AR glasses as theoretical smartphone replacements

shleepy

Limp Gawd
2FA
Joined
Oct 23, 2008
Messages
179
TLDR: What features of AR glasses actually have practical value for average consumers? Do AR glasses make sense as an eventual successor to smartphones, or are they more like optional accessories with specific niches?

We're 12 years past Google Glass, and it still feels like cheap, useful AR glasses for average consumers are going to be available "any year now." Lots of rumors, small companies releasing interesting but ultimately failed products, etc. I guess Apple Vision Pro had a brief moment, but I don't know anyone still using it or interested in it; but also, they're way too expensive, heavy, and big to be the kind of AR glasses one can comfortably wear on the street.

I think many techno-futurists envision AR glasses as the "next smartphone," in terms of the primary piece of portable technology that people will use to interact with the world, once the technology gets to a certain point. But is that really likely? And what would need to happen for that to work?

Without thinking too hard about it, I can see how - theoretically - having a screen immediately available in front my face, when needed, would be nicer than having a smartphone. No need to reach in my pocket and bend my neck, "bigger" UI/screen than a smartphone, readily-available eye-level camera, etc. But a couple of questions, if they're really going to be the next smartphones:
  • Are there good ways of interacting with them, at the same level as a touchscreen?
    • Voice sucks (e.g., when on public transit)
    • Eye movements are too simple and awkward
    • Maybe something with hand gestures, but that has similar problems as voice...
  • Does the AR element actually add much?
    • Showing walking/driving directions, in the form of giant arrows or whatever, would be kind of cool (something I'd use once a month, perhaps)
    • Having a pinned recipe in the corner of my vision as I'm cooking - also a nice little thing
    • "Big screen for movies/games when sitting in my economy seat on a flight" - already possible with the current "XR" glasses, but that's not really AR
    • But most of the interesting use-cases I've heard are very specialized - e.g., surgery or architecture

What am I missing? Convince me that it's the technology of the future! I want to be excited for them, but I don't know why.
 
Makes more sense as an accessory, as its hard to put them in your pocket.
 
  • Like
Reactions: noko
like this
I think the idea is that you can always keep them on your head (if they're light enough to be comfortable).
Doesn't really feel realistic idea that people will want to keep glasses on 24/7 everywhere, especially indoors.

But I could imagine wearing the accessory when needed, for example the Nokia Lumia AR navigation system through glasses when navigating in a new city. It was a bit awkward through the phone camera but would work wonderfully with glasses.
 
We're 12 years past Google Glass, and it still feels like cheap, useful AR glasses for average consumers are going to be available "any year now."

I used Google Glass for a minute here and there. Obviously it wasn't cheap, but it also wasn't very useful or usable. Very small field of view, akward to read, akward to interact with, impossible to integrate with in a useful way at the time. I was supposed to be mentoring an intern to integrate our messaging app wirh Google Glass, and it was supposed to have a Bluetooth API, and at the first meeting I had with them, they were like yeah, that's not available; so that was a big waste of time --- the only thing they would do is push through a web service witha maximum of one push per 20 minutes... I don't even know what you could integrate well with that; of course, their companion app would push SMS over bluetooth, but I didn't know how to reverse engineer that, and my intern barely knew how to show up to work sober *sigh* ptsd.

Integration can be improved, and one hopes the field of view / visual placement can be improved, but I think voice is realistically the only viable input method for these. Voice input works for a lot of people, but not me, and that's fine; but it's also harder to use voice in a crowded room situation too.

If you never used Google glass, it was basically what became Google Watch, but instead of looking at your wrist, it was up and to the right (i think you could also adjust it to be up and to the left instead). And if you need corrective lenses, good luck!

Walking / cycling / maaaaaaybe driving directions could be a good use case for this, but people who don't wear glasses hate glasses and it's not that compelling. People who do wear glasses need something that works with prescription lenses or it's not worth it.

Navigation can work pretty well audio only, and there's lots of reasonable ways to get audio to you (headphone jack, bluetooth, speaker). Visual navigation can be easier, but unless it's really tricky and really fast paced, audio would be good enough.

And if you've already got to talk to the thing, it may as well talk back to you.
 
Doesn't really feel realistic idea that people will want to keep glasses on 24/7 everywhere, especially indoors.
I mean, I wear glasses basically all the time as is, so would not be weird to me. But not 24/7, I don't know of that many people that try to sleep with glasses on.

But if they don't work with prescription lens, then yes it's still quite useless. If they do work with prescription lens, then I guess it wouldn't be much different for the glasses wearing population. We're already used to it, so not really any harm in extra features on our glasses, unless it gets distracting.

I don't think driving is actually a good use of it though, not unless it actually tells you where to lane change and turn at, like a heads up display. I don't think Google Maps is made for that yet. Hell Google Maps doesn't even tell you what lane to actually be in (ironically Apple Maps does, which is something I found weird; Apple Maps is actually better than Google in some ways). If it's just displaying a map in front of you, I feel like that would be distracting. Thing is I can glance over at my head unit and see the navigation display briefly, then come back to looking at the road with 100% focus. With a map and other information in my face, I feel like I would be distracted as it's up in front of me all the time (also probably with varying opacity due to lighting conditions, which is going to be even more hard on the eyes). There's a benefit for having something like that separated from your vision.

Now if it showed a breadcrumb/arrow style of navigation kind of like in a video game... that would be pretty neat. But highly doubt that something that small has enough processing power to do Computer Vision on that scale yet. Something like that would be far more likely to end up in some later model Tesla first, as a feature of the windshield, than anything else. And it would be incredibly cool.

Actually, I really feel like that's going to be a weakness with this glasses stuff all around. Sure people right now are caught looking down at their smartphones and stumbling into people, but with glasses there's no tear-away, and it's going to be noisy even for the thing you're viewing. It's fine while sitting down, but I feel like while you're walking around, it's going to basically feel like stumbling around with half-VR going all the time. Maybe there's a way to get used to that, I don't know. I've never tried anything like that. For me the neatest thing to have with AR glasses is uh... lying in bed going to sleep so I can watch a movie or something as I fall asleep. Or maybe slacking off at work, stationary at my desk and looking serious while in reality... watching a movie.

That's about it.

Edit: Actually shit, I need to look into that now. Do these glasses that show stuff in front of you look pretty obvious to an outside viewer while you're watching something in them?
 
Last edited:
. Hell Google Maps doesn't even tell you what lane to actually be in (ironically Apple Maps does, which is something I found weird; Apple Maps is actually better than Google in some ways).

Google maps can do this, sometimes it does --- iirc, most often if you drag the map to see a bit forward. Waze (which is Google for many years now) does often, but it's usually unhelpful --- like when the lanes change at least 3 times before your exit, it tells you what you need to do for your exit, but not what you need to do now so you'll be in the right place for your exit.

I haven't had the pleasure of Apple maps while driving, so I dunno. From what I hear, it's great for driving near an Apple office and pretty decent near Apple stores, but that's old news. It'd have to be really good for me to switch because it was better given my opinions of Apple's opinions on UI.

Here's an article from 2014 about 'Lane Assist' in Google Maps 3.0 https://www.techlicious.com/blog/google-maps-3-0-lane-assist-uber-savable-maps/ I think this is just generically called lane guidance now. Of course, Google is terrible at consistency and sometimes it feels like Google Maps and even Waze are developed by people who have never actually driven a car or even been the navigator while someone sitting next to them was driving a car.
 
Last edited:
If it's just displaying a map in front of you,
Good augmented reality would paint on the street (like when you play gran turismo) the suggested path, lane, turn in advance, speed limit, at least that where it could go as a tech, it would let you purely look at the road in a least distracting way than a second screen ideally.

AR screen could be at least at first more an companion that phone replacement, you can leave some of the compute-storage on the phone to make the glasses lighter and better battery life, still have a touch screen to click-on, not sure if the muscle control type-hand in the air would be good enough to send text when you do not want or able to talk out loud and do all that a phone can do via the touch screen easily.
 
Last edited:
Good augmented reality would paint on the street (like when you play gran turismo) the suggested path, lane, turn in advance, speed limit, at least that where it could go as a tech, it would let you purely look at the road in a least distracting way than a second screen ideally.

AR screen could be at least at first more an companion that phone replacement, you can leave some of the compute-storage on the phone to make the glasses lighter and better battery life, still have a touch screen to click-on, not sure if the muscle control type-hand in the air would be good enough to send text when you do not want or able to talk out loud and do all that a phone can do via the touch screen easily.

Yes, that's what I was saying in the post, but again also I really doubt a pair of glasses could actually do that. In order to paint said HUD on your display, you need a lot of computer vision, which is visual data processing. I guess that field has probably gotten a lot better with AI, but that's still probably way too much power for a tiny processor in a pair of glasses. So, again, this is much more likely to be a feature in a Tesla car that gets projected onto your windshield, rather than an AR item.

Projecting onto a windshield would also be much easier because you can much more easily do math on what's visible through the windshield at the driver's current sitting position, vs a pair of glasses that could effectively be located anywhere in 3D space vs what they're viewing.

Google maps can do this, sometimes it does --- iirc, most often if you drag the map to see a bit forward. Waze (which is Google for many years now) does often, but it's usually unhelpful --- like when the lanes change at least 3 times before your exit, it tells you what you need to do for your exit, but not what you need to do now so you'll be in the right place for your exit.

I haven't had the pleasure of Apple maps while driving, so I dunno. From what I hear, it's great for driving near an Apple office and pretty decent near Apple stores, but that's old news. It'd have to be really good for me to switch because it was better given my opinions of Apple's opinions on UI.

Here's an article from 2014 about 'Lane Assist' in Google Maps 3.0 https://www.techlicious.com/blog/google-maps-3-0-lane-assist-uber-savable-maps/ I think this is just generically called lane guidance now. Of course, Google is terrible at consistency and sometimes it feels like Google Maps and even Waze are developed by people who have never actually driven a car or even been the navigator while someone sitting next to them was driving a car.
Google maps provides its "lane assist" way too close to the actual turns to be meaningful. It's near useless. I always have to be paranoid and scroll ahead a little on the route to see which way I'm turning on the next turn and how far it is ahead. Apple maps does it way better, though it's been a while since I've used it. This isn't a new factoid, you can google around and see tons of people saying that they prefer the lane assist in Apple maps by miles. If you have an old (lol but not too old) iPhone lying around, you could maybe tether it to your Android and try it out.

Unfortunately I just prefer Android too much in terms of flexibility and what I'm free to do with it, so there's no way I'm switching back.
 
I think the biggest issue with AR glasses that have see through lenses, the switch between real and AR is not there yet. You have to either focus on what is on the screen or your surroundings. Unless they can figure out how to project the image so it is in focus when you are looking past its projection then it will not be usable as a interactive device. I have the original Nreal glasses and use them to watch media when on the road or traveling. One area where they came in handy was when I was using an endoscope to look for something in my walls. Being able to have the glasses on right in the spot where I was searching was a big help in terms of dexterity and orientation
 
you need a lot of computer vision, which is visual data processing. I guess that field has probably gotten a lot better with AI
Not sure at car speed pace, but at walking space that something current phone-phone camera can do, apparently never tried them.

Apple when they were working on maybe making a car had some plants for having it with the windshield and car camera like you describe, Audi and others seem to be trying it:

View: https://youtu.be/-FFi1cJXmDA?t=100

The bigger distance between the eye and the windshield than a glass lens, maybe reduce the issue of very different physical focus (even if it fake being far to the brain)
 
hardware still isn't there. need good eye tracking and better screens/refresh rates. Remember how long it took OLED become mainstream vs what was predicted? Probably still 10-20 years away from AR/VR being mainstream unless AI accelerates the tech advancements
 
hardware still isn't there. need good eye tracking and better screens/refresh rates. Remember how long it took OLED become mainstream vs what was predicted? Probably still 10-20 years away from AR/VR being mainstream unless AI accelerates the tech advancements
Apparently Apple VR kit level of head and eye tracking was already good enough, not sure how long to make that fit in an glassed too... Meta working level tech would have been extremely expensive right now
 
VR is one thing because it's a completely controlled environment, with 100% consistent viewing distance, and the hardware has complete control to decide what the picture is. And also put cameras in between you and the screen very easily due to the dark void.

Real life is a chaotic mess and the only thing you can sort of rely on consistently is how far the viewer is from the glasses. The frame is much smaller so the hardware has to be that much more compact, the processor needs to do calculations based on how far your eyes are away from the lens (not everyone wears glasses the same), etc etc etc. Way too many challenges for glasses vs a full VR headset. You could offload the processing to another unit in your pocket or something, but I think it would still take a while to get there. You need to combine Computer Vision calculations for eye tracking with calculations for tracking the environment.

Not sure at car speed pace, but at walking space that something current phone-phone camera can do, apparently never tried them.

Phones are pretty powerful and usually have AI built in on the chip. I guess if you don't mind your glasses being a plug in to your phone, and needing the most flagship phone possible for any gen, that approach could work. But the latter requirement is going to make the cost balloon far beyond the entry point price of the glasses. The phone battery is probably also going to get burned out with going full throttle doing all of those calculations constantly, since outputting to glasses with a weird resolution and doing the transform matrix calculations to decide where elements go on the actual hud is going to get complicated. I mean who knows, maybe they have this figured out better than I thought, but I just don't see it as working reliably without a LOT of money dumped into it, and I don't know what company would do that considering the returns are going to be so uncertain.

As you linked, I see luxury car manufacturers, and especially Tesla, as being much, much more likely to adopt this. Plenty of space and battery for the processing within a car, AC system can kick in to cool the chip, and it really just makes sense as a windshield projection.

Glasses are kind of going to be "next, next, next gen" imo. I can't really see there being enough market interest.
 
Every time I start thinking about this subject my mind goes off in weird directions. Like using vestigial organs for local compute. Except that the most practical application for AR is in places like pawn shops or a market place where all the good stuff is off the sales floor.

Or coming from a different direction using it as a jumping off point for minimally invasive vision corrections. Like a hearing aid for the eyes
 
Back
Top