For the last decade, AR has been solely in the domain of the enterprise. Google Glass and Magic Leap have both tried and failed to bring the tech into the hands (or rather heads) of consumers. Google Glass was perhaps too simple a device, whereas Magic Leap tried to do too much. I think Qualcomm's reference design is, as Goldilocks might say, just right.

• Wearing an entire computer on your head isn't going to allow for an appealing design (Looking at you, HoloLens)
• A device that's too simple won't be able to address enough use cases to make it appealing to consumers (what's up, Google Glass?)
• A single device that tries to do everything is going to be prohibitively expensive, pricing itself out of the consumer market (hello there, Magic Leap)
• AR's real differentiation comes from mobility use cases so the smartphone is the perfect partner

By pairing it with a smartphone, vendors can offload processing power, connectivity, and the sensors needed for AR use cases like SLAM. They can make a sleek, lightweight, appealing design - while still offering a wide range of functionality and have appealing graphics. But this comes at a price - the device has to be tethered. While the Nreal Light is one of the best AR headsets, it suffers from using a USB cable. Qualcomm's split processing design puts some processing power on the device itself, which would allow it to function without the cable. Great for those mobility use cases, like navigation or running apps. It's also important to note that in future, "tethering" doesn't necessarily have to mean "cable". For something simple, like GPS data, Bluetooth may be more than adequate - in that example, eliminating the need for a GPS module on the headset, lowering that BoM and the price point.

As predicted in our AR headset forecast, we are anticipating a major smartphone OEM to launch consumer-friendly AR glasses this year, and reference designs like this edge that prediction closer to reality.