How Against the Storm managed to mix city-building and roguelite play

This interview is part of our Road to the IGF series. The IGF (Independent Games Festival) aims to encourage innovation in game development and to recognize independent game developers advancing the medium. Every year, Game Developer sits down with the finalists for the IGF ahead of GDC to explore the themes, design decisions, and tools behind each entry.

Nominated for Excellence in Design, Against the Storm mixes roguelite elements and city building, challenging players to build and thrive despite ever-imminent disaster.

Game Developer sat down with Michał Ogłoziński, the game's designer and programmer, about how the game was born from thinking on why some city builders felt so endlessly replayable, how thinking of the city as an avatar helped them weave roguelite elements into a city-builder, and how a tiny fictional addition—labeling a certain system the Queen—would turn a possibly-divisive mechanic into a beloved part of the game.

Who are you, and what was your role in developing Against the Storm?

Hi, my name's Michał Ogłoziński and I'm a designer and programmer at Eremite Games. I'm working hand-in-hand with a team of 5 people on Against the Storm, a roguelite city builder set in a world tormented by apocalyptic rains.

What's your background in making games?

My game development career started in mobile games where I worked as a Unity programmer and lead for around six years before ditching them for the sake of working on Against the Storm full-time. Developing Against the Storm has been my dream job every day for over 2 years now.

How did you come up with the concept for Against the Storm?

Back in 2018, we had just finished developing our first full-fledged mobile game (Shattered Plane) which we worked on after-hours for two years. It helped us grow and bind the team together, but we still couldn't allow ourselves to go full-time indie.

We knew we wanted to leave mobile game development behind, so we were looking for game ideas and toying with different prototypes for a new PC game.

We had a few clear goals in mind. We wanted a game with a twist unique enough to make it easier for us to break through the sea of other PC games; a game that would be highly replayable to allow us to work on it in an open development format (we were really hyped on this idea after working for years in a very player-disconnected mobile industry); and a game that would not overwhelm us in terms of scope (this was our first PC game developed as Eremite Games).

At some point, while working on a prototype for an ARPG, we realized it was absolutely out of our scope and ditched it. The whole team was starting to get tired and desperate for some fresh ideas.

For some time, I'd dwelled on the problem of replayability in city builders and tried to figure out why some older strategy games with city-building elements were so replayable (games like Stronghold or Kingdoms and Castles). I vividly remember playing Stronghold: Crusader's "Skirmish Trail" as a kid, where different maps had only some set of the total resource pool. With the trading post at hand, the puzzle was very simple: extract whatever you could, sell it, and buy everything else you needed. I remember it as immensely fun and this feeling of "cheating" the game (I knew very well it was designed that way but somehow I was still feeling very smart nonetheless) stuck with me.

It all clicked together in one evening. We had some experience in city builders, so we could at least assess the scope. The ideas of "Banished meets Slay the Spire" and "a city builder where you play a different game each time" were bold enough and the run-based aspect would allow us to develop the game together with the community. I made a quick pitch, included the world of everlasting rain we were tinkering with in previous prototypes, and got the rest of the team on board.

What development tools were used to build your game?

The game is built in the Unity engine. We had years of experience in it, so it felt very natural and safe. Apart from that, we've built a vast set of internal tools to allow us to make rapid content iterations. The two week update cycle turned out to be quite a beast with an unbridled appetite for quick changes and improvements.

Against the Storm is a roguelite city builder. What thoughts went into bringing those two genres together? How did this mixture shape the game?

Initially, there was a lot of doubt if we'd actually pull it off. One of the other designers working with us at the time kept saying we were trying to mix oil with water. At first glance, both genres stood quite in opposition to each other.

One of the key moments was the coining of the concept of: "the city is your avatar." We quickly realized it couldn't be a game about particular people, but instead, it had to be a game about the town itself—its buildings, production chains, resources, and population.

Reframing it like that allowed us to open a floodgate of mechanics that are very familiar to the players but are not common in the city builder genre. If your town is your avatar, what if it has quests or health? How would it level up? How would it trade, fight, and explore? Most of the key systems like orders, exploration, resolve, and cornerstones originated as the result of this approach.

That last bit led us to another key moment when we came up with the idea of a thick forest surrounding the village. You could now venture into the wilds and explore the world around you. While it took us more than a year to refine this system, it became one of the foundations upon which we built the rest of the game and its unique flavor.

Stability is often important in finding success in a city builder, but Against the Storm is about rolling with constant change. How did that affect how you designed its mechanics and world?

This is actually one of the key bits of design in Against the Storm. Stability is the ultimate goal of city builders and once the player achieves it, we as designers usually have two options: we can either poke the player out of this balanced state or let them get bored. In Against the Storm, the seasons and events do the poking, but ultimately we do what all roguelites are destined to do: we congratulate the player, give them their rewards, and send them away before the sweet taste of steamrolling through the challenges can turn into bitter boredom.

The option to do it is probably one of the most joyful things about designing run-based games. Knowing the player will only have access to a given tool for mere minutes, we can allow ourselves to make this tool ridiculous, overpowered, and very situational. We can encourage players to make risky decisions and push them into trying things they would never dare in a more linear game.

Players will need to care for multiple races with various needs throughout the game. What drew you to create these diverse desires and requirements for players to work with?

Probably the most significant design challenge was to make each game different from the previous ones. To this day, it's our main motivation behind most of the systems and changes that we introduce to the game. At the time, we were on the lookout for some larger system that could become a tool to diversify the experience later into development. Something we could build upon.

The revelation came when Damian Ziomek, our brilliant graphic designer, presented a sketch that depicted five different creatures alongside each other. One of the creatures was a beaver and we instantly fell in love with it (it was late 2019 and we had no idea that Timberborn, another city builder with Beavers, was also being developed here in Poland). The idea that you play each game with a different set of species who have different sets of needs and quirks came very naturally after that.

What drew you to create the demanding queen for the game? What do you feel she added to the player's experience?

At a very early stage of the development, we realized that the game needed to check if your town (your avatar) was doing well. There has to be someone or something that will come and judge you. In strategy games with combat, at some point, you face an invasion. An ill-prepared village in Kingdoms and Castles will be pillaged. On the other hand, a well-fortified town will chase the Vikings away.

I remember I kept rambling about "the King" impatiently waiting for your results while someone shouted "the Queen". It matched perfectly and stuck immediately.

After we released the first demo in September 2020, we realized that players liked this idea very much. By aligning fiction and mechanics, it somehow hid the crude nature of the latter. Mechanically, those were merely two bars: blue for doing well and red for doing poorly. But I vividly remember watching one streamer screaming "Queeen, please!" at, well..., a timer.

If not for this thin layer of fiction, he would probably be screaming at us—the developers—for not giving him enough time or balancing the game poorly. But instead, it was the capricious Queen demanding results. I was stunned by how much can be achieved with such a small change.

There are several different roads to success in the game. What thoughts went into making these varied routes (winning over the people, exploration, serving the ruler) and making them all into viable possibilities?

Getting enough Reputation Points (from various sources) is key to winning a city-building "run" in Against the Storm. Like so often in game development, we initially overscoped and planned six equally viable ways to gain Reputation, but quickly realized it would require many more systems than we could deliver. We settled on three instead and built upon those.

The Queen's Orders system came first, and even now feels like the most natural way to gain the Reputation Points necessary to win. The Resolve, a system that, in a nutshell, passively generates Reputation upon reaching a certain "happiness" threshold, was second. While it took much more time to refine, I believe we managed to make it deep enough to be actually fun and challenging. The third one, gaining the Reputation through Exploration and completing glade events, was initially introduced as a way to ensure the player is never stuck.

No matter how resource-starved the settlement is or how unlucky the rolls are, the player can always open another glade. It can become the final nail in the coffin or it can be a thrilling comeback. Either way, there is always some way to break from limbo.

We are still in the process of improving the Exploration system as it relies heavily on caches found in the forests and tools the player owns. We are also in the long and delicate process of introducing trade as a viable, if more exotic, way to win the game.

The design is still far from perfect and we are well aware of it. We need to improve specialization and find better ways to devalue other sources as you lean into, for example, Orders or Resolve. Otherwise, players may be tempted to get the Reputation from the easiest parts of each source instead of committing to one certain playstyle. This is something we would love to improve upon in the future.

The blights can cause disaster for the player, but they also cause drastic changes in the world. What thoughts went into creating these blight effects to be devastating, but bring about compelling new challenges?

Like in many other cases, it was a combination of a design need to dress up some problematic mechanics and our eagerness to manifest the world we created.

From the very beginning, we knew the player would play games spanning around 2 hours and that there would be a need for a place "in-between". We also knew we wanted to show our world as a breathing and living place. The World Map—with the Smoldering City (the population's last shelter against the Storms) in the center—was deeply ingrained in our fantasy. We just couldn't let it go, even as we realized we would have to find a way to reset this map periodically.

We had plenty of heated debates about this problem with lots of back and forth. No one enjoys watching their 10-20 hours of gameplay go down the sink with one quick animation as the cycle ends and reshapes the World Map. It feels very unnatural in a game about building. In the end, we veiled it with a layer of fiction, well aware that we would have to come back and find some better solutions in the future.

You can also carry things forward from one failed expedition to the next. What ideas went into the things you would carry forward? How did you capture a persistent sense of progress through these elements?

One of the early design pillars of the game was to have a strong aspect of "rebuilding"—bringing something forgotten and ruined back to its former glory. While in the end, the other two pillars (diversity of experience and becoming part of the world) overwhelmed this one, the Smoldering City still echoes this idea. The idea of "City as an avatar" helped guide us here as well. If the city is a hero returning from a dungeon, it feels very natural to spend its hard-earned coins on an upgrade to an attack speed or new shield. Or, in our case, a permanent boost to goods production or villagers' speed.