Thoughts & impressions about upcoming MIX headset

I've been following the MIX AR/VR headset from ANTVR. They have a pretty unique optical setup that gives them a very wide (96°) field of view in a pair of surprisingly small/lightweight glasses.



They're also remarkably open and engaging of the community — they just did a live interview with someone who I believe is their chief engineer. He seemed very frank and sincere to me. I respect that.

From a 50,000 foot level, here's my impression of the product itself: great optics, but still quite immature on tracking technologies. They're selling the headset in conjunction with a variety of add-on options for things like hand tracking or controllers. As for the head tracking, which is so crucial to stable & persistent AR, you can either add Nolo for an outside-in solution, or they are developing their own inside-out (vSLAM) solution. But it's clear from the interview that this is a very difficult problem, and they are still looking for a better implementation. They would be open to partnering with someone else.

So, it's hard for me to plunk down $500 for what is essentially only a half-solution. AR without good head/world tracking, and without at least a good 6DOF controller of some sort, is really not AR at all; it's just a fancy see-through display.

But on the other hand, it's clear the ANTVR guys get this, and they're working very hard to find a solution. I hope they do, because this could be a real competitor to Magic Leap (with a noticeably wider field of view). And since I'm frustrated with how long Magic Leap is taking to release anything, there'd be a certain justice in a scrappy young upstart beating them to the punch.

However, the other difference between MIX and Magic Leap is that MIX does not address the VAC problem. The engineer went on about this at some length, suggesting that in the future they might have multiple displays at different focal depths to mitigate this issue. Whereas Magic Leap, with their fancy (light-field?) tech, supposedly eliminates the problem altogether.

But I'm not convinced that VAC will continue to be a serious problem once you get used to it. I think separating accomodation from vergence is something most people will learn to do with a little practice; the only reason we're not naturally good at it is that it simply doesn't come up in the real world. So, if they can solve the tracking an input problems, I wouldn't hesitate to jump on MIX regardless of VAC.

Anyway. Very interesting gadget.

They are going to be at AWE 2018 in Santa Clara (CA) at the end of the month, so if you are in the area, please check them out and let us know what you think!