Skip to main content

Fantastic Contraption: Why A Cat Is 100 Times Better

Virtually building

While at GDC I had the opportunity to sit down with Colin Northway, one of the creators behind the excellent VR remake (and the 2D original) Fantastic Contraption [official site]. We talked about how VR worked differently compared with traditional game design or, to put it less mundanely, why a walking cat dispensing sticks is better than a menu system. While at GDC Adam really loved Fantastic Contraption and, I think Graham was having a similarly good time with it back in the UK. We also made it one of our Games Of The Half Year so it's high time we put this interview up!

Pip: What was the moment where you said we should do Fantastic Contraption but on Vive?

Colin Northway: We first tried the Vive at Valve so we played The Blu (an underwater experience), Tilt Brush (a VR art toolset) and Job Sim (a tongue-in-cheek job simulation). It was fantastic. We went in being lukewarm on the idea of VR and then came out being like 'Okay, yeah, I could dedicate my life to this, I guess!'

At some point I was thinking about Tilt Brush and how Tilt Brush has these really nice big motions where you're moving your arms all around and you're walking round the space a lot and you're moving in to do details and moving out to do big - I really liked how it used the space. I was thinking about Fantastic Contraption, the game from 2008, that was 2D, thinking that would have the same kinds of motions.

Watch on YouTube

[Fantastic Contraption is about making these contraptions out of sticks and wheels to accomplish particular tasks]. If you had to have a stick that was as tall as you were you would sweep your hand from the bottom of the floor to the top. If you're working on a contraption like a machine you need access to all sides of it so you'd be walking around it the whole time.

The things I liked about the play experience of Tilt Brush seemed like they would be in Fantastic Contraption a lot so we got really excited about it. I managed to get my friends Andy Moore and Lindsay Jorgensen and Kim Voll - they got excited about it so we banded together and started cranking.

Pip: In the process of making it what has been the unexpected challenge of VR? Your GDC talk is about menu systems so was it that?

The interesting thing about menus is the same thing as what's interesting about a lot of game design in VR and that is - you've played videogames your entire life, you know a voluminous amount about videogames, you have all of these interfaces and mechanics and games stored away in your brain that you've played for hours so when you go to design a game you're always drawing from this depth of knowledge. When we started working on Fantastic Contraption it was the same way. We started drawing on 'What is the videogame solution to this?'

My moment of awakening was, so we started working on the game and it's about building stuff. In the original 2D version of the game there would be a menu where you would select wheel mode and then you would click to add a wheel. You'd select stick mode and click to start a stick and then drag and let your mouse up to end the stick. It's a very videogamey interface so when we go to make the new one I'm like, 'Let's do the same thing, that's how videogames work'.

Some other team members who had played more in VR were like, 'Maybe we should make a bucket with sticks in it so you can go over and grab the bucket?' I was like, that sounds like a pain in the ass! You've got to walk over to the bucket all the time, why don't we just do the videogamey thing?

We did both, we put on the Vive pad a menu so you could click to select a different mode then click and drag in space. We also did the bucket, although when I complained I didn't want to walk over to the bucket all the time somebody - I think Lindsay - was like, why don't we make the bucket a cat and then the cat can follow you around? That was a pretty great idea.

So we made the cat and the videogamey idea and worked on it for a while before we got access to a Vive. Then when we were in our world for the first time I tried both of the interfaces and it was just immediately obvious that the cat was a hundred times better. That's the lesson we're all going to have to learn.

Adam: The cat is a hundred times better?

Colin: That's right. A lot of Contraption has been that. We tried as hard as we could to take all the menu stuff out of it. We have a level select that's like a table in 3D with little models of every level on it. If you want to load a level you look at the models and decide which one you like the look of then you pick it up - you grab it with your hand and put it down on the model of the level you're currently playing and poof! it becomes the new level.

You save and load the same way. If you want to save a contraption you've got a little model of the contraption you've built. Pick that up and put it down on the table. Then you can come back the next day and it's still there on the table.

Those are things we've never had to think about before because we could never physically do them. That's where the biggest challenges and also the biggest rewards come in working in VR.

Pip: Will these lessons help us simplify or improve 2D interfaces or is it so intrinsic to VR that they just don't apply?

Colin: I think it's unlikely that it's going to feed back into 2D screen games. The reason I think that is because there's this fundamental problem. It's also the reason sometimes people are like, 'You should release Fantastic Contraption as a 3D normal screen game with a mouse and keyboard'. [As in, the VR remake should get a 3D version for non-VR platforms.] There's a fundamental problem where pointing to a point in 3D space on a 2D monitor is really hard.

Look at modeling software. They're non-intuitive to use and it takes a lot of training to be able to use them well. So making games around non-intuitive interactions is almost never a good idea.

Really, the whole reason we can do these different interfaces is we have this ability we've never had before and it's never going backwards because there's never going to be an intuitive way to pick out a 3D point in space on a 2D plane. Actually that's why Infiniminer is such an important game. Zach discovered a way to basically do that. It's the minecraft idea of [you're always acting] in relation to one block.

Adam: Do you think VR is something players are going to have to learn? One of the appeals of the Wii was people could pick it up and play it. They didn't have to have been playing games their whole life. Is VR the same?

Colin: I think there's a ton of learning to do but mostly that's on developers. Certainly, VR has the opportunity to be much more intuitive. It all comes down to how good we are at realising that. If you look at Job Sim, it's such an intuitive - you drop in and you throw stuff around, press buttons, mess with stuff, use blenders and stoves. It doesn't need an instruction manual. If you look at Tilt Brush the basic action of drawing is intuitive but you still need some - some more videogamey verbs or computer tool verbs. Here's a menu, you click through the menu and point and click to select. It depends how hard you work at it and what game you're making.

Pip: With Job Simulator I just did all the things it was asking me to do, I was really well behaved. I was a good employee. It didn't occur to me to mess about until I was watching someone else!

Colin: [much laughter]

Adam: Thank you for your time!

Read this next