Today’s VR headsets may have relatively high resolutions on paper, but when the pixels are stretched across a wide field of view, the effective angular resolution is far lower than what you might expect from a typical 1080p TV or monitor. Unfortunately, a suitable VR display capable of achieving both a wide field of view and retina resolution isn’t readily available yet. Until then, Finland-based VR startup Varjo is using a combination of macrodisplays and microdisplays to put high density resolution to the center of your view without giving up the wide field of view.

Update (5/30/18): At AWE 2018 this week I had a chance to see Varjo’s latest ‘Beta’ prototype, which demonstrates clear improvements over its predecessor. Compared to the Alpha prototype, the Beta headset is using a higher resolution ‘context display’ (the screen which makes up the peripheral view) at 1,440 x 1,600 per eye, while the ‘focus display’ (which offers retina resolution density at the center of the view) remains at 1,920 x 1080 per eye (60 pixels per degree). The company has also confirmed the headset will support SteamVR Tracking 2.0 (as well as 1.0).

Photo by Road to VR

Notably, the size of the central focus display (the area of high pixel density) is slightly larger on the Beta prototype, and the blending between the two displays has been improved. It isn’t invisible, but it’s smoother and cleaner than before, largely eliminating the appearance of a slightly darkened band around the edge of the focus display (while still showing some glaring and distortion). That makes the transition from the high density portion of the view and the lower density portion less obtrusive. Varjo says they’re now using hardware and software tweaks to smooth out the transition.

The improvements in the blend between the displays have now reached a point where it’s definitely worth putting up with some artifacts around the edges of the focus display for what you get in return: a truly retina resolution area at the center of your view with absolutely no visible pixels or screen door effect. Looking through the focus display reveals sharp text which would otherwise be unreadable with the context display alone. Photogrammetry scenes were at times breathtaking—miniscule textures are revealed which are simply invisible without the level of pixel density provided by the focus display, making the scene look truly photorealistic (when looking through the center of the display).

Looking through the headset’s focus display feels like looking into the future of VR itself—toward a time when the entire field of view will offer retina resolution. But therein lies the big question about Varjo’s inset display approach: other companies are working on larger high resolution displays for VR, and eventually it seems we will achieve retina resolution displays which can cover the entire field of view. So is Varjo’s headset just a stop-gap, or something more? Time will tell, but from what I’ve seen from the company so far, I doubt their inset-display headset will be their only contribution to the VR space.

Photo by Road to VR

The company’s CMO, Jussi Mäkinen, tells me that the Beta prototype has also made strides toward manufacturing readiness. For the headset’s initial launch the company has decided against hardware foveation (moving the focus display to keep it at the center of the gaze). R&D on that front continues (we covered several of the company’s potential approaches here), but early access partners using the headset said they would rather have the fixed-focus display headset in their hands sooner rather than wait longer for a hardware foveated version, Mäkinen told me. That said, the initial fixed-focus display version will include eye-tracking which could be used for foveated rendering, and potentially aid in further smoothing the transition between the focus display and the context display.

Varjo plans to launch the commercial version of the headset later this year, and is exclusively targeting professional/enterprise use-cases, with prices expected between $5,000 and $10,000.

Photo by Road to VR

Original Article (2/26/18): The Varjo headset makes use of what the company calls a ‘context display’ and ‘focus display’. The context display is a large macrodisplay with a 1,080 × 1,200 resolution spread across a 100 degree field of view. Alone, it would look almost identical to the fidelity you’d expect from the Oculus Rift or HTC Vive. Varjo’s trick however is putting a microdisplay (the ‘focus display’) with a 1,920 × 1,080 resolution at the center of the headset’s field of view. Although the focus display isn’t tremendously higher resolution than the context display by pixel count, it’s pixels are packed into just 35 degrees horizontally, making it incredibly pixel dense.

Photo by Road to VR

At MWC 2018 I got to check out the Varjo Alpha prototype headset. Inside I saw an extremely high quality image at the center of my field of view which had no noticeable screen door effect. Beyond that 35 degree rectangular area, the resolution drops to the same levels you’d expect from first-gen consumer VR headsets. At the boundary between the focus display and context display, there’s an imperfect transition between the high resolution area and the low resolution area, which looks like a blurry rectangular halo, but it was actually somewhat less jarring than I was expecting.

A rough approximation of how the focus display looks against the context display. Relative fields of view are not to scale. | Photo by Road to VR, based on images courtesy Varjo

The company is using an optical combiner, essentially a two-way mirror, to composite the two displays. Going forward, Varjo hopes to further smooth out the transition between the two displays, the company’s CMO, Jussi Mäkinen, told me, using a combination of both hardware and software refinements.

Understanding Pixel Density & Retinal Resolution, and Why It's Important for AR/VR Headsets

The difference in quality between the focus display and the context display is truly night and day. Not only does the focus display not show any noticeable screen door effect, the jump in angular resolution turns otherwise blurry smears into perfectly legible letters, as a virtual standard eye chart placed inside the demo experience made clear. Textures benefited immensely from the improved angular resolution, revealing detail that simply isn’t visible on the lower resolution context display.

In the video above, keep your eye on the lower lines of the eye chart to get an idea of the difference in resolution between the focus display and the context display. The resolution jump is less noticeable here since the camera isn’t capturing the headset’s full field of view. You can’t quite see the boundary artifacts in this video.

The Vajor Alpha prototype was tracked with SteamVR Tracking, which the company plans to continue using going forward. Since this was a handheld demo (no strap on the headset yet), I didn’t do a comprehensive test of the headtracking tracking in its prototype form.

– – — – –

In its current state, even with a static focus display, the benefit of the extra resolution is apparent, and you can almost fool yourself into thinking the entire display is sharp, as long as you consciously try to keep your gaze pointed through the very center of the lens. Of course, course, in practice, your eyes are not always looking perfectly through the center of the lenses, so you won’t always be looking at that super sweet spot of the focus display. And you’ll have to contend with your eye crossing back and forth over the transition point (and back and forth between high res and low res).

Varjo’s long term hope is that they’ll be able to move the focus display in real-time, creating a sort of hardware foveated display, such that the focus display is always at the center of your gaze no matter where you’re looking. That would of course require excellent eye-tracking (and then some) but if they can pull it off, it would make the headset even more compelling because your eye would never have to cross the border between the displays (and always be in that super sweet spot), and your brain might even do a good job of ignoring the border if it’s always a fixed distance from the center of your view.

Achieving an active focus display isn’t likely to be an easy task, though Varjo has patented several potential approaches, mostly involving quickly pivoting the optical combiner about two axes, which we examined recently when we explored the company’s key technology.

Though Varjo CMO Jussi Mäkinen tells me that the company already has prototypes with an active focus display, it’s still up in the air whether or not Varjo’s first commercial product, which is planned for release later this year, will use a static or active focus display.

Photo by Road to VR

Mäkinen told me that Varjo is for now focusing exclusively on enterprise and commercial applications for the headset, which the company initially expects to price between $5,000 and $10,000. He said that the company is actively listening to feedback from its partners about what aspects of the headset are most critical for improvement.

Understanding the Difference Between 'Screen Door Effect', 'Mura', & 'Aliasing'

If Varjo can develop an effective and reliable mechanism for active foveation, their headset could be great stopgap for making near retina quality VR headsets with a wide field of view before any single display (per-eye) is ready to ready to deliver such an experience. But we know that both macrodisplay and microdisplay makers are working toward that goal, so where does that leave Varjo once a single display can do it all?

Mäkinen says that Varjo doesn’t just want to make a headset, they want to pioneer the productivity use-case of VR, using a mix of hardware and software. One example he gave was using a totally virtual workspace without the need for physical monitors, something that can really only happen once headset resolution is high enough. He said we can expect more from the company on the software side in the future.

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.

  • We all know it is cool, but the problem, as the display expert Karl Guttag has said in his blog, is that having the moving display is damn difficult. Because it is very hard to made it without jittery and without your eye noticing the transition are between the two displays (with one moving). So, fixed one is “easy”, but I don’t get that much the use and moving one is hard

    • Laurence Nairne

      I feel like they’ve taken the wrong approach to the problem, and will develop themselves into a corner. I don’t know the numbers, but I’m certain a software solution to foveated rendering would be cheaper to deliver on mass than a hardware one. They may get there first (though even that’s doubtful), but as soon as eye-tracking is standard and marketable software exists to make use of it for foveated, these guys lose.

      • Thoemse

        100% agree. Using a display with ridiculous resolution (if available) and only rendering max res in the main focus FOV sure has to be easier then this. Considering that Pimax is allready using 2x 4k displays I am sure we are not all that far off from 8K displays. Imagine a pimax with 2x 8K dispalys and foveated rendering in the main focus. Changing that focus with eye tracking would be the icing on the cake.

        • cataflic

          It’s ok to render foveted areas with eye tracking, but to cover an entire human eye fov, a total display solution with the same ppi of the Varjo central display must be multiplied for 20… i think that for many years will remain echonomically impossible to commercialize this solution

          • Jerald Doerr

            All this talk about true eye resolution… my god it will be 10 + years before we even get close to that with displays even video cards that have the power to render games at that resalotion and still look good artistically… True 4k + Hi-FOV + software solution + Duel GPUs + Sli/Crossfire direct HMD suport would be killer.. is the only hope for the next 3-5 years and I will be happy for a long time…. leave the rest up to GPU upgrades.

          • Hivemind9000

            I thought Varjo’s screen was about 6x the pixel density of current offerings from Vive/Oculus (3000 ppi vs 450-ish ppi)? Which should mean a 6-7k screen (per eye) should give a similar resolution to Varjo at a 100 degree FoV. Not that big a step compared to today’s high-end offerings (Pimax at 4K/eye).

            Personally I want more FoV (like Pimax’s 200 degrees) which would make the requirement about 13k per eye to match Varjo’s fidelity. That’s more of a stretch I guess.

          • cataflic

            I think that 8k per eye with a 16:10 aspect ratio will be ok…
            Human eye with a 120° overlapping fov and 150° fov per eye will be ok.
            2 years to produce such compact supercrisp panels at “human” prices
            1 year to deliver alpha product ideas with sw foveted rendering
            1 year to hype
            …5 year to market at high prices
            If Varjo can do it in 1 year has got a 4y andvantage…

            I think that ar solutions will come first…

        • josh

          Thats what Pimax hopes for but everyone highly doubts theyll achieve this because they STILL haven’t actually gotten 4k to work reliably and even when it does almost work it takes TWO hdmi plugs, one per screen. I could also buy the best components the market has to offer and I wouldnt have enough available pc power to render 4k VR, the components just don’t exist for consumers.

          • Hivemind9000

            Not sure why you think Pimax haven’t got 4K (well 2 x 4K) working reliably. Care to share your source? They still need work, but from what I understand it is more around the lenses and tracking at this stage.

            The two HDMI plugs are more an issue withe the state of HDMI standards, which I am sure will change over time with new 8K panels starting to make an appearance (Dell’s 8K display needs two HDMI cables to drive the screen at 60Hz).

            The main challenge will be having the hardware to drive them at close to 90Hz. With Pimax’s 4K->8K upscaling you can do this today on a decent rig (GTX 1080ti etc). Sure, they are pushing the high-end of the market, but someone has move the technology forward. That’s definitely the end of the market I’m interested in.

          • G-man

            They are havhav trouble getting above 82fps or something like that on the full 8kx. Gpus right now theoretically can do two 4k res at 90fps as their absolute maximum resolution. But thats based on full 16:9 8k at 60fps. So it may require better gpus to get 90fps.

      • Jistuce

        The problem remains that displays with the resolution needed simply aren’t available in the sizes needed. That’s why the microdisplay insert exists, because the optimum solution can’t be implemented yet.

        Yes, it is a temporary solution that will be rendered obsolete as technology progresses, but right now it makes sense.

        • Cynthia McBride

          Gℴℴgle giving every one $99 per-hour to complete few task from the comfort of home . Labor for only few hours daily & fun more time together with your own relatives .. Any one can also get this best career!!last Thursday I got a new Fiat Multipla after I been earnin $7934 this-past/six weeks .no doubt it is my favourite-job however you wo’nt forgive yourself if you don’t test this.!oh851i:➥➥➥ http://GoogleFeedSmallBusinessOpportunity/getpay/$97/per/hr ♥♥v♥♥b♥m♥♥z♥♥a♥♥♥q♥♥z♥♥b♥♥w♥f♥♥d♥♥k♥♥q♥♥♥g♥♥s♥♥♥w♥♥♥r♥♥♥f♥♥b♥♥v♥♥j♥d♥s♥u♥♥♥k:!lf241a:cahlf

        • Laurence Nairne

          That’s a fair point, but in its current state it isn’t a full solution. It’s a static prototype with possible avenues to explore.

          By the time they build a working, commercial product, proponents of the software angle will have caught up. As will the displays to power it.

          I’d also say that Varjo can’t be looking at this as a temporary solution as they are using investors’ money to finance it!

          • benz145

            It depends on the economics. If Varjo can nail a solution for moving the image of the focus display, and also eliminating the boundary transition, it’s possible that their solution could have a cost advantage compared to using a single high density display that covers the whole FOV.

          • Laurence Nairne

            Hmm, I’m still not convinced (though admittedly it’s a theoretical argument based on numbers that haven’t been decided yet!)

            It’s also more prone to breakdown and I expect there’d be issues with weight distribution, but that depends on the mass of the moving display and it’s housing.

            I just can’t see the longevity in such an approach, though I’d be happy to be wrong. Display technology exists and packing more pixels in is an iteration, this is virtually an original concept with no clear solution to making it work in practice, reliably, every time.

          • Lucidfeuer

            Even with foveated rendering, you need a very high-resolution adaptive display, which simply don’t and won’t exist for a long-time.

            The “eye resolution” is about 6K per eye so 12K ambiguiated (between the two eyes) at most, but that’s only for a single fixed focus point. Apply that to the whole FOV (~220°) and you have to multiply that by as many focus point as possible which I saw somewhere amounts to 70 mp, which means you’d need a single 32K display to cover the whole FOV at eye resolution…

            Their solution might be the only way to get there at first, especially as the augmentation of FOV is the priority (which means resolution gets spread further accros the screen). And as Ben said above, since the eye resolution really only is 4K in the focus area, 1K around and little less in the rest of the viewpoint, this solution might be more economical than an overkill 32k display.

          • Laurence Nairne

            Well sure if we were only going to ever accept an eye resolution solution then you’re both absolutely right. I guess perspective dictates whether this survives – if they secure enough enterprise level customers then it’ll be a dream. It’s certainly not a solution for the masses (but I’m aware they don’t currently intend to be with that price bracket).

            Even in enterprise it only hits a niche market of potential customers as many will settle for lower resolutions when they have to consider buying HMDs in bulk.

            I do hope it succeeds as it’d be great to have something like this in the ecosystem – I’m sure I’ll never get hold of it in my line of work but it opens to gates to finding more cost effective means of delivering the same results if nothing else.

          • Lucidfeuer

            Well, the truth is I don’t see a mechanical solution being for mass VR market either, but since there’s no mass market yet and only niche VR, they could very well find a cost effective solution if the incentive and implementation of that solution delivers…alongside Foveated rendering.

          • Laurence Nairne


          • theonlyrealconan

            People tend to forget: pimax is only 1440 upscaled. Those that have tried the vive pro and pimax 8k say the pro is sharper. So we are even further away then people think.

    • Lucidfeuer

      I believe if they had the idea and competence to do this prototype, it clearly show promises in at least researching and implementing solution. I remember someone mentioning piezoelectric moving polarised mirror (meaning it only reflects on the side the microdisplay is projected on), that’s one exemple.

    • dk

      yep obviously to be great it must work almost as well as a headset with no moving parts

  • bill carson

    So the obstacle to using the microdisplay throughout the lens is cost , is that it ? More costly than finding tech for the eye tracking and smooth transition ? I think by the time they figure all this out , the costs will come down and just microdisplay will be used .

    • elev8d

      Yeah, that’s what I’m trying to understand as well. I was expecting them to be much cheaper than several thousand dollars.

      • bill carson

        Actually I forgot .The issue is the graphics power needed to drive the micro display resolution .

  • Nate Vander Plas

    One thing I noticed that’s missing from this conversation is that, in order for the moving/foveated microdisplay to be useful, you have to first fix the inherent lens issue. With all lenses on VR headsets I’ve seen, there is a sweet spot in the center and the rest is pretty blurry. It wouldn’t matter if the portion of the screen way off to the side is retina resolution – it would still be blurry until you moved your head to put it back in the center. This non-moving solution seems like it would be enough to solve the current problem. But ultimately I hope that lightfield displays like Magic Leap (as much as many here hate that company) replace the single-plane display altogether.

    • brandon9271

      ^This. I have a Samsung Odyssey and a Rift. The resolution of the Odyssey is “good enough” but the sweet spot is a pain in the ass. The Rift has a much better sweet spot but the god rays are TERRIBLE. Serious work needs to be does with optics. I’m very disappointed that Vive Pro is rumored to uses the same optics.

  • Andres Velasco

    Meh, don’t care

  • Jason Gorman

    Could micro lens arrays (MLAs) that accomplish the same thing without needing two displays be used instead or is this a more expensive approach (Nanoscribe)? Could an MLA also be used in the same way that the current HMDs use the fresnel lenses?

    • Downvote King

      Even if they just scaled the image instead of producing the native high resolution needed to truly take advantage of this idea, the lack of visible pixels would still make a huge difference, and provide a degree of future-proofing as more processing power becomes available.

  • daveinpublic

    I think the main issue with this solution is the same issue with regular foveated rendering… they haven’t been able to track people’s eyes well enough. Facebook talked about this a year ago in their Oculus event. So, these will be ready at about the same time the software solution is ready, and the software version will probably be cheaper than a motorized high res micro display. Higher res screens are in development already… from RoadToVR in 2017, “Earlier this year, Clay Bavor, VP of VR/AR at Google, revealed a “secret project” to develop a VR-optimised OLED panel capable of 20 megapixels per eye.” If they can follow through on this, and create a little higher resolution than a software solution, more power to them, but I don’t think anyone is going to follow them into what appears to be a money pit.

    • Dave

      True and what little advantage they may have performance wise could easily be short lived and all they’ll be left with is an expensive device which was once upon a time was quite good.

      This work though is important as other will learn from this and ultimately it contributes to advancing VR which is amazing. Who knows they may do really well but selling devices for 5K to 10K US dollars even in the commercial sector with the relatively low uptake and high investment in VR just sounds like a really bad business plan as it will soon be overshadowed by much better cheaper devices.

  • MarquisDeSang

    I don’t care about definition, give us 140fov+

    • G-man

      just one pixel for each eye that you have a 140 degree fov of…..

  • $5000 to $10000…. BaHAhhahahahAHahahahahaha

    • dk

      chinese knockoff ….500

  • dk

    “Today’s VR headsets may have relatively high resolutions on paper” … today’s res is a joke …especially the rift/vive res….because without eye tracking or other optimisations driving more pixels is pretty hard

    • MosBen

      The point of that sentence is that most people have TVs or computer monitors as a point of reference for resolution, so the displays in current HMDs seem, on paper, to be relatively high resolution, but in the specific application of VR it turns out that you need high resolution to achieve the same perceived clarity because our faces are so much closer to the screens and the pixels are stretched to a much wider field of view.

      • dk

        and my point was there isn’t a headset with good angular resolution on the market…and the 1st gen was especially bad…… because they haven’t implemented anything saving resources in a massive way

        • G-man

          your judgement of “Good” is based on what? some fantasy land you live in where there are 8k 3inch oled panels everywhere?

          • dk

            so u don’t think that the angular resolution of the first gen(lower than 1440p gearvr) looks like crap?
            …………aaaaaand the point was as long as there aren’t great ways of saving the resources they will keep going for whatever resolution is convenient to work with

          • G-man

            the resolution of the screens on gen 1 was what is was because thats the screen technology that was available. there sno reaosn to assume future headsets will implement no resource saving methods. go uses lower frame rates, and foveated rendering because the panels are too high res for the soc. there is zero reason to think companies will stick with picking a resolution that is “convenient”, just as theres nothing to comapre good, or “looks like crap”. for the first version of a product. the rift nad vive were the best looking and highest resolution vr headsets you could buy. they looked like THE BEST. you are comparing them to looking at a 4k tv from 6 feet away or something in your own head and complaining it doesnt look as “good ” as that

          • dk

            they could have had a bit higher res on the 1st gen even without eye tracking or something like that …..but there were plenty of reasons not to…..I’m not saying next gen will be the same ……just that it will be disappointing if it is

          • G-man

            the screen technology to do higher resolution for gen 1 in mass production quantities didnt exist.

          • dk

            what technology didn’t exist …having just slightly higher pixel density …..sure fine …..super impossible got it

  • Francis Prado

    People saying software foveation and higher density screens are easier/more likely to appear in the future don’t seem to understand two things:

    1) Even if you want to display a say, 2K content on a 16K screen by upscaling it, that STILL requires quite a bit of processing power.

    2) More importantly, with sofware foveation, you have to keep increasing the megapixels as you increase the FOV. With this approach, the rotation of the foveal view has to be no more than how much our eyeballs are able to rotate (60 degrees to the sides and this is at most and not for all people).
    In other words, 16K per eye may solve screendoor as well as pixellation for 110 degree FOV. Now imagine trying to get it to 200 degrees, or the human FOV, which is 270!

    Im no fan of Varjo, I think they are patent trolls and will ultimately either keep selling pricey devices to the pro market or go out of business and sell themsleves/their patents. I even think the chose the dumbest method of implementing a foveated screen that has to be dynamic by using a large beamsplitter plate which is hard to rotate and noisy. But their approach is far better and future-proof at this point than keeping adding megapixels for the above two reasons.

  • David

    A headset needs to account for the eye not always looking at the center of the screen. If you get high resolution ONLY at the center of the screen, I think this might limit the usefulness of this headset to some niche tasks.