Game Dev Insight: Manifold Garden
Manifold Garden is the highly acclaimed first-person exploration game by William Chyr. This installation artist turned game dev has spent 4 years on the project; and though Manifold Garden has roots in art and architectural design, it also challenges players with physics-defying puzzles. It’s also a natural evolution of his work as an artist:
“I really saw, and I still do see the game as an extension of the installation art; and that’s about experiencing and perceiving space.“
We recently spoke to William in our Get to Know a Game Dev series, and he had a lot to share about his upcoming game as it enters the first round of Beta tests. Besides his daily dev log, it offers a great glimpse into the development, iteration and final testing of a unique exploration game like Manifold Garden.
It really feels like you’re building an experience. Is there any particular inspiration for the game or how you approached the build?
Well, there’s a few ways to answer this. Regarding the playability of the puzzle, in terms of development I very intentionally decided at the beginning that I would only focus on puzzles. I knew that was something that would take the most amount of time, and would be harder to integrate further along. It started with this idea that you’re an architect and you go inside people’s brains. But I found myself throwing out ideas because they didn’t fit into this narrative – it was restrictive and was not letting me explore. So I thought, I’m just not going to do that; forget about that and just make sure I have a really strong puzzle.
I think it took about two and a half years for that to be solidified, before I was able to start thinking about any of those other things, like narrative. The narrative aspect of it did not even really come until the beginning of this year, so it’s been three and a half years! It’s only a couple months ago that I understood what the narrative aspect would be.
Did that ever make it difficult to keep working on the game?
On a purely production level, being that I was involved for so much of it, I knew I couldn’t do all of that. I thought, ‘I know the most fundamental part was going to be the puzzles, so that’s all I’m going to focus on and later we’ll add stuff.’ I just underestimated how long the puzzle aspect would take.
Haha. So did the vision for the game change much, as the puzzle structure evolved?
When I first started out I wanted to do a typical sci-fi style, (you know, gooosh (slidey-door-noise)) and I cannot draw like that. So it ended up forcing me to work solely with boxes, and I knew – I mean the entire game is boxes, I just pick a face to work on – and I knew that shaders would be easy-making. Shaders is mostly programming, and you can do a lot of really interesting stuff in terms of art style, and what’s great is once it goes on the camera? Everything’s affected. So from a cost-benefit analysis, that is the most bang for the buck. Just have a bunch of boxes, put [shaders] on the camera, and all those boxes look good. And I knew that because I’m not good at some objects by hand, it’s easier for me to do it this way.
Did it still feel like making art, at that point of game development?
I’m an artist, I think of myself as an artist in the sense that what I create is art. But not in the sense that I can draw. I’m an artist in the thinking sense, but not in the usual sense. But what I am good at is I can tool the numbers in the shaders and do some refining, so I was able to work to my strengths.
After 3 years of iterating and refining that, I feel now the game looks really great but a lot of that is because of those choices I made early on: we’re going to stick with an art style that I know I can do well, and just going to refine that instead of finding another artist. So modelling it in that pipeline allowed it [to look great]. Those constraints early on can seem weird, because it just seems like every studio has an artist.
It sounds like your pipeline decisions were made for strategic reasons…
Haha, that might be generous. Budget constraints, strategy… yeah, I got lucky in a lot of ways. Like it’s a first person game, it’s very mechanics focused, and the mechanics mean everything has to be 90 degrees – it’s a lot of angles – which eliminates a lot of the issues. Because of that I’m able to make this massive game with such a small team.
Have you found being your own team to be a personal challenge?
Now it’s not so bad. I think maybe the 2nd and 3rd year of it were the worst where it, you know there was that first beginning – the first year, you think it’s only going to take a year and there’s a lot of excitement. And I’m still very excited about it.
Now I’m in a phase where the game, I was actually just thinking about this yesterday. It doesn’t feel like I’m making the game anymore. It feels like the game has legs now and it’s running, and my role is to just keep feeding the fire. That’s a great feeling. Because of that, I would say the last 10 months I’ve taken a much more balanced approach: I don’t work all the time.
And you’re seeing some payoff: a lot of interest from press, already a few awards?
Yes, I know what the game is about finally. I know what I’m making, and it’s so weird to be saying that but it took three and a half years to understand what this is. It’s like I had ideas, but they were swatting at something; it wasn’t fully there. Now it just was like, I remember earlier this year that one moment where it just felt like everything came together. I could understand it; it was just like crystallized, I could see it. There are still a lot of questions and unknowns about the business side of it, like the launch and what’s going to happen then, what’s the best way to approach that, but in terms of what the game is, if I look at the game in a vacuum I do know what it is.
I just finished a video; I’m not calling it a trailer because sometimes you say that and people are like, where’s your gameplay? It’s more of a point-scene experimental thing, and then you’re not working with the system of the game but you get to still exercise your game. It’ll be on the Steam store page.
The footage from the game is so fun to watch, even if it’s not actual gameplay just seeing the aesthetic being exercised is rewarding.
Yeah, you know with architectural photography and videos the difficult part is, if you were to do that in real life, it would be incredible physically demanding. You’ve got to pan, get stairs… so what’s kind of nice in the game, not that I’m against physical exercises, is that you can do these grand sweeps very easily.
What’s next in terms of the game plan for Manifold Garden?
I’m going to publish it, so not getting it to publishers. Where we are now is we’re almost done with all the mechanics. There’s like two or three there; a lot of what’s been difficult about making this game is so many of the mechanics affect one another. So you’ve got the gravity changing, and you have overlapping of these boxes and trees, but the boxes and trees are tied together with the overlapping, and early on when you’re prototyping one, you’ve got to put them in but neither are fully finished. It’s like building a house but everything is changing, so you’ve got to tear down and rebuild.
I’ve watched probably a dozen people playing the game in the last 4 years. But I think I’ve been to even more conventions over the last three and a half years, and at each one you get 50-100 people on the floor: that’s easily 1,000 who have seen it. That’s been incredibly helpful in figuring out the design of the game, but now [testing] is going to be 6-8 hours (or even 20 hours or more) – and that’s just not feasible for me to watch someone play the whole thing, but we still need that data.
On the plus side, here’s the map of Manifold Garden so far. ~60% of levels are now connected. Game is pretty big. pic.twitter.com/yEbd6t6zVW
— William Chyr (@WilliamChyr) August 20, 2016
So we are moving into rounds of beta testers. We have our first round signed up, about 200 people, and we’ve been setting up a pipeline. So there’s a bug reporting system in the game, interplay, so they can have a title for the bug and type in a description and take a screen shot. So we’re still finishing that up, and the idea is that this would hopefully be used by the programmer. What I want this to do is get us into that practice. We want to do this once a month up until release, so that that means by the time we release the final game, it shouldn’t be this ‘Holy cow, we put this out and haven’t thought about it’, but it’s the drill – it’s just another [bug] and we’ve done it before. Ideally we’d identify as it was built, but I want it just to be that [easy].
I’m sure you’re learning a lot, but your dev logs really help others to learn from your process.
That’s kind of why I keep sharing so much, like every time I talk about dev scenes, I’m like you should try streaming, to build a pipe for the game, like I was just telling my friend about that – and his game’s coming out in a month – he’s like, what can we do to hype it and will streaming get results? But that takes time and you need critical mass.
Last year, I started streaming knowing that I’d need it. And now I’m in such a great spot, I got partnered recently, and every time I stream I get up to 70 concurrent viewers at anytime. We still have a good ways to go, and it’s at that phase now where you’re starting to see it’s easier to get more people to watch as you’ve got this decent crowd. But it took that time to get there.
I was sent some Manifold Garden fan art today! 😀 pic.twitter.com/puZpl6t32t
— William Chyr (@WilliamChyr) August 10, 2016
Some people get held up in well, they keep pushing back when they want to start streaming because they want to make sure it’s really good, but I’m a big believer in, just put something out. Once you get it out there, like this trailer – it was on my to-do list and two nights ago, I was like… I’m doing it; because if I don’t do it tonight, it’ll never get done. And once it was done it was really easy, and I can always give it another shot. It’s so easy for me to take that, post it on Discord and get feedback.
And I want to do the same thing for [beta] launch. Just get something out there, that’s a safe space to fall – they’re people excited about the game, they know it’s our first beta test so if there’s major mess ups? It’s totally fine. That means by the time we launch on Steam or launch on PS4 we’ll be – I don’t want to say experts, but we’ll have gone through these processes.
You’re going to make mistakes, just try to get those early ones out of the way. The dev log, you can see, I try to share everything; and it doesn’t matter because no one’s going to judge me. Yeah, your game looked terrible at the beginning; yes, I know! It can’t hurt me.
Is there anything unique about Manifold Garden that people might NOT know?
I changed the name from Relativity to Manifold Garden, because one problem with sharing a lot of early work is sometimes press will just Google your game and take whatever screen shot they find, and it ends up being an old one. But fortunately with the name change, we were able to just be like – now the old pictures, they don’t find.
That’s pretty clever; it’s better than fighting SEO all day long.
It was intentional, and I was kind of worried about that because I thought I’d lose a lot of momentum, but it ended up being the biggest saturation I had up until that point. It was definitely the right decision; I mean I love the new name.
Because I’ve kept all these old builds, one of the things I’d love to do closer to launch is a streamed evenr: a 12-hour feed where I just play through all the old games and talk about them, just like a trip down memory lane. I’ll be like, this is what I was thinking at the time, you know just to see how the game has grown. It’d be really interesting I suppose for me, and it’s such a rare thing to see – not too many people share that.
You mentioned Steam, PS4 — any reason for one platform over another?
I’d love to put it out on as many platforms as I can. I mean there are some that… mobile isn’t the right one. From a performances standpoint, the specs of that don’t seem to be, for how the game would play, wouldn’t make sense. In terms of console, it’s really just that Sony approached me, and I’ve had a very wonderful time. We’re just two people, and it’s hard for us to take on… I see some people doing a game on PS, Mac, Ps4, Vita; each one of those is very challenging. I just want to make a really solid game. Everybody asks me about VR all the time, but I know it’s not going to work – we have our hands full to just do two platforms. At the end of the day, we are two people making our first game.
Well if your game does well, who knows – a time for VR may come. (He laughed, but I think William hated us just a little for saying this 🙂
To follow William Chyr‘s progress, be sure to check out the dev log on Twitch every weekday at 4pm EST, as well as the Steam store page below. We’re looking forward to seeing the development of Manifold Garden in Beta testing and will be sure to check back with William along the way!
- On August 22, 2016