IIRC, this has been done, and it's not hard (magic leap 2 does it), but Abrash doesn't like it for some reason. I'd imagine it would make the holograms look weird, and you'd have to darken a pretty big halo around the lit pixels to make it work, creating some odd visual artifacts. Alternately, they could just dum the whole thing - I have seen a lot of selfies from Orion team people today, and the lenses are clearly pretty dark, so either they're in a sunglasses kind of situation, or they can dim the entire thing either electrically or chemically (transitions lenses).
Face-wearables - will any survive or thrive (Rift / Glass / HoloLens / Vive / Apple Vision etc)?
- Thread starter neye_eve
- Start date
There's a guy on Threads doing an AMA, so you could ask him:
https://www.threads.net/@lukehurd/post/DAcPbJuPLzU/
https://www.threads.net/@lukehurd/post/DAcPbJuPLzU/
So it’s sort of an anti-blooming effect. Or make them sunglasses in which case you’re trading real world contrast to get some hologram contrast. That does little to dissuade me from the opinion that the best case scenario for optical pass through AR has a pretty fatal flaw.
Vergecast was talking about this, didn't see an obvious roadmap to productize this any time soon.
But they said it's a pet thing for Zuckerberg, who's consumed with not being dependent on Apple or Google.
He believes or hopes that AR will be the next big computing platform, where everyone will wear these things and ditch their phones.
Supposedly they have or are trying to line up deals with Italian designer sunglass companies -- Luxottica? -- because it would have to be a form factor that people want to put on their faces all the time and so far, designer sunglasses are one such type of product which has a proven sales record.
But they said it's a pet thing for Zuckerberg, who's consumed with not being dependent on Apple or Google.
He believes or hopes that AR will be the next big computing platform, where everyone will wear these things and ditch their phones.
Supposedly they have or are trying to line up deals with Italian designer sunglass companies -- Luxottica? -- because it would have to be a form factor that people want to put on their faces all the time and so far, designer sunglasses are one such type of product which has a proven sales record.
Essilor-Luxottica isn't really "a sunglasses company". They're a near-monopoly that owns practically everything in the optical space, or if they don't own the brand outright, they're the license holder for the brand. That goes for both sunglasses (e.g. Ray Ban, Oakley) and prescription and a large chunk of the rest of the prescription and non-prescription verticals as well-- Sunglasses Hut, LensCrafters, Pearl Vision, Target Optical and that's just in the US-- they have similar coverage in other countries.
It sounds like both parties have a vested interest in this space-- if it ever does become mainstream.
Meta owns something like 5% of EL, basically making them a solid lock for future glasses of any kind.
They bought GrandVision, owner of Vision Express in the UK and a great swathe of European high street optical chains. I have no idea how that passed by various anti-competition authorities.
But the alternative is a bulky headset that is basically a non-starter beyond niche.
In order to make things look sensible in bright conditions without overall dimming, you not only need to selectively block real-world light behind virtual objects, you then need to also be able to light up pixels brightly enough to make virtual objects match the brightness of their surroundings. My guess is that even if you can do the former perfectly, no display tech that can practically be accommodated in a mobile device is going to be able to do the latter for daylight exteriors anytime soon.
Probably what you want to do is use whole-lens dimming to knock everything down to the highest brightness your display tech can match, and then use more selective dimming to make virtual objects more opaque.
Probably what you want to do is use whole-lens dimming to knock everything down to the highest brightness your display tech can match, and then use more selective dimming to make virtual objects more opaque.
The difference is that there’s a path toward incrementally improvement with camera passthrough. Miniaturization of the cameras, display, etc. Incremental improvements to the housing. You can imagine in say 4-6 years a version of the AVP that’s maybe 30-40% lighter. Move out to 10 years and you can imagine a really compellingly thin and light camera passthrough through device.
What does the equivalent runway look like for ameliorating the flaws of optical pass through AR. Can you imagine a device in the 4-6 years time frame that looks less washed out hologrammy?
Here's a difference question for you...do you see ANY path forward for pass through to be worn all day? How about for pass through that looks somewhat like Orion? I see no path forward to all day wear for pass through.
Physics won't allow for a camera passthrough device that's as thin as an optical passthrough without using waveguides. So yeah, a 30-40% lighter AVP might be possible, but that's still going to fail social acceptability. Even if it were 90% lighter (100g incl. battery), I don't think it's a viable device beyond intentional wear, whereas something like Orion, which gets out of the way and acts as just regular glasses, while too heavy right now (98g) has line of sight to all-day wear (50g).
I personally don't think there's line of sight for AR (optical passthrough) to ever have true MR capabilities like VR has, but that's OK, because the use cases are different. I do think in 10 years, AR tech will be at a point where it can replace your watch and phone for nearly all use cases, plus it will unlock a bunch of new ones.
I’m doubtful that 2 wearable devices are going to coexist. And I have a hard time imagining holograms being an acceptable solution. Given the just overwhelming importance of visual quality to all things computing it just doesn’t seem likely that people will accept that tradeoff. I don’t disagree with you that camera passthrough has its own limitations though.
My guess is that both will remain relatively niche until you can have your cake and eat it too. That probably looks like a waveguide based system with an ability to selectively fully block the real world. The technology isn’t there for it in the next 10 years but maybe in 20?
Even if they can get the size and weight down to sunglasses form factor, would people want AR projected all day or however long you're suppose to be wearing those things.
If it mostly projects the notifications which now ping you on your phones, is it worth it?
It could be useful for say navigation, labeling the streets around you, projecting arrows towards your destination. That's a nice UX but is it that much better than current mapping apps? Well I guess you never have to take your eye off the road if you're driving.
If it mostly projects the notifications which now ping you on your phones, is it worth it?
It could be useful for say navigation, labeling the streets around you, projecting arrows towards your destination. That's a nice UX but is it that much better than current mapping apps? Well I guess you never have to take your eye off the road if you're driving.
I think what you're looking for is impossible. Full passthrough technology that is socially acceptable, comfortable enough to wear all day, and light enough requires replacement of the human eye.
Yeah, I guess if the FOV were good enough on waveguides, they could fully replace VR headsets - I'm not sure this is a useful case though. I think both categories of device will continue to exist, but VR devices being more and more niche. I think that Orion + 5 years probably gets you most of what smartphones do today, except for photos and videos. I'm not sure how to solve those though - I guess you could go full dimming, but shutting out the world to look at photos and videos seems like a bad compromise.
You wouldn't have it projecting anything most of the time. I think its usage pattern will be similar to a smartwatch, at least at first, so occasional use for a few minutes per hour. That lines up with the battery limitations too.
I actually don't think that notifications will be the primary use case - the glasses should require a user gesture to project so that they do not interfere with potentially safety-critical activities.
This is one good, obvious use-case. However, I think this would draw a ton of power, since the projectors have to be on all the time. I haven't figured it out yet myself, but I can imagine use cases that are currently phone based - looking at QR codes, translating signs, live captioning/translating conversations, etc. Stuff that lasts for a minute or 2, then lets the device go back to sleep.
What’s socially acceptable today wasn’t always. If an otherwise compelling enough product appears and it’s at least ‘socially acceptable adjacent’ I think norms will change.
That’s a pretty severe shortcoming when photos and video are the beating heart of phones at this point. Texting is also a core function and it’s not clear how you make texting work well in any kind of face wearable while you’re on the go.
Texting works well with dictation (though social acceptability and privacy are real concerns), and there is certainly an OSK you can summon. I have also seen demos of using EMG for typing on any surface. I can totally see a world where you can have a virtual keyboard and type on it, either like a desktop keyboard or a thumb board.
I wonder if it wouldn't make sense, in the long run, for people to learn SEE (Signed Exact English) or their local equivalent. This seems like it would feel quite natural as a silent input method for on-the-go AR. Or maybe we can just read people's minds.
I think that doesn't do well from a social acceptability standpoint. It's like hand gesturing in front of the cameras writ large. I think there's a possibility for using some sort of EMG gesturing, which can be done inside your pocket, or with your hand at your side. Like Orion demoed DPad gestures, so as the tech gets better, maybe you can do more there.
Microsoft Halolens 2 is reportedly dead, with no replacement coming:
https://www.uploadvr.com/microsoft-discontinuing-hololens-2/
https://www.uploadvr.com/microsoft-discontinuing-hololens-2/
Not just social acceptablity, but also privacy. If you rely on visible input (like Vision Pro) what you are gesturing is in principle readable by someone 30 feet away. If you rely on voice input it's broadcast to everyone who can hear you. Gesturing with hands hidden isn't consistent with the idea of how headsets work. It has to see your hands or you need another device on your hand to make it an input device when your headset can't see it.
Honestly, I don't see them replacing handhelds, even though handhelds have drawbacks of their own. Also you can put them away more easily.
Yes I can imagine in 5-10 years that the Orion type glasses will be tremendously better and will look less washed out and brighter, along with smaller, lighter, and more "regular" looking.
Face recognition...so when you meet someone (first time or multiple times) it pops up their name and relevant info about them.
But brighter isn’t the issue. The issue isn’t that the hologram isn’t bright enough. The issue is that the pass through light behind it is too bright. Absent a way of lowering the lumens of the real world, increasing contrast by brightening the display will just lead to searingly bright light on top of the real world.
What’s the thing that is going to make this problem better over the next 3-5 years?
There's a front page article on why this is bad. Maybe if it requires consent from the other person it would be OK.
straight out of Minority Report then? Probably with annoying ads popping up in every blank space too.
I saw the doxing article. I'm not convinced it is really that bad at least in public.
It almost feels like it must be intentional with how many times this has happened:
- Invent something way before its time
- Fail to continue pursuing it to any significant degree
- Eventually cancel it after letting it wither for years
- See competitor come out with largely the same thing soon after
- Freak out and try to catch back up
Hardware and the OS ecosystem along with small powerful chips come first with iteration over time leading to slightly smaller sizes, fashion does not. You can see that with the Apple Watch and now the Apple Vision, the Swiss watchmakers will be history by the middle of this century if they don’t incorporate, learn master the tech aspects (they are running out of time with every iteration).
Meta has already wasted $50 billion dollars and they are not good with hardware and most of the OS level software work that needs to be done. You can’t buy everything off the shelf particularly the chips and the OS both have to be done in house in this race.
One other thing and probably one of the big reasons why Apple is looking to replace the Qualcomm modem long-term is the fact that you need to miniaturize to get to that so-called pair of glasses everyone keeps talking/crying about but never seem to consider what it’s really gonna take to miniaturize all of those hardware items combined with a real OS.
The breakthroughs needed in this area going forward is not going to endear Apple or Meta to the EU who will label it uncompetitive at every turn.
Don't forget the Nadella add-on:
Steps 6-10: sell back-end services for new device on Azure.
https://www.macrumors.com/2024/10/23/apple-may-stop-producing-vision-pro-by-end-of-2024/
According to the article the AVP is rumored to be ending production soon and development of the next "Pro" version is suspended for at least a year. A total of about half a million were produced instead of an original plan of 8 million. Instead they will focus on a lower cost headset.
So what are the options to produce a lower cost version? The obvious first cut would be the outer facing screen. Beyond that you could reduce resolution slightly, reduced field of view, take advantage of cost reduction from CPU process improvements or increased yields. But even a "low cost" version I don't expect to debut for less than somewhere between $1000-$1500 so they'd still be nowhere near the range of the Quest 3 or 3s and way outside "impulse buy" territory for most people. So then I have to ask if they are even going to release anything or is this a scenario where they tried to enter the market but will eventually decide there just isn't profit to extract and completely exit the market?
Personally I still see the market for AR/VR as basically limited to professional uses such as: architects, engineers, scientists, etc. mostly for special purpose visualization with another relatively small market for gaming. I just don't think a market that focuses on watching movies and doing business meetings or opening spreadsheets or presentation files actually exists. Even with billions of dollars I don't see Apple forcing the market they want into being and I don't think they are very interested in the much smaller markets that may be reasonable for such a product.
According to the article the AVP is rumored to be ending production soon and development of the next "Pro" version is suspended for at least a year. A total of about half a million were produced instead of an original plan of 8 million. Instead they will focus on a lower cost headset.
So what are the options to produce a lower cost version? The obvious first cut would be the outer facing screen. Beyond that you could reduce resolution slightly, reduced field of view, take advantage of cost reduction from CPU process improvements or increased yields. But even a "low cost" version I don't expect to debut for less than somewhere between $1000-$1500 so they'd still be nowhere near the range of the Quest 3 or 3s and way outside "impulse buy" territory for most people. So then I have to ask if they are even going to release anything or is this a scenario where they tried to enter the market but will eventually decide there just isn't profit to extract and completely exit the market?
Personally I still see the market for AR/VR as basically limited to professional uses such as: architects, engineers, scientists, etc. mostly for special purpose visualization with another relatively small market for gaming. I just don't think a market that focuses on watching movies and doing business meetings or opening spreadsheets or presentation files actually exists. Even with billions of dollars I don't see Apple forcing the market they want into being and I don't think they are very interested in the much smaller markets that may be reasonable for such a product.
100% agree. I've used PS VR, Quest VR and also had the opportunity to view Hololens at the Seattle museum of industry a few years ago when they had an exhibit of some kind of castle with fuck all of whatever going on in the visor. Made me realize all the MS talk was BS. The HL was not a good product (for the one test I had). Uncomfortable, shit resolution, small display to see, color was subtle, not bright enough.
No consumer will ever wear a helmet for more than a few minutes. All of these techs will sell a few millions to nerds and thats it. VR is cool for a minute then you move on and don't mention it again. AR is for businesses, and even then the tech isn't there unless you can spend$$$$$. What was the issue with Hololens? The display was too low resolution to be useful for businesses. And the military thought it was shit.