One of the more understated titles on E3’s show floor this year was Manifold Garden, an upcoming first-person puzzle game which challenges players to think outside the box — in a major way. If you’ve played Antichamber or Portal, you can get the general gist of how it plays: three-dimensional spaces, while initially seeming ordinary, unfold into complex, non-Euclidean structures like something out of an Escher drawing. Walls become floors. Pillars become staircases. The entire universe loops inside itself, an action in one plane affecting all identical planes outward into infinity.
Manifold Garden is, in short, a game that’s incredibly easy to grasp visually but nearly impossible to pin down in text. Many of its early levels are designed around a triple torus, in which the three dimensions we’re most familiar with — length, width, and depth — loop back around into themselves. Keep going in one direction, and you’ll end up where you started. In games, we’re accustomed to seeing toruses like this working in two dimensions (think walking to the top of a map in an RPG and reappearing at the bottom), but the third is a little more elusive. The first thing that came to mind when I first saw it demoed was actually that half A-press video which became a short-lived meme a while back, in which a glitchrunner sends Mario to successive “parallel universes” by exploiting a level’s outer boundary. Honestly, I feel I’m belittling this game by even making a comparison like that.
Sorry if you didn’t realize you had vertigo before this moment.
But unlike the glitch video, Manifold Garden‘s “parallel universes” are plain to the naked eye — and they’re incorporated into puzzle solving. You may need to send a device or a stream of water from one version of a structure to an adjacent one. A building that is unapproachable with the gravity pointed in one direction may become accessible when you tilt gravity by 90 degrees. The level design is so complex and fascinating that its lead architect, William Chyr, gave a talk about it at the Game Developers Conference last year.
ZAM: What was the seed of what would become Manifold Garden?
William Chyr: I was an installation artist before entering games, primarily creating large-scale installations out of balloons for science museums and art centers. The problem was I got typecast as “the balloon guy.” I looked for other mediums to work with that would break me out of that mold, trying glass, metal, but none of them I really connected with.
A friend of mine was really into videogames, and showed me a lot of the work that was being done in the indie game space. He showed me games like Journey, Flower, and Braid. When I played Tale of Tales’ The Graveyard, that’s what made me think I could make a game. Mostly, it was because of the size of that — a short 5 minute experience, which felt very doable at the time.
‘Relativity’ (M.C. Escher, 1953)
The game initially started off as a small project to learn Unity. The idea was to make a game based on M.C. Escher’s work, with a focus on puzzles like Portal. I had studied physics in college and had worked in a nuclear physics lab before, so the idea of exploring physics thought experiments through games was really appealing.
The project was initially called Relativity, based on the Escher print of the same name, and it centered mainly on the mechanic of changing gravity and walking on walls. The first version actually had the environment rotate instead of the player, but that led to a lot of design problems (e.g. after the first rotation, all the objects ended up in the corner). I changed the mechanic to having the player rotate, because the metaphor [of the name Relativity] is that the problem (the environment) stays the same, but now you’re seeing it from a different perspective. Changing the environment felt more like changing the problem.
World wrapping and architecture were not introduced until much later. I’ve actually always been interested in topology and architecture, one could argue those ideas were always there, just didn’t find their way into the game until much later.
When it comes to making these puzzles work in looping three-dimensional space, what are the the biggest concerns? Was there anything which emerged in playtesting you didn’t expect?
The biggest difficulty has been signposting, or guiding the player to where they need to go without using words, characters, or a map. This is difficult even without the looping 3D space, but that just makes it even harder.
During playtesting, I’ll often see unexpected ways of solving puzzles. This is part of the process, and much of the time designing a puzzle is about removing the “wrong” solutions. Some puzzles are meant to be solved via different ways, but there are also some that are supposed to teach a specific concept that players need to know later on. For the latter, it’s important to make sure players arrive at the right conclusion so that they have the correct mental model and the necessary tools to move forward.
Are there other works that you would consider important influences?
Starseed Pilgrim is probably the biggest influence in terms of game design. Not many people see it right away, but the meta structure of Manifold Garden is heavily based on the meta structure in Starseed Pilgrim.
A panel from Blame! (Tsutomu Nihei, 1998-2003)
Blame! by Tsutomu Nihei is a manga which has a had a huge influence on the visuals and architecture of Manifold Garden. It features a lot of mega structures that the characters traverse. It does something really wonderful which I’ve never really seen before in manga — you’ll see some tiny spec in the distance, and a few panels later, the characters end up there. It create a really strong sense of scale and atmosphere.
We’re to a point now where you might consider “non-Euclidean building navigation game” its own subgenre, with Portal, echochrome, Monument Valley, Antichamber, and now Manifold Garden as well. Do you think it’s a matter of the technology getting better, or are players just getting more receptive to games which (openly) break with “realistic” architecture?
I think exploring worlds that break “normal” physics is something that appeals to a lot of people. I think the success of a film like Inception is a clear indication of that. I remember when I saw the trailer for Inception for the first time, and was immediately drawn to it. It didn’t even matter what the plot was. Getting to experience a world that was so fundamentally different — the basic rules of physics were re-written — felt incredibly exciting. The fact that so many people love Escher’s work is also a sign. So I think on some level, it’s just human nature to want to experience different worlds.
I’m not sure if technology getting better directly contributes to that, other than making the tools of game development more accessible, and thus easier to experiment with more unusual ideas.
Manifold Garden is due out later this year for PlayStation 4, Windows PC, Linux, and Mac. The above interview has been lightly edited for space and clarity. More screenshots, animations, and development updates are available at manifold.garden.
