Last week I was fortunate enough to be in San Francisco for the anual Game Developers Conference. Part of the conference (though not the most important reason for going) includes an expo floor for companies to show off things like engines (Unreal, Unity, etc.) and various other middleware products (Havok physics, Speedtree, etc.). This year one of the booths was dedicated to giving a demo of the Oculus Rift VR (virtual reality) headset. I had heard so much hype from journalists, developers, and technologists that I decided I had to go see it for myself.
After waiting in line at the booth, I finally got to try the headset on, and I was very impressed with a number of things:
Firstly, for something that has the potential to be bulky and cumbersome, the headset was surprisingly light and comfortable to wear. More importantly, it was able to fit over my glasses with no problems whatsoever.
Secondly, the head tracking was very responsive and accurate while looking all round (including behind me and while rolling my head from side to side), which is key to providing a high quality VR experience. The human brain is very good at detecting anomalies in positioning if we are perceiving the images shown as existing in real 3D space. For more reading about the importance of low latency in tracking and updating a VR display, you can check out some (highly technical) articles from John Carmack (ID Software) and Michael Abrash (Valve) that go in depth with the complex set of problems that VR involves.
Lastly, the stereoscopic 3D effect was very convincing without being overwhelming or nauseous. The game being played in the demo was Hawken (a multiplayer mech game currently in development), which features a jetpack on the mechs. The person giving the demo told me to fly as high as I could, look down, and then go into free fall. The images were convincing enough to generate a sense of speed even though I was sitting comfortably in a chair.
Unfortunately, the Oculus Rift features a relatively low resolution panel with an extremely low pixel density. The device uses a 1280×800 panel that is split in half for each eye. In order to achieve the incredible field of view angle that completely covers your peripheral vision, the already low number of pixels are spread apart, creating a “screen door” effect from being able to see the gap in between individual pixels.
Also, as a result, the image became much blurrier than normal with some slight ghosting effects while moving my head but tracking a static object with my eyes. This is due the object snapping to integer pixel coordinates every frame, giving the appearance of the object shaking back and forth from the eye’s perspective.
Problems aside, I do believe that the Oculus Rift is a significant breakthrough in VR technology. The company plans on using a higher resolution display when the consumer version is released (probably at least 1-2 years from now) after they overcome the latency issues with pushing more pixels to the display per frame, so for now I’ll remain cautiously optimistic for the future of VR.
I’m glad to hear good things about the Oculus Rift. I learned about it last semester in my COMM 190 class and it looked really cool, but I was afraid that it had the potential to fail pretty hard. I was also excited to hear that it fit comfortable over your glasses as I wear glasses too. I hope that VR is really the next generation of gaming and not something that will just end up flopping. Once the technology gets here it will create huge potential for new game ideas and mechanics.
As for the downsides you mentioned, I am not surprised, but hopefully as technology catches up, they wont be as noticeable (for example, a higher pixel resolution will hopefully fix the “screen door” effect you mentioned). Like most people, I also hope to get my hands on one of these when it becomes available to the public so I can try it out myself.
I whole-heartedly agree with Samuel’s post. I’m looking forward to more virtual reality technology and can’t wait to see us move on from this cycle of simply improving graphics as a means of better-immersing players into the game. The reality is that you are sitting somewhere staring at a monitor. Until you can replicate real-life exactly and touch, feel, taste, smell, and hear in the environment; I don’t think a player will ever truly be fully “immersed.” Ideally, we won’t experience this sort of immersion for sometime until someone invents some sort of Holodeck as seen in the Star Trek franchise; however, although it is still supremely primitive compared to what could be, Oculus Rift does seem like the logical next step and I can’t wait to try it out.
Another Valve employee, Joe Ludwig, just posted the slides from his GDC talk on lessons learned from porting Team Fortress 2 to VR. They’re less technical than the articles I linked in the original post and go into some of the issues related to making games for a VR headset, especially with regards to UI and input.
I’m really looking forward to this kind of tech. I can’t really complain about the first iteration of real consumer-level VR not being perfect.
After watching some demo footage from the PS4 and some upcoming PC titles (like BF4), I feel like standard graphic fidelity is plateauing in terms of immersion. The visuals look amazing, but no matter how good they look, or how good they’ll ever look, nothing will change the fact that you’re sitting in a chair or on a couch. The next step is actually putting you in the game instead of just giving you control of it through remote devices, which is what VR is designed to do. I watched an interview recently where one of the designers hinted at features they’re still working on, like gloves and other attachments that can put the actions of other parts of your body into the game. How cool would it be to look down and see your own hands in the game world?
All-in-all, I’m hoping the Oculus Rift gets a lot of developer support so that this corner of the market becomes competitive and starts reaching mainstream retailers. I personally can’t wait to try one out.