Support NeoGAF

[Dreamwriter]

Amusing that to produce the ultimate Augmented Reality, you have to remove as much visible reality as possible.



Given the quote above, that seems to be way off the mark. Unless you whip that dimmer way, way down to serious mood lighting.

I was talking about AR, not the HoloLens. In fact, you cut off my very next words

Note, I am talking about the future of AR, not the current HoloLens - HoloLens' limited FOV and how it makes it holograms translucent, that wouldn't be cable of obscuring the view of your room. From what Magic Leap has shown, theirs might be capable of it.

 

[LukasTaves]

It takes more power because to fill a larger field of view they have to generate a more pixels. It's the same reason why you would need a more powerful graphics card to show a 4K image instead of a 1080p one.

I'm no expert on this but I have looked into it enough to understand the basic problem with the FOV. For AR to work you have to create and project an image. So where do you put those projectors? You could put them directly in front of your eyes but then that would block the view of the world and defeat the point of AR. Btw, VR doesn't have this problem because it is assumed already that the outside world will be hidden.

Anyways, for AR to work it has to project the images from the side and then somehow redirect the light to go in your eyes. There are two main ways to do this. They can use tiny mirrors but that makes the glasses very thick, and I believe heavy and fragile. The other way is to bend the light like a lens but there are physical limits on how much you can do this. Different colors of light bend at different rates. If you try the bend the light too much the colors start to separate. This is why there is a small FOV. You need to be able to bend the light more to increase the FOV.

There are many different techniques used to try to overcome these problems but they all have some kind of drawback. Some entirely new way is needed to get a large FOV with AR glasses and to the best of my knowledge MS is just using existing known techniques and thus are stuck with their limitations.

That was actually my point, you have one setup of projector/lens with a narrow field of view, but if you have multiple projectors placed at different spots another set of lenses could cover another area and so on.

Of course, that creates another problem on itself with overlapping images from the projectors, but if they can lit each pixel individually it could be possible to align the projected images perfectly.

That in no way means it's feasible today, just the precision to have all that setup perfect aligned could cost a fortune, then there's the size of the setup.

But I don't think it's a matter of waiting a new tech to present itself, or come up with a whole new solution to the problem.

 

[panda-zebra]

I was talking about AR, not the HoloLens. In fact, you cut off my very next words

Similarly, I was talking about AR and not hololens.

 

[pixelation]

That's... not good.

 

[Dreamwriter]

Similarly, I was talking about AR and not hololens.

It sounds like you were talking about display technology. You realize AR can be done with normal LCDs or OLEDs, right? Heck, people have done AR with Oculus Rift DK2 by attaching two cameras and a depth sensor to it, so the user sees the real world with things added to it. Magic Leap uses an entirely different display technology than HoloLens, it uses light fields, which mimic how actual light enters the eyeballs (thus allowing you to do things like focus on an AR object just like you focus on other objects). Since your eyeballs can see dark colors (and light field cameras are able to take pictures with black in them), it seems like HoloLens would be able to as well.

I'm not totally sure about that, since the Magic Leap demo didn't have anything in it that was black, but it seems like it would make sense.

An example of AR done with an Oculus DK2

 

[Opus Angelorum]

I think this has a very narrow chance of success.

 

[pixelation]

I think this has a very narrow chance of success.

lololololololol

 

[Soi-Fong]

And what's horrible is I've read that it's literally physics itself that is preventing them from expanding the FOV further. The refractive index they're using is at the very limit of what's possible with glass.

 

[visiblemode]

They sold this tech on it's potential, without letting people know that 1st gen wasn't even going to get close.

I just don't know if this tech will ignite real interest with consumers until it's fully immersive. Seems to me like something they should have waited on, until they had it perfect.

 

[Alo0oy]

AR is a far better technology than VR in my opinion, I don't think I'll ever spend extended amount of time behind a VR headset, but I can see myself using AR for a long period of time because it doesn't take you out of the real world.

The problem is that MS kept lying about it, & now I'm really bummed out that we won't see anything like it in years, damn it MS.

 

[Grover]

Took out a tape measure and can confirm this guy's method of describing the field of view is accurate if you're struggling to picture it:

 

[justsomeguy]

They sold this tech on it's potential, without letting people know that 1st gen wasn't even going to get close.

I just don't know if this tech will ignite real interest with consumers until it's fully immersive. Seems to me like something they should have waited on, until they had it perfect.

It's not a consumer product.

 

[wondermega]

I'm a developer on a competing product. There's tons of issues with AR aside from just the fov, which (that particular case) you get over fairly quickly.

All things considered it's just early days, but the race is on. I'm very thrilled to be working on this stuff and having run the gamut over the years (pc, console, mobile development) there is really nowhere else I'd rather be.

I do understand the fear and frustration though. People have expectations, and being let down by bull shots/ etc is not an uncommon thing in this industry. Just sit back and watch as it develops, don't be too quick to judge with AR/VR and other new tech yet, until it's out for a bit, you've seen it yourself, and it has had a bit of time to mature.

 

[Zaph]

It's not a consumer product.

Would have been nice if MS went with that rather than having it's major debut on a consumer-focused stage show, with a bunch of consumer-focused entertainment demo's, using popular consumer brands like Minecraft and Halo.

It's almost like this problem wouldn't exist, expectations wouldn't be inflated, and the incredible work of that team wouldn't be overshadowed by what the tech can't do.

 

[DeepEnigma]

It's not a consumer product.

They should say that whenever they bring it on stage for an Xbox conference.

 

[Alx]

Would have been nice if MS went with that rather than having it's major debut on a consumer-focused stage show, with a bunch of consumer-focused entertainment demo's, using popular consumer brands like Minecraft and Halo.

Hololens debuted at CES 2015 (January), and was further demoed at Build conference (April). E3 was only the first time most gamers got aware of its existence, but it was already 6 months "old" and well known of all relevant actors of the industry.
But I do agree that showing it there wasn't a great idea if they have no real gaming prospects for it.

 

[Mr. Giggles]

I'm a developer on a competing product. There's tons of issues with AR aside from just the fov, which (that particular case) you get over fairly quickly.

All things considered it's just early days, but the race is on. I'm very thrilled to be working on this stuff and having run the gamut over the years (pc, console, mobile development) there is really nowhere else I'd rather be.

I do understand the fear and frustration though. People have expectations, and being let down by bull shots/ etc is not an uncommon thing in this industry. Just sit back and watch as it develops, don't be too quick to judge with AR/VR and other new tech yet, until it's out for a bit, you've seen it yourself, and it has had a bit of time to mature.

What is the competing product? I wasn't aware that there are other AR headsets being produced

 

[Darth Smurf X]

They sold this tech on it's potential, without letting people know that 1st gen wasn't even going to get close.

I just don't know if this tech will ignite real interest with consumers until it's fully immersive. Seems to me like something they should have waited on, until they had it perfect.

Am I in a Project Natal topic? -checks thread title- Nope. Just history repeating itself.

 

[xxracerxx]

What is the competing product? I wasn't aware that there are other AR headsets being produced

10 Forthcoming Augmented Reality & Smart Glasses You Can Buy

HoloLens isn't even listed on there.

Another: Augmented Reality Glasses: What You Can Buy Now (or Soon)

 

[nOoblet16]

And what's horrible is I've read that it's literally physics itself that is preventing them from expanding the FOV further. The refractive index they're using is at the very limit of what's possible with glass.

Virtual Retinal Display can't come soon enough !

I think it might be possible to use this with AR
https://www.youtube.com/watch?v=lsFMjMt_oGo

 

[Stare-Bear]

Also, it costs $3000...

 

[nOoblet16]

Also, it costs $3000...

As it has been mentioned multiple times, it's $3000 for the devkit. There is no consumer version in the market.

 

[wondermega]

What is the competing product? I wasn't aware that there are other AR headsets being produced

Daqri Smart Helmet, as mentioned there are several AR-enabled devices hitting the market in the near future.

 

[Raist]

No assuming needed. All hands on reports have the tracking and smoothness of the overlay as being quite nice.

Any report saying "yep, it's exactly like what's shown with their custom camera'?

 

[EventHorizon]

That was actually my point, you have one setup of projector/lens with a narrow field of view, but if you have multiple projectors placed at different spots another set of lenses could cover another area and so on.

Of course, that creates another problem on itself with overlapping images from the projectors, but if they can lit each pixel individually it could be possible to align the projected images perfectly.

That in no way means it's feasible today, just the precision to have all that setup perfect aligned could cost a fortune, then there's the size of the setup.

But I don't think it's a matter of waiting a new tech to present itself, or come up with a whole new solution to the problem.

That still leaves the problem of where you put the projectors. In order to be able to see the world, no projectors can be in front of your eyes, but you would need the projectors to be there if you wanted to reduce the angle that was needed to bend the light. You can't get around the fact that the projectors have to shoot the light in from the side and then somehow redirect the light into your eyes. That simply creates too big of an angle needed to bend the light in order to create a large FOV.