During the DICE Summit in Las Vegas last month, Oculus VR VP Jason Rubin said that in his eyes, the future of VR game design isn't all about hand controllers like Oculus' own Touch; gamepads have their place, and are here to stay.
That's a bit of an intriguing position to take, given how popular controllers like Touch and the HTC Vive's wand controllers have proven among VR game devs and players.
For example, the first VR game to top the Steam charts, Survios' Raw Data, isn't designed for a gamepad; the winner of the Best VR/AR award at this year's Game Developers Choice Awards, Owlchemy Labs' Job Simulator, is also a game built to be played with hand controllers. Oculus' own Brendan Iribe went hard on the pitch for the Touch controllers' hand- and finger-tracking capabilities, telling Gamasutra in 2015 that "this is the right path for VR input."
At GDC this week Gamasutra sat down for a chat with Rubin, and he opened up a bit about what he thinks the future of VR game input will be -- and why devs shouldn't overlook gamepad controls for certain types of VR games.
"Here's what I think: we don't know what the ultimate input device is for VR. And it's likely there are a lot of ultimate input devices, just like there are in the real world," said Rubin.
"I believe that the magic of having your hands in VR is incredibly important for new users, and will be important forever. But there will be times, after VR matures, where people come home and go 'I just came home from eight hours of work, and right now what I want to do is this [leans back in his chair and mimes holding a gamepad in his lap.]"
It's a question of comfort. Rubin believes VR games that ask players to stand and wave their arms around are appealing, but that appeal wanes when people are comfortable enough spending time in VR that they're willing to pop in even when they're tired and play for long stretches. Right now, this is something most developers aren't building for; Rubin believes that when people get to a point where they're playing VR games regularly for long stretches, they'll appreciate the option to relax and just play VR games with a gamepad.
"Right now it kind of seems like that's old-school, but it's old-school with forty years of evolution to get the perfect gamepad controller," said Rubin. "There's a reason it's shaped that way. And I believe that over the long run, people will go back to that. Steering wheels aren't going anywhere. Spaceship controls aren't going anywhere."
Rubin is also convinced that third-person character action games work in VR, and that they work best when you design them around a gamepad rather than gesture controls. As an example he trots out Playful's Lucky's Tale, a VR game "basically made for a controller", and suggests that 30+ years of evolution have ensured the gamepad is more precise in some ways than any hand controller.
"When I designed Crash [Bandicoot], we were given Sony's controller that they were releasing with the PlayStation. It was a D-Pad; it didn't have an analog stick," says Rubin. "Most people forget that the PlayStation did not launch with an analog stick. And the reason that Crash, which was our first 3D character action game, went in, out, left and right, was because we had a D-Pad."
He contrasts that with Nintendo's Shigeru Miyamoto, who he believes ensured that the Nintendo 64 gamepad included an analog stick to more precisely replicate 3D movement in Mario 64.
"I thought Crash was a fantastic game, but ultimately Miyamoto's controller changed the way that input in character action games would work, going forward," says Rubin. "And so that kind of game, sitting back and controlling a character in third-person, really is married extremely closely to that analog stick. And I'm not sure that Touch makes that any better. And the precision that you get, even though we've put two sticks, one on each hand, the precision that you get without your hands joined together, it just doesn't work as well. The connection and stability of holding one thing in your two hands, that's what gives you precision. So when those things come together, I think you end up with a controller, a gamepad controller, and not a Touch controller."
While Rubin is careful to say that "we haven't even scratched the surface of what Touch can do," he's confident in recommending devs don't discount gamepads for VR games -- especially if you're at all interested in trying to do a third-person VR game.
Incidentally, he had one other piece of advice for VR game devs in 2017: don't go all in right away.
"Enter slowly. Prototype. Look at everything that's out there. Trial and error," says Rubin. "Too many developers just jump in and say 'here's my idea, I'm going to go with it', never having developed in VR before. And they don't realize it's not a good idea until they get to the end of it."