Google’s annual I/O conference kicks off tomorrow and we’re expecting VR to be a major part of the event. Here’s a look at everything we know and what we expect to see from Google’s virtual reality initiatives. Android N Bakes in VR Android was built as an operating system for more phones, not for VR headsets. The latency between input and response that’s acceptable for tapping and swiping is far removed from the sub-20ms threshold required for a great VR experience. Keeping latency low between a user’s head movement and the corresponding movement on the headset’s display is absolutely critical. [caption id="attachment_46143" align="alignright" width="325"] See Also: The Past, Present and Future of Google and VR[/caption] One of the biggest reasons why Samsung’s Gear VR provides a much better Android-based VR experience than even Google’s current Android-based Cardboard VR offering is that Samsung and Oculus have created custom Android firmware for Gear VR-compatible Samsung phones which is specifically tuned for virtual reality. This custom firmware allows Gear VR to cut through Android’s inherent latency in several important technical ways, such as Asynchronous Time Warp rendering. Today, standard builds of Android lack such VR-specific features, but the next version (Android N) is likely to be the first that’s tuned for VR performance. References to a ‘virtual reality mode’ have already been spotted in early builds of the new software, and we’re expecting to hear much more about about Android N’s VR-specific features at I/O 2016. Ultimately, we expect that Android N will bring the default Android software stack up to par with where Gear VR is today (hardware withstanding). Android VR as an Ecosystem Mentions of an 'Android VR' plugin in Unreal Engine development documentation surfaced recently, raising questions about whether or not it has anything to do with Google's VR initiatives. I believe the answer is yes, but I’m putting my money on the term simply being the next step after Google’s Cardboard initiative (rather than referencing an individual headset/product). As it stands now, Cardboard is precisely that: an ecosystem descriptor. It describes a class of ‘dumb’ VR viewers (which do nothing more than hold your smartphone in front of some lenses), and the software ecosystem that goes with them (both the tools and the apps that are made for Cardboard). Android VR as a term is likely to encompass both the new VR-specific software features of Android N along with a new class of VR viewer headsets and apps which go beyond Cardboard's current capabilities. No All-in-one Headset While some have speculated that the rumored ‘Android VR’ is an all-in-one headset being developed by Google, I don’t see Google taking such an approach (at least not at I/O 2016). It simply doesn’t make sense from a strategic standpoint. As the facilitator of Android, Google’s goal is to help its hardware partners sell phones. Android’s biggest competitor is iOS and Apple; the end goal of anything Google does with Android is to increase Android’s marketshare over iOS, not to steal marketshare from its hardware partners that fuel the successful Android ecosystem. Creating an open ecosystem of smartphone-based peripheral headsets (as they’ve done with Cardboard) is the exact same successful strategy Google used for smartphones: supply the software (and occasionally, hardware reference designs) and let hardware partners furnish the hardware side with many competing choices for customers. Opening the door to allow any of their partners to create smartphone-based Android VR headsets (rather than Google creating their own an all-in-one) encourages the key objective of selling more Android phones across the ecosystem. Look at it this way: if Android has great VR and iOS has no VR, it’s going to be a huge opportunity for Google to lure iOS users to Android (such opportunities don't come around that often, so Google is going to want to push this hard). Selling an all-in-one headset wouldn’t encourage an Apple customer to join the Android smartphone ecosystem (because they could then get the Android VR experience without converting to an Android smartphone). Making the smartphone the key to Android’s VR experience gives customers a major reason to switch to an Android smartphone and serves Google’s key objective of helping hardware partners to sell phones (and soon, headsets). Continue Reading: Android VR Headsets Bring Straps and On-board Hardware & Cardboard Lives On (for Now) Android VR Headsets Bring Straps and On-board Hardware Take a look at the sea of VR viewers out there that have the official ‘Works with Cardboard’ designation (a little orange badge with QR code), and you’ll notice a curious oddity: unlike the Rift, Vive, Gear VR, and others, none of them have straps. This isn’t an accident, it’s a mandate. Google won’t certify a VR viewer as ‘Works with Cardboard’ if it has a strap. A major reason for this move is that requiring the user to hold the device up their their head with their hands limits the rotational speed of their head significantly as it essentially requires you to rotate your entire torso as you turn. This was done to combat the latency inherent in current versions of Android, and to keep Cardboard VR experiences aimed at shorter sessions. But now that Android N is likely to be tuned specifically for VR (ideally getting under that sub-20ms latency mark), the restriction can be lifted. And for that reason, we suspect that Android VR headsets will offer two things that Cardboard doesn’t: straps and the possibility of on-board hardware. Why on-board hardware? Modern phones don’t have sensors on board that are fast or accurate enough for great VR. Another one of Gear VR’s tricks is that it has a high performance IMU built into the headset which takes over from the phone’s own IMU, speaking to the phone through micro USB. Even if you cut through Android’s latency on the software side, millions of existing phones out there will be lacking sensors that are ready for VR. Sensors on board Android VR headsets could fix this in the interim, while smartphones of the future will likely have the necessary sensors built in. Other hardware that could be quite useful on-board Android VR is Google’s Tango sensors for positional tracking and other computer vision applications, as well as an input mechanism like the touchpad on the side of Gear VR. Cardboard Lives On (for Now) There’s already more than 5 million Cardboard viewers out there as of January 2016. By all accounts, most people who try Cardboard quite like it, despite it being notably inferior to more expensive platforms. And rather than phasing Cardboard out, Goolge only seems to be pushing it further. Ever since launching the initiative, the company had open-sourced the design and allowed third parties to make all manner of Cardboard variants. Only recently has the company finally begun officialy selling their own first-party version of the folded cardboard device (though they have certainly handed out many for free up to this point). Cardboard will likely continue to exist as a super low-cost VR option for older devices that won’t ever see an upgrade to Android N. With the latest version of Android (6.0, Marshmallow) only deployed on 7.5% of devices (despite being released 8 months ago) and a majority of devices running Android 5.0/5.1 (initially released 18 months ago), there’s going to be a huge portion of Android smartphones that will likely only ever be Cardboard-capable, and this will likely continue to be a segment that Android VR developers want to target. Importantly, Google is likely to continue its Cardboard infiltration of the iOS App Store, threatening to achieve some user lock-in before Apple even announces any official VR offering. With I/O 2016 starting tomorrow, it won’t be long before we see how close our predictions align with reality. We’re on the ground at I/O 2016 this week and will be bringing you the latest on Google’s virtual reality initiatives.