Pico has unveiled its next XR headset, the Pico 4 Ultra, which is slated to launch in China next month starting at around $600.

It was reported early this year that Pico was preparing some sort of XR headset based on the Pico 4 platform following a supposed leak.

Now Byte Dance’s XR subsidiary Pico Interactive has finally announced Pico 4 Ultra, the promised ‘next-gen’ platform refresh of its Pico 4 headset, replete with color passthrough cameras and depth sensors for mixed and virtual reality, and a second gen version of Qualcomm’s Snapdragon XR2 chip—notably the same chipset used in Meta Quest 3.

Image courtesy Pico Interactive

Here’s a quick look at Pico 4 Ultra’s specs:

  • Processor: Snapdragon XR2 Gen 2
  • RAM & Storage: 12 GB RAM + 256 GB Storage – LPDDR5 + UFS 3.1
  • Wireless Connectivity: Supports Wi-Fi 7 (802.11 a/b/g/n/ac/ax/be), Bluetooth 5.3
  • Mixed Reality Sensor: 32 MP color passthrough camera ×2, iToF depth-sensing camera ×1, Environment tracking camera ×4
  • Display: 2.56-inch screens at 2,160 × 2,160 pixels (× 2), 1200 PPI (pixels per inch)
  • Rendering resolution: 1920 × 1920 (× 2)
  • Refresh rate: 90 Hz
  • Optics: Pancake lenses at 105° FOV, 20.6 PPD (pixels per degree)
  • IPD Adjustment: 58 mm–72 mm
  • Audio: dual stereo speakers, 4 microphones and supports spatial audio recording
  • Battery: 5,700 mAh rated capacity, 5774 mAh typical capacity
  • Charging: supports QC 4.0 / PD 3.0, 45W fast charger
SEE ALSO
'Mobile Suit Gundam VR' Trailer Shows Off Story for Interactive Anime, Coming to Quest Fall 2024

Pico 4 Ultra is now available for pre-order in China, priced at 4,300 RMB (~$600 USD), with shipping slated to begin September 2nd.

It’s uncertain whether Pico will bring Pico 4 Ultra to additional countries, however it has been certified in South Korea, UploadVR notes. Pico currently sells its original 2022-era Pico 4 in many countries in East Asia as well as Europe, however not in the United States.

We’ll be keeping an eye on Pico’s global site in the coming days for greater indication of launch regions.

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.


Well before the first modern XR products hit the market, Scott recognized the potential of the technology and set out to understand and document its growth. He has been professionally reporting on the space for nearly a decade as Editor at Road to VR, authoring more than 4,000 articles on the topic. Scott brings that seasoned insight to his reporting from major industry events across the globe.
  • ViRGiN

    I know at least one YouTuber who will "love it!" and never use it again.

    without Meta OS this thing is dead on arrival.

    • kakek

      Yup.
      incredible that they can't understand the basic reality : mobile VR headset are basically consoles.
      Consoles are sold by games.

      • Christian Schildwaechter

        (Much)TL;DR: XR is moving AWAY from HMDs as game consoles towards general purpose devices, mostly due to the much larger audience; there have been lots of media consumption focused HMDs in China for a decade that were never released somewhere else; Pico is very well positioned on their home market to go after productivity XR, better than Meta in the US, due to special characters of the (huge) Chinese market.

        Current mobile VR headsets are basically game consoles, but have never attracted enough gamers to make the market self-sustainable. Lots of theories why, with "too much friction" being the most likely reason why even gamers don't go for the more immersive medium. Game consoles are often used to relax in the living room, lounging on sofas while playing with friends. The current sweaty (local) single player VR HMD pressing against the face and requiring to clear up space, and secure pets/kids, running out of battery after two hours even if you had remembered to charge them, apparently still cannot compete.

        Which is why pretty much nobody besides Meta is currently bothering with building VR HMDs focused on gaming. Instead we now get technically similar XR HMDs not focused on immersion, where the differentiation is the resolution/quality of passthrough or the amount of RAM to run multiple productivity apps in parallel. Both things you don't want to see during an immersive game session.

        Unlike Meta, but like Apple and Google and Sony and Lenovo and others, Pico now (long term) targets XR productivity use, with games just being one, but not the most important aspect. They apparently scrapped the Pico 5 to go back to the drawing board, with the Pico 4 Ultra an interim solution almost identical to Pico 4, but improving passthrough and RAM to make it more suitable for the use and development of productivity and media XR apps.

        They are also launching their own Android store in China, and since the Google PlayStore isn't available there, can expect much higher user acceptance than Meta would see for this outside of China. Meta's Horizon OS, competing with Apple's visionOS and Google's AndroidXR, is in trouble, as it will not get access to PlayStore productivity apps, and most Android users won't accept anything else. Pico doesn't have to compete with Meta or Google in China, where the Android app store market is open. They even made a deal with Apple that allows mirroring the iPhone display to a Pico HMD, something Meta is still trying to get. So while Meta built/bought the largest VR gaming community, they are somewhat locked in there, having troubles to expand beyond it.

        Starting around a decade ago, an astonishing number of China-only HMDs with weak 3D performance unusable for gaming were released just for watching movies, partly driven by people living in tiny and/or crowded flats without large TVs. And huge streaming portals like iQIYI offered HMDs very similar to Quest 2 at low prices with subscriptions for their portals. That's basically like Netflix releasing a Quest clone at half the price just for watching Netflix, instead of for years not updating the Quest Netflix app limited to 480p and then just killing it. A significant market for HMDs focused on "media consumption" existed in China for years before AVP brought virtual cinema to the attention of the masses.

        Consoles, gaming PCs and Quest HMDs are sold by games to gamers. Desktop PCs, laptops, smartphones are not, and sell a lot more to a much wider audience that may value having access to WeChat, Weibo, Douyin or iQIYI. Everybody besides Meta is going for that market instead of gaming, and may actually succeed there (in a couple of years). Even if VR gamers believe that VR is for gaming and everything else will fail.

        • Gradius

          In that case, do you think something like the Immersed Visor will succeed because it focuses on productivity and media consumption in a smaller form factor?

          • Christian Schildwaechter

            Depends a lot on your definition of success, but in general I'd say no, mostly because of their business model based on subscriptions.

            They are basically trying to sell a virtual monitor replacement at a subsidized price, making the money back from a somewhat mandatory subscription. There will be a free tier, but it should be apparent that the company will be losing money in the beginning. Compared to current PCVR HMDs that will continue to work even if the product is discontinued (and you don't update Windows if you use WMR), the Visor would probably become useless if Immersive's runs out of money. And they are most certainly not profitable.

            They are targeting professional users, who might be fine with a subscription. I personally don't want any monitor that only works as long as a) I keep paying monthly and b) the company behind it keeps finding investors willing to give them more money. They are too small for the hardware to be really special, so most of it will be based on Qualcomm's current XR reference HMD (that includes eye tracking) plus microOLED from BOE (that Apple so far rejected due to quality issues). Hardware that everybody else will have access to too, so it is very likely that whatever Samsung releases will be able to do anything the Visor does and more. And (ignoring the USD 400 bait offer) a professional would probably rather give Samsung USD 2000 than Immersive USD 1000 (incl. life-time subscription) for extra features and being more future proof.

            It may find a special super-slim/-light niche like the Bigscreen Beyond, but Immersive is too small to really win based on hardware. They currently benefit from being the first to offer such a device, and (still) have enough money to make the offer look very attractive. But they'd need a lot more to either grow the Visor to the large numbers where the production actually becomes cheap, or to keep subsidizing it similar to Meta. Meaning they could be snuffed out anytime by a larger company deciding they want to own that market. Usually smaller companies only survive this by pricing the devices accordingly high, similar to what HTC has done in professional VR for years. Immersive tries to play the big boy game with subsidizing and making money back from other channels, but doesn't have the resources to back that up for a long time.

        • Skeltal

          Productivity XR or AR being more successful a product category than gaming VR is primarily a delusion by those in the XR industry. Just because the industry has a hard time making good VR software doesn't mean a cop out answer like "productivity" is satisfactory to the end user. Be specific as to what value there is to be had here, otherwise it's another colossal waste of time and resources like the Quest Pro, like the Vision Pro, Magic Leap, etc.

          The worst selling Quest headsets are MR headsets, btw.

          • Christian Schildwaechter

            TL;DR: Apple makes more profit from gaming than Tencent or Sony thanks to their iOS platform designed for productivity use that only does gaming on the side; if Meta would have given a USD 3500 AVP to each of the 10mn currently active Quest user for free instead of spending money on XR themselves, they would have saved about 2/3rds of their money; gaming is a niche, VR gaming is a tiny niche in that niche that only looks big or important if you look at it from inside this bubble.

            The Gear VR had horrible retention. Not because the experiences were bad, but because it occupied the user's Samsung phone, and if you wanted to answer a call or read a notification, you had to remove it from the viewer, then later put it back in. People valued always having access to their phone a lot more than the VR experience it could also provide, and quickly abandonned the viewer.

            There are 6bn smartphone active in the world, that their owners look at 60-150 times each day (depending on country), compared to less than 20mn active VR users across all platforms, mostly used for gaming. The first smartphone by todays definition was the 2007 iPhone, and we are only a few years away from the number of active mobile contracts exceeding the number of humans. It's probably a decade too early to tell if people will replace phones with (future, much lighter) HMDs, but it is easy to see why the XR industry is going for this "delusion".

            We are now 11 years (DK1)/8 years (CV1, Vive) into "consumer VR", with very slow user growth, and Quest 3 selling a fraction of what Quest 2 sold. VR gaming turned out to be a small niche, burning billions in investments, and the majority of people in that niche very apparently value it at USD 300 or less, as observed during the different price phases of Quest 2. Meanwhile a lot of people use their phones so much that paying USD 1000+ has become acceptable.

            "Useful" tech sells a lot more than niche gaming devices still troubled with a lot of comfort and usability issues. By the time the iPhone was 3.5 years old, Apple sold as many each quarter as Meta has sold HMDs since they acquired Oculus in 2014. And now they have the AVP for "spatial computing", positioned as a head worn iPad with giant screens that runs regular phone software. Which VR gamers of course decried because it lacked the controllers needed for gaming. The phone software "accidentally" made Apple the most profitable gaming company in the world. In gaming revenue they only trail behind Tencent and Sony (and now probably Microsoft-Activision), but exceeded all of them in actual earnings.

            So a company widely known for ignoring gaming won the gaming (money) market just by creating a popular productivity platform, where people occasionally also wanted to play games. Show this to any MBA, then tell them about Meta investing USD 50bn in MRL/USD 100bn total in XR, or about USD 5000/10000 per currently active Quest user, and they'll declare anybody investing into mobile VR hardware for gaming a lunatic.

          • Skeltal

            The reason why mobile gaming brings so much profit is the combination of low friction, high production value, and unregulated gambling. The XR talent is not well suited for an equivalent. In the promise of lower friction we've seen some of the most frustrating UX imaginable. In the quest for viewing the medium as raw tech and not an artform, the talent reflects that, and as such there are difficulties selling microtransactions or even building a unique IP in VR.

            Also, the monetization practices of Apple and Google are largely incompatible with XR and the quality of software people expect out of it. The ad model becomes incredibly frustrating when overriding your vision, even as a tiny banner. Omit the gambling, omit the adverts, and you have the mobile ecosystem without the profit.

            The problem with being the next computing platform is that unlike the smartphone and unlike the PC, we already have those. It's like having a novel idea for a portable all-in-one computing device, that's better than anything we have, but having the iPhone still in market going strong. The journey towards this dream will take so long that Apple and Facebook will long be out of market by the end of it.

            If the idea is to generate profit, Gaming VR is a dumb idea, but Productivity/Lifestyle is even worse, with a much less attractive track record. Many of the "wins" have user numbers botted or embellished to all hell, with any surface level peek (Decentraland, Immersed). The grassroots wins that did happen were directly cloned or shut down by platform holders (Virtual Desktop, YUR Fitness), making any competent developer unwilling to touch these platforms. This isn't even getting into the Apple policy change that was rolled back, leaving a bad taste.

            The Pico 4 Ultra isn't going to suddenly sell because it's MR, its fate is entirely in the hands of its value as a gaming platform.

          • Christian Schildwaechter

            The problem with being the next computing platform is that unlike the smartphone and unlike the PC, we already have those. It's like having a novel idea for a portable all-in-one computing device, that's better than anything we have, but having the iPhone still in market going strong.

            We have phones that can run thousands of apps everywhere as long as you hold them in one hand and everything fits on a single 6" screen you have to look down at. We don't have phones running those apps, but leaving your hands free while showing the content on a 100" display right in front of you. Or several of these displays, placed wherever you want them, relative to your body or the room. That's what AVP does, and it's a pretty novel feature for a portable all-in-one computing device that can do most of the things your smartphone or your PC can do.

            HMDs are currently heavy with still too low resolution, but I can pretty much guarantee that within a few years, the idea of pulling your phone out of your pocket and looking down on its tiny screen, or having to schlepp around a laptop just for the decent display size and keyboard, will look very quaint, when a set of high tech ski goggles allows you to turn everything into a display and interact with documents by just grabbing them in the air.

            It doesn't even have to do more than smartphones or computers. It would be sufficient if it does the same in a more convenient way. Add a simple bluetooth keyboard, and the sole disadvantage of using an XR HMD compared to phone/laptop will be that it won't fit as easily into your pockets as your phone. But since you don't have to hold it like a phone, there is no need to put it away all the time to free your hands, and instead you can just wear it all the time .

          • Arno van Wingerde

            Hm… I agree in principle, but on that basis just about every other technology would not be worth investing in "because it is not a phone and will never sell in those numbers".

            For me, it is still unclear whether users are prepared to actually do stuff on the AVP, rather than on a laptop/tablet/phone. Watching a film in an airplane: well, if it is a long flight and I am prepared to carry such a bulky/heavy device on top of everything else: then maybe. Putting up your headset to have a teams conversation: maybe. Productivity stuff: only if it offers better advantages than multiple screens and even there I would only use it if I need 3D.

            Gaming: we all know that VR gaming is inherently different from everything else. It can be nauseating, and the graphics typically suffer, compared to a desktop/console, but the experience is incomparable. I am still amazed that people do not flock to VR gaming, but you're right: interested people had the chance to try this and most people thought "nice, but no".

      • Andrew Jakobs

        Except with the Ultra they also target XR and using it as a desktopscreen replacement.

    • Arno van Wingerde

      I do not know how many "good" games have been transferred to the Pico 4, but I saw a number of the games I actually play. If the hardware is on par with the Quest3 or better, and software support is good this thing might be a hit in China, enough of a market by itself. Given that Pico laid off staff that did just that makes me wonder though…

      • ViRGiN

        There is basically no vr market in China, except for like vr arcades for one time off customers.

        the games that were ported to pico previously were far lesser and often incomplete games compared to quest, like contractors without mod support, or golf vr without dlcs. The devs i know only ported their games to pico cause pico foot the bill in the past. The userbase is far too small for anyone to bother. Picos in the western world serve as wireless pcvr essentially.

        • salamancer

          Here in Australia, we had access to Pico headsets before Quests so maybe it could see sales here

  • LP

    >Optics: 105° FOV

    Meh. When will they stop making diving masks?

    • ViRGiN

      When physics cease to exist.

      • LP

        You're don't know well HMDs that are on the market, so you're talking stupid things.

        HTC have 120 FOV.
        And that 20 degrees IS matter.
        Not to mention Pimax's experiments with extreme FOV.

        • ViRGiN

          I wonder why they no longer experiment with extreme fov then? Why pimax still hasn't captured even 1% of steamvr despite making like a dozen headsets already?

          you call 120 not a diving mask?

          • LP

            I was talking about pimax experiments, you're talking about percentages in steam. I said 20 degrees IS matter, you're asking if they matter.

            You don't even read what's written, how can you even expect to have a dialog?

          • ViRGiN

            Go block yourself.

          • LP

            I'm still of that opinion. You are inadequate, embittered and only capable of hate.
            Keep hysterically disliking, that's all you're capable of.

            What makes you think anyone would even communicate with you after something like that? https://uploads.disquscdn.com/images/21b0840116f01542e6ee0c39370533776f188d25eca6e2c6d26c9cb5361738af.png

        • Andrew Jakobs

          Oh, which headset from HTC has 120 degrees FOV? But I do agree, wider FOV is something they should work on.

          • LP

            Vive/Focus/Elite

            As far as I can remember, they always made a higher horizontal fov than the oculus . Not counting Flow or Comsos.

            The problem is more about price, cheap standalone pico and oculus will have more market penetration, even if it's 100~ degree field of view.
            And it ruins the VR experience actually. People feel uncomfortable and cramped wearing these headsets. Which is an industry problem that is underestimated.

          • Andrew Jakobs

            Focus seems to have 120, elite seems to have 110.. well, let's hope HTC's coming headset will be at least the same as the focus 3 with a pricerange below $1000. But I highly doubt it will be anything to write home about.

  • Andrey

    I wonder why Pico (or Bytedance) still continue to release their headsets, especially after layoffs in their AR/VR department. Just like in Meta's/Sony's/HTC's/even Valve's case, everything is based on software library and Pico doesn't even have those rare Meta exclusives, while those tiny spec's upgrades really doing nothing. Like, did they released EVEN JUST ONE exclusive based on something really popular (at least in China/Asia in general) for their own platform? I just love when big companies with a lot of money from other sources can't (or don't want to?) finance the creation of at least SOME AA(A) games for their own hardware to make it more popular and "sell" it to a wider audience…

    At the same time, surprisingly, they actually did something right. They released (or plan to release soon) two foot trackers that will work with their headsets for only 60$ that will not only have IMUs in them, but also IR markers for much better accuracy when trackers are in sight of HMD's cameras. You can say what you want, but I, hater of "social VR", still want my VR legs for years! I want to be able to kick doors in shooters, or zombies to ward them off, or just enemies in something like Blade and Sorcery or Battle Talent. New doors for full-fledged dancing games, soccer and so on will also be opened. Imagine how much cooler new Batman game would be if it included support for such kind of trackers and leg kicks! Still can't understand why Meta won't release something simple as that for those who need it, especially being the leader in the industry. Lately they outsource a lot of *&^% to others ("Made for Meta"), but rotating chairs and vests are one thing, much more niche, while leg tracking is still missed yet essential input method. Ironicly, once again, almost everyone already did it (Sony's Mocopi, HTC's VIVR [Ultimate] Trackers and now Pico too), but Meta is still cooking. Hope to see it being availible with Quest 4 at the very least.

    • ViRGiN

      I dont think Meta outsourced anything, its more of a certification that the device is compatibible with Meta headsets, and likely won't be completly bricked with future system update, perhaps they will also include support for these haptics, chairs in SDK so actual developer support might become a reality.

      We are yet to see how well the leg tracking will work. We saw optimism with the htc inside out trackers, and when it did came out I only briefly heard that they are nowhere near as good as something like Quest Pro controllers, which on it's own, seems to be not that good of a product (tho i love build quality). And still, putting these on, charging is really only for the vrchat weebs. I envision some sort of a webcam, kinda like CV1 sensor on a stand, that you put somewhere in your room and with its wide fov stereo lens, it sees your room in 3D and will track your body. Meta is expert in AI, and this seems like an AI job, not sensor work and occasional IR tracking when in range. I never enjoyed any of these trackers, even expensive motion cap suits are really bad, and for good results, not in real time, the data after the capture has to be filtered.
      Even with HTC trackers with lighthouse, the result is far too basic, and it's been around for years. You know the type of users who call this "full body tracking" when it's not tracking body at all.

      This could also serve as sort of holodeck teleconference, as it could capture your room and allow people to visit other people houses. Imagine removing the wall, it would be like seeing your neighbors.

      They could also go step ahead and make it mount on the ceiling and capture 360 view of the room so you can invite others to your place. I think for wide adoption these things must be really functional. When not in use it could double as home security.

      Another problem with all these trackers is that they will naturally shift on your body, making aligning very tedious. AI camera solves this issue completly.

      Leg tracking would make great content for commercials, but then again it would look like they are leaning into more of Kinect while everyone is waiting for "real" games (Batman appears to be great).

      • Andrey

        Man, if after reading this comment someone will say that you are a “troll” that can’t normally participate in the discussion – they are insane. Though all of them probably already have blocked you, so they won’t be able to read it and see how wrong they are, lol. You have my respect!
        About “outsourcing” – sorry, it was my mistake. I meant that they, instead of R&D and then produce/sell their own Meta branded accessories decided to give it up to everyone who wants to do it. It is indeed not a real outsourcing, more like just “we will let you sell it with “Made for Meta” sticker if your product will meet our standards”.
        I agree with most of your points and, personally, won’t care how exactly my “body” (as you correctly mentioned, legs only) will be tracked. Accuracy indeed is much more important, so as the price and if one (or even two/three and so on) cameras will cost less or at the level of one Vive tracker, I am all for it! And if it will also deliver some other cool features like mentioned teleconference (imagine this + Meta’s upcoming Codec avatars together…) that some people actually asking for on Meta’s forums and Reddit, it will expand the usability of the device even more.
        But in reality I don’t think Meta will do something like this in general. After closing Ready at Dawn and finally officially “canacelling” GTA SA VR… I think their whole paradigm about VR and, especially, AAA VR games has drastically changed. And so, if they are not adamant on bringing more software to the platform anymore, imo there is no real chance that they will add new hardware. I really hope they will, but I already hoped for years for Lone Echo 3 and how I will ride in my lowrider car back to the Groove Street with Big Smoke, Rider and Sweet sitting next to me – and look where my hopes brought us…

        • LP

          >Though all of them probably already have blocked you

          And they were right.

          • ViRGiN

            You came back. You're obsessed.

        • Somerandomindividual

          The amazing thing is that you waste such time writing such long responses to people you know are trolls. Just block them and move on!

    • Zack71

      I quote every single word!
      I would never buy a Pico 4, because there will never be a game optimized for its hardware, as was the case with the Pico 3: anyone remember "Just Dance", which was supposed to be a Bytedance exclusive?
      Let's hope Meta takes the hint, leg trackers would be very useful not only for fitness apps like FitXR or Les Mills Bodycombat, but could bring standalone VR to a new level of immersion!

    • Andrew Jakobs

      I like my Pico 4 as a standalone as my PC is still only a RTX2060super and I have an old router not even capable of Wifi6. Pico already has had leg trackers for a while, I bought those on aliexpress at a sale for $25.
      once I have a new PC/router, hopefully Q1 2025, I'll certainly keep the Pico 4 Ultra in mind, even if it's only for development purposes.

      • Andrey

        As always, you are free to like and use whatever you want! It’s just that I personally can’t see even a single reason to buy Pico over any Quest headset. At the same time if we are talking about disadvantages, then I can name a few (from no “only on Pico” exclusives and, especially, no support for Meta exclusives to it’s price – it was/is not cheaper than Quest 2/3, at least not THAT cheaper for it to make a significant difference; plus I heard that Pico’s pancake lenses are worse than Meta’s in clarity, though I never used a Pico headset with this kind of lenses, so for me it’s just a rumor, but I tend to believe that it may be true). If we are talking about standalone headset in 2024 and someone would ask for my recommendation, I would immediately answer: “Buy Q3 or, if you are on a budget, wait for Q3S”. If this person was a any kind of simulator enjoyer, I may recommend Pimax (though I would mention about all of it’s disadvantages), but I would never recommend anyone to buy a Pico, because, imo, there isn’t a single point in doing so.

        • NL_VR

          i know a couple of people that bought Pico 4 because it was cheaper than Quest (at least at the time) and you can use it wireless for PCVR.
          They were not interested in the standalone library but i think all of them got some standalone games also.

      • You're using wind-powered equipment.
        You're not allowed to have an opinion.

  • eadVrim

    What I don't like about the Quest 3 is the grainy low quality pass-through and instability of the objects in mixed reality, where they keep moving a few centimeters. I don't know if this is the case for the Pico 4 Ultra.
    In terms of the revealed specifications of the Q3S it seems closer to the Quest2S or Quest 2 MR than a Quest 3.

    • Andrew Jakobs

      The Q3 seemingly has 2 4MP camera's, and the Apple Vision Pro 2 6.5MP camera's, where the Pico 4 has a single 16MP camera, and the Pico 4 Ultra has 2 32MP camera's. So one might think that the passthrough on the ultra should be freaking sharp.

      • Christian Schildwaechter

        The Pico 4 Ultra will use 2×2 pixel binning, combining four sensor pixels into one actual pixel, thereby reducing the effective resolution to 8MP per eye. Which is still more than enough for the 4.7MP per eye resolution of the HMD, helps to significantly reduce sensor noise by averaging the four pixels, and doesn't waste compute on excess resolution not actually needed. The sharpness of the passthrough will still be more limited by the display than the cameras.

        Qualcomm Snapdragons contain a Spectra ISP (Image Signal Processor) that is probably also included in the XR2 and so far hasn't been used a lot there. Having it pre-process the hires sensor image to get to a lower/more sensible resolution with improved stability that will make processing the image easier at basically no extra compute cost would be a good use case. Similar to Meta delegating all the image analysis for room and hand tracking to the Hexagon DSP (Digital Signal Processor) on Quest, keeping CPU and GPU free for user applications and OS.

    • ViRGiN

      Name two things Pico really did surpass Quest headsets, other than raw hardware specs.

      • eadVrim

        I didn't try a Pico headset to judge, about the Q3 I like its software (library, games, improvements) even somtime this device finish in the trash if it is overheated or an update goes wrong.

      • kebo

        Pico 4 owner here (PC). It has the battery at the back so you only have 300g at the front which makes it feel super light. Even the Quest 3 seems quite heavy and uncomftable to me.

        I really hope all future headsets will do that. It's such an easy thing with massive impact.

        • ViRGiN

          But you can’t replace headstrap. Battery doesn’t weight much. I don’t have comfort issues with q3 with elite strap.

          So, not really a better feature.

      • Carlos

        They have a great IPD range. I had a Quest 2 and I hated it (regreted upgrading my Q1) because the poor 3 IPD setups were terrible to me.

  • Foreign Devil

    Isn't the XR3 chip already available? Why didnt' they use that?

    • Andrew Jakobs

      Because it is probably much more expensive? The XR2 gen2 is only just made available to manufacturers other then Meta.

    • Christian Schildwaechter

      The SD8 Gen 3 is available, but no XR2 Gen 3 ("XR3") based on it. The XR1 (smart glasses)/XR2 (VR/MR/XR) lines come with a core configuration optimized for sustained loads instead of the typical burst/idle use on phones. And we'll probably only see the next XR2 generation 2026 in a Quest 4, then based on SD8 Gen 5.

      The Pico 4 Ultra uses the current top XR2 SoC that became generally available in April after half a year of exclusivity for Meta. Pico added more active cooling to the Pico 4 compared to Quest, allowing to run the SoC faster. Meta runs esp. the CPU cores a lot slower than the chip would allow to gain thermal headroom for the GPU and prevent throttling. On Quest 3, the CPU performance cores run (AFAIR) at 1.4-2.05GHz, depending on app. The exact XR2 specs aren't public, but the same cores run at 2.8GHz on the equivalent SD8 Gen 2.

      Combined with extra RAM, the faster clock makes Pico HMDs technically faster than the matching Quest, but as developers optimize mostly for Quest, users won't see much if any performance gains. More expensive XR HMDs from Samsung and others will use the XR2+ Gen 2 expected this Fall. Qualcomm usually releases a Plus version of their top SoC about a year after release, using the same chip design but roughly 10% faster due to process improvements with slightly higher clock rates.

  • David Cano

    Thanks Pico for existing and save us from using meta.

  • xyzs

    "Ultra" but no eye tracking…

    Pass.

    What's needed is a big tech OS (HorizonOS or Android XR), and eye tracking optimization if they want to steal shares from Meta's hardware.