Best of Quest 2022: Designing ‘The Last Clockwinder’ and its Time Looping Puzzles

Knitting. That was the unlikely inspiration for Pontoco’s time-looping puzzle game The Last Clockwinder, winner of Best Game of 2022 in our annual Best of Quest Awards.

“Knitting is essentially doing repeated actions on a loop,” said John Austin, Game (Co-)Director of The Last Clockwinder, “and that immediately led to this idea of making clones of your own actions. We liked this idea of playing with short loops of just a few seconds and seeing all of the clones working together in harmony.”

The knitting framework eventually unraveled, but Pontoco kept exploring and refining the cloning and time-looping mechanics. Add in a touch of Studio Ghibli and Moebius, as well as a poignant story about the perils and benefits of automation, and you’ve got one of 2022’s standout games.

To celebrate their Best of Quest win, we sat down with Pontoco’s John Austin, his fellow Game Director Matt Blair, and Art Director Anita Tung to talk about why The Last Clockwinder made sense in VR (and how that choice helped guide and reinforce the story’s themes), the tension between automation and individualism, and where the team sees VR heading in the near-ish future. Read on for more behind-the-scenes details, as well as some beautiful concept art created during The Last Clockwinder’s development.

John Austin: Ever since I can remember, I wanted to be a computer programmer. It strikes me as a bit odd now that I was so confident, but I feel like there’s something very attractive about programming as a kid. Everything else available to you at that age is very fake and sugar-coated! Programming on the early internet was one of the only things that was truly real—here are a bunch of people who can do incredible, actually useful things using just their own brains. Anything you can dream up you can make! Also, on the internet nobody knows you’re a 10-year-old! [Laughing] I actually remember taking great joy in hiding my age on early internet forums.

Eventually I received a copy of Macromedia Flash, taught myself Actionscript, and made a couple of flash games. I suppose that was my first step into programming. After that, I just never really stopped. I went to college, took a tour through big tech working at Google Brain, but all the while I was making games in some shape or form.

In college I met Matt Blair and Crystal Ngai, the other co-founders of Pontoco. We made several games together, one of which became Gathering Sky. I took my final six months of school off to finish that game full-time. I didn’t quite have enough runway, so I ended up running our launch operations on my laptop during my employee orientation at Google! We’ve come a long way since then, haha.

Matt Blair: I was getting my degree in mechanical engineering and a friend of mine convinced me to take an “Intro to Game Design” course for fun. I liked programming and I had spent an embarrassingly large fraction of my childhood glued in front of PlayStations so I thought “Sure, sounds fun.” Two years later I dropped out of an engineering PhD to pursue game development as a career. That friend of mine utterly destroyed my mechanical engineering career and for that I am forever in his debt.

John Austin: The original concept came up when we were tossing around ideas for a knitting game. Knitting is essentially doing repeated actions on a loop, and that immediately led to this idea of making clones of your own actions. We liked this idea of playing with short loops of just a few seconds, and seeing all of the clones working together in harmony. We didn’t end up continuing with the knitting framing, but cloning was just too good of a gameplay mechanic to pass up.

We had a very long prototyping period (nearly a year and a half). At first we imagined the game as a fairly standard puzzle experience, like Portal, but as we built it, we realized it was much more interesting to design creative challenges: loose puzzles that could be solved in any number of ways, rather than levels that had a single solution.

As an engineer, I’ve always had a soft spot for Zachtronics’ programming games. They can be very unapproachable, but they inspire a rich form of player creativity that you just don’t see anywhere else. So we drew a lot of inspiration from that genre, and used our cloning-your-actions mechanic as a way to make the “programming” aspect more approachable. In private, I’ve always considered this to be a “hidden Zachtronics game.” It doesn’t necessarily show on the surface, but it’s there deep at the core.

John Austin: This isn’t my area, so I’ll pass this off to Anita, who art directed the game!

Anita Tung: We knew early on that the whole game would take place in one main environment, with different platforms that swap in and out of the same space. So knowing this, it became really important for that environment to be a pleasant place to spend a lot of time! It had to feel bright and open, but also cozy and livable. The right balance here seemed to be a room with lots of natural light and windows, but with a few good reading nooks. Most of the design after that was just guided by what I would personally find relaxing and homey: books, plants, pillows, rugs, tapestries, and lots of natural materials like wood and plaster. Alongside the actual content of the environment, we wanted to push things stylistically to feel as much like a painting as possible. For us that meant using painted textures, focusing on the overall light and color, and using some post processing volumes to tweak the final look.

In terms of artistic influence, a lot of people have rightfully pointed out the Studio Ghibli similarities—that’s a major one! Another one that I was looking to for inspiration on the mechanical assets is Moebius, he’s got such an interesting sensibility around hard surface designs. A lot of the technology in his comics has this odd mixture of organic shapes with more traditional geometry that I wanted to emulate.

Anita Tung: Visually, I wanted to create a sense of harmony between technology and nature. There’s a theme running through the game of the benefits and drawbacks of automation, how some tasks need human involvement, and how desires for mechanical perfection don’t always work out well. But we also want to strike a balance there—the game is definitely not anti-automation! So it was important for the mechanical elements to have a sense of cohesion with the more natural forms of the plants. When designing the technology in the environment, I’d imagined three somewhat distinct “generations” of technology. The oldest and most arcane is very integrated with the tree itself, and consists of a lot of organic vine-like wiring that flows in and out of the trunk itself.

John Austin: Unlike many VR studios, we’re not a VR-exclusive studio. We pick the platform that best fits the game. And after we hit on this idea, we *knew* it had to be a VR game. Being able to see your own subtle gestures reflected in the clones was simply the best possible instantiation of this mechanic.

Much of the story comes from my experience as an engineer. As we approach a world where nearly everything can be automated, the question is “Should we?” What is the value of having humans in the loop? I think VR is a very interesting spot to be asking this question. It’s moving so quickly—sometimes it’s important to take a step back and ask “Why are we doing this?” and “Where does this lead us?” I think that good games can push us to think about these kinds of questions.

Matt Blair: Over time, the medium of VR really influenced the themes of Clockwinder. Once we were able to see the clones acting out all the subtleties of our movements and behavior, we got really interested in the tension between mechanical perfection of the robots and the human imperfection of the player that they mimic. That in turn guided the story towards exploring perfectionism and self-criticism. I feel that the game and the medium are truly inseparable in this case.

John Austin: VR is definitely having a bit of a moment to be sure. The Quest 2 took VR mainstream in a way that no other device had really been able to.

The challenge, of course, is how we can improve from here. There are some incredible improvements to fidelity (wider FOV, high-lux displays, higher refresh rates) coming soon. How do we manage to fit these new expensive features on a standalone device that’s already pushing the limits of what’s possible with a mobile chipset?

As we run into these fundamental computational limits, the only real answer is to just “do less work”. Machine Learning can absolutely enable that, and I expect it to be critical to the future of VR. Many of us are familiar with ML upscaling like Nvidia’s DLSS, but that’s really just the start. Over the next 5-10 years I would expect to see ML become much more integral in the way games are built. To the point where most of the rendering stack may be “hallucinated” with ML, rather than rendered directly. It will be a wild new world!

Matt Blair: From what I’ve seen over the last few years, the only thing I can say with certainty about the coming years of VR and AR is that nobody knows what it looks like yet. The experiences that people are drawn to the most strongly are the ones that aren’t possible outside of VR, the ones that don’t have a direct parallel in other media. We see this with VRChat, Beat Saber, Gorilla Tag, and to a smaller extent with The Last Clockwinder too. So my cheeky prediction is that whatever the next big thing in VR is, it will be totally unpredictable.

John Austin: As a creator, the response has really been incredible. People are solving the puzzles in ways we never would have dreamed up ourselves. I’m particularly fond of the one person that created a shovel using our bonding mechanic.

John Austin: It was incredible, and a bit overwhelming! There were some incredible games released this year, and I’m absolutely honored to have been selected. The nature of the award meant Meta wasn’t able to tell us until the morning it was announced. We found out when we opened the storefront that morning, and had to do a double take. “Is that us?”

John Austin: It’s been really great. The Cyan folks have more than 30 years of experience launching puzzle narrative games, and being able to tap into that knowledge and their community has been fantastic. Plus they were early adopters for VR as well, so it was essentially just a perfect match. One of the major benefits is that Cyan is also a developer, so we can come to them with all sorts of arcane technical challenges and they understand what we’re talking about.

Doug (our contact at Robot Teddy), Hannah (Cyan), and I had known each other for a little while before we started this project. In a lot of ways this weird dual-publisher partnership was more a matter of “Hey wouldn’t it be great to do something together?” Of course, the game had to be good, but once we showed them our final build, it really just all fell into place.

John Austin: Make beautiful games!

Congratulations to The Last Clockwinder and Pontoco for the Best of Quest 2022 win! When you’re ready to start cloning yourself, you can find it on Meta Quest 2 + Pro for $24.99 USD.