Rumors are swirling that Google is developing a truly-standalone VR headset — one that doesn’t require either a smartphone like Cardboard and Gear VR, or a tethered PC, like Oculus and Vive. If it succeeds, the device will be more akin to Microsoft’s upcoming HoloLens AR headset in providing an integrated, untethered solution for enthusiasts willing to spend the money on a separate, complete device.
Google certainly has as much or more experience as any of its competitors with all the pieces needed to make this happen. The troubled Google Glass project provided valuable experience in what works and what doesn’t in head-mounted wearables, as well as in how to minimize power, weight, and size of a mobile computer. Google Cardboard’s 5 million units shipped dwarfs the sum total of every other VR device ever — giving Google a natural advantage in getting content developers on board, and a user base that may be interested in trading up. Google’s Tango project demonstrates real-time SLAM (simultaneous location and mapping) in a mobile form factor, and it isworking with Lenovo and Qualcomm to bring that to market in a consumer phone.
Breaking through the power bottleneck
One of the biggest remaining issues for Google or anyone else looking to build a mobile solution for AR, VR, or vision-intensive applications is power consumption. GPUs are much better than CPUs at chewing through the computations required, but usually at the expense of consuming great amounts of power. So, for now, custom silicon looks to be part of the solution. Google has already announced that it will be partnering with startup Movidius to incorporate its custom vision processing and machine learning chips in future phones. So it would make perfect sense that Google would also use those chips in a VR headset — although so far both companies aren’t willing to comment publicly on the speculation. Microsoft is addressing this issue on its own so far, with what it calls a custom Holographic Processing Unit in its HoloLens.
Better Android integration than Gear VR or HTC Vive
One advantage that Google has over Samsung and HTC is its control of the Android operating system. As a Gear VR user, one weakness of current devices is clumsy integration with the native device OS. A Google-branded, standalone device could benefit from a more-seamless integration with Android. I’d expect it to run existing Cardboard applications, but it might add a complete VR interface to Google Play — so that content and apps could be managed directly from the device, unless the user is expected to manage the headset’s applications from their smartphone.
Separately, it is likely that Google will follow-up the mass-market success of Cardboard with a slightly higher-end version, perhaps to be announced (or even given away) at its I/O conference this May. I’d expect that product to look more like a Gear VR — with lenses, sensors, and perhaps even audio — but be dependent on your smartphone for most of its processing and its display. Of course, it would also support a wider variety of smartphone models than the few the Gear VR does.