I think this can be solved with electro/photochromic lenses. I don't know though.
Face-wearables - will any survive or thrive (Rift / Glass / HoloLens / Vive / Apple Vision etc)?
- Thread starter neye_eve
- Start date
Would glasses need the same R1? My understanding was that the R1 was responsible for gaze tracking and then compositing the overlays with the pass through video hyper-fast so that the pass-through video is so low latency it’s perceived as real time. But with AR glasses there is no compositing of multiple video streams and there may or may not be gaze tracking.
I would expect gaze tracking, hand tracking, SLAM, etc., most of the XR stuff that R1/XR2 chips do that aren't in the basic cell phone chips. Not passthrough, sure, but most of the rest.
I have found the craziest corporate mashup of all time Razer (the LED-festooned mouse people) x ResMed (the CPAP/Sleep Apnea people) make a facial interface for the Quest 3:
View: https://www.amazon.com/Razer-Facial-Interface-Meta-Quest-Comfortable/dp/B0CW2BBHZ6/?th=1
Honestly though, I hear this thing is amazing, but I dunno about spending $70 on a FI.
View: https://www.amazon.com/Razer-Facial-Interface-Meta-Quest-Comfortable/dp/B0CW2BBHZ6/?th=1
Honestly though, I hear this thing is amazing, but I dunno about spending $70 on a FI.
I could see a light sealed but fully ventilated FI being useful. Eyeball dryness/sweat limit in headset time and I think current ventilated FIs leak light or suck at moving air. Still, open periphery ftw.
I gotta say...those Resmed devices are nice. My CPAP has literally changed my life.
Yes. INSTANT quality-of-life changer.
I went from constantly feeling whacked out and underpowered to AWAKE, literally overnight. I didn't know that state existed outside of adrenaline rush.
I found this interview with Boz where he describes some of the new tech required to get AR glasses to work, and I guess I can kinda see where the billions of dollars are going, or at least a few tens of them
www.matthewball.co
Interviewing Meta CTO Andrew Bosworth on the Metaverse, VR/AR, AI, Billion-Dollar Expenditures, and Investment Timelines — MatthewBall.co
What it will take for headset sales to cross 100 million annually, the specs of Boz’s dream headset, Meta’s spending on Reality Labs, whether and when developers might get access to the Quest’s raw camera feed, the many inventions required to ship optical AR glasses, the role of AI, and more.
www.matthewball.co
I wonder if someone at Razer has a CPAP or someone at Resmed liked VR! LOL
Apple Vision Pro U.S. Sales Are All But Dead, Market Analysts Say
Anyone in the U.S. who wanted a $3,500 Vision Pro may already have one. The only hope for increased sales is a rumored cheaper model in 2025.
gizmodo.com
It looks like demand for the AVP is dropping even more dramatically than projected. We're not quite to "everyone who wants one has one" but it's getting close.
"Projected" = "we made up some numbers". Did we expect any different? It's a $3,500 first-gen device. I'm keenly interested in the space, but not at that price point for the bleeding edge first model.Apple Vision Pro U.S. Sales Are All But Dead, Market Analysts Say
Anyone in the U.S. who wanted a $3,500 Vision Pro may already have one. The only hope for increased sales is a rumored cheaper model in 2025.
It looks like demand for the AVP is dropping even more dramatically than projected. We're not quite to "everyone who wants one has one" but it's getting close.
From Apple? Honestly yes. They have one of the best market analysis teams in the business. So either they got hit by their own RDF, or the reality of the device tanked further demand.
Yeah 75% below projections is bad (and they're only forecasting through August, but don't expect holiday sales to make up much if any ground.)
What isn't clear, and to be fair-- is whose projections and when were they made. Apple did already revise down their expectations so is that off of the revised number?
They also talk in nebulous terms about units returned since nobody other than Apple seems to have hard data on that.
What isn't clear, and to be fair-- is whose projections and when were they made. Apple did already revise down their expectations so is that off of the revised number?
They also talk in nebulous terms about units returned since nobody other than Apple seems to have hard data on that.
Bolded for emphasis. And that's a bingo; no one on the outside has any clue whatsoever what the internal projections were and if they've met them, are slightly short, or way short. Nor does anyone outside of Apple have hard data on sales figures or returns. It's all guess-work, and likely with huge error bars on a new device.
I don't disagree. I'm just noting that reporting is amping up around the idea that Apple is running low on people willing to pay for an AVP. If the pricing structure were different, the demand curve would probably look different-- but that's not the case.
Apple Vision Pro sales seem very reminiscent of Macintosh sales following it's introduction.
Reporters gonna report.
Well, of course? I'm in that bucket. $3,500 is a hard no, $1,999 would have been a soft no, $999 would have been an immediate yes. That's kinda why I'm so skeptical of any reports around "missing sales targets!!!". We have no idea what the actual sales targets are, and what the sales figures are. If Apple miscalculated badly and the market was softer than they anticipated, then bad on them. If it's 3rd party projections and prognosticating that they've now "missed", then... who cares?
This may be another Newton or Cube, an overpriced product that failed.
With the Newton, it at least presaged the migration of computing to mobile devices.
Will the AVP also be predictive of HMD devices that the mass market adopts or will it's predictive value be that the category has no future?
AVP could be priced at $1000 and it could lead to a lot of impulse purchases, get out tens of millions of devices out there.
But people might relegate it to a part-time or occasional entertainment device, not a general computing device.
Or people might stop using it altogether, which is apparently the case with a lot of Quest devices.
With the Newton, it at least presaged the migration of computing to mobile devices.
Will the AVP also be predictive of HMD devices that the mass market adopts or will it's predictive value be that the category has no future?
AVP could be priced at $1000 and it could lead to a lot of impulse purchases, get out tens of millions of devices out there.
But people might relegate it to a part-time or occasional entertainment device, not a general computing device.
Or people might stop using it altogether, which is apparently the case with a lot of Quest devices.
As I've said before, for the AVP, I don't think sales numbers really matter. It's getting to PMF. Now that the thing has been in-market for about 6 months, it's probably a good time to assess how that's going.
Has the AVP found PMF? No. - Obviously not, but that's fine, even the Apple Watch didn't find PMF until the second revision, and that was incredibly fast for any hardware product.
Is the AVP moving in the direction of finding PMF? I would say no as well. None of the directions they dangled in their initial release have had any meaningful traction, and certainly not enough to make product decisions for a v2. This is a really rough place to be in - I always talk about 2 of Apple's other pre-PMF products, the Apple Watch and the HomePod, and where the Apple Watch moved in the direction of finding PMF immediately, the HomePod did not, and to this day still hasn't, and thus has languished.
Is the AVP making moves to bend its product trajectory towards one that could find PMF? Yes, this is the second derivative, and no, I don't think so. At WWDC, Apple opened up some of its APIs to business/industrial cases, hoping that maybe they can open a new direction and find PMF in business scenarios, but this is speed running the Hololens playbook, which MS also did once they realized that all of their directions had failed. They tried targeting business/industrial first and found some traction there, like Airbus, but the market is niche/small. They landed on military in the end, which might work, but then they've so narrowly defined "Market", that no matter how lucrative it is, it won't really move the needle as a product category. As a consumer product company, I doubt this is the way Apple wants to go.
Has the AVP found PMF? No. - Obviously not, but that's fine, even the Apple Watch didn't find PMF until the second revision, and that was incredibly fast for any hardware product.
Is the AVP moving in the direction of finding PMF? I would say no as well. None of the directions they dangled in their initial release have had any meaningful traction, and certainly not enough to make product decisions for a v2. This is a really rough place to be in - I always talk about 2 of Apple's other pre-PMF products, the Apple Watch and the HomePod, and where the Apple Watch moved in the direction of finding PMF immediately, the HomePod did not, and to this day still hasn't, and thus has languished.
Is the AVP making moves to bend its product trajectory towards one that could find PMF? Yes, this is the second derivative, and no, I don't think so. At WWDC, Apple opened up some of its APIs to business/industrial cases, hoping that maybe they can open a new direction and find PMF in business scenarios, but this is speed running the Hololens playbook, which MS also did once they realized that all of their directions had failed. They tried targeting business/industrial first and found some traction there, like Airbus, but the market is niche/small. They landed on military in the end, which might work, but then they've so narrowly defined "Market", that no matter how lucrative it is, it won't really move the needle as a product category. As a consumer product company, I doubt this is the way Apple wants to go.
We've actually had a couple of decades of VR games and content and over a decade of AR content.
Nothing has really taken off and driven sales of headsets other than newer capabilities or performance levels being available at lower price points or lighter, untethered form factors.
The lack of content which would drive sales of devices is a sign that there may be no such killer app or apps which would make people at least adopt HMDs as supplementary devices, let alone as replacements for their desktop OS or mobile devices.
That may ultimately be the resolution to the question posed in the title of this thread.
Nothing has really taken off and driven sales of headsets other than newer capabilities or performance levels being available at lower price points or lighter, untethered form factors.
The lack of content which would drive sales of devices is a sign that there may be no such killer app or apps which would make people at least adopt HMDs as supplementary devices, let alone as replacements for their desktop OS or mobile devices.
That may ultimately be the resolution to the question posed in the title of this thread.
Well, one face-wearable, (Meta smartglasses) seems to be on a path to PMF if not there already. The first generation did not find PMF, but the directions (always there camera, always there headphones) did point towards PMF, so with the second generation, they leaned into that and seem to have come up with a genuine success.
I still say it's much more Newton-like, and the Mac-like comparison isn't all that close.
The Mac was introduced into a 1984 market where Apple already had established products (Apple II series). While they were cheaper, all personal computers were expensive, and a ballpark price of $1500-2000 including monitor and accessories for a IIc ($6000 in adjusted 2024 dollars) was certainly cheaper -- $2500 base, $2800 bought you a "Fat Mac" 512k (about $8k adjusted), which was the first really useful model.
It was that the Mac was such a radical departure from the mass-market Apple II both in hardware and software, and all of the software and accessories to make it work took time to emerge. The AVP at least superficially does most VR headset/AR enabled tasks that other products have done before-- arguably much better in lots of ways-- but isn't something that feels all that exotic. Apple can leverage iPadOS and software, etc. even though it's not nearly as rich of a software ecosystem as Apple would probably envision a mature AVP to have.
It's newish (I mean it was new a few months ago), and new for Apple in this segment, but they're not dropping a product for which there was no existing mass-market competitor the way they did with the Newton. If Apple had a ~$999 VR headset that (for example) didn't do AR in a meaningful way, and generally had a feature set more limited than the AVP, then the 1984 Mac comparison might be closer. It's also Apple entering into a market where the competitors have largely decided the pricing structure and set expectations.
It was also not clear that the Mac was going to be the only "way forward" for Apple in 1984. They put a ton of marketing behind it, but the Lisa had already failed and they were printing so much money with the Apple II series that they had no real intention of dropping it cold in favor of the Mac. The IIgs came out two years later in 1986, even.
I'm well aware of the history having been a software developer and having purchased, and receiving, my first Macintosh in April '84 and my first Macintosh II inJune '87.
Congrats? Are you going to have a constructive response about why you think the AVP is more Mac-like than Newton-like in terms of market positioning? Even then the Newton comparison isn't ideal, as competitors in the VR space exist when the Newton tried to forge a new product category entirely. Competitors in the AR space are much more limited in terms of capability with the arguable exception of HoloLens and its military offshoots which are a complete other AR methodology, but VR is not remotely new.
So if, say, Apple saw a lot of enthusiasm around content consumption on AVP, that would be a signal to pivot and make something more like the XREAL Air glasses? Or if people seemed to be using it mostly to play games, that would be a signal to make something in the price range of a game console?
I think you're coming at this from the wrong direction. Apple is interested in HMDs because they might be the next important form factor for general-purpose computing. We already know what general-purpose computing platforms are good for, and it's pretty obvious how adding unbounded volumetric display capability has benefits for multitasking, media consumption, gaming, shopping, navigation, multiuser collaboration, and other existing use cases.
So Apple doesn't have to divine how to proceed here through careful reading of early-adopter behavior — which can be a trap as often as not if your goal is to build a mass-market product, since early adopters are atypical. They know what value they're trying to deliver. Really, they're already delivering a lot of it, it just comes with too many downsides right now. High price, discomfort, lackluster battery life. The path forward is clear, at least in broad strokes. Address these limitations as tech permits, keep refining the UI, add spatial features to apps where relevant. Eventually they'll have something broadly compelling.
The appeal is not in being a general computing platform.
iPad can edit Microsoft Office files. That’s nice, but not why the platform exists.
Apple needs to assess what people will primarily be buying AVP for — over or in addition to existing general-computing platforms. So that they can focus on those strengths.
Being able to do all those other things is beneficial, but it may not even be the most important reason for the product, even though that may push people over the edge (think: professional buying it for the virtual Mac display(s), but being able to also use it as a stand-alone entertainment center when on the road). There will be a number such scenarios.
iPad can edit Microsoft Office files. That’s nice, but not why the platform exists.
Apple needs to assess what people will primarily be buying AVP for — over or in addition to existing general-computing platforms. So that they can focus on those strengths.
Being able to do all those other things is beneficial, but it may not even be the most important reason for the product, even though that may push people over the edge (think: professional buying it for the virtual Mac display(s), but being able to also use it as a stand-alone entertainment center when on the road). There will be a number such scenarios.
It would have to be electrochromic. Photochromic are too slow and don't work well when you're behind glass (e.g. in a car) because what they respond to is UV. They also go much darker and respond really slowly when they're cold. I don't know if there are good enough electrochromic materials to really work well in low light though. In low light conditions, you really need them to pass almost all light. In full sun, they need to block probably 70% or more to make overlays visible. In terms of technical performance, you probably can't beat video passthrough like Apple does, but I think that kind of overlay is not really tested enough to use in real world dynamic situations because of the risk of completely blocking what you need to see. Apple seems to think it's not ready/not safe enough at this point and I agree.
Probably not. I think there's a lot less computational demand in that case. But possibly there are different processing architectures that can do it all with less computing even in the video pass through approach. The Vision Pro was Apple's first version, and I'm guessing they did what lot of companies do when trying out a new technology - throw all the resources at it you can fit and then let the SW people make it work. Find out how much resources it really needs by either running up short or seeing that loading never reaches more than X% then maybe save money on a cheaper processor and less memory because you've proven you don't need it.
Market research is more than just making up numbers. It's based on interviewing potential customers. And your complaint is definitely one they look at. The market for a $1000 device is typically a lot bigger than for a $3500 device. IMO they're kind of trapped by iPhone pricing on the low end. If they were to price it too close or below an iPhone, they'd have people asking why an iPhone costs so much when an AVC (Apple Vision Consumer!) costs only a little bit more and obviously has much more expensive/advanced hardware. But a company that sells no phones or much lower priced phones could get in at the $1000 level, presuming they could build it and make a profit at that price without harming their own phone's pricing structure.
This is the Apple pre-PMF playbook that they pioneered and perfected so yes. But in the case of the AVP, for content consumption, they wouldn't pivot, they would lean into that direction as it's one of the directions they launched with to get signal on. Gaming was not a direction, so that would be a pivot, but that was never in the cards anyhow so Apple's developers didn't really go that way.
The signal from AVP has not shown those things to be viable though. This is good data as well. Still if you're saying AVP is mostly a dev kit/technology demonstrator/time machine (in Boz's parlance), and therefore should not seek PMF or sales numbers, then it's not a product and should not have been marketed or sold as one.
To be clear, by general-purpose computing I don't just mean office documents and web browsing, I mean everything people commonly do on iPads and iPhones, including media consumption.
The Watch and the HomePod never had the potential to be general-purpose computing platforms, due to limitations inherent to their form factors. They're not useful references. It would be a huge mistake to shift the Vision platform away from general-purpose computing. Something narrower, like TV glasses, wouldn't be of meaningful strategic value to Apple, and any narrow product in a form factor that can, in principle, support a general-purpose computing, is very likely to eventually fall to a bullet point on the feature list of a general-purpose platform, as we see from the many products that have been subsumed by smartphones.
Maybe if Apple were a startup that desperately needed market traction with anything to validate their existence as a company, a temporary shift toward something less ambitious might make sense. But they're not. There's no reason for them to take their eye off the prize here.
As I've said before, it's the Macintosh Portable. It's a conceptually sound product, the first version of which is too clunky and expensive to find a large audience. Should Apple have concluded five months into selling the Portable that notebooks weren't a viable product category?
Even as someone who is pretty bearish on HMDs, I'd have bitten at $999. The problem is that these devices are worth less than their bill-of-materials. I wouldn't buy a Quest 3 at any price.
If $70 is too much to ask for for a more comfortable FI, then there is no real market for these things.
Macintosh is not a mass market device, though it's US market share is up to 14%.
Apple's playbook doesn't say to narrow the function of the product, it says to invest in the directions that are moving in the direction of PMF. Look at the Watch where they saw health as the main thing to lean into - the other stuff didn't go away (well luxury eventually did). Also look at Apple's other pre-PMF success, the iPhone. Customers wanted apps and they really went for it there. The other directions were still there and got love, just less.
This is a goalpost move. I wasn't talking about a product category, I was talking about a product. If Apple has no intention of seeking PMF from the AVP, it's not a product. As for the product category, I have said I think that VR/MR headsets are an evolutionary dead end, but Apple may disagree, however to be right, they need to find PMF somewhere, and if they're not trying to do that then AVP is a data gathering apparatus.
I don't recall the Mac Portable. Did it show signs of finding PMF in its first 6 months in market?
Mac Portable was DOA more or less which is why you don't remember it. It had some niceties, but it was a very expensive portable (not notebook!) computer at a time when clamshells were becoming more common. Supposedly it entered development in 1986 and wouldn't have been so out of place had it not seen delays.
AVP isn't behind the curve like the Portable per se, but I don't see how a anything that costs as much as an M2 Ultra Studio can survive without having a professional use case. I work with some people that handle in-house VR software, nothing too exciting, and AVP doesn't support OpenXR which makes it a total non-starter for them. I would be surprised if there aren't other organizations that have the same issue. The hardware is very nice, but the fact that Apple can't or won't work with Khronos due to some prior legal entanglement (It's mentioned at the bottom of page two in these WebGPU meeting minutes) limits the appeal.
AVP isn't behind the curve like the Portable per se, but I don't see how a anything that costs as much as an M2 Ultra Studio can survive without having a professional use case. I work with some people that handle in-house VR software, nothing too exciting, and AVP doesn't support OpenXR which makes it a total non-starter for them. I would be surprised if there aren't other organizations that have the same issue. The hardware is very nice, but the fact that Apple can't or won't work with Khronos due to some prior legal entanglement (It's mentioned at the bottom of page two in these WebGPU meeting minutes) limits the appeal.
Supporting a vast ecosystem of third-party apps is the opposite of leaning strongly toward any particular use case. The framework you're applying here just doesn't work well for analyzing classes of device that can host general-purpose platforms.
I don't understand this. Apple's thinking isn't, and shouldn't be, "Well, we shipped this particular hardware, how can we get as many people as possible to use it?" Their priority is, and should be, the platform. Making their version of spatial computing a widely appealing prospect. This is a multi-year project that will inevitably unfold across a series of hardware products. AVP is the best version of it they were able to implement in 2024. One could argue, perhaps, that they should have waited until they could ship something more refined. I disagree. This class of device could eventually replace the smartphone. Trillions of dollars of market cap hang in the balance. Any amount of cash Apple is plausibly burning by being 'too early' is worth it to mitigate the risk of error in the other direction.
It was the first Mac notebook. It was $7300 in 1989 (~$18K today) and weighed 16 pounds. So, of course, almost nobody bought it. But the proper course of action wasn't scrutinizing what the few users who did buy it were doing with it, discovering that (hypothetically) it seemed to be especially popular among insurance adjusters, and figuring out how to lean into that. That would have been nonsense. The proper course of action was to simply work toward more refined versions of the same basic thing. Apple did precisely this with the PowerBook 100 series, which was much more successful, and today, of course, ~90% of Mac units are notebooks.
It's not my framework, it's Apple's, and it works, should they find a direction that works, and doesn't when it doesn't (HomePod).
If they're not trying to get people to use it or learn what it is that people eventually will use it for, then it's a directionless product, and has no purpose. If its purpose is, as you say to make their vision of spatial computing a widely appealing prospect, then as I said above, it's a dev kit/tech demonstrator/time machine, and not a product. So yeah, they probably shouldn't have sold it to the public, though Apple is special, and they can ship science fair projects to their fans. I don't think Apple is doing this though - the idea that this is a marketing exercise to buy mindshare for their vision of spatial computing seems pretty un-Apple, to me.
Lening into insurance adjusters wouldn't make sense unless that was one of the directions they were teasing. Also, I don't know if they had invented the pre-PMF playbook in 1989.
What is an iPhone for? Or a Mac Pro for? Or an iPad Pro for? It’s a weird question to ask about a general purpose computing platform because the answer isn’t a sentence it’s a long bullet list. You wouldn’t think of any of those products as directionless so much as multi-direction.
That’s Apple’s goal with AVP specifically and spatial computing generally - build a platform for which there’s no one answer to “what is it for” because there are a plethora of answers to “what is it for”. Now to get there you need to actually make some sales. So while it doesn’t make sense to pigeonhole AVP development to just one or two use cases, it absolutely makes sense to spend extra effort in those areas that seem to offer early returns. Going back to the Mac Portable example, you don’t want to base all future development on insurance adjusters but it wouldn’t hurt to add a a bell or whistle explicitly for them and then set up a booth at the insurance adjusters convention to sell the things.
Didn't they reduce production? That would be a pretty clear sign that they are below projections.
But what would the price have to be for mass adoption of any AR/VR/etc?
I imagine that price point is different for different people (probably different for different generations too...Boomers and Gen-x can more readily afford to spend more, but I suspect they are less inclined to)
Yeah, that’s not gonna work. This isn’t just “a better mousetrap”, this is “a different mousetrap that costs a lot of money, makes you look like an absolute dork, is uncomfortable for extended wear, but that’s OK because the battery life is ass, and has no obvious use case right now”. Apple needs to show the world what the AVP can do that outweighs its colossal disadvantages when compared to other, none-“stupid looking hat” platforms.
The iPhone, the Mac(Book) Pro, the iPad - these were all things for which the utility was obvious and if it wasn’t, Apple spelled it out for you, and if it still wasn’t obvious, the product itself was at least cool as fuck.