Survios has been toiling away in virtual reality since 2013. The Los Angeles company has seen the hype cycle around VR since before Oculus was acquired by Facebook, and it has persevered through the slower-than-expected growth of VR in the past couple of years.

Other companies have pivoted into video games or enterprise VR, but Survios has stayed true to its roots in consumer VR games. And it has had strong titles such as Raw Data, Creed: Rise to Glory, and the newly launched Battlewake. It also teamed up with HBO to create Westworld Awakening, it launched a VR arcade, and its games are now played in 2,600 VR arcade locations around the world.

I met James Iliff, cofounder and creative director at Survios, in the early days of Survios, when it was showing off a demo called Zombies on the Holodeck. I recently interviewed him onstage at the Greenlight Insights VR/AR/XR strategy conference in San Francisco.

Here’s an edited transcript of our interview.

GamesBeat: We go way back.

James Iliff: 2012, 2013? Some of the stuff was already out there.

GamesBeat: Yeah, with your Zombies on the Holodeck demo. Can you talk more about your background?

Iliff: Certainly. I’m the chief creative officer at Survios. We’ve been around since about 2013. We focus on making software for virtual reality experiences, first for home retail environments, and now we’re building a lot of stuff for arcades, which is a big new business for us. We’ve brought about seven products to market. The latest one is Battlewake, a pirate ship game, as well as Westworld, a game we did in conjunction with HBO. That’s more of a story-driven, linear campaign.

We have a wide range in our portfolio of games. We like to do different genres where we can, different types of mechanics. A big piece of my work I focus on is working with the engineering team to create new interaction technologies, new UX technologies, so every one of our products has a new piece of tech that pushes the industry forward on the software side. For example, we had a racing game with a new locomotion method. With our latest game, Battlewake, we had a new vehicular locomotion method. There’s a lot of different things we like to do to expand the industry where we can.

GamesBeat: You have your own VR arcade now. What do you call it?

Iliff: It’s the Survios Virtual Reality Arcade in Torrance. It’s our testbed.

GamesBeat: Was that always part of the plan, or did that represent a late-breaking pivot, doing a location?

Iliff: It certainly wasn’t at the beginning. I think probably around 2016, 2017 we started looking at the market and realizing that although home users–there’s a large retail market and steady, slow growth, but the arcade market, when we really caught wind of it and paid attention and started researching, was a smaller market, but it was having exponential growth. It’s growing much faster than retail.

This is around the time where we wanted to start making games that functioned well in arcades. We were building games for home users that had very advanced mechanics and very layered mechanics. That’s fantastic for someone playing at home who wants to get a lot of hours out of a game, but it’s not necessarily something that a new user in an arcade can pick up and play very easily.

We started to adjust our thesis and started to build some of our content for arcades first. We knew we needed a testbed to actually make sure these games are what users want, so that’s what we created. We opened up an arcade that’s around 2,000 square feet in Torrance, at the Del Amo Fashion Center, which was a great place for us demographically and geographically.

GamesBeat: There’s a lot of different kinds of VR arcades out there right now. What was the design philosophy for what you wanted to do?

Iliff: With this one, we wanted to establish a strong foundation, a strong baseline for research. We could funnel all our games through, get user testing done, understand the users in a natural arcade market. This is in what are essentially bays with full motion VR hardware. We’re starting to explore free-roaming as well. A lot of arcades are doing that, and also larger, fully immersive experiences, like what The Void is doing. For us, those aren’t standard technology stacks. We wanted to focus on one that was a standard set we could put our games through, get feedback, and ultimately optimize our games so we could sell in the arcade market as well as at home.

GamesBeat: You’re in about 2,500 different VR locations now, right?

Iliff: The one in Torrance is our one owned and operated location. We also have about 500 direct relationships with arcades here in the U.S., and we have about 2,500 arcades in a network in China through our joint venture with NetEase. We’ve also been opening up other joint ventures in Japan and Korea.

GamesBeat: I didn’t know there were that many places in China. Is that everywhere you can be?

Iliff: Essentially yes. Every major city, every corner of every city. It’s a lot. Part of that–it seems weird to us. There’s, what, 500 Chuck-E-Cheeses in the U.S., but there are thousands of these iCafes, PC bangs and whatnot on the other side of the pond. What’s interesting is they never had the console generations. They never had that home user focus. A lot of their market — teens and millennials over there — they’re very accustomed to going to internet cafes or PC cafes and playing their games there, in an arcade format. Because consoles never existed, really, until just now — obviously the rules have changed recently — this has just been growing ever since. They have a lot of arcades.

GamesBeat: Running people through gets you all this information, but if you’re in 2,500 locations elsewhere–can you extract the kind of information you really need only from what the owned and operated location elsewhere gets you?

Iliff: We have a lot more fine tuned control over the owned and operated one. We can change layout. We can change how customers interact with the hardware and how onboarding works. We still get a lot of customer data, demographics data, from all of our arcades. We learned a lot about how birthday parties are a really big thing, and corporate events. Building a game that has multiplayer components to it, that allows you to have a group of 10 people playing a game together. That’s a huge thing.

GamesBeat: Can you get there with Battlewake, then?

Iliff: Exactly. Battlewake was one–that’s a game we made for arcades first. We did sell it at retail, but the focus was selling it into arcades, and it’s doing really well. We have several multiplayer modes. We have a four-person co-op mode where you can go through the campaign, and we also have an infinitely replayable proceduralized mission with four players. We also have a 10-player deathmatch mode called Plunder. You can have 10 people on the open seas, all in different boats, all battling it out with different crazy weapons. That’s a lot of fun for parties.

GamesBeat: Is it turning out to be a game that people will go back to?

Iliff: From what we’ve seen so far, yeah. There’s enough varied mechanics, enough different boss battles, enough different weapons. With this game, there’s a lot of stuff online about it, but the crux of it was we wanted to streamline the experience of piloting a ship without making you nauseous. We actually used some smoothing algorithms to make it more comfortable when you’re sailing on the ship and wrecking into other objects.

Anywhere you point–I can point forward and it’ll shoot the forward guns. It could be machineguns or whatever. I could aim left and it’s going to shoot cannons. I can aim left and it’ll shoot a mortar or flamethrower. You don’t have to pick up interfaces or look at different buttons and all these things to choose what you’re shooting at any given time. You just aim and it fires a different weapon based on where you’re aiming. That was an example of a streamlined user experience that was really vital for arcades.

GamesBeat: If you play any sailing game on PCs or consoles, you can’t learn it in just one session, right? How did you think about enabling people to learn this right away?

Iliff: Thinking about it from the ground up, making it as streamlined as possible. For our more legacy content, like our big action combat game Raw Data, that wasn’t originally built for arcades. That was made for the home. But we wanted to sell it into arcades, so we had to do a variety of changes.

This simplifies the difference in the customer set, but it’s a very different customer set. The one at home is a hardcore gamer. They’ve spent a lot of money on a PC gaming rig. They’ve bought a VR rig. They’re playing games, and they want core titles. The person walking into an arcade is likely a casual gamer, or even a non-gamer. How do you handle that?

For example, with Raw Data, although we had this very long profile progression across many missions for home users, where you can get upgrades and upgrade your character over time and get special abilities, instead we switched that to a match-based progression that we automatically rewarded to you in arcades throughout the course of a 20-minute match. I’m playing a match, fighting a bunch of robots, and within 2 minutes I automatically get my next upgrade. My character’s leveling up constantly. You’re slowly learning how to play the depth of the game over a short period of time because it’s introducing it to you in an easy way, instead of having to go through the longer, denser retail experience, where you have longer profile progression.

That’s one example of streamlining. Also the direct user experience, making it easier to pick things up and interact with items, just making some assumptions about what 90 percent of the users want to do. You have less optionality, but more focus.

GamesBeat: As far as the revenue side of things, what would you be willing to say about what you’ve learned from when revenue comes in? What are some lessons for people about that?

Iliff: We’re seeing right now that the arcades are growing at about three times the rate that we’re seeing in the retail market. Depending on how many partnerships you can have, that can become pretty profitable fairly quickly. We’re lucky enough to be able to expand rapidly, especially with our joint ventures. We’re not selling onto distribution platforms. We have direct licensing agreements.

At retail, the best thing we can say is to keep your budget in a certain place, where you can make your money back and then some. We’ve seen a lot of folks overbudget their titles. We try to hit somewhere around the $3-4 million range for our own original IP. If we’re doing licensed IP or work for hire, we can go much higher than that of course. But one thing that’s changing now is that I’m starting to see some companies go into traditional.

This is an interesting aspect of the industry that’s just now emerging. There are some examples of this, like Payday 2 VR. That’s a game that was traditional, but now has a VR expansion to it. Some folks are building games up for VR and for traditional at the same time. What this allows you to do is, if you’re selling to to the traditional flat screen market, you can have bigger budgets, because your revenue is going to be much larger and you can support unique, interesting, in-depth VR features that you otherwise wouldn’t be able to make if it was strictly a VR title.

GamesBeat: How much is each person paying per session in arcades?

Iliff: That varies widely. It totally depends on the location and the partner. For us, we’ve done a lot of different pricing tactics. It changes every couple of months to test different things. Right now it’s around $20 for a half hour and $30 for an hour to play whatever you want. We just do it by time. Some folks have had success doing it by title, although that’s not what we’ve found for ourselves.

GamesBeat: Would you rather have a popular game in 3,000 arcades, or however many units sold at home? How would you balance that question?

Iliff: Honestly, the real golden goose is going to be the game that can do both. A game that can crush it at retail, that core users love, and also one that is streamlined and easy to use enough so that Grandma could play it at an arcade. If you can magically make that happen, there’s very few examples of this happening. We’ve seen it with Beat Saber and Super Hot and some others that have accomplished this.

You start to see a few aspects–this is going to be superficial, because this isn’t a substitute for good game design, but you start to see a few external aspects that are in common. For example, the game is often 180 degrees facing one direction. You’re not looking all around. You’re not exploring a larger world. You’re facing one way. Think about Super Hot and Beat Saber and how that works. For the most part, it’s all facing one way. Also, you’re not locomoting around. Things are coming in to you. The game comes to you. It’s also invading your personal space, so you get really strong depth perception up close. You have a more visceral experience.

I could go on. There’s a lot of aspects to these types of games that work great in arcades and also work well at home. The danger you run into, though, is making a product that’s lukewarm in both markets. It’s really important to focus. If you can find something that works great in both markets, awesome. But we highly recommend, at least with your first title, starting out and choosing one market or the other and drilling down on that.