With Apple Vision Pro out in the wild, it’s likely that Meta’s Quest Pro line will evolve to become more of a direct competitor. And we’ve actually got an idea of what an even higher-end Quest Pro 2 could potentially look like.

Apple’s Bar

With Vision Pro, Apple was clearly not aiming to make an affordable headset. Their strategy was to create the minimum viable quality bar for what they wanted the mixed reality experience to look and feel like. After packing in all the tech & features they thought was necessary, Vision Pro landed at $3,500.

That makes it only an indirect competitor at best to something like the $500 Quest 3. But clearly there’s a market of people willing to drop substantial money for a high-end mixed reality headset.

The Future of Quest Pro

So it makes sense that Meta’s Quest Pro line will evolve to compete in that high-end price range. But what could a Quest Pro 2 look like if Meta wanted to go toe-to-toe with Apple?

Well, it turns out we might have a pretty good idea.

Douglas Lanman is the Senior Director of Display Systems at Reality Labs Research, the R&D arm of Meta’s XR division. Late last year he gave a guest lecture at the University of Arizona, a school from which Reality Labs regularly recruits researchers.

Lanman spent most of the one-and-a-half hour session overviewing the research his team has worked on in recent years. They’ve built various prototypes separately exploring holographic optics, high-dynamic range, retina resolution, reverse-passthrough, and more.

While most of the headsets presented were purely research prototypes, Lanman wrapped up the session by talking about his vision for ‘Mirror Lake’, a concept headset that could incorporate many of the technologies his team has been exploring over the last decade. He presented the following rendering of what the headset could look like if it was actually built today.

Video via Mixed Reality News

And while he stressed this was just a concept dreamt up by him and his team—not a product roadmap—he said the concept was “practical to build now” with parts that could be readily sourced from the market. Whether or not it could be mass produced in the ballpark of $3,500 is an unanswered question, but Lanman said he thinks the industry is ready to reach “a new plateau” with a device like Mirror Lake.

He went on to explain that Mirror Lake would include compact holographic optics, multi-view eye-tracking, a varifocal display, reverse-passthrough, and baked-in prescription correction. That’s quite a mouthful… so let’s break each one of these down.

Holographic Optics

A huge challenge in making XR headsets compact is the need for optics to be placed between the viewer and the display. While Meta’s latest Quest Pro and Quest 3 headsets have impressively compact ‘pancake’ optics, Meta researchers think they can go even thinner. Their solution is a ‘holographic lens’, which is exactly what it sounds like: a lens baked into a hologram that can be practically as thin as a sheet of paper.

Multi-view Eye-tracking

Eye-tracking works best when you have multiple views of each eye. This means better source data for determining exactly which direction the eye is facing. But more cameras means more overhead for processing, heat, power, and cost. So Meta researchers dreamt up an interesting method of getting multiple eye-view of an eye from just one camera. By using a holographic element, the camera can look at the lens of the headset and see multiple reflections of the wearer’s eye. That could provide more views for better eye-tracking without adding more cameras.

Varifocal Display

Every XR headset on the market today uses stereoscopy—the overlapping of two similar images—to display 3D imagery. And while this is similar to how we see in the real world, it leaves out an important visual cue; the distance from which light originates changes how our eyes focus on what we’re trying to see. But because XR headsets have a fixed display, the light is always coming from the same distance. That means you can only set one focal distance, which would be like your eyes only being able to focus at once specific distance from you. This is known as the vergence-accommodation conflict (or VAC).

A varifocal display is any display which allows variable focus, thus solving for VAC. Meta has researched several solutions to this problem, but the one Lanman says could be part of Mirror Lake is using a stack of electronically controlled polarizers to allow dynamic alteration of the lens focus. Combined with eye-tracking, this would allow the system to focus the light for the specific part of the scene that you’re looking at.


‘Passthrough’ is what we call it when you put cameras on the outside of a headset and use them to show the wearer a view of the outside world. It’s kind of like your vision is ‘passing through’ the headset. Reverse-passthrough is what we call it when you put cameras inside the headset to show the outside world a view of your face. This is the same thing you may have seen on Vision Pro (Apple calls it ‘EyeSight’).

But it isn’t as easy as just sticking a display to the outside of the headset. Because the front of your headset sits quite far from your actual eyes, showing an image of your eyes that far out would look very unrealistic—like your eyes were somehow glued to the front of the headset.

Instead, you need a way to make the eye appear sunken back into the headset. To do that you need some kind of light-field display, which is a display that shows different views depending upon the angle that you’re looking at.

Lanman said that in the Mirror Lake render they actually did a ray-tracing simulation to show an approximation of what the reverse-passthrough on the headset would look like, given the components the headset would theoretically be made from.

Prescription Baked-in

For those who need glasses, having a headset that can support your prescription is important. Lanman says that given the compact nature of Mirror Lake, there isn’t much room for glasses. Instead he suggests that such a headset would be customized for each user’s specific eyesight needs. On Mirror Lake that could be achieved by manufacturing the holographic lenses to accommodate vision correction for each individual. That’s clever because it means correcting the headset’s visuals for each user without adding any additional bulk or components to the headset.

– – — – –

Meta’s first ‘Pro’ headset, Quest Pro, had an awkward launch to say the least. While it had some impressive optics, face-tracking, and improved mixed reality capabilities, it still end up feeling like a souped-up Quest 2. It played all the same content and the few things it did uniquely didn’t quite feel like they justified the $1,500 price point. Meta seemed to agree, because it dropped the headset to $1,000 not long after release.

Quest Pro | Image courtesy Meta

Making Quest Pro’s launch even more rocky, Quest 3 was announced not long after and brought many of Quest Pro’s improvements down to a much more attractive $500 price point.

Apple Vision Pro Will Soon Let You Convert 2D Photos to 3D & Share Them Live via SharePlay

If Meta is going to have a ‘Pro’ headset line, the headset needs to do substantially more than its consumer-focused headsets. Vision Pro—packed with tech and priced at $3,500—has seemingly opened the door to Meta further differentiating future Quest Pro headsets with a higher-end offering.

And considering that one of Meta’s senior researchers think it’s possible to build a Mirror Lake-like headset with parts that are already available, it’s not a leap to think that some of that new tech could find its way into a Quest Pro 2 or Quest Pro 3.

Newsletter graphic

This article may contain affiliate links. If you click an affiliate link and buy a product we may receive a small commission which helps support the publication. More information.

Ben is the world's most senior professional analyst solely dedicated to the XR industry, having founded Road to VR in 2011—a year before the Oculus Kickstarter sparked a resurgence that led to the modern XR landscape. He has authored more than 3,000 articles chronicling the evolution of the XR industry over more than a decade. With that unique perspective, Ben has been consistently recognized as one of the most influential voices in XR, giving keynotes and joining panel and podcast discussions at key industry events. He is a self-described "journalist and analyst, not evangelist."
  • MackRogers

    My take away from reading this article is that Apple ate Meta Labs lunch and they are back to the drawing board.

    They can not release a “pro” variant now that is day 1 inferior, and they are unable to compete with Apple.

    They have been able to putz around for years with zero competition and small iterations in the Quest 3(which is selling terribly btw versus the quest 2).

    They have hitched their entire future to QUALCOMM, the dumbest decision they have ever made. Qualcomm is the king of “slow rolling” innovation. Every single one of their chips throttles under thermal load.

    Meta is in a really really bad spot. I suggest Zuck fire EVERYONE associated with VR starting with Abrash and focus completely on the ray bans.

    • Xron

      Wonderfull that people like you aren’t managers of big tech companies, the innovation would be grounded at the start under you.
      Its always good to have competition, so I’m rooting for both companies, it might bring AR to casual people after some time.

      • Benjamin Räder

        Also the high attention on VR, created by Apple, will most likley push the whole XR-Market. So everything that pushes the tech is a welcome push in my eyes.

    • Gabe Zuckerwell

      This is what happens when you worship weebly brad.

      • Gabe Zuckerwell

        It goes without saying, I entirely retract this statement. Entirely.

        • ViRGiN

          i wonder who hurt you.

    • MeowMix

      wow… someone forgot to take their haloperidol this morning …

    • Christian Schildwaechter

      All chips throttle under thermal load, that’s a feature, not a bug, to get most performance out of the available power envelope. The MacBook Air and MacBook Pro use the exact same M2/M3, but the Pro adds active cooling, causing the SoC to throttle later and sustained performance to be higher.

      And while Qualcomm is an unpleasant partner to deal with, they are still the king of the hill when it comes to freely available high end mobile SoCs. MediaTek is now gunning for them with the Dimensity line, but achieve high speeds there by dropping the efficiency cores and instead use only performance cores, which increases the power draw and reduces battery life.

      Qualcomm is expected to respond with a similar configuration, and the XR2 Gen 2 already differs from the SD8 Gen 2 it is based on in the core configuration. SD8 features a 1/2/2/3 high performance/performance/efficiency/low power config, while XR2 uses 2/4 performance/efficiency cores, because it needs sustained performance, not occasional high load with long idle phases like a phone.

    • Andrew Jakobs

      Ah, which manufacturer of chips should they have used then? At the moment Qualcomm is the only one making highend socs, Apple only makes them for themselves. It’s simple to just develop your own soc, it took Apple many years and many buyout of startups/companies.

    • Mradr

      Was around 15% last time I check – that isnt “selling terribly” considering they drop the price of Quest 2 so low.

    • Kass

      Oof bro, a mix of some solid takes, but mostly brutally bad. Yeah meta lacking competition has sucked, I’m so glad apple made its way into the space. Also, I agree, I think meta is heading back to the drawing board for their high end headsets, and seeing how awful the quest pro was, I am thankful.

      Yeah, apple trounced meta like mad, the high end apple headset makes the high end meta headset look like a joke. However, pricing their headset an amount that people can feasibly pay, with only a mildly inferior experience, can go a long way. Day 1 inferior doesn’t matter if no one can afford the superior product.

      The quest 3 isn’t selling poorly. It just isn’t. Not as amazingly as the quest 2, but thats to be expected. Not sure where you got that idea, but it’s just factually not true.

      Meta hasn’t hitched their entire future to Qualcomm, I foresee them someday soon switching to making their own SOCs like apple and Samsung. And id argue that the innovation in Qualcomms chipsets is just as slow as everyone else lol. You just seem to have a lot of really fiery opinions, and not a lot of fuel to burn them on.

  • xyzs

    Just ignore the reverse pass through. Nodoby cares.
    However the holographic lenses coupled to MicroOLED and varifocal, this we care.l!

    • Christian Schildwaechter

      You cannot pair holographic lenses with current microOLED, as the lenses need coherent light that microOLED backlights don’t provide. Current backlights are actually white instead of red/green/blue as would be needed, and the light is emitted at a random polarization, causing 90% of it to be lost even in pancake lenses. Varifocals are similarly unpractical today.

      You may not like EyeSight/reverse pass through, but compared to holographic/varifocal lenses, this is a much easier to solve problem at low cost, even with the lenticular lenses Apple added to solve the issue of multiple viewer perspectives.

      • Mr. Fox

        So wouldn‘t that theoretically be possible with eMagins uOLEDs and their direct pattern technology ?

        • Christian Schildwaechter

          No, direct patterning just means that the OLED layer itself emits RBG instead of white that later has to be separated into RGB with color filters. They are not emitting coherent light, meaning the phases of the light waves are shifted randomly. In coherent/laser light, the waves align except for a constant shift, allowing their amplitudes to add up and the “beam” to carry a lot of energy.

          A hologram is basically a film with small, reflecting particles that are oriented in a way that light of a specific wavelength passes exactly one way. This allows to create holographic lenses, by having light go the same way as it would through a normal lens. It actually reconstructs intensity and phase of a wave field, you need light of the same frequency and phase to display what is recorded in the hologram. There are different ways to get coherent light, but most of them rely on filtering out unwanted frequencies/phases, making them very inefficient. The only practical light source for use with holograms/holographic lenses are lasers.

          • Mr. Fox

            OK, thanks for clarification.

    • Mradr

      Agree, personally, I would love to see it as more of an “add on” than something that has to be built in. For people that do want it, but for everyone else, there is no reason to pay extra for the reverse pass through. There are far more important areas that could see the extra money go to than some fashion statement. Eye tracking for sure DOES need to be there. If anything, help figure out way to remove the need for space for glasses or extra lenses for people that need vision correction.

  • g-man

    Of course even thinner optics would be great but if you look at the iFixIt QP teardown the optical modules are already quite small. It’s the whole rest of the chassis that makes it a face brick.

    What I’d like to see in an AR-first design is a rear battery halo with just those modules hanging down in front of the eyes with no light blockers. Mount everything else above them in the forehead pad and in the arms.

    Varifocal would be awesome but there’s so much to gain just by moving away from the VR-centric face brick design. I don’t think anyone cares about creepy eyes.

    And for the love of god copy the AVP UI.

    • xyzs

      The AVP optics issue is not only their size (that is still too big regarding the poor FOV), it’s their weight, thick glass is not your neck friend. Holographic lenses would be so much lighter in comparison. And they would bring the center of gravity closer too.

      In the ifixit video, we can also see that the fans takes lots of space and push the electronic away, electronics that look badly over engineered and made of too many parts.
      They should have put the compute unit with the battery to leave the headset to a simple sensor and display role maybe.

  • Cl

    While there are some cheap android phones there are also quality expensive ones. Not sure why you think VR can’t be like that too. You don’t really make sense.

    • ASADVD

      This is the way most apple fanboys think and it’s really annoying.

  • Christian Schildwaechter

    Most of the tech listed isn’t ready. Holographic lenses need coherent light/lasers, still far too large/inefficient. Varifocals exist in labs, but VAC seems to be a minor problem on AVP, with testers reporting more eye fatigue from looking at a monitor. So Meta’s Vision Pro competitor will be technically similar to Qualcomm’s new XR reference HMD. And that will be wise, a lot cheaper, and enough to compete with AVP.

    AVP cost estimates are USD 1400-1800, with USD 700 microOLED displays, ~USD 150 M2, USD 100+ for assembly (teardowns by iFixit/Phone Repair Guru show why) for~50%-65% of the total cost. Future AVP will use cheaper BOE/SeeYa displays, and Samsung/Meta could launch with these. Other parts aren’t expensive, the EyeSight OLED ~USD 15-30. Apple is a high margin company, so USD 3500 isn’t a baseline. They can only build 450K AVP in 2024 due to Sony’s display production limits, a high price will reduce demand and thereby shortages and scalping.

    XR2+ Gen 2 is slower than M2, but supports the same 12ms passthrough, 4K per eye. The M2’s power is needed to run iPad apps inside MR, as iPad Pros use the same SoC. The UI or virtual cinema need a lot less performance. ETFR may still be an issue, and it’s essential on AVP, where disabling it reduces edge blur, but causes stuttering. Pretty sure Meta/Samsung can build a close competitor for USD 1000 using cheaper microOLED, pancakes, Qualcomm SoCs and similar components. Not quite as fast/polished as AVP, but creating the software for a smooth experience plus content deals will be the harder part.

    • Newlot

      wow so apple sells vision pro at more than twice the hardware cost? that would make profit margins extremely high, can we expect future vision pro versions to be sold closer to hardware cost?

      • Christian Schildwaechter

        Apple’s gross margin for the last decade was ~40% (45% in Q4’23). So on average a USD 1000 device cost Apple USD 600 to produce/sell/ship, with USD 400 profit. With USD ~1600 production cost for AVG, the gross margin would be 54%, significantly higher than Apple’s average, compared to ~0% margin for Quest 3. Margin varies with product type, basic models have a smaller margin, so USD 599 for an M2 Mac Mini 8GB RAM/256GB flash is a good deal. But Apple charges ~4x of what you’d pay on Amazon for extra RAM/flash.

        Upgrading AVP from 256GB to 1TB adds USD 400, the same as for Mac. So AV(P) pricing will probably follow the Mac, reasonable bases model with very limited RAM and storage, and A LOT more for configurations for more serious use. Assuming they reduce the (still very experimental) AVP production costs by about 1/3rd for a follow-up, non-pro base model with 25% margin, USD 1300 would get you an Apple Vision with 256GB, USD 1500 512GB (~35% margin), USD 1700 1TB (~45% margin). The AVP M2 is the 8 CPU+10 GPU core variant. Upgrading the same on Mac Mini to 10+16 adds USD 300, 12+19 USD 600. A maxed out Apple Vison with 1TB, 12+19 M2 Pro would cost USD 2300 (~75% margin).

    • Sofian

      VAC seems to be a minor problem on AVP

      I am sure it is when pretty much all you can do is watch 2d screens at a fixed distance…

      • Christian Schildwaechter

        At some point we will have to switch to varifocals. But AVP and competitors currently struggle with weight comfort issues when used for more than 30min to a few hours, and are clearly not yet feasible to be worn all day while doing lots of 3D tasks requiring precise depth perception.

        There are a lot of problems we still have to solve before VAC has even a chance to become the main problem that stops people from using XR HMDs. Which is good, because adding the varificals we have today would increase the other comfort problems a lot more. So letting them mature first and live with VAC issues for the time being makes more sense.

      • Traph

        Oh wow I didn’t even think about that. 75% of the time, when I’m having to heavily focus on a software keyboard in VR to type something long, I go absolutely cross eyed for a sec when I shift my focus back to something elsewhere in 3D space.

        “Focusing on UI elements” is like 90% of the interaction on the AVP. I’m demoing it tomorrow, will be interesting to see how much of a problem it is for me. Hopefully it’s not some hand held guided “experience” thing where they’ll shy away from anything that can cause issues.

    • lhl

      In Lanman’s talk he mentioned the Mirror Lake concept design included a BOM/component list, all of which could be sourced today. He also stepped through the CAD diagram which showed the the display element labeled as a “laser backlit LCD” – I did some searching and sure enough, JDI was showing off a 2K laser backlit display prototype for VR HMDs back in 2021 (as a free bonus you get 97% BT 2020 color gamut), so I don’t think production of HOEs is quite as far-fetched as you posit.

      I think VAC eye-strain/cognitive load is very under-appreciated, since there have been decades of research showing that it *is* a big deal for a lot of people, especially if the goal is to spend all day in a virtual workspace.

      Personally, I’d rather have electronic varifocal (DeepFocus was open sourced *6 years ago now*, AFAIK the PDL/half-wave plate system is pretty well baked now as well) and I’d be OK skipping the holograms/reverse pass-through if we could get to 60ppd and varifocal (eg, I’d be happy if a Butterscotch Varifocal were available for sale, assuming you could get a proper windowing system and the ability to run Desktop apps in more than a single monitor).

      • Christian Schildwaechter

        Haven’t seen the talk yet, but “tech isn’t ready” doesn’t mean it doesn’t exist. I have a laser cutter with a 5W output blue laser diode. You’d need green/red too, probably two sets, but those exist small, powerful and cheap enough (USD ~30 each) for use in HMDs. The issue is the whole module drawing 20W. The ~8*8*5mm³ diode sits in a 35*35*75mm³ aluminum heat sink plus electronics/fan to remove those 75% of the energy turned to heat. Efficiency is better with lower power, and creating at least a prototype doable. But that’s not the same as “ready” for a product to be sold.

        Apple could offer AVP with M2 Pro or Max, for improved performance. M2/Pro/Max have about 20bn/40bn/67bn transistors, but Pro only adds ~4.5mm, Max another ~5.5mm to each die edge, with less difference for the SoC, making size not the issue. Power consumption and head dissipation are, these go up significantly. The existing cooling might even handle an M2 Pro, with cranked up fans making AVP go from silent to PS4, but battery life would tank. Still feasible and much easier done than holographic lenses with laser backlight, yet Apple doesn’t offer it because the “tech isn’t ready”, as in practical for an actual product. Still speculating, but that’s my guess why that they “could” build Mirror Lake today at reasonable cost, but don’t sell it. VAC will become an issue with longer use.

        • kraeuterbutter

          when you see the power-consumption of the Vision Pro, it seems, the M2 is by far not running at its full potential..
          so i think: M2 pro or Max ar not realy needed sofar

          • Christian Schildwaechter

            You can disable ETFR in the AVP developer settings, which leads to the image getting less blurry towards the edges, something a lot of people notices and clearly visible in screen recordings. But if you do this, the image starts to stutter. Foveated rendering reduces the GPU load, so the M2 with 10 GPU cores is already taxed enough that some (minor) compromises regarding peripheral image quality had to be made.

            It’s doubtful there will be a lot of performance issues, esp. since most existing apps will be from iPad, which features at most a somewhat slower M2. But just like on MacBook, the Pro/Max models would be interesting for some tasks to be faster and esp. to allow for more complex graphics, so I’d expect Apple to offer more SoC tiers on future HMDs, once this becomes technically feasible.

          • kraeuterbutter

            when the AVP is running 2 hours with its 35Wh battery, thats 17W powerdraw in average…
            the M2 is rated with an TDP of 20Watt i think
            this 17Watt needs to power: the M2, the R1, all the sensors and cameras, the fans/cooling and the 2 Displays, Sound…
            so thats what i meant: i think the M2 is dialed down, not running at its full potential because of heat and energy-reasons…
            so wondering if a bigger prozessor (more powerconsumption) will find its way into the goggles.. or we will see more powerfull processors because of newer versions, with higher efficiency

  • Duane Aakre

    Why can’t Meta just marry the best features of the BigScreen headset and the Apple headset?

    1) The on-head weight should be limited to 200 grams. BigScreen shows it is possible, if you move all the compute off the headset.
    2) Apple’s headset will normalize the use of pucks. Put the compute and the battery in the puck. Since the puck doesn’t have any real weight constraints, have a minimum of four hours’ worth of battery capacity. And put some fans in the puck so you can run the compute chips without throttling for thermal reasons.
    3) Why hasn’t Meta teamed with Roku or Amazon to get streaming stick functionality into their headsets? If a $30 streaming stick can stream all the media sources in 4K, a headset should be able to do it, too, with a nice simple UI.

    • ViRGiN

      Bigscreen is as dumb headset as it gets. Just lenses+screens.
      Physics still apply to XR. Longer paths means higher latency. Wiring on-headset sensors back to external puck also adds complexity. The cable will wear out.
      Noone has really done it so far, so there are plenty of reasons.

    • Christian Schildwaechter

      Because reality is driven by physics, and we haven’t come up with time machines to bring us technology that will still take years to miniaturize and optimize. Apple only externalized the battery, while all the processing still happens very close to the sensors, which have to stay on the headset. And Bigscreen most likely came up with the Beyond only because Meta made special deals with Netflix and Fandango, reducing their fee for selling/streaming content on Quest, while Bigscreen was still charged 30%, making their business model unsustainable.

      • kraeuterbutter

        well.. all PCVR-Headsets work with cables as well..

        so.. it should be possible
        the AR-Glasses with “pucks” are also there, they also do not only provide batterypower to the glasses but also do all the hard work..

        • Christian Schildwaechter

          No, the Beyond can only be so light because it’s a VR HMDs with almost no onboard sensors and data flowing mostly from PC to HMD. It connects via USB-C with DisplayPort ALT-mode to reassign the superspeed data lines to video transfer, leaving only a USB-2 speed data port back to the PC. Most IR tracking cameras are lowres b/w, the data which could be squeezed through USB-2 with minor preprocessing, so adding inside-out/hand/eye tracking might work

          But data for 20 AVP sensors with hires color passthrough (that MagicLeap with puck doesn’t have) cannot, you need processing close to the sensors. XR2 Gen 2 can connect 12 cameras directly via MIPI/CSI on the PCB, each with several lanes, and the Quest Pro required an extra FPGA for combining camera to work with only 7 ports. You could move the M2 with RAM/flash into a separate puck/PC to render the image there, but only because all sensor data is handled by the R1, enabling passthrough/tracking even while M2 restarts. And of course the sensors have to stay, so you’d save maybe 100-150g. AVP is amazingly small for what it does, and to get it lighter you have to cut features or improve technology.

        • Andrew Jakobs

          No they don’t all, you can use HTC Vive headsets with a wireless module, and it is possible to move the battery for it on top of the headstrap so you don’t have that annoying batterycable (still hate that cable, but it’s a major improvement over having the wire to the PC, which I already had suspended on the ceiling with kiwi pulleys).

          • kraeuterbutter

            all headsets (also the HTC Vive) runs over cable… still.. also with the wireless module (which is not the point in this discussion) transfere the date to the displays, gets the data from the sensors through cable

          • Andrew Jakobs

            Uhh, how is it running over a cable when used with a wireless module? It’s just like a Quest 3 using wifi wireless.

          • kraeuterbutter

            the Vive has no Wifi-Modul..
            so: the wifi-Modul is connected via cable to the headset… its a short cabel, but its a cable..
            you remove the original cable, and plug in the cable from the wireless-modul

          • Andrew Jakobs

            What a crap explanation, the module makes the vive wireless as it sits on top of your head connected to your headset, the only (annoying) wire required is the cable to the battery, but hee, if that’s the case, the Apple Vision Pro isn’t a wireless headset either. And with your stupid explanation neither is the quest a wireless headset, as it got wires inside of it, connecting the battery to the mainboard or the headphones.

          • kraeuterbutter

            and my point is not hat we need a computing unit 5m away…
            my point is: get as muchweight out of the face…
            put the computing on top or at the back or to the sides.. but not in the face…
            a quest3 has with orginal strap 90% of the Weigt or even more in and on the face…
            seeing how much more comfortable the pico4 can be compared to the quest3 its clear for me how important a good balance is

          • Andrew Jakobs

            Oh yeah, I agree completely, the Pico is out of the box way more comfortable as the Quest 2/3.

    • Andrew Jakobs

      Bigscreen doesn’t show it’s possible, it needs external tracker, is wired and not a standalone. Sorry, but no way I want to go back to wired crap. Everything should be in the headset/strap, unless it’s all completely wireless.

      • kraeuterbutter

        WMR-Headset showed it:

        computerpower 5 meters away
        tarcking in the goggles…

        • Christian Schildwaechter

          The nIR tracking cameras are usually 400×400 pixels and only used for occasionally correcting drift from the IMUs inside the headset and controllers, so they don’t even have to run at high frame rates. The distance isn’t the issue, with data moving at about the speed of light over cable, the required bandwidth of all the other sensors is.

          • kraeuterbutter

            welll you will right..
            still: a Varjo XR4 has a higher resolution than the AVP and maybe also a better Passthrough
            and: its with cable..

            i also dont mean to use a 5m cable…
            maybe you just need 30cm cable to the back of the headset, so it its more balance..

            like pico4: its little heavier than the Quest2 AND STILL feels WAY WAY lighter on your head..
            so when a headset could put the cooling (question: does only the proc need cooling, or do the displays itself need cooling ?), prozessor, logic-board, … to the back -> that would safe weight on front and add needed weight on the back for balance

            maybe it would even be possible to “hardwire” it with only about 10cm
            meta i think showed a prototyp were some of the componets were not put in the front-visor, but left and right on the sides…
            for all headsets: its needed to get weight away from the front.. its the main cause for discomfort

          • Christian Schildwaechter

            In principle you can put the compute wherever you want to, the question is how practical the result will be. I don’t know how Varjo realized the passthrough. AFAIK it uses two 12MP cameras, compared to 6MP on AVP, leading to a much improved quality. And since sensor chips are actually very small, my guess is that Apple stuck with the lower resolution because doing all the required processing for two 6MP images within the 12ms they offer is all the R1 can to within a reasonable power limit. Those limits do not exist on the XR-4, which also aims for less than 20ms photon-to-photon latency. The XR2 Gen 2 in Quest 3 can do 12ms passthrough according to Qualcomm, but that’s apparently just pumping the data from the camera to the screen. According to Brad Lynch, measured passthrough latency on Quest 3 is typically 50ms do to the extra processing needed.

            Varjo could solve this by using a cable combing DisplayPort and USB-3. They may have to do that anyway to use DP2 with higher resolutions than possible with DP1.4 over USB-C on the Beyond (not sure about that). They could also have included some SoC or FPGA on the HMD itself, as they don’t have to care as much about weight, power consumption or thickness as the AVP. And yes, not only the processor needs cooling, but e.g. the bright dual layer backlights of the microOLED in AVP will generate a lot of heat too, and usually also get less efficient when running hot.

            In the end it’s about balance. You can create a VR power house like the XR-4 by using a fast desktop GPU, rather thick and well shielded cables with lots of bandwidth, hot/heavy components on the HMD itself, you just have to deal with the consequences, like needing a counter-balancing headstrap and a high price. You could do all that on the Beyond too, but of course you’d lose the main benefit of the ultra low weight. AVP crams everything into a very complex device, making it very expensive, Quest 3 tries to use less components for an affordable price, but cannot provide the same feature/performance level.

            Someone could go and take the sensor set of AVP and combine it with a belt worn compute unit and battery connected by a thicker cable, or split it so that passthrough/tracking remains on the HMD with generic compute and power external. Technically its possible, but it doesn’t really fit any particular use case. If you really need the power and quality, you’d want a PC/XR-4 with much higher performance, but not portable. If you really want it light, you want a Beyond with everything not necessary stripped out. If you want it affordable, you have to go the Quest 3 route with everything in a single device with low build complexity. The AVP is the really odd one here, because it is technically a very untypical Apple device, very heavy with external battery, and prices so high it will only sell in low numbers. I still consider it mostly a developer/first mover device, to test what will work, and a 2nd/3rd gen consumer device will be lighter (though not Beyond light) and somehow integrate the battery for a much smoother experience.

            A HMD with an external compute puck may be useful for a specific use case where you need more power than can be crammed into the HMD itself, while at the same time staying mobile, so e.g. an industrial AR HMD. I just don’t see anybody releasing such a device for consumers due to the involved extra hassle of having to somewhere attach the puck (AVP assumes you use it stationary with the battery pack on the table/sofa etc), and extra complexity due to the split device, making it more expensive. Even putting the battery at the back of the head leads to a number of realiability issues. Maybe someone will release a puck device combined with a HMD part similar to Beyond with passthrough and eye tracking, but I somehow doubt it, even though it is technically feasible.

        • Andrew Jakobs

          But still needed a cable to provide the video…

  • eadVrim

    Why Meta would imitate Apple Vision Pro!! I think the opposite is what was happened.

  • eadVrim

    Why would Meta imitate Apple Vision Pro!! I think the opposite is what happened.
    For me I prefer the current competitor Quest 3.

  • John G

    Agreed. It’s a “I’ve never seen XR” attitude.

    When people say “Headsets are too isolating”. I reply “What are you taking about?” When you put one on, you’re anything BUT isolated. You’re suddenly in a huge world! It’s only isolating for people on the outside who can’t see the world you’re in.

    That’s what screen cast is for, if someone else is in your play space.

    Once more people are used to spending hours in an HMD, they’ll know the person they are looking at is indeed in another world, and won’t need the eyes to remind us they may be in pass through.

    I mean, no one gets electronic dark sunglasses.

    “But dark sunglasses are so isolating!”

    == John ==

    • Christian Schildwaechter

      The whole point of “spatial computing” is that it happens in the same world as we live in today. Not in another world you enter using an HMD with other people somewhere else, projecting your view onto a local TV for those present to watch.

      And electronic dark sunglasses are for noobs, the pros use Joo Janta 200 Super-Chromatic Peril Sensitive Sunglasses to deal with the daily issues of real life.

      Joo Janta 200 Super-Chromatic Peril Sensitive Sunglasses have been specially designed to help people develop a relaxed attitude to danger. At the first hint of trouble, they turn totally black and thus prevent you from seeing anything that might alarm you.

      Douglas Adams, H2G2

      • John G

        LOL on the sunglasses, but my point was that no one complains that dark sunglasses are isolating – yet they are. You can’t see people’s eyes in them, or tell where they are looking. No one cares. It’s neat, don’t get me wrong, it’s really cool, I’m just not sure it really justifies the price tag. Time will tell.

        I think the EyeSight is a transition thing.

        “Spatial Computing” is a buzzword (and not a new one, nor an Apple one). I get that the AVP isn’t “VR” (although they do sell a light blocker), it’s more XR, but the use case isn’t firm yet. Time will tell.

        My prediction was that the M2 chip isn’t really powerful enough to do full VR, which is why Apple stresses XR (cough, ‘Spatial Compiting’) for the AVP.

        I’ve done some benchmarks on Apple Silicon on desktop to answer a Quora question on why SteamVR pulled support for MacOS, so I’m not convinced, but haven’t looked really hard for AVP benchmarks. I honestly hope it’s better than the desktop M2 silicon.

        • Christian Schildwaechter

          People are bothered by others wearing sun glasses inside. Casey Neistat wears sunglasses in all his videos even inside, so people won’t notice him looking at the camera display instead of the lens, as people find this very irritating.

          IMHO spatial computing is different from VR, but both are just names for certain function sets. Technically all HMDs are smartphones with added lenses, and smartphones small computers with added batteries. Whether you call it spatial computing, metaverse or the end of civilization doesn’t really matter.

          AR/XR is much more compute intensive than VR due to added sensor data processing, causing battery life to tank on Quest Pro with passthrough/eye tracking enabled. And the AVP M2 is way more powerful than anything Qualcomm. The version in the AVP is the 2nd lowest config with 8 CPU and 10 GPU cores. Desktop M2 Ultras go up to 24 CPU, 76 GPU cores, M3 Max (=1/2 Ultra) for MacBooks now to 16 CPU, 40 GPU, easily keeping up with gaming GPUs or beating them in special use cases.

          MacOS SteamVR support was intended for VR content creators, not users, never got stable and ran only on Intel CPUs with miserable iGPU performance or dGPUs. M1/2/3 are different beasts, leading to the first Macs with enough performance for modern PC games.

    • Sandeman21

      “When you put one on, you’re anything BUT isolated.”

      Yeah, welnot sure why you scrub it off, but obviously, they mean isolated from the people around you and looking at someone’s face is THE #1 social feature of humans.

      Someone not looking at you in the eyes while you are talking is considered problematic behaviour and these people who do not look people in the eye often become social outcasts.

      People do not feel like they are connecting with you, and it feels awkward.

      It is the top advice communication coaches give for interpersonal relations and building rapport.

      But the same happens when you talk to others and they are looking at their phone or computer screen etc. With the headset on, it feels more.

      But there is no way fixing it. Certainly not by plastering someone’s face on the headset. It is what it is.

      btw, sunglasses are isolating too, and they are considered rude in social circumstances unless the sun is really in your face and can’t help it.

      They have just become a bit more acceptable as a mild nuisance because the person wearing them is still looking at, and talking to you.

  • ApocalypseShadow

    Instead of pushing articles that will help promote AR and VR, we keep getting these articles that push “this against that.” How Facebook can compare to whatever headset name is for the day. How Facebook can ride Apple’s coattails.

    Apple not only is not competing in VR, they haven’t even mentioned Facebook at all in their promoting of Vision Pro. Truthfully, not any VR headset. But we’ve seen Facebook copy Apple and do snide remarks against Apple. You don’t even have to like Apple to see the narratives. The slant on how Facebook can keep their modest lead when it’s only in VR over others that aren’t spending billions and taking losses to gain market share.

    The question remains on if the public is interested in Facebook hardware when Apple has created good standing with the masses even at expensive prices. Are businesses going to accept Facebook hardware over the likes of Apple? Is Facebook hardware even able to keep their value like Apple products do? Even old Apple hardware holds value sold used. Not that it isn’t possible for them to succeed in creating a fan base of their hardware as well. But they are way behind Apple in this area besides in software excluding VR games.

    Point being, how about promoting all hardware to be successful than pushing articles to promote one particular company. Where there’s definitely good hardware. But we keep getting the dangling carrot about their software or their future hardware. There’s more than one company trying to grow AR and VR.

    • Put down the Kool Aid. Leave the Cult. Apple hardware is trash and it litters the landfills of too many 3rd world nations, mostly because it’s lack of repairability.

      It’s HYPE. Nothing else. Don’t be a sheep.

    • Ben Lang

      Competition leads to improvements across the board. It’s better for all parties involved: companies and customers. This is what we’re seeing in action right now.

  • STL

    Please stop attempting to incorporate heavy front-weight displays into the design of everyday sunglasses. This design approach is quite frustrating. Looking ahead, either the glasses need to be exceptionally lightweight (perhaps in 10 years) or they must be supported by a truly ergonomic head strap. As of now, there’s no superior head strap option available than the halo head strap used in PSVR/PSVR2.

  • Gato Satanista

    I use glasses and their proposed solution for this is a very bad one. In my quest 2 I buyed an adapter for prescriptions lens and i think this is the most versatile solution because…
    – It allows me to easily share my headset with family and friends.
    – if my prescription changes… with is very common, I can just buy new prescription lensses.

    A fixed correction on these holographics lenses is a very, very, very stupid idea. Please, dont do this. Just sell the thing with adapters already

    • Christian Schildwaechter

      Changing holographic lenses isn’t possible, but they are very cheap to produce, much cheaper than the adapters you bought for Quest 2. So as long as they can easily be swapped, it should be affordable/no problem to have a set for your prescription, a generic one for visitors and even individual ones for other family members, and also replace them every time your prescription changes.

      Holographic lenses are actually pretty cool, very flexible, light and cheap to produce. Not quite as cheap as stamped plastic spheric or Fresnel lenses which cost just a few cents, but way cheaper than pancakes or the extra lenses you had to buy to correct the Fresnel or pancake lenses in exiting HMDs so far. The complicated/expensive part is the backlight required for holographic lenses.

  • fcpw

    What Zuck doesn’t get is they can have the most magical headset OAT- I don’t trust Facebook and their desire to sell my personal info for pennies. Fix that and the rest will follow.

  • The prescriptions baked into the holographic thing would be great for the owner, but would make a pain share the headset with family and friends

    • Christian Schildwaechter

      Since holographic lenses are basically just holographic film that stores a “picture” of the light path, they should be rather cheap to produce, much cheaper than e.g. pancakes that have to be very precise and stable to reduce ghosting from multiple reflections/refractions through several physical media of different thickness.

      So the easiest solution would be buying a lens set for every family member wearing glasses, which could be sold just like Apple’s Zeiss inserts, with a lower price for standard “readers” and a higher price for prescription lenses. Only the production costs for either should be less than USD 10 for a pair, even with custom lenses that would require “writing” the optical path with lasers, basically film exposure, which takes a fraction of the time it takes to grind and polish prescription glass lenses.

      I’m not sure how hard it is to properly align the required laser backlight with a holographic lens to be useable in a display, esp. when using RGB, so this may be a limit. But the lens is just thin sheet of plastic with small reflective particles inside, oriented with coherent light and later permanently fixed like photo film.

  • This is… so … dumb. The entire article. Just the basis alone, absurd!

    The Quest is the dominate item. The Quest is superior in all ways. This bad joke from Apple is a paperweight. I’m so sick of these Apple fanboys, instantly promoting the latest Apple turd as the “Market Leader” of whatever market that absurd company chooses to bumble into next.

    PUT DOWN THE KOOLAID! Just because some noob at the Apple store patted you on the head for your latest $10,000 purchase doesn’t mean you know anything at all about technology, nor have you actually used your money well.

    You know what Facebook should learn from this exercise in hype? Ignore it. Or even better, just point out the ONE BILLION THINGS the Quest does better… for 1/8th the price. Steal that hype train of stupidity and reroute it back to sanity.

    Seeing this article here disappoints me greatly. You are hyping up GARBAGE.

  • What “Quality”? Did I miss the hand controllers? PCVR link? The software library of solid VR titles? What “Quality”?

    Oh, you mean HYPE. Absurd, sheepish, HYPE. Yah, not much hype. Just quality for a great price. But you keep spending that money. China needs YOU!

  • Joshua Kariakin

    q3 has been a big leap forward but it still feels like we need a 30-50% breakthrough on the core techs of resolution, eye tracking, hand tracking, weight, and fov to hit a really solid sweet spot that will ensure VR finally sees mass adoption. And at a similar price point or maybe a bit higher but also subsidized even more. What really boggles the mind is that Meta seems to think their UX and limited offerings of quality experiences are good enough, or at least they have not been quick to bring them up to speed, even given the billions they pumped in already like W…T…F, truly awful decisions and a lot of low hanging fruit imo for improving comfort and ease of use. It’s like they believe some medical killer app will compensate or the hard core VR true believers, or chat VR people, will make up for that deficit? The lack of a better head strap and battery pack is beyond stupid , and given their engineering prowess surely they could have figured out how to suspend the weight evenly to get it off your nose and cheeks, if third parties did a halfway decent job for way less.
    But nope, nikel and dime is early adopters with half ass straps and battery charging stations – a really bad look and a big turnoff to first time VR people who already may have been hesitant and needed absolutely every barrier smoothed out. At the very least create a vision and roadmap to get people excited Meta, for the day we have a super comfy screen replacement that has eye tracking , super high res, but also a way to organize and customize my UX and build things right off the bat – a virtual home say with 3d objects and portals are links to experiences. Take lessons from rec room for God’s sake, and ditch the preachy lame 3d videos and anything less than 4k on your video offerings. I can’t sort or find shit, it’s scattered around, I can’t bookmark things all in one location, I can’t access my chrome bookmarks, I can’t arrange my games or remove forever my old oculus go games from the list, there is no consistent way to rotate my view when in a 3d 180 or 360 video, like seriously curate some shit ! The killer “app” might well be just the ability to leisurely wander through worlds and experiences and to build and share them – at least for a big segment of us; so make THAT easy to do and less expensive please! If u support that, the community will build out an amazingly rich multiverse that aligns with their interests I think. We need more cross platform with PC flat screen in terms of spectating and interacting with VR so it doesn’t feel as isolating and so curious parties can get a taste of games before buying.

  • Sandeman21

    Yeah no.
    Inverse pass, is rumored to be abandoned by Apple too. It is a stupid gimmick that they used to market differentiation but obviously abandoned the thought. You do not see the feature anywhere.

    Also No. The 3500 is a ridiculous price. Meta should NOT follow it. 1500 is a great price point for a pro device.

    The rest is evolutionary and Meta is already on its way there, ahead of everyone else.

    The only thing that they absolutely must do, but unlike Apple they don”t have any control of, is the SoC. Apple makes their own awesome chip and Qualcomm really needs to step up. They do not seem to care much.

    Standalone VR needs seriously powerful chips and sadly Qualcomm has not been able to produce one although they are trying for years to accomodate Microsoft for Windows Laptops and other portables. This needs to change because if the price of the next Apple headset drops significantly, Meta could lose their Pro market.

    Yes you can connect the Quest on a PC, but the standalone visual quality is deplorable.The GPU is nowhere near where it needs to be.