How Firewatch was created

The game designer of Firewatch told what difficulties he faced when creating the game. And why the term “walking simulator” cannot be applied to this project.

The designer of “one of the most successful walking simulators of 2016” wants people to stop calling him that. His lecture entitled “Limitations of game design in story research games”, which was held as part of the National Game Summit in Montreal (Montreal International Game Summit this week), Nels Anderson from the studio Campo Santo was preceded by criticism of such a name for this genre.

“In my opinion, the term [walking simulator] is very wrong. It makes me sad when people use it,” Anderson said. – Partly because it’s such an inside joke. Initially, it was used as a derogatory reference to games like farmer simulator, or trucker driver simulator, or something like that. And now people are kind of trying to rehabilitate this term. But if you don’t understand this, then you don’t need to use it to try to describe the game, it makes zero sense. Given that, as a rule, simulators in general are more accessible to people who usually do not play games, especially first-person games, it is not necessary to use this stupid and false term. It’s kind of dumb. Because the point of these games is still not that you have to go to them.”

Regardless of what to call this genre, it imposes certain restrictions. Designers need to take them into account. Anderson highlighted five key points from his experience working on Firewatch.

Genre limits the freedom of a game designer

The first of them is that game designers will definitely feel that the genre limits them. Other developers can rely on gameplay, the need to shoot enemies or solve riddles when creating a story. In the case of a story research game, the gameplay is mostly limited to the plot.

“Exactly how players will interact with the gameplay is determined by the limitations imposed by the genre. In most games, it’s pretty easy to retroactively add something to the game world to improve gameplay. At the same time, the added elements do not necessarily have to fit perfectly. The fact is that players are tolerant of even the most logical decisions, if they improve the gameplay momentarily. But with a story-based research game, this number will not work. Faith in the circumstances offered in this genre is a much more fragile substance than in others.”

But developers who create games in this genre have access to such themes and settings, which are usually not accepted to address within the framework of more traditional genres. In Firewatch, this is a Wyoming State Park, where a connection is gradually emerging between Henry’s colleagues and Delilah, who communicate by radio. However

Improve the gameplay after the fact – it will not work

Anderson warns that developers can’t apply ideas that improve gameplay if they don’t fit into the plot.

For example, it was originally planned that Firewatch would be similar to metroidvania. Players had to find equipment that would open access to new areas of the game world or a shortcut to previous locations.

“It turned out that all these seemingly optional things in metroidvany are actually very important for the structure of the game world. All these jetpacks and first aid kits provide new opportunities, which means they open up new areas. From the point of view of game design, it’s not cool when you get a new thing, it opens up new areas, and there’s nothing in these new areas.”

Campo Santo eventually added some additional areas to the game. The player, having reached them, received new dialogues (and a raccoon attack) as a reward. But very few have seen them. And logical inconsistencies prevented the idea from being implemented on a larger scale.

In the case of metroidvania, it doesn’t matter when you find a first aid kit or a jetpack. Both in the first hour of the game and right before the final boss, these items work the same way. But in Firewatch everything is different. If the objects had been found at the beginning, when the two characters were just getting to know each other, the communication between them would have looked absolutely different than if it had happened at the end, when the relationship had developed and certain dramatic events had already occurred.

“We didn’t expect this, it wasn’t easy for us,” says Anderson. – We tried to apply the usual game structure, and suddenly found that due to genre restrictions, everything changes and takes a very strange, almost unrecognizable form.”

Making a player a participant in events is much more difficult than telling him a story that has already happened

Another major difficulty was that the narrative in the game is conducted in the present tense. Anderson noted that in most story-driven research games, the player tries to piece together a picture of what has already happened. And finds out the reasons for what happened. In Firewatch, developers tell a story in which events unfold directly around the player. In other media, this is very common, but there were few examples in games.

The game shows the events of one day after another. At certain points, the narrative is interrupted and jumps forward in time. The main difficulty, according to Anderson, was that the game was interrupted at interesting moments. This technique not only helped to make the narrative more cinematic, but also made it more difficult for players to explore areas of the game beyond the main plot. In addition, the developers had to figure out what to do if the player gets to some areas of the game world before it is provided by the plot.

“We didn’t want to do an adventure game like the old LucasArts games, where the player has to click on something not the way the character starts: “I don’t want to do that yet,” Anderson shares.

However, he admits that there was a moment when the team did not come up with anything suitable.

“We’re not as lucky as the game Gone Home, where the action takes place inside the house and the characters have all sorts of notebooks, personal belongings and all that.” As a result, the team had to figure out what to do with the player when he is where he shouldn’t be, or doing something he shouldn’t be doing. For example, there is a moment when the player gets a new walkie-talkie out of the box. The game brings the player to it, but does not force him to pick up an item. If the player missed this part for some reason, then he will have to go through “a good piece of the game” in a different way, with different dialogues and plot.

“We had to do this because we just couldn’t think of a more elegant way to get the player to get a new walkie-talkie without completely destroying the fourth wall,” says Anderson.

In the wild forests of Wyoming, you can’t really walk around. So the team, in order not to destroy faith in the proposed circumstances, had to use typewritten documents and other evidence of human presence very carefully. The information was mostly relayed by Delilah. But she couldn’t play the role of an omniscient oracle and constantly supply the player with all the necessary knowledge. And since the story unfolded in the present tense, there were events that the player saw and that Delilah could not know about unless he told her about them.

The team found some ways to get around this limitation. Sometimes Henry just talked to himself. And towards the end of the game, when it was necessary to bring the player to a certain location without Delilah’s intervention, the developers gave the character a fictional technological gadget that pointed the way.

Even developers could not rely on the usual techniques such as radars and reminders. If important information came across in the dialogue, it was necessary for the player to clearly grasp it from the first time. It was impossible for him to start a new conversation before he received this information. And the characters could not repeat twice, because it would destroy the immersion in the story.

“If you are going to make a story-based research game, then think carefully about how exactly you are going to give the player information. And what are the potential weaknesses of this method. Because this genre, unlike others, leaves very little room for maneuver,” said Anderson.

The player must see the reaction of the world to his decisions

Another key point in the development of such a game is that the plot is the only thing the player is focused on. As a result, it is difficult to achieve the feeling that the game world reacts to your actions.

When the Campo Santo team created a dialog system for Firewatch, it was inspired by the game Left 4 Dead. When certain events occur, the game remembers them and in the future the choice in the dialogues is based on what happened before. So if the player utters some remark or does something, then the characters can mention it later. The more specific the situation is described in the dialogue, the more lively and authentic the game world seems.

But it is not necessary to refer to any events in the dialog.

Talking, Henry and Delilah come up with a playful name for one of the hills. The player can choose which one. And then, when he looks at the map, he sees that Henry signed this hill with the chosen name.

A playtest of a game in which voice acting plays one of the main roles is a difficult task

The creator of Firewatch saved the last and perhaps the most frightening lesson for last.

“It’s very, very, very hard to play a story-driven research game like ours,” Anderson shared.

If something didn’t work, if the playtester didn’t like a certain scene or a certain moment, it was very difficult to understand why this was happening. Is it because there is no voice recording of the dialogue yet, and the text version does not convey the whole atmosphere? Has the playtester forgotten or not understood what happened before? Or is it a technical problem, and part of the dialog is simply not visible?

Anderson admits that he still hasn’t figured out how best to solve this problem. The Campo Santo team had to bring a significant part of the project to an almost ready state just to understand if they were moving in the right direction.

As a result, the game was made more or less consistently. Before starting an in-game day, it was brought to a certain level of quality. And only after that they moved on to the next day.

Because of this, the development progressed slowly. But as a result, the team came up with a lot of important moments for the game. When developers saw that playtesters were doing something unplanned, it inspired them to new ideas. So there were, for example, rude comments from Henry or Delilah on some unexpected actions and hints about what will happen in the future. And if the first was very easy to do, since the voices of all the actors were recorded in home studios, and they could be added to the game at any time with minimal effort, then it was not so easy to enter hints, since future parts had not yet been assembled.

It was also very difficult to test the final stages of the game, given that the testers were forced to go through the previous parts at the same time. Either people would have to be forced to go through a playtest marathon, despite the fact that feedback about the previous parts of the game was not really needed, or call back the previously involved playtesters and hope that they remember the whole plot that they saw before.

“Looking back and summing up, I understand that there were a lot of difficult and difficult things,” Anderson said. – To be honest, even more than we expected when we started developing Firewatch. But despite the work done, I am sure that in creating games of this genre, a huge layer of design information still remains unexplored.”

Translated by Irina Smirnova

Source: Games Industry