This device is a volumetric display, similar in product placement (if not exactly the same illumination technology) to the volumetric dome display from Actuality Systems about 20 years ago.
Volumetric displays have their place, but they can't display general occlusion of far objects by near objects. That restricts their application to non-photorealistic scenes that often look like clouds of points.
Since occlusion is one of our strongest depth senses (much stronger than stereopsis), that's a significant restriction.
Big spinny things are also hard to scale.
While other autostereoscopic display technologies like the parallax barrier displays from Looking Glass Factory give up the ability to walk around the scene, they work with no moving parts and can display photorealistic scenes (either synthetic or photographic).
Again, different display technologies have their place, but volumetric displays have historically struggled to compete in the 3D display market.
So, they show mostly correct applications: CAD / 3D-modeling work, schematic visualizations, etc.
Also, for an individual seated user, it's relatively easy to adjust the direction of gaze, and do occlusion at the rendering level, giving a rough idea of occlusion. They even show it in one of the segments of the video on the product page (with some color balls).
Sure, it's possible to approximate occlusion for a fixed viewer. At that point, though, the display is competing with displays like looking glass (lenticular, parallax barrier, or stereograms) that are higher resolution, cheaper, and have a rendering pipeline that's more compatible with standard computer graphics.
Occlusion could be done with head tracking for one observer (not both eyes though); but defeats the point. I'm guessing AR glasses will work for much of this.
About “can't display general occlusion”, wikipedia calls this a misconception. But it does later state that it requires sacrificing vertical parallax.
> It is often claimed that volumetric displays are incapable of reconstructing scenes with viewer-position-dependent effects, such as occlusion and opacity. This is a misconception; a display whose voxels have non-isotropic radiation profiles are indeed able to depict position-dependent effects.
> the ability to reconstruct scenes with occlusion and other position-dependent effects have been at the expense of vertical parallax, in that the 3D scene appears distorted if viewed from locations other than those the scene was generated for.
"non-isotropic radiation profiles" is probably better expressed as the ability to produce light that can be modulated differently in different directions. That's what's required to show proper view dependent shading on a surface, and what's required to turn off a voxel that's behind another voxel from a particular view angle.
That's arguably a different class of display, and an extremely complex one.
To build it you'd likely want to sacrifice vertical parallax (to allow the use of a spinning lenticular sheet or something else, and to reduce the information capacity required), but there's no fundamental reason that prevents a full parallax display of this type (a spinning integral photograph).
From an engineering complexity point of view, it's the hardest of all worlds.
I worked for people that were making these kinds of things (in 2009-10[0]) for advertising (8ft high, 6ft diameter spinning cylinder[1]; 8ft high, 4ft wide flat display with 8 interconnected spinning discs) and yeah, they were absolutely terrifying. Especially the flat display - sounded like a jet taking off when starting up and wasn't much quieter at full pelt[2].
[0] I suspect these things are easier now with lower power LEDS, etc. given you can cover a cylinder in directly addressable LEDs pretty cheaply and don't need the massive spinning death machine.
[1] Deployed at a few places around the UK but now sadly removed.
[2] They were intended for roadside displays, not shopping malls.
It's neat and all but I can't help being skeptical that, beyond certain narrowly specific use cases, volumetric displays like this will measurably increase actual utility enough to be worth the increased cost, size, weight, power and complexity. I'm assuming "utility" here to be something like "usefully actionable increased comprehension of a 3D object or terrain".
My reasoning is that the human visual system already has a bunch of neural wetware that's evolved to be really good at turning visual cues like highlights, shadows, reflections, specularity, depth of field, etc from a 2D scene into an internal 3D representation. When you add extra cues over time like object/light source motion, moving POV and parallax it gets even better. And all those cues are already "free" with common 2D displays + motion. Adding a bit of additional hardware to 2D commodity displays enables things like stereo binocular views from 3D glasses, head tracking and interactive control which further deepen scene comprehension at only slightly more cost, space and complexity. That's a lot of pre-existing, cheaper, easier alternatives that are as good as volumetric displays for most use cases and nearly as good for the remaining use cases.
Beyond a few specialty use cases, research labs and military trials, I suspect the majority of these displays will be deployed as a novelty to attract attention or as social signaling (eg trade show booths, high-end retail, corporate hospitality, etc). Unfortunately, those kinds of use cases tend to have a shelf life only about as long as the novelty and high early adopter prices last.
On one hand there are light field displays like https://lookingglassfactory.com/ not to mention plain stereoscopic displays as in the Nintendo 3DS and maybe someday synthetic hologram displays that do the same thing as the light field displays except using a wave interpretation of light as opposed to a ray.
Then there are various headsets such as the MQ3, Vision Pro (basically VR with some video passthrough) and the other kind such as Hololens, Magic Leap that have an optical combiner.
For that matter you can make a 3-d print of an object that you want to inspect.
Practically they stand or fall together on being able to exchange content, there is no "3d economy" unless I can make a model with some standard format and view it with all of those displays.
It feels like this is the barrier to that 3D economy. The width of an iphone is approximately the interpupil distance of an average adult, moving one camera to the other side would make stereoscopic image/video capture a widespread capability. It just lacks an application.
I was a massive doubter (like a few of the folks here) when I met co-founder Gavin 8ish years ago.
Unlike many of the other founders in our little startup ecosystem, he and Will have stuck with it long term, and I’ve watched from afar as this project has grown, technically and commercially to what it is today.
From the outset, this project appeared to be doomed. The earliest iterations of their product seemed far too limited to be of any use (The view area was hardly bigger than a large grapefruit, the enclosure was massive, and the resolution/clarity was … not great), and speaking of use, who would need such a thing?
Against all odds, they’ve prevailed. The hardware is way better, the integrations have improved significantly, and while I’m still not sure what their customer are using this for, clearly there’s enough revenue coming in that they can retain a team of software engineers (they were hiring recently too).
This is speculation but I wouldn’t be surprised if it was mainly or even entirely bootstrapped with the exception of one or two government/university grants.
They’re not even based out of a “good” city for entrepreneurship, Adelaide (as much as I love it) is not exactly known as a hotbed of innovation.
This product shouldn’t exist, and yet it does. It’s almost like they’ve willed it into existence. I’m not sure if it’s passion or persistence, but I have to admire how they’ve stuck with it, defied the odds, built a business and turned sci-fi into reality.
I’d bet good money that they’ll still be here, growing in 10 years time.
P.s. I don’t know if he’s still there but Ken Silverman was/is on the team (He's the creator of Duke Nukem 3D's Build Engine and is highly regarded by John Carmack)
Oh wow, I'm from Adelaide originally and didn't know they where based there too (in my defence, I haven't lived there in 10+ years). Maybe I'll try and drop in and say hello when I'm back.
It's a LED matrix spinning at 15 RPS. That's why animations are a bit jittery and why the center is always darker / not illuminated. That said, their examples with anti-aliasing look amazing. I'd say this will be a great tool for doctors to analyze x-ray / CT data.
> why the center is always darker / not illuminated
I don’t understand that part. Why is that? I’m familiar with the theory of persistence of vision displays, or so i thought. Wouldn’t the center be brighter denser rather than darker? If you have the same led density but the leds “move less” because of the lower radius that is what i would expect. What am I thinking wrong?
I believe there is a plate with LEDs on both sides that is rotating. When you look at that plate from the side, it obscures the stuff behind it. That's why you see more "shadow" the closer you get to the center. In this video, you can see the effect:
Around the 10 seconds mark, the camera rotates around the display and then you can see a glowing laptop screen move behind the VX2. And there you can see that the center of the VX2 has something blurry but solid which obscures the background. That's the rotating plate with the LEDs on it, I believe.
No, the enclosure keeps the air contained and moving at a constant rate with the armature. The only noise comes from the brushless motor, which is under 50dB at 1m.
Does comfort beat out detail? It seems like AR, of any stripe, can do better.
An LED matrix seems fairly fragile and very time-consuming to assemble. I'm not an expert. Could the price come down over time? I'm having difficulty imagining the market fit. I don't mean to sound too negative. It does look very cool! I'd love to see it in person.
As a v0.1 it seems great. Reminds me of the jaws ad in Back to the Future 2. Mall advertising is probably the only thing it is suitable for now. As it gets higher res I could see it replacing TVs.
>It seems like AR, of any stripe, can do better.
Only if everyone in the group has AR goggles, I could see this being better for group activities where everyone can be around and reference the same "thing".
Readit News
Volumetric displays have their place, but they can't display general occlusion of far objects by near objects. That restricts their application to non-photorealistic scenes that often look like clouds of points.
Since occlusion is one of our strongest depth senses (much stronger than stereopsis), that's a significant restriction.
Big spinny things are also hard to scale.
While other autostereoscopic display technologies like the parallax barrier displays from Looking Glass Factory give up the ability to walk around the scene, they work with no moving parts and can display photorealistic scenes (either synthetic or photographic).
Again, different display technologies have their place, but volumetric displays have historically struggled to compete in the 3D display market.
Also, for an individual seated user, it's relatively easy to adjust the direction of gaze, and do occlusion at the rendering level, giving a rough idea of occlusion. They even show it in one of the segments of the video on the product page (with some color balls).
> It is often claimed that volumetric displays are incapable of reconstructing scenes with viewer-position-dependent effects, such as occlusion and opacity. This is a misconception; a display whose voxels have non-isotropic radiation profiles are indeed able to depict position-dependent effects.
> the ability to reconstruct scenes with occlusion and other position-dependent effects have been at the expense of vertical parallax, in that the 3D scene appears distorted if viewed from locations other than those the scene was generated for.
https://en.m.wikipedia.org/wiki/Volumetric_display
That's arguably a different class of display, and an extremely complex one.
To build it you'd likely want to sacrifice vertical parallax (to allow the use of a spinning lenticular sheet or something else, and to reduce the information capacity required), but there's no fundamental reason that prevents a full parallax display of this type (a spinning integral photograph).
From an engineering complexity point of view, it's the hardest of all worlds.
Sure, if you replace each pixel with a tiny light field display pixel you can do occlusion.
I worked for people that were making these kinds of things (in 2009-10[0]) for advertising (8ft high, 6ft diameter spinning cylinder[1]; 8ft high, 4ft wide flat display with 8 interconnected spinning discs) and yeah, they were absolutely terrifying. Especially the flat display - sounded like a jet taking off when starting up and wasn't much quieter at full pelt[2].
[0] I suspect these things are easier now with lower power LEDS, etc. given you can cover a cylinder in directly addressable LEDs pretty cheaply and don't need the massive spinning death machine.
[1] Deployed at a few places around the UK but now sadly removed.
[2] They were intended for roadside displays, not shopping malls.
My reasoning is that the human visual system already has a bunch of neural wetware that's evolved to be really good at turning visual cues like highlights, shadows, reflections, specularity, depth of field, etc from a 2D scene into an internal 3D representation. When you add extra cues over time like object/light source motion, moving POV and parallax it gets even better. And all those cues are already "free" with common 2D displays + motion. Adding a bit of additional hardware to 2D commodity displays enables things like stereo binocular views from 3D glasses, head tracking and interactive control which further deepen scene comprehension at only slightly more cost, space and complexity. That's a lot of pre-existing, cheaper, easier alternatives that are as good as volumetric displays for most use cases and nearly as good for the remaining use cases.
Beyond a few specialty use cases, research labs and military trials, I suspect the majority of these displays will be deployed as a novelty to attract attention or as social signaling (eg trade show booths, high-end retail, corporate hospitality, etc). Unfortunately, those kinds of use cases tend to have a shelf life only about as long as the novelty and high early adopter prices last.
On one hand there are light field displays like https://lookingglassfactory.com/ not to mention plain stereoscopic displays as in the Nintendo 3DS and maybe someday synthetic hologram displays that do the same thing as the light field displays except using a wave interpretation of light as opposed to a ray.
Then there are various headsets such as the MQ3, Vision Pro (basically VR with some video passthrough) and the other kind such as Hololens, Magic Leap that have an optical combiner.
For that matter you can make a 3-d print of an object that you want to inspect.
Practically they stand or fall together on being able to exchange content, there is no "3d economy" unless I can make a model with some standard format and view it with all of those displays.
"The frame holds two P1.875 128x64 LED panels"
P1.875 refers to the pitch (distance between LEDs), seems to be a standard HUB75 panel.
Unlike many of the other founders in our little startup ecosystem, he and Will have stuck with it long term, and I’ve watched from afar as this project has grown, technically and commercially to what it is today.
From the outset, this project appeared to be doomed. The earliest iterations of their product seemed far too limited to be of any use (The view area was hardly bigger than a large grapefruit, the enclosure was massive, and the resolution/clarity was … not great), and speaking of use, who would need such a thing?
Against all odds, they’ve prevailed. The hardware is way better, the integrations have improved significantly, and while I’m still not sure what their customer are using this for, clearly there’s enough revenue coming in that they can retain a team of software engineers (they were hiring recently too).
This is speculation but I wouldn’t be surprised if it was mainly or even entirely bootstrapped with the exception of one or two government/university grants.
They’re not even based out of a “good” city for entrepreneurship, Adelaide (as much as I love it) is not exactly known as a hotbed of innovation.
This product shouldn’t exist, and yet it does. It’s almost like they’ve willed it into existence. I’m not sure if it’s passion or persistence, but I have to admire how they’ve stuck with it, defied the odds, built a business and turned sci-fi into reality.
I’d bet good money that they’ll still be here, growing in 10 years time.
P.s. I don’t know if he’s still there but Ken Silverman was/is on the team (He's the creator of Duke Nukem 3D's Build Engine and is highly regarded by John Carmack)
I don’t understand that part. Why is that? I’m familiar with the theory of persistence of vision displays, or so i thought. Wouldn’t the center be brighter denser rather than darker? If you have the same led density but the leds “move less” because of the lower radius that is what i would expect. What am I thinking wrong?
https://www.youtube.com/watch?v=2I0sjF9NuTU
Around the 10 seconds mark, the camera rotates around the display and then you can see a glowing laptop screen move behind the VX2. And there you can see that the center of the VX2 has something blurry but solid which obscures the background. That's the rotating plate with the LEDs on it, I believe.
An LED matrix seems fairly fragile and very time-consuming to assemble. I'm not an expert. Could the price come down over time? I'm having difficulty imagining the market fit. I don't mean to sound too negative. It does look very cool! I'd love to see it in person.
>It seems like AR, of any stripe, can do better.
Only if everyone in the group has AR goggles, I could see this being better for group activities where everyone can be around and reference the same "thing".
Either way, pretty cool, but I don't see it replacing TVs any time soon.
[1]: https://en.wikipedia.org/wiki/Surface-mount_technology
[2]: https://en.wikipedia.org/wiki/Pick-and-place_machine
I don't think there are only two options here. There are other approaches like Google Starline.
No. AR is limited to one focal plane (HUD instead of full 3D), unless you want to round-trip the entire image through the goggles.