The difficulty of making a successful game for the virtual reality was underlined by Oculus VR's Matt Conte in an informative session at the Reboot Develop conference. The size of the market in 2018 is smaller than its key stakeholders had hoped, he said, but developers must focus solely on its users to create a competitive produce.
For much of his talk, Conte - Oculus VR's head of development engineering - went through a string of points about effective VR design. Solutions for locomotion, restrictions on play space among users; all important things to consider, though a lot will have been familiar to longstanding VR developers in the audience.
One thing that does appear to be shifting is the thinking around what it takes to succeed in the market. Given that Oculus originally shipped its high-end headset, the Oculus Rift, with an Xbox One controller in the box, Conte's near-definitive stance on the need for focus on VR alone was striking.
"Lucky's Tale, which was an experiment in making a third-person platformer... nobody's making games like that any more"
"You want to build exclusively for VR," he said. "There are some exceptions to this rule, but they are very few. Your title should be built so that your users can do something that they can't do in any other game."
In that respect, motion controllers are almost as important as the headset itself. Conte cited Ready at Dawn's Echo Arena as an example of how hand presence - which is only possible with motion controllers like Oculus Touch - allows entirely unique experiences to be created. The sensation of grabbing an opponent's head while floating in zero-gravity to throw an object into the goal "makes perfect sense in VR", where it would not on any other gaming platform.
Ensuring hand presence is now a "must do" for any VR developer, he said, and that idea is not always been so plainly stated. PlayStation VR, for example, does not come with motion controllers as standard, while Oculus itself put a third-person platformer front and centre of its early content catalogue.
"Most people are gravitating towards first-person experiences," Conte told the crowd. "Lucky's Tale, which was an experiment in making a third-person platformer... nobody's making games like that any more. As charming as that was, most people want to put on the headset and just be in a different place.
Conte added: "The difference in engagement between motion controls and gamepad games is night and day. We don't even put the pad in the box any more. Touch controllers are standard with the Oculus Rift... There's a pretty good degree of compatibility between motion controllers on all the major platforms, so you can make something cross-platform that works."
"I'm not going to mince words: VR is still small. There's not as many headsets out there as we thought there might be a couple of years ago"
The importance of developing exclusively for VR was counterbalanced by Conte's insistence that developers should build their games to work on every available platform. Oculus has been criticised in the past for its stance on exclusives, folding periods of timed exclusivity to the Oculus Store in with the funding deals it strikes with developers. However, Conte's talk indicated that the company is not as interested in the concept as before.
"You want to reach the broadest audience," he said. "I'm not going to mince words: VR is still small. There's not as many headsets out there as we thought there might be a couple of years ago. It's growing, and it's actually growing at a pretty decent pace, but every decision that you make you should be thinking about: How does this get my title into the most users' hands as possible?
"Some people find it weird that we tell them to ship on all platforms. We don't want exclusivity. We want VR to thrive. But VR is a niche, and you don't want to be a niche within a niche. Ship everywhere: Oculus, PlayStation VR, Steam, mobile, if you can. Do whatever you can to get as many eyes on it as possible."
Perhaps the biggest barrier to shipping on every platform is mobile. The differences between mobile and PC VR surfaced again and again in Conte's talk, from the details of the hardware to the behaviour and expectations of the audience.
"I think a couple of years ago, when I started in VR, the prevailing wisdom was to keep your levels to eight to fifteen minutes long," he said. "We have data that disproves that. Play sessions that last hours are not uncommon. You should build content that you think people will stay in the headset for. When I use VR I stay in for a couple of hours. I think that's generally what we're seeing these days. Don't nerf your level designs or the size of the content."
That advice largely applies to PC games, however, and Conte indicated that not only does mobile VR have a host of different problems, one of the very biggest still hasn't been solved yet.
"Mobile customers look very different to PC customers," he said. "PC customers are a classic gamer-base. Mobile is a bit more challenging. Free-to-play doesn't really work - or it hasn't been shown to work just yet - and most people don't want to pay premium prices."
"Mobile is a bit more challenging. Free-to-play hasn't been shown to work just yet, and most people don't want to pay premium prices"
Ultimately, VR game developers in 2018 are dealing with a smaller addressable market than Oculus and its parent company Facebook expected, and yet content really ought to be made for VR alone. Similarly, a VR product must be shipped on every existing platform, even though the platform with the most headsets, mobile, has yet to land on a working business model. If there was one consolation in this complex business picture, it is that brands and existing IP don't hold the same power in VR as they do on more traditional platforms.
"Established titles and brands are non-critical to success," Conte said. "You don't need an IP to excite the VR audience. Superhot wasn't well known, until [Superhot Team] created themselves as the biggest VR IP. This is great for indie developers and small teams."
Conte also left the crowd with some advice on the kinds of products that are most likely to succeed in the market today. "The day of the zombie shooter has passed, for sure," he said to a chorus of relieved laughter, though he noted that the enduring appeal of Epic's Robo Recall suggested that "action shooters" in general may still be a smart bet. What the VR market lacks, however, is something very different.
"Everybody has been saying that a tabletop, real-time strategy game will be the killer app, but it hasn't happened yet," he said. "But Echo Arena, and the Lone Echo franchise - competitive multiplayer online gaming - I think that's the biggest open space right now.
"Social presence is hugely important. The problem right now is concurrency. Going into the lobby of a multiplayer game is a sort of frustrating experience on a bunch of games, because there may only be a small number of people who are actually playing that game at any one time.
"Being in VR with someone is way better than being isolated - for a lot of reasons, a lot of experiences... Right now the limiting factor is the number of people in VR."