Alex Schwartz is the Chief Scientist and Founder of Owlchemy Labs, and he talks about the process of developing Job Simulator with Valve’s SteamVR and the HTC Vive. They wanted to create a series of mini games like the WarioWare of VR, and he shares a bit more of the backstory for Job Simulator. They have a number of different experiences planned beyond the chef simulator and bartender that include engaging two-handed interactions.
Some things that didn’t work include locomoting the player without them walking around, and so they had to design the spaces so that they could walk around in them. They had to figure out how to grab objects, and simulate reality like object penetration without feeling like it’s fake, the downfalls of infinite chopping, and adding conservation of angular momentum to objects that are thrown.
I asked about the Chaperone System, but he couldn’t talk about more details about that yet. Some of the other open questions are how to adapatively scale the room size, but also account for the ergonomic and structural design problems that come up in walkable VR experiences. The goal is that it should be fun in a wide variety of spaces.
Alex said that the experience and crunch leading up to GDC felt like a war with all of the API updates, iterations, and fixing various bugs. It was like “a band of brothers coming out of the other side coming out of a traumatic experience”, but that at the same time it was the coolest thing that he’s ever done in his development career.
Alex also helped to organize a positional tracking VR game jam, and Valve came out to preview some of their hardware prototypes.
He’s very grateful to have been a part of the process of developing a demo for GDC, and he sees the potential for VR to change every industry and in the end change the world.
Theme music: “Fatality” by Tigoolio
Subscribe to the Voices of VR podcast.
Rough Transcript
[00:00:05.412] Kent Bye: The Voices of VR Podcast.
[00:00:12.070] Alex Schwartz: My name is Alex Schwartz. I'm the chief scientist, founder, CEO, and janitor of Alchemy Labs. I started the company about four years ago. We do independent games. We have four games on Steam, a number of titles, and the one that we became notable for in VR space was called Oculus, which we then brought to Google Cardboard and called it Cardboard. You can see a pattern there, right? And so we've been close partners with Valve when it comes to bringing content to the platform. We ran that Boston VR Jam back last year where they brought out their positional track headsets. And so dot dot dot, that led to the point where basically we were contacted out of the blue. And they said, hey, we want to bring you in on a little secret VR thing we're doing. So we flew out, that was about five months ago. and they told us about HTC and the Valve initiative for the headset and we basically came up with and started working on Job Simulator as a new IP from the ground up for the next generation of VR hardware with basically us developing on the Vive headsets from the ground up.
[00:01:17.893] Kent Bye: And so tell me a bit about, you know, that process of trying to figure out the gameplay and the mechanics involved with, you know, having fully positioned tracked hands and then how you sort of ended up with doing the Job Simulator.
[00:01:29.623] Alex Schwartz: So basically what happened was we realized that in 2015, which we're assuming that there's going to be a number of consumer VR things coming out by the end of the year or around the end of the year. First off, getting a deep multi-hour experience in VR requires a lot of content and a lot of the demos that have been seen now are tech demos. And so we were thinking about how can we make a fully-fledged product that we could charge a premium price for. And the idea of doing something that was similar to the WarioWare of VR came up. And we really caught on to this idea of doing a series of minigames that were tied together via some thematic element. And the more we started thinking about what would be really fun to do when it came to two-hand manipulation, which is, I think, the most groundbreaking part of the new VR headset is to have tracked hands in an absolute space. the door is blown wide open when you can just pick up things, manipulate. So we thought about, you know, these various mini-games that we wanted to do, and all of them seem to have some semblance of a job associated with them. So that's how Job Simulator was born. We came up with the story, I don't know if you've seen the pitch, but basically, in the future, in 2050, robots have taken most of our jobs, and so, you know, they've automated it for the better of humanity. So in order to make sure that our children and our children's children understand what it was like to have these professions around and now that they're gone, there was this job simulator initiative was born. And so you can go and take these antiquated job tapes and put them in the mouth of the job bot and he brings you back to this simulation of these various jobs. The first of which we've created for the GDC demo, which was a gourmet chef or a sous chef role. Now, in there you can chop vegetables, you can make soup, you can stack sandwiches. But the other part of the concept is that the robots kind of got their history wrong a bit. So it's kind of like a skewed lens on the past. So things are kind of off in a kind of hilarious way. So that's the game.
[00:03:28.194] Kent Bye: Nice, and so did you continue to build out it as a minigame? So you have multiple scenes, so there's other things that you're doing other than just, you know, being a chef in a kitchen?
[00:03:37.283] Alex Schwartz: Oh yeah, yeah. We wanted to make sure that people didn't think we were building the chef game. Our first demo that we wanted to make sure it had a lot of great interaction and depth to was building that kitchen, so we really focused on that. We have a number of really, really fun experiences planned. I don't want to spoil too much, but one of them is fairly similar but a little different. We have a bartender simulator where you have this very slippery bar and you're trying to fill orders really quickly. So you've got your tap and you're pouring bottles in here and there's people waiting for orders and you're sliding beers down the table and things are crashing and breaking and it's kind of this frenetic action madness. So that's one of them. And then I think we'll stray from this, you know, table in front of you concept for a number of the jobs. And there's a lot of really crazy things we're thinking about. So yeah, expect to have a number of various jobs in Job Simulator, not just cooking.
[00:04:23.871] Kent Bye: And so what were some of the things that you tried that just really didn't work at all with having two-hand interactions?
[00:04:30.400] Alex Schwartz: That's a good question. Let's see, things that did not work at all. Well, in general, it doesn't have to do with hands. Locomoting the player without them actually walking around was terrible. So we created our spaces in a way that you can just walk around and that's kind of the magic of the new Lighthouse tracking mechanism is that you just have your room and you can walk around in it. So, in our kitchen for the GDC demo, you walk over to the fridge, you take things out of it, you put the items on the plate, which is on the opposite side of the room, and you traversing feels so good. But when you're, you know, holding forward on a controller and it moves you instead of you moving you, that's where it all falls apart. Let's see, other things, there was, I mean, there's a ton of experimentation about how to grip objects. how to make it so that you don't feel like you have this weird Superman hand that can penetrate any object. So like if you're holding a knife and then you smack it on the table, the knife actually breaks out of your hand. It breaks your grip like it would instead of sending the knife directly through the table with no consequence. object penetration through solid other objects is a really complex thing to solve and we spent a lot of time trying to make it feel right. The right amount of realism with the right amount of silliness and playfulness and let's see other things that didn't work. We ended up making it so you can infinitely chop any item as many times as you wanted with no consequence and it basically made more and more and more objects in the scene and bogged you down to Basically getting a low frame rate which makes you feel sick. So we had to we had to remove that you know the concept of just infinite chopping. I mean we had a number of Troublesome interactions with like joints and hinge joints and just took a long time to get it, right? Throwing we didn't initially have rotation on the throwing So it would just kind of stay straight in the air as you threw it and realize that that would never work. And so we needed to write this convoluted conservation of angular momentum formula to make it work and went back to textbooks and figuring out how to do it. So there was some math that, you know, they told me back in school games would have math. And I was like, yeah, yeah, yeah. And then, crap. We had to look that up. But yeah, in general, we're so proud of the way that it came out in the end. People are raving about it. And yeah, it's awesome.
[00:06:41.520] Kent Bye: What was the the lighthouse? How does it determine how close you are to the walls and figure out like what sort of the UI elements that make sure that you just don't like run into the wall?
[00:06:51.527] Alex Schwartz: They haven't announced the whole system that protects you from hitting walls. So I can't really talk about that It's safe to say that valve will not allow you in the consumer version to just slam your face into a wall without Properly letting you know that you're about to do that
[00:07:07.178] Kent Bye: I see, okay. Yeah, because I heard that there was some elegant sort of grid systems that, you know, would warn you. And I guess, you know, in terms of like, you know, how do you design a game when you don't know, you know, is there a minimum size that you need in order to have the lighthouse? And then how do you adapt the room based upon that? Can you make it so that you design your level so it's adaptively altered so that it fits the room?
[00:07:29.135] Alex Schwartz: I mean, it's not even just the room size. What if you have someone who is four foot tall playing the game and then we have a table that's three and a half feet tall and they have to interact with items on it? Are we gonna make it so that someone who's eight feet tall also at the table rises up? You know, there's so many ergonomic, structural, level design problems in VR that, honestly, we're there right with Valve figuring out how the hell this is gonna work. And so, I guess we'll see. Maybe they'll be min-spec. I can imagine a future where the back of the box says, hey, you need a 4x6 space to play this game. That's not out of the question, you know, that you can't play it in a phone booth that's 2x2. I don't know how they're going to solve it, but it's definitely something to think about is that the game should be fun in a wide variety of spaces, and you're going to have obstructions like couches and things like that. So they're certainly working on it, and they really know their shit. So I believe that they will solve these problems.
[00:08:23.553] Kent Bye: And what was it like to collaborate with all the other developers who were also, you know, in the midst of trying to figure this out and, you know, sharing information and feedback with each other?
[00:08:33.059] Alex Schwartz: Well, you know, we were in communication with both Valve and the other people working on the software. And honestly, by the end of this, you know, with months of working on it and iterations and API updates and having to fix this and that for GDC and the crunch leading up to GDC, I feel like we all went through a war together. The band of brothers coming out on the other side, having lived through a traumatic experience. I saw Chet and some of the other guys who made content at GDC for the first time in person in months. We hugged and it was understood that, holy crap, we did it. We came out the other side alive and didn't die. So I guess that accurately describes our process. It was the coolest thing I've ever done, development-wise, in my career.
[00:09:16.858] Kent Bye: Wow, and so it sounded like, talking to Denny, that Valve was kind of identifying skills and qualities that these different development shops were having. What was Alchemy Labs bringing to the table?
[00:09:27.426] Alex Schwartz: We were super good at chopping objects on a table, so we were brought in day one. No, no. Honestly, I feel we're really lucky to be invited. We're super, super lucky to be invited by our industry heroes, Valve, to come work on some early stuff. So we obviously did something right to catch their eye. And I think, like I mentioned, doing Ah! early and being there right at the beginning and working on that. positional track VR jam and just being close partners with Valve and proving that we, not like we knew more than anyone else about how to fix these problems, but it's trial and error. Everybody's figuring it out together. So we had done a lot of that trial and a lot of that error and had kind of figured out what could and could not work. So I think anyone with, you know, one to two years of VR experience is like, At this point, it's such early days that you're a veteran if you've been working on VR stuff for that long, so it's really inspiring to be in a room full of people who have this much knowledge about VR. Like, the smartest people in the world are working on VR right now, and it's really exciting from a high level just to think about the fact that 30 years from now, this will be a moment where, you know, VR is starting to come mainstream, and we're there making little early fun stuff for the first headsets that came to consumers.
[00:10:42.842] Kent Bye: So this positional tracked game jam that you had out in Boston, was this catalyzed because of you knew this stuff was coming from Valve, or was this something that you had on your own initiative, or how was it that Valve ended up coming out and kind of giving a sneak peek of some of their hardware?
[00:10:56.148] Alex Schwartz: Just good relationship and saying, hey, do you want to do this and see it happen? I think, you know, at that point, no one had any concept, at least on our end and publicly what they were saying, that this was ever going to be a consumerized thing. They'd been doing VR research for years and years. altruistically trying to make VR a thing and make it happen. And just knowing some of those folks, Doug and Chet and others, and just inviting them out and saying, hey, this could be cool. And them saying yes, that's really indicative of our relationship with Valve. They're really open to trying new things. They're really great partners. Everything we've done working with them, it's been an awesome experience. So yeah, that's kind of how it went out.
[00:11:33.652] Kent Bye: And so what do you want to experience in virtual reality?
[00:11:37.135] Alex Schwartz: Oh, loaded question. We don't know yet. If I knew what my killer app was for VR, I'd be building it right now. So I'm waiting to be surprised by the billion dollar VR experience.
[00:11:49.206] Kent Bye: And finally, what do you think is the ultimate potential for what virtual reality could enable?
[00:11:55.332] Alex Schwartz: I mean, it'll revolutionize every industry. Period. I have a lot of first-time people do a VR experience, and I get the privilege of watching someone come out of that experience. And, you know, first is lots of swearing, and then lots of starry-eyed looking around, and then begins the brain dump. People start going, oh my god, what would this do to education? Oh my god, what would this do to travel? And how would this change medicine and simulation, you know? every single industry will be affected positively by this stuff. So, you know, you're getting into, like, soundbite quotes of, like, futurists. It'll change the world. But it will. I think it'll change the world. Great. Well, thank you. Thanks for having me.
