Analysis of Magic Leap 2 at SPIE AR/VR/Mr 2022 | Hacker News Reader

Mr 2022

kguttag.com

250 pointskguttag130 comments

Thanks for the article! It's very interesting to see what Magic Leap is up to. Any rumblings about how Microsoft is progressing on Hololens 3?

Also, a side note—I develop [Unity libraries for displaying web browsers in 3D][1], including a package for Hololens because I received many customer requests for it. In contrast, I've received relatively few requests to support Magic Leap, which leads me think that the adoption of ML1 has been quite low in comparison.

[1]: https://vuplex.com

Thanks,

The rumor mill is that Microsoft has slowed down development of HL3 development with many people leaving. Most famously, Bernard Kress, and optical architect of Hololens 1 and 2 has left for Google.

There were very few Magic Leap Ones ever sold. The number I hear is between 5,000 and 10,000 units sold for money with about as many units given away.

Hololens 2 seems to be shipping about 70K to 100K units per year but it is not clear if that has peaked or not. The display quality is very poor as I have written about extensively, but the ergonomics are good enough for some industrial applications.

I get the sense that ML2 was too far gone and with money running out to react to ML2 and thus still has terrible ergonomics. With blocking 80% of the light, not supporting ordinary glasses, and having a cable which is a snag hazard, the future looks bleak for Magic Leap 2 as it was for ML1. I don't know if they would have been a success, but at least they would have had a shot in the same space as HL2 if they had reworked the serious ergonomic problems. The dimming feature is a differentiation that resulted in them missing the market altogether.

Thank you very much for the info! The detail about Kress joining Google is fascinating, as is the contrast in sales numbers for Hololens 2 and ML1. Hololens 2 is the most advanced AR headset that I've used so far, so it's helpful to hear from you that its display quality is poor compared to others. Despite the display quality, I've been blown away by what it's capable of, and I look forward to seeing how this space evolves over the next 5 years.

I was tentatively excited for the prospect of the magic leap 2 being a successor or alternative to the hololens 2 in industrial / job site environments so am definitely disappointed to hear it has the same problems in those contexts as the magic leap 1. The ML1 was a complete non starter for us because it’s a physical safety hazard in any risky environment.

Thanks as always for your coverage on this stuff.

This looks like a really great tool! Is it possible to control the contents of the WebView through script and turn off the interactivity / keyboard?

Thank you! Yes, apps can control the webview through the IWebView interface:

https://developer.vuplex.com/webview/IWebView

In most cases, keyboard input is sent to a webview programmatically:

https://support.vuplex.com/articles/keyboard

So, to disable the keyboard, the app can just omit the Keyboard prefab from the scene and leave the WebViewPrefab.NativeOnScreenKeyboardEnabled option disabled (for mobile):

- https://developer.vuplex.com/webview/Keyboard

- https://developer.vuplex.com/webview/WebViewPrefab#NativeOnS...

If you want to disable clicking and scrolling, you can set the WebViewPrefab's ClickingEnabled and ScrollingEnabled options to false:

- https://developer.vuplex.com/webview/WebViewPrefab#ClickingE...

- https://developer.vuplex.com/webview/WebViewPrefab#Scrolling...

We designed a number of ar, an vr glasses.

Engineers of most buzzwordy brands don't seem to have had "optics 101" during their engineering education.

https://news.ycombinator.com/item?id=28839541

> Holographic waveguides, and such are a dead end development with a substandard image quality which you cannot do anything about, as well as its inefficiency (these waveguides waste 70-90%+ of light.)

> The fundamental problem is that the amount of image light transferred is inversely proportional to the amount of light obscured. — they are extremely inefficient, and there is physically no way around this.

Very nice in-depth analysis, thanks kguttag.

My personal bet on this space is that kura.tech's product is going to be the real deal; but it is exciting to see that the AR field in general is starting to enter its own plateau of productivity and we may start seeing great products soon.

I think we need to see Kura actually demonstrate that their technology works.

You might want to see this video (queued to his discussion of Kura at CES 2022: https://youtu.be/S0heZVN5NCs?t=1328

>I think we need to see Kura actually demonstrate that their technology works.

Indeed.

Back in October, when Facebook made a lot of "metaverse" noise, I wrote a thread looking at the history of stereoscopic 3D from the 1850s to the present as a way of explaining why I think facehugger VR is going nowhere. Somebody turned up to complain that going back to the Civil War era wasn't sufficiently big picture for him:

"You need a long view. All the serious players are still in the Mercury phase. The Gemini and Apollo phases of XR are coming this decade and into the 2030s. An $800b company will likely spend $100b+ getting there. You need a wide aperture to understand what is going on."

https://twitter.com/williampietri/status/1454135053697519619

What I didn't realize until later was that the rando reply came from no less a presence than Rony Abovitz, Magic Leap's exiled founder, who apparently now spends his time vanity searching for "Magic Leap" so he can reenact XKCD 386. There are many things I find jaw-dropping about his views, but my biggest question is: Why did he raise and burn $3.5 billion dollars if that's only 3.5% of the cost of "getting there"?

Which is a roundabout way of saying I think kguttag is right when he says, "ML2 looks like a product designed for the consumer market, but it was seen that it would be far too expensive for that market, it was recast as an 'enterprise' device." To me it's a not-great solution desperately looking for a problem.

Quick link to XKCD 386, for others - https://xkcd.com/386/

The same Abovitz that did this abomination of a TEDx talk ... https://youtu.be/w8J5BWL8oJY

2001: A Waste Odyssey

omg I'm in tears, its like a low budget David Lynch skit

What the fuck was that?

The kind of person who can raise $3.5 billion without a working prototype, let alone any products or revenue

This iteration of IT bubble is particularly into charismatic founders.

What, and I mean this sincerely, the fuck?

I have been to a bunch of experimental theater over the years, so I was prepared to accept the initial setup. But he didn't pull it together at all!

Frank Zappa wrote a song about the Magic Leap investors:

https://www.youtube.com/watch?v=K_WgCGOWae4

https://www.google.com/search?q=cocaine+decisions+lyrics&oq=...

I wonder if he's singing about what he named his kids.

Man the 70s were special.

The only drugs Frank Zappa did were nicotine and caffeine.

I just love the youtube comments on this. They were satirical even just after the talk posted.

"This crazy art form corresponds to the crazy significance MAgic Leaps technology will have on mankind. IT IS THE NEXT BIG THING WE HAVE ALL BEEN WAITING FOR."

...

> Which is a roundabout way of saying I think kguttag is right when he says, "ML2 looks like a product designed for the consumer market, but it was seen that it would be far too expensive for that market, it was recast as an 'enterprise' device." To me it's a not-great solution desperately looking for a problem.

Microsoft did the same thing with Hololens. First it was supposed to be an entertainment device for the home, where you could enjoy interactive experiences and mess with little Minecraft models on your coffee table. Well, that didn't pan out. Next they made a bid for the creative professional market, who might get a kick out of being able to manipulate 3D models in realtime through a headset. Oops, you can do this with a $300 Oculus Quest and a free Blender plugin. Okay, how about the enterprise market? They've got much deeper pockets, and we can just toss together a few productivity apps to make it seem attrac... what do you mean they don't want it?

Fine, we'll take it to the one place that will burn money on a nothingburger no matter the cost. We could bring these people a gilded pile of scrap metal and they'd pay 10x market price. There's only one customer that meets this criteria: the United States military!

Rony sez: "You need a long view. All the serious players are still in the Mercury phase. The Gemini and Apollo phases of XR are coming this decade and into the 2030s. An $800b company will likely spend $100b+ getting there. You need a wide aperture to understand what is going on."

He should know, he went to Nova and he's been as deeply familiar financially in this exact effort as anyone for 10 years.

But from 1850 to 2020 that's 170 years so it looks like about the year 2190 would be the lower end of a not-big-enough-picture. That does make the offhand estimate of 2030 look a little optimistic. But optimism is the name of the game. You don't go into this if you thought it would take that much more than a lifetime. That's just the kind of thing you find out later. Sometimes later than others.

Looks like in "today's dollars" the Gemini program was over $10 Billion and Apollo over $200 Billion. Plus it took about a dozen years to spend it, seems like a long time to get to the moon but by comparison you could say sheesh NASA went through that money fast. And what do they have to show for it?

You do need a long view. Nothing happens overnight.

So I guess $100 Billion for the total AR dream would be a good deal since it would only get you halfway to the moon anyway. Wouldn't buy much else either, at today's prices.

>Why did he raise and burn $3.5 billion dollars if that's only 3.5% of the cost of "getting there"?

Do you really want my answer? Well if you did I could probably come up with something . . .

I don't know. I would say the historic trend is computing becomes more and more ubiquitous. Truly ubiquitous computing would need an always available display and AR wearables seems like the solution there.

"Its too expensive right now" isn't really a damming argument, is it?

The thing with trends is that they all end eventually. At the start of the automobile era, cars were quite slow, and they got faster over time. If we just draw out the dotted-line trend, we'd all be driving around at supersonic speeds.

It's true that computing has become more and more ubiquitous in recent decades. Will that continue? We'll see. Will it continue with VR and AR? Maybe, maybe not. Your approach to this is fine for asking "Is this SF short story plausible?" But it's not at all sufficient to justify billions of dollars in spending.

Read your thread, but I don't think you have quite the right idea.

There's certainly a long history of static 3D imaging with photography and then movies. But that's not all that related to the modern tech started by the Oculus Rift. Yeah, 3D doesn't make photography or movies fundamentally different for the most part, because with some rare exceptions we already can fill in the blanks.

Consumer VR itself is a pretty new development and does introduce new things to the table that weren't there during the civil war and whatnot. Eg, things like Robo Recall and Beatsaber have gameplay that just doesn't work without VR. You could sort of try with something like the Kinect or the Wii I guess, but it'd be much more awkward and clunkier.

And some tech indeed takes a while to develop. Mobile telephony is a tech more than a a century old, if you could the very oldest prototypes. Analog, consumer mobile phones first appeared in the 80s, 42 years ago. The first smartphone is from 1993, and the first iPhone was released in 2007. I'd say at this point a smartphone morphed into something completely different, to the point that it's really a very portable computer that sometimes happens to make calls.

The Rift CV1 is from 2015, so that's pretty recent. And it takes an amazing amount of high end tech to make a headset work, so I do expect there's still lots and lots of room for development.

Now on the Magic Leap, I don't know if that's going to go anywhere or not. The company itself, probably not. But unless there's no way whatsoever to make it usable, I think somebody will eventually come up with an affordable and useful version.

Sorry, but I'm not persuaded. Does some tech take a while to develop? Sure. But some tech never develops. If you're only including happy-path examples, that's a biased sample. If you're trying to do analysis and not propaganda, you have to look at both routes and see what the differentiators are.

If you want to suggest this time it's different, you have to explain why it's different. My point is pretty simple: Stereoscopic 3D is an attractive nuisance. There are many times historically people have confused its admitted novelty value with actual utility. There is every reason to think this is one of those times.

Facehugger VR is a marriage of two concepts: 3D virtual worlds and stereoscopic imaging. There is lots of proof that the first is hotly desired; anybody who has tried to pry a kid away from Minecraft or Roblox knows that, and I was the same way with Doom and Quake. But there is very little evidence that stereoscopic 3D has more than novelty value. And there's 150 years of evidence that people, even very smart people, confuse that novelty value with something that will last.

> If you're trying to do analysis and not propaganda, you have to look at both routes and see what the differentiators are.

Well, the thing I see immediately is that you're not really doing an accurate analysis. Stereoscopic 3D from a fixed point of view is indeed an old concept, and definitely not enough. But we've moved well beyond that already.

I think the actual value proposition is immersive stereoscopic 3D + good in-world body-based controls. That changes things in a way something like a 3D TV can't.

Eg, for games, you can't really get more immersive than acting out your character's movements yourself. Something like Superhot lets you do Matrix-like moves in VR that just nothing else does. You can find games with slow-mo like Max Payne, but they don't make non-VR games in which you can actively move your head out of the path of a bullet and have that work naturally.

Or something like Racket NX is a very real workout that works very naturally.

The downside of course is that the tech has considerable limits and constraints that will still take time figuring out. Normal computer games worked out their mechanics over decades, so if you go back far enough to something like Dune 2, it feels very clunky.

On a longer term, I'm hoping for the day I can replace my monitors with a VR helmet and just display anything I need anywhere in an arbitrary position and amount. One can't ever have enough terminal windows.

I again don't think you're grappling with my point. There are certainly differences between each of the 5+ waves of excitement over stereoscopic 3D. Each time people argued that really, those differences made the difference. And each time they were wrong, because they personally found the idea exciting. Instead of addressing the pattern seriously, to me it seems like you are repeating it.

But it's not about just stereoscopic 3D, I'm saying. That's old. Immersion is what is new. Eg, take this:

https://www.youtube.com/watch?v=O1wStc0m86M

That's a VR specific experience. You're not just clicking a mouse button and watching a soldier die, you're sticking your sword in their eye socket with your own arm. Even though that game is lacking polish still, the experience of doing that is unsettling in ways that no PC game is. Because it's really you doing that, without any abstraction over it.

I agreed previously there are differences. And clearly these are differences you find intellectually exciting.

But that was true about previous generations. 3D TV was going to bring the experience right into the home! It was a game-changer! Not just movies! Not just TV shows! But live events like sports, where 3D perception would make a big difference!

And still, it sank like a stone.

There have been VR-specific experiences since the 1990s. I have tried many of them. They are neat! But take Superhot and Beat Saber, two "only in VR examples". I rented a Quest and they were indeed cool. But Superhot sold more non-VR copies, so obviously the experience works well enough without VR. The kids really loved Beat Saber, but they played it by sitting on the couch, staring straight ahead, and twitching their wrists. And when they got tired of having a sweaty, heavy thing on their head, they went back to their Switches and the PS4. When I sent the Quest back, they never even noticed. Whereas if I'd gotten rid of the PS4, it would have been armageddon around here.

The truth is that existing games are already very immersive. Getting the kids out of Minecraft or Roblox or Horizon takes a crowbar. if you want do demonstrate that this new technology is truly more immersive, you'll have to show not just that you think it's cool, but that a mass audience actively prefers it and won't go back.

That has certainly happened with entertainment technologies in the past. Look at color film and later color TV: people were willing to pay up, and the new tech almost entirely drove out the old in short order. Compare that with 3D movies and 3D TV: people care at best a little.

So far, everything I see suggests that facehugger VR is the latter category. If you have data otherwise, I look forward to it.

> The kids really loved Beat Saber, but they played it by sitting on the couch, staring straight ahead, and twitching their wrists.

That's completely missing the point of it. If you're not swinging your arms around like a maniac and aren't lying in a puddle of sweat after a few songs, you're not really doing it right. Also, Beat Saber penalizes you for such minimal movement.

> The truth is that existing games are already very immersive. Getting the kids out of Minecraft or Roblox or Horizon takes a crowbar. if you want do demonstrate that this new technology is truly more immersive, you'll have to show not just that you think it's cool, but that a mass audience actively prefers it and won't go back.

Immersion to me is not addiction, it's immersion. Feeling like you're inside the game, for however long that happens to be.

Thanks, yes, I understand the theory of VR. What I'm talking about is the reality of VR. You're like a guy explaining the theory of why 3D movies are better even though most people clearly do not care very much, and many actively don't like it. Data trumps theory.

Immersion is not addiction, sure. But I feel plenty immersed when playing Minecraft. I lose track of the outside world. Or watch the way people move physically when they play an action video game on a screen; if they're into it they will move their heads and bodies to dodge. They'll physically jump back at a surprise or a scare. Immersion happens just fine on 2D screens.

Does Facehugger VR feel immersive? Definitely. Does it it feel more immersive at first? It did for me, although that feeling wore off over time. Does that matter enough to get people to stop using flat screens and game only in VR? Not for me or my family, and that appears to be the general reaction.

I'm going to say one last time that you do not appear to be getting with or grappling with my main points. If all you have is more argumentative nitpicking, this will probably be my last reply, as it looks to me like you're in the grip of the exact problem I'm describing.

This is both really impressive and really disappointing. The technology has been refined to an incredible degree but it sounds it's still nowhere near the fidelity that would make it useful. It may the biggest VC money pit in SV.

In Florida, not SV

Funded by SV, vision alignment with SV which is probably more important than where it’s made. However it is certainly impressive that they’re making a lot of progress in a tropical latitude, a feat that hasn’t been replicated by many successful tech companies.

> Funded by SV, vision alignment with SV

Actually apart from Google, their investors are nontraditional: https://www.crunchbase.com/organization/magic-leap/company_f...

The “usual suspects” all passed. Even google’s venture fund passed but Sergei decided to do it.

I think ML is an example where not being from the valley was kind of an advantage for them on the fundraising side.

Not just any part of Florida.

Although today things are culturally much more homogenous, this particular town 50 years ago already had a long tradition of starting companies with the express intention of becoming comfortable to begin with bringing in money from investors only. The dream is that the venture could actually pay off someday and then everybody involved will really profit phenomenally when the company is sold.

Not so much pursuing product sales where value needs to be added. You're selling a dream that naturally appeals to investors more so than retail customers who would want something tangible sooner. Who has more money anyway?

"Real estate" was usually the financial vehicle but you had to be able to sell "false estate" (Everglades) as if it was really a good investment. Well it truly had no place to go but up ;)

Quite unlike nearby Miami or Palm Beach even, and nothing like Tampa or Orlando and the other places.

Much more prominent dream-selling element in company founders (among the smaller number) by far than SV at the time. You've got to really be persuasive. SV sure has come a long way in this regard now.

This time it's not real estate but technology, crafted to appeal to SV investors.

A number of years before Magic Leap got attention, there had been a very embarassing "Digital "Dream" up the coast in Martin or St. Lucie where they were going to make it like a little SV. The idea was to create a ground-up "ecosystem" where young people can have the kind of opportunities in software & game development they could only dream about.

There was a real estate component in that one so they broke ground, started developing a campus and everything, but once the investments quit coming in they were done. Plus real estate had become uncertian in ways that had never occurred there. Some respected investors did lose a lot plus I think even some quasi-government development agency went down. After the details came out the general consensus was they were taken for a ride the whole time.

Magic Leap has offices all over. They had one in Mountain View just a few blocks from where Rengstorff crosses Highway 101, over by Google. That was a few years ago, dunno if it’s still there.

They seems to have done some good things. Sadly in AR as well as many high tech markets, it is possible to have a great technical achievement but miss the market requirements. The dimming feature caused them to lose at least 70% of the real-world light off the top. Nothing else they could do would made up for that mistake. It looks like they were focused on novel features over utility.

I think a large part of this is that they were building the ML2 for the high end consumer video game market when one day they they were told it was too expensive and it would now be an enterprise product. I hate the display quality of the HL2, but at least they improved ergonomic and interface issues over the HL1, whereas the ML2 repeats the mistakes of the ML1.

I’ve loved your posts for years on piercing through the smoke and mirrors on Magic Leap. Thanks for another great breakdown!

Thanks

https://philip.greenspun.com/humor/media-lab.html

>Generation of Bits

>Tales of shame and degradation in the Big Idea Lab

>by Hunter S. Negroponte

[...]

>The Big Lie

>Media Lab critics (there have been a few) have occasionally questioned the practical application of our work. Well, have you heard about the Holographic Television? No longer a device found only in the back of comic books, we've actually made this sucker work. An honest-to-god motion-picture hologram, produced in the Media Lab basement on a 2000 pound holography table by computers, lasers and mirrors spinning at 30,000 RPM. It's real! It works! Life Magazine even came in to photograph it in action (of course, they had to fill the room with smoke so the lasers would show up on film). Practical application? Sure, it requires a 2000 pound air-suspended rock table and a Connection Machine II to run, but hell, everyone knows the price of computing power and 2000 pound rock tables is cut in half every year. My point, however, is more mundane: we have created a demo literally from smoke and mirrors, and the Corporate World bought it. Even my good friend Penn (or ``Penn,'' as I call him) Jillette would be proud.

>In fact, I'm a few points up on Penn. You may have heard of the Interactive Narrative work that is proceeding in the lab. Folks, I'll be honest with you for a moment. I know as well as you do that it's a stinking load of horseshit. Roger Ebert said ``Six thousand years ago sitting around a campfire a storyteller could have stopped at any time and asked his audience how they wanted the story to come out. But he didn't because that would have ruined the story.'' You think Hollywood would have learned this lesson from the monster ``success'' that Clue, the Movie enjoyed several years ago. But no! I've repackaged the ``Choose your own Adventure'' novels of childhood as Digital Information SuperHighway Yadda Yadda crap, and again, they bought it! Sony right this minute is building an interactive movie theater, with buttons the audience can push to amuse themselves as the story progresses. Dance for me, Corporate America! I'm SHIT-HOT!

>Why, just the other day I listened to a member of my staff explain to potential sponsors that we had spent \$US 4,000,000 of Japanese sponsor dollars to construct a widescreen version of ``I Love Lucy'' from the original source. And HE SAID IT WITH A STRAIGHT FACE! CAN YOU FUCKING BELIEVE THAT? Boy, I bet those Nips wish they had their money back now! Earthquake? No, we can't do much to rebuild your city, but we SURE AS HELL can give you a 1.66:1 cut of Lucy to fit all those busted HDTVs of yours! HA HA HA!

[...]

Optical AR is a dead end. Passthrough AR with VR goggles and 8K cameras on the outside will be the way that wins.

No one wants to wear a headset if it makes them look stupid. If AR glasses are a dead end then mass adoption of AR outside of gaming/enterprise is likely DOA.

Bluetooth headphones were "stupid looking" and endlessly mocked.

Now airpods are in vogue.

Social perceptions easily change, especially when there are billions of dollars invested in making them change.

Hmm, I don't think so.

I do think the AirPods metaphor is good but for the exact opposite reason: AirPods were the first "true wireless" stereo earbuds on the market with both a working microphone and reasonably good pairing and activation capability (at least with iPhones).

Previous Bluetooth headsets were either mono, for phone calls only, which quickly and probably correctly got associated with obnoxious salespeople yelling about deals in airports, or connected stereo affairs, which were generally strange looking and unwieldy. And the 2 or 3 "true wireless" offerings on the market ahead of Apple were hampered by hobbyist quality and poor Bluetooth behavior. AirPods offered a meaningful and substantially improved utility over their predecessors - not just a glossy style or some kind of marketing driven social change.

In this same way, true augmented AR offers genuine advantages over passthrough AR, in terms of comfort, safety, and utility. If the technology can be made to work augmented AR is likely to win over any kind of passthrough AR for reasons of utility, not just fashion.

What are you talking about? Good Bluetooth headphones had existed long before apple did something.

headphones != earbuds

Airpods were one of the first truely wireless earbuds from a well known company. They were successful because of their looks, not in spite of them.

Before airpods wireless earbuds looked like this: https://www.lg.com/us/bluetooth-headsets-headphones/lg-HBS-7...

Do we forget that AirPods solved a real pain? No more tangled wires, and they were the first in-ears I actually found comfortable.

Before that wired headphones were "stupid looking" and endlessly mocked. Before that, boom boxes.

In all cases, I think it was more the “look how I’m more interested in my own world than in people around me” than the actual looks of the device.

Possibly, the fact that early portable tech typically is fairly impractical (for example in battery life or in base ergonomics) added to that.

Apple is great at many things, but exceptional at fashion.

Last time I went skiing I noticed ski goggles kind of look cool, and are the perfect form factor for pass through ar. I think that’ll be the form it finally takes when Apple makes it cool.

Show up to your next work meeting wearing ski goggles and report back to us.

AFTER the Apple ads..

I've always thought the winning format would be akin to ski goggles as well. If Apple does it well, I can see it becoming popular.

Absolutely. When you look at the compromises involved with optical using today’s technology, the rumors of Apple pursuing passthrough for their first set of goggles make perfect sense.

I wouldn’t have believed this until I used the Oculus Quest. But now it seems fairly possible.

This seems like a tech-community-only opinion to me, the same thinking that is currently killing Magic Leap. If you want to build a mass-market AR headset, you should have one priority: make it look like glasses. People like eye contact. People don’t like looking like alien telemetry pods are eating their eyeballs. Once you have that, then you can worry about things like resolution, fov, and color reproduction. The only companies I’ve seen understand this are Snap and Amazon, with the caveat that they forgot to make the glasses do anything useful.

I suspect this is why Apple is having such a hard time with their headsets. Above all else, Apple doesn’t release products that are dorky.

> Above all else, Apple doesn’t release products that are dorky.

https://www.apple.com/nz/airpods/

Friendly reminder that AirPods alone are a bigger revenue line item than Netflix.

Yes, and they're dorky as all hell.

Can you recommend an audio device that you are sticking in your ears that differs significantly from AirPods that is not dorky?

“Being cool” is the biggest appeal of AirPods. As pointed out elsewhere on the thread, they made wireless Bluetooth so fashionable celebrities wear them in Zoom calls. Before AirPods wireless Bluetooth was not cool. AirPods are a carefully engineered phenomenon that Apple will be seeking to replicate with VR/AR as much as possible.

Airpods shows what Apple marketing can pull off as far as defining what cool is. No one would have though Bose Airpods were cool.

I don't think there is a market for AR on a consumer's face except like a side display like Google Glass. I think the market for AR is in industrial operations, military, etc. where people won't care what they look like.

Besides which, the military is historically a leader in tech that sometimes makes its way back to the civilian world. Notably, the internet was originally ARPANET, funded by the DoD, and the TCP/IP protocol was designed with military considerations in mind (e.g. designing a protocol that would be resilient to failures in parts of the network).

A company that secures a steady contract delivering AR equipment to the military will then have the runway to polish the packaging until it's palatable for consumers.

I want product reviews as I look at items on shelves in stores, without holding my phone in front of my face the whole time.

Many people already more or less have that: Amazon.com, at home on their couch.

I wonder whether there still will be stores selling things you want to spend time to choose between options by the time that tech exists.

> If you want to build a mass-market AR headset, you should have one priority: make it look like glasses.

I so vehemently disagree with this because it's the same category of thinking that keeps XR in the dark ages.

The capital "P" Problem in XR is content. Behind every major innovation in content consumption technology has been content to consume. As of yet, there is no such content for XR, and the one thing Magic Leap did (and this was a few years back) was to invest in people to create content for their platform and dev tools to make it.

You can have the ugliest piece of garbage tech tethered to a desktop PC, and people will pay to use it if they get to experience some piece of content created by a team of artists that can't otherwise tell the stories or convey emotion as effectively in any other medium.

Throughout the history of consumer media you find that a successful medium is determined by what art is created upon it, not by its technical aptitude. There is always something better, but what makes it useful is what artists can do with the limitations of the medium today.

That's why we see this pivot to the nebulous "enterprise" (without real success). It's because they haven't found artists that can create real content yet, at least nothing people want to experience. The most successful so far was LBE (location based entertainment) which was seeing good growth pre-covid.

I agree with you where VR is concerned, just not AR. This is why I don't like the term XR, it conflates markedly different devices. The Quest 2 is good enough that content is becoming its biggest problem, and if it gets some really killer games or social apps it could gain serious traction. It already has way more satisfied users than the entire AR industry, because AR hasn't reached that "good enough for content" milestone.

I follow VR more closely than AR. It has a ton of examples as to how people will not pay to use ugly garbage tethered to a desktop PC. This is why Facebook's pivot to standalone headsets has been so successful. PC tethered VR apps look better, are more interactive, and are vastly larger and more in depth. No one plays them, because a slick single device is more important than every other factor. When I used to demo PC VR apps to people, they'd be blown away. They'd ask how to get one, I'd explain the PC and the base stations, and their faces would fall. No one ever bought one. When I do the same demo with the much worse apps on the Quest, people buy Quests.

Magic Leap's investment in content is another example. What good did that investment do? None, because the Leap itself was too big and clunky for people to buy.

Why not just put up a head tracking 3DTV at that point? Only caveat is it'll look like a window to VR rather than an overlay to real world, but from psychological standpoint it might favorably communicates that the local and remote is physically separate with a substance inbetween.

And what about seeing peoples' expressions? Optical AR is an order of magnitude more likely to be acceptable in a social situation.

If you are talking for many "game playing" applications you may be correct.

But it be ridiculous to thing that people are going to be going out in public with a VR headset with cameras (known as Passthrough AR). On top of a number of human factor issues, it would dangerously block a person's view of the real world.

Just crank the FoV slider up to maximum.

> Passthrough AR with VR goggles and 8K cameras on the outside will be the way that wins

None of the big companies are interested in what that would be though.

They want the equivalent of the next smartphone revolution so it needs to be like a pair of glasses.

Why not just augmented glasses? I mean with a transparent display that is just a display, showing the time and notifications in one corner. A larger view if pulled up through a menu.

I bet they could make them look very slick, since the hardware requirements would so much less arduous. There was a set of 'augmented' bicycle glasses here on HN some months ago and they honestly looked pretty compelling.

For years I haven't been excited by anything in the VR space. I tried them, even with the running apparatus and the slippery shoes, and it was fun. But that's it.

Nor the Leap, the Quest or the Hololens interest me very much. It doesn't bring much into my life, but promise to take a lot from it. Besides, the fact it's backed by companies with a not so good track record at respecting their users, and some of them, at making closed proprietary platforms, doesn't make me want to dive in.

But recently, 2 projects started to get me excited:

- the Tilt 5: https://www.youtube.com/watch?v=Fzr-IeyWj40

- the Simula: https://simulavr.com/

They are both trying to be very good at a very specific use case, which I think is a better way to approach the problem right now since we don't have VR solved. The vibes their creators give is also much more human, in an "internet from the 90" kind of way.

The Tilt 5 already works well, and people using it all report that it's the best VR experience they had to date. They also make it easy to port games from existing platforms, and actually talk to communities such as war-hammer 40k players and board game lovers, instead of staying in their ivory tower wrapped in secret to finally come down as our lord and savior. Plus I love board games.

The Simula is only a concept, and it may very well end up being vaporware like many kickstarter projects. But I want it to succeed so badly. First, it's just linux. No proprietary platform BS. It tells me it wants me to use my current apps, and works with FOSS and standards instead of trying to spoon feed me some expensive locked down metaverse crap. And I can really see the value of unlimited screen space and a full immersion work experience, especially since the device is optimized for text readability.

Just added myself for notification when their kickstarter begins (Simula). I wish they went with an AMD APU, instead of Intel Xe graphics. The youtube gameplay comparisons seem to give AMD a big edge in well... everything. I'll definitely be backing this. I want to play Tabletop Simulator on all of my walls and ceilings. :-)

(Thanks for telling me!)

An APU is on the table for future iterations :)

The Simula webpage made me smile, if you spend too much time with your neck moved like it is shown, you'll regret it sorely at the end of the day.. It's probably much better for the neck to "move the world" than to move your head to look at another window!

The Tilt5 looks fun but seems like an expensive toy. But the simula I would actually buy. This was the exact usecase I wanted when the first headsets were coming to the market. Hope they succeed

Toy indeed, but that's what VR headsets have been until today anyway.

And actually 3 times less expensive than the competition.

I have been super skeptical about VR since the 1990s, but I see your point about the Tilt 5. Board gamers are a serious market, one willing to spend. And designing for such a narrow use case lets them dodge a bunch of the hard things about VR. I'll be very interested to see if they can create a sufficiently large niche market to sustain the company.

Karl Guttag is a very experienced and interesting person. "The AR Show" has two very good interviews with him the cover a lot of ground on displays and history of computing.

https://www.thearshow.com/podcast/111a-karl-guttag-part-1 https://www.thearshow.com/podcast/006-karl-guttag-pt2

I dont see how this ends up any differently than ML1. Relatively good headset that has some technical advantages, and is obviously impressive at some level for what was produced with smaller amounts of money, but for what? It still has major disadvantages, and with the stability of ML compared to Microsoft, I wouldnt particularly put long term faith in the former if I was a company looking to purchase these.

The whole thing definitely looks like a ML1 holdover from when they thought they were going to somehow hit the consumer market.

Personally I think that a simple birdbath design with a decent display is one of the better ways to enter the market if you want to move a relatively large amount of units.

Also Karl Guttag is an amazing blogger. His articles are the sole reason why I got seriously interested in AR and NED hardware. Now Im looking to make part of a career out of it.

Wow, talk about a tech news blind spot: I honestly though Magic Leap wasn't even around anymore. Last news I heard about them was they burned through $billions, their product woefully (almost fraudulently) failed to meet any of the hype they set out, they sold a handful of units, and then the company folded. How on earth are they still in business?

>I honestly though Magic Leap wasn't even around anymore. Last news I heard about them was they burned through $billions, their product woefully (almost fraudulently) failed to meet any of the hype they set out, they sold a handful of units, and then the company folded. How on earth are they still in business?

AR/VR in general has hit the "slope of enlightenment" portion of the hype cycle [0]. Talented people are still diligently working on this stuff behind the scenes and advancing the state of the art, even though mainstream adoption has a ways to go.

[0] https://hobi.com/gartner-hype-cycle-emerging-tech-potential/...

In my experience when someone tells me where something is on the hype cycle (especially if it's on the left hand side) I immediately know it's going absolutely nowhere. My hit rate with this heuristic is upwards of 95%.

I'd like to believe that's true, but has any meaningful progress been made?

VR games seemed to have plateaued in novel immersive experiences - people show off their Oculuses and Vives with the same tech demos.

The Meta VR presentations are embarassingly primitive, marginally better than Second Life from 2 decades ago.

The AR demos I see on Android and iOS (measuring the length table; having a dog run around on top of it) have not improved in fidelity in 3-4 years.

I think what Microsoft is doing with Hololens for the industrial sector is VERY cool, and is a meaningfully significant branch: Having a mechanic receive plans remotely and projected on top of an engine or a door and get overlaid instructions for repair is going to be either revolutionary or a dystopian collapse of skilled trades as every company decides to replace training and high paying trade jobs with Hololenses and remote experts. I'm cynical, but at least it's going somewhere impactful.

But what's happening on the consumer side? Facebook's vision is nauseating. Snapchat, Google, and Apple have gone totally quiet. I assume they're in stealth mode waiting to unleash something and blow our minds? That would be pretty great I suppose.

For sure. if it is in the plateau phase, it seems to me it has plateaued at a very low level. And indeed, I think the hype cycle chart is misleading, in that it suggests that all hyped technologies eventually have productivity that approaches the hype. But things like jet packs, undersea living, and flying cars all had periods of hype and never went anywhere. Or more prosaically, 3D TV was intensely hyped and then vanished like it had been covered up by the Illuminati.

I think the lack of XR adoption is especially obvious given the pandemic. Remote working? Video calls? Ubiquitous. People snapped up a ton of Switches early on, and then it's been PS5s that nobody could get enough of. But VR? I know a lot of game players from 13 on up, but I don't know anybody for whom facehugger VR is a daily driver.

> But what's happening on the consumer side?

I’m biased but I think Tilt5 has the most interesting AR approach. And then brightest display, of course.

> people show off their Oculuses and Vives with the same tech demos

Do they now?

I mean the most significant VR game is Beat Saber (exercise is a good niche it turns out) but by now the show-off game would be Half-Life Alyx rather than something tech-demo like.

Beat saber looks fun as hell, but I don't think it's a VR seller. It reminds me of Wii Tennis - we had motion controls for 15 years now.

Half Life Alyx is to me a perfect representation of the problem - all the best VR games are basically First-person shooters. Which also happen to be the most popular games in general, I agree.

But it's a dead end. It's where we've optimized because it's where the technology limitations are the least compromised.

The games where true VR immersion would be beneficial - flying, swimming, exploring complex 3d spaces - are not immersive enough with the technology we have it seems.

I remember a very famous VC (who is still amazingly rich and famous) claiming the Magic Leap was a "bigger leap in tech than the iPhone."

Surprised nothing happened to him.

Wasn't everyone (including Steve Jobs) crazy over the Segway?

I think [the design] sucks. Its shape is not innovative, it's not elegant and it doesn't feel anthropomorphic. —Steve Jobs

https://hbswk.hbs.edu/archive/steve-jobs-and-jeff-bezos-meet...

Actually I think it was Jeff Bezos who liked it, and Steve Jobs thought it was terrible..

I've barely cracked the surface of this and it's already one of the best analytical breakdowns of new tech I've seen. Thank you, kguttag!

This is obviously a very hot market. Where do you see AR vs VR in 2 years? Five years? Do you think these companies will have supply chain issues? Do you think AR will face adoption issues? It seems like the lack of light, "glowing eyes", eye tracking cutouts, and other issues will be big hurdles for all companies in this space.

I'll come back with more technical questions after I finish reading. This is great.

While the market is "hot" in terms of awareness, it is still a very small market facing major technical challenges. I think AR while useful in enterprise markets (measured in the 100's of thousands a year) is a long way from being ready for the mass consumer market.

I am seeing a lot of progress in some areas and will be publishing an article later this week on them. In particular both Dispelix and Digilens have made considerable progress on the "glowing eyes" issues (Dispelix all but eliminates it). Avegant has a very nice small LCOS light engine that pairs nicely with the Dispelix waveguide.

I think the biggest problem for AR is that the expectations are very high and the physics is very tough. Many physical optical features within a few wavelengths of light were diffraction ruins everything.

It seems to me like a regular VR headset + dual cameras on the front might be able to get much better results than actual AR?

Even if it did, people won’t wear them in public. Google Glass looked half normal and people were still getting accosted in restaurants.

Yes, but that problem was much more “Google” than “glass”.

Perhaps so. I’m just skeptical that the viewed would accept such VR-pass through style goggles even if the viewer finds them innocuous.

I was working on enterprise Google Glass apps in 2014. The problem was lack of applications. You could do very few things with it beyond showing some text and pictures.

That doesn't matter for industrial applications though.

People freak out at the idea of cameras of some random dude constantly watching them.

This is a bit funny because they'd be watched by multiple cameras in any restaurant or supermarket. OTOH the viewpoint of the overhead monitoring cameras is very distinctive, and the resolution is usually barely enough to see a face. The Glass's camera gave more "normal" and higher-resolution footage.

There is no irony here.

People freak out because it's worn by a _person_ who, specifically, is watching _them_.

You'd probably react similarly if somebody, during a party, for no reason kept pointing a recording microphone at you, even though there were voice assistants like Alexa in the room.

People are discussing whether AR device could be worn in public. But I’m wondering: should it? I mean, I’m the last guy to question novel technology. I was desperate to get the first smartphones. But why would you want an always-on screen in public?

I think this is a "grass is greener" type argument. There are also massive problems with pass-through AR (VR with cameras).

I'm not entirely sure it is. Pass through has serious engineering hurdles to get over first with latency, power consumption, weight etc.

But the waveguide method has limitations rooted in physics and math that won't change until 30%+70% stops equalling 100%.

I know which one I'd put more Hope's in.

Did you get to try the Avegant prototype? It sounds quite promising.

Yes, I took the picture of Avegant used in the article. The Avegant and Dispelix waveguide combination looks pretty good. It is only a prototype without any tracking/slam. It is a display only demonstration by a component company.

I was impressed by the size of the Avegant engine and the transparency and lack of forward projection by the Dispelix prototype. They are claiming they will get 2,000 nits to the eye out of the design which should be good enough for outdoor use IF they have some form of clip-on sunglasses (2,000 nits is not enough for outdoors in full sunlight without some help).

The current Avegant engine has an optical component in that was depolarizing the light from the LCOS display and losing contrast so they wanted me to wait to take through the lens pictures. The image from the current prototype looked sharp but did lack contrast.

I realized a few days ago that AR is more for robots and automation than for humans.

Imagine a company supporting ar in there product for maintenance.

They need to describe there Maschine to a point that AR can create an overlay and a step by step guide.

If you do this so that your technician can do it so can a robot.

If we ever make the effort to export 3d models from things and make them available, I believe we will see ML extracting repair steps and robots executing them before humans walking around with AR gogles.

Why would a robot need AR at all? And the challenge with building a repair robot is not to provide it with information about the item, it's in understanding the information and physically manipulating the item.

A robot doesn't need ar.

That's my point I tried making.

As soon as you actually use ar for people, at that point you have enough data for a robot to just do it instead.

As soon as you start creating ar support for it, you no longer need it.

I'm saying that this is not correct at all. There is no lack of data, but it's very very difficult to make a robot that can autonomously manipulate things well enough to repair things. It's not lacking data, it's about physical control.

Do you think if they focused on creating a monochrome AR display first for applications in engineering and medicine it would have been adopted creating a path for the company to later research and develop a multicolor device?

Seems like they lost. Wouldn't it have been better to start simple? They never released a MVP.

I don't think the issue is color. Their original product was trying to differentiate on Vergence Accommodation Conflict (VAC) and did a poor job of it. They were also going after the consumer with a product that was always going to be way too expensive for consumers with an image quality that was going to be too poor. They also made what I think is a bad set of trade-offs in terms of ergonomics and human factors.

The problem with medicine is that is it a tiny market. You can't justify the type of money they were trying to raise for that market. Under Abovitz, they were always a swing for the fences company.

All the above said, nobody I know of is making money selling AR headsets today. They are either subsidized by VC or other investment money or by big companies funding R&D efforts. The market is not other product areas where you can make money with an MVP and grow the product.

I vaguely remember Rony and friends reversing the concept of the endoscope product they developed prior to have the moving fiber optic strand project light into the side of a lens which would channel the light and reflect it into eyes instead of capturing light and projecting it onto a light sensor in the case of the endoscope. Everything they did had fiber optic strands for RGB on that original idea. Having one color would have been an easier place to start. Nonetheless, it seems like they had other more difficult unrelated problems such as Vergence Accommodation Conflict.

There used to be a number of good monochrome AR projectors for industrial and military use last few years.

They look pretty unlike a sleek consumer device, and are expensive, say, $1000 for just the display. For a business, it's peanuts; for a consumer and even prosumer device, it's a pretty significant sum, for a monochrome display not good for gaming.

Karl, do you have any thoughts on the Simula One project? I see that you specialize in AR, but it looks like there is massive crossover in tech. The Simula team pitches this as the next popular form factor in business computing. Anything to that?

As you wrote, I don't follow VR.

That said, I'm very skeptical that a VR headset will work in place of a computer monitor for long term use with business computing.

Nice article.

btw, ML has gotten $3.5B in funding!

https://www.crunchbase.com/organization/magic-leap

I assume in this case your acronym “ML” references the company Magic Leap?

I treat any and all Magic Leap announcements with a pinch of salt these days. I'm old enough to remember their then stealth mode claims to have invented a device that would revolutionize society, put a stop to world hunger, prevent war, abolish capitalism, etc. etc.

> The more I think back about the presentation and other aspects not covered in this article, such as their ability to design and manufacture their own waveguides, the more I agree with the people that think the presentation was more of a “For Sale” sign.

Which differs from other such presentations how?

From the article: "Sometimes a very open presentation is given by a company trying to sell the company."

Reply on news.ycombinator.com