Lucky's Tale interview: Paul Bettner on crafting a 3D platformer for the Oculus Rift
The most surprising, charming game I played at E3 was on the Oculus Rift, but it wasn't bullet-dodger Superhot or fright factory Alien Isolation. It was platformer Lucky's Tale, which looks and plays a whole lot like 3D Mario. But man oh man, does VR make a difference. The sense of depth it adds is immediately helpful and immersive, and I knew immediately VR would have been a killer feature in a game like Super Mario 3D World on the Wii U. But Lucky's Tale will be a PC, Oculus Rift exclusive, launching alongside the consumer version of the Rift at an unannounced future date. The demo was so exciting, I had to find out more.
In my last E3 appointment, I spent 30 minutes talking with creator Paul Bettner about how his studio stumbled onto the idea of doing a traditional platformer in VR, the technical challenges of nailing the VR camera, and the future of Oculus hardware.
Wes Fenlon, PC Gamer: How's your show going? Exhausted?
Paul Bettner, CEO/Founder, Playful Corp: Yeah, I'm pretty tired, but it's been really awesome to see people's reactions. We came to the show and we had something that I don't think anyone was expecting. And I really enjoyed it, but that didn't mean other people were going to like it. The reaction has been overwhelming.
PCG: It seems weird, in hindsight, that it's something nobody was expecting.
Bettner: I know! Once you see it...
PCG: It seems like such a simple game--oh it's in 3D, of course that extra perspective is smart--but to my knowledge, no one with a profile has tried to do a platformer yet. Nate Mitchell, [VP of Product at] Oculus, was saying you did a bunch of prototyping before you made this game. Can you talk about what you were making?
Bettner: Sure. I told Nate when I was heading into this, the only way I know how to do this is I just have to build a ton of stuff. VR is so new, I have no idea what's going to work. I couldn't tell him, I couldn't tell you. So we just need to prototype. We'll do that for three or four months, and from that we'll see what we think works. So we built like 30 games with the Oculus. We were like, literally putting out a prototype or two every week. We built an aquarium simulator, a version of Zuma...
PCG: Is an aquarium simulator just like...you're looking at an aquarium?
Bettner: You're inside the aquarium, and the fish are eating at your head! [laughs]
We built a music game where you're kinda just bobbing your head to the beat. Half the things we built didn't even have a controller, they just used your gaze to determine what was happening. It was all over the map. We had a game where you were a chef, but the frying pan was attached to your face, and the food would dump into it, and you'd have to flip it up...crazy stuff!
Through that experience we started to see these things emerge that were the best moments we were having in VR. And then one of those prototypes, finally, we stumbled on this third-person platforming idea. When I first heard the idea I was sort of excited about it because those are some of my favorite kinds of games, but I wasn't very hopeful that it would work at all, or be worthwhile to make. But the first time I saw the prototype, we actually built it at a game jam inside the studio, so it was like in two days we built this thing and had the character in front of you...I remember putting it on and I saw the character and I was like "well, it looks like any other platformer I've ever seen. This is kind of cool."
But then I looked up and I saw the whole level just stretching out. And that was like, "whoa, wait, what is that." It was this new moment I'd never felt before in a platformer. At that point I approached Nate and the team at Oculus and I said "this is what we have to make." Not only is that the coolest thing, but a platformer gives us an excuse to have every level be a new type of experience.
The level we're showing here at E3 is a very standard platforming thing. It's such a new idea, we wanted to give something to people they could easily digest. It wouldn't be too hard. But inside the studio we've developed dozens of levels, you're on a hot air balloon and throwing bombs, we can put you under water, all these minigames and boss battles where every single level is like a different ride you're going on. If we weren't doing a platformer, we wouldn't be able to put all that in one game.
PCG: From that perspective it sounds different from something like Mario. And this game felt a lot like Mario. Just sayin'.
Bettner: Yeah, well, that's good! It's one of my favorite games of all time.
PCG: But the idea, the core of Mario, is that Mario jumps a certain way and butt stomps on something, and he has these acrobatic moves. There are sometimes mini games and gimmicks, but it sounds like you're going more in that direction, having more variety.
Bettner: I think we will, but only because VR encourages us to play around with that stuff. The thing I love most about Mario is that they don't try too hard to tell you a lot of story or give you a lot of complicated mechanics. Mario is fine with each level being a playground full of toys. I love that, and that's certainly what we want to embrace with this game. You saw the bouncy mushrooms. They're just fun to sit there and watch in VR, the character bouncing up over your head, watching him do that. I think it probably actually will be closer to Mario than further away, the game we end up making.
PCG: I wall jumped on something in the demo, and was like "Oh, you can wall jump." And Nate was like, I didn't know you could wall jump!
Bettner: [laughs] There's actually a ton of secrets that we've hidden throughout the level that you can only get to by wall jumping, but it's hard to see that in a short demo. There's lots of stuff hidden in the environment, actually.
When you were doing the game jam, was there someone on the team who for like a month had been like "Seriously guys, what if we just made Mario or Crash Bandicoot?"
[laughs] Yes, actually! His name is Evan Reidland (sp) and he's the lead programmer on the project right now. He was the guy who was like seriously, this will work, just let me build it,I gotta show you guys. I was like, I don't know, but it's a game jam, so you can do what you want. He built it and it was phenomenal.
PCG: What else have you looked at for using the Oculus in the game? For example, there's a part where you throw bombs and aim with your head.
Bettner: That's my favorite, because again to use Nintendo as an example, when they build a piece of hardware like the 3DS or the Wii, they're not content to just port an existing game onto that. They always want to do something that could only exist on that hardware. It's leveraging the hardware in some totally unique way, whether it's a touchscreen or whatever else.
With this, I've wanted to do the same thing since we started this project. I want to create mechanics in this game that only work in VR. The throwing mechanic is the first good example of that. It almost feels like this cooperating between you and the character. He picks up the bomb, but he's waiting for you to look where you want him to go and then he'll throw it there. Certainly that experience of playing with your hands and your head at the same time isn't something you can have on a TV. I'm hoping we can find a lot more moments like that.
We've tried lots of little things that are all fun. The bomb throwing one is the one that feels the most natural so far. But when I look at that mechanic, really what I want that to turn into is like a power that the character has that's so much fun, something like Mario's fire flower. LIke it becomes an iconic moment in the game like the cat suit in the latest Mario, something that people relate that character to.
We have one little power where you can shoot water out of your mouth. It's fun because you can stun the creatures and freeze them in place while doing it, but what I love is you can do it and Lucky and he's just like [sputtering and shaking his head].
PCG: One of the things I really liked in the demo was the voxels that would pop out of the cubes. I didn't know if that was unfinished or an intentional look.
Bettner: I'm glad. That's the look we're going for. That stuff in the world kind of doesn't look like it belongs, it's rendered in a different way, and I like how that ended up. It pops that stuff out.
PCG: From the perspective of the graphics, how far along are you? Is this what it's going to look like?
Bettner: I think this is pretty close. We've intentionally gone for something...VR is pretty intensive on the hardware, you need a pretty high end machine to run it in the first place, and we have to run at 76 frames per second. We can't ever dip below that. So because of that, we ended up initially creating a sort of world that didn't rely on a lot of high end graphics techniques. But, luckily, that creates some of the most compelling-looking stuff in VR.
It's funny, a lot of the tricks that we've used in games for many years--a perfect one is texture mapping--we're used to putting textures on everything. In VR, textures end up looking like wallpaper. It's like your brain can see the trick a lot easier in VR. So when we pulled all the textures out and end up with these surfaces that just try to mimic real surfaces like wood and plastic and stuff like that you see in the demo, and instead we spend all our computational time and graphics time on lighting, that tends to be more comfortable in VR. It gives you a better sense you're grounded in the environment. But we're still not going over the top with those effects, because we do want to make a game that will run on most people's hardware once it comes out.
PCG: I'm guessing the draw distance is a fairly big significant deal too, just being able to show the whole level.
Bettner: Yeah. It is, but it's so nice when you can look up and see the whole thing. We wanted to push that all the way out. We barely add any fog at all because it's so neat to see everything in the distance. But it is expensive to do that. It's worth it.
PCG: When I was playing I didn't really have any problem with the way the camera follows Lucky, but a couple people I talked to said that it either felt unnatural not having that control, or that it made them a little bit nauseous. Is that something you've experienced in development? How have you tried to nail down the camera?
Bettner: It used to be way worse. Now we've gotten to a point where a good amount of people think it's comfortable, but not everyone, like you were saying. It's been probably the hardest part of the development of this game, getting that right, feeling good and comfortable. When you play it, I think it's easy to take some of those things for granted. When you're playing it looks like it just works and you don't think about it most of the time, but the way the level is laid out, the way the camera works, it's all carefully orchestrated so the camera never runs into anything, doesn't clip through the geometry, there's always enough space around you so if you lean forward and around you don't' run into anything. And yet you're just close enough to the character to feel like you could still reach out and touch him.
Getting everything to work that way, it's not even so much about the code that runs the camera, as it is about the level design itself. It's this back and forth. A lot of people say that they love when stuff gets really close to them, like the fish that fly off the water wheel and hit you in the face, but if we do too much of that...it's this careful balancing act.
If we were to look at the game in your dev tools, would the camera be this swooping thing, following this rail?
Bettner: It would. You remember Lakitu on Mario 64 on the cloud? That's basically what you're like, just a camera floating on this cloud above the character. But it's the other stuff that makes that work. We shrink the whole level, so the distance is--there's like this comfort zone in VR. I have a sense for what might cause this.
I think that evolutionarily, our brains are wired to have the greatest sense of depth right here where your hands are, because that's where you're operating tools and stuff. Stuff feels most satisfying when it's in the range of one foot to maybe eight or nine feet away. So by shrinking the whole level we basically bring all the gameplay into that sweet spot that's happening right there. That just ends up feeling the best. We just kind of stumbled our way into that, but that's what we found.
With the camera, what got you to the point that makes it feel good? Was it making it move more slowly?
Bettner: Yeah, it was. And the effect of shrinking the world, means that the camera is only moving at a few inches a second as it follows the character. Whereas when he was really big and the level was huge, it had to race at many miles per hour to catch up with him.
PCG: And people threw up.
Well, it was about the same as you find in most VR games. Most VR games are running at a regular scale, 1:1. You have to move fairly quickly through the environment. Here, because everything is so small, the level itself is literally only a couple dozen feet across in real units. And so it's almost more like you're leaning around rather than running through it as Lucky moves.
Some of the other things are, if you notice when you go through those tunnels, those are really neat moments, but they don't last very long. What people describe is that they want that sensation every once in awhile. They sorta want the roller coaster to drop every once in awhile, but they don't want it to last very long, and after that they want to be in a calm place. They don't want it to keep going, because then it's like, this is too much.
Did your team start working with Oculus stuff before Crystal Cove, with the original dev kit?
Bettner: Oh yeah, yeah.
PCG: What have been some of the big changes for you as a designer?
Bettner: Positional [tracking] is obviously huge, but we're not leveraging positional as a gameplay mechanic too much. The biggest thing positional adds, it's just so much more comfortable. The small movements of your body, fidgeting around in your chair, because that's tracked now, it's just hard to describe, but it's fundamentally more comfortable because of that. But resolution, the increase in resolution, has made the experience so much better. With the Rift DK1, it was almost like a simulation of virtual reality, but you still knew that you weren't really feeling like you were there as much. But we've just now with DK2 crossed over this threshold, where nothing is really in-your-face wrong about the experience. Mostly you can just let it go and be like, yeah, I feel like I'm here. But they still have a ways to go. I think they still want higher resolution, better tracking.
PCG: I feel like 4K is where it's really going to hit the sweet spot.
Bettner: Michael Abrash says it's actually 8K that we need, which...[laughs]
PCG: Well, he's got high standards, Michael Abrash.
Bettner: [laughs] He does. It's interesting though. We were starting to get to a point with PC hardware, where you could definitely go and buy the Titan Black, but you were only getting small incremental improvements in the way the game looked. And now VR comes along and our PCs can barely keep up with what VR requires. It kind of resets that whole thing, and now there's plenty of reason why you need dual Titans in your machine, or whatever, 'cause VR is making you render each eye at 8K and you have 16K of total resolution...that'll be nuts, and I think the race will continue for a little while just because of that.
PCG: When you were prototyping other ideas, were you looking at PC games and going, there's no good Mario on PC? Platformers are really underrepresented on PC.
Bettner: I know, I know. But that's kind of a double edged sword. I think so, and I think that PC game market is totally underserved there. I think they're super hungry for a great platforming game, or more than one. But at the same time, people will point to that and say, aren't you afraid that nobody is going to buy it because the PC market isn't that way? And I'm like, I think gamers love great games, and I don't think the demographic works that way. If you make something that everyone's going to love, then by definition everyone's going to love it! I think just because it's the PC gaming market, that I have to make hardcore, gritty, whatever. I think those people love a Pixar movie just like I do.
PCG: How much more do you have to build in the game?
Bettner: We're only four months into this, so it depends on the launch date for the consumer version of the Rift. That is what determines how much time we have to spend on this.
PCG: Is your plan to launch with the consumer Rift?
Bettner: Yep. We want this to be one of the first games people can get for the Rift when they first buy it. I don't know any more than you do about when the exact date is going to be.
PCG: I was going to ask!
Bettner: But I do know when it's not going to be. Some people are speculating certain dates, and I'm like, at least I know that it's not going to be that. So we have a certain amount of time to be able to build this game to a certain scale, and I'm happy we've got it to a point where it's mostly comfortable. We still have work to do there. Now I want to create these incredible made-for-VR moments, like the bomb throwing, I want to find more of those things and build a lot more levels. I hope to deliver a game that feels like the same scale as something you might find as a high-end PC or console title.
PCG: Can you tell me when the Oculus is not coming out?
Bettner: [laughs] No. All I can tell you is that the DK2 will be in everyone's hands really soon, and really, that's so good, it's' even better than I thought the first version of consumer VR would be. I thought it was going to be more like DK1. And now DK2 is here, it's way beyond that, and it's still not the consumer version. Oculus has had so many chances to take shortcuts, they could've capitalized on all the popularity and hype of DK1, put it up for sale, tried to get as many people as possible to buy it. But they did the opposite. If you go to try to buy one, you have to click through all these pages that are like "Stop! If you're not a developer, don't do this, don't buy it." And they still sold like 80,000 of them. I think that takes a lot of courage on their part, to be able to keep the long-term vision in sight.
Thanks for Paul Bettner for fitting me in for a last minute interview. After I stopped recording, I strapped the Oculus on for a second run through the Lucky's Tale demo. Bettner gave me a few more interesting tidbits as I played: the demo level, which is very laid-back, will probably be one of the first in the game. The dev team has already designed some much more challenging stages, and it's actually harder for them to be restrained and develop easy levels. The overall structure of the game isn't set yet--they haven't decided if levels will be hub-based, like Mario 64's castle, or use a world map, or take some other form.