Apple's AR Headset Coming Next Year With 'Mac-level' Power, Report Says (theverge.com) 63
Apple's first AR headset will be released in the fourth quarter of 2022, according to a research note from analyst Ming-chi Kuo. The Verge: Kuo predicted back in March that the headset would be released sometime next year, and is now also providing more technical information on the device. The headset will have two processors, according to Kuo, one with "the same level of computing power as M1" and one lower-end chip to handle input from the various sensors. For example, Kuo says that the headset has "at least 6-8 optical modules to simultaneously provide continuous video see-through AR services." The headset is also said to have two 4K OLED microdisplays from Sony. Kuo cites the headset's "Mac-level (PC-level) computing power," its ability to be operated untethered, and its wide range of applications as factors that will differentiate it from competitors. Various reports on the device have disagreed as to whether it will be wholly independent or rely on an iPhone or a separate processor box to stream content.
Cool. Can I afford one? (Score:2)
Will it be affordable?
Re:Cool. Can I afford one? (Score:5, Funny)
You won't be able to afford the cleaning kit.
Re:Cool. Can I afford one? (Score:4, Insightful)
Re: (Score:3)
It might have an M1 and technically be able to perform at similar levels to an M1 Mac, but battery life and cooling capability will probably hold it back. It's basically a trade off between computing power and bulk, and nobody wants a bulky headset.
Still waiting to see what the killer app for this is. Apple used to launch with something to wow fans with, but lately they haven't really shown anything magical on the software side. The Google magic eraser really stood out this year, but what did the iPhone get
Re:Mac Level Power? (Score:5, Interesting)
Well, I can't speak for anyone else, but what I would pay good money for is some kind of AR system that can learn to recognize real-world locations and objects and let me hang notes, links, photos etc. not just on the objects, but sub-features of the objects. And to be able to download others' collections of same. This has obvious applications in repair and maintenance that could be genuinely useful to me in the really real world.
I anticipate AR eventually providing some kind of object recognition system such that anything you focus on can be described and labeled. Every tree species identified, for example. Or step into any building and the routes to the offices, checkstands, and bathrooms are always displayed on the floor. Navigation routes are drawn similarly. Pick up a car part and it's identified. Look at a mushroom, same. Some kind of quick toggle to shut off all the overlay crap (except for emergency notifications.)
However I am a bit picky and also want the system to be an eyetap, I don't want any parallax whatsoever, especially when taking photos or recording video but also when labeling items. And so far nobody seems to have come up with an affordable and effective way to create those, eyetaps are so far expensive and/or bulky.
I should also add (Score:2)
I'm not going to be buying one from Apple even if it's great, because if I'm going to wear a computer it's going to be FOSS. Even then there's likely going to be closed modules with firmware blobs that present enough of a privacy threat that I don't want to compound it with a closed system on top of them.
Re: I should also add (Score:2)
Re: I should also add (Score:1)
Look, my first smartphone was a Visor, after that, I had an iPaq
Similarly, my first VR Headset was a Cybermaxx, later Vuzix, Oculus and now we have to use Facebook.
AR wise, Magic Leap and Nreal Light - both duds...
As much as I despise Apple and their iSheep community, these guys at least are able to disrupt the market. So, Let's wait ans see where the market will follow.
Apple is not an innovator, Apple is a signal that the market is ripe...
Re: (Score:2)
...Similarly, my first VR Headset was a Cybermaxx, later Vuzix, Oculus and ...
How come nobody uses the xkcd 3-D headset?
https://xkcd.com/941/ [xkcd.com]
Re: I should also add (Score:1)
Man, i rather like to skim the clouds in an airplane. Gives you the real feel.
But, cool idea for a build...
Re: (Score:2)
One of my more mundane ideas is to attach cameras to stalks which stick up off of a vehicle in a "V" so you can get an overview of backing a trailer, operating a fork lift, etc. You'd get a less godlike view, more like a child playing with toys.
Re: (Score:2)
I'm not going to be buying one from Apple even if it's great, because if I'm going to wear a computer it's going to be FOSS. Even then there's likely going to be closed modules with firmware blobs that present enough of a privacy threat that I don't want to compound it with a closed system on top of them.
Right.
Because you're going to take the time to pore over millions of lines of source code in one of these things...
Ya know, your previous post had a lot of good application ideas for AR; but the ridiculous FOSS screed in the Parent just completely devalued your opinion. There won't be a FOSS AR headset for another decade, minimum, and there won't be one worth having, ever, sorry!
Re: (Score:2)
Because you're going to take the time to pore over millions of lines of source code in one of these things...
What you seem to have forgotten, and I'm flattered really, is that there are other people on the planet besides me. I really am honored to have you so obsessed with me and what I'm saying that you forgot about all of them, though.
Ya know, your previous post had a lot of good application ideas for AR
Yes, I do know. I've thought about this a lot.
but the ridiculous FOSS screed in the Parent just completely devalued your opinion
Your claim that insistence on freedom devalues my opinion completely devalues your obsession with me as a slashdot user. Now I feel gross, like, don't get any on me. Ick.
Re: (Score:2)
Because you're going to take the time to pore over millions of lines of source code in one of these things...
What you seem to have forgotten, and I'm flattered really, is that there are other people on the planet besides me. I really am honored to have you so obsessed with me and what I'm saying that you forgot about all of them, though.
Not obsessed. Just directing my attention to your comments proclaiming your POV; since I was, er, responding to your edict, albeit admittedly not specifically addressed to me.
But, silly me; I thought that's how these "Discussion Forums" worked. . .
Ya know, your previous post had a lot of good application ideas for AR
Yes, I do know. I've thought about this a lot.
I could tell. That was a sincere compliment!
but the ridiculous FOSS screed in the Parent just completely devalued your opinion
Your claim that insistence on freedom devalues my opinion completely devalues your obsession with me as a slashdot user. Now I feel gross, like, don't get any on me. Ick.
LOL!
Not saying it devalues your opinion. You have the right to insist on anything you wish; but you shouldn't use that as a dodge against actually addressing my point (that, in a purely practical sense, insisting on FOSS
Re: (Score:2)
insisting on FOSS in this particular instance Is a complete non-starter
What makes you imagine that? If I really wanted, I could already go forth and build my own eyetap. It would look like shit though, compared to a finished device. If AR really takes off then I guarantee you'll see AR projects on Kickstarter et al.
Re: (Score:2)
insisting on FOSS in this particular instance Is a complete non-starter
What makes you imagine that? If I really wanted, I could already go forth and build my own eyetap. It would look like shit though, compared to a finished device. If AR really takes off then I guarantee you'll see AR projects on Kickstarter et al.
With all undue respect, as an embedded systems designer myself, I highly doubt that you could create a viable wearable AR/VR "goggle" device, "looking like shit" or not.
A good bit of this class of device is simply not off-the-shelf components. Unfortunately, us mortal engineers (without an Apple-sized R&D budget and team of potentially dozens of embedded developers) are pretty-much constrained to use stuff you can order through normal distribution channels, and no Kickstarter campaign is going to raise
Re: (Score:2)
Well I'm fairly sure their killer app isn't going to be that, because if you look at how half baked Siri is you quickly realize that they can't reliably identify stuff or provide accurate information to the user. Google is way ahead but even so regularly makes mistakes with image recognition.
Re: (Score:2)
Well I'm fairly sure their killer app isn't going to be that, because if you look at how half baked Siri is you quickly realize that they can't reliably identify stuff or provide accurate information to the user. Google is way ahead but even so regularly makes mistakes with image recognition.
So, you automatically know that Apple's AR image-recognition will be based on the same code that drives a voice-based digital assistant?
Idiot.
Re: (Score:2)
It doesn't have to be perfect as long as the display code is intelligent. Don't display any of the tags until the image is recognized with high confidence. Also let me dial the threshold for that up and down temporarily on a case by case basis. Then I can decide what it is if my brain is better at it than they are, which I imagine it will be sometimes.
Re:Mac Level Power? (Score:4, Interesting)
Re: (Score:2)
What's going to hold this back is that the main thing people are generally interested in is other people, and virtually hanging private notes on people is creepy.
Sure, the whole thing is going to seem creepy until people get used to it, and that's going to take a long time. I have a hard time remembering stuff about people no matter how hard I try, make notes whatever. If I make notes that come up when I look at 'em then I'll be back in the game with everyone else who can just do that, a social creature again. Maybe other people want it so they can abuse it, I just want it so I can function in society at the same level as others.
Re: (Score:1)
Re: (Score:2)
They could probably put the high-power chips near the front panel, in front of the wearer's eyes, and make the skin out of a material that will help spread and dissipate the heat. (People won't often grab the headset there, so getting toasty is less of a concern.) Head motion should help the airflow. Figure that this will use TSMC's N3 process, which advertises a significant power reduction at the same processing capacity, and the claims are not too far-fetched.
Re: (Score:2)
They could probably put the high-power chips near the front panel, in front of the wearer's eyes, and make the skin out of a material that will help spread and dissipate the heat. (People won't often grab the headset there, so getting toasty is less of a concern.) Head motion should help the airflow. Figure that this will use TSMC's N3 process, which advertises a significant power reduction at the same processing capacity, and the claims are not too far-fetched.
So, no built-in OptiGrab?!?
https://www.google.com/imgres?... [google.com]
Re: (Score:3)
Still waiting to see what the killer app for this is.
I think the killer 'segment' for AR goggles could be in repair / manufacturing / medical. Imagine watch repair where each part can be defined and labeled while you're taking the movement apart, or the AR overlay can show you exactly how to orient a specific component to make sure that you're installing it correctly.
Re: Mac Level Power? (Score:1)
Re: Mac Level Power? (Score:3)
Re: (Score:1)
Re: (Score:1)
You don't need realtime video game pixel shading, high res textures, and all that post processing stuff to make AR useful.
Geometry and collision detection is almost exclusively done on CPUs anyway.
Re: (Score:1)
No it doesn't. That's just flat out marketing propaganda.
The M1 isn't magic, it doesn't defy physics. Like any other processor on the market it has to trade off performance against power, and is limited by the efficiency curve of available silicon. Macs were previously overheating, and had abysmal battery life if you actually did anything with them due to them putting design over practicality. As such they decided to switch to the M1, because they chose to sacrifice performance for better battery life. They
Re: Mac Level Power? (Score:1)
Apparently you have not seen any benchmarks.
Re: (Score:1)
Your comment could just as well have been written:
"Apparently you haven't read the Apple marketing materials that reinforce my pro-Apple bias"
The only benchmarks where Apple outperforms are consistently comparing Apple's top of the range M1 against Intel's mid-range processors, and even then the M1 never wins in all categories. The fact the M1 can't out-perform a mid-range Intel cheap in every category is laughable.
Apple's best M1 can't even come close to beating Intel or AMD's best processors and has to be
Re: (Score:3)
Re: (Score:1)
Macs were previously overheating, and had abysmal battery life
Considering that the rest of your post is completely wrong, this nails it: rofl.
an Intel or AMD processor wipes the floor with Apple Silicon any day
You are extremely misinformed.
Re: (Score:2)
No it doesn't. That's just flat out marketing propaganda.
The M1 isn't magic, it doesn't defy physics. Like any other processor on the market it has to trade off performance against power, and is limited by the efficiency curve of available silicon. Macs were previously overheating, and had abysmal battery life if you actually did anything with them due to them putting design over practicality. As such they decided to switch to the M1, because they chose to sacrifice performance for better battery life. They effectively decided to pursue the same path as mobile processors; much longer battery life, much less wasted heat, but achieved by reduced performance.
You have to literally be full reality distortion field retard to believe M1s are magic. They're not - an Intel or AMD processor wipes the floor with Apple Silicon any day, they just also push the limits of current silicon more and generate more heat and use more power as a result.
You do realize, of course, that Apple's mobile SoCs are at least 2 generations ahead of everyone else, both in power efficiency and performance, don't you?
Of course you don't.
Re: (Score:1)
Re: (Score:1)
So less powerful than a PC equivalent then?
What, we were supposed to believe "Mac Level Power" is a good thing, rather than a bad thing? Er, no. Mac's have always been under-specced for their price. This is a dumb advertising pitch.
You haven't been paying attention to the Apple Silicon Macs.
Re: Mac Level Power? (Score:1)
The M1 processor beat out the i9 when it was released, the new M1 Max beats out many desktop machines and even many Xeonâ(TM)s. Most VR/AR headsets right now are powered by the equivalent of a Raspberry Pi 3.
Predictions! Predictions! Free predictions! (Score:2)
> Kuo predicted back in March that the headset would be released sometime next year ...and that the latest Apple Watch would have glucose monitoring. And lots of other fabulously wrong predictions. Why does anyone post this guy anymore?
Gross (Score:2)
But sadly Apple is probably the only company that can pull this off without repeating the whole glasshole stigma.
Re: (Score:2)
Gross.
But sadly Apple is probably the only company that can pull this off without repeating the whole glasshole stigma.
But luckily, Apple is probably the only company that can pull this off without repeating the whole glasshole stigma.
FTFY.
Make it crazy Apple (Score:2)
What I think Apple should do here, and probably has to do anyway due to hardware constraints, is to avoid trying to look like regular sunglasses.
It's not going to look regular, so Apple may as well go a bit bold with the design, to make it very clear these are not normal glasses.
What I'm really curious about is if the whole device will sit on your head, or if some part of it will somehow be carried elsewhere, maybe even over a shoulder or two.
I think AR has a lot of existing potential so I look forward to s
Re: (Score:2)
What I think Apple should do here, and probably has to do anyway due to hardware constraints, is to avoid trying to look like regular sunglasses.
It's not going to look regular, so Apple may as well go a bit bold with the design, to make it very clear these are not normal glasses.
What I'm really curious about is if the whole device will sit on your head, or if some part of it will somehow be carried elsewhere, maybe even over a shoulder or two.
I think AR has a lot of existing potential so I look forward to seeing what they are cooking up! I normally go to WWDC anyway (when they have it) but this being being announced might be a reason to go and check it out in person.
Since Apple is in control of the entire stack, both hardware and software, I would suspect this will be an Application-Specific SoC, not just a repurposed Ax or Mx-series chip.
And it will rock.
Replacement parts. (Score:2)
I wonder if you have to pay Apple more than the price of the whole thing to replace the iBalls.
Only sort-of AR (Score:2)
If this is correct, it won't really by an AR headset along the lines of HoloLens: a transparent display that lets you directly see the world, but projects things on top of it. Instead it will be a standard VR headset that completely blocks your view. It will rely on cameras to feed images of the outside world to the displays. You can already do that with lots of VR headsets. It probably won't provide as satisfying an AR experience, but it's also probably an easier and cheaper technology to implement.
If
Re: (Score:2)
It probably won't provide as satisfying an AR experience
I agree with what you are saying about what they are probably doing, but why do you think the AR experience would not be "as satisfying"?
Since Apple has a ton of camera expertise at this point, if it's totally came based I'm thinking the quality of the external view would be really good, given the Oculus Quest 2 is already OK
Also way more advanced cameras would probably mean the real world tracking, modeling, and presentation would be better than most
Re: (Score:2)
4K per eye is pretty good resolution as VR headsets go, but it's less than the resolution of the human eye. Your view of the real world will be less sharp. I also worry about the effect of the cameras not being at exactly the same location as your eyes. You'll see the world as if your eyes were in slightly different places, which could impact the experience. It will also change the experience of interacting with other people. HoloLens looks like a fancy set of sunglasses. You might get surprised looks
4K per eye (Score:2)
Sorry, 4K per eye isn't revolutionary enough to be game changing -- even if they greatly reduce the gap between pixels and use a diffusing sheet. OLED is great though (if it isn't the cheap kind that gets burn in.) I thought they were going to do 8K per eye using microLED technology and foveated rendering so the GPU could handle it .. THAT would have been a game changing paradigm shift.
Re: (Score:2)
So macbook thermals in an iPhone form factor? (Score:2)
Also, wouldn't the batteries get pretty heavy?
I guess sooner or later, we will find out, but now I a
Re: (Score:2)
OK, so you have a headset with an M1 in it, presumably somewhere on your face. I'd imagine that gets uncomfortable quickly. Even an entry level iPad gets pretty uncomfortable when playing an intensive game and it much larger and much lower powered. Assuming a design like an occulus, dimensions-wise, wouldn't that be like putting MacBook, which requires active cooling fans, into an iPhone form factor.
Also, wouldn't the batteries get pretty heavy?
I guess sooner or later, we will find out, but now I am more nervous than excited.
I think an AR processing load will be significantly-lower than a VR processing load. Plus, Apple will sure to be including one or more task-specific subsystems to take much of the load off of the CPU.
You do realize, for example, that an M1 Max MBP outperforms a Xeon Based 2019 Mac Pro with an Afterburner Card at encoding and decoding several video formats. It also outperforms that same Xeon at single-core compute-tasks.
So yeah, I think Apple can pull this off, if anyone can.
Not questioning the muscle, but he weight & he (Score:2)
You do realize, for example, that an M1 Max MBP outperforms a Xeon Based 2019 Mac Pro with an Afterburner Card at encoding and decoding several video formats. It also outperforms that same Xeon at single-core compute-tasks.
So yeah, I think Apple can pull this off, if anyone can.
I'm not questioning the performance, I am questioning the physics, specifically the weight of the battery and the amount of heat an M1 would produce. I am imagining an M1...say an iPad pro, but shrunk down to an iPhone shape...all that heat, only at less than 1/6 of the surface area.
I am also imagining a laptop or iPad-grade battery. I am now imagining wearing those.
If the rumor is true and this thing needs a muscular M1 processor, it would be reasonable to assume it needs a decent-sized battery an
Re: (Score:2)
You do realize, for example, that an M1 Max MBP outperforms a Xeon Based 2019 Mac Pro with an Afterburner Card at encoding and decoding several video formats. It also outperforms that same Xeon at single-core compute-tasks.
So yeah, I think Apple can pull this off, if anyone can.
I'm not questioning the performance, I am questioning the physics, specifically the weight of the battery and the amount of heat an M1 would produce. I am imagining an M1...say an iPad pro, but shrunk down to an iPhone shape...all that heat, only at less than 1/6 of the surface area.
I am also imagining a laptop or iPad-grade battery. I am now imagining wearing those.
If the rumor is true and this thing needs a muscular M1 processor, it would be reasonable to assume it needs a decent-sized battery and will produce a lot of heat. Of course, time will tell.
You keep making this a CPU-intensive task. I submit Apple will make their AR goggles work smarter, not harder.
Therefore, we have no idea what size battery will be needed.
Steve Jobs passed away 10 years ago (Score:2)
Will be crippled so that you still need an iPhone (Score:1)