As I chat with Bossa Studios designer Luke Williams on the E3 show floor, there is a guy standing beside us in front of a TV, trying earnestly to manipulate a living piece of bread into a toaster, using a PlayStation 4 controller.

“It does look like he’s struggling,” Williams says.

Bossa’s latest game is I Am Bread, a physics-based game that has players move a piece of bread across various environments with the goal of becoming toast. Moving around requires a lot of wiggling and swinging, and four controller triggers are used to grip and release at each of the slice of bread’s corners.

If that sounds awkward, I can confirm that it absolutely is. But Williams says the purpose of Bossa’s awkward, hilarious games like I Am Bread and Surgeon Simulator isn’t to frustrate or troll players, but simply to try new things with game controls and concepts.

Here’s a condensed version of our chat.

Why bread? Why not a hot dog or something else? A piece of bologna maybe.

I like it when the logic and the mechanics go nicely together. I can take the premise of bread wanting to become toast, and put that mechanic in any environment. The player is always aware that they’re to become toast. And we can extrapolate that idea, that same goal, into all kinds of settings, add complexity. That tied it all nicely together for me, along with the whole idea that “the floor is lava.”

Basically I wanted to try all kinds of control iterations. We wanted to have bread move in an interesting way, or how you might imagine bread would move. If you just take a control stick and shuffle the bread around as an animation, “A” to jump, that’s not very interesting to me. We like to keep it a little unconventional, and let players kind of make up their own way of moving around. That’s why we use physics and stuff.

When you make controls quite hard like we do, it’s hard to tell when they’re good, because you start playing with them. Are they hard for you, or is it just me? You want to be able to get good at this. We don’t actually introduce new skills or new abilities, you just actually get better at the core controls, and then we throw more complicated environments at you to push you further.

How do you playtest a game that’s kind of meant to be frustrating?

It’s hard, because we’re actually not trying to make it as frustrating as possible. We have our controls in the way we move bread, we try to make that as intuitive and easy as possible. It’s just that wrapping your head around a whole new logic of controls is going to be very hard.

We didn’t deliberately go, “Haha! Let’s throw in some random gestures!” It’s all very consistent. When you see people doing the 13-second heart transplants in Surgeon or the 20-second Bread speed runs, it’s because they’ve played it enough that they know exactly what the controls are going to do.

Is it a misconception that Bossa’s games troll players?

Yes, absolutely! It’s a weird one because people say that we’re just trying to deliberately piss people off, or make it so hard. But from my perspective, I just like playing with strange control schemes.

We have this device, this input, and everyone is adhering to the same stuff all the time. If you think of a concept and what that game will be, and get the controls to fit alongside it, rather than saying “We’re going to make a first-person shooter” or “We’re going to make a third-person shooter.”

We ask, “What is the kind of game we want to play and make, and how do you interact with that,” and build back from there. We try to make the controls intuitive and as fun as possible, and not just be like “Haha! We made it so deliberately hard!” When  you go the deliberately hard route, you end up with these impossible games, and those are easy to do. People feel like they can get better [at our controls].