2019年8月20日
Solo dev Gwen Frey explains how she developed puzzle game Kine using only Blueprints
To see how Frey nearly single handedly developed the game with virtually no coding experience, we interviewed the solo dev. In our discussion, she talks about the challenges of designing a game for the first time, details some of the skills she had to pick up along the way, elaborates on how she leaned on Kine’s narrative to create puzzles, and talks about how she created the game’s sketchy hand-drawn comic artstyle.
Towards the end of the interview, Frey reflects on lessons learned, shares her favorite UE4 tools, and provides advice to aspiring solo developers.
As someone who has stepped away from working for major game studios to become a solo indie developer, can you provide us a high-level overview of your background in the industry?
Gwen Frey: Sure! I got my first gig working as a tech artist on an unannounced MMO. After that studio closed, I joined a brand new start up where I was one of five employees. I had a blast helping to grow that studio into Secret Identity Studios and doing the early development for Marvel Heroes Online. After two years there, I left to join Irrational Games. I worked at Irrational for years developing BioShock Infinite and various DLCs until they eventually shut down.
After that, I co-founded The Molasses Flood, where I was the CFO and one of six developers. I helped launch our flagship title The Flame in The Flood as the sole animator, tech artist, and generalist. Finally, I left The Molasses Flood to establish my own solo studio so that I could develop Kine.
Kine features a novel narrative puzzle concept with a unique music thematic. How did you come up with it?
Frey: In the beginning, I wasn’t making a game at all. I was experimenting with an animation idea where a character would somersault around in the world and move by kicking off of walls. In order to prototype this system, I was using a cube. But I just absolutely fell in love with how the cube moved! We were pitching different game ideas at The Molasses Flood for our next project, so I made a small prototype where you were this cube thing that pushed off of walls and rolled through obstacles like some kind of weird 3D tetris.
Were there any particular games that inspired you to make Kine?
Frey: Yes! The first prototype of Kine was clunky and not fun. Keep in mind I’ve never been a designer and I never really played puzzle games when I was making this. Kine was just a tech demo and I abandoned it to work on other things. While Kine was on the back burner, I played a game called Stephan’s Sausage Roll (SSR), and fell in love with it. I realized this was how I could make the tech demo I had crafted into a game. It was immediately clear what I was doing wrong - the first version of Kine had massive rooms that you’d move around, whereas SSR had these very small puzzles (relatively) where you had to plan ahead and carefully consider your movements. I loved that game and many months later, when I went back to make Kine again, that game was still very much on my mind.
Kine features a really cool 3D comic book art style. How did you come up with that aesthetic?
Frey: I personally love sketchy, drawn artwork, but I mostly made the game in this style because I knew it would best service the game design. Computer screens are 2D and it is very difficult for people to sense where they are on a 3D grid. With my first blocky-prototype, I was having a hard time judging distances in 3D space – I couldn’t tell if something was up one unit and back one unit or simply back three units. I work in 3D viewports for a living - if I was struggling then other gamers would certainly struggle as well! So I tried using a hand-drawn style. This way I could author lines to overhang the edges of objects and more clearly indicate gridlines over edges and gaps. I personally loved the idea of having a sketchy hand-drawn world, and I probably would have crafted the game in a similar art style to this no matter what. However, the design needs made it mandatory.
In lieu of complex textures, Kine uses simple solid colors. Can you speak to this design decision?
Frey: Some part of this was necessity. I started this game intending to finish it solo and I am not great at texturing. I’m really good at messing with materials and I knew I could make the materials seem vibrant and dynamic without a lot of texture work, so I leaned into that. Also, the Switch had a popular launch two years ago when I was coming up with this art style. I wanted to develop a look that would be beautiful on a 4K monitor, but just as readable and beautiful on a smaller screen. I figured using solid colors or gradients was the key to making that work.
With a background working as an animator, can you delve into how you created the impressive animations in Kine?
Frey: Ha! My background as an animator is probably wasted on this game. There are no complicated characters – the characters are cubes with faces. The three characters have two facial expressions and one-to-two idles each. I was working a full-time job as an animator when I started Kine and the last thing I wanted to do was come home and continue animating. I kept the character moveset very simple to help me stay interested and motivated on the project.
However, as I programmed things, I couldn’t help but use curves to dictate how the characters roll around and I couldn’t help but tweak the ease-ins and outs from different states. I don’t know if you can call that animation, but the characters do move around the world in a jaunty, playful way. None of that was keyframed animation, that is just coded into how they move. I also had a lot of fun adding motion and flourishes to the background elements in the world. Also, in the end, I made the levels that you aren’t on fly up and out of the way and fall back down with a squash. So I guess my background does show through in this game. I couldn’t help it!
The later puzzles in Kine really test one's mental fortitude. How did you conceptualize and develop the game's various puzzles?
Frey: I started out just playing with how the characters move. I tried to come up with puzzles that forced you to move the character in a novel way. I wanted you to have to think and figure out how to get from place to place, rather than “lucking” into it. Most of the middle puzzles in the game were about exploring this. As I got used to how to design puzzles for Kine, I started branching out a bit more for inspiration.
There is one part of the game where two characters are working a day job that they don’t want to do. One character, Quat, desperately wants to quit and go back to making music full time. I made these levels while I was still working a day job and it reflects how I felt at the time. When I first designed that area, each puzzle was the same as the last one with a very slight variation. This ended up being a difficult area to polish later on because I want my players to have fun, but I’m trying to portray monotony and boredom in the puzzle design. Those two ideas are at odds!
I also tried leaning into the story for inspiration on most of the puzzles. I would think things like, “These characters are in love. They should go dancing. What would it look like if they were dancing?” And I’d try to make a puzzle that felt like a choreographed dance. For one set of puzzles, I was watching pairs figure skating and I tried to capture that feeling in the motion of the characters. For a different set of puzzles, I was thinking, “This character is injured. How can I design a puzzle where you feel helpless? Can I make a puzzle where you are forced to limp?” And I explored that idea. I think the puzzles where I was explicitly trying to tell a story are the best ones in the game. I enjoyed that experience immensely.
How much coding experience did you have going into the project?
Frey: None really. I have always been a bit technical and I’ve always tinkered with things, but I’ve never had any schooling or training in programming. I taught myself a bit of Python when I was in animation school so that I could craft my own rigs, and I developed that into a tech art career. I have a lot of experience making Maya tools in Python for different art teams. I also taught myself C# at one point to build a small standalone audio tool early in my career. I honestly don’t remember anything about C# anymore, it was just a means to an end. I have had to pick up other scripting languages over the years (I built animation graphs at Irrational Games in Morpheme using Lua, scripted our early prototypes at Secret Identity in UnrealScript, etc.) but that was all scripting languages and it was always to achieve some goal.
Is it true that the entire game was made using Blueprints?
Frey: Yes. I don’t have Visual Studio installed on my computer. I don’t know C++ and probably couldn’t edit it if I wanted to. A month from now when Kine ships, the PC version will be 100 percent Blueprints.
Can you share any tips on how to learn Blueprints?
Frey: I have always learned best when I had a goal. Think of a thing you want to do and then try to do it. If you can’t do it or don’t even know where to start then Google it. There are thousands of people using Unreal and there are tons of resources out there. It is mind-blowing just how much information you have access to these days. You just need to have a goal and the motivation to reach for it.
As your first major solo project, were there any particular skills you felt like you needed to pick up or sharpen throughout the course of Kine's development?
Frey: There were tons of things I needed to learn. I don’t naturally put myself out there or promote my work very well. That is something I’ve been practicing over the last few years with mixed results, and it is something that is very important for a solo developer. Also, I taught myself how to be a designer the stupid way. I think the puzzle designs worked out well, but I really, really spun my wheels figuring out what the world map and UI should look like. In the end, I arrived at something pretty simple, but it is embarrassing how long it took me to get there. Another skill I probably needed to pick up and sharpen was basic programming concepts. I’m pretty ashamed at how cobbled together the undo system is under the hood. For the next game, I’m going to do a little more research on how to properly save and undo a game state.
Also, this year I had to transition from being a solo developer to directing a few outsourcers – I was funded and wanted to polish up the game’s environment art and music. That was surprisingly painful and difficult! When you plan to make a game solo and you work on it for years, it can be challenging to give up any creative freedom. I have always collaborated very well with others in the past, but actually directing someone to achieve my personal vision was a new experience. For example, I generally have an easier time showing an artist what I want by doing it… but that is time-consuming and people don’t really want you doing their work for them. Directing audio was insanely difficult because there were often times I knew what I didn’t want but I wasn’t really sure what I wanted and I had no idea what words to use to explain what I was looking for. Fortunately, everyone I worked with was super understanding and lovely to work with. I am very lucky to have worked with the professionals I did and would like to thank Mitchel Wong, Rick Nath, and my buddies at Surface Digital. Y'all rock!
What has been the biggest thing you've learned while developing the game?
Frey: Patience and persistence matter far more than skill and resources do.
What has been the biggest challenge making the game and how did you overcome it?
Frey: Honestly, the biggest challenge wasn’t anything technical – those challenges are terrifying sometimes, but they come and go. The business stuff wasn’t too challenging either – this was my second time starting a business. The biggest challenge was getting myself into a position where I could work on Kine full time. Early on, I was burning myself out trying to work a job and developing Kine at the same time. When I started development, I was still working at The Molasses Flood, and we were very close to closing the studio. I told myself that after the studio shut down, I’d take a bit of time and make Kine as a solo project before getting my next “real” job. When things started going really well at The Molasses Flood and it looked like we weren’t going to fold the studio anymore, I was surprisingly sad! I was supposed to be happy that we were going to be stable and funded, but in my head, I had already planned out how I’d develop Kine. I really, really wanted to do it! Leaving the studio that I helped [establish] right when things were starting to go well seemed crazy and stupid. Also, I didn’t want to betray my friends and business partners by leaving to do my own thing. In the end, though, I talked with my co-founders and we came to an arrangement. I left on good terms, but it was one of the hardest things I’ve ever had to do. I still love those guys and I really hope their next game is a massive success!
How long did Kine take to make?
Frey: I started making the prototype after work on nights and weekends. I did that for around four months and then set this project aside for about nine months. At one point, we went down to working part time (20 hours a week) at The Molasses Flood. I started working part time at TMF and part time on Kine for a year. Then I officially quit and started working on Kine full time in September of 2018. So that’s one-year full time, one-year part time, and several months of nights and weekends.
What made UE4 a good fit for the game?
Frey: There is no other engine I could have used to make Kine. Dammit, Scott, I’m an animator not a programmer! And I can’t code my own game from scratch.
Do you have any favorite UE4 tools or features?
Frey: My favorite parts of UE4 are the Material Editor and the animation graph. These systems are far more sophisticated in Unreal than any other engine I’ve seen. This is where Unreal Engine really shines.
Did you leverage the UE4 Marketplace at all?
Frey: Yes. I needed a cartoony hit effect for one specific level in the game. This was only going to be seen a few times and I didn’t want to spend too much time coming up with something. It was easier for me to just pay $10 and get a splat effect off the Marketplace than craft something myself.
Do you have any advice for aspiring solo developers?
Frey: It is far easier to scale up something simple than to scale down something that is wildly out of scope. Start off making a game that you think you can finish in six months. If you find a part of your game deserves to be explored further then start exploring it, and scope your exploration so that you will ship the new version of the game six months from that point. Do this repeatedly until you seriously can’t afford to do this anymore, or you run out of things that need exploring. At that point, force yourself to ship in six months. You should never start a project intending for it to take many, many years. Successful projects that take many years are usually started with the intention of shipping much sooner than that.
Thanks for your time. Where can people learn more about Kine?
Frey: You can pre-order it right now on the Epic Games Store! Also please follow me on Twitter @diregoldfish. I can’t say what comes next, but I can say that Kine is only the beginning!