Chandana Ekanayake is a game director & art director at Uber Entertainment where he’s working on a VR adventure game for the PlayStation VR titled Wayward Sky. I caught up with Chandana at the Seattle VR Expo where he shared with me some of the highlights from his presentation about “Unlearning Game Design for VR.” He talks about his lessons learned from creating a third-person, single player adventure game focusing on atmosphere and storytelling and why they decided to focus on the PlayStation VR as their initial platform.
LISTEN TO THE VOICES OF VR PODCAST
Become a Patron! Support The Voices of VR Podcast Patreon
Theme music: “Fatality” by Tigoolio
Subscribe to the Voices of VR podcast.
Rough Transcript
[00:00:05.452] Kent Bye: The Voices of VR Podcast.
[00:00:11.920] Chandana Ekanayake: My name is Chandana Ekanaike. Most people just call me Eka for short. I am an artist and designer and the project lead on Wayward Sky, which is, we call it a look and click VR adventure game. It's currently slated for PlayStation VR launch. It's essentially a third person for locomotion and first person for interaction. So we're trying a hybrid approach. I've been working in VR for the last year or so. I've been in traditional development for 18 years, from console, mobile, PC games. I work at Uber Entertainment. We're mostly known for Monday Night Combat and Planetary Annihilation, RTS games, and a shooter. And Wayward Sky is our first VR game.
[00:00:46.700] Kent Bye: And so you're giving a talk here at CVR on unlearning game designing for VR. So what are some of the big points that you're going to be making about how do you do game design for virtual reality?
[00:00:56.548] Chandana Ekanayake: So I saw the Valve room demo, I think almost two years ago now, early on. We're across the street from Valve, so we get to walk over there and see demos on occasion. And I saw an early version of what became the Vive. And it was a really simple demo. It was essentially, I was standing on a box, looking over another box, mapped with a flat color. And one of the Valve employees said, try walking off the edge. And I could not. And at that point, I was like, I'm sold. This is amazing. I want to be in VR. So we had our DK2. Oculus, early on, sent us a Gear VR headset, so this was last year, last August or so. So, the Gear has a single button input on the right side, so it's like, what can I design for this thing? That was our sort of first approach. So I started playing around with locomotion ideas. I tried a bunch of demos. Anything that comes off of Reddit Oculus, I'll download and try, and I find I'm pretty sensitive to motion sickness, and I still am, even a year, and people talk about VR legs, and... getting over it, but I haven't. And I think it's a good thing. I don't feel like we should force our audience to get VR legs or force them to get better at VR. There's a varying degree of opinion on that, but at least for this first generation, make our players and grow the community. Because again, if our players are turned off by getting sick or a friend tells them they're going to get sick, then they're going to be turned off by VR. And that's not going to help anybody. So we started with that. I started prototyping about a year and a half ago, and I really wanted to do an adventure game, see what it would feel like to do. And with the gear, with the limitations of the gear, using your head to point and click was a decent idea, so we tried that approach. So exploring a space, great. Okay, now how do you actually traverse? So we tried putting in connected rooms with triggers that let you walk through a space. We do camera cuts. I'm like, okay, this sort of works, so we can see, explore a bigger area than what's in front of your periphery within your view by doing camera cuts and third-person movement. So for movement, we try third-person, looking with your head, pointing an area, and the avatar moves at that location. And there's hot spots in the world where we want to design puzzles, and I was like, well, we should do that in first-person, just change the scale. So you click on a hotspot, the camera switches to first person, and now you're 1 to 1 scale, human scale. In third person, you're essentially 60 feet tall looking down on a diorama. And I really like the play of scale in third person to first person. So we tried that. There's a bunch of other tests we tried. Just like what most people end up doing going from traditional dev is, I want to do an FPS. I want to explore the world. I want to move. And we tried that initially, and I could not hack it at all. It's bad. Like, once you get a little bit of motion sickness, it kind of lasts with you for the day. And sometimes it's worse, depending on the day, too. So yeah, so we approached it third-person, first-person. Again, with the limitations, the design sort of fell out of what the hardware can do. So, I like stylized art myself, personally, but because the gear is also, it's a mobile device, there's only so much you can do on that, and we decided to do the art style for Wayward Sky in a more simplified approach as well. So that's one of the first lessons, locomotion lessons. Then once we got our hands on motion controls, I was really sold. I got what was called Morpheus at the time, the PlayStation VR demo at GDC this past GDC last year. I was really taken aback by how good the motion controls worked and how The head presence was already there for me for a lot of things, but once I had like hand presence, I was like, okay, this is a game changer. So once we had that, we decided to partner up with Sony for Wayward Sky and introduce motion controls and also positional tracking, because that's the Gear VR, you know, it's really hard to do that or you can't do that yet because it's a mobile device and there's no external camera. So we decided to switch to PlayStation development. At the same time, we got our Vive kit, and we messed around with the Oculus. And for me, the big lessons, the things that I look forward to, what I consider first-generation devices with presence, the PlayStation VR, the Vive, and Oculus with the touch controls, move controls, and the Vive controllers, for me, that's really what VR is interesting, where it gets really interesting, where you have a world that you feel like you're in, and then you also have hands or wands or things to manipulate the world with. So we integrated those kind of motion control ideas in A Wayward Sky. In third person, we essentially give you a flashlight laser pointer to, instead of using your head to point, we let the player point with their hand with the wand controller. And in first person, instead of using your head to look at puzzles and rotate objects, we give you two hands or a hand to manipulate. And as we playtested, we like doing shows like PAX and these kind of things where we just let people play it and we see how people react. And just getting hand controls and motion tracked hands one-to-one, people just naturally reach out and try to touch things because it just feels good. You know, as we're developing the game, we'll keep trying different things out or what works well, what work doesn't. How do you exit out of first person into third person? And doing straight camera cuts was a little jarring. So we fade to black between transitions super fast. You notice it more as you're watching the game, but when you're playing it, most people really don't even notice that there's a transition. The other things we learned doing third-person interaction is when we transition between shots, it could be disorienting to the player. So we use the main character, the avatar, to make sure that she's always in view. And she's sort of the linking element between each camera transition. And there's some cinematography lessons we learn from how much of an angle do we rotate the camera, how much we don't. You know, we never try to go beyond 180 degrees either way, because you can be really completely disoriented when you cut between cameras. Same thing when you cut into first person. Because it's third person avatar, we cut the camera to the direction the character is viewing from, not the player, and that seemed to work pretty well too. So we tackled locomotion with third person and first person. We tackled interaction through the motion controls. And then the camera, you know, how do we get the player to look at what we want them to look at? We tried an experiment where, like, how can we do a cutscene in VR? How does that work? When the player is the camera, we don't want to do any sort of cinematography things like track or zoom and move it around or screen shake, things like that. We just used old theatrical tricks, essentially. So if you're looking at a scene with two characters, how the characters are interacting, the direction they're looking, you know, a plane that flies by with a longer trail so the player naturally looks at that. Or when the characters are in a fight, we throw objects in the direction we want the player to look at. So there's a lot of those kind of tricks, visual tricks, where if we want you to look to the right, we make sure there's something interesting over there. And then we use sound. We use 3D positional sound to direct the player's attention as well. So yeah, cutscenes in VR worked well as long as you let the camera always be the player's head. You know, we teleport between cameras in the scene, but the player has full control of the head all the time. We did mess around with movement. So you can do movement without getting people sick, especially getting me sick. as long as you do stuff like it's linear movement and there's no acceleration on the camera. So we have sequences where you get in an elevator, it goes into first person, the player hits the button and there's an audio feedback of the button clicking in and the elevator mechanism starting. So if the player knows that's going to happen, they're way less likely to get sick. So we do linear movement vertically. We also have sections where you fly through a space, but it's slow, linear, no accelerated movement. That seems to work well.
[00:07:45.703] Kent Bye: And so what were some of the gameplay mechanics that you found fun in VR?
[00:07:50.364] Chandana Ekanayake: So touching things in VR is just, it's so insane. It's so fun, like just picking stuff up and throwing it around or manipulating, because it just feels like you're grabbing the element in the world. So the main menu that we have right now is essentially a physics playground. As soon as you boot up the game for the first time, we want you to understand what's interesting about VR. So we let you play with objects in the world that are characters in the world that are essentially little toys that get unlocked as you progress. But if you buy a headset and you want to show it to a friend, the first thing you want to do is like, hey, try this and try grabbing things in the world and try it out. So physics interaction is still very fun as long as you can, you know, physics can go a little crazy depending on the performance and what you do with ragdolls and stuff like that. But we try to go with like what feels good. If I was in this space, what would I want to do naturally? you know, what could I manipulate? So anything we put in the world with just puzzle knobs or doors with complex mechanisms, like, okay, that's the thing, we make sure we put a handle on it so that the player can grab it and rotate it, and there's feedback. There's visual feedback, there's audio feedback, and there's feedback on the controller as well. So touching the world still feels pretty good. So our approach is, yes, let's do puzzles that are interesting, but, like, the interaction and interactivity should trump everything else.
[00:08:56.598] Kent Bye: I think that's a challenge that a lot of VR tech demos is that you may have something that may be novel and fun but lasts for 10 minutes and then you get bored. So what is it that you did in the design of your game that made it a little bit more interesting or hard where a player could fail and then there'd be a challenge where they're actually cultivating some sort of skill?
[00:09:15.502] Chandana Ekanayake: Right, so going from a tech demo to a full game, there's pacing, right? You don't want to make your player tired by constantly giving them things they have to touch and grab and throw and things like that, so we pace out. So the third-person movement to first-person interaction worked out really well for pacing as well. So, essentially in third person you look and click and you click in the world and there's a little hotspot that gets drawn and the character moves. At that point the player is actually free to look around and explore and figure out where I want to go next and they get to a first person hotspot. Now it's like, okay, now I'm a more active participant of this. So there's a little bit more passive to active pacing we do throughout the game. In between that, we also pace each level. So one thing we don't know, that a lot of people don't know, is how long are you comfortable being in VR? I mean, I spend every day in Tilt Brush, but I'm comfortable with it, but I'm not sure if you've never tried it before, are you going to be okay with keeping a headset on for three hours? So one of the design aspects we were also trying is our level breakage is there's 15-20 minute chunks and then between that we have minor saves so if you want to play for five minutes you can. So we're trying to pace things out and if you want to sit and play through a two to four hour game, You know, you can do that too. So we try to pace it out differently. As far as challenge, Wayward Sky is a light action adventure game, so we don't give you game-stopping puzzles. We want you to enjoy the whole story and see the field experience and see what's interesting about the world. It's also set way above the clouds, so doing verticality things are really fun too. So it's mostly about just how many puzzles are in this level, what's the break in between, you know, with the movement, and where's the story elements. We do sort of an intro set up for each area. Halfway through the level we'll do another story bit that's in cutscene form and at the end we'll give you a payoff, so. It's weird thinking about an elevator as a payoff, but in VR it feels pretty good. It's like you traverse through an area, you get around some enemies, and you get in an elevator, you hit a button, and you get to relax and take in the vista, and we try to preview sort of the next area that's coming up. So those pacing moments in VR where you can enjoy the view and then balance that with, I feel like I'm doing something, but I understand my bigger goal, and that's sort of the things we're trying to solve.
[00:11:19.364] Kent Bye: I see. And with the PlayStation VR, it feels like it's likely going to be like a sit-down experience, and then the Vive, you can do both room-scale and sit-down, and Oculus is sort of in that middle, where in the beginning, it seems like a lot of the Oculus CV1 experiences will be sit-down, because they don't have the touch that's going to be launched, but eventually, in the second quarter, maybe the touch controllers will start to land, and then maybe there'll be good enough tracking to do walking around. As you're designing these VR experiences, what are the constraints that you've found for that sit-down experience with the motion controls?
[00:11:54.650] Chandana Ekanayake: Sure. So for one thing we learned early on is if you're doing sit down, you know, the player heights can vary, chair heights can vary, and it's okay to do a camera reset once a player is ready to go. If you're doing a standing or walking room experience, the floor is always the floor. You don't want to mess with that. You know, you want to go from ground up and build your levels, build your experiences from the ground up, assuming the ground never changes, which is fine. You know, if that's what you want to do, you want to make sure that you do that because as soon as you start messing with the ground height and if the scale is slightly off, a player will feel it. And they won't know why, but they'll feel off. If they try to reach down and grab something off the ground and they end up hitting the controller on the ground, I mean, that's bad for the controller, too. Bad for the player, bad for the game. So those are two big things that we try to maintain. For Wayward Sky, it's running on PlayStation VR. We have a version that runs on the Vive and runs on Oculus as well. We're running in Unity, so we have different sort of setups that, depending on the build you're working on, we try to maintain those rules. Floor, don't change the floor. And then you can reset the camera if you're seated.
[00:12:52.896] Kent Bye: And so you've got Wayward Sky, are there other prototypes or experiments that you've been working on and what have you been finding with VR design?
[00:13:00.480] Chandana Ekanayake: Sure, I have more ideas than I have time for, so we try to do game jams as much as we can. We went down to Austin for a VR game jam in the summertime that Valve was doing with Alchemy Labs. That was fun. For that, I wanted to try out... I'm really fascinated by third-person or God-view into a space, especially since we can scale the player to be a giant looking down in a miniature tabletop. So one of the things we tried out was a Pikmin-type game where you have a laser pointer that you're looking at the world, and we have these robotic cats. The cats run around to wherever you point in the world, and there's resources to dig up. So they go and dig it up and then you have essentially a gravity gun and you can pick up the resources and put her in a little chamber that prints out more cats. So I was like, you know, how fun is that? And we tried that for the Vive. So we put you on a plateau where the bounds are the bounds of the room. So you can walk around, suck up the cats with the gravity gun, throw them in the air and see them fall down, point them to a resource and then they dig it up. And then we had a little boss fight where you do the same thing where you throw the cats onto this monster. and they dig it up. So that was neat. It felt pretty good. And then for traversal, we would essentially concede as your drop pod gets dropped into a world, and then that's the area you can see and interact with. And if you want to go to another place, we bring the pod up into the sky and then drop you back, essentially teleporting you to another spot in the world. So that was fun. For that, we didn't have our Vive kit, so we prototyped with the Hydras. The Hydras are the closest thing to a Vive you could get. I mean, it's way different. The Vive works way better than the Hydras, but we could at least just figure out what to do with the controls. We did a game called Beat Zombie for Game Jam, for LDJam, and that was essentially three days with my VR team, a small team, from scratch. We did essentially a boxing game with zombies and dancing to rhythm, which I'm going to revisit because this is before we had motion controls, and we essentially used an Xbox controller. And the thing that I find with doing, like, I'm surprised there's no boxing games, but I think the issue is When you punch something with a motion controller, or you hit something hard, like a glove, you want to feel the impact. And there's no really good way of feeling the impact of that kind of motion, so it feels off. Like, you know, you hit something with a bat, you should feel some sort of impact. The closest thing I've seen to that, or felt to that, is there's a SteamVR Valve created demo where it's a longbow demo, and as you put an arrow on the bow and pull it back, there's a haptic feedback, which tricks you into thinking that you're actually pulling the, there's some feedback on the strength of the bow. So I'm going to try something more in that line. Punch out in VR sounds awesome, and I would like to try it, but how do you get over the fact that those impacts don't feel good, right? I mean, in traditional games, it's going to be like a little bit of screen shake and audio and rumble, and maybe that's enough. But in VR, it seems like you feel like you're there more, and then not having that impact just feels way different, not as good.
[00:15:52.733] Kent Bye: And what do you think is the ultimate potential of virtual reality and what it might be able to enable?
[00:15:58.827] Chandana Ekanayake: Oh man, so I spent a lot of time in Tilt Brush and I got a chance to play Medium. So for me that is amazingly therapeutic and fun because I'm an artist by trade and being able to just like sit and then just sculpt and do that in VR which feels so natural and be able to walk around the thing and So I really like that aspect. I like to see more creation tools and content creation tools and painting tools and sculpting things. That's really interesting. Job simulator is amazing. And that's just like one of the first demos I show people. My kids love it because they just mess around. Alchemy Labs did such a great job with that. So right now we're seeing a lot of vignette games, you know, smaller scale games, smaller teams, essentially one or two main mechanics and not really full games. What's really interesting to me is when there isn't more of a market you'll see the Ubisofts and the Konamis and the Microsoft's the world jumping in for bigger titles that cost a lot more money and you need a bigger market to make that money back. See what kind of things they try. I think you'll see games like Assassin's Creed that'll have a VR component initially before it's in the full game, just to try things out. And as soon as the market grows, which I think it will, you'll start to see more experiences. There are things VR does do really well right now, as far as we know, and that could change. So the exploration of what that means, I think it's changed over the last two years. You know, what can be done, what's interesting, what's not. So I'm really hopeful for VR's near future and long term, see where that goes. And when AR is ready to be in the mainstream, I'm excited for that too, because a lot of the sort of UX issues we're solving in VR, that translates really well into AR as well. I want to see a wireless headset at some point. That'd be ideal, but we're not there yet.
[00:17:30.788] Kent Bye: You mentioned Tilt Brush a couple of times. Do you find yourself doing prototyping in a 3D environment in a way that using your artistic skills to kind of like flesh out either character design or maybe talk about like how you're actually using Tilt Brush as an artist?
[00:17:44.901] Chandana Ekanayake: Yeah, so Tilt Brush has mostly been a toy for me at this point. I just try different brushes and try different things. We did a contest at PAX. It was like a bunch of six artists from various companies. We had 30 minutes to paint something, and it was really fun. There are tools that I would like. I'm curious what Unity would feel like in that space where you're level editing. And that's something I hope somebody does. Maybe Unity will. Unity, if you're paying attention, that'd be awesome. Like, I want to walk around the space and edit my level and then play it right there.
[00:18:07.555] Kent Bye: Yeah, I did an interview with Timani West, and she's talking about those tools that they actually are developing. So yeah.
[00:18:13.403] Chandana Ekanayake: Awesome to hear. I love that. So Tilt Brush isn't really made to be a productive tool. It's great for sketching stuff out. But I still use my Wacom tablet and stuff in Photoshop for production stuff. But yeah, I'd like to, if it had, if there's a tool that's more like a Unity thing in terms of like simple primitives that you can transform and scale and move around, that'd be great for layout. And to be able to export that super easily, or just run it in Unity, that's fine too.
[00:18:36.852] Kent Bye: Anything else left unsaid that you'd like to say?
[00:18:40.561] Chandana Ekanayake: With all the things we know, the community is great. There's a lot of people sharing stuff on Reddit and I love going to talks and meetups and things like that and game jams. There's a lot we're still learning, almost on a month-to-month basis. So I'd hope for all the big players to keep working together. Until there is a market, I'd love to see more games come out on all the platforms. I think it'll be interesting where things are going to be a year from now. I think it's going to be different in terms of what type of games we see. It's been a pretty accelerated couple years, you know, from super simple passive experiences to really active experiences. I got a chance to play Bullet Train recently, and that was phenomenal. After five minutes, I was comboing, teleporting, and shooting, and reloading, and throwing stuff. Like, that just felt like it had a really good flow. And that's something I haven't seen till that in VR, where it feels like how you would get into a zone playing an FPS or, you know, Counter-Strike or things like that. That was awesome. I hope more people copy that and just do more of that or take that a step further and see what happens.
[00:19:40.874] Kent Bye: Awesome. Well, thank you.
[00:19:42.356] Chandana Ekanayake: Thank you very much.
[00:19:43.798] Kent Bye: And thank you for listening. If you'd like to support the Voices of VR podcast, then please consider becoming a patron at patreon.com slash Voices of VR.
