Dying Light 2, and the importance of music design
Olivier Deriviere is an award winning composer and music designer, most famous for his work on Remember Me, Dying Light 2 and A Plague Tale: Requiem. He has worked with world class musicians and ensembles such as the London Contemporary Orchestra, and is passionate about interactive music and its integration into games. He talked to me about the importance of good music design over the content of music itself, the need for game studios to recognise the power of adaptive music, and why there has been a lack of innovation since the days of midi-based soundtracks.
.
Hi Olivier! I’ve seen people refer to you as a pioneer, an advocate, a campaigner, an innovator, even evangelist when it comes to adaptive music and the way that music is implemented..
Provocateur. You're forgetting provocateur!
Yes! And I've noticed a sense of dissatisfaction and frustration from you about how underappreciated and underutilised interactive music is in games. First of all, can I ask, why is it important to you? What are the benefits of a heavily interactive score? And why is it more interesting to you than linear music?
Okay. What matters to me beyond anything else? I'm a gamer. And as a gamer, what drives us all the time, you always hear, is graphics. Physics, lighting, textures, animation, you name it. All these features which are amazing and are praised and honoured by, number one, the industry, and number two, the players. But what about audio? What about music? You know, people don't really talk about it. When we do, it’s mostly about the renderers, such as Atmos, Sony 3D etc. It's not the content, it's the rendering. Which is interesting in some ways, but it doesn't change the way we're actually using the music- it’s just the way we listen to it. As a gamer, I felt sort of frustrated that MIDI-based music from the past, e.g. NES/SNES, Amiga, Atari, it was very reactive. Then the PlayStation came out, CD came out with that, and we started to have very passive but orchestral music. And everybody was praising this, saying, this is the new era of games, like it somehow felt like an advancement for music because it sounded as it did in the movies.
But I didn't feel it this way. I felt it was a mess. We were missing everything that made videogame scoring different and special. And we had to wait until Halo, the first one on Xbox 2001, to really get back to that idea that music can be more than just, like, a nice background, but actually participates in the gameplay. And since then, we’ve been stuck. Even though we had multiple new ways to use music, because of technology. Once again, many people would tell you about layering, like, “oh, we can layer music”. But that can really hurt musicality for instance, because now everybody was doing everything in one key so you could easily go from one layer to the next. But overall, yeah, the musical experience in games hasn't been as groundbreaking as lighting, AI, all these other things. So that's why I'm interested in this. Because as a gamer, I want fresh experiences and I think audio is a big part of that. So why don't we try to improve this?
I think it’s interesting what you said about everything ending up in the same key just so it could easily transition. So I know one of the pieces of music that people talk about a lot in games is in Red Dead Redemption, there's a part where you cross over into Mexico and people often say how emotional it was, how much it affected them. And I think one reason it stood out so much is because that's one of the few pieces of music in that game that isn't in A minor. And I think everybody has been sitting playing this game for dozens of hours, listening to music in the same key, and then suddenly they break free from it, just as the character breaks free from that territory and that situation.
Well, I could not agree more with you about that. The normalisation of the experience for 20 hours or so, to get broken by this. You're normalised and therefore you don't pay attention anymore. And then something completely out of the box comes in, which is, as you mentioned, music that is in a different key, but also, it's a song. And as a song it's a thousand times more appealing than anything else because there is a vocal, there's somebody singing. So the impact of that is very telling about how you use music. Because this is music design. Maybe it was an accident for them or maybe it was perfectly thought about, but the thing is, it worked. And the thing is, I always use this example. So I'm happy you're talking about this! Like, as you said, everybody speaks about that moment, which is crazy. Yeah, how funny.
Crossing into Mexico in Red Dead Redemption
There's a tweet that you put out a while ago about Dying Light 2. You said, “I dare you to not mute the music before you max out 500 hours”. And that resonated with me because the idea of people muting music in a game is a bit frustrating, I think. You'd never mute a film soundtrack while you're watching a film. But I also understand that if you're playing a game, the huge amount of time that you're playing it for, especially if the soundtrack hasn't been well thought out, it means listening to a small number of tracks again and again. It can be quite annoying. So what measures do you take, or would you take to encourage people to not mute the music, other than the music being fantastic to listen to?
Well, the thing is, games are not movies. You can never compare the two. And of course, games have tried to become movies, and funnily enough, movies have tried to become games. You know what I mean? We need to emancipate from this idea so we can be in our own sort of ground. The thing is, it depends on the style of the game, for instance, Skyrim. Skyrim I couldn't play without the music. It's just impossible. It would kill the experience. For me, it's an amazing score and I love the melodies, I love everything. So, you know, a thousand hours with this music is great for me. Dying Light 2 is an action based game where the music sort of pushes you. But when you're doing stuff, the music grows and grows and grows. And I can definitely understand that at some point you're like, "hey, I know, I got it. It was great. But now I get bored about this idea or it's obnoxious”. I don't like this idea of having this music all the time. This I can understand. Recently, after a survey, we found that people really do still listen to the music after hundreds of hours of playing, which is for me, a surprise. So it depends on the style, it depends on what the music is doing. People ask me, “Aren't you mad, when people mute the music and put their own music in the game?”, I'm like, what game are you talking about? Because if it's A Plague Tale Requiem (the first one), then they're missing out, they're missing the experience. But if it is Dying Light 2, after you're done with the story and you're just messing around with your friends, that's fine. I’d like to play it with some metal music in the background. It'd be fun.
Would you argue that the music in Dying Light 2 benefits you as a player to the point where if you took the music out, you would lose not just the experience, but you would not be as good at the game? Like there's information you're getting from the music that maybe you need?
Yeah, so that's definitely what we wanted to do. So that's the challenge, that’s the goal. Did we achieve this? I guess you’d need to ask people other than me. Some were skeptical at the beginning, even me. I think that we did succeed to the extent that if you stay in the main quest and do some open-world stuff during the main quest and all of that, it's really good. It helps you become part of the story, become part of the world. But as I said, 500 hours.. I don't know, it's a lot. Every activity has its own music. That was something that I debated with myself- should I do different music depending on the region, taking into account cost, time, other things? But then it makes more sense to have the same music because players will feel like home. It's like, oh yeah, I know where I am, what I'm doing. It provides markers for players, so they're not lost.
The parkour music is different. You go to an activity, then it doesn't switch to the music of the activity. But if you start doing the activity and the game knows this, then it's back to the activity music and then back to the quest. I mean, this is crazy what we did. This is like really crazy and people don't really understand; it’s one thing that people don't realise and it’s why I'm making a series of videos to explain exactly what we did. But the goal is not to be like, hey, you see, we've been working our ass off to provide you this. The best reward is to have them, as I said to you, after six months, not turn off the music and say one of the best thing of the game is the music. That's the best that we aim for.
It must be slightly frustrating that so much of what you do has a subconscious effect on people. Like you say, you have to make these videos to show people what's under the hood of the car. But people talk about the music a lot and say it’s is amazing and they love it. Surely the interactive nature of it is part of why they love it, even if they don't know it themselves?
You're pushing a good button here because I think that the interactivity of the music, people got it. Like, when you're parkouring, the music, you feel it doing something, but you don't really understand at first. But then you become better and better and then you feel the music is growing. And then it becomes like this huge thing and you're like, "okay, I know”. And if you stop running, the music stops. It's as instant as this. So everybody felt it. But two things. Number one, what you said about the subconscious nature of those things; I must say that this is true for any department. Like music, graphics, any department is doing subconscious stuff like this- all the details in the city, and everything that nobody cares about. You don't stop and look at those things, but if they're not there, it doesn't work. So yes it's very frustrating sometimes for artists or animators or programmers because what they're doing is not in your face, but without this, what is in your face wouldn't work. So that's number one. Number two, if people don't feel the interactivity, this is where we fail. And that's when people could say, why would you bother working so hard if people don't feel it? And that's very key. It's like, I've been very much trying to tell the young generation, you must sort of, like, kill musicality sometimes so people understand what's going on with the music. Because if your transitions are too smooth, or there’s no contrast between the layers, people won't feel it.
It reminds me of this example when I was working on Remember Me?, and every move you were doing was triggering a different music cue and everything, and the creative director put down the controller, looked at me and the audio director and said, “I can't wait for the music to be interactive”. And that's a huge problem. It's like, oh, boy. And I had to sort of reformat the way I've envisioned interactive music in a sense, that it's not about being fluid and sort of seamless, but rather unusual, very obvious feedback for players. Like the invincibility star in Mario, you know, it's not just a little layer of flutes on top. You know what I mean? So that's the real job for people that want to advocate for interactive music, is to make the interactive music matter for the gameplay experience, not to have to ask, “when the player did this, and then the music did that: did you feel it?”
It reminds me of Grant Kirkhope, he said a very similar thing to me, that when he was doing Nintendo games, Nintendo 64 games, these transitions were very obvious, very in-your-face compared to most games today.
Everybody remembers them. And these days, people are obsessed with being very subtle with the transitions, and so people don't tend to process or notice it. And as a musician, you have so many tools now to make these smooth transitions. But it doesn’t have to be like, for instance, when you jump and the music changes when you land and you're exactly on the beat. It can sometimes be better to not be on the beat, you just land and the feeling of that in itself is so enthralling that you just forget that the music didn't match the downbeat or anything like that. You don't care. But I'm not saying that to be seamless is wrong. It's also good to be seamless. As long as the music gives you an impact- that's what matters. It's like, “oh, I feel it totally.”
I think maybe an example that is quite similar to your parkour jumping is in Doom, the new 2016 Doom. When you go for a special kill, and there's a low pass filter on the music, and you get a sting when you do this melee hit, and it's not exactly in time with the music, but it doesn't matter. I mean, it just has that impact, and it feels amazing and gives you power and all that. It's great.
Well, Mick Gordon. Yeah, we know each other a little bit, and yes, I do respect him exactly for this, because I know he's doing this great music design. I know that he knows. And I was so happy that players felt it in 2016. When I see this, I want to celebrate that, because I think this is where us, the composers and the people, we should push and be like, look the players. Look at them. Look at what they’re saying.
I found a scientific paper online recently that someone had written about these impacts on players. They were trying to prove scientifically and categorically that adaptive music has a real emotional effect on the players. And I think it was borne out of this frustration that people aren't seeing it enough, how powerful it can be.
That's amazing, because I've been giving talks at GDCs and over the world, talking just about music design and all these things, and the reception was always very mild or negative, almost because I was a composer, but I was not talking about the music itself. And the thing is, it's very difficult to convince people about those things if they're not part of the project and see that happening. As I said to you, with Dying Light Two, they didn't believe in this. But at the end, they were like, “that's amazing”. Same for any games that have those things. I'm pretty sure that prior to the game, they didn't know, but after it was done, they're like, we cannot go back. And so that's the thing. We need to have more and more people doing that, so it becomes the default. I'm not saying it's bad to not do, let's say, complicated or advanced music design. It's not bad because some games don't need it. And so the thing is, we have celebrated so long the fact that we have great composers who write music for games to an extent that we forgot that it's not the music that matters, it's the game, you know? And even more frustrating now is that most of the composers that get celebrated don't even have a little part in the implementation of the music. Meaning that someone else is making their music work with the game, and the composers get the credit. I don't like this much. When I'm working with any developers, I'm there with them. I'm part of the studio. From the bottom to the top, and I can tell you the love is there, the frustration is there, the difficulties are shared. The brotherhood and sisterhood is there. So the thing is, if we start celebrating all these people as much as the composers, to be like, okay, it's a combination, then maybe the push would be even better for these kinds of things.
Do you think the reason we're in this situation (where we have people writing music and then other people editing and implementing that music), is that because the composers themselves aren't interested in the systems? Or is it because of established studio structures? Or something else?
It could be exactly what you said. I know composers who are more than capable of doing all the music design, but have not been alowed to, the stuido would say “Send us the music, we'll do the music design”. But it can also be the other way around.
The thing is not it's very easy to make games, okay? It's really difficult. So when we're talking about music and all of that, it really shouldn't be an aditional problem. As a developer, there are so many problems that music should not be one of them. If it becomes a problem, then you're in trouble, okay? That's bad. So this sort of safety that we as an industry had for ages starts to fade away. But the tools now are so easy to manipulate and so useful that it doesn't cost much more to do these things. So basically nowadays there is no excuse. You could say there is no excuse. So if we tell developers there is no risk in doing this, and we tell composers that they can do it safely, then composers don’t need to learn everything about it, but at least the basics. And the people inside the studio should help in terms of budget and tech and that kind of stuff.
Have you personally had any situations where you've had ideas or concepts that you haven't been able to implement? Where you’ve been told, that's going to cost too much, or we don't have the tech, or anything like that?
Well, I must tell you this. I've never, through my whole career, met any developers that I went to and asked for features who then said “no, never”. They’ve always say yes. And at the beginning, we didn’t even have those tools that I talked about. So we had to design the tools ourselves. So programmers needed to do it. For PlayStation Two, PlayStation Three, Xbox 360, etc. And it's a good example of two things. Number one, that developers are open. They are open to this as long as you know what you're doing. And that's the second point, is like, how many of us knew what they were doing as a composer, asking for features? Because me, I come from a background of music, computers. I've been coding, I've been very interested into those things. So discussing with programmers and looking at hardware was always to me, a pleasure, an interest. And I can totally understand that some composers, they don't care about this. So the thing is, if you convince the developers that, yes, you're up the task, they will open everything. I mean, once again, for Dying Light Two, that was crazy what we did. The thing is, Techland really opened everything for me in terms of programmers, not only for music, because we went to Abby Road, London Contemporary Orchestra, all this. But on the other side, it was like, maybe five programmers helping with audio, that's crazy. Plus a team. So this is something that I think will be more and more accepted, as the tools develop. And let's say the ease of use gets better and better, hopefully the young generation- I mean, I just hope the young generation will be better at this than the old guard, let's just say.
So what else can we do? How do we convince people to take it more seriously and increase the adoption of these things?
It's almost 20 years that I've been pushing for this. I've tried everything. I've tried going on conferences, talking with composers, talking to developers. It felt impossible to convince them, really, because if you talk to developers, they don't see it because they have so much more than just the music to worry about. They don't care. As I said, they don't want this to be another problem. And the composers, they pretend they do it. So it's very difficult for me to say, how do we advocate? And I found myself to be like, okay, the only way I can advocate for it is to make it myself and to show people. But I’ve never done any hit games, like 90 Plus Metacritic games. I've never been there. And we know what that means in terms of impact on the players, on the industry. People wouldn't know my work- I would ask them, have you played this? They hadn’t played it. They don't know it. So it's very difficult for me to advocate more than that because basically, I've proven that it's working, even on Remember Me, Assassin's Creed- I've proven it on those games, but they were never something people were interested in. I mean, players, yes, don't get me wrong, the players who play them were like, wow. But there were not many of them. But you know, the industry is very weird. For instance, The Witcher Three, which is a huge hit, got completely snubbed by some awards and things like that, it's very weird. I don't understand it myself, but I tend to just not mind anymore and just do my stuff and my little thing on Twitter.
I'm about to give a speech at Gamescom and the title is “Music Won't Change Your game. Music design will.” That's the title. So I'm back to doing these things. So some of us are doing it. When Picasso and all these painters were all together, or the French composers together, and they created this movement, I wish we would do this. I wish a bunch of us would be like, oh, we're part of the music design movement, which is a movement. It doesn't mean that you need to follow it. It's just like, oh, we're just trying something there. Is it worth it? Is it good? We don't care. But we're trying. We're pushing and we'll see.
So you don't want to go as extreme as Dogma, for instance, and lay down the ground rules?
No, it's not about that. No, because I think that everything is valid in a sense that you have many games where the music design is very varied. Like, I care about everybody, what they're doing, and I respect what they're doing. But some games, if the music design would have been different, let's say more integrated within the gameplay, it would have been a different game. A different game.
You mentioned Skyrim earlier, which has a fairly non-interactive score. I've put hundreds of hours into Skyrim, and there have been moments where, for instance, you're running up a mountain and get to the top and see a whole new area stretch out below you, with a dragon in the distance, and at that moment, the music just happens to swell. It's just coincidence, but the impact is huge because, you know, it's not prescribed. You know it's a coincidence. And so it makes it feel even more special somehow. I always thought that was really interesting.
Yeah. So your argument, if it's coincidental, is it good? In a sense that if they had done that with some music design so when you see the dragon, you feel this, well, is it better or is it worse?
Yeah, in this case it was the knowledge that it was a coincidence. It's like when you go for a walk and you get a double rainbow while you’re thinking about someone, or something like that happens. That is just a crazy coincidence. And it feels, I mean, I'm not a religious person, but it feels almost like it's spiritual and significant.
Yeah, I understand. I agree with you, and I think that it makes it better. But who knows? Because that's the thing you're pointing out. If I had to debate and be on the other side, I'd be like, oh, what you're saying is so true, because if we were listening to this French guy over here, everything would have to be calculated and prepared, and so somehow it would be intentional, and you would expect this to happen every time. But here it's random, and so it creates this sort of like, as you said, this spiritual moment, almost, that you know you're not going to get again. So that would be a counter argument. But then I would answer, because now I'm becoming crazy, I'm the French guy again, I'd be like, yeah, but we can randomize. So that's the thing- we need to treat each game with a very special care. And there is no recipe. That's the thing in music. As you said, Skyrim, you and I, it feels like we really like the music. It really made an impact on us. Some people don't like it, some people just don't. And the bottom line is, like, it's always something that is very personal. As composer, you can write “likable” music. You know what I mean? You know the codes, you know how it works and it's easy. But now, is it something that everybody wants to do? No. Gary Shyman, for instance, the way he writes, it's very tortured and everything and it's beautiful at the same time. So it's very personal.
One more thing about Skyrim. For me, the beauty of the music, it’s not just the composition; it’s the arrangement, it's the orchestration and it's the performance of it - I don't know who's responsible for that exactly, for some reason there are no performance credits on the soundtrack, but it's beautifully done. And I wonder, is that something you have to sacrifice when you make interactive scores? Can you retain that musicality? Is it always a trade off?
Skyrim is an exploration, a sort of wandering-type of game. There is not much happening, you know what I mean? So the big swells, etc., it works because the music can develop and have this sort of structure- the theme starts, then there is this sort of B section and then the theme comes back bigger with the choir and then very nuanced dynamics. It's amazing. I mean, this is one of the rare scores that I really appreciate. Like, wow. I wouldn't say it's like the best orchestration production at all, but I don't care. I like the music, which is like, what matters to me. And then the thing is, you cannot do this on a game like, I don't know, Mario. You can't. Because the type of game is not the same. So going back to Plague Tale 2, now we have this sort of like stealth fight system. And the way I did it, the system is always the same, but the way I dress it is always different. And when I say dress it, it's not like it's a different music using the same sort of structure. The structure will be used differently depending on the situation. So I'm breaking my own system, but for good reasons. And the thing is, as I said to you, you don't feel it and it's very musical. We were with the other directors recently because we went through the whole game and it felt like the music is like perfectly, let's say, musical. So it's not the same as Skyrim with the themes because the gameplay is not the same. But now you need to look at the meta structure of Plague Tale compared to the metastructure of Skyrim. Plague Tail is a narrative driven game. So basically, what Jeremy Soule was doing on one track, I'm doing it on one level. It's sort of like, I expose the theme and you have this sort of progression.
Gary one time said this to me; sometimes you need to let music express itself. And I said to him, you're right, but you're saying this, not knowing what we can do with interactive music. And interactive music doesn't kill the music. It's a form. Just as a concerto, fugue, symphony has form. You make music within these rules. So it's musical, you know what I mean? That’s the way to think about it.
So one final question- how do you personally go about taking one of your interactive scores and making it into a linear OST? How do you resist releasing 20 hours of music when you could quite easily create any permutation you wanted? Is that a challenge for you? Is that something you enjoy or don't enjoy?
Well, let's put it this way. I'm not writing my music to be on a CD or on a soundtrack or for a concert. I'm doing it for the game. Once it's good for the game, then I think about a soundtrack. I'm trying to focus on content that has a discourse; something that when you're listening to it, you cannot do anything but listen. You know what I mean? Not like an ambient track. I mean, sometimes I include some of the ambient tracks because it's good to rest a little bit if you're listening to the whole album. But most of the time the soundtracks I put up are very intense, like one emotion after another. So I just get the substance of the work and compress it into something that is coherent in itself for the listeners. But I don't want the listeners to put this into the background of, I don't know, doing a workout or drawing or browsing. I want the music to take their full attention.
Dying Light 2 is out on Xbox, Playstation and PC. Plague Tale: A Requiem releases on all platforms on 2nd November 2022. The soundtracks for both are available now