Designing world-spanning & momentum-based platforming in Sylvie Lime

This interview is part of our Road to the IGF series. The IGF (Independent Games Festival) aims to encourage innovation in game development and to recognize independent game developers advancing the medium. Every year, Game Developer sits down with the finalists for the IGF ahead of GDC to explore the themes, design decisions, and tools behind each entry.

Sylvie Lime is a platformer that asks you to play around with momentum, movement, and some wild abilities to get across its deadly hazards. And sometimes you're a lime.

Sylvie, the game's creator, took some time to speak with Game Developer about the Nuovo Award-nominated title and how the creative platformer was inspired by an oddball Famicom game and a box of citrus, the multi-game spanning ideas that would inform its varied cast, and how some unexpected chaos came out of adding some really mundane tools into the platforming title.

Who are you, and what was your role in developing Sylvie Lime?

I'm Sylvie and I developed Sylvie Lime together with my friend Aria. For the most part, it was my project, but Aria helped with some things related to the characters and story. The non-player characters were sort of co-designed and belong to both of us, and the setting of the story comes from one of Aria's works. I programmed and designed the game, created the graphics and audio, and wrote the dialogue.

What's your background in making games?

I have been making video games for around 20 years. I started out using a tool called Project FUN created by the DigiPen Institute of Technology. It's extremely difficult to find any information about this tool nowadays, since DigiPen reused the "Project FUN" name for something else, but there is an article about it (from the August 2002 issue of Nintendo Power) available on the Wayback Machine.

Since this tool became available sometime around August 2002, that means I probably started making games in late 2002 or early 2003. I was only 11 or 12 years old and I found Project FUN kind of hard to use, but I managed to create a weird, simple game with it. Unfortunately, when I burned it to a CD, it didn't run on my friend's computer. I created about 0.5 more games with Project FUN before I discovered GameMaker in mid-2003. It was a lot easier for me to use, and it could make games that ran on other people's computers. Now I've created 120+ games mostly using different versions of GameMaker (but sometimes other tools like Bitsy).

How did you come up with the concept for Sylvie Lime?

Earlier last year, I lived in an apartment where the kitchen and common area were shared with four other people. I went into this area and there was a box of limes on the couch. That's what gave me the idea of Sylvie turning into a lime. That's how I remember it, but when I think about this story, it seems kind of unbelievable. I don't think it's common for people to buy an entire box of limes. The box definitely said "limes" on it, but it might have just been an old box that was being used to store something else.

What development tools were used to build your game?

I used GameMaker (the current/modern version, which used to be called "GameMaker Studio 2" but is now just called GameMaker, which makes it confusing to tell people which version you mean). Some of the pixel art was done in the built-in GameMaker sprite editor, but I also used Aseprite for a lot of it, particularly for character art and larger sprites. The music was made with PxTone and most of the sound effects were made with LabChirp. Some were just recorded with my mic and edited with Audacity.

Sylvie Lime is a platformer that offers some silly mechanics that get players to think about platforming differently. What appealed to you about finding new ways to jump around dangerous spaces?

I think I could have made a more standard precision platformer using Sylvie Lime's base mechanics—something closer to Celeste where you use the lime to navigate through a series of tight obstacle courses. However, making exploration games with an open-ended, non-linear world design is a long-term interest of mine.

Shortly after I started making Sylvie Lime (when all I really had done was a test room for the lime mechanic), I started playing the Japan-only NES game Atlantis no Nazo. It's a game that is full of treasure chests that are largely useless, obscure invisible secrets, a web of branching paths, and strange items whose purpose is never explained. There is an item that lets you press Up on the D-Pad to increase your score by 3 points. Shortly after that, by coincidence, Atlantis no Nazo appeared on a pledge drive stream for the Indiepocalypse game anthology, and I commented in the stream chat that there should be more games like Atlantis no Nazo.

Shortly after that, Andrew (the Indiepocalypse creator/curator) asked to commission a game from me for a future Indiepocalypse issue, and I decided to use the commission as an opportunity to finish Sylvie Lime. It appeared in Issue 34. So, I think it was Atlantis no Nazo that convinced me to make Sylvie Lime an unusual open-ended exploration game instead of a linear challenge gauntlet. Sylvie Lime ultimately isn't that much like Atlantis no Nazo, but I wanted to create a game with the same kind of strange energy and magic.

Sylvie's lime power turns into a bouncy citrus that preserves momentum in neat ways. What thoughts went into designing this mechanic?

I can't remember when or why I started thinking about this, but the basic idea is that, in a game where you have both horizontal acceleration and variable-height jumping, your character can move at different "angles" by coordinating your jump arc and horizontal movement appropriately. Recreating different angles is a skill you can practice. I vaguely remember having an idea for a game where this "movement angle" is used to fire projectiles in different directions, but I don't think I ever made it... Maybe there was a game by someone else I played that made me think about "movement angles" in this way, but I can't remember.

Anyways, the main idea of the lime is that you can use this angle to launch yourself. One concern was that if the player can make very small and precise adjustments to their movement, then there are too many different possible angles, and it becomes difficult to recreate specific angles that might be needed for a cool trick. That's why the game runs at 24 FPS: to reduce the amount of precision available.

Also, what thoughts went into the design of the silly platforming tools (Sylvie's house, hammer, etc.)? What ideas went into creating these? How did you want them to shake up the possibilities for the game's platforming?

When coming up with the items, I wanted to have six "active" items which have an immediate effect when you press a button, six "passive" items which have a constant effect you can toggle on and off, and six "quest" items that are part of the game's story and don't really influence the platforming.

I basically tried to make each item in the "active" and "passive" groups have a very different purpose. One of the "passive" items (the gem) kind of feels like an active item in disguise, though. This is because its effect was changed late in development. The original idea was that turning on the gem would add a lyrical track to all the songs. But it was too hard to come up with lyrics for all the songs, so I had to think of another idea.

The ideas for items came from various places. For the house, I already made a game called JIGGLY ZONE where you can place a checkpoint at any time while you're on the ground. I wanted this game to have a similar amount of freedom with checkpoint placement, but I didn't want to just reuse the same system, and that led to the idea where you throw your checkpoint. I think it was also a little inspired by the PC88 game Dragon Slayer where the character has a house that you can push around the map.

The hammer was inspired by the hammer in the NES game Milon's Secret Castle, which reveals secret doors hidden in certain background tiles. The snowball serves as a basic projectile weapon, but due to the nature of one of the storyline quests, it ended up with a certain unique feature that was needed to make the quest work. Overall, I took a lot of inspirations from other games (including my own past games) but some aspects were just improvised based on what parts of Sylvie Lime I'd made so far.

Some of the items I didn't really think about that carefully. One of the most powerful items, the chair, was basically designed for a joke puzzle. There's a chest that you can't open because it's sitting on top of a fake block, and you can only open chests when you're standing on (non-fake) ground. The solution is to place the chair underneath the chest.

That was the original purpose of the chair, but after adding it, I realized it was actually really useful all over the game. I thought that was cool, but then playtesters discovered that the chair completely breaks about half of the boss fights, since you could just drop a chair on the bosses to lock them in place. I didn't like that, so I made the bosses interact with the chair and push it out of the way. You can still kind of cheese some bosses using the chair, which I think is good. I like having some cheese available, but being able to lock them in place and completely prevent them from attacking felt too extreme.

What drew you to show all of the collectible tools on the side of the screen? What effect do you think that had on the player?

I think this was a pretty arbitrary decision that spun off of the resolution I chose. For a while, I used 16:9 widescreen resolutions and I would pick a base resolution that evenly divides 1920x1080 because I had the impression that this was a very common resolution, and this would allow my game to scale up nicely and fill the screen completely for anyone who uses this resolution. However, now I have a 16:10 laptop and I don't care anymore, so I just choose whatever resolution I feel like.

For some reason, I went with a 4:3 resolution for Sylvie Lime, probably because I saw some person on Twitter say something like "4:3 is just more cozy!" and I couldn't get it out of my head. However, I think I was also into square games at the time, so I decided to do a 1:1 resolution for the game area and use the remaining space to the right for a HUD. This is probably a little inspired by old computer games that have a square play area and use the right side of the screen to display various information, like the aforementioned Dragon Slayer (you can see screenshots here).

Early on, I didn't know what was going to be on the HUD, and it could have been things like a health bar or a death counter or a playtime clock, but once I had a better idea of the game's direction, it made sense to just show the items you've collected there. I actually decided to show the items on the HUD before I came up with the system where each item is tied to one letter key. You may have thought it was the other way around—that I listed the items on the HUD so people wouldn't forget what letter key they're tied to—but that's not the case.

What happened was that I programmed the system where the items get displayed on the HUD, and then my next task was to implement some kind of "item select" feature that lets you choose which item to use with the action button, toggle items which have passive effects, examine quest items, etc. But this seemed really annoying and I didn't want to do it, so I just assigned every item to a letter key.

I didn't think that much about what effect showing the items had on the player, since showing all the items on the right seemed normal to me, but I guess it's actually a little strange these days to not have a separate "inventory screen." I've noticed that players of Sylvie Lime often forget what tools they have available, because there are so many options. I feel like this would be an even bigger problem if you had to pause and go to a separate screen to look at what items you have.

Also, while it's somewhat awkward to remember the correct key corresponding to each item, I think if you had to "equip" items from your inventory, it might actually be even more difficult to use items freely and to combine the powers of items. I guess this arbitrary decision is surprisingly important to the game's design and feel.

With so many possible platforming mechanics and tools, what thoughts went into weaving platforming puzzles into the environments to make players put the tools to work?

I designed almost the whole game around the lime power. When I designed the world, most of the items weren't actually implemented yet. I had some ideas for items, and this influenced a few of the screens where I designed something that's difficult to do with the lime but would be easy with a particular item. Once the items were added, I played the game and noticed a few areas where a certain item made it "too easy," so I adjusted the levels a bit. But for the most part, I didn't consider the tools other than the lime when designing the levels.

Many of your puzzles reach across several screens. What thoughts go into creating platforming puzzles that spread far across the world? In getting the player to think beyond just the screen they're on?

One thing that helped is that the whole world is contained in one "room" in GameMaker terms, which means I can edit the entire world at once—it's not divided into separate parts. I also created an overlay for myself that shows the screen boundaries so that I can see how each screen connects and design multi-screen scenarios.

Maybe I'm forgetting something, but I can only really remember one kind of multi-screen puzzle in the game, which is to launch your lime at a certain position and angle (often a flat horizontal angle) so that it travels through several screens and ends up at a new location. There are some areas where it's more obvious you can do this than others, so I think I wanted the player to get used to this idea and then start thinking about whether they can use it to make progress in other places.

Likewise, some of the puzzles extend into the ground in ways that feel like you're doing things the game doesn't intend (even though they're clearly supposed to be possible). How did you create that feeling of finding secrets and doing things you weren't supposed to be capable of?

I think this is a direct consequence of designing most of the levels around the lime power rather than creating scenarios where you're expected to use a specific item. The "unintended" feeling probably comes from the fact that people can see how, in theory, they could get past the screen with some difficult and precise lime moves, but instead they use items to make it significantly easier.

There are also some areas where it's not really obvious how to do it with the lime alone, and there are multiple alternative solutions involving creative item use. When a player finds one of these "creative item use" solutions, it probably feels "unintended" (even if it's something I predicted and allowed for) because there isn't really a sense that the area was designed to hint towards that solution. With some rare exceptions, the level design generally does not "guide the player" towards solutions involving items because everything was just designed around the lime.

At the same time, clearing the game lime-only (or even lime-and-house-only) is obviously really difficult, so avoiding items doesn't feel "intended" either.

Part of the driving force of the game came from meeting people and helping them. What ideas went into creating fun, charming people to meet and giving them interesting things to need help with?

Well, the characters in this game have a strange and long history because they weren't originally created for Sylvie Lime at all. The initial versions of the characters were created for a detective/mystery game that me and Aria were thinking about but never made much progress in developing. We both really like mystery stories but don't know how to write one.

As practice, we thought about creating a mini-mystery (or maybe a series!) that focuses on just five characters who all follow distinct archetypes: a mature, perceptive, no-nonsense detective; a cool, mysterious girl who likes teasing people and seems to know more than she lets on; a sweet and gentle girl who's easy to manipulate; her combative friend who'd do anything to protect her; and a confusing girl who seems to act with the sole purpose of sowing chaos and discord. These archetypes were basically developed with the goal of making interesting characters for mysteries. Their personalities clash in various fun ways, and except maybe for the detective, it's easy to come up with motives for why they would commit or assist with a crime.

These archetypes are a bit different from the characters that ended up in Sylvie Lime. Since the original mystery game will probably never be made, I can reveal one of the plot twists. It turns out that the characters are actually part of a video game world, and each one is being controlled by a real person. So, there was going to be a second set of characters who would be the "players" of the characters from the first set. It's this second set which evolved into Sylvie Lime's NPCs. Basically, Sylvie Lime's characters were born from thinking about what kind of person would roleplay each of the characters from the mystery game.

Now, the conceit of the story in Sylvie Lime is that the five NPCs you meet are part of a computer club, and together, this club developed the game "Sylvie Lime" and they inserted themselves into the game as NPCs so they can speak to the player.

So, this introduced another twist. Now instead of thinking of the characters as people who roleplay as criminals online, I had to think about what they'd be like as video game developers. So, of course, I drew on a lot of my own experiences. A lot of the things the characters say are just true statements about what I did when making the game, and a lot of the design philosophies the characters express correspond to my own opinions. Some of the "arguments" between characters represent things I felt some internal conflict about.

At the same time, the characters are not just mouthpieces for me or a representation of my inner monologue. Because the characters started with their own basic personality, a lot of them ended up saying things that are counter to my opinions, or behaving in ways that I wouldn't behave. Some characters are like me, but some are more like a person I would look up to or a person who I used to be like in the past.

I guess this is a long way of saying that my goal was not really to create fun, charming people to meet and give them interesting things to need help with, or at least not in an explicit way. For the mystery game, I approached it by trying to think of characters who would have interesting interactions with each other, as opposed to the player, and also could be credibly accused of crimes. When those characters evolved into the Sylvie Lime characters, each character's personality became influenced by my own in a complex fashion.

The heart-pounding romantic events that happen in Sylvie Lime are kind of a combination of half-baked plot ideas from the abandoned mystery game, and my personal feelings about love and experiences with love. Ultimately, I think I was just relying on the hope that I am fun and charming and there are interesting things I need help with.