I played an amazing looking game this week, Fragments of Euclid by Antoine Zanuttini, a short first-person puzzler that appears to be set inside the art of M. C. Escher.

For me, however, it’s more like a dry run for William Chyr’s Manifold Garden, a game I’ve been looking forward to for a while now. Manifold Garden is also a first-person puzzle game with Escher-inspired impossible geometry. It should be no surprise to hear I discovered Euclid through Chyr’s Twitter feed.

Can Euclid tell me something about Manifold Garden?

Euclid really nails the Escher style, but it’s fake Escher. Take a look at the opening level, “Relativity”:


And compare it with Escher’s Relativity:


Without doubt, it is a stunning reproduction… but of form only.

In Euclid, the player can change their relationship with the world, shifting into an alternate perspective where gravity defines a different down. Through this modest gift of agency, the player can solve puzzles. However the player selects from a discrete set of baked-in perspectives – of which only one is real at any time. Escher’s optical illusion blends multiple perspectives into one beautifully incoherent image. Euclid allows you to switch orientation and invoke a different reality, but you can’t exist within the incoherenence of Relativity. An optical illusion is the art of the impossible but Euclid is the art of the possible.

The original title of Manifold Garden was Relativity, leaving no doubt as to the game’s source inspiration. However, Manifold Garden evolved its own style, less Escher and more of an ascetic take on NaissanceE.

manifold garden
Manifold Garden

Optical illusions only work because the artist controls what the eyes get to see and such illusions rarely survive when the eyes are given the freedom to explore. Manifold Garden accepts this and creates something else. In reality, Euclid does this too, but sticks closely to its inspiration for thematic value. This is where things get a little hairy because Euclid allows the theme to overpower the puzzle gameplay.

Put Escher aside and you will soon recognise that Euclid is a riff on Portal. The difference to Portal is that all portals are predefined and passing through one will change the orientation of the world. The problem comes in the dense, monochromatic and repetitive nature of Euclid‘s structures because the environment is just plain difficult to recognise once reorientated. Stepping through a portal does not feel like stepping into the same room with a different pair of eyes – it just feels like stepping into a room you’ve never seen before. This is a serious problem because the whole game is about the player’s conscious search for the correct puzzle-solving perspective.


In this way, Euclid‘s first “tutorial” level becomes a frustrating gauntlet of confusion. The developer expects you to follow the path up the stairs and enter one portal which leads to the level’s exit, oriented correctly. It is not obvious you’re supposed to do this and you’ll likely end up doing what I did – jump around the wacky architecture and explore. It won’t take long before you’ll encounter a warning “not to wander” and a command to press the R key to reset. You’re always one jump away from landing on a surface from which there is no escape.

This conditions you to treat exploration as a dangerous activity and slow down, focus on the environment. But it takes real commitment to notice you’re actually in the same room when you pass through one of the portals. This fractured, chaotic first level sports no strong visual cues to keep you grounded.

Take a look at the trailer for Manifold Garden. On first glance, it might seem to suffer from the same problem.

But I’ve watched William Chyr’s GDC presentation on the Manifold Garden‘s design challenges where it is clear that Chyr is aware of the potential issues. In Manifold Garden, the player explores small bubble universes. It might look like the level consists of a structure repeating into infinity, but space has been bent back on itself, and you’re actually seeing the same small structure that you’re standing on, again and again. If you fall off a platform, you’re going to land back where you started, not on another copy of the platform. Of course, this isn’t to say that some of the Manifold Garden‘s challenges might turn out to be a little too unwieldy for our brains to parse, but there’s a generous simplicity here than Euclid resists with its opening showpiece. The irony is that Euclid‘s first level is probably the most confusing of the game – the next few levels are back to basics affairs.

However, there’s one problem I do not think Manifold Garden will tackle any better than Euclid. Whenever I encounter a puzzle game with blow-me-away visuals, I always feel a pang of disappointment when I discover, in the end, it’s just another game about pressing buttons and unlocking doors.



  • More clear visual cues would fix some of Euclid‘s problems but getting rid of the portals completely would also do this. Remove discontinuity and the player is far less likely to get lost. The player would then need some other mechanism for changing perspective but then we’re getting even closer to Manifold Garden in which players are free to choose world orientation at will (from what I’ve gathered). It would also change the nature of the game entirely.
  • I don’t mean to imply that Manifold Garden will be disappointing. But after seeing many exciting screenshots of Manifold Garden I was cautious what the “game” actually was. I’ve tried not to get too spoiled in advance but it did look like locking the player inside a room with a puzzle, which I found a little deflating. Naturally, Manifold Garden promotes its awe-inspiring visuals first. Contrast with The Talos Principle where we were never in doubt it was a game about lasers and doors. And The Witness was about line puzzles. By default I’m more jaded about great graphics these days – I assume the mechanics will be something traditional, tried-and-tested until proven otherwise.
  • Ya know, I couldn’t tell you what I wanted instead of puzzles with buttons and doors. I do know I’ve had enough of “figure out the correct sequence” puzzles which turn up everywhere from Tengami to Sword & Sworcery.
  • Euclid does have some other issues such as a sensitivity to dropping cubes in exactly the right spot. These problems are not impossible to solve and could be worked out in later iterations of the concept.
  • I’ve talked a lot about its frustrations but TRUE STORY I liked Euclid!

Download my FREE eBook on the collapse of indie game prices an accessible and comprehensive explanation of what has happened to the market.

Sign up for the monthly Electron Dance Newsletter and follow on Twitter!

9 thoughts on “Art of the Impossible

  1. “Ya know, I couldn’t tell you what I wanted instead of puzzles with buttons and doors.”


  2. Oh you *know* what I mean. If you’re going to have a puzzle game, then what kind of puzzle game will this beautiful virtualscape offer? Buttons and doors.

  3. I was really entirely serious. I played the beginning, and got the message “You should not wander around,” and I was like “What the hey? First of all, you haven’t given me anything to do but wander around, and second of all, wandering around in this amazing scape is what I want to do.” It wouldn’t even be a not-puzzle game if it were a Bernband; exploring in a non-Euclidean space like that is a puzzle in itself. One looks for the hidden places in Bernband, does one not?

    (You could make a collect-all-the-trinkets puzzle but I suspect this is not what you’re looking for.)

  4. It is strikingly hostile to exploration and I think that’s as a band aid because you can get stuck way too easily. “Do not wander” is definitely a line I’d cut on a later version!

    For it to be Bernband, I think it needs to be safer. But you might be right – the Relativity level is so difficult to navigate that it is a challenge in itself.

    And there is the secret room, of course…

  5. Amazing! Coincidentally (very!) I played yesterday the game by the same author of “Fragments of Euclid” – “Non Euclidean Room” ( “NaissanceE” memories were immediately here, since that game changed my life perception and everything. The difference was that unique genius loci of “NaissanceE”, which became the perfect narrator. The place, the dimension told story, perhaps the most cryptic and most interesting I’ve played so far. I missed such narration by location in “Non Euclidean Room” and hope, it will be in “Fragments of Euclid” and in “Manifold Garden” as well, otherwise it will be just a nice piece of interactive manierism. Visually all works impress me, especially because I always dreamed to walk inside Escher imagination.

  6. Merzmensch, alright, sounds like I should try out Non-Euclidean Room! Have you played Kairo? That is one of the best examples of storytelling purely through environment. It’s not easy to unravel and many were confused. I have a whole post dedicated to explaining it 🙂

  7. I bought Kairo on a Steam sale right after reading HM’s wonderful article about it. Based on this conversation I will have to get started with it soon!

    Amazing what art style can bring to these experiences. The pencil shading in Fragments of Euclid evokes a certain mood. The game would be the same had it been clean cel shading or “traditional” 3D, but the mood would have been totally different. It takes skill, also, to present so starkly monochromatic a world; in this case it looks like the developer doesn’t always succeed perfectly, given the sense of disorientation you apparently feel when passing through some portals. Whether through use of color or specific architecture cues, something needs to anchor the player in conceptual space or else you’ll run into a situation where some players don’t realize they’re in the same place at all. I can see that not ending well.

    Very interesting find, HM!

Comments are closed.