debate is incredibly abstract and it’s not worth getting in angry arguments over it because the lines get incredibly blurry once you take player behavior into account.
Since the start of the genre in the late 80s, these games have been built on top of roughly the same RCI model. RCI stands for Residential, Commercial, Industrial. People live in residential areas and make trips to obtain goods from commercial areas. The commercial buildings get the goods from industrial buildings, which is also where people work. It’s the circle of consumerism. This model does a pretty good job at presenting the different forces at work in a modern city. The RCI model was right there at the start in 1989, and it’s still part of our city-builder games here in 2020. The gameplay has gotten more nuanced over the years, but the same basic design has persisted.
The challenge that the player needs to overcome is that industrial areas are ugly and sometimes create pollution, so people don’t want to live near that. So your first instinct is to zone these three different kinds of buildings in isolated enclaves and connect them with roads. The problem is that those roads will become choke points that will bring your city to a halt as the road system funnels everyone down a single highway or over a single bridge. Add in the need for city services like police, fire, garbage collection, and education, and you have lots of complex decisions for the player to ponder. They need to balance these systems, while also balancing their budget and adapting their design to work with the given terrain.
Developer Colossal Order took this idea and leaned really hard into the traffic management puzzles. Computers finally have enough memory and processing power to simulate individual cars and pedestrians on a citywide scale, and we wound up with a game where puzzles would emerge from the underlying systems. It felt like the genre was finally complete. Like, THIS was the kind of simulation that the games had been suggesting all along, but it wasn’t until the last few years that we had the memory and processing power to make it work.
And before you jump down to the comments to shout at me that SimCity 2013 simulated individuals as well. Yes, it did, but the simulation was brain-dead and nonsensical. A sim would leave their house in the morning and drive around until they found a building with an available job to do. Then they’d leave the job at the end of the day and drive around until they found the first house with an available bed. Also, electricity would invisibly wander the streets, looking for places that need another bucket of energy. It used a lot of processing power to simulate something that made no sense and was of no benefit to the player in terms of gameplay.
In Cities Skylines, every person you see has a name and a home and a job. The game tracks their income and education levels, which dictates where they can live and what they can do. They’re born, they grow up, they get an education, they choose a job, and they eventually die. And along the way they participate in a massive city-wide traffic simulation.
And that traffic system seems to be the real star of the show, here. Sure, you still have to contend with zoning and budgetary problems, but the most interesting problems are the ones that real-world traffic engineers have to contend with.
This game has introduced me to the weird subculture of traffic nerds. You can visit the City Skylines sub-Reddit or the Steam forums and find people discussing the finer points of roundabouts, diverging diamonds, cloverleafs, windmills, and traffic lights.
The game is even a bit educational. I lived in Boston around the turn of the century and I got to experience roundabouts for the first time in my life. At the time I found them really irritating. I thought that trying to read exit signs while also driving in a circle and continuously merging with the traffic around me was absolutely harrowingI doesn’t help that I needed glasses, but the degradation of my eyesight was so gradual that I hadn’t noticed. I was having trouble until reading signs at a distance, which made everything that much more chaotic and stressful.. It’s easy if you’re familiar with the area, but if I’m new to a town then my first trip through the big roundabout is going to be a little scary. At the time I thought they just hadn’t gotten around to modernizing the Boston road system. It wasn’t until I played Skylines 15 years later that I realized just how magical roundabouts are. Sure, that first trip through is nerve wracking, but if it saves me from five minutes at a traffic light every morning, then sign me up.
The game doesn’t deliberately teach you this. It’s just an emergent result of a really well-done traffic simulation. You see a clogged intersection, you make it a roundabout, and you watch as the traffic magically clears. A roundabout isn’t always the answer. Sometimes you need a different solution, and sometimes you’ll have to invent your own.
So that’s why I love this game. Cities: Skylines is a fantastic simulation with just enough detail and scope to have real-world applicability, but also simplistic and playful enough that the player doesn’t get bogged down in the details. At the same time, it also works as a creative building tool with endless modability. It can be either of these things or both at the same time. And on top of all that, the sheer scale of this simulation is staggering. There’s nothing remotely like it out there, although I guess we have the dunces at EA to thank for that.
One last note here is that everyone seems to be assuming that Developer Colossal Order is working on the sequel. Since 2015 we’d been getting DLC every 6 months or so, but as of last year that has slowed quite a bit. The reasonable assumption is that the team is putting their efforts into Skylines 2 now. I hope that’s true. Five years of DLC has left the game feeling a little cluttered and there are some mechanics that could use a refresh. At the same time, I hope the sequel doesn’t come too soon because I don’t know if this site can survive another relapse.
 I doesn’t help that I needed glasses, but the degradation of my eyesight was so gradual that I hadn’t noticed. I was having trouble until reading signs at a distance, which made everything that much more chaotic and stressful.
The Biggest Game Ever
How did this niche racing game make a gameworld so massive, and why is that a big deal?
Raytracing is coming. Slowly. Eventually. What is it and what will it mean for game development?
Let's ruin everyone's fun by listing all the ways in which zombies can't work, couldn't happen, and don't make sense.
C++ is a wonderful language for making horrible code.
Good Robot Dev Blog
An ongoing series where I work on making a 2D action game from scratch.