Would it be Smart for Augmented Reality Glasses to focus being Display Interfaces Instead of All-in-one Systems?

A huge thing that sets apart most of the current generation AR headsets in the market from most Virtual Reality headsets is that they often integrate as an all-in-one system instead of acting as a headless display interface.

Phones are often what people upgrade the most. The upcoming Samsung S10 has a 7nm processor and graphics. It’s not only powerful, but also hugely efficient. Why don’t more AR glasses use phone tether capabilities to use it as the brains and power, something like the Samsung DEX? Why don’t we use the widespread adoption and billions of dollars of iterative technology in current generation phones to power AR glasses, kill two birds one stone and avoid looking like Cookie from Ned’s declassified?

Cookie from the show Ned’s Declassified

It would make AR glasses cheaper since you don’t have to design a processor and battery from scratch. More R&D can be focused on stuff like precision and reliable anchoring or display technology. Adoption rates and developer interest would increase with a lower price.

Prototype of Magic Leap courtesy of Business Insider

Why do we need devices like ‘Focals by North’(which I tried out and reviewed a while back) that heat up significantly and have to add bulk to its frame to accommodate the battery which also doesn’t last long? Not to mention its price, how it’s not upgradeable and being closed source. It’s a glorified smartwatch.

Magic leap is a step in the right direction, albeit they still made their own proprietary system. A tether to the back of the glasses would be non-intrusive and would make sense in current generation.