Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
October 24, 2014
arrowPress Releases
October 24, 2014
PR Newswire
View All





If you enjoy reading this site, you might also want to check out these UBM Tech sites:


Four ways to design for horror, from Amnesia dev Frictional Games Exclusive
May 8, 2014 | By Kris Graft




For a man who conceives of video games that scare the living daylights out of me, Thomas Grip seems like a pretty affable guy.

Grip is creative director at Frictional Games, the Swedish independent game studio that developed the horror adventure games in the Penumbra series, as well as its breakout title, Amnesia: The Dark Descent.

The studio is currently at work on Soma, which, if the trailers are any indication, will crank up the intensity and fear found in the studio's previous games. With such successful designs from the Frictional team, we recently chatted to Grip about a few aspects of horror game design. This isn't a definitive best practices list, but these are just a few things that have worked for Grip and Frictional.

Frictional acknowledges that horror isn't about "finding the fun" first

Designing a game for horror is unique to designing games in other genres, says Grip. "The biggest thing is that when you're doing a horror game, you're going for a special emotional punch that you're not doing in other games in the same way," he says.

A shooter or a platformer, for example, have a basic, core gameplay loop - there is a focus on "fun" which relies on the effectiveness and subjective quality of that loop.

"But a horror game, which is very different, I think, from all other genres, is about 'Is this game scary?' You're not as much looking into figuring out how to make your gameplay mechanic better or more fun, but rather 'How can I make the game scarier?' The gameplay mechanic is almost sort of a side thing."

"There's a basic design tip - that you don't start with the story, you start with experimentations, prototypes, then you make that into a game," Grip says. "While horror games can come from that direction, that's not really, I think, how you want to do it. You want to start with how you want the game to be, the type of horror you want to do."


"You want to start with how you want the game to be, the type of horror you want to do."
With Amnesia, the team at Frictional did decide on the most basic mechanics, such as the first-person perspective and physics-based interactions. The developers and designers built their horror game from there. From that point, additional mechanics were added to serve the ultimate emotional goal.

Grip acknowledges that horror isn't the only genre in video games in which designers begin with a high-level concept, then create mechanics with the purpose of conveying that concept. Developers on games like thatgamecompany's Journey and Minority Media's Papo & Yo also start with an emotional goal first.

"It's this high concept that is dictating what you should do in the game, and what you shouldn't do [from a mechanics standpoint]," says Grip. For example, weapon upgrade mechanics might seem like a good idea in a Frictional horror game. But even though that might make the game more "fun," says Grip, "it will destroy the horror element."


Grip working at home, creating horrifying video games

Frictional iterates and playtests for maximum horror

The way Frictional designs for horror is very much based on the gut instinct and experience of the game's developers. You won't find any Excel action matrices or biometric scanners here. Or even very many external playtesters.

"For Soma, we have a sheet that just details the very basics of every level, stuff that's going to happen there," says Grip. The designers plan out levels' use of light, the kind of moods that level is meant to convey, enemy density. At this stage, levels are being put in, or thrown out.

After developing playable slices of the game according to those plans, the designers basically try to see if they can be immersed and frightened by the same scares that they work on as their day jobs.


"You have to play it yourself, and put yourself in the position of a new player, and just feel it."
"You sit down, late at night, and play a big chunk of it in one go," Grip says. "Then you get a feel for it." At this point, the dev team becomes horror editors -- snipping, adding, moving around game elements and scenarios that don't click during their play-throughs.

Frictional makes sure to get to the point of iterating for maximum horror early on in the game's development. "I feel it's sort of unnecessary to do anything too detailed at an early point, because by the time you're finished, everything's going to change," he laughs.

For Grip, there is good reason to have developers playtest the game, instead of outside playtesters. For one, games like Soma in particular are slower-paced and not focused on a mechanic such as shooting, which players can more easily give concrete feedback about. Playtesting for horror relies a lot on what's happening in player's brains on a subconscious level. Drawing out those emotions and making players describe them so developers can improve the game is difficult, if not impossible.

"It's very hard to judge how successful you are, based on the feedback you get back from people," Grip explains. "When it comes to how you need to tweak [certain aspects of the game], you have to play it yourself, and put yourself in the position of a new player, and just feel it. It's super hard, you're constantly making mistakes, but it's sort of the only way to really get inside their head. It's hard to ask people to explain subconscious stuff that's happening."

Frictional lets the players scare themselves

Like a good horror movie director, Grip relies on players' mental modeling of situations and environments - basically creating an unsettling space in which players' imaginations can run rampant, creating their own objects of fear. This Lovecraftian approach to game design tends to be highly effective, even on Grip.

He talks about a recent example in Slender: The Arrival, where his mental modeling for fear was firing on all cylinders. "You're approaching this house, it's getting darker, you have these leaves rustling, capturing the mood perfectly," he says. "I knew this was a Slenderman game, but I didn't know it'd start Gone Home-style, with nothing happening. ... I know the whole Slenderman meme or myth, and I came into that game just filled with what I know about Slenderman. I'm using that to project on everything - like the rattles in the trees - 'Shit is he in the trees?!' The possibilities are limitless."

"It got really creepy, and I felt this cold chill down my spine, which has never happened before in a game. And I wasn't sure if I'd be able to continue playing this game," he laughs. And it was all just me scaring myself - no mechanics!"

This reliance on players' imagination - their mental modeling - does introduce a game design problem, however: How do you keep that modeling going throughout the course of the game? "It just breaks down very, very fast, and it's hard to see how to mold that, other than just having empty environments where nothing happens. That just gets old after a while." Grip says he's still trying to figure it out.

Frictional adheres to horror in all aspects of the game (and that's really hard)

While Grip chooses to start from a high-level concept at the beginning of development, as opposed to prototyping "fun" mechanics, he concedes that the high-concept approach can be incredibly taxing.

With a mechanics-based game, a developer, in theory, can "find the fun" in a few days -- perhaps even in hours. But "finding the fear" takes a lot longer, and a lot more time. For Amnesia, Frictional had to throw away large chunks of the game, very late in the game's development, which was demoralizing (though definitely not unheard of in regards to story-centric game development).

"It'd be cool to get away from that and work on something that's fun to play in the first week," admits Grip, when asked if Frictional would explore genres outside of horror. When you need to serve a high concept, everything about the game is constrained to that concept.


"We have our own niche, and I feel that it's sort of stupid to go out of it too much."
"You have to [adhere to] the horror theme, and the adventure theme, and the puzzles -- oh, it's so boring to design that!" he laughs. "The story has demands, the puzzle has demands. You have to figure everything out, how everything fits together."

Grip says Soma wasn't even going to be a horror game at first. Originally, it was meant to be a game that explored the idea of consciousness. But then horror just happened to be the best vessel to express that idea. Over the course of two years, Soma was increasingly identifiable as a horror game. It's as if Frictional couldn't escape from it.

Even though Grip has this fleeting fantasy of doing something that's "fun in the first week," one can tell that he's still excited about the potential of the horror genre.

"We have our own niche, and I feel that it's sort of stupid to go out of it too much," he says. "I feel there's tons more to explore there. ... And if Soma goes well, I know exactly what I want to do with the next game. So there would be at least another horror game after that."

"Then after that, we'll just make a top-down shooter or something," he laughs.

Soma is slated for a 2015 release on PC and PlayStation 4.


Related Jobs

University of Central Florida, School of Visual Arts and Design
University of Central Florida, School of Visual Arts and Design — Orlando, Florida, United States
[10.24.14]

Assistant Professor in Digital Media (Game Design)
The College of New Jersey
The College of New Jersey — Ewing, New Jersey, United States
[10.24.14]

Assistant Professor - Interactive Multi Media - Tenure Track
Bohemia Interactive Simulations
Bohemia Interactive Simulations — Prague, Czech Republic
[10.24.14]

Game Designer
Next Games
Next Games — Helsinki, Finland
[10.24.14]

Senior Level Designer










Comments


Alex Belzer
profile image
Though--didn't Amnesia have a core gameplay loop? The constant cycle between expending resources to create light to restore sanity (but where the player is able to be spotted by a monster), to hiding in the shadows where sanity depletes (but is safer from being spotted) is kind of the very definition of a loop. Though one that creates tension instead of fun.

As Grip pointed out, the audio-visual elements in a horror game go a long way towards playing with player's minds, and was expertly done in Amnesia-- but without this clever system I don't know if it would have been as special as it was.

Would love to hear from Grip about these kinds of systems. I'm sure he'd have something fascinating to add.

Dan Felder
profile image
The difference is that the mechanics are driving the experience of tension, paranoia and powerlessness - the same way that the audio-visual elements. They're contributing to the overall emotional experience, not creating a fun loop for the sake of making a fun loop and building a game around that loop.

Greg Scheel
profile image
Dang, you guys sure start employee training early...

:P

Theresa Catalano
profile image
When you talk about "adhering to horror in all aspects of the game," unfortunately that's not really the case with Amnesia. Here's some things that Amnesia could do better:

1. One aspect of the game mechanics that is vitally important to creating tension is the idea of consequences for the player's actions. The player has to fear death for a real reason, not just because the screen gets bloody and says "you died." So, you need some sort of penalty for death. Unfortunately, Amnesia does the opposite of this... it actually rewards the player for death, because not only do you get to keep all of the progress you made but often the monster that killed you will despawn. If you want to adhere to horror in all aspects of the game, then this is a vital one you can't ignore.

2. A very important thing that all horror games must do is strive to be constantly unpredictable. Unfortunately, Amnesia doesn't seem to try very hard. There are only two enemy types in the whole game, and they all pretty much act the same, simply charging straight for the player. Next time, think about having more types of threats, not only in appearance but also in how they react to the player. Keep the player guessing.

3. Also along the lines of predictability... think about varying the setting more. You want to be constantly surprising the player. A dark castle is a good setting for horror, but there's ways to mix it up more. Try not to have all the areas that the player explores feel the same.

Also, though this is personal preference, consider doing the next game in third person. I find that third person lends itself better to horror because you're better able to connect with your avatar if you can see them. You also might want to think about using surrealism more. And with regards to the soundtrack, while you want it to be atmospheric, it would be even better if it were both atmopsheric and memorable.


none
 
Comment: