New report on Apple’s VR headset: 8K in each eye, potential $3k price tag

I'll add when the iPhone came out it was way way more than it's competition. It was effectively the first $100/month w/ 2 year contract phone bill as compared to other devices where $35/month was common.

I don’t know in what world would that be affordable.

Maybe if companies buy it for their employees which makes this an enterprise product like the Mac Pro or software like Logic Pro.

It would be insane of Apple to invest billions in development to end up with a product that is not targeted for regular consumers.

They had a partnership with Valve back in 2017 for Vive support on MacOS... Just three years later it was canceled. (There were actually more VR games running on Linux at that point than on MacOS, thanks to Proton.) I'm not expecting Apple to try to enter the VR gaming space again anytime soon.

It's this very weekend: The Super Bowl.

Well, not the Super Bowl precisely, but in experiences like sports or other time/space limited events. Sitting ring-side for the next big match. Or at the 50-yard line at the Super Bowl. Or at the finish line of the 100-m dash. All for, I dunno, $15? And the cost for the company is just rigging a google-maps-like camera ball and streaming the data out. Especially as the quarantine grinds onwards, a better form of escapism that AR/VR provides may have a big market for time/space limited events.

The early iPad days replaced by the Pro suffix. Or even worse. AirPods Pro Max.

Apple’s product line those days has a lot of fluctuations in their pricing. (On purpose) and they seem to have consumers for both.

This right here is Appleism.

And why wouldn't Occulus, HTC, Magic leap or big boy Microsoft have a hit?

I'm pretty sure Apple's gaslighting field has yet not found a way to bend laws of physics.

Everybody wants a sleek, everyday wearable device, without usability corner cases, but such are physically impossible to make.

Portability, visibility, usability. Choose one.

First, power requirements demand either external power, or extreme power austerity cutting into display, and graphics.

Recon Instruments had first really practical battery powered HUD goggles, and they barely lasted more than an hour in real life use.

The lowest possible power at which you can provide just any much rich graphics is 250mw — the lowest end of most energy efficient SoCs. Possibly, Apple can bruteforce it with 5nm custom ASICs, but not by much, 100mw at max possible with CMOS cells.

Even if you have a magic chip with hypothetical 0 Watt power usage, you will not improve the situation by much as your display will still eat at least 1W, or usually up to 2.5-4W if you use any waveguides.

Daylight visibility is essentially impossible without displays outputting at least 100000+ nits. There are no tricks around that, that's just physics.

There are very few technologies on the horizon which are physically capable of achieving anything better.

Monolithic devices are one, and only ones which can offer sub 1W power use at any much good visual quality.

But, they are very expensive to make. All existing makers are manufacturing on lab scales only. Getting manufacturing out of the lab, and into fabs is impossible without the process technology, getting out of the lab too. And this shortens the list of credible contenders to just 1 company in the world, and guess, it's nor Apple.

Third, even if you agree to a wired up design, where do you get 16k video from? Have you ever seen how thick DP 2.0 cables are?

How do you get latency down?

How do you make even simplest AR interactions not require the wearer to also wear a Quad SLI videogaming PC?

How do you transport 40 gigabits per second of video without having the I/O PHY eat more than the system itself? Only optics can do it under 1W.

Apple can surely put its silicon to good use here, but even a purpose made ASIC video system will be on the edge. It will limit them to the most basic graphics, and pretty much hardcoded, and handcoded use cases to extract maximum power efficiency, and workaround hardware limitations.

It will be a basic HUD, maybe with some GFX, and good video playback options. Essentially an Apple Watch, except you wear it on the head.

I bet they will intentionally limit its functionality to not to let users see performance limitations, and corner cases too often.

i would love forcing or otherwise pressuring my employees to have this strapped on their face for work 24/7.

What exactly do you mean by this? Are you saying that right now, the most popular use of VR headsets is for pornography? Or are you speculating on how VR headsets will be used in the future?

Speaking of, I'm going back to beat saber. You might have been misled by its name if you think that's a porn app. :) Email's in my profile if anyone is up for some multi.

--

PS. It's pretty obvious to see the people in the comments here who haven't used VR headsets. Eyestrain and headaches are mostly a non issue with recent models, but what they still are is clammy. They get warm. Super uncomfortable for work purposes. Apple's design could solve that I suppose.

As a non-apple-user, I'm excited to see if they can produce something as lightweight as the Quest 2 that is much higher end. I would buy that in a heartbeat, even at a 3k price tag. (Yes, mostly just for beat saber...)

Also, it sounds like this would be more of an all-in-one device, where it’d handle the rendering instead of connecting to a separate computer, so the rendering performance is more likely to be the limitation than any kind of transmission limitations anyways.

This seems to be describing foveated rendering, which is reducing the image quality in your peripheral vision, because you are less likely to notice it there. It requires tracking where the eye to so you know what part of the screen the eye is looking at.

The RTX 3090 is likely rendering the whole 8k screen at a consistent quality level, whereas foveated rendering would mean that only the part of the display that the eye is actually focused on would be rendered at full quality. If Apple could pull off the tracking well enough (accurately, with low latency), they could probably save a lot of GPU power by lowering render quality outside of what you're looking at.

https://en.wikipedia.org/wiki/Foveated_rendering

It'd also explain the price tag, if you have to buy both a high quality display and a speedy GPU that's been glued together.

b) VR uses variable rate shading because we have a lower visual acuity in our peripheral vision.

c) VR rendering typically "shares" a significant amount of the work between the two viewports. E.g.: one set of "commands" are rendered simultaneously into two buffers with different view transforms. Textures and meshes are cached once and rendered twice, so the bandwidth requirements aren't actually doubled.

d) Display stream compression (DSC) and similar technologies would work well for VR because the viewport is always in motion with a high refresh rate. One could even imagine sending a H.265 compressed stream wirelessly at a mere gigabit, which is fantastically high bitrate video but well within current WiFi capabilities.

e) There will be future developments as well, we're not stuck with current technology. Keep in mind that current era flagship GPUs are manufactured on silicon processes that are about 3 generations old! By the time this VR kit hits the mainstream market, GPUs could be manufactured on a 3 nm TSMC process and easily put out 90fps in 8K resolution.

If the product is legitimately better than everything else on the market, it doesn't matter how expensive it is on initial launch.

That detail seems a little suspect to me in this whole story

Back in 2015 Michael Abrash said:

In order for it to have retinal resolution across a field of view of 180 degrees it would have to have something on the order of 16K by 16K resolution [per eye]. [1]

--

[1] https://youtu.be/UDu-cnXI8E8?t=4232

180 * 135 * 60 * 60 * 100 * 48 = 419904000000,

420 billion bits per second, without compression.

Example headset that used a small display with this pixel density.

https://arstechnica.com/gadgets/2019/02/retina-resolution-he...

Because 2k per eye seems enough until you realize that it gives you less than 20 ppd (pixels per degree), while 2560x1440 27'' monitor at 50cm gives you 40 ppd

To get same ppd on 27'' monitor at 50cm it would be 1111x625, for 24'' at 75cm its 700x400!!

My point is that while current ppd might be fine for some, its very low for anything that requires detail. Any small text or gauges or similar is simply not readable in VR while being crystal clear on FHD monitor, just because the field of view is stretched so wide in VR headsets.

You can say that small text isnt important in VR, but that would be dismissing the fact that small text and higher feature resolution can increase information density dramatically, which is always desired. After all, you aren't browsing this site on a screen with 640x480 resolution, are you?

By the end of the first day I had my tape measure out moving my couch, coffee table, and clearing out the living room. It's the VR space now.

I would be shocked if this is anything but a gimmick, with all sorts of upscaling and rendering tricks.

I very highly doubt apple is all of a sudden the producer of the best gpu in the world, with a wide margin. Nvidia claims their 3090 ( a $1500 card ) is capable of 8k 60fps. In benchmarks, this is only possible in certain easy to render games.

The idea that apple can double that, and push a higher refresh rate for smooth vr (usually 90fps minimum) is wild.

The question of the effect on vision of shining screens into your eyes for hours at a time still stands.

Edit: whoops, guess I was wrong about that!

It's not entirely clear what problem VR is trying to solve in a lot of situations where's it's claimed to apply (like office use). There's obviously some places where it could be quite useful, but light-field fixed screens might offer a lot of similar benefits for a lot of these applications and at this point at least, that tech is a bit further along (and might be more comfortable for users).

There are many circumstances where AR is far better than VR, even with perfect VR

To get this you need outward facing cameras, which people will balk at for privacy.

Personally, I think that VR will win out, even for AR applications (or maybe my semantics are incorrect). I can imagine a VR system that uses external cameras to mimic a person's natural field of view and overlays additional information in the process of displaying it. It seems like a more available route to a fully immersive "augmented reality" than the current glasses based prototypes.

They can make an amazing VR headset right now, a comparable AR headset is many many years if not decades away.

I suppose it's the issue if your expectation of VR is resolution-sensitive. But good VR apps will just treat that as a limitation and factor it in. GBA apps of the time were pixel-art, and resolution was never its "main problem" (batteries were), because you weren't using it to watch movies or something.

IMO a lot of the true VR issues have been fixed by now. Pass-through, standalone wireless, baseless hand tracking, no more headaches, etc… A lot of the work that remains is incremental improvements (such as resolution, battery life, FOV, weight, tracking accuracy). I would personally like to see some very-low-latency wireless earbud support, too. Once again the Quest 2's over-the-ear system is great.

The most severe issues I have with VR right now are very much not software/hardware related, but rather have to do with sweating: Band getting dirty, lenses fogging up. The fact a lot of VR apps are fitness based makes this a serious problem, at least for me.

The real "true issue" of VR currently is probably the lack of apps. There's fairly few killer apps, especially on the much more restrictive Quest line.

But I agree with you there's a huge potential for "retina" VR.

I think the bigger challenge is not screen resolution but lenses. The G2 has quite a narrow sweetspot so although the centre of your vision is basically monitor quality and text looks perfect, the quality drops off towards the edges so it doesn't really work for reading where you want to scan left to right with eye movement. It does work well for simulations where moving your head is a more natural way to look around.

This is basically the plot of San Junipero but with the vast resources going into ML, neuroscience, and anti-aging, it's not hard to imagine this being a possibility within the next 100 years.

A surprising aspect of VR is that this is actually the complete opposite, just how much your real body actually matters. VR is exhausting. 10 minutes of Thrill of the Fight and I'm physically incapable of continuing. An hour of Rec Room or Echo VR and I have to stop from the sweat buildup. Even hanging out and just talking VR Chat is so more physical (standing, actively interacting with both arms, potentially full body tracking) than sitting at a desk during the day job, more comparable to something like being at a conference where at the the end of day you really feel it, all the standing around and slow walking, since in total it was more than an average day.

There's a full lecture on the subject here: https://www.youtube.com/watch?v=gpdX9jkhv2U

Jokes aside, I agree - I think they're taking their time with this and it might bring something interesting to the scene.

All the "serious" HMDs on the market today work just fine - Valve have put some serious engineering into their tracking technology, for example.

That said, this excites me - I wouldn’t be buying a gen 1 at 3k, but down the road I could see replacing a laptop or two with this kind of thing.

Apple products aren't inherently addictive. It's the applications that are.

Also they do provide Screen Time and other features to control your usage of their devices.

So how would a VR headset work with everything?

1) the iphone has the best facial tracking camera and software (better even than desktop and webcam and GPU)

2) like the Occulus, you should only need a phone to connect.

3) only 1 vr headset currently has an internal camera able to track faces and eyes. Facial expression, iris movement and gaze tracking is hugely missing currently, but trivial with the iphone.

I would have the iphone being the base station and do the full body tracking "just put it on your desk". I would also put their existing face tracking and eye tracking tech inside the vr headset. I'd make it so the headset only needs an iphone to connect. I'd make it cordless only.