Picture by gratisography
In regards to this month about “failure”, I wanted to bend the subject a bit by talking not about human failure but handling failure as part of an audio designer job for a video game. For this, I decided to ask people who worked on games where character failure is prominent or even a central point. With no further addition before delving into the actual article, I would like express massive thanks to Emeric Thoa, CEO of The Game Bakers and Creative Director of Furi, Gordon McGladdery, Founder of A Shell in the Pit and Sound Designer of Rogue Legacy, Kevin Regamey, Creative Director of Power Up Audio and Sound Designer of Darkest Dungeon and Martin Stig Andersen, Sound Designer and Composer of Limbo and Inside, for their cooperation and the time spent to answer my questions.
While time and resources are sometimes very very (very) limited, one might need to refrain themselves on achieving their perfect vision for handling a failure. On the other hand, a creative mind can often be able to overcome this and offer a clever solution to what appeared to be a complex situation, such as handling a character death in a way relevant to the game’s mood. Or, as Gordon put, “I don’t think one needs a SpaceX budget to create a good fail-state! It really depends on the game. I was very happy with how the fail state in Fossil Echo worked out (more John Robert Matz‘s doing than mine). It’s a game where you die a LOT. If you die for the first time in a while you get a big musical sting, but the following stings are less dramatic if you die consecutively in a short amount of time”. Other examples, quoted by Emeric, are Prince of Persia: The Sands of Time and the Uncharted series, where the characters themselves are giving the requested sonic information to the player, either after dying (with the Prince breaking the 4th wall and stepping in as a narrator mentioning to the player “That’s not how it happened”) or just before (with Nathan Drake and his signature “Oh no no no no no!” in case of a bad situation).
Communication with the rest of the team is also the key of handling failure (or any situation, really!), as Kevin stated, in order to be sure to “have access to all the parameters the game is using – health, what type of enemy killed you, the environment you’re in, etc… It depends hugely upon the gameplay, but colouring the fail state with the situation and environment can lead to wonderful results.”
Taking a bit of a meta-approach on the subject, Martin noted that “it’s hard to talk about failure in-game because it’s not really a failure, it’s part of the game but I think a lot of games don’t really cope with that. For instance, when you die in a game, the game might stop and reloads; some games even take you out to a menu to load the game again. But of course it also depends on the kind of games you’re making. If you die maybe once in an hour, it’s not really a problem. But for games where you die a lot, you want the player to stay in the experience, you don’t want to take them out.” And to continue on his experience on Inside, “one of the things we did in Inside was to let the sound continue across death-respawn, along with having some kind of music synchronized to the game going on. Then, during a death-respawn scenario, we would let the music play so that, when you spawn (in an arbitrary point in time) the game would look at the music and sync to its current point (…) But if you work with dialog, there will always be the problem of respawning and hearing the same joke playing every time (…) As a player, you want to be in the game so if you are suddenly taken out and everything reloads, it’s like if the medium would reveal itself and you would became aware that you’re playing a game. So if you can handle that properly, it becomes more magical in a way. That’s what I liked in Inside, the ability to play the musical parts which continue even if you die. It’s almost like a musical video playing no matter what you do.”
Of course, failure also has an influence on a player’s state of mind and willingness to keep playing and it’s a designer’s (sound-wise or not) responsibility to prevent this state to become too frustrating, otherwise the players might end up rage-quitting a game and never touch it again. One of the key sonic points to keep in mind according to Gordon, is variation: “Repetition can be inherently annoying, and getting the same fail state over and over again can have a “Chinese-water-torture” effect. In Rogue Legacy the death sting itself was always the same (while frequent, deaths in that title aren’t as common as in Fossil Echo but we did have variation on the death vocalizations. Fail states not being too frustrating also relies heavily on overall game design. A really good loop without any awkward pauses, which Rogue Legacy has, is vital. Super Hexagon also has a really good, non-punishing loop.”
Those awkward pauses mentioned by Gordon are also something Martin noted, replying that “death can be quite rough but it has to be fast. I don’t like games where you cut out to some cutscenes when you die. Death can be really rough and scary but the time before you’re back in the game has to be quick. You should just have time to realize you’re dead before starting again. Even if you just play a small video: as soon as you die you just want to play again, so that video would just make you feel more detached from the game. Both in Inside & Limbo, as soon as you die, you have the choice just to push a button and you’ll reload immediately.”
While having a transition as seamless as possible seems like a shared opinion amongst everyone, Kevin also mentioned that the way of handling failure in a game really depends of the game itself: “Is it Dark Souls or Candy Crush? If it’s a hardcore game, defeat can be a little more punishing because the players are differently equipped to deal with that situation psychologically. They’re okay with challenge, and trying again is a reasonable hurdle. For a casual game on mobile or similar, the defeat needs to be “aww too bad!” but with a positive twist, to encourage players to try again. One thing we do regularly is create a win and lose stinger that both seamlessly transition in to a neutral-positive interstitial loop – that way, whether they win or lose, they end up on this positive note and don’t feel entirely soul destroyed.”
Another role of a sonic signature in terms of failing is the ability to give players a feedback about their action, as noted by the Prince of Persia and Uncharted examples above quoted by Emeric, who also noted that “[The sound feedback] should almost be like a reward sound: a strong feedback, giving value to the moment and information about the reason of the failure. Sound-wise, it should say ‘The challenge was tough, you failed, here is why. Now stand up and try again.’ (…) In Furi, we designed what I called the ‘Tension mode’. It’s a sound mode with filters, that only triggers when 1) the player has little life left 2) the enemy has little life left. If the player has a lot of life and the enemy is close to dying, the player is easily winning. If it’s the opposite, then the player is getting crushed. There is little tension in both cases. But when both characters in a duel are hurt, close to dying, but also one hit away from winning, the stakes are very high. That’s when we trigger this ‘tension mode’ that changes all the sounds and tones down the music. I like this refinement in modern games. It’s adapting the direction to what is actually happening and it’s where interactivity is rich, compared to more traditional entertainment.”
And while failing is obviously bad news for a player, Kevin also mentioned that being really close to failure without actually failing is where a lot of fun can be found: “One game that excelled at this was a student project out of DigiPen – a survival racing game called ‘Nitronic Rush‘. Your car can overheat and explode, and as the heat meter nears the fail mark, an alarm fades in. This alarm is actually an additional layer in the music, so while a typical alarm might be annoying, this alarm just gets the player more amped up.”
“The tricky part is not to add too much to the game.”, added Martin. “As soon as a sound is not useful to the player, they will start to filter it out. If there are a lot of sounds with no other meaning than making the soundscape beautiful, the player will automatically start to filter those out and think they are not important. My way of doing it is by taking more minimalist sounds and work with sounds that are related to the gameplay. So even if I end up with a lot of sounds, they’re all somehow related to the gameplay, telling you something on how to survive. The tricky part again is, in order to make the player listen, you have to play sounds that are valuable, that have valuable information for surviving. There’s also specific ways of dealing with this, of course, but it often becomes a bit cliché, like using a heartbeat or something else that sole purpose would just be to warn you, which is a bit ‘undynamic’. It’s nicer to have information that will help you survive rather than just being told that things are going bad. That’s much more interesting that just starting to play heartbeats or some muffled sounds.”
Immersion is really the key and it’s something that is currently discussed in various new ways because of the emergence of technologies such as AR or VR. Taking that into consideration, Gordon states that “In a lot of games, the character VO is a great dynamic way of ramping into a fail state, but that will no longer be very useful in VR since putting a voice in someone’s head is weird. It’ll be cool to see how processing binaural sources to make the user question their perception of reality comes into play.”
VR is not yet accepted by everyone as “the next big thing” and while there is already a lot of work on it, some people would rather adopt a “wait & see” attitude, fearing an innovation that might just fell flat or not deliver as much as dreamt. And while Kevin don’t think VR will change a lot of things in terms of designing a fail state, Emeric and Martin admitted being part of this “wait & see” crew.
“I think the closer you try to approach reality, the more fragile it becomes.”, stated Martin. “If you want to suggest something being real, the illusion will fall apart as soon as the player will encounter even the tiniest mistake. (…) If you’re sitting on your couch and playing a game on the TV, you don’t see yourself as being there, there is some kind of abstraction layer between you and the experience, some kind of ‘contract’ that the world presented to you is on a screen. One thing discussed a lot is the question of how to use music in VR because as a player, you would ask yourself “Where is it coming from?”, which is a legitimate question. And I think that’s really interesting because I rarely work with a purely abstract music, I like to use elements that are represented in the world. If you take the soundscape for Limbo, for instance, I think it could actually exists in such an environment because the soundscape builds up from it. It’s related to the environment and to the game mechanics. If I have to say something positive about VR is that it’s something that will make sound and music more interesting, in my opinion. Because I think that, intuitively, people will start to look for elements in the game and the environment and use them musically to create moods and atmospheres, almost like an electroacoustic approach. You will be tempted to work with the musical values that are in the sounds in the environment. I think there’s a good chance that’s going to happen because it’s going to be difficult to play a symphonic orchestra in something that you want to present as reality. That said, in terms of VR, I’m more interested in purely abstract things. As a medium, I think that would be the most obvious way to go, creating abstract worlds, visually & sonically; it could be really beautiful. But it seems like every time we have a new media, we just try to take what we have from the previous one and put it into that, which seems to be what’s happening now. The closer you want to get to total reality, the further you get away from it.”
Despite new technologies arising, some design tropes travelled through the years and will probably still be there as years go by. Game over jingles (“iconic, fun [and] meaningful”, according to Emeric), low pass filters for Gordon, dissonant sounds for Kevin or a certain sense of roughness for Martin (because of “hard cuts in the sound and picture”) are some of the ideas that arose during the creation of this article.
Obviously, all of these things can’t go with every type of games and when asked about their design process during the creation of, respectively, Rogue Legacy, Darkest Dungeon, Furi or Inside, they were kind enough to give an overview of their thoughts and goals about the player failures. For instance, the main preoccupation of Gordon, on Rogue Legacy, was to “make sure everything was covered and flowed properly and there weren’t any gaps to complement what was a brilliant gameplay loop”. As for Kevin, he shared his design tip for Darkest Dungeon with a two words concept: “Blunt brutality. The death blow is not celebrated much visually, nor is it celebrated much sonically. The game itself doesn’t actually have a fail state – you just churn through characters until you win. These heroes are not your friends – they are your tools.”
As for Furi, which Emeric directed, it’s a game where “you never die (…) You are either knocked down, and you stand up, or you are taken back to the jail you start in (the boss takes you and brings you back, and the game over screen is the character back in the jail). But failure is a big part of the game concept for sure. The goal with Furi was to make a game where 1) you always knew why you failed 2) you always understood it was your fault 3) you alway thought you would do better if you tried once more. In addition to the slow-mo on the action & sound when you fail, on the game over screen, The Voice, the narrator character, coaches you and encourages you to try once more, with a sort of pep-talk. It’s not a game that tries to punish the player, it’s a game that rewards perseverance.”
As for Gordon’s studio A Shell in the Pit on Rogue Legacy, Martin worked both as a sound designer and a composer on Inside. A double role that allowed him to blur the lines between both arts, thanks to his background in electroacoustic music and musique concrète: “While creating the sound design for Inside, I didn’t think in terms of music and sound design as two separate elements. For me, even the footsteps, how I create and mix them, are musically, in a sound design way. I consider the whole mix to be the real music. It’s just a personal preoccupation I have, the way the sound can be blended with the image. Also, I like how you can transform something very concrete into abstraction and not just swapping between the two states of having music or no music. It’s interesting to subtly transform from one to the other. Without noticing, suddenly you have a musical soundscape but you didn’t really realize how that came into play. In Limbo there is one place where I used a humming because you are into the city and I thought of air conditioning and stuff like that. And this humming gave me a fundamental pitch, which I was then able to use as a base to build my soundscape on.
Failure can be nice in the case of live performance. In a theater play, you have real actors there that can respond to failure somehow and interesting things came come from that. You often see that. That’s obviously harder in a game because the game itself is quite dumb, I don’t know if you can teach it to make a clever response to failure. Because the fragile nature of arts is nice in a way, it could be interesting to see, to have this kind of sense of being fragile in a game. But I can’t quite see how to accomplish that. That would be interesting.”
This fascinating approach made possible by creating a bridge between art forms was also for me the opportunity to wonder if other representations of failure in other arts can somehow inspire the development of design techniques in video games. Emeric, for instance, quoted one of his musician friends who told him that “sometimes, during concerts, one player would somehow screw up a little, fail a solo or something, and the other musicians would notice it and try to cover for him, give him time to catch up and get back on track. I think it’s a cool game mechanic when in a game, players who have not failed can help a failing player to come back into the game. It creates a nice interaction and cooperation feeling.”
Not mentioning a specific art form, Gordon mentioned that “it would be really cool to have a fail state that really drove home some kind of hubris. Most character failure in other narrative media is hubris-related, yet in games it’s almost always just the player making a mistake.”
Asked about how they overcame and learn from a failure in their careers, here are the answers I got from Emeric, Gordon, Kevin and Martin:
Emeric: Once, I was really pissed by a colleague who had used all the patience I had, and I punched the wall and broke it (unfortunately, it was in my appartement). I still think I was right to be pissed off, but punching a wall is a serious fail in terms of anger management and team management as well. I should have handled the situation better. After this incident, I wrote a post-it “Describe – Explain – Chill” and stuck it in front of my desk. It meant that I should describe more how I feel and why I feel this way, so that people can understand better when they are going too far. And also that it’s just stupid to get this angry when we are making a video game. I didn’t punch anymore walls since then.
Gordon: I did a pretty poor job managing audio on a game the month prior to my wedding. It was simply too much stress and as a result I gave poor audio direction and feedback to my outsourcer, and didn’t communicate nearly as well as I like to with my client. I wasn’t happy with how it wound up. I’ve been trying hard to improve communication, timeliness and simply not taking on more work than our team can handle.
Martin: Before, when I was a composer of instrumental music, my process was quite different. You know, you write a score, then you hear it and only then you can judge it, establish if it’s a failure or not and learning from that – and of course I learnt a lot from that -. Then later on I moved to more electronic compositions and what I liked there is the direct response to what I create. I play a sound and directly hear it, which means I can play with it, transform it and hear it again, and take my decisions based on that sound. That way, the environment I’m working in is more controlled. So everyday I make a lot of failures but – if it’s a good day – there’s also something good coming out of them and then I can collect that and get rid of the rest.
Kevin: I worked at another studio before co-founding Power Up Audio, and had absolutely no involvement with the industry at large. The studio itself didn’t do anything in terms of game jams, dev meetups, school outreach, etc. – but I also failed to involve myself personally for a very long time. I was burning out and losing interest in what I’d thought was my passion. Eventually I began attending the Full Indie meetup here in Vancouver. I attended game jams. I released my game Phonopath online. I went to GDC. All these things catalyzed my departure from that studio in very profound way. Now that we’re out on our own, our whole studio makes a continuous conscious effort to find new ways to involve ourselves in the industry: attending conventions, helping at charity events like Summer Games Done Quick, sponsoring game jams, and even supporting up-and-comers in-game audio with our weekly demo reel review segment, “ReelTalk”.
And go figure, we love our jobs again.