Interview: The Dream Machine’s Anders Gustafsson

Since 2010, Anders Gustafsson and Erik Zaring have been releasing their claymation point-and-click adventure The Dream Machine in installments. The game, with visuals built by hand using cardboard and clay, guides us through the world of newlywed Victor Neff, who solves puzzles and collects items as he overcomes obstacles. Having recently moved to a new apartment with his wife, Victor sets out on a voyage through several human minds, stumbling upon the secrets of his new home. With the sixth chapter due out on May 11, NerdQ’s Erik Meyer talked shop with Anders.

Erik Meyer: You’re using physical materials, animating with Maya, building in Flash, and creating a unique aesthetic in the process; the website relates the project to dreams and voyeurism. How did you arrive at this particular Anders Gustafssonway of working and creating an experience for players? Why these particular tools and methods?

Anders Gustafsson: Both me and Erik [Zaring] have backgrounds in traditional animation. I’ve worked on animated television series for kids, and Erik used to run a stop motion animation studio. When we started talking about doing a game together, Erik suggested we’d do it as a traditional stop motion project. I resisted the idea, because I thought it would take too much time, but he built four or five sets overnight just to prove me wrong. When I saw those, there was no turning back.

But doing the whole thing in strictly stop motion would’ve been too crazy, so we ended up with a compromise where some things are animated digitally, like the characters, etc. But the overall ambition is to do as many things as possible directly in camera.

Once the project received some funding, we needed to produce a proof-of-concept prototype very quickly. By that point, I’d already done several adventure games in Flash, so I already had an adventure game engine up and running, and we just used that.

If I started the project today, I would make different decisions, but them’s the breaks.

EM:  We’re coming up on The Dream Machine’s sixth installment; as you look back, what have been the biggest implementation challenges, and what are you proudest of?

AG: I’m a shitty programmer. So that’s been a bit of a challenge to overcome. I’ve become better, sure, but there’s so many ideas I’ve just had to table because implementing them would take too long for me. Luckily, I’m both the programmer and designer on this project, so I can always tell the designer to come up with something that’s just as good but less technically challenging.

I don’t think a man’s reach should necessarily extend beyond his grasp. It’s easy to get caught up chasing sexy tech, but sometimes it’s better to just settle for the low-hanging fruit and call it a day. If the end result is good, who cares how it was achieved?

Am I right, people? Can I get an amen?

The thing I’m most proud over is that it looks very likely that this project will get completed. I’ve been working on this game for so long now that for a moment, completion started to sound like an abstract theoretical construct. But the deadline is less than a month away, and I can’t see what would jeopardize it substantially.

EM: Unlike games that result from a weekend game jam, The Dream Machine has been released over many years. Describe the effect that an elongated passage of time has had on the project.

AG: When you live with a project over so many years, you start injecting a lot of your own life into it. You really get to know me and Erik when you play. It’s our fears and fascinations put on display. There’s parts of it that almost feel like a diary.

Editing the game becomes an ongoing part of your life. I tend to get obsessive about what I’m working on, so to get to keep massaging already-completed content is a bit of a dream come true.

Your social life takes a bit of a hit. I’ve been answering the question “So, what are you up to nowadays?” in exactly the same way now for eight years running. The eyes of my best friends immediately glaze over when somebody asks me about the game. They’ve heard it all a million times at this point.

You start trusting the process. There’s really not much you can do about it. At this point, my will is subservient to the game’s will. I just execute and try not to mess it up.

EM: What makes a two person studio (be it relating to games or animation) different from a larger development company, in terms of process? What does your typical day look like? (If you have such a thing…)

AG: As a small studio, you have to wear of lot of different hats. I do everything from being the director of the whole project to being the tech support guy for our players. I try to compartmentalize as best I can. I usually do technical stuff in the morning when my brain is still fresh and sparkling. Then I put my design hat on and work out a puzzle or two. In the afternoon, I try to do things like writing. I like to do that outside if possible (I live in Sweden), preferably in the park with a wad of sticky notes and some cans of weak, shitty beer (what we in Sweden call Folköl). It allows me to sip on something without getting drunk.

When I get home, I pin the sticky notes to a wall in my apartment. Once the wall has been filled to capacity, I go through all the notes and throw away the ones I don’t like. Whenever I find myself confused as to what I should be doing, I consult the wall.

Sometimes, I leave small offerings in front of it.

EM: Weigh in on the decision to publish in an episodic, chapter-release way. Many games have downloadable content or serialize main storylines, but how do you feel this format resonates with players and complements/reinforces your ability to produce quality work?

AG: In our case, episodic was the only available option at the time. We wanted to do a very ambitious, long format game but didn’t have the upfront money to do it. So we divided the story into chunks and used the earnings of the previous episode to finance production of the upcoming one. That’s obviously quite risky as you might end up completing an episode with not enough money to do the next.

To mitigate that risk, we started selling the whole damn game right off the bat (which ended up working out), but it also meant that our lives were taken hostage by our own unfinished game. Once you start selling the whole thing, you better complete it.

There’s been moments when I wished we hadn’t done it this way, but I also attribute this situation as the single most important factor as to why we’ve managed to get this far. Once we started pre-selling episodes, we simply didn’t have much choice.

Do I recommend episodic development? It depends on the game. For a story-based game, having year-long breaks in between episodes is not ideal. No question about that.

On the positive side, you get to build a community over a longer period of time. You can ongoingly tweak your project.

On the negative side, it’s harder to get attention from press for an episodic project. They can write about the first and last episode, but drumming up interest for the middle of the game is very hard. If you limit the game to two episodes, that might work in your favor. Three episodes is pushing it. Six episodes is not ideal.

And unfortunately, there’s still a slightly negative connotation to episodic games. People tend to be more skeptical of episodic games, since they’ve been burned in the past. Some of the first games to try the format never got past the first episode.

EM: You’ve mentioned previously that your process at Cockroach Inc. involves taking on other work to float the studio while working on The Dream Machine; in the last five years, alternative streams of revenue like crowdfunding, Steam sales, and indie bundles have arisen as options for indie developers. Out of curiosity, what do you think of these sales mechanisms and how they impact gaming?

AG: We don’t need to take on external work anymore. We do every once in a while, but it’s mostly because we believe in the project. Not so much to keep afloat. As you mention, there are more streams of revenue today, and finding an audience is easier. If you make something people appreciate, it’s easier to make a living off it today than 10 years ago. No doubt.

The new sources of funding are changing what types of projects people start, for both good and bad. With a platform like Steam that almost has a monopoly position in digital game sales, filtration is a huge problem. There’s so much on there, it’s hard to find the gems.

But who am I to complain, right? Steam is one of the strongest reasons as to why The Dream Machine is going to get completed. If they hadn’t offered us a slot, we’d be in trouble.

EM: What are your UI/UX criteria when designing puzzles? What things do you look to accomplish in player challenges, and how much do you leave to people to stumble upon?

AG: I go for clarity and brevity. I want the components of a puzzle to be as clear as possible.
The challenge should come from how you put them together.

It’s easy to create artificially difficult puzzles. You can just hide items and write vague clues, but that’s a phony way to create challenge. If the player needs a key to open a door, put the key as far away from the door as possible and BAM, you’ve made the puzzle more difficult. Does that make the puzzle more enjoyable, though?

As a puzzle designer, you’re always trying to evoke the AHA moment in the player, when they’ve percolated on the puzzle and suddenly manage to see the connections between the individual parts. Increasing the distance between two elements of the puzzle does nothing to enhance the AHA moment. It’s just an artificial way to make the puzzle harder to solve.

When you playtest an adventure game, it becomes very obvious that people go through a cycle of thought when they’re trying to solve a puzzle. They try to identify the components of the puzzle, they make a hypothesis of how the components fit together, and then they test the hypothesis. If they succeed, they move on. If they fail, they start the cycle over again, hopefully identifying more components and making more informed hypotheses.

As the game designer, it’s very important to give feedback when the player fails and has to start the cycle over. Otherwise, the game quickly becomes tedious. Since we partly deploy the game on the Internet, we can use metrics to see how people are playing, what item combinations people try, etc. Whenever we see enough players try a combination we hadn’t thought of, we simply add that to the game, sometimes as a valid solution to a puzzle but most often as an option to give feedback, nudging the player in the right direction.

Stuck players often resort to key chaining. That’s the brute force tactic where they try every inventory item with every object in a scene. It’s not a fun way to play an adventure game. It’s not rewarding. Even if you stumble upon the solution, it feels undeserved and arbitrary. When we detect that players are key chaining, we know it’s time to add a hint or rethink the puzzle.

The difference between a good hint, one that sets off a chain of new ideas in the player’s head, and an obvious hint, one that just gives the puzzle away outright, is often minuscule. Changing a word in a sentence can often be enough.

I like to decentralize hints as much as possible, scattering them randomly throughout the game and letting the player discover them piecemeal. I like to approach storytelling in games that way, as well.

EM: In dreams, we often easily accept things that would be unsettling or jarring in real life; as you’ve charted transitions in Victor’s mindscape, how have you kept a sense of consistency (the right kind of feel) between otherwise outlandish locations and experiences?

AG: Being only two guys kind of solves the consistency problem by default. Even though our game takes place in various outlandish dreams, there is a certain sense of oddball cohesiveness to the entire thing. At the end of the day, everything was created by only two guys, and I think the players can feel that. There’s a maker’s mark to everything in the game. If we were a bigger team, one of us would have to do more supervision, and the cohesiveness would probably get diluted.

EM: Big projects are an endurance sport, and The Dream Machine represents a huge undertaking. What keeps you excited in gaming and developing, and as you look ahead, how will this project inform future project decisions?

AG: We’re trying to push the medium a little bit. Not the form, necessarily, but the content. The ideas we bring on stage are rarely seen in games. There are parts of the game – especially Chapter 6 – that I don’t think I’ve ever seen anywhere else. That’s exciting. I can’t wait to hear what people think about it.

Over the course of the game, we’ve gradually prepared people for the fact that the game can get pretty weird. When we put it in fifth gear, things get trippy. And in Chapter 6, all that preparation and foreshadowing hopefully reaches a suitable climax.

For future projects: Trying to create something this ambitious when you’re only two guys is insanely risky. Anything I do in the future will be more sensibly scoped. I hope. I don’t think we’ll ever get the chance to work on a project like this again, so we’re giving it everything we’ve got.


In case you missed it, here’s the announcement trailer for Chapter 6:

Author: Erik Meyer

Erik edits content, writes articles, conducts interviews, and draws silly things for The NerdQ. He also produces Planning Session, a comic showcasing dev discussions.

Leave a Reply

Your email address will not be published. Required fields are marked *