Facebook CEO Mark Zuckerberg spoke about the idea of a ‘Quest Pro’ headset back in May, saying that it could include sensors for both face and eye tracking. The v32 firmware release for Oculus Quest last month ostensibly confirms that the company is moving forward with the device, as it contains hidden reference to an “Oculus Quest Pro” headset along with text mentioning both eye and face-tracking.

Reddit user ‘Reggy04’ did some deep-dive sleuthing and has posted a number of strings found within the Quest v32 firmware which not only includes Quest Pro by name, but also notes a few features that would mean a dramatic increase in the number of sensors embedded within the alleged headset.

Sensors are said to improve hand tracking, something that feels a bit like a hit and miss on current-day Quests.

“QUEST PRO ESTIMATES YOUR HAND SIZE AND HOW THEY MOVE SO YOU USE YOUR HANDS INSTEAD OF CONTROLLERS IN VR.\””

Eye-tracking sensors, which can be used for a number of tasks like foveated rendering and smarter UI navigation, are also supposedly onboard. This string of text below refers to an eye-tracking setup which, like many such calibration tests, requires you to look and hold your gaze on virtual objects so the system can correctly model your unique eye movement.

“\”FOLLOW THE TARGET WITH YOUR GAZE\””
“\”EYE CALIBRATION FAILED\””

“\”EYE MOVEMENT ESTIMATION COMPLETE\””

Mentions of facial tracking, which is important to social VR interactions, are also found in the v32 firmware. It seems a calibration test is needed here too to get a good bead on how you smile, frown, bare your teeth in anger, and act surprised.

“COPY THE DIFFERENT EXPRESSIONS\””

“\”SMILE NATURALLY UNTIL THE CALIBRATION IS COMPLETE. YOU MAY NEED TO HOLD THIS POSE FOR A FEW SECONDS.\””

“\”SHOW AN ANGRY EXPRESSION UNTIL THE CALIBRATION IS COMPLETE. YOU MAY NEED TO HOLD THIS POSE FOR A FEW SECONDS.\””

“\”FROWN NATURALLY UNTIL THE CALIBRATION IS COMPLETE. YOU MAY NEED TO HOLD THIS POSE FOR A FEW SECONDS.\””

“\”SHOW A SURPRISED EXPRESSION UNTIL THE CALIBRATION IS COMPLETE. YOU MAY NEED TO HOLD THIS POSE FOR A FEW SECONDS.\””

“\”FAILED TO CALIBRATE THE EXPRESSION. RETRYING…\””

“\”FACE MOVEMENT ESTIMATION COMPLETE\””

SEE ALSO
'Song in the Smoke' Now Releasing "a few weeks later" Than Planned

Further in the firmware, Reggy04 also find mention of sliding lenses, which could denote that Quest Pro will include some form of interpupillary distance (IPD) hardware adjustment that, much like the original Oculus Rift, will require you to move the lenses to get a clear picture.

“\”SLIDE THE LENSES CLOSER TOGETHER OR FURTHER APART TO IMPROVE VISUAL CLARITY.\””

There’s mention of depth adjustment too, which sounds similar to how it’s done in Rift S. Reggy04 notes that the string below was specifically added in v32, so it’s possible it’s also referring to Oculus Quest Pro too.

“\”PRESS AND HOLD THE DEPTH BUTTON AND MOVE THE HEADSET CLOSER OR FURTHER AWAY FROM YOUR FACE. THE LENSES SHOULD BE CLOSE TO YOUR EYES, BUT NOT CAUSE DISCOMFORT.\””

“\”ADJUST LENS DEPTH\””

“\”TURN THE WHEEL LEFT OR RIGHT TO ADJUST TIGHTNESS. THE FRONT PADDING SHOULD FIT DIRECTLY OVER YOUR FOREHEAD.\””

An all-digital version of Facebook Connect, the company’s AR/VR developer conference, is coming October 28th. We’re hoping to learn more then since Connect has been a historic venue for insights into the company’s research.

We’re also hoping to learn more about the company’s Project Aria AR glasses, future plans for its new Ray-Ban Stories camera glasses, and how experimental pass-through AR capabilities are coming on Quest 2.

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. See here for more information.


  • xyzs

    I hope they release it soon.

    VR hardware activity is so sparse… it’s really boring and frustrating.

    If smartphones improvement were the same speed, we would have the iPhone 4 as the super flagship in 2021…

    • MasterElwood

      Are you kidding me?
      Quest 2 vs 1:

      120hz vs 72hz
      RAM +50%
      Resolution +50%
      and a waaaaaaay faster SOC

      in just 18 month? That’s not fast?

      • One every Christmas would be nice.

      • xyzs

        Loss of custom IPD, loss of full black level, crapy plastic build finish, sound less good, still same fresnel lenses, still same tracking accuracy, no improvement in FoV, no improvement in comfort, no improvement in battery life, yeah I confirm what I said: Evolution suck in VR even if they put a better chip in it.

        Not to mention that the core tech didn’t evolved since the very first CV1 model or even since the dk1….. more than 8 years ago. It actually had a better field of view and better no Fresnel artefacts lenses.

        • kebo

          Less weight = more confort, bateries of controllers work longer, less to no SDE, better sound (in my experience).. actually the SDE in Quest 1 was an absolute joke and for me the single reason why that headset is a failure.

          What do you mean with “core tech didn’t evolved since the very first CV1”? That thing had light sensor tracking with 2-3 base stations and now we have best possible inside out tracking. Plus much better pixel density and Hz. So core tech (tracking and display) did evolve. Not to mention the missing cable.

        • david vincent

          Non-fresnel lenses are not necessarily better since they go with A LOT of eye-position dependent distortion, which can cause discomfort and motion sickness.
          Between the fresnel lenses artefacts (which can be reduced) and the higher risk of motion sickness, the manufacturers made their choice.

          • Alergic to Soy

            PSVR has non fresnel lenses and they AWESOME.
            Frensel is trash and shouldn’t be used in VR

          • david vincent

            Ok kiddo. I guess you don’t understand what is eye-position dependant distortion.

          • Alergic to Soy

            It works fine for me grandpa, so i like it better.

      • Jistuce

        I mean, I remember when we joked that the rapid advancement in processors and RAM prices meant a computer was obsolete the moment it hit store shelves. To say nothing of the 3D accelerator wars.

        Eighteen months feels very slow sometimes, but I’m honestly glad we haven’t leapt back into monthly hardware advances.

        That said, it’d be nice to see less evolution and more revolution.

      • Alergic to Soy

        You are wrong and just proved his point.
        The SOC is off the shelf part and not even the newest of the release year like gaming phones got but older.
        So this kills all the RAM/CPU/GPU/WiFi discusion and just leaves the screen
        The thing is, the screen is also off the shelf and nothing new, its not a OLED [and i mean REAL RGB OLED like in PSVR] its not using any of the VR dedicated micro screens that got released in last 4 years , we saw all of them here on this web site a coin sized 4K OLED screens for VR or somethig else in the middle, doesnt have to be 4K right now.

        Quest 2 is lazy update, they did ZERO work on it,

    • Alexander Sears

      Progress has been slow for VR. Frankly, the difference between recent Vive’s new Pro 2 and the first Vive Pro is unremarkable. However, I’m not 100% sold on the exact comparability of an established market such as cell phones with that of a nascent industry such as VR. Businesses are conservative organisations, I reckon once they see potential, if any (I hope there is), in VR, then we’ll see more significant year on year improvements.

    • Rolotor

      It’s not that surprising, really. Touch-screen smartphones have been around since 2006. VR on the other hand is a relatively new technology. Obviously, there’s going to be a lot of experimentation involved and that will slow down releases considerably.

  • I honestly don’t give a **** about more stuff that’s mainly just for “social” features, truth be told, and would rather they focus on things like improving the contrast, enlarging the sweet spot, increasing the field of view, and reducing the weight considerably. Of course, if the eye and face tracking comes alongside those things then that’s all good.

    • Sven Viking

      If their eye tracking allows for foveated rendering, that’d be useful for increasing FOV, resolution and graphics quality with realistic processing requirements.

      • Ad

        Even Carmack said that would have less than 50% benefit. A fan could accomplish more.

        • Sven Viking

          So add a fan as well and get >100% benefit. The benefit increases with increased FOV and resolution though.

          • Rogue Transfer

            The overriding problem is the delay between detection of where the eye is looking and the added time it takes for that to be sent and an update to happen to the rendering. This, according to John Carmack causes significant blurring, which limits what can be achieved.

            The smaller the region or the more you move it, the more noticeble the problem becomes. So, increasing the FOV will exacerbate the issue with dynamic foveated rendering using eye-tracking.

            There’s also added power and performance needed to do continual, fast eye tracking. That drains the battery more puts more limits on what’s possible in a standalone device. Plus, extra heat which on a standalone headset can quickly add up. Hence, why Mark Zuckerberg repeated warned about not wanting to ‘burn your face off’ adding more sensors/eye tracking. His words.

            Adding extra resolution also increases battery drain significantly, just like with any mobile display, it generates more heat too with the additional higher speed switching circuitry to cope with the extra pixels.

            We also need to remember that the figures often given are in the ideal case scenario, which nearly all software & scenes don’t allow for. This means, the benefits in actual games will be much lower than those touted by researchers or with test software.

          • Sven Viking

            You make good points but things like increased resolution are going to happen either way, so may as well have something to help with it. Also a lot of the eye tracking processing is likely to be done on dedicated chips rather than the main CPU/GPU.

            Increased FOV shouldn’t exacerbate the problem of noticing the foveal area since it’s not relative angular size/movement but absolute angular size/movement that’s the problem, and although a larger FOV would allow for larger maximum movements, large eye movements trigger a significant period of saccadic masking and allow the stop position to be predicted ahead during the deceleration of the eye, making them actually easier to deal with than smaller movements. (You do need a method of displaying the low-res area that doesn’t result in flickery changes in the periphery but it sounds like some existing solutions are working pretty well already.)

            I’ve seen quite a few people report not being able to notice the updates with existing eye tracked foveated rendering systems. Likely that is because the foveal region is pretty large, but it’ll be able to be shrunk down as the tech improves.

        • Blaexe

          That makes no sense. There’s already a fan in Quest and the GPU is already running at near maximum clock.

          • Rogue Transfer

            In dynamic environments or those with more materials and draw calls, CPU is very important. Most games can’t utilise the GPU fully with a slower CPU, because most scenes, assets and animated characters aren’t designed for minimal draw calls or material switching.

            This is why the Quest 2 having up to half speed CPU(main core) causes nearly half the performance results with benchmark software compared to the Vive Focus 3, which has a stronger fan cooling system allowing for a faster CPU speed than the Quest 2.

          • Blaexe

            Of course CPU performance is important. So is GPU performance. Foveated Rendering would only impact the GPU headroom though and I doubt higher CPU performance (which you could get out of a bigger cooling system) would enable completely different experiences.

            HTC “buys” the higher performance with a noisy fan, external battery and needs to combat the additional weight with expensive materials. facebook just decided to go for different tradeoffs (slower CPU clock but therefore silent fan, lower weight without using exensive materials or designs).

          • Rogue Transfer

            If we consider a little more, higher CPU allows for more processing all round, that allows for more predictive calculations & sensor data interpretation.

            One aspect missed about dynamic foveated rendering, is it is highly dependent on processor intensive, fast eye image tracking. Hence, it requires a significant CPU use as well as GPU.

            Thus, higher CPU would be very beneficial, all round.

          • Blaexe

            It “could”, not it “would”. As we can assume now, Quest Pro will have various new sensors that need the CPU performance but won’t necessarily provide better graphics or improvements in games.

            Dynamic Foveated Rendering is still a complete wildcard.

            If the dual cell display with the same resolution turns out to be true, the Foveated Rendering is imo unlikely as it benefits more the higher the resolution and FoV is.

          • Ad

            That’s not true at all. He said the opposite. The Vive Focus 3 has added cooling and it has a big bump in performance with the same chip.

          • Blaexe

            There’s a bump in CPU performance. The GPU performance is basically the same. Both what I said is true, do your research.

          • Ad

            This is avoiding the main point which is that the entire idea of DFR in the short term being this massive windfall of performance is completely unrealistic.

          • Blaexe

            I was responding to “A Fan could accomplish more.”

          • Ad

            The fan was a baseline for something simple. A good upscaler or obviously DFR would need more CPU power anyway.

      • But it probably won’t (at least not initially) and will likely instead just give you more “social” features.

        • Sven Viking

          That’s very possible considering Zuckerberg’s focus on social features and ad-relevant user data. I’m personally guessing they’ll also have it working for foveated rendering, though. Hopefully we’ll find out one way or the other in a couple of weeks.

    • Holdup

      Fb is all about social, unfortunately, it would be cool tho to atleast get a oled display with slightly better res or fov or even both

    • Arno van Wingerde

      But maybe that is not what this for! how about VR videoconferencing, sitting at the same table, seeing the expressions of you colleagues/customers, without flying to another continent first! That would also explain the “pro” moniker. Sure, no direct use for gamers, but maybe great for VR as a whole.

  • Ad

    They’re going to advertise like they’ve never advertised before.

    • ViRGiN

      Just like you’re pushing all the steam shovelware indie shits on Reddit

  • 3872Orcs

    I hope this marks the shift to eye tracking in all consumer headsets from now on. While I would like more FOV I think we need to get eye tracking on board first for all its benefits; mainly performance and for faster and more accurate input while navigating and selecting stuff in VR. Hopefully with eye tracking I can have my classic strategy games in VR, the genre I miss the most from pancake gaming and where most of my gaming hours are spent.

    That said I’ll never again buy anything Facebook. It ended with the Quest 1 for me and deleting my FB account.

  • Rupert Jung
    • Sven Viking

      Eye tracking is necessary for varifocal so if anything this news makes it more possible. Something not being leaked in a firmware reference doesn’t mean they’re not doing it — also these references are mostly related to UI messages and users probably wouldn’t need to calibrate varifocal.

      I still think the chances of varifocal may not be high for this headset though.

  • LiquidKaos

    So, should I get a Quest 2 for Xmas, or wait until the Pro is released? How long until the Quest 2 is too slow/old to dance with?

  • DeanVega

    Damn, eye-tracking that can tell everything about a person when correlated with similar data? Yeah, are you all sure about Facebook knowing where you look when you’re visiting VR porn sites like this or private stuff or… you know?
    It’s kinda scary, can’t be the only one to think about this.

    Or maybe work interactions in new virtual company environments or whatever there is to come. Imagine Facebook knowing where you look at any moment… ew

  • johann jensson

    The only reason i’d be interested in eye tracking is if we could finally have dynamic foveated rendering, so we can up the framerate while at the same time having better visuals in games.