Mixed Reality with Passthrough | Hacker News Reader

Mixed Reality with Passthrough

developer.oculus.com

147 pointsbemmu142 comments

Neat, but I still wouldn't touch oculus with a 10 foot pole especially when cameras in my home are invovled

Remember the wars between VHS and BetaMax, and VHS won because (supposedly) it allowed porn?

Well, now we could see the same thing happening. Oculus doesn't allow porn (because who in their right minds would watch it with their own personal Facebook ID tied to it, with cameras running and connected to the internet?). But a competitor that can dethrone Oculus isn't yet on the horizon. Where is our "VHS"?

This isn't a complaint. Just an observation :)

>>because who in their right minds would watch it with their own personal Facebook ID tied to it, with cameras running and connected to the internet?

I'm willing to bet that a sizable portion of the population would watch porn on this device even after you explain this to them.

I sizable portion already do

You must mean interactive porn on the official store? Because there are several players on the official store, plus the browser, that play porn fine.

I think you didn't read the entire comment.

I did. Still don’t know what you mean.

You can still do porn, just not in the store. You just need a website.

Or do you prefer your desktop porn to be via a .exe as well?

Ah, that comment again

I don't see how it's a weird opinion, I'm also speaking for myself others are free to enjoy the platform if they want I don't care

No your opinion isn't weird. I agree with you in fact. But we also dont need to share every opinion always (see: "Inside" by Bo Burnham).

Obviously some people like this technology and are happy to trade the choice of using their FB account, nor does your comment contribute any discussion to this cool feature or tie it back to the privacy discussion in any meaningful way. It might be worth letting people just enjoy that news.

I think you're right, looking back on it my comment was kind of pointless

I am so glad I invested in an Oculus Quest 2 recently. So far, I've been having so much fun just exploring the various things to do in VR.

I initially bought this to explore UI/UX for VR, but with AR, it opens up the technology and my personal research to a whole slew of challenges.

AR has been in the consumer space a lot longer than VR. In fact I’m more surprised that AR had never taken off in a big way than I was about VR. Sure we’ve seen some games like Pokemon Go take advantage of it. But AR should have been a game changer.

AR will become a game changer once the hardware is there, i.e. glasses than can be worn comfortably and anywhere. Everything before that is just practice for the software and UI.

The hardware has been there for a decade already. You might have AR tools already in your pocket (eg Google Translate)

I agree with the idea but not that we need comfortable glasses that we can wear anywhere. AR will be a game changer on my bicycle if my helmet can integrate it. AR will be a game changer when I'm bombing down a mountain on my snowboard, if my helmet can accommodate it. My belief is that we will get deep market adoption in these verticals before any practical general purpose AR is possible

> But AR should have been a game changer.

VR turned out to be the easier of the two to turn into a viable consumer product - mainly because you can get away with a big bulky headset for gaming around the house, but you can't wear it outside for a walk.

The hardware that you could wear outside for a walk (e.g. google glass) hasn't actually been AR, and has just been like a tiny heads-up display in the corner of your eye.

I disagree. AR mobile games have been around for a few years and VR first hit the market 2 decades before AR. If anything, VR has been a harder technology to break through. Which, in my personal opinion, is because it’s more likely to make people feel sick plus is less of a social tech.

It depends what you mean by AR - As this term is used to describe showing an overlay on video on a phone, a heads up display in the corner of your eye in a pair of glasses, or technology like hololens where you can make objects appear to float or overlay the real world.

These things have all been referred to as AR, and are all very different, but I’m specifically talking about the latter personally (which is the one the article is talking about).

It’s kind of the same with VR by the way - you used to be able to watch 360 videos on your phone and the viewport would change as you moved it, but it’s obviously very primitive compared to something like the VR oculus offers. These two things obviously are very different technologies, which is the same as in-phone AR and the AR described in the article.

All of the AR examples are legitimately AR. It’s a massive field.

VR, on the other hand, isn’t such a broad field. Take your 360 videos for example, there’s no interaction with the content.

Even 90s era VR was very specifically referring to interactive worlds. Whereas Augmented Reality has always just meant having our real world senses enhanced with digital technology. That means phone apps are legitimate examples. Google Glass is a legitimate example. The barrier for entry is much lower yet the possible utilities for AR are much higher than with VR. Which is why I’m surprised it hasn’t taken off in a much bigger way with all the hype that VR has.

The technology stack to support Google Glass is almost entirely distinct from the technology stack which supports in-phone AR. Maybe both are AR, but then all that means is it's not a particularly useful term, because it can mean things that are entirely different from both a technology and user perspective.

So I think we need to define terms - Otherwise if we are comparing overlaying a tape-measure on the camera on your mobile phone to being able to plug on a headset and playing beat-saber, it's not a particularly useful comparison and I'm not surprised that VR has bigger hype!

(But moving about a bit of furniture through the camera view on your mobile phone is very different to something like magic leap, or oculus passthrough).

> The technology stack to support Google Glass is almost entirely distinct from the technology stack which supports in-phone AR. Maybe both are AR, but then all that means is it's not a particularly useful term, because it can mean things that are entirely different from both a technology and user perspective.

Bullshit. The technology cinemas use for their stereoscopic 3D is totally different to the technology used for Samsung’s 3D TVs, and different again from the Nintendo VirtualBoy which is also different from Google Cardboard. Yet all of the aforementioned are stereoscopic 3D.

Not every technical term depends on a technical implementation.

> So I think we need to define terms

It is a defined term. You just seem intent on moving the goal posts for some reason.

> (But moving about a bit of furniture through the camera view on your mobile phone is very different to something like magic leap, or oculus passthrough).

I agree. But 30 years after the first time I tried VR and the whole industry is still basically a novelty. Whereas AR is already providing practical value to millions (eg Google Translate) simply because it is more accessible and has more scope for useful utilities.

Which brings us full back circle to the very point I opened with. ;)

AR has been a game changer, via face filters. More stuff is possible with it everyday, but it’s already getting a lot of usage

Well, it doesn't help that nobody actively releasing apps seems to understand that AR is a lot more than just drawing things on top of a camera feed.

Almost every social video platform has some firm of AR filters. Snap probably has the most pervasive use of AR, with tools available to develop new filters (Lens Studio).

As of right now, these are the only examples of apps that do something with the tech that can't be done without the tech, and probably done better.

You can see why in the games market and Pokémon Go is a good example of why. The AR camera feature distracts, not adds, to the gameplay. Everyone I knew playing the game turned the camera feature off because it drained battery fast and limited play time.

Now, I would still call Pokémon Go an AR game, even without the camera feed, because I count the map overlay tracking your real work location a form of AR. It is a feature that augments your reality. And that feature adds to the gameplay. People enjoy traveling to play the game.

I think AR games will never be a big thing because AR design is fundamentally at odds with game design. With AR, you have to design an app the works in a context that the user brings in. In contrast, with VR, you're designing an app that provides a context to the user. Game design--especially as done by most game studios--is usually focused on providing a wholly contained experience for the user.

And that AR design challenge is just fundamentally more challenging. First of all, we can't get a lot of the interesting information about a user's personal context. The most we can get right now is geographic location, rough estimations of surfaces in their area, and maybe some very rough object classification. We don't know things like whether or not the user has a TV, or where that TV is located. Even if we did, we'll probably never be able to know what brand of TV they have, so now you have a problem of none of the stakeholders caring to ever fund a "TV classifying" project, because they'd never be able to sell the branding tie-in.

And I think that's the real problem with both the AR and VR industries. There's little incentive to create a product that users will care about, outside of games, which work better in VR. Any funding your going to find for any use case outside of games will want it 100% married to their brand, but brands have all gotten so indistinguishable from one another that there's nothing to really do.

Same. Best purchase in a long time. Can’t believe I was wondering if it would be worth it.

The VR UI paradigms are really nice.

I haven't been following Oculus since I bought the first one on Kickstarter (which I had a lot of fun with); the Quest 2 has cameras to support AR or is that an add on? It looks like it on pictures, but there is little mention of it on most pages about it (at least at a quick look).

The first oculus headsets used camera base stations to track the headset and controllers, all of the recent ones have used cameras on the headsets itself to track everything, these can be used to see the real world, however since they are infrared cameras meant for tracking there is no colour and the resolution isn't as high as it should be for AR to be useful.

I had the first oculus, and it didn't have any base stations. It didn't track anything at all, in fact... it only detected rotational motion of your head, it didn't detect any other motion at all. There also were not any controllers.

Yes but those were developer kits not retail devices.

Even that I forgot: mine was the dev kit indeed.

Sure, but they were still called an Oculus Rift.

It sounds like you’re talking about Oculus Go, which isn’t the first oculus.

The original DK1 was 3DoF, and the CV1 launched with a single sensor and an Xbox controller. The CV1 didn't get proper 6DoF until after launch.

The first oculus is the rift though

They released a dev kit that was called the oculus rift that came out before the one you are thinking of.

Dev kits don’t count though. Nobody mentions them when we talk about the nintendo X or the playstation Y. Why would it be different for the oculus?

The "dev kits" were freely sold to everyone and also bought by consumers, not just developers. Very different to sell-your-soul-NDA'd console devkits given out to select studios. "early adopter edition" would have been just as fitting.

Because even though they called it a dev kit, it was still a commercial product in many ways... there were a ton of games released for it, and a whole community of users who were not oculus developers. I got one, and I never wrote any software for it nor had any real plans to. I got it purely as a consumer.

DK2 had a separate camera for outside-in tracking. I have one in a box in my closet.

No, I had the Dev Kit 1. It was called Oculus Rift.

The Oculus Rift definitely had more than what you’re talking about, and it was the first oculus

I have one sitting next to me. It doesn’t have positional tracking… you can read about it here: https://en.wikipedia.org/wiki/Oculus_Rift#Development_Kit_1

I believe he is referring to the first commercially available headset, which is indeed, the CV1 "Oculus Rift"

The dev kits were commercially available… anyone could just order them. That is what I did.

Once upon a time the original Rift only had 3DOF tracking. This was back in the "Palmer Luckey just met John Carmack" days.

Yes, the Quest 1/2 and Rift S do "inside out" tracking. They have 4 cameras (5 on the Rift S) on each corner of the front face plate that tracks the headset and controller positions in 3D space (in addition to internal accelerometers and gyroscopes).

This is, for the most part, what enables the Quest to be fully wireless. Because there is no base station or fixed position camera, you're not limited to where you can use it, nor how big your play area can be. If you're in a new space, you just draw a guardian border around your play area and you're good to go.

The "passthru" feature mentioned above lets you see the camera feeds composited together in the headset display. It's very helpful for things like moving a chair out of the way or finding your controllers if you put on the headset without grabbing them. This new API allows developers access to this camera feed and to interact with its display for AR applications.

How long can you use it before feeling dizzy?

I don't really get dizzy or disoriented unless it's a game where you move with the joystick control or something. Games where you move around a table or teleport from place to place haven't bothered me at all.

I’ve only used 2 headsets. The first PSVR made me queasy quickly. I’ve used the Quest 2 a ton without feeling dizzy or sick for even one moment.

I can use it for 2 hours or more. The only games where I get dizzy within 15 minutes are the roller coaster and driving games that spin you upside down where it feels like you have no control over movement.

I get motion sick easily. So I was worried about that but most popular games are designed in a way that there is no movement that would cause me motion sickness. Also they list games’ comfort levels and I avoid anything that is intense.

The biggest issue for me is videos where there is no rating and one slow camera pan or zoom will make me almost throw up.

This changes everything!

I would find it interesting to see how a layer on top of reality would allow non-technical people to perform technical tasks.

For example I don't know a lot about car engines. I can imagine being guided through minor repairs by utilizing these types of overlays.

I can imagine this kind of overlay system also to provide a big boost in diagnosis for problems and systems such as car engines or even medical diagnosis.

I find it hard to imagine to use the current bw only and grainy cameras of e.g. the quest 2 for such tasks, it really doesnt feel natural. But might be easy to improve this in future generation hardware

Feels like a test. I think it’s a solid bet Quest 3 will have better cameras!

I'm sure they chose the ones used for better CV tracking & properties, especially in low light etc, but could be easy to add a pair of true-color ones, especially for passthrough. Just checked again and the main issue IMO currently is the low framerate / blur when using it. The movement of the hands e.g. is quite uncomfortable this way.

It works if the "AR" is just using the real world as a kind of backdrop and not the main visual elememt.

If you are using it the other way around as a kind of real world view with only a few AR elements on top it gets old pretty quicklY

It’s hard to see why VR headsets with pass-through would win out over AR headsets (like the Hololens) in the long term for such uses. But I suppose if you need something right now, VR would be more available and cheaper.

AFAIK, AR headsets are still additive displays, ie. no way of displaying black. That’s the only advantage I see in using VR headsets for AR

Also waveguides have other limitations, like image fidelity, FoV, needing darker environments and significant display artifacts.

One reason is that with hololens / magic leap you can’t create black. The display is like a ‘screen’ layer in photoshop and can only add additional light, so you can’t get great contrast and anything you add to the scene has a ghost-like quality.

All those demos you’ve seen of AR glasses making screens on a wall are faked. Magic leap paid a special effects house to make a lot of their ‘real’ demos.

Also - field of view. Magic leap has an AWFUL field of view.

I think it's interesting to think about adding elements to what we experience in reality via AR.

Versus importing elements from reality into a VR experience.

Based on your comments I would say one of the big advantages of importing from reality into VR would be to be able to move around without colliding into physical objects in the real world.

That would be really cool. Like a game that had virtual objects but in the same places as the physical world, but hey don’t need to be the same thing.

Long term I think the idea of a computer almost permanently between us and our number one most trusted sensory organ is a fast recipe for Utopia / Dystopia.

You basically described the Hololens concept from years ago. What changed? This feels like old territory and the excitement doesn’t feel justified.

Edited.

It's a HN guideline not to accuse others of shilling without evidence. (It's such a common pattern that it makes for boring reading, among other reasons.)

It a practice platform for when the tech catches up with the dream. Anyone who isn’t excited by augmented reality tech is just.... I dunno... bored of life?

Its also a great chance, to virtualize large swathes of the "non-touch" economy - and thus be another step stone to a carbon reduced world. No need for a plastic gnome on the front lawn, just place it. No need for birthday deco, just place it in virt, no need for other deco, just place it in virt. No need for a fancy car display, just use virt-hud, with a basic-backup.

> Anyone who isn’t excited by augmented reality tech is just.... I dunno... bored of life?

I want invisible technology, the one that does everything and I don’t have to interact with it.

I am waiting for AR lenses.

>Anyone who isn’t excited by augmented reality tech is just.... I dunno... bored of life?

lol isn't this just the most ironic claim? i'm not excited by augmented reality exactly because i'm not bored with life (i.e. unaugmented reality).

I’m not excited for AR at all. Conversely, I’m extremely excited for VR and have only become more so after getting my first headset (Quest 2 primarily used with SteamVR).

Possibly this is not a great analogy but if there was just one car company in the world for example Ford, I don't think cars would be as versatile or as commonly adopted.

Therefore I'm seeing any progress on the part of multiple companies signaling that this is more than just a one-off.

Another example would be the very early handheld iPad predecessors (Palm Pilot). Look like nothing was going to happen there. But after several failures it started to take hold.

> What changed?

Hololens price is $3,500 where as Quest 2 price is 300$. Quest 2 has the potential to become a mainstream device with that price.

Okay that's a perspective I can understand a little better. Making AR more accessible by using cheaper VR technology to implement it seems like a good reason to be interested.

I'm still not seeing the excitement as a consumer because the apps aren't there yet. From a development perspective it also just feels… iterative.

>I'm still not seeing the excitement as a consumer because the apps aren't there yet.

Why would the apps ever be there if the hardware isn't? Wouldn't that statement be like complaining "the apps aren't there yet" for the iPhone 1 or Android launch? We know how that movie plays out.

>From a development perspective it also just feels… iterative.

But "iterative" is generally by far the most important part of technical success where mainstream adoption is required to get the most out of it. Nobody builds Rome in a day, it takes years reaching a critical mass of iterative improvements.

Incrementalism is the name of the game.

Cheaper, better devices... about an order of magnitude here. A larger user base, by more than that. A better understanding of UX paradigms. ETC.

The majority of successful products of the last generation were conceptualized and attempted during the dotcom, circa 95-00. Smartphones, social media, online dating, ecommerce, travel, cloud computing.... The user base, infrastructure, technology, culture and such just weren't there yet.

If you wrote off everything that was attempted but failed to catch fire in the 90s, you would have written off almost every tech product created since.

I’m a fan of the potential of these technologies, but I’m guessing that sadly FB’s subsidised/closed VR/AR systems will have a ‘hidden’ (& frankly creepy) cost that is unfortunately characteristic of their business model: https://itif.org/publications/2021/03/04/balancing-user-priv... https://carrcenter.hks.harvard.edu/publications/reimagining-... https://www.reedsmith.com/-/media/files/perspectives/2017/06... [pdf]

Fundamentally different devices.

The Hololens is a waveguide display with a very small FOV. It's also 10x the price.

Hololens (and most other AR headsets) is an additive display on a transparent waveguide. The Quest2 is passthrough video on a full face LCD.

The display technologies are fundamentally different. HL is additive so you can't show blacks. You have to make due with tricking the eye with grays. Its a lot harder to get realistic lighting on objects. The FOV on the Quest2 is much better.

The Quest2 is a couple SOC generations ahead of the HL2.

The Quest2 has a much bigger install base.

The Quest2 is 1/10th the price.

Do you still need a fb login to use oculus ?

Yes, you can make a fake account just for oculus. So?

Do you not risk being banned for being a fake account if the Facebook AI doesn’t think you’re social mediaing enough?

Yeah you do.

No you don't (officially), I also use a throwaway account and had no issues. There were a few unrelated isolated incidents in the beginning that were heavily publicized by the media but in the end you're at the mercy of the same AI ban bots just like with YouTube, Google or any other big platform so you have the same chance of finding yourself locked out if your behavior trips up the algorithm.

I recently created a throwaway FB account for "work stuff", and the amount of spam mail I'm getting now each day from Facebook to "engage" is shocking. Also: I didn't register the account with my real name and a different email address, yet Facebook magically featured out who I am and is sending out invites to the friends on my (long dead) real Facebook account. It's all creepy AF.

I've been using a "fake" account.

I don't do anything dodgy with it, and my name is mostly the same as the credit card.

They are much more reluctant to kill a dormant account if you attach it to hardware.

Oculus does sell a version without the login restriction for around $700

Iirc those still need a Facebook for work account.

Only to manage the fleet of devices. The users of the devices do not need a Facebook account. In fact, there isn't even anywhere to enter one.

those are locked out of any apps on the Occulus store though, right?

They are enterprise licensed headsets. You're not just going to walk up to Facebook and say "here's $800, give me a headset with no pre-installed apps".

You have to get approval to get access to Oculus for Business, then you purchase the headsets through a partner company (CDW, in my case).

But when you eventually get everyone to wake up long enough, concurrently, to do their jobs, yes, you essentially get A Quest 2 with no apps other than the browser installed.

Is pretty cool. The browser is built on Chromium, so it has the full WebXR API. I have a WebXR/WebRTC app that I've built for my company. We have a fleet of 15 headsets right now. We send them to our students taking our foreign language classes and they use them to practice with their instructor in 360 photo environments of popular tourist places in their target country.

>We send them to our students taking our foreign language classes and they use them to practice with their instructor in 360 photo environments of popular tourist places in their target country.

Man this is actually my number one idea for a VR app: VR language learning for more powerful immersion. Looks like I'm not the first to think of this I'd better hurry.

Imagine you're learning French and when you get to the word "voiture" a car materalizes in front of you, when you get to the word "cheval" you're suddenly riding a horse, etc. Now apply some SRS soup and you're learning vocab in an unforgettable way. I have no idea where to start but I want this.

There are apps on the market that kind of do that. They are in various stages of development, but most of them aren't really going anywhere. I believe the approach is fundamentally wrong, as it focuses on the low-hanging fruit of the problem of teaching people language.

We're the only company that makes the VR a part of an existing program of instruction, with instructors. All the other apps are self-driven, which doesn't really work for the vast majority of people. We are also the only app not leaning on speech-recognition, which also doesn't really work for language training (too many false-positives and false-negatives).

I'll second this. The vocab is probably the easiest and most straightforward bit to train right now. Flash cards have already all but solved that problem as they provide the simplicity, an easily optimizable metric, and a dopamine hit from numbers going up.

The part that's harder is getting people to fluent conversations, both on the comprehension and construction side.

Do sidequest and wireless virtual desktop function?

Any blog posts you recommend from the hacking/developer perspective?

I don't use Sidequest myself, but you can install any APK you acquire yourself, so it should work. All of the system features are the same, so hand tracking, Link, etc should all work.

No, I don't have any links. Everything that is out there is almost universally written for Unity and I don't do Unity anymore.

You could dip into GodotEngine they have pretty decent VR support. Our hand Tracking ng based fitness game VRWorkout is made with godot

I'm pretty happy with Three.js and my own code.

Since that's a violation of the TOS, you're taking a bit of a risk with that approach

Was about to buy one to play with, especially as they have new hardware (to me, I wasn't paying attention).

Then I noticed this on the terms of service when clicking 'buy':

Requires wireless internet access and a Facebook account.

And, uh, heh... no. Just no. I don't even need a Sony/PS account to use a playstation, and at least a PS account has to do with, you know, the playstation.

Meanwhile, Facebook has to do with a bazillion other things, has all this baggage, I've never had and ever ever want an account, and... what? I'm going to get an account for a barely usable piece of expensive hardware?

No.

Yup. Seems like they walked back the forced migration for existing users, but new Oculus users must sign up with FB it seems.

https://www.oculus.com/blog/facebook-accounts-on-oculus/?loc...

How do you lend the device to your friends? Do you log out and let your friends log in?

Can you sell the device without your personal ID on it?

I have an Oculus Quest 2 - it's really not intended to be multi-user (even within a single household), which seems like a big miss.

The answer to both of those questions is: "hit the factory reset button in the Settings menu" which blows away any and all state. That's realistic for selling the device but not really for lending it.

Facebook has allowed multiple account logins on a single headset for a few months now.

Oh, a tech from a decade ago adapted on the closed SDK of a closed device!

The fact that they have to write a paragraph promising that everything is local and that videos don't get sent online, curiously, makes me even more suspicious.

And if they didn't write that paragraph? It's Facebook, so you'd be suspicious anyways (and rightfully so).

What tech from a decade ago are you talking about? This is one of the first devices,at least the first in the consumer price point, to have this functionality at this high fidelity.

Yeah... I don't remember any kind of low-latency high-res VR video-passthrough from 2011 (or anything even close to that year)... not to mention anything at a consumer price point.

I suppose FB now falls under the usual umbrella of Apple criticism: they implemented something well, for which the possibility has been known in the industry for some time, and so we gotta slag on them for not inventing it?

I continue to have strong reservations about an ad company (especially that has such a dodgy history with user privacy) running this platform, but the technical achievements from FBRL are quite (pun intended) real.

As a Quest 2 owner I'd love some competition but nothing comes close to its feature set at its price point. Valve index or HP Reverb may have marginally higher resolutions, but Quest 2 is both standalone AND does PCVR/Steam completely wirelessly. For what its worth it's easier to sideload apps on a quest than an iphone, and I've gotten random android utilities to run on mine, most importantly Mullvad VPN. You can install a custom app launcher and a utility that lets you instantly access it by double clicking the side button, letting you skip all of facebook's stuff. You can browse the open app catalog here:

https://sidequestvr.com/

Between the oculus store, sidequest, Steam, and dropping in android APKs the device is a lot less closed than you think. Would I prefer if it was facebook account free? Yes but I dont know who else has the cash to compete with facebook on this one.

Quest 2 has higher resolution than Index. And if you look at pixels per degree then Quest 2 has almost twice the resolution of Index.

I own a Quest 1, and I've got every bone to pick with Facebook imaginable.

However, the Oculus Passthrough implementation (and their 'guardian' feature, for that matter) is unparalleled. It's many times more dependable than the SteamVR safety features, and surprisingly low-latency. I'm not at all surprised that it's local.

It's using https://www.khronos.org/openxr/

If you dig the idea of using VR for productivity then you should check out the SimulaVR project, which brings wayland/X support to Linux VR.

https://simulavr.com/

I hope competitors will deliver similar hardware but with more open ecosystem, check out https://lynx-r.com/!

I am interested by the technical aspect. Pass-through is not as obvious than sending the camera feed to the screen.

If the field of view, eye position, focus and latency are not right, the world will feel wrong. If it is way off, you are as good as blind. Close range focus is already a lost cause on current gen headsets. I don't know how tolerable the other factors are and how Oculus dealt with the problem.

> Pass-through is not as obvious than sending the camera feed to the screen.

Pass-through is sending the camera feed to the screen.

And we've had a limited pass-through for a while in that you can configure the headset to enable it when you leave the play boundaries. You can also enable it at will with a double-tap to the headset. I've walked around my entire office floor with it, greeting people. It works fine.

I lie awake at night and wonder, what if valve made competitive VR hardware, instead of the steamdeck?

It's clear the r&d is heavily in oculus's favor, and nobody is showing up to play. Even valves new headset is marketing fluff- "reading brain waves" doesn't work how they advertise it.

Valve has the advantage of not being a feeding tube for a bloated, predatory social network.

Because it already is one.

The other major advantage is that they’re not publicly traded which forces you to do unsavory things to chase analyst quarterly forecasts

Oculus sells their hardware at a loss, and make up for it by selling your data.

Valve doesn't care about being competitive. They're not market driven. They hire the best engineers possible, and let them work on what they want.

Sometimes that comes together in something beautiful (Alyx, Index) sometimes it results in something out of touch (steam machines) and sometimes it results in unending development hell.

> Oculus sells their hardware at a loss, and make up for it by selling your data.

Source?

What facebook does is worse than selling your data, but they rely upon people saying they sell your data, which they can deny, to hide what they actually do, which is analyze your data to sell behavior modification products.

Can someone explain how they can keep the passthrough video private? Sounds like they're saying the data is processed on a separate worker that can't access the network, but that worker obviously needs to pass data back to the main thread so that UI/content can be rendered in response to the video.

Perhaps they don't allow "custom" processing by the dev? E.g. can only use FB's face/pose/etc detector? That would be quite limiting though.

Maybe they only allow passing back a small amount of information per frame? That'd still make it possible to leak information over several frames (if GAN latents are anything to go by, there aren't that many bits of information in a human face, for example).

Like much in this announcement, the details are still to be revealed; but this is from the Developer blog :

   Apps that use Passthrough API cannot access, view, or store images or videos of your physical environment from the Oculus Quest 2 sensors. This means raw images from device sensors are processed on-device.

[1] https://developer.oculus.com/blog/mixed-reality-with-passthr...

The social aspects are pretty interesting. I game occasionally with my brother, who lives on a different continent... Chatting about life while shooting zombies or whatnot.

One of the best parts is hanging out, post game. Basic avatars and directional sound make the whole thing feel very different to video chat. A real world setting would give this a different dimension.

VR is in an interesting place. This initial hype has died down a little. Users no longer haul their headsets around for show and tell. Meanwhile, a bonafide juvenile market for hardware and content exists and incrementalism gets to do its thing.

Honestly, I'm kind of surprised there aren't more entrants into the hardware space.

Hardware projects are massively challenging even for the most well-funded operations in the world. Something about retailing, returns, support, compliance, certifications, internationalization, contract manufacturing and the rest make it difficult to enter. VR hardware is especially problematic as you have issues like vision prescriptions, nausea, HMD related injuries/RSI, safety of people and property around the players, and all kinds of other interesting avenues of liability. It’s capital intensive, and a lot of capital got tied up in Magic Leap, which appears to be a dud. I hope I’m wrong and they flourish, but my father instilled in me a strong suspicious of all Florida based businesses and they have yet change my mind.

Very cool innovations coming out from the Oculus team. I just wish the headset was more comfortable! I get immense pain on my face after using it for a few minutes, it seems to be a trade-off between fitting it snugly or having it slosh around my head.

The tech is great but the comfort is poor. Surprising because the Rift is decently wearable. Maybe Oculus 3 will have some strong improvements with the comfort level.

All my friends with the quest upgraded the strap. Oculus doing a crappy upsale, just a cost you need to swallow

I got a third party halo strap for my Quest 2 and it vastly increased comfort and ability to wear it for extended periods without fatigue. Might want to look into that.

I can second all these criticisms, as well as the advice below to upgrade the strap. I use the elite strap with the VRCover face pad and strap pad and it’s a massive improvement. I also find the VRWave strap pad makes it easy for me to run my link cable through the strap and behind my head which is also a comfort improvement in some games.

The strap upgrade is totally worth it. Much easier to find a comfortable, good fit.

I have experimented with cross-fading different 3D scenes / stereo 180 videos into each other and it's truly something that cannot be explained without experiencing it. Closest thing I've experienced to like, magic, since the first time I saw an automatic door at a grocery store as a child or my first time in an airplane.

I find wearing a VR helmet for a significant period of time to be mood-depressing. I think with AR the problem would go away, but I find it unlikely that passthrough AR will be much better than VR. I will try these new demos when I have a chance.

Can I now get a Spaceship "lounge" overlay for my flat in "AR"?

Check out Custom Home Mapper on SideQuest. He was experimenting with an "AR" mode back when it still required the menu glitch to activate passthrough

It would have been better if we could've started experimenting with this stuff more than two years ago, but Facebook kept access locked up for their use only. I'm glad they're slowly doling out more access, but the industry is now set back by over two years, and any experimentation will now happen on Facebook's terms. Of course that's all for your protection. Of course it is.

Jittery, I can already feel RSI making a come back.

Still, I’d love to have a bunch of documents open, just like paper

this page was definitely not GDPR compliant.

I just gave up.

What did they do that wasn't compliant?

Opt out instead of opt in.

Opt out next to impossible.

I couldn't think of a good use for a low resolution, laggy grayscale mix-in video of my real environment.

The page does show one use case that I'm very interested in--showing the rough position of my hands, keyboard and mouse. If there's to be advancements in VR development environments, it has to at least have an input mechanism that's as good as what we have in the real world. Having the same in VR is as good and can be improved upon.

Reply on news.ycombinator.com