Stephen Danton talked about the 2D side-scroller he is developing together with his wife Sara Kitamura: combat system, world development, ambiance, VFX, and more.
At the university, I studied Computer Science with a focus on game design and human-computer interaction. After school, I joined Microsoft where I was lucky to work in teams that were doing “v1” or next-gen projects. That shaped me to become someone who focuses on crafting experiences rather than visuals or specific interactions.
At Microsoft, it was all about tools or software that made people smarter and more capable. With games, it’s about systems and mechanics that allow people to feel masterful, brave, accomplished or even alone and scared.
At the end of the day I see myself as an experience designer rather than an interface or game designer, and I think that allows me to tackle such projects as software tools or video games in similar ways.
Sara has a background in painting, photography, color theory, and interior design. She’s a natural artist and has great instincts. She has a hands-on approach and likes to ensure we invest time evaluating concepts in-game rather than on paper. Sara has an excellent eye, especially when it comes to the overall composition and level of detail. She’s also masterful when it comes to editing a scene or noticing when subtle things are off in something like a character's movement.
We share so many of the gamedev tasks, collaborating on every aspect of Unto The End. We’re super lucky that we have very complementary skills and both like wearing a bunch of different hats.
Start of 2 Ton Studios
Our studio is small, just Sara and I, a husband-and-wife team. We’re Canadian, grew up outside of Toronto and met at University. After school, we moved to Seattle - I joined Microsoft while Sara did non-profit work and then opened an interior design business.
After leaving our careers, we traveled for a few years in Europe and South America. Most of that time was spent learning as much as we could about gamedev, trying out lots of ideas and showing them to people to help make them better.
One of those ideas was the concept project North. It’s always fun to look back at old stuff to see how you’ve grown in terms of design, art or animation (especially animation!). Those first tries at Unto The End’s art style and game feel were definitely born in North.
Despite hundreds of side-on games, we felt there was still more to be “juiced” from a 2D side-on presentation. We wanted to avoid anything derivative, like Dark Souls copy-pasted to 2D. Also no homages to old favorite games. Everything should be there for a reason, not just because another game did it.
Speaking of the combat system, our goal was to capture the feeling of melee combat (its intense, visceral nature, balanced with the need to keep your nerve), without being a sword fighting simulator or a fighting game with elaborate combos and intricate inputs.
We realized that 2D lets the player do a few things better than 3D:
- Judge the high-low height of a threat relative to your avatar
- Assess left-right distance between two objects
- See behind your avatar
We started prototyping in Unity and kept our prototypes simple. At first, there were no AI-controlled opponents, just Sara and me fighting each other in a player1 vs player2 setup. The very first prototypes didn’t even have two controllers, just Sara on the keyboard and me on the controller.
We started with the most basic actions (move, attack, block) and simple animations in a classic 1v1 fighting game setup. We played it over and over, tuning things until they started feeling right. During each step, we’d break things down and ask questions:
Should guard be timed? Should there be guard and parry? After you guard what should happen? Should your guard stay or do you need to re-guard? Should movement be fluid or should we put players in a “fight mode”? How fast should attacks be? Should attacks be cancelable? Can attacks be interrupted? What happens when you miss an attack? So on and so on.
We removed anything that felt too rigid or complex. Everything that stayed, fit into our goal of delivering the feel of combat, rather than a simulation. Players' skill and their ability to read and react in the right moment was key.
Then we’d add new actions for depth and only kept those actions that had distinct reasons for being there. Nothing is there just because it’s “cool”.
From there, we started adding AI-controlled opponents. We invested a ton into them. Every opponent is intelligent, there are no trash enemies. Opponents will tactically surround the main character (father), use range weapons if they have them, use terrain, lure players into traps, set up ambushes, etc. Opponents can go anywhere the father can, they’ll jump gaps, drop into caves, climb ledges, use ramps and so on. They also follow the same rules as the father, blocking high or low, attacking high or low, dealing the same amount of damage with similar strikes, being vulnerable to friendly fire, etc.
The world comes entirely from our imagination, mostly inspired by the time Sara and I spent in Iceland, Scotland, Patagonia (Southern Chile), and Canada, where we grew up. We wanted it to feel grounded, yet it’s a mythical place, not bound by history or established fiction. There is no magic in the game and we stayed away from traditional fantasy tropes like wizards, orcs, elves, dragons, etc.
We like to use side paths and loops with “ah-ha” moments at intersections, where you emerge from a cave or tunnel and have that great feeling of knowing where you are and how to continue, but also feeling smart for doing so.
The world is closest to INSIDE in terms of structure and flow - we have no levels or loading screens, it’s one continuous world. We do have more “forks in the road” than the Playdead games though. There are various places in the game where you have to choose which path to take.
This was important because we wanted to capture the feeling of being alone in unknown lands and struggling to get back to your family. Part of that feeling is making decisions with imperfect knowledge and then dealing with the consequences.
Characters are drawn in Illustrator as a set of pieces: shoulder, arm, hand, etc. and then composed as a “2D rig” within Unity. We use an approach similar to Mark of the Ninja: adding in “breakdown pieces” for arms, weapons, and so on, when needed for a particular attack or movement.
The world is crafted using a combination of sprites (done in Illustrator) and planar polygons rendered in Unity. The game is hand-drawn - we don’t use room kits or set building blocks.
We don’t really think of our art direction from a technical optimization perspective. It was a style inspired by what you often see in movie concept art and from artists like John Harris or Genndy Tartakovsky.
We want our games to capture a feeling or experience rather than trying to mimic the real world. Our goal is to have just enough detail to immerse players while still leaving room for them to superimpose their own ideas and emotions.
Unto The End is about a father trying to get back home to his family. He gets stranded in the mountains with limited supplies and must figure out a way back. He’s a normal man, not a super warrior or a famous knight, he’s alone and often outmatched, and it’s up to the players and their skills to see him home safely.
We wanted our sound design and music to support the emotions and psychology of the father, without distracting the players from their feeling of accomplishment or mastery.
For instance, we don’t have combat music like Dark Souls. Instead, the focus in on movement, footsteps breathing, the clash of weapons.
We also don’t have predefined music tracks. Music is designed to emerge as a result of the player’s actions. Running over a bridge might have some baseline music, and then other tracks are layered in dynamically, based on the player’s actions. Rolling away from a falling rock might cause a “stinger” track to play. Jumping off a ledge as it crumbles away might bring in a new instrument or harmony. Everything is layered together in real-time so it feels like the game is scored for your specific playthrough and actions.
Lighting & Ambiance
Much of the landscape the father traverses is stark and has a foreboding. The player should feel alone and out of place like the world is bigger than the father and he’s simply one of many intelligent creatures trying to survive.
We like to play with light and dark because it’s something we feel all people can relate to. Most people find the dark uncomfortable, so we use that as much as we can. Alternatively, well-lit spaces feel safe and welcoming. We use those to give players a reprieve, but also change things. Sometimes, we’ll drop the audio in a daytime space to create suspense or tension.
A lot of these things are inspired by our own exploration, walking through pitch black lava tubes or hiking in mountains with no set path or plan.
Technically, we use all the standard Unity tricks like unlit sprites, colored lights, light cookies, etc. That’s how we do things like the cave exits and light shafts you see in the trailer.
We also made something that we call a “negative space cookie”. It acts like a Unity light cookie, but we can draw it in the world to block and shape our light. It helps give the ground a better sense of depth and solidity.
In addition, all the tools, FX, camera positioning, etc. are previewed live, just like they would be in the game, within the editor’s game view during editing. When we move the father around, the game view updates to show the lighting, camera positioning, depth of field, fog, etc.
Something we worked really hard on was having effects with the same style as our characters and world. There are lots of tricks out there on how to make such things as realistic fire or blood, but keeping things consistent across all aspects of our art was a fun challenge.
Our falling snow is done using a combination of two-particle systems. The first is a standard Unity particle system where each piece of snow is a particle. The second is a particle system that uses “cards”. Each card is a single texture with dozen or so pieces of snow on it. The particle system moves and slightly rotates the card. It saves a lot in terms of performance. We render the snow on 3 different depth of field layers (background, main, and foreground) so we get nice blurring and parallax.
The blowing snow, fire and fog breath are standard Unity particles with a shader that slightly enhances Unity’s built-in Additive particle shader. We just keep adjusting the size and rate until we get something we like. We also make pretty heavy use of Unity’s wind zones and spend a lot of time adjusting how much a particle effect inherits from them. We find it helps give the effects a more natural feel.
Blood is done with a set of standard Unity particles for the blood spray and a Unity line renderer with a dissolve shader applied over the time to sell the motion of the sword and blood falling away as it moves. We animate the position for the “blood arc” within the animation timeline for each attack and then do a bit of smoothing between the frame positions, otherwise, it comes out looking too geometric.
Our favorite bit is something people probably won’t even notice: if the father stands under falling snow it will slowly collect on his hair and clothing. When he jumps, attacks or does a sudden movement some of it will fall off. If he rolls in the snow he’ll come up with bits of snow on him. Right now it’s only on the father, but we’ll add it for all characters by the time we ship.
Speaking of challenges, I think the desire to do something original and then facing the reality of never really knowing if you’re doing the right thing or if anyone is going to care is most problematic.
When we take time to think about it, we know it’s pretty crazy. 2D side-on combat games have been around for at least 20 years and the idea of a “guy with a sword” comes with a whole mountain of expectations. Trying to reshape those expectations and deliver something new to players has been an incredibly interesting challenge.
We knew early on that we needed to show players that Unto The End isn’t just another hack-n-slash game. It rewards skill and mastery in distinct ways, and despite sharing some mood and tone elements, it’s NOT Dark Souls, Sekiro or For Honor, copy-pasted into 2D.
That reshaping of expectations starts from the first time the player sees the father and is reinforced through every aspect of gameplay. The way the father moves, attacks, jumps, falls, climbs and breathes. The way we present help and training, the way opponents react. How we show damage, what happens when you pick up supplies, and so on. Everything is there for a reason, most of it to help you “get” the game and in the hope you'll have fun playing it.
We can’t wait to show it to more people and ship the game next year.