Finland-based Varjo has announced a new display technology for AR and VR which the company is calling “the first human eye-resolution VR/AR/XR immersive display.” Varjo claims an “effective resolution” that’s nearly 70 times greater than the Rift and Vive.

We’ve recently seen new VR displays from both Samsung and Kopin with an impressive ~3.25x increase in resolution compared to the Rift and Vive.

Varjo however claims an effective resolution that’s nearly 70 times greater than today’s headsets. They’re doing so with an interesting display layout which combines typical VR displays with higher density micro-displays to make part of the image hyper-sharp. The company says that portion of the image achieves retinal resolution, meaning that the pixel density is so great that you can’t tell individual pixels apart.

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

To achieve this, Varjo is projecting a high-resolution microdisplay into the center of a lower resolution display. The effect, as reported by The Verge, is like having a super high-resolution window right in the middle of your field of view, with lower quality everywhere else; that’s why they’re calling this “effective resolution” rather than “resolution.” Although the super high-resolution area is presently limited in its field of view, the visual fidelity it offers looks very impressive.

Above: Through-the-lens view of a limited portion of Varjo’s prototype headset. Below: the same view (and limited portion) of the Oculus Rift | Image courtesy Varjo

The Verge’s Sean O’Kane explains what it was like to look through the company’s prototype headset, which is built from an Oculus Rift:

Looking at them through that window in the center gave each of these scenes new life. Textures that were obscured by the Oculus’ dual 1080 x 1200 displays could now be seen in more lifelike detail. I could read individual filenames in the folders on the virtual desktop. The cockpit of the plane was especially striking. Looking at it through the Oculus displays surrounding Varjo’s tech, I couldn’t understand any of the labels on the many knobs and switches at my virtual fingertips. Looking “through” those microdisplays, though, I was able to read all of them.

O’Kane notes a few issues with the current prototype, like a notable border separating the high-resolution and low-resolution parts of the image (rather than a seamless blend between them), and issues with a mismatching framerate. According to the report, Varjo says they can fix both of these issues, and has an aggressive timeline set for what they say will be a professional-focused VR headset coming in 2018 which will be priced at “thousands of dollars.” The company also says the tech could be used for AR applications.

 

Image courtesy The Verge

 

Combining multiple displays to form a single view in a VR headset is not a new idea, but it is an interesting one. One of the biggest challenges is integrating each display in a way that forms one cohesive image. With Varjo’s present tech, the high-resolution area sticks in the middle of the field of view no matter where you move your eyes.

SEE ALSO
Vive Eye-tracking & Corrective Lens Add-on Confirmed Compatible with Vive Pro

With advancements in eye-tracking, it could be feasible to move the high-resolution area to wherever your eyes are pointing, and it seems this is where Varjo hopes to head 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.


  • Timotheus

    Ok. So basically a very high resolution microdisplay image is projected onto the lense by a mirror. If they achieve to change the projection to follow some eye tracking you would basically have retina resolution everywhere.

    Atm such a microdisplay may cost 800$ and the resulting HMD might be in the spheres of 1600$ or something. Not really cheap.

    But I would pay that much money, if I would have the ultimate resolution now instead of waiting 20 years for it. Given it already has the eye tracking built in it and the border polished such, that like in the Nvidia foveated rendering, you don’t even notice, that the peripheral view is low detail.
    People hope for 2nd gen VR to have some 4k at best and Varjo already could provide 70x the current resolution?

    Of course a game in ~16k(?) can’t be rendered natively, but even with upsampling it would hugely help in getting rid of the screendoor effect and the HMD theoretically is top notch the next couple of years, with the graphical power raising up to finally achieving that 16k resolution.

  • Mei Ling

    What happens when you start rolling your eyeballs around looking to the sides? You’re going to end up with significantly lower resolution and the experience would feel very strange. Foveated rendering would draw power to where you are looking but in this case it won’t improve the resolution because their customised display is fixed and constantly at the centre of field of view.

    • Master E

      I could be wrong, but some applications are already using foveated techniques i.e. The Batman game… if you look straight ahead it’s much clearer than subtly checking the periphery and it works great. If they are hitting resolutions this high and clear hopefully he’ll be on top of clearing up the periphery a bit in no time.

      Regardless… weren’t most of us thinking 4k VR was like 5 years away? Things happening much faster than I personally expected. Which means HTC and Oculus are hard at work with their competition. Only a matter of time (and way less than previously thought) they are going to make these dreams come true

      Holodeck in my lifetime seems attainable

    • RavnosCC

      Agreed, seems like an oversight… no pun intended.

  • Lucidfeuer

    Genius idea. Your overall ~220° FOV outside of your focus point and it’s outer ring doesn’t account for more than “4K” in resolution.

    However you’d need a micro-display or mirror capable of following your eye movements while projecting a ~4k image per eyes. That could be the mechanical solution to achieve eye resolution, while foveated rendering would make it possible GPU/Software-wise.

    Again, some people are doing the job of corporations which are failing at the VR conception and thus market.

  • Master E

    Aim for the stars and at least hit the moon

    So blown away by announcements such as these and happy VR/AR/MR is getting such support and motivation behind it

    I’d imagine if they can do 70x the current displays scaling it down will be the easy part for games and applications

    Would easily pay $1k for this

  • Toffotin

    Torilla tavataan.

    Sorry, ahem… Yep, seems really interesting. Reminds me of Magic Leap. Or at least what I have understood Magic Leap is. I haven’t read the Verge article yet, only this one and one in Engadget, but it seems like very few people have gotten a change to try it out, so we’ll see. I work in Helsinki and also am a game dev, so I wonder if I could get a demo from them :)

  • Tim Suetens

    “…professional-focused VR headset coming in 2018 which will be priced at “thousands of dollars.”

    Thousands. Nothing to see here, move along.

    And unless your computer was recently stolen from NASA, you won’t be able to run the thing anyway, even if you DO have the cash to spend on the headset.

    • Toffotin

      Well Microsoft Hololens devkit is 3000$, the Commercial Suite is 5000$.
      These are mainly meant for developers, I’m sure the price will come down eventually. By how much though, I can’t say.
      Also, in the Engadget article it said “The company adds that the system has low computing requirements thanks to “foveated eye tracking.””. It probably still requires a pretty kick ass computer to run smoothly, but nothing that has to be stolen from NASA.

      Engadget article: (https://www.engadget.com/2017/06/19/varjo-promises-a-vr-headset-with-human-eye-resolution/)

      • Tim Suetens

        Yes, my point is that at this price point and with these system reqs, this company and its products will be completely irrelevant in helping VR break through to the mainstream.

        Almost no one will be able to afford it.

    • Toffotin

      Well Microsoft Hololens devkit is 3000$, the Commercial Suite is 5000$.
      These are mainly meant for developers, I’m sure the price will come down eventually. By how much though, I can’t say.
      Also, in the Engadget article it said “The company adds that the system has low computing requirements thanks to “foveated eye tracking.””. It probably still requires a pretty kick ass computer to run smoothly, but nothing that has to be stolen from NASA.

      Engadget article: (https://www.engadget.com/2017/06/19/varjo-promises-a-vr-headset-with-human-eye-resolution/)

    • Chugs 1984

      “And unless your computer was recently stolen from NASA, you won’t be able to run the thing anyway, even if you DO have the cash to spend on the headset.”

      lol

      I can see system requirement now:

      CPU: Cray XK7, Opteron 6274 16C 2.200GHz, Cray Gemini interconnect,
      Cores: 560,640
      RAM: 710,144 GB
      GPU: NVIDIA K20x
      TFlops: 27,112
      Power: 8,209.00 kW per hour (lol)

      But hey, at least you’ll be able to tear a hole in reality with this thing

  • bud

    best resolution I have seen outside of real life is lucid dreaming, any one that has had some proper LD wishes these could be able to be tapped into. I guess they will one day.

  • bud

    best resolution I have seen outside of real life is lucid dreaming, any one that has had some proper LD wishes these could be able to be tapped into. I guess they will one day.

  • Jason Mercieca

    Thousands dollars wow no way …

  • Rep Rep

    Well, if they can incorporate foveate rendering and eye tracking and bring the cost down to $1,000….then we can start talking about a consumer product. Otherwise, it’s no use.