The Gamasutra Deep Dives are an ongoing series with the goal of shedding light on specific design, art, or technical features within a video game, in order to show how seemingly simple, fundamental design decisions aren't really that simple at all.

Check out earlier installments, including creating the gorgeous voxel creatures of Fugl, designing the  UI for VR strategy game Skyworld, building an adaptive tech tree in Dawn of Man, and achieving seamless branching in Watch Dogs 2’s Invasion of Privacy missions.

Who: Emil Kraftling, Game Director at Avalanche Studios

My name is Emil Kraftling and I’ve been at Avalanche Studios for the past 11 years in various narrative and game design positions. While Avalanche Studios is best known for making AAA open world action games like Just Cause, Mad Max and Rage 2 -- we have also made several self-published slower-paced experiences including the very successful the Hunter franchise and most recently our 80’s themed guerilla action game Generation Zero, for which I was the Game Director.

What: Environmental Storytelling

Generation Zero was something new for Avalanche Studios. While we had released two self-published games before, they were both fairly niche hunting games and not actively competing with more mainstream action games. But with Generation Zero we had this crazy idea of taking one step in the direction of the more action-packed publisher-funded projects we have made. Turning you from the hunter into the hunted, with a setting inspired by our childhood memories and fantasies growing up in Sweden in the 80s. 

The idea of using our home country Sweden as the setting for the game originated from the fact that the environment assets in Call of the Wild reminded me of the Swedish countryside where I grew up. And as soon as we started talking about it, we just instantly fell in love with the idea of recreating the setting of our childhood. Both in terms of the shadow that the cold war cast over the nation -- the preparation and sense of resistance that became ingrained in Swedish society as a whole. But also in our ignorance and naivety as young people and our shimmering nostalgia of those days. We were living in a world on the brink of destruction, but that didn’t stop us from running around in the countryside playing war with the sense that it was all a bit cool and we could take on anything.

There were risks with the setting however. Our personal experience of Sweden in 1989, a setting which we felt fit perfectly with an atmospheric stealth action "guerilla" game, is not very familiar to the international audience. We would need to not only recreate the setting but also in some sense introduce and explain it to the players. That’s not uncommon of course -- but Generation Zero is and was always a very small scale project. Nowhere near the budget or team size of our AAA titles. We knew we would not have access to the standard storytelling devices of previous games. No cinematics, no NPCs, no scripted cameras and very limited voice over.     

What we did have, however, was environmental storytelling. At Avalanche Studios we often refer to our game worlds as the "main character," and seeing as there were almost no other characters in the game, and you play a non-named character, it felt more relevant than ever. And we definitely had a story for the world. Going by "truth is stranger than fiction," we built the backstory of our world based as much as possible on real-life history - and very often we found that those real-life facts supported the gameplay.

As an example, during the cold war, Sweden built over 50,000 underground shelters, bunkers, factories and military artillery-, radar- and command complexes as a result of fear of WMDs. Portrayed in the game they became excellent “dungeons” and points of discovery for the player. The Swedish military placed isolated readiness weapon storages all over the country in sometimes remote locations or hidden away in barns, in order to quickly arm conscripts where they were and not require timely rallies to exposed military bases. Perfect locations in a guerilla-themed game where you traverse the world looking for weapons and ammo. 

So we followed reality as far as we could before making the inevitable sci-fi jump. Obviously, there weren’t big killing machines roaming around in the '80s --  but if there had been, we’d like to think they could have come about the way they do in Generation Zero.

A key direction from the start of the project was that we wanted players to feel like underdogs. Like they are on a dangerous trek through a world that is familiar and not in ruins but at the same time ominously hostile and where potential foes lurk around every corner. The guerilla gameplay aspect would then come in through weapons, equipment, environments and coop tactics that when deployed could bring you from being an underdog to having the upper hand in any encounter. We wanted you to feel alone, isolated and initially completely in the dark of what has happened in the world.

The game starts with you returning after an excursion and finding the world abandoned. The very first house you see show signs of struggle, and the remains of some kind of dead machine, which is the first time we hint at all to what you will be fighting in the game. 

People seem to be missing, which is an element of the story that we initially included because we didn’t have the budget to do NPCs, but which then took on its own life and provided a strong mystery aspect to the story while also increasing the sense of isolation.

The story of the game was divided. The main mysteries tied to the origin of the machines and the absence of people are handled in missions that are linked and are always provided at key locations in the world we call Command Bunkers. But a lot of other backstories are provided in isolated narrative missions that range from telling of pivotal moments that occurred during the recent invasion, or more mundane events that happened even before but can still lead you to some item, location or weapon that can be useful in the present. In both of these instances, the narrative is either conveyed via left behind notes, documents and recordings or through the dressing of the world itself. An early example   a church where people had sought refuge. Before leaving, they left a note behind on a whiteboard (yes, they actually did exist in the 80s, we have the images to prove it), letting any stragglers know where they had gone to. 

We purposefully avoided overuse of mission markers and breadcrumbing as we wanted players to look at the world and the environments and use them to deduce what had happened and where to go next.

Another mission example have you finding a machine head with a bullet hole in it. You can pick it up and the mission text indicates it was hit by a high caliber sniper round, prompting you to investigate the nearby area for a sniper. Most people would then probably look for elevated positions. There is a high mining tower just nearby, and venturing there you will find the remains of a sniper with some supplies indicating he was held up there for a while before succumbing to one of the nasty small machines (which is still around). 

Aside from these isolated mission narratives, there are also places in the world that are just propped to tell the story of what transpired. Skirmishes that took place in between the army and the machines (with the army losing), cars crashing or queuing up at bridges as they tried to reach safety during the invasion. 

It was important for us to also think of the game world as a very real place. More than just building a map and placing content in it, we designed it as a place where all settlements and villages had names, logic and an element of recognizability. Even our bank of environment assets wasn’t massive, we wanted people to be able to still sense a difference to each of the games’ seven regions. We created some major and minor characters, who while not necessarily shown in the game, still had designated houses where they lived, occupations and workplaces, and relations to other characters in the world. We made sure to put up actual road signs along our roads, indicating distances and directions to other settlements. Both for player guidance reasons, but also for narrative reasons and making the world feel a bit more like a real place.

A final thing worth mentioning is that we decided to commit to authenticity by having everything in the world stay true to the Swedish setting. Thus, signage you find or newspapers and written notes are all in Swedish. 

Recordings you pick up are also in Swedish. Now you get localized subtitles if looking at a Swedish text, and there are both subtitles and English options for the VO of course, as we need everyone to be able to understand - but we still felt that it would have broken the immersion to not have everything in Swedish as the default.

Why: Versatility and value

The main reasons for going heavy on environmental storytelling were obviously partially budget-related, as we didn’t have the tools or resources to produce cinematics. But the decision was also related to the desired experience of you feeling like sole survivors trying to fight your way to some answers. In many ways it made sense to have the world be this big place frozen in time where you have to backtrack the sequence of events by exploring, looking at and deciphering the world.

We also knew that we were making a live game, something we would continue to work on and add to long after launch, so we needed both a form of storytelling that could be easily expanded upon, and a world that allowed “room” for expansion. That didn’t marry well with either ripping narrative out or forcing more classical narrative devices in, whereas it is relatively easy for us to add new environmental storytelling, both in adding and propping locations, but also through adding new missions to the existing areas, buildings and settlements we have established.

Result: An iterative success

For the most part, the environmental storytelling was successful in achieving what we wanted. However, we have learned a lot about how to improve the approach. For instance, the world storytelling in missions wasn’t always successful. In the aforementioned example with the note on the whiteboard, a large number of players spent way too much time trying to find clues in the church, running past the whiteboard as they couldn’t distinguish it easily from the rest of the location propping. It didn’t have a constant sheen or highlight but forced you to look specifically at it to highlight it was, in fact, something you could interact with.

Similar experiences were had in some of the other missions as well, where people were used to mission objects being highlighted more obviously. But while we got feedback about the lack of direction from one group of players, at the same time we had opposite feedback from a lot of people that they liked not being handheld by objective markers in missions and appreciated the slower paced, less breadcrumbed type of missions. Fortunately, due to the game being a live service, we can have a dialogue with the community, listen to feedback and measure it against metrics to address it in a way that makes it an experience that is easier for everyone to progress through, even if it isn’t what they are used to.

We’re particularly happy with that a lot of players have highlighted the atmosphere of the world and the authenticity of the Swedish setting as areas where the game succeeded. The Swedish 80s setting that was both internally and externally highlighted as a possible risk due to being unfamiliar, ended up being a success and seemingly something that felt fresh and interesting to players.