Virtual Reality is one of those ideas that has been part of the tech folklore for ever, but that haven’t quite taken off. Popularized in the 80s, VR has seen a number of ups and downs, hits and misses, hypes and backlashes — the most recent one being the rise and fall of Second Life. With VR, it’s always been two steps forward, one step back. Suddenly, a programmable, kickstarter-funded $300 headset falls in our laps from out of nowhere, and… it’s another two or four steps forward!
All this time, the consumer-grade VR hasn’t really been true VR: it’s simply 3D environments displayed on 2D screens. This is fine, to some extent; those environments are becoming incredibly realistic, so when the action is intense, we let ourselves believe that we are in those environments, even though, of course, it’s just a computer screen and the real world around us is right there too. True VR, with goggles that block the real world while tricking the brain into a simulated 3D world, has been confined to very specialized niches, such as the military, experimental medical treatments, and research labs. The hardware has been very expensive.
Things have suddenly changed last year, when a kid in Southern California, a fan of both games and VR, who had collected and reverse engineered a large number of vintage VR paraphernalia over the years, put together a headset with commodity hardware resulting in a price point lower than smart phones. With the kickstarter campaign, Oculus VR was able to raise enough money to develop a usable prototype and a relatively simple SDK, gifting all game developers and gamers with an affordable path into immersive VR.
The response has been intense. Just about every game studio out there is experimenting with the Rift, and many have been distributing Rift-enabled versions of their games. The OpenSim/SL community is no exception. At least a couple of people started to adapt SL viewers for the Oculus Rift. There are rumors that Linden Lab is also working on Oculus support. So, encouraged by these news, about 1 month ago, I finally got my Rift headset. I was prepared to be under-impressed, but… I’m in awe! Having worked in and with OpenSim for 5 years now, I obviously have fantasized about having access to hardware that would let me be truly immersed in my vLab and other spaces like it. I just didn’t expect it to happen in my lifetime… And yet, there I was yesterday, having a real meeting in my vLab! (picture above) The Oculus Rift is every bit of what it promises to be!
Don’t get me wrong: it’s far from perfect! In fact, there are many problems to be solved, some of them quite serious.
First, the resolution of the headset that they are currently selling is pretty crappy. You see the screen door effect quite clearly. Thankfully, this is the least of all the problems, and it will go away soon: Oculus VR already has a high-def prototype in the works.
More importantly, simulator sickness is a real problem that affects everyone, some people more than others. I certainly can feel it, especially when certain movements happen. A couple of days ago, Steve LaValle, a Principal Scientist at Oculus VR, gave a talk here at UC Irvine where he explained, among other things, some of the underpinnings of simulation sickness, and some techniques that Oculus and game developers can use to minimize it. I was impressed to hear that they recently hired two experimental psychologists specializing in visual perception whose main role is to try to figure out how to minimize that effect. Let’s just hope that this incredible push towards commodity VR hardware will energize work in the area of psychophysics!
Third, a new, interesting set of problems arises related to input and UI. Once you put the goggles on, the real world disappears, meaning keyboard and mouse are mostly useless peripherals. With immersive VR, we need a whole new user interface paradigm that still doesn’t exist as such. Parts and bits exist — Microsoft Kinect and the Leap Motion are examples of input hardware that can be used in VR, and many people are already on it — including the OpenSim/SL viewers that target the Oculus Rift. But for the most part, everything is still unexplored territory. Even though research in these kinds of interfaces has existed for years, there hasn’t been any pressing urgency to use them, so no one knows yet what are the best “idioms” and metaphors for gesture interfaces. But it’s not just gestures. These virtual environments exist to be visited, walked, run, flown. We need ways to use our entire bodies as controls, including walking, crouching, etc. Recently, another Kickstarter project got funded to develop a walking base as a complement to the Oculus Rift. I suspect a whole new breed of peripherals will emerge within the next couple of years. Again, unexplored territory!
All of these problems (i.e. the psychophysical effects and the UI) are both serious and seriously interesting! They are the kinds of problems that become important only when a new door of exploration is suddenly cracked open for everyone to see what’s beyond. That’s what the Oculus Rift has done. I have no doubt that once the novelty washes out, there will be steps back here too. But what an incredible feeling it is being able to go to places that are out of this world and to do it together with people who are half way across the globe!