I was recently interviewed by Chris Mackin from Bristol’s dBs institute as part of an article about specialising in sound design.
You can read the piece here, where I talk briefly about discovering that sound design was actually a thing, and how finding out that the sound on TV programmes like Blue Planet would be added later in post (and not recorded live at the scene) was a big and inspiring revelation!
Monstrum - the very cool procedural horror game made by Team Junkfish - released on home consoles today. Congrats Junkfish!
To mark the occasion, I thought now would be a good time to re-post an interview I did a few years ago with Jaime Cross (@speedyjx). Jaime is the Audio Lead at Team Junkfish (@TeamJunkfish), and back in 2015 he very kindly took the time out of his day to chat with me about Monstrum shortly after the game originally launched on Steam.
As you can see from the length of this post, we talked about quite a bit of stuff! Topics we covered included Monstrum itself, the formation of Junkfish as a studio, Jamie’s audio influences, and, of course, his awesome sound design and music compositions.
Obviously, a lot can happen in 5-years’ time (cough, cough, Monstrum 2, cough) and a lot of things have changed between then and now (remember Steam Greenlight?). Still, I thought it would be a neat idea to put the interview back up online on the lustrum of the game’s original launch so that more folks can check it out.
Read on…if you dare!
Tom: How did Team Junkfish originally form as a studio, and what’s the story behind the name?
Jaime: Team Junkfish came together in 2011 at Abertay University. In the third year we had to do a big group project - Abertay have a lot of big briefs from Microsoft, Sony, Disney etc., but students also have the opportunity to form their own team and present their own ideas. Students can pitch their ideas to the lecturers and if it’s good they’re given the go-ahead. Back then, there were currently nine members of what would become Team Junkfish, and they wanted to do this big idea but they were told they were probably not going to be able to do it unless they got a sound guy. So they pitched for a sound person in front of the entire year, and I just said “I’ll do it, aye!” and that was kind of it.
There’s been ten of us ever since, and it’s coming up on four years now in total that we’ve been working together. It’s scary when you think about games companies and the way they can come and go, but yeah we’ve been doing this for four years and we officially became a company two years ago now. The Team Junkfish name came from when we did our first big prototype, and it was basically a giant flying mechanical whale, and we just went “Yeah, let’s call it Junkfish!”
The team’s first projects FooFoo and DinerSaur were aimed at young audiences. What prompted the shift to horror games, or was it just a natural evolution of the team’s creative urges?
Originally the FooFoo game was part of the Samsung Student Developer Challenge, so we put it together mainly for that competition. We did quite well – we got a lot of press out of it, which was very cool. DinerSaur was a cool augmented reality arcade shooter made for the Dare to Be Digital competition a couple of years ago. The game itself was technically made just by six members of the team (then known as Prehistoric Spatula) over a period of nine weeks. That included three programmers, two of the artists and me doing the audio.
We had thousands of people come by and play it, and after that we sort of decided “So what do we actually want to do now?” Everybody pitched in with ideas, and the one that we thought seemed the most interesting and doable was Monstrum. It was more of a diplomatic way of moving forward rather than an impulsive choice of “Let’s just do horror games”.
Where did the original idea for Monstrum come from, and what were the main influences throughout the game’s development?
The original pitch was basically what would happen if you mixed The Binding of Isaac with Amnesia: The Dark Descent. We ran with that concept for a roguelike horror game, something that would be replayable over and over and where you’d never get the same jumpscare twice or anything like that. That was the main emphasis that we were working on – to make a really replayable horror game.
What was the reason for situating Monstrum in the time period of the 1970s? What was it about that time period and aesthetic that was crucial to the look and feel of the game?
With regard to the game’s setting, the designer decided that with landlocked locations – such as the usual horror game tropes of hospitals and asylums – not only are these environments overused, but theoretically you could also just break a window, escape and off you go. However, if you’re stuck on an abandoned ship out at sea then you have to escape in a more specific way, because otherwise you’re probably just not going to meet a great fate otherwise!
As far as the ’70s aesthetic went, it was down to a mixture of things. In the original concept for the game, the player was supposed to be exploring an old ship but one that’s set in the current time period. It’s since shifted from that obviously, but generally we just thought that it was an interesting aesthetic that hasn’t really been done all that much. We thought it would be something that would make the game stand out that little bit more.
You’ve previously described the Monstrum experience as ‘Alien on a boat’ but interestingly you suggest that the player’s experience is more closely aligned with the character of Dallas rather than Ripley. Can you elaborate on this distinction some more?
It was one of those weird things where we thought yeah, ‘Alien on a boat’, that’s a really great way to describe the game. Then Alien: Isolation got announced and we were just like “Fuck!” It kind of put a bit of a dampener on us all; we thought we were screwed. This new game is coming out, it’s going to look amazing – we’re done. However we gradually started to pick ourselves up after that and reminded ourselves that actually no, Monstrum is probably going to be quite different from Isolation – we’re aiming at different markets, let’s just keep going. We were at EGX Rezzed in 2014 and Creative Assembly were also there with a big Alien: Isolation booth set up with VR headsets, so we got a good chance to compare the two games there. That was quite re-affirming, as it showed us that they were different enough.
The focal point of the Ripley/Dallas character comparison was based around that sense of impending doom that surrounds your character in Monstrum. Namely that Ripley survives her alien encounter and makes it through the film as a survivor, whereas Dallas doesn’t! We wanted to apply those feelings to the players; that you’re not Ripley, that you’re probably not going to make it out of this. You’re just going to have to get out of there as best you can and try. I think that was the basis for comparing the two characters.
You prototyped Monstrum as a board game before moving ahead with the bulk of production. How was that as a design experience, and did it highlight any issues about the game that you hadn’t previously considered up to that point?
It was very, very loosely tied into the game and really strange, but it did give us a feel for how the real game would eventually play out. Essentially it was a singleplayer board game where you had to move through the randomly placed corridor pieces and rooms to find the random items potentially hidden in them. We had all these different systems going on, but some of them didn’t carry across all that well because they were based on dice rolls and stuff like that, but in terms of generally planning out whether all this would work as an actual game, it really helped us look at different issues we might not have properly considered before.
How should things be placed? What speed should this monster travel at? How easily should it be able to find you, or chase after you? Little concerns like that eventually become much bigger things once you actually start to develop the game and are really important, so having that sort of stuff trickle down into the final thing was pretty useful. It’s stuff that we’re doing again now as we’re prototyping new ideas – paper prototypes are still very important.
How is the Oculus Rift VR integration coming along? With the game already out at this point, how do you plan on making VR functionality feel like an integral part of the experience?
We’ve been working on the VR integration right from the beginning of the project, developing for Oculus Rift alongside the game since we started in September 2013. It’s not like we finished the game and just decided to stick the Oculus stuff on it and then start again from scratch. So in that sense it’s been really good, we’ve been aware of any issues that we need to address, and we’ve been going back to fix them as we’ve been moving along.
Our programmer Stephanie [Bazeley] has been getting all the VR stuff sorted; there’s a lot of weird things that you might not consider when designing for VR, such as user interfaces and post-processing effects. Elements like that don’t really work the same way in the Oculus because of the stereoscopic screen. Getting that sort of stuff working has been quite difficult, so she’s done a lot of iteration on the UI systems and solved some weird clipping and animation issues. It’s all these little weird problems that you might not consider all that much until you run into a wall and you can suddenly see through the entire ship – that is a bit of a problem!
The concepts of player agency and responsibility are particularly well implemented in Monstrum – how did you manage to balance the game’s difficulty so that it works well for both brand new players and expert ones?
A lot of this stuff came about from simply doing lots of playtests to see if people wanted another shot – yes or no? If players came from a background of having already played roguelikes such as The Binding of Isaac and knowing how those sort of games work, then they could quickly adapt, learn and experiment through dying. The main issue came from people coming in from the opposite end of things who hadn’t played a roguelike before. They would go “What do I do? I don’t know what to do with this thing! Nothing is telling me anything!” We didn’t want things to be hand-holdy; Monstrum is very much a difficult game, and we say up front that it is hard and you will die quite a lot. So we iterated quite a lot on the initial tutorial room which basically just says “Here’s some stuff, this is what you use it for, press these buttons for interactions”, but beyond that you’re on your own. You have to explore, you have to check your journal to see “How do I solve this situation? Ah, I need to need to find X, Y and Z to escape – okay, got it.”
The main thing that we were looking to build into the game was that as long as the player learns something from each death, it doesn’t feel like a complete failure on their part. It’s stuff like that where if you’re new to the game and run into a new monster, you don’t know what it’s going to do so it’s about picking up on its behaviours. On your next run, if you get the same monster again, you’ll then have a better idea of how to deal with it. Picking up important information like that is very cool, and interesting to see how different people do it.
Have you been surprised by the ways in which players go about tackling the various monsters and escape routes? Have there ever been instances where you’ve seen players do something totally unexpected?
Yeah, there’s been quite a lot of those instances just because of the way the game is built. The procedural generation is one thing, but the monsters are all AI-driven, so there’s no real pre-set things or scripted events that happen as a result.
There have been a few instances where I’ve seen players carefully following a plan, and then suddenly BANG! A door suddenly just flies through the corridor and you see them freak out! It’s been fun to see how people take to that and all the different instances that can happen.
I don’t think I’ve seen anybody have the same exact playthrough, which has been our main thing, our overall objective. It’s a very emergent experience; we hear people telling their own stories about how they would be trying to find a particular item, and then out of nowhere this intense chain of other events kick off. I think that’s an interesting thing in games in general. Being able to give people that opportunity to have their own version of the same game is pretty cool.
The way you have designed the monsters according to various forms of fear is a really interesting concept; the animalistic rage of the Brute, the claustrophobic paranoia of the Hunter and the cruel mirth of the Fiend all come across really effectively in their designs. Can you talk more about the inspirations behind each monster and what went into designing their unique AI behaviours?
In terms of the monster designs, we’ve always had those three basic archetypes in mind. The Brute is big and chunky, triggering your primal fear response when it’s chasing after you. Then you’ve got the Hunter which is the thing lurking in the dark making creepy noises, and then the Fiend which is the sadistic psychological one. Even when we didn’t know exactly what the monsters were going to look like originally, we’ve always had these design concepts in mind; we basically thought about what each one’s going to do, and then expanded out from that.
The Brute was the first one we built because technically he’s the simplest. We built a lot of the backend AI work around him, and then adapted it for the other monsters to make sure that it worked for each of their unique behaviours. It was interesting to see it change up, especially when it came to stuff like the Hunter, because he’s completely different from the other two monsters in that he’s not really available on the ship and instead he pops out at you from the vents. It was interesting to get those systems all working together, especially from the audio perspective, to make sure that they all tied into the ship and the environmental sounds. You might hear a rumble and be able to identify it as the monster, or you might not and think “Oh no what was that? What do I do now?”
On that topic of encountering a new monster, you ingeniously put the Fiend into the game disguised as part of a routine lighting update. I’m guessing you must have some amusing stories of people encountering it for the first time?
We snuck it in just to basically beta-test it and see what we thought. Then we saw people on forums just going “Why are the lights going weird? What the hell is this thing?” and all that sort of stuff. We saw videos where people were just freaking about this strange new thing that had just killed them, and calling us sneaky bastards – we were just like “Yes, yes we are!” So yeah, it was pretty cool and it gave us a decent opportunity in Early Access to see how people took to it and tighten it up a bit more before release.
You’ve been using Ableton as your primary digital audio workstation throughout the project. What is it about that software in particular that appeals to you over Cubase/Logic/Pro Tools etc. and what VSTs and plug-ins do you tend to favour when working?
Ableton was one of the things that I was taught when I was at college alongside Pro Tools. I was mainly using Pro Tools going into university but it just kept crashing too much, so I decided “No, go away I’m using Ableton!” That was basically it, there was no big overarching decision to it and since then I’ve been using Ableton for the past four years nearly exclusively.
As far as VSTs and everything goes, for a lot of the sound design in both the music and the game itself, I basically used a lot of Ableton stock stuff. Their granular delay is really cool – I’ve used it quite a lot on various different things, and it’s just a really weird sounding delay which stretches the sounds out. It’s weird to describe without hearing it, but it was one of the ones I’d always go back to. It was just basically a lot of the standard Ableton tools that I used, nothing super fancy or expensive.
I’ve got some Waves stuff that I’m sometimes using in newer projects, but when it comes to stuff like EQs or compressors, I’m still finding myself going back to Ableton because I know their tools so well. I know exactly how they affect sounds so I can get them to work really quickly. I suppose that’s me being lazy and not learning new tools to a degree, but it’s also down to familiarity of speed. There’s a trade-off there I guess.
The game’s soundtrack is an interesting blend of atonal textures layered with creatively melodic noises and effects. How did you approach the composition process for Monstrum?
A lot of the music wasn’t necessarily instrumentation as much, but rather a lot of sound design and manipulating samples. All the monster themes were built around that idea.
For example, with the Brute’s theme, I really wanted to focus on it being driving and pulsing because he’s this big physical thing charging after you. That’s a lot of percussive elements with a strong drumming pulse going on, which also ties into the creature’s fire elements too. There’s actually an engine loop playing on the track which I basically warped out of time, then pitch stretched it in Ableton to make it into a constant triplet rhythm. So you’ve got these two pulsing rhythms running through the track which give it this chaotic chase feeling. Then there’s other sounds like steam screeches and stuff like that going off in the background which are basically samples where I was pitch shifting them to have them sound melodically in tune.
Another VST I made use of was Camel Audio’s Alchemy. That one was really cool, but unfortunately it doesn’t really exist anymore. It allowed you to sweep through various presets, and it had the two XY pads which are similar to Ableton as well so it meant you could do really quick automations and stuff like that. It was very handy, especially for the Hunter’s themes where I used it quite a lot.
For the Fiend’s themes I ended up using one of the Max for Live plugins called Granulator. It’s an interesting granular synth that reads little bits of a WAV file which you can then stretch out and control how many times they’re repeated. Once I’d composed the Wander theme I simply dragged it into that synth to use it as an instrument. I ended up using the Wander and Chase theme as three separate instances, so that was interesting.
You’ve mentioned before that Silent Hill‘s composer Akira Yamaoka was a big inspiration for the soundtrack. What is it about Yamaoka’s music and compositional style that appeals to you?
There are a few different things which come to mind. Looking just at his music on its own, it’s the way he uses sound and samples to create that signature feeling of his. Like with the original Silent Hill soundtrack, it’s dark, gritty and very industrialised, whereas the later ones sort of become slightly more melodic. You can sort of see a different musical feel in the other games from that point on, but he still keeps that familiar really oppressive feeling across the whole series.
When you’re hearing his music in the context of the game, it just works so well with all the stuff that’s going on. One of my favourite Silent Hill moments is the final Pyramid Head encounter from Silent Hill 2. There’s those big screeches and crunching sounds playing which mirror the movements of his knife and helmet, but then you’ve got this eerie choir coming in behind all that which gives the scene this sort of otherworldly, god-like feeling. It’s moments like that where he uses those sounds and contrasts really well together which I find pretty cool.
When designing the diegetic sound effects, did the procedurally-generated room reverb systems make your mixing process easier or more complicated?
It was a mixture of both really. Some things we made quite dry, such as footsteps, but because of the way our system worked, in some cases we had to pre-bake the reverb onto specific effects. The reverb in the game engine is mainly for spatialisation – making an effect sound like it’s actually in that area and space you’re currently in. There were some sounds which didn’t play well with our in-game reverb system, such as the environmental rumbles I mentioned earlier for example, so in those cases I had to pre-bake all of the reverbs onto those sounds and give them all different distances and drop-offs and things like that. Sometimes you just have to do one or the other really!
I particularly enjoy the small changes you make to the music as the game progresses, such as the way the main Wandering Alone on a Ship at Night theme will irrevocably change once you’ve encountered the monster for the first time.
It’s one of those things where I didn’t think it felt right for the music to sound the same after the player experiences the monster for the first time in a game. The way you’re now perceiving the game world is different, and so the music should reflect that change.
It was one of those things where musically you’re almost starting from a blank slate, but then after the first monster encounter it should change and morph into something else. You know what you’re up against now, and you want to carry that feeling across in the music even when it’s not there onscreen. That’s what I wanted to accomplish with changing up the wandering themes.
You wrote a really interesting blog post on the fallacy of the ‘game audio as 50% of the experience’ argument. Why do you think that audio design is overlooked in a lot of games design, and what can be done to better integrate the process with the other design disciplines?
I have this personal philosophy about world building when it comes to games. I think everybody should work together to make a cohesive role as opposed to everybody doing their own little thing separately and hoping it’ll all come together and work somehow. In other words, there needs to be plenty of back and forth on everything – conversations like “What’s the monster design in terms of the art team’s perspective? What does it do design-wise?” I see what I can take from those discussions and make music from them basically. Rob Bridgett talks a lot about this design approach in his book, Game Audio Culture, which is definitely worth reading. He writes about this sort of stuff and how we should improve game audio workflow, and just game workflow in general. It’s a really interesting concept and I really hope it carries forward.
It’s one of those things where people will have a lot of appreciation and nostalgia for things like game music, but they might not necessarily understand how it actually works in the context of a game. A lot of the time, other members of the team might just say we need a sound effect for a specific thing, but it’s very much a black box situation. They’ll tell the audio designer what they need, but not talk about what they’re doing, and the audio designer will just make the effect and say “Here you go, does it work?” Again, communication is the big issue, and it’s essential in order to have programmers, artists and designers understand audio designers workflows and vice versa.
It’s about seeing how you can come in from the audio side of things and how you can influence your fellow designers, how they can influence you, and making sure that when you’re trying to explain stuff to them, they’ll actually understand what you’re saying and not get confused with really weird technical jargon.
I think that’s where a lot of the issues lie. With artists and programmers and designers, they have this shared lexicon where they can probably speak to each other roughly but they might not know the technical nuances of everything. However if I started speaking about things like parametric EQs, they’d all just be like “What is that? What does muddy mean? I have no idea what you’ve said…but okay!”
Yeah I suppose it’s hard to express some of those sonic qualities accurately from a linguistic perspective. It’s similar to trying to describe the minutiae of a particular part of the colour spectrum to another person; what might appear as a bright red to my eyes might look more like a reddy-brown to yours.
Yeah, it’s all about understanding the implementation process. When it comes to getting the right ‘feel’, a lot of people will say they really want an element to feel a certain way, but getting it to that point is not necessarily down to just making the right sound effect. You can make the effect so that it sounds good on its own, but when played in the game, it might jar with everything else that’s going on.
Things are getting a bit easier with middleware programs like FMOD and Wwise, which are opening things up a bit more. We actually didn’t end up using anything like that in Monstrum for the final game, but we did use it for prototyping early ideas, which made it a lot easier for me to explain the various systems to programmers. I just had to set up all the logic and explain that this is how it works; these are all the music things, if you trigger this element, then this will happen. Even using it for non-traditional methods like that, it’s still really handy, and learning those tools is really useful for anybody who wants to do game audio and game music.
With the rise of streaming platforms such as Twitch and YouTube, do you see traditional horror games becoming something of a rarity in the future? In other words, do you see traditional solo horror experiences giving way to more community-based spectator sport experiences?
I’m not too sure. An interesting game to watch out for, especially in that regard, would be SOMA. I’m hoping it does really well and it does still prove a point that you can make these horror games that are primarily singleplayer experiences that you’ll want to really immerse yourself in and go through by yourself. At the same time though, I don’t think group/spectator-orientated horror games are necessarily bad. Things like Five Nights at Freddy’s have done really well, and every time I’ve seen [creator] Scott Cawthon speak about the games he’s basically said look, this is my work – if my games don’t appeal to you, they will to someone else.
A lot of people complain about the rate at which he’s producing his games, but from a game developer perspective I think it’s really clever, because he’s got all this extra stuff that he adds in with each new game. It’s not just another churned out sequel with the same content, there’s more things going on in each new one; he might need to develop extra systems and stuff like that, but a lot of the same signature backbone is there every time. In that respect, I think he’s done really, really well off the back of it, and the way he’s kept the continuity across all four games is impressive as well. Reading all he’s posted about the series, he comes across as very humbled by it all and grateful for his fans – a genuinely nice guy. People are obviously engaging with that series and really enjoying his games, so there’s room enough for all types of horror games to co-exist.
So Five Nights at Freddy’s might be a great horror game for one audience, whereas Silent Hill, Amnesia, or say something like Clock Tower might be more to the tastes of another. They’re all different horror games but they’re all horror games in their own right.
I suppose they aren’t mutually exclusive categories are they really – like you say, horror is now such a broad genre that there’s now games available for pretty much every particular niche.
I think the more interesting thing to look at is the different types of horror that will come in to the genre from other cultures. Japanese horror games are obviously quite big and they’re based on their own culture’s thoughts and considerations of what horror is, and the same goes for their films too. When films like Ring and Ju-on get remade as for American audiences, they didn’t really have the same sense of horror to them. They still work as basic horror films, but not in the same way; there’s this feeling that some crucial part was lost along the way.
It might just be people insisting the Japanese versions are better because they were the originals, but other people might say that there’s slight translation issues and influences that aren’t as apparent culturally to western horror audiences as they would be to Japanese ones. That’s why those films might be way much more terrifying for Japanese audiences than western ones. In that respect, I’m sure there are horror things things we have in British culture which aren’t necessarily applicable to America or anywhere else in the world.
Seeing new horror games exploring different cultures of horror will be really interesting, especially with the advent of major game engines like Unity now going for free. This democratisation of game engines allows smaller teams or even individuals to make their own games – Digital Happiness, the Indonesian developers who made DreadOut are a good example, and seeing titles like that coming out of countries you might not expect is very exciting. There might already be a really thriving scene there, I’m not entirely sure, but it’s going to be really cool to see all these different aspects of horror, or even brand new genres coming out of these different cultural elements.
A lot of people moan that this means we’re just going to get a load of random people flooding Steam trying to sell a load of random crap. While it’s kind of true on one hand, on the other it enables people who might never have had the opportunity to make their own games now have the means to get stuck in. Instead of just rushing something half-baked out, they could take years working on their magnum opus before finally releasing it; it might be this really good game. Look at people like Tom Happ with Axiom Verge for example. He was doing everything himself on that game, and when it came out people just thought it was absolutely amazing. People always seem to look at the negatives rather than the positives when it comes to things like an abundance of Unity games appearing.
We used Unity for Monstrum, and other companies much bigger than us have also used Unity to great effect, it’s a powerful engine. Lots of people don’t seem to consider that however, instead just writing off anything on the engine as just another random Unity game. It’s a shame. I guess it’s one of those things – if you don’t like these games, then don’t play them – but people like complaining I guess!
What’s next for Team Junkfish then?
Right now, we’re prototyping two different projects. One of them is completely different to Monstrum, while the other one has a couple of similarities, but is still quite a bit different. I’m not entirely sure which one we’ll be pushing ahead with, but hopefully we’ll have that sorted out in the next month or two. If these prototypes pan out then we’ll hopefully announce something next year. One of them is very similar to Monstrum, so hopefully we can capitalise on our success as well as learn from our mistakes.
Are you looking at going through the Steam Greenlight process again for these projects, or have you considered crowdfunding it through Kickstarter etc.?
Greenlight is a weird thing, in that Valve have been saying for a long time that they’re planning to get rid of it, so we’re not really sure what the deal with that would be. It might be a case of having to go through the Greenlight process with new projects anyway, or it might already be gone by that point – we just don’t know. It’s the same with Kickstarter – we don’t know if we need to do a Kickstarter, and if we did that then there’s so much planning that needs to go into that. Now that’s a scary thought! (Laughs)
Monstrum is available on PC, Mac, PlayStation 4, Xbox One & Nintendo Switch.