Behind the Development of Hand-Drawn Strategy Game Don't Kill Them All

Could you please introduce yourself and tell us about the studio and its creative vision and vision for the game Don't Kill Them All, particularly? 

Vincent Rochette: My name is Vincent Rochette. I'm the creative director for Don't Kill Them All. I work at Fika Productions, which is a company that we started about eight years back, and we started coming out of school, so none of us have worked much in the industry before setting out on Fika Productions, apart from our Art Director and Concept Artist Ariane Petitclerc, who had worked a little bit previously in video games. 

The studio was founded on the concept of fika, which is a Swedish word for a coffee break, so the idea is to take things slow and 'take time to take time' basically – to have a coffee break, to relax, and to joke around with friends. Behind Don't Kill Them All, it's a little more specific, I guess. I personally really like strategy games. I think when strategy games shine is when we let players care for their units, and Don't Kill Them All is kind of an extension of that with the orcs, which are these cute little annoying creatures that you're supposed to take care of and watch grow and become useful units on the battlefield. 

Please tell us how the concept of the orc training system came about. I played the demo and found it really interesting – it combines such laid-back calm gameplay and intense battles, where you have to think. How did this combination appear, what inspired you, and why orcs? Also, how did you come up with that training coaching system?

Vincent Rochette: The idea behind Don't Kill Them All actually doesn't come from me. It comes from Francis Beauchemin, CFO and Community Manager at Fika Productions, and the initial concept was the following: say you're managing an army and you're going out to war; is it really a good idea to kill everything in your path? Don't you want to have something to loot? We thought the orcs were a good shorthand for that kind of logic where we're used to orcs and media screaming "kill them all," "behead everything," and "destroy everything in your path," and then we thought it'd be an interesting twist for a strategy game – you're in charge of these super brutal monsters, but their brutality alone won't get you the resources you need. You've got to try to temper them a little bit, which is like a reverse power fantasy, whereas usually you're playing a game and you're trying to get higher damage numbers and higher stats; while there is a lot of that to our game, how you use that power is way more important than just accumulating it mindlessly. So it stems from there and builds upon itself, and a lot of the design is informed by the themes that we want to do.

We didn't think up a design and then skin it over; we thought lore-wise first and like what would build this up, and then modified our design so everything meshed together thematically.

When creating the concept and the game itself, how did you plan to balance tactics, base building, and combat? What part of the game did you consider the main and prevalent?

Vincent Rochette: We had that conversation a lot, especially during early production where we weren't sure if we wanted to create the game horizontally – to make everything and see how everything builds up together, or vertically, where we focus hard on one part and then spread out afterwards. And in the end, we realized that a lot of what we do is fed by the encounters and the combat system. So we attacked that first. The idea is: how would an orc manage those combat situations to get the most out of it? What kind of tools are at their disposal? There's node protection, there's node gathering, there's managing hostile AOE patterns – all that stuff was built first, and then we just asked the logical question: okay, so now that we have all these resources, what do you want to do with them? And the base building came through itself. It's a game where we're inspired by a lot of different titles that have this "go on an expedition and then come back and use what you've gathered on that expedition to boost something."

The base building part was a logical step for us. We're showing that the way that the orcs are managing all these encounters without you, where they're killing everything and razing everything and not really caring for the resources – it's not really working. And the best way to show the player that it's not really working and that what you're doing now – adequately strategizing is working is by having the player construct stuff themselves. If you're seeing the camp you're building grow before your eyes, it's a pretty strong way to show that the player is having an impact — the player thinks, 'I built that mud pit, I built that workbench,’ and if you didn't save those resources, if you didn't alter your strategies, if you didn't protect those resources, you wouldn't be able to build that. So, I guess it came from the encounter system and then fed through the base building.

You mentioned that you drew inspiration from some other strategy and turn-based games. Could you please mention specific titles? What inspired you? What did you use as a reference?

Vincent Rochette: We've got a lot of comparisons to Cult of the Lamb, and that's definitely one of our inspirations. There's a lot of Stardew Valley as well in there, especially in early versions, I think we put it forward a lot, and now we're building a lot of systems that merge well with that core progression, where in Stardew Valley you can farm and do other tasks, then do something and then come back and macro loop the way that time works; we were heavily inspired by that.

Into the Breach, of course. Everything with threat management and seeing your battlefield before you start managing threats and whatnot. 

Stardew Valley was unexpected to hear in that list.

Vincent Rochette: I don't think our game screams Stardew Valley, but I do think that there are a lot of macro design decisions that were heavily inspired by Stardew Valley, and we're still working on the game, there are a lot of systems that have yet to be developed but I think once you do actually see the final product those inspirations are going to be a lot more clear because it's a lot in the macro. It's a lot in the year system and how time goes forward. There are some days that you have different activities you have to prepare for that kind of stuff. It’s a pretty big inspiration for that. Since we're not really there yet in our... well, we are there in our design process; we're just not there in our production process yet. We're working on those kinds of systems for the foreseeable future. Darkest Dungeon, big one. That was a big inspiration for the raid system – you're opening your map, you're selecting access points, you're selecting places you want to go raid based on some objective, and then going over there, doing something, and coming back.

We were also inspired by Wartales. Wartales has this passive trait system that we were pretty inspired by as well. What I feel Wartales does really well is that you've got your party of mercenaries, and the way those characters craft relationships and whatnot is super organic. It's not necessarily story-driven; it's action-driven, and that, in the mind of the player, creates a narrative for that character. It's like, "Oh no, I don't want Bertha to die 'cause Bertha is super cool – she killed three people in the last encounter, and I found her at this tavern."

Your units are building narratives that the player can latch onto and extrapolate for themselves. XCOM as well – pretty big inspiration, of course. It's one of the first strategy games I've ever played, and I think a lot of its design permeates my vision of the thing.

Moving on to the technical part of the development, which engine does the team use and why?

Vincent Rochette: We use Unity; we started using Unity on our previous game, Ship of Fools, and since we know the ins and outs of the program, it wasn't super strategic to completely change the engine for the second production. We thought that it would be nice to capitalize on what we already know from Unity, so there wasn't much of a real push to change engines. I think we're pretty Unity-driven for the moment.

Could you tell us about some of the biggest technical challenges the team encountered during development? Maybe you came up with custom tools, or there were some interesting points in the technical process – could you mention some of those?

Vincent Rochette: One of the things that we've, I think, really taken advantage of is that from the get-go, Don't Kill Them All wanted to preserve the visual identity of Fika Productions, with Ship of Fools and now Don't Kill Them All. I think that if you look at both art styles side by side, there are definitely some comparisons, and we wanted to pursue creating a game in 2D, but one of the things that we realized in Ship of Fools is that everything looks flat, which is fine because it's an orthographic perspective, but we wanted to push a little bit further and we've started using projections, which means we're drawing the asset in an isometric perspective and then we're creating meshes from there and extruding and pushing back according to the camera, which makes it so we don't need to make models for everything. It's like taking the whole modeling job and shortening it to preserve the artistic vision. That's a big tool that we've started using in this production that we didn't use too much in Ship of Fools, only in a couple spots, but we innovated a lot more for Don't Kill Them All, especially because the camera zooms in and zooms out, it moves around all the time, so we wanted to give the aspect that there's some depth.

Because of the design, we had to innovate our dungeon construction system. So our raids aren't procedurally generated, but they're built with a bunch of different systems that are talking to each other. There's the playable area, which is on the ground floor, and then there's everything that's on an elevation. For the canyon, that's everything that's on the sand on the side, and then everything that your orcs can walk through is the ground layer. Everything that's on the side is handcrafted in these tileable environment set dressings. So we're going to take a 20x20 layout of tiles and then place a couple of trees, make sure it makes sense with the visuals, make sure there's no clipping or whatnot. And then the environment builder is going to go take that and tile it everywhere that the dungeon is going to be imprinted on, and the dungeon is basically your ground level. It's a tree where you decide what goes after what, and then, through an algorithm, it'll place rooms side by side according to what you decided goes after what and before what. The area that it's going to place things in is always random. So the player doesn't do the same three access points over and over, and doesn't start to feel the exact similarities between environments. It's a little bit more hidden that way. So that's something we definitely had to figure out – a big part of it.

What was the inspiration behind the distinct hand-drawn style of the game? How did the art direction process go? What were the tools that the artists used?

Vincent Rochette: Ariane and I are co-art directors; she has more of a design background. My background is more in animation, and I think we complement each other through those two sorts of branches. I think one of the important things for our art style is that when we start designing something, it's always like, what are the constraints first, and then how can we still achieve our vision with the constraints, because maybe the style could have been ultra complex but then it's unreadable and whatnot.

The art style is built by constraints. One of our constraints is that we don't have someone who's a dedicated modeler on the team. I guess it's a constraint, but it also pushed us to innovate with the 2.5D style and projections. We've got some textured line art, which is pretty thick, and that's again driven by necessity because we're zooming out and zooming in from the orcs; we want to make sure they're readable against a lot of different backgrounds. They have very iconic shapes – we're trying to stay away from overly complex design decisions. If you're looking at five orcs, you're not overwhelmed with a bunch of visual information; you can parse each thing.

We're putting a lot of emphasis too on customizability and making the player feel that the experience that they're having is unique to them and it's really their own, which is why we designed the orcs in a way where they give you a blank orc canvas where you can put a bunch of different customization options. And sometimes we had fun just switching over facial hair, or switching over hair, or eyebrows, whatever. And since the faces are so simple, changing one cosmetic option can really give the entire character a whole new look, I guess.

I noticed that the orcs have different abilities and different weapons. Could you tell us in more detail how these individual traits change the combat and the story in general? As far as I understand, different characters can shape the story and the fighting process differently. Could you tell us exactly how this happens?

Vincent Rochette: That stems from the core notion of our game that we want players to really connect with their orcs, which means that every orc needs to be unique and different in their own way. If we go more into a game design aspect, the orcs are different, and they serve as your backbone for your build. The orcs have a synergy with themselves and also with the team that you're composing, and the player has to keep that in mind.

An orc that has the agitation trait just comes with that, right? So they have the agitation trait — that means they have three-range tolerance and seven movement. So they're pretty speedy orcs to move around the battlefield, and it's maybe a good idea to pump up their movement so you have a really agile orc on the battlefield.

And then you've got your archetype of the orc that's, whenever you need an orc in a certain place, that orc is the one you're going to turn to, because you've built them a certain way, you've bought equipment that gives them maybe special movement options, maybe you've given them a leap or something similar, and you've maybe changed their weapons. So, like the war scythe, which does extra distance. Thus, you've got that figured out and your build is crafted to that, but one orc alone isn't going to cut it, so you're going to have to create different archetypes.

The archetypes are not things we're telling the player. The player makes up their own archetypes based on what they think will be useful, which I think is a pretty fun way to play with the class system. You could have someone with the agitation trait with a huge hammer as much as with a war scythe. Those two builds are equally viable, depending on what the other orcs that you're bringing in are going to bring to the table.

It helps the players differentiate their orcs when they've built them a certain way, and it will hopefully help them create their own sort of narrative, like what I talked about in Wartales, where the player builds their own meta-narrative on the actions that happened, and then, "Oh, I remember this orc — this orc got tangled in the spider webs earlier, and now they behave a certain way." They have their identity, and I care for them because they're an individual. That's how those two kinds of systems work into each other.

The cosmetics of your orcs are going to differentiate them, and then the cosmetics are going to help you remember that that's the build I gave this guy, because it's Bruce's build, and Bruce has that build, and I've noticed that it's not because Bruce wears this thing.

The game has a demo now, and I'd like to ask what feedback you've received – did it shape some of the development process? Did you change any decisions because of the feedback you got?

Vincent Rochette: That's a good question. We haven't made radical decisions because what I think is really inspiring and fun to see is that a lot of the feedback that we've gotten is super positive. People are really responding, which is super validating for us as a team to know that we're connecting with our community, and most of the feedback that they're giving is the kind of things that we're working on or that are planned or whatnot. We're getting extra feedback that is telling us that we're heading in a good direction because people are logically anticipating the features that we're making. A lot of that is pretty cool.

I think one of the things that we've really taken for granted but that's helped a lot is that a lot of players say on our Discord – they're reporting bugs, and it's super helpful to have people reporting bugs because then we can use them to prioritize if we've got like 40 people reporting the same bug, obviously we can look at our bug chart and decide, which one we're going to fix first. So, that's a really big thing. There's a lot of good feedback too that's not necessarily changing our direction completely, but it's altering how we're thinking of making it – "okay, so they're telling us that they like this aspect of the game especially, so we can tailor what we're making already to come play with that feature more than other features."

As it says on Steam, the game is scheduled to launch in Early Access by the end of this year. We're looking forward to the full release and to all the new features and systems that will be added. I really enjoyed playing the game, and I would like to continue trying all the features.

Vincent Rochette, Creative Director at Fika Productions