Originally published at: Creating Cocoon | Developer Interview - XboxEra
Earlier this year, I got the chance to sit down with Jeppe Carlsen and Erwin Kho from Geometric Interactive to discuss the mind-bending worlds of their latest game, Cocoon, and how it came to be.
I was pretty blown away after I finished Cocoon – if you’ve yet to play it, and want a broader look at the game, you can check out my review, right here:
This interview has been edited for brevity/clarity.
Jon: So, Jeppe, Erwin, thank you very much for taking the time to have a chat with me today. First of all, congratulations on the launch and the frankly very positive reception of the game you guys must be very pleased. After 6 1/2 years of hard work?
Erwin: It’s kind of nice. (Editors note: They were very humble)
Jon: As I saw these little bits of cocoon appearing over the last couple of years and the build up to it, I already kind of went in with very, very high expectations and, you know, I was pleased to say they were all met. But when I think of 6 1/2 years for a game like this, how did that process start? So, you know, you’ve got an idea in your head to get to that end result. Was it always going to be worlds within worlds? Was it like “I want to build a puzzle game”? What was the initial impetus for the idea of cocoon?
Jeppe: Yeah, it is. It’s a long time and it changes a lot also over the course of developing a game, but the sort of first seed is I want to make a puzzle game with Worlds within worlds. That is where it begins. And me and my co-founder Jacob in Geometric, we made a prototype of this idea and it’s just very purely gameplay prototype.
You could play the game for around an hour, it had many worlds, it was too complicated for its own good, but it also proved very much that we are on to something and when we did that I had already been thinking about the game for quite some time also. So that was like just a lot of ideas that every time I thought about them, it got a bit extra refined or something new came up that I just got more and more fascinated by it, so I was like, let’s sit down and do this, when we founded the company and just blast it away. Just building person after person after person, just extremely quickly. And then you sort of land a little bit afterwards and figure out what are the next steps here.
But the game changed in many ways into something very different than what I had imagined, and that’s largely also due to Erwin’s work. So, we sort of made an early decision that we wanted the worlds to be not like 140. There’s just something about the nature of the idea of having worlds within worlds that it’s just more fascinating if the world seems more real. And that’s why we reached out to Adam and luckily he was super interested and got on board. But then I think the real process, the real heavy one begins which is that we now have to figure out like exactly what this game is. And that’s not me having like a master plan for what this game is. I think when you make a game like this in a team of, let’s say, 10 people – I think my role is to try and listen to the game and figure out like what the game wants to become because, I don’t have control over Erwin, he’s way better at what he does than me. So my job is to look and listen in a way, the way I see it. So I do have a clear idea in many aspects of what the game becomes. But then, all the different choices and turns we make along the way, is much more team based and innovative. A very long process where I think my role is more to listen, not necessarily only to Erwin. Just listen to the game in a way, if that makes sense. That’s at least the way I see it, because there’s too many parameters for me to control it, so I’m also a little bit of an observant, in the process of trying to steer it in a way. That’s the way I see it.
Jon: Yeah, it’s interesting you say that because the world of cocoon, you know the creatures, the insect-like nature of it, the different biomes that are part of each world, suggests some form of alien and insect like technology. And I understand that originally it was just an astronaut and he had a cape and then you kind of like snipped that cape in half and it turned into….
Erwin: Right. I mean I saw a sort of stylized art style was something that I was already doing in my freelance career and yeah, then Jacob saw a series of prints that I was producing in that time and that was very much like my fascination with, like miniature biomes and and terrariums. And my fascination for biology. So that’s more like on a basic visual level. And then for the game, in terms of the design language of the insect and the industrial merging with the organic. After we figured out like, oh, this astronaut, the astronaut with the cape, oh there’s something very insect like about it. And we even had a little guy like, running around. But that that design evolved a little bit into the final version that you see in the game. And then the rest of the style of the architecture and technology sort of worked in tandem with that. And I would say it’s a very, very iterative and organic process to develop and yeah, it was lots of conversation back and forth about, OK, where is this thing going? Because I also had like, no idea! The options, right? Like your even the possibilities are huge. It’s like, oh, ****. Like, what is it going to be?!
Jon: As I played through it, I mean, there’s moments that make you feel like kind of icky and it’s not real, but… again, I refer back to AAA games and they have this obsession with realism, and have to be looking the ‘best’ and I often find the games that live forever are the games that are stylised and unique. You know that Cocoon’s going to look fantastic in 10 years. Just as you know, the Last of Us Part 2 for example is going to look really old and weird in 10 years. You know what I mean?
Erwin: They’re remastering it now, right?
Jon: Indeed! And this this is one of the things I love about games, right? I’ve played games for 30 years of my life, right? So, I know the unspoken ‘rules’ of video games, but my partner is not a gamer. I finished the game for the review and I was like, “that was stunning” and as I talked about it, she was intrigued. She understands what a game controller is, but that is as far as her knowledge of games goes. And I was amazed to pass her the pad and see her just figure it out. I said, look, there’s only two buttons, right. I’ll let you figure it out without…
Jeppe: One button. Yes, yes.
Jon: Well, move stick and A, but you know what I mean! But, without a single word, without narrative, without a UI, without a ‘here’s a hint’ system, it was just animation and sound all just combining to just instruct and teach her exactly what to do. And she’s, I don’t know, she’s about 84% of the way through right now. I’ve had to help like twice with some timing stuff, but I think is the beauty of this medium and I have to wonder – when you guys are building this kind of thing, is that in the back of your mind? Like the way that can communicate to the player without saying a single thing – without this like ‘click this’? It’s just: you’ve only got 1 button to worry about guys. So, try it out and the player will just naturally do it. What’s the process for that? If there is one or is it just iteration after iteration?
Jeppe: Lots of innovation is one of them but it is also sort of the game design angle I’m coming from, but a game like this is to try and sort of teach things without saying anything. But I’m instructing you very well, you just don’t know of course! That it’s very precisely laid out for you to understand everything, but that is the test. And I think with this game in particular, compared to some of our previous, the things I’ve either worked on or my own little games also knew that what I eventually had to teach you here is of a more complicated nature. Like it’s more complex. Which is specifically more the latter half of Cocoon, which was the plan all along that it sort of needed to get there. And for me it just made sense that sort of the more complex I want it to be then the more simple it also has to be in a way for it to still be approachable. So there was very early on, sort of mindset from me is like accessible complexity or like you know, simple and complex at the same time. And then of course the controls and all that as a natural way to start by making it very accessible.
Jon: Brilliant. The sound design plays a part in this as well. Kudos to you guys for helping me out with the soundtrack for my video review, because I always like to use the music that’s there. They’re kind of these ethereal pieces, and I almost noticed that through sound design, when you are on your way to a solution, somehow the game knows that you’ve figured it out and there’s just a little motif that occasionally will play to say, yeah, you, you’re on the right path. It’s really subtle, but am I right in thinking that that is there and I’m not imagining it?
Erwin: That is there! Yeah.
Jeppe: Extremely late addition as well, right? It was like almost the last night and the producer was shaking his head a little bit when I said I think maybe that could be great. Also, the team did a little bit, I think. Also, if that motif was playing like for every single person, I think you would get really tired of hearing it. So the way I sort of laid it out was that every time you solve something that has specifically to do with you having figured out how to utilise the hierarchy of worlds in a new way, it’s like visualisation. Now your mind has expanded just a little bit every time and every time I said register, I think you are there and then then it happens. So it’s always if you play the game again, you would notice that these motifs play specifically when it has to do with those puzzles. And if it is the same… sometimes I repeat a puzzle, just to be like, hey, do you remember how this works? Kind of thing. Then I don’t play it. But then when you do it in a new configuration kind of way, something new has clicked, then it plays. It is specifically also intentionally playing it. I think one interesting thing is that some of the process in cocoon we could also see that when we play the game, players are pretty certain that they got it right. But they are not 100% certain and you have this moment in the game where you’re walking around with the orb, with the solution sort of packed in it, and you’re in this anticipation mode to see, I really hope this works. And then when we kick in the motif, it’s to give you a little confidence boost like this. This will work. Trust. Trust yourself. This will work. Kind of. That’s the intention.
Jon: I saw an interesting post, I think it was on Reddit where someone had created some sort of insane Visio graph of the game laid out in all of its sections and layers, and I was looking at it…I mean, I couldn’t make head nor tail of it at first, but then as I started to really look at it, I was like oh, yeah. And then it expands and then we can go back here and it grew and grew. Was it kind of crazy to you after 6 1/2 years worth of work? You come back and you see some person has just completely laid out the game: like here it is. Here’s the design, this is how they did it. Does that blow your mind a little bit, people are that into what you’ve made to make that sort of effort?
Erwin: Right. Like it is amazing. I think we should have something like this in the studio on like a big poster. Like our first cool fan art.
Jeppe: We should print it, I get ashamed of my diagrams. Because they look like ****! I’m way, way too impatient to do something that neatly. Erwin can confirm my documents are only for myself to read and they make zero sense. This looks amazing. It’s like, when you have games, you always know what fan art is like. People put very little characters, but this is. game design fan art. That’s the way.
Jon: From a design perspective Jeppe, all of your previous games have been 2D – yes, they’re puzzle games, but they exist in in one linear line, right? What was the transition like to thinking in 3D?
Jeppe: It’s a little bit 2D, just a plane, but it’s also like left to right platform is almost also like 1 dimensional in some aspects of the design. I think the main reason was more again, the nature of the idea, I think you could in principle make this as a 2D side scrolling idea, when you also jump into worlds or whatever, I just don’t think it would make sense. I think it’s about the feeling that when you’re landing there that’s a world to explore. It doesn’t feel like that if you land on the line and can go left or right, I think that would just be very strange. But I think with this game, it was a careful step into 3D. It’s still very controlled and from a gamer’s perspective it’s taking place on a plane occasionally also like using 3D, instead of being a story that’s structured on a line. And I think it was nice for our team, like I’ve done enough games that are from left to right. Also, I think our next game is most likely not going to be a left to right game.
Jon: I’ll see you in 6 1/2 years!
Jeppe: Hopefully, hopefully not!
Jon: But it’s interesting you talk about the transition of worlds. One of the things I really loved, the animation and the art style that combined together that first time that you hit a pad and you fly up, you buzz for a bit and then the World just pings out. I mean from an art perspective, it’s extremely slick. Like you never get bored of doing that, even when puzzles have you putting one world down and swapping them over and going in and coming back out and realising that things are affecting things, in other worlds. What was that like from an engine perspective? Because obviously you have got to have these running concurrently and have this instant load across a myriad of devices from an engine perspective. How tricky was that to pull off?
Erwin: For us as the artists, it was actually the easiest because the character had existed, right? So although we need to make sure is that as you pull out, you do pull out really fast and it gets like motion blur, but all we need to do is make sure that there’s like enough landscape in the far distance, as you’re being pulled out and most of it was just technically an interesting challenge that, yeah, the the animators and our lighting guy had to figure out, like, how to do the rendering and how to make the animation, the pool and all the particle effects and all that.
Jon: That’s amazing. It’s really cool and that’s across all platforms as well. So it’s not using any magic. You know the SSD’s of the future or any of this nonsense that the marketing teams like to push? It’s just solid forethought and design.
Jeppe: No, I think that aspect of it that you are pointing or hinting towards, which is like loading of content. I’ve also seen like other podcasters or whatever play the game in a let’s play a video or something where they talk about it like this it must be sort of, because we can build something. It’s not that complicated. In reality, it’s an illusion. It’s that complicated in all those, let’s say, left to right games. You have to load in the next room that you’re progressing on. So here basically we have to make sure that because we know where you’re going to land, right? So if you pick up the green orb, and you’re approaching a place where you can jump in. We have to start making sure that we load in basically that part of the world and computers are also fast nowadays. So we can technically have a lot of what is around a respond point running. In the beginning we had the solution, we thought we could just have the entire game running. That was an illusion. We have lots of loading systems and lots of colouring systems, making sure that performance is tight and but it’s of course more complex than if you have a game where you can just run forward, but I don’t think it’s that as insanely complex as you might think. I think there’s also just an illusion thing happening.
Jon: I know that we’re coming up on time here, so I want to get in a last couple of quickfire ones. I am aware of a ‘secret ending’. I think it has been, believe it or not, cracked already. I think I’ve stumbled upon the beginnings of it. But for the achievement completionists out there, this is an optional secret ending, and this is for someone who really likes complicated clues and puzzles to go and dive into and spend some more time in Cocoon. There isn’t an achievement that’s in the game that’s locked to this secret ending. Is that correct?
Jeppe: That’s correct.
Jon: You’ve already mentioned or hinted you’re working on your next project. You’ve already got some ideas in mind. Are we staying in the world of isometric 3D or are we stepping back to 2D?
Erwin: Yeah, I can’t say if it’s going to be like an asymmetric thing or not, but I’m a 3D person. So, it’s highly likely that’s what it’s going to be. 3D models are going to be involved in some shape or form. And highly likely it’s not going to be from left to right.
Jon: I really appreciate you spending some time with me to chat about Cocoon today – it’s a wonderful game and congratulations again for all the critical acclaim and great reviews.
Jeppe: Thank you – yours was the first review that we saw and we were extremely excited.
Erwin: We read and watched your review on the big screen. In the office!
Jon: What, my review? Oh, wow. And you know the fact that you guys were kind enough to help me improve what I had made when you guys lent me the soundtrack that was very kind of you. So thank you.
Jeppe: Lots of respect for even reaching out and doing that, like most don’t do that. They just put on some music or whatever and then… Yeah, very, very professional video you were a part of -aside from giving us a good score. Well done!
Jon: We do our best! Thank you both for your time today.
