Waveguide optics are increasingly looking like the best near-term solution for creating AR glasses that are truly glasses-sized. Lumus is one such waveguide company and recently let their latest waveguide prototype, the 50° FOV ‘Maximus’, be examined under the proverbial microscope. Through-the-lens photos of Lumus Maximus reveal impressive ‘retina’ resolution of 60 PPD, along with excellent brightness and leading image uniformity.

Lumus is one of a handful of companies building waveguides, but their approach is unlike most. Rather than diffractive waveguides, the company builds reflective waveguides which bring some advantages, especially in light efficiency, which means greater brightness (important for outdoor use) and more transparent lenses (important for being able to look like ‘normal’ glasses rather than darkened sunglasses).

Near-eye display expert Karl Guttag got to examine the Lumus Maximus waveguide up close, including getting permission to take through-the-lens photos with known test patterns for measuring effective resolution. Guttag came away impressed with the performance of the Maximus waveguide across a range of metrics, though he notes that it’s yet to be proven whether or not the company’s manufacturing method is scalable.

Image courtesy Karl Guttag

The Lumus Maximus prototype that Guttag examined was merely a head-mounted display built to show off the waveguide; it doesn’t currently include any of the on-board compute, sensors, or batteries that would still need to be packed into the glasses to make them proper standalone AR glasses—as DigiLens has done with its Design v1. Still, the Maximus was functional from a display standpoint, allowing for a detailed look at real-world visual performance.

SEE ALSO
Google's Project Starline is a Light-field Display System for Immersive Video Calls

Effective Resolution

The current prototype uses a 2,048 × 2,048 LCOS microdisplay (per-eye) and spreads those pixels across a 50° diagonal field-of-view. Even with all those pixels you could wind up with poor image quality if you can’t precisely control the light as it makes its way from the display, through the optics, and into the user’s eye.

The Lumus Maximus waveguide seems to do great job of preserving the display clarity, even after all the bouncing necessary to expand the image from the tiny microdisplay to the full 50° field-of-view.

Image courtesy Karl Guttag

Guttag found that the Maximus waveguide managed an angular resolution of 60 pixels per-degree at the center of the display, which meets the ‘retina resolution’ bar. In theory, that means if you were holding an augmented reality book in front of you at a normal distance, you’d easily be able to make out the letters on the page (assuming the image processing pipeline applied during head tracking doesn’t sacrifice any clarity).

For comparison, no consumer VR headset on the market has achieved 60 pixels per-degree, not even the upcoming Vive Pro 2 with its class-leading 2,448 × 2,448 resolution. This is because VR headsets spread their resolution out over a much larger area, which trades angular resolution in favor of a wider field-of-view.

Brightness

Brightness is hugely important in AR glasses because many use-cases are imagined to happen on-the-go, which includes outdoors during sunny days. If you don’t have enough brightness, users will hardly be able to see what’s shown in the glasses when the backdrop is bright. Many AR headsets struggle to achieve brightness that’s even sufficient for indoor usage, which is why they tend to tint the lenses like sunglasses.

SEE ALSO
'AGAINST' is a VR Rhythm Game Aiming to Fuse the Best Parts of the Genre with Its Own Gritty Flair

As a reflective waveguide, brightness seems to be one of Lumus’ key advantages. Guttag reports that the company claims 3,000 nit brightness for the Maximus prototype. Equally important as the brightness of the resulting image is the efficiency of the entire optical pipeline. If your optical pipeline loses 90% of the light put into it, you can theoretically add a brighter light source to reach your desired output brightness—but this comes at the cost of more bulk, heat, and cost. Thus, a minimally bright light source combined with a highly efficient optical pipeline is ideal.

Image courtesy Karl Guttag

To that end, Lumus claims it expects Maximus will reach an efficiency of 650 nits per-lumen, which Guttag points out is more than ten times as efficient as the WaveOptics waveguide which sits at 50 nits per-lumen (which is why we aren’t surprised to see the new Snap AR glasses with a dark sunglasses tint). Not all diffractive waveguides may be as inefficient as WaveOptics however; in our recent hands-on with DigiLens’ latest (diffractive) waveguides, the company said they’ve reached 300 nits per-lumen.

Image Uniformity

The Lumus Maximus prototype also shows impressive image uniformity, which is the consistency of brightness, color, and resolution across the display (colloquially called the ‘sweet spot’ in the VR space).

Image courtesy Karl Guttag

Above we can see a comparison between Maximus and HoloLens 2, the latter of which has long suffered from poor color uniformity, causing a rainbow-like haze over the view. While HoloLens 2 is a worse-case example for comparison, Maximus still appears to be ahead of others as well. “The color and brightness uniformity of the image, while not perfect, is vastly better than any other waveguide-type optics I have seen,” says Guttag.

SEE ALSO
Quest 2 Continues to Grow by Leaps and Bounds Among VR Headsets on Steam

Granted, there’s still room for improvement. While the Maximus above clearly looks better than HoloLens 2, you can still clearly see the color uniformity and brightness start to fall off at the corners of the Maximus image. While this wouldn’t be as much of an issue with an immersive field-of-view, recall the image here is only 50° diagonally, which means the corners of the image are not far into the peripheral view where poor uniformity would be less noticable.

– – — – –

For all of its impressive optical performance, a huge question remains at this stage in the development of consumer AR glasses: can Lumus’ reflective waveguides affordably scale to large manufacturing volumes?

Image courtesy Schott

Last year Lumus announced a strategic partnership with glass specialty company Schott to grow manufacturing capacity and achieve what the companies call “favourable costs.” Guttag figures that Lumus’ approach could be “at least as cost-effective as diffractive waveguides” given the appropriate investment in manufacturing.

For an even more detailed analysis of the Lumus Maximus prototype, check out Guttag’s complete article.

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.


  • I thought they looked pretty slick … until I saw them on that dude’s face and realized they look kinda like wearing horse blinkers or something. Until AR glasses look basically just like real glasses, with, importantly, the lens part not having some visible crap looking dorky and obscuring part of it–which may be an impossible aspect of how such glasses even work–I do not believe they will ever catch on beyond mostly business uses and a bunch of hipsters willing to look like muppets to try the next hyped-up gadget/tech.

    • brandon9271

      Looking ridiculous hasn’t slowed VR down any :)

      • Because you don’t have to wear VR headsets outside and the like.

        • brandon9271

          you don’t have to wear AR outside either :) but generally I’m more concerned with doing my job versus looking cool. Welding googles, hard hats, safety glasses etc.. They serve a purpose but aren’t exactly fashion accessories. People should view AR in the same light… as a tool, not a fashion accessories for dickbags to try and look cool at the coffee shop. lol

          • Well, personally, I don’t give a crap about using these devices to be a better slave to the system–but that’s just me. I like to use such devices for playing games and watching movies and enjoying entertainment in general. I don’t care about a productivity tool; I care about and entertainment device. I already have my smartphone for all that lifestyle crap, and I don’t want to have to shove a stupid pair of glasses on my face to add another layer of adverts and popups and stats and graphs and step counters and weather reports. But I’ll happily shove a VR headset on my face to escape into some amazing gaming world that simply cannot be experienced as such in any other medium created to date. So, VR has afforded itself the “luxury” of being bulky (although it will get smaller), but AR can’t really afford itself that “luxury” imo, outside of really specific business uses–which I don’t give a crap about and are already being served their crappy experimental AR devices as we speak anyway.

  • silvaring

    At MWC 2019 in Barcelona Lumus hosted a talk, the CEO said that Production (helped by Quantus, to start in 2019) would make 100,000 lenses per month (with high and repeatable yields). What has become of this? Another company unveiled their own display at the time, LetinAR. Their wearable glasses prototype uses a single, vertical mounted OLED (640 x 400 24-bit RGB color with around 800cd/m2 & 10-50% light efficiency) built into the frame. More importantly the low resolution display is somehow going into a 4mm thick lens, with 7 small pinhole mirrors, for a FOV of 20 degrees per eye with objects in focus from 25cm to infinity (a virtual retinal display). I would like to see Karl talk more about this kind of display in comparison to what Lumus are doing.

    • kontis

      Anything under ~80 deg FOV which is more or less the casual consumer acceptance threshold (and not coincidentally it’s also where ortho-stereo starts being useful) doesn’t have to be mass manufactured and instead can be sold as expensive enterprise-only device, like Hololens.

      These things might be valuable as tools for some pros, but people who want to just play with virtual cats need something MUCH closer to natural FoV of normal glasses to even bother with these things as free toys, let alone actually pay for them. We are probably a decade away form that.

      It look like this will be another fusion energy – it’s always a decade away. In 2014 Facebook bet on Smartglesses in a decade and now they officially admitted it’s not happening as they aren’t even close to their spec targets.

      Zuckerberg’s theory of “new platform every 10 years” was a total BS.

      • Not expecting to see competent “heavy glasses” consumer AR until well into 2030’s…

        • Lulu Vi Britannia

          Five years from now, remember you guys said that ^^. We’re much closer to consumer AR than what people think.

          Software is already here, 6DoF is already here, displays are already here, all that’s left is figuring out proper waveguide on a wider FOV.

          • silvaring

            You’re missing one key element, fashion sense. If something cannot be hidden or is ‘ugly’ it will not get the ‘smartphone’ level of investment to take it mainstream. That’s just how it is. Fashion-Usability-Practicality in that order.

  • xyzs

    How much the army put in microsoft Holocrap already ? Ahaha
    Looks like they bought some xbox1 the price of the ps5 now…

    • Andrew Jakobs

      Hololens 2 is already out for a long time, and it’s more than only a visor. No cameras, no onboard SOC.. All they shown now is a display.

  • Andrew Jakobs

    All nice comparing it to hololens 2, which is a headset almost 2 years on the market already, whereas Lumus still doesn’t have this prototype in production for the foreseeable future, so by the time this is in actual production, Microsoft might also have a new hololens.

    • Hivemind9000

      You’re right, Hololens 2 does have a lot of other features going for it, but the actual article was comparing only the display system to a number of current products, including Hololens (i.e. it wasn’t a product-product comparison – just the lens/projection subsystems). The Maximus is only a prototype to show off the Lumus display tech. It will be interesting to see what products will adopt the Lumus display in the near future.

      That said, the article did throw a little bit of shade at the Hololens display tech choices:

      “They should have used a better LCOS device in HL2. The HL1 used one of the worst available at the time of its design, and by the time of the HL2 design, there were much better LCOS devices available.”

      Maybe the HL3 will have Lumus displays?

      • Phyllis Dvorak

        Lyla previous six weeks l earned $19562 by working on my computer staying at my home in my free hours.(r1290) I’m able how to do it by working few time in whole day with my laptop. Realy it’s very easy & everyone definetly can join this work. >>> http://www.Belifestyles.com
        .

    • dk

      yep plus there was recently a report that they r planning on a consumer version …but who knows how long it will take and how capable it will be …could be 2 years or whatever

    • Beatrice Beck

      Lyla previous six weeks l earned $19562 by working on my computer staying at my home in my free hours.(r1321) I’m able how to do it by working few time in whole day with my laptop. Realy it’s very easy & everyone definetly can join this work. >>> https://krowdy.co/9tMf

    • Nice consideration, Andrew

  • Looks very interesting… Also Karl is usually very critic… if he likes this, it means it is really a soild product!

    • Cassie G. Boruff

      I basically earn about $21,000-$22,000 every 30 days through the internet. I got rid of my job after working for the same organization for a long period. I required trustworthy earnings. I was not researching for the “get rich overnight” home programs you see online. Those all are type of ponzi multilevel marketing plan wherein you have to first get prospective buyers and then sell a product to friends and family members or anybody so that they will be in your team. Working over the internet has many advantages like I am always home with the little ones. Here Is what I do> https://linkshort.co/SKd

  • Sven Viking

    Regardless of other considerations, this is very impressive tech.