StarVR is an in-development VR headset from Starbreeze Studios which revolves around a massive 210 degree field of view; after looking inside their latest prototype, you may not want to go back.

Technically, you could say that first time we saw this headset was back in 2013, when a small French team was designing it under the name InfinitEye. The company went on to be acquired by Starbreeze Studios and debuted its new StarVR identity at E3 2015.

starvr headset e3 2016 starbreeze (6)

A year later, Starbreeze is showing the latest prototype, and you literally have to see it to believe it. The 210 degree field of view is an eye opener in the virtual world and makes other headsets feel like binoculars. In fact, when demoing the headset, the company has a button they can press to bring in a silhouette which approximates the field of view of ~100 degree headsets like the Rift and Vive; the difference is obvious and astounding.

When we saw the 2015 StarVR prototype, we liked the ultra wide field of view, but the resultant visual artifacts were not worth sacrificing the clarity offered by other major headsets (‘light ray’ artifacts withstanding). But with improvements to the lenses and rendering made over the last year, the company is starting to make that trade-off look rather appealing. If the company can continue to improve at this rate, StarVR could be in hot demand in the near future.

Hands-on: Latest StarVR Upgrades Highlight Ultra-wide FoV & Nearly Invisible Pixels

starvr headset e3 2016 starbreeze (3)

No one is arguing that a wider field of view isn’t better. The view inside of StarVR is truly impressive; you feel like you’re putting your head into the virtual world in a much more significant way than other headsets, but when it comes to optics, it’s always a balancing act. Yes, you’ve got a massive field of view, but achieving it requires a complex set of lenses which, in this case, combine normal and Fresnel elements which bring their own unique challenges to clarity.

starvr-headset-e3-2016-starbreeze-(9)And while there’s the unavoidable constraint of optical physics (i.e. you may never be able to make a photon emitted from one point in space look exactly as if it was emitted from another), there’s still an optimal performance for any optical design. And while the view through StarVR isn’t perfect, Starbreeze has demonstrated clear improvements since the last iteration of their headset, and the hope is that it’ll only continue to improve.

The optical engineering isn’t the only thing that’s improved. The current prototype is markedly lighter (now at 380 grams) than the former and seems to fit to the face better. Starbreeze has removed the bulkier strut mount from the older version and replaced it with flexible straps like those seen on the HTC Vive.

starvr headset e3 2016 starbreeze (5)

The massive per-eye resolution of 2560×1440 is greater than its competitors, but those extra pixels are stretched over the wider field of view (which is why the industry needs to start talking about pixels-per-degree). Without being able to do a side-by-side test, the best I can say is that StarVR’s PPD feels similar to the Rift and the Vive.

Acer to Invest $9 Million in Starbreeze to Help Manufacture StarVR Headset

What the headset currently lacks next to those headsets is low-persistence; a display technique which significantly reduces blur during head movement. Low-persistence requires an OLED-based display which Starbreeze says they’re working on and expect to have for the final headset.

starvr prototype (4)
See Also: StarVR Prototype Mini-Museum at E3 Shows Glimpse of Development History

While the view through the headset is becoming increasingly impressive, thanks to that monster field of view, Starbreeze still has a long way ahead of them to reach the head-tracking accuracy and precision of major consumer headsets now on the market. While the prior headset had fiducial markers printed on it for optical tracking, the current prototype is using Sixense STEM’s magnetic tracking. The company tells us that they’re continuing to explore tracking solutions.

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.


  1. Good stuff, I’d like to see them move on to low persistence OLED’s, also if they moved to DP1.4 they could probably send the signal through a single wire… though two wires could potentially allow for a card per eye…

    • Only if they can make the lenses better, in my opinion. The lenses in the Vive really are terrible, we’re overlooking it at the moment because of how awesome the experience is otherwise. But I can’t stomach the thought of the visual distortion/blurriness/fresnel artifacts getting WORSE because of the difficulties of a larger field of view.

      Actually, I think that if I had to choose between greater FOV, greater pixel density, or better lenses, I’d choose better lenses for the next gen (I sure hope I can get two out of three though!). I want DK2 level clarity and color reproduction but Vive level FOV, and I want better focus outside of the narrow central “sweet spot” than both of them.

  2. Awesome article! Can’t wait!

    What is more important for our next HMD upgrade, FOV upgrade like this set? Or a PPI upgrade(pixels per inch), which the Star is close to the current Vive/Rift and I see as a limitation. Would a 4K 100 FOV be better for consumers?

        • 720p projected already has far less pixel structure than an LCD/OLED 1080p screens, so if it would be possible to use 1080p projection, it might be just good enough.. Let’s face it, current GPU’s aren’t up to the task of even pumping out dual FullHD for VR at affordable prices..

          • That’s only if you want games playing with everything at max. Not all games need real life visuals. People keep talking about this 4K 90fps benchmark but it’s perfectly feasible with graphics that aren’t as demanding. It’s all dependent on the game being played.

        • Have you tried the consumer Rift or Vive? With their small distance between pixels, the pixel structure is barely visible at 1200p. With the same technology but at 1440p, it seems possible to me that the pixel structure will no longer be visible.

      • That will, I think, be very immersion breaking, because it will look like there is something to see on the periphery when you are looking straight ahead, but as soon as you take the natural motion of looking sideways you’ll see nothing but blur.

        I would generally say that greater FOV is what I want, however, it’s really hard to say without having experienced either greater FOV or greater pixel density. I do notice that the lack of pixel density makes distant objects look less “real” and to some degree destroys the illusion that the space is large. I don’t know if greater pixel density would actually help that or if it’s something else, maybe the convergence/accomodation mismatch.

      • I think we need to get every parameter to the point of “good enough” before we start focusing on “even better”. Right now, resolution is “good enough”, but the FOV isn’t quite good enough, and the comfort and lens quality (“god rays”) are far from good enough. I think the correct priority is: 1.) Comfort, 2.) FOV, 3.) Lens quality, 4.) Pixel density, 5.) Everything else

        • ” I think the correct priority is: 1.) Comfort, 2.) FOV, 3.) Lens quality, 4.) Pixel density, 5.) Everything else”

          That’s about right although I would argue No. 3 and No. 4 are of equal importance.

          • Yeah, the order of 3 and 4 are debatable. In my opinion, the Rift lenses are pretty bad and definitely need fixing soon, while the Vive lenses aren’t so bad except in rare circumstances (I’ve tried both extensively).

          • I think you’ve never tried DK2 lenses; if you had, you’d probably notice how bad the Vive lenses are. I never thought I’d say this, but the DK2 lenses were pretty good (in comparison to what we have now). The color, contrast, and lack of fresnel reflections is a big advantange to DK2 style lenses.

            I played Lucky’s Tale all the way through on my DK2 when it came out. Tried it later on the Vive and couldn’t believe how much worse the colors and glare were. It was to the point that I went from “I’m so happy that revive exists so that I can play my Oculus Home games on my Vive” to “meh I don’t even want to play them anymore because the difference in visual quality depresses me”.

          • I had a DK2 for a year. I agree that the DK2 lenses are much better than the Vive lenses. And the Vive lenses are better than the Oculus lenses. I just think “red mark reduction” and FOV are more important than improving on this. Maybe I’m wrong though – I haven’t done a side-by-side comparison of the DK2 and the Vive or Rift. But the red marks really bother me. And the FOV limitation is very noticeable and does not generate as much presence as it should.

          • How is it better? When tried the rift was clearer visuals and they both had artifacts just different. Personally I prefer my friends Rift over my Vive except they don’t have controls yet.

          • The Rift’s “god rays” are much more noticeable than the Vive’s, and the very-non-overlapping FOV causes some people (myself included) to have a very noticeable and uncomfortable black area between their eyes (regardless of IPD setting). I bought a Rift, noticed these problems and was very bothered by them, then tried the Vive at the store – the Vive had no issue with a black space, and the “god rays” were much less pronounced. Now I have a Vive and I’m keeping it.

          • “X lenses are much better than the Y lenses”
            Don’t think your situation is the norm.. For people not bothered by god rays, Rift lenses are much better (other headsets are very blurry in the periphery).

          • “For people not bothered by god rays, Rift lenses are much better”
            It’s not just the god rays – like I said before, depending on facial structure, some people (including myself) see a black gap between the eyes / black bars on the edges, no matter what adjustments are done. Vive does not have this problem for me.


          • For you, sure. Other people are not bothered by this and appreciate the better comfort and clearer image of the Rift.

          • That Black Gap really isn’t good enough, I guess they did that in order have a wider FOV.. they just to create wider ratio screens. FOV was one of the Selling points of the Kickstarter Oculus and for them to fall short on it isn’t good enough in my opinion.
            FOV is crucial to creating the Illusion that they are literally in another space.

          • That red mark reduction will be difficult. The headsets, if worn for a long time are causing low level pressure sores and blanching due to them being fixed in one position to your face for a long time.

            I can see there being issues where those that play for very long periods of time frequently could start getting higher grade levels of pressure sores due to the same positioning on the face of these masks.

            Look up pressure sores or pressure ulcers to see the mess they can make of the human body.

          • Damn, I wish I hadn’t looked that up…. OK I’m rooting for actual comfort to be number 1, 2 and 3 on that list. Weight reduction would go a long way so that you could wear them more like glasses perhaps but that’s surely 20 years off. And by then we might be moving into implants instead.

          • Yes, they’re pretty horrific. I used to work in a hospital and they’re real easy to get when pressure is applied to the skin for a long time and it’s one of the main concerns of patients in a hospital, especially for those that are immobile and can’t move. They have to be turned and moved regularly. Repeated pressure will start the skin literally rotting away due to lack of circulation and the real severe pressure sores can happen in days – it’s not something that happens over months or years.

            It wouldn’t surprise me to see people wearing these for a long time with quite bad red rings on their face.

          • Well, comfort is good enough for me, and for literally *everyone* I have ever put in my Vive so … not sure I’d put comfort at 1. I’d put it in with “everything else” at this point.

          • Well I didn’t mean “comfort while wearing” so much as “ability to wear it comfortably and securely without getting crazy red marks all around your face that don’t go away for hours and probably cause permanent skin damage”. Search the Vive forums and you’ll find all kinds of crazy ideas people came up with trying to fix this. I actually made my own weight-redistributor based off someone else’s idea and it works pretty well, but it’s kind of annoying.

          • As you’ve highlighted, user experience is extremely important and is priority number one when you are dealing with HMD’s. You could theoretically have the most advanced VR headset with superb resolution, fov, lenses but if this product is not comfortable to wear at all then it all goes down the drain.

          • True. And I’d pay $200 more for a much-more-comfortable / much-less-marking Vive that’s otherwise identical. I’m not rich, but I think it would be worth it just for that.

          • >> without getting crazy red marks all around your face that don’t go away for hours

            you almost certainly have your headstrap set way too tight.

          • Only reason everyone says the VIVE is better is because of the Brightness levels, which has an effect on the level of immersion. But the Oculus has more Clarity and is certainty more comfortable, it’s just a shame they couldn’t accommodate glasses slightly better.

            The Oculus Strap Solution is lot better and doesn’t require the tightness of the older head strap model. Proof the VIVE was rushed through development.

          • Depends on your definition of “good enough”. What I meant was that it’s good enough that the lack of detail / SDE can be ignored with minimal effort. Note that aliasing is NOT the result of low screen resolution but low RENDER resolution, which does not require a better screen to improve.

          • My definition of good-enough is, not getting distracted by the big pixels and screen door effect.
            For a 100 degree FOV
            2k x 2k per eye is acceptable

            4k x 4k per eye is “good-enough”
            8k x 8k per eye is great
            I’ve built my own HMD because existing ones are crap. They compromise resolution way too much, on the altair of FOV.
            I have 1280×1440 per eye, but I use high-end achromatic optics that don’t require CA correction, and they have a longer focal distance, so they are not just showing the center 50% of pixels, on an already low-res creen, like Oculus and Vive.
            I’m sorry, 1080×1200 per eye, when you are only actually seeing 50-60% of those pixels, then you are applying lens-correction, is:

          • DUH, on AA…
            I’m a VR / 3D graphics expert, BTW…

            Have been doing VR since the late nineties, built my own HMDs, wrote my own 3D modeling / animation / rendering / CAD software, my own game engines etc…
            And I’m sick of seeing pixels / screen-door in VR…
            It ruins the illusion for me.
            Maybe, I have higher standards, because I’ve been doing this stuff for a living…

          • Nice. Well I guess different people have different priorities. Obviously I would love more resolution. I’m just bothered more by the other things I mentioned.

    • Better GPU’s at affordable prices.. 4K is nice, but if it requires a GPU of at least $700+ than it’s just doomed, and only available for highend users.. First make sure the displays themselves are better (there are other ways to get rid of the screendoor effect, without having to really up the resolution, or using DLP to actually project the image directly into the eyes has shown that even at 720p the screendoor effect is already lower.. Also make sure tracking is better, and the headsets are lighter/more comfortable.. Also better optics… and if that’s done, by that time the GPU’s might have advanced to such a degree that 4K VR is available at $200-$300, and only THEN start upgrading the resolution..

        • oh comeone, you call $600-$800 ‘insanely low’? you must be f-ing insane… At least a GPU like the 1080 is needed to drive 4K at 90fps (minimum), and it’ll take a few years before that one is around the price of $200.. because less then $200 is what is needed for it to become mainstream..

          • There is no such thing as 4k VR at the moment so there’s that. Also the current resolution is fine for now. You’re just a senseless pixelwhore no one should care about.

            The most important thing that needs to be improved are the lenses and the sweetspot. After that we can look into higher FoV and after that higher Pixel desnity.

          • I’m not a pixelwhore, as I still enjoy my VFX-1 with 320×200 per eye, and I still think the DK2 and the current CV1/vive are good enough for now, except the GPU’s need to increase in speed and decrease in price for now..
            I agree with you completely, first other things have to be resolved perfectly before going to higher resolutions..
            even dumping oled displays for DLP projection might solve some of the screendooreffect problems (as on 720p it already is much less than with the current CV1/Vive, so upping the resolution to 1080p (which I think isn’t available at the moment for micro DLP yet), would even make it less of a problem..

          • where the hell did you see true 4K VR ? this one have dual 1440p , 4k will need 12.5% more power.
            so if RX480 can run 4K just fine, can’t see any problem to run this VR.

          • RX480 is faster than GTX980 and my GTX980 run games just fine on 4K @ 60FPS ( mixed on High/Ultra ), can’t see any problem to run dual 1440p @ 90FPS

      • what you say makes a lot of sense for graphically intensive games, but that’s not the only use of VR headsets, and you don’t have to render at full resolution.
        I think even with a GTX970 you could benefit from a higher resolution HMD : games would still be rendered at a lower resolution and be downsampled (which is what Gear VR is doing most of the time), so you may lose a bit of sharpness, but you’d also have less SDE since each physical pixel would be smaller. And when using your headset with an app like BigScreen to watch a movie, or use your desktop, or play a game that’s less demanding, you could use the full resolution, which would make the experience a lot better.

      • True. You don’t need the same PPI across the whole FOV. With eye-tracking, dynamic rendering featuring cascade of PPIs around fovea is probably the best balance between experience and performance.

      • Won’t REQUIRE top-of-the-line, but even with foveated imaging you’ll
        want as good a PC as possible to maximize supersampling, in order to
        eliminate aliasing. But at least the requirements won’t go up relative to non-foveated 1200p as foveated screen resolution goes up.

      • I often play for 2 or 3 hours at a stretch. I’d be lying if I said my face isn’t totally sweaty after I take off the Vive, but its never sore…. and I’m even using the extra skinny foam pads to get better FOV.

  3. StarVR has all the time they need to improve – the current video cards (with MAYBE exception of GTX 1080 ti) cannot handle this kind of resolution with good level of detail and 90Hz framerate needed for immersion. The resolution is nearly 8K and is twice what Oculus and Vive have right now – which already require GTX 970/980.

  4. Just go with steam and lighthouse then not only will you have the best tracking system for VR… You also become an option for Vive owners which would expand your market! Lighthouse tracking is the best at least right now and I expect it will also improve over time!

    • I agree 100%. SteamVR could also support this HMD so developers can easily make games for it. (not sure if this is OSVR, something inhouse, or even SteamVR). I could see these guys making “premium steamvr headsets”

    • “Lighthouse tracking is the best at least right now and I expect it will also improve over time!”

      Yes to the first, maybe to the second, but ultimately Lighthouse might be usurped by technologies such as the Sixense STEM magnetic field system.

      Magnetic field tracking has some very nice properties that Lighthouse and other systems will never overcome. STEM has no occlusion problems and potentially very long range. If it can also match other tracking technologies in terms of price, availability, accuracy and battery life, then there’s going to be less and less reason for HTC and Oculus to keep pushing their IR optical tracking solutions.

      My only fear here is that STEM is already a little too late to the party, and it may have problems that Sixense simply hasn’t reported. Hopefully we’ll find out when they ship it later this year (hopefully).

      • I would be completely fine with that if they can beat lighthouse in performance, accuracy, and battery life… I also ordered a Stem system and although I own the Vive am interested in how well they will perform against it..

  5. The most serious problem for VR at this moment is not resolution or FOV… its motion sickness. As long as there is no solution for that one, the possibilities to make games are extremely limited, if you do not want about 50% of all people to not beeing able to play the game without becoming seriousely sick and turn their back forever on ANY VR (even non-sick making one).

      • Of course.

        Its because 50% of all humans DONT have the motion sickness problem. Then of course something else is THEIR problem.

        But for the VR industry as a whole (and if VR fails, it will also be YOUR problem ;) ), the problem that 50% of the humans on this planet cant play VR games at all (if there is artificial locomotion), even with 8K per eye and 220 degrees FOV, is a far worse problem.

        And there is solutions for this. One of them is suposedly already in use with helicoper simulators for the US-Militay and involves faking the vestibular system to believe that it feels the motions, that the eyes do see.

        • Ah, now you add “if there is artificial locomotion”. The reason I haven’t had a problem with sickness is because all the games I’ve played are designed for the hardware – they use teleportation (or are set inside vehicles). Just design games around for the system rather than porting over 2D games and it’s not a problem.

          • There is nothing wrong with free-roaming locomotion if you have the right hardware for it, which not many people do right now. For the present, I think developers should include options for both free-roaming and teleportation. Locomotion is not an unsolved problem, the solution is just not widely available yet.

          • It’s a solution but not a convincing one for your inner ears. For the longer term a more “mechanical” solution is needed such as the InfinaDeck although we are at least a decade away from these transforming into “usable”and affordable consumer-grade devices. A combination of roomscale and teleportation should suffice for the majority of games for now and the foreseeable future that comes pre-packaged with the VR device and without requiring a purchase of additional motion platforms.

          • Hmm, not sure about the Infinadeck. One of the problems with motorized treadmills is their latency. After you stop walking they take a while to bring you back to the center, which your inner-ear can detect (when it stops). On a passive treadmill like the Omni, stopping and starting is practically zero-latency as it’s all you. Your inner ear does not detect movement through space, only acceleration. You can see in the gif the little head accelerations and decelerations accompanying each step which are equivalent to what the head feels in real life. It’s that (along with the correct leg movement) that eliminates sim-sickness, and will make first-person exploration games like Fallout 4 VR and Resident Evil 7 VR playable – a combination of teleporting and room scale won’t work as well for these very important types of games.

          • The InfinaDeck is still in development so there’s plenty of time to improve on latency but it cannot possibly be compared to a passive treadmill until the technology reaches it peak. When it does, it will be miles ahead in simulating the most natural gait possible. We could talk about this all day long but it’s not worth discussing at the moment especially when the individual in question most likely has never tried a passive option such as the Virtuix Omni. Now this is all thinking on a longer term meanwhile the most natural solution to locomotion in VR is roomscale. A combination of both is the most effective solution for those who are prone to motion-sickness, those who want to achieve the most natural gait possible without having to establish any sort of pattern of leg movement and with the freedom to traverse hundreds of virtual miles and yet it is extremely effective for first person exploration games, that is, of course, if you have tried this method yourself to start with.

          • By a ‘combination of both’, do you mean roomscale and teleporting? Most people barely have space to take a couple of steps in any one direction, which means the amount of teleporting that is needed to play an open world game is far too much. I do understand that as motorized omni-directional treadmills are still in development they shouldn’t be written off. For the time being though, passive treadmills combined with 6DOF head and hand tracking is the high end of exploration based VR. At the same time it’s compact enough to make it into people’s homes as well as arcades. That gives me hope that the VR industry will be able to establish itself over the next 12 months. AAA content has been announced for VR, and that’s good news for everybody.

  6. With the wider field of view available in the StarVR, I expect there will be a trade-off between stereo overlap and peripheral vision.
    I would have thought stereo overlap in the central vision area was more important for immersivity than wasting pixel space at the margins.

  7. I wish we’d just go with PPD for all displays period. I’ve been calculating PPD for my monitors and matching them up for years. It’s the actually important ergonomic measurement. PPI is not very helpful most of the time.

  8. StarVR + Star Citizen. This is going to be the Ultimate dream of gaming for a lot of people. But which 2 GPUS would give me a stable 120FPS minimum at such high resolution coupled with ultra high fidelity visuals in Star Citizen at launch? Do we need a miracle in the GPU market to drive this headset for AAA titles? Will this be possible by 2020?

  9. I love my Oculus but I hate the FoV. It really interferes with my enjoyment of the “Virtual world”. I’m looking forward to getting my hands on a next gen Oculus with wider FoV of this, StarVR.