You can't do good AR. You can do a nausea-inducing version with significant lag and poor real world image quality. The whole benefit of real AR is to remove the lag associated with capturing the environment.
Trying to replicate AR in VR makes VR a stop-gap to true AR. It does not make AR a subset of VR.
What's your definition of "good" AR - Lag is a camera issue, put a 120fps camera on the front of a VR headset, you're well over the nausea threshold. I get seasick looking through the electronic viewfinder of a mirrorless camera (hence I shoot with DSLR), because they only use a 60fps refresh and I'm very lag-sensitive, but I can work in VR all day - you just have to have the refresh over 90fps, which all good headsets do.
I maintain, AR is a subset of what VR can do - VR systems can do AR tasks, AR systems cannot do VR tasks, that makes AR a subset, and VR a superset. If Apple can call AR on a phone screen "AR", then AR through a VR headset is just as much AR. If you're going to insist on a
"no True Scotsman" definition, then what you're fetishising is the tool's aesthetic and methodology, not its utility.
Realistically the only distinction between the AR & VR, is one of them can block out the outside world. Whether that is achieved by a VR headset that has passthrough video, or a fitted AR headset that has a lightsafe electrically opaquing visor doesn't really matter - the point is the size of the thing is going to be similar either way.
Wide field of view, and high quality graphics, necessitate a certain size of lenses and length of optical path, regardless of what sort of headset they're in.
Why use a AR? When you can't reasonably have a screen in front of you. Anything that requires you to be moving about, be in a confined space, or always have access data without looking away. Surgery, mechanic in a tight spot, training for things like driving cars, trucks, or boats.
If you look at the environmental requirements for Hololens, you basically need to build an environment around it. I don't buy the "just use it while you're active" argument. It sounds like it's just as tethered to specialised, controlled locations, as a Vive is to its lighthouse-designated playspace.
As for surgery, this is not the dark ages, we don't do most surgery with naked eyes. It's mostly keyhole, and mediated through cameras already.
And training is already done through VR - you absolutely do not want learners operating vehicles with their vision obscured by anything if they still need so much hand-holding that they need in-vision training materials.
That's why we use VR for vehicle training.
Again, it doesn't have to be in a pair of raybans. Society won't be very accepting of hidden cameras, so it doesn't need to be a solution that is always on you.
The hololens 2 shows that AR exists in a form that is functional.
So, it's the size of a VR Headset, it's as face-obscuring as a VR Headset, but it's able to do less than a VR headset.
People generally prefer their tech to disappear. VR forces you to make the world disappear. Which is great, for the short periods you want that to occur. But as soon as you need actual collaboration with other people, or access to anything else in your environment it's no longer useful.
Unless the technology is so compelling that using it is the point of the tech - all these AR tasks you're talking about, you're making incremental improvements to other tasks, at the cost of a bulky head appliance. It's a dispensable addition. What VR provides is workflows and capabilities that simply can't be done with AR, or any other way. Again, VR hardware's capabilities are a superset of those of AR hardware.
That doesn't exist in consumer products. Also you can't capture the facial expressions of people not wearing a headset. Remember, in no situation is everyone wearing AR or VR, but you still need to be able to work with them.
I mean it's being done right now with consumer headsets that have inside-out camera tracking - A person in a hololens can only be seen from the nose down - so it's the exact same facial area that's exposed.
For 4 hours. I adamantly disagree that the average person can use VR that long. Your sample is biased because it's based on people who have trained themselves to do it. Not everyone can overcome all the issues of VR to get to that point. More importantly most people will not want to do it.
I literally put old-age pensioners with no prior exposure to VR in it, and after a 15 minute demo, they work in Tilt Brush for hours on end. Their only limits, are standing endurance, not the use of the headset itself, which would be the same limits an AR user would encounter. So, your opinions don't match my observed experience.
Here's where things stand IMHO:
- AR-Only headsets are not inherently any more compact than VR Headsets, outside of the science fantasy of "AR Glasses".
- AR-Only Headsets are just as tethered to controlled environments, as VR headsets are to their playspaces.
- VR Headsets can do AR tasks just as effectively, even moreso because they can process passthough video, eg low-light image amplification, contrast boost, darkening tint etc.
You're welcome to have the last word, because I think we're chased this about as far as it can go - I don't agree with your fundamental premises, because I've seen counterexamples to your assertions about the advantages of AR-Only headsets, and your assertions about the disadvantages of VR Headsets. I think both of these tools are going to be relatively niche compared to something like smartphones, and more importantly, Apple is not going to deliver "ordinary glasses" that have AR capabilities,
ever.
And, while we talk about whether things are niche, and what "ordinary" people are going to want to use, we need to remember,
ordinary people get surgery on their eyes, to avoid having to wear glasses at all. Glasses of any type are not a thing "ordinary" people want to wear,
ever. You say people want their tech to disappear, but an appliance you have to wear on your face is as intrusive as tech can be.
I also think a major motivator for the AR fantasy has some pretty creepy class issues bound up within it - people who live in a 1%er fantasy world already, are inherently more likely to think about ways to augment that world. Ask someone in a single-room studio if they want to take up space with extra virtual objects, or if they want to have a limitless vista / empty distraction-free workspace, Vs. a wealthy person whose house is bigger than the Steam Home, you're going to get different answers.