Magic Leap's AR Demo Video

First time accepted submitter iMadeGhostzilla writes TechCrunch reports: "Magic Leap is showing what it might look like to use its hardware for augmented reality gaming in the future, with a new demo of what the team is apparently 'playing in the office' right now. The brief video shows examples of interacting with YouTube and Gmail apps, along with browsing a menu system for OS-level interaction. The person in the video from whose perspective it's apparently shot then selects a shooter game, tests out a weapon after choosing from a variety of options, does some tower-defense style stuff by placing a current and fights some visually impressive but fairly generic baddies. [...] The video was posted with an apology for Magic Leap's absence at TED." Commenters on reddit and elsewhere believe the video is fake. Magic Leap recently came into the spotlight with its recent $540M backing by Google and others.

Comments

It's not a real demo

[Anonymous Coward]

Save this junk for gullible VCs.

First person shooter?

[ArcadeMan]

Come on, we all know what we want: first person hookers.

Fake, not practical

[Dan East]

First of all, of course it's fake.

The biggest flaw is the interface itself. A full 20 years ago, when Johnny Mnemonic (the movie, not the short story) came out, and I saw Keanu Reeves using that VR internet type access, I knew that those interfaces would never work in the real world. I'm sitting here browsing the internet, typing this message, etc, by moving my fingers maybe an inch most. We are currently stuck having to do physical interactions to interface with a digital world. As long as we are stuck using this kind of interfacing (IE it's not plugged directly into our mind), then the less physical effort is required to interface, the more efficient, faster, accessible, and convenient computers will be to access.

Really, I'm going to lift my arm up to head level, and make a huge gesture like I'm pulling laundry out of the washing machine, to look at a message? LOL Suuuuuure. Anyway, you can spot it in a minute when these interfaces are generated by animators / artists / movie effects people, because they could never gain widespread usage when the majority of people would be fatigued after a mere 10 minutes using such a system. Kind of like every movie with a computer makes beeps and blips every time you interact with any widget, which in reality would drive the average human out of their mind in exactly 20 minutes. Yeah, looks cool as a prop, but annoying in real life.

Oh, and I like the way that gun somehow turns into an actual prop in that guy's hands. They've invented a transporter and holodeck to go along with their VR headset apparently.

Re:Fake, not practical

[Dan East]

Here's there scene I'm talking about.
https://www.youtube.com/watch?...

This new Magic Leap video... same poor interface concept, slightly better graphics.

Re:Fake, not practical

[nathan s]

I'm reasonably sure that the gun isn't supposed to be "AR." I'd say it's an actual prop that their AR system is presumably "recognizing" and generating specs on, like other systems we've seen that use game-specific props for AR through cell phone cameras or whatever. That said, who knows how much of the rest is "real" - without some way that the entire room is part of the AR system, there's no way it should know to avoid rendering the lower part of the robots behind that divider...that's the obvious big question mark.

Re:Fake, not practical

[Anonymous Coward]

In theory you could figure out where and how high the divider is by generating a depth map with a laser (like the Kinect does). The only problem then is that you have to then make a lot of assumptions about it's depth, and the height of the floor behind it etc. You could also input the office geometry into the game beforehand, but then if things like chairs are moved the game wouldn't work right any more.

I've done a fair bit of VR and augmented reality development, and I didn't see anything that is impossible in theory. The hardware to do it would have to be a lot better than the consumer stuff we have at the moment though.

Having said all that this video looks way too much like it was made by a bunch of animators. If they had anything even remotely approaching a real prototype they would probably be showing off something a lot more convincing.

Re:Fake, not practical

[nathan s]

Right. I wasn't saying that any of it is impossible, just looks like a bit more is involved than a headset.

Re:Fake, not practical

[Bite The Pillow]

It's fake because my Xbox 360 can basically do the same thing, other than the dual display, which is supported in a few select games, so it basically can do all of that?

My last generation, not current generation, non-upgradeable hardware, from 2010, can already do pretty much this.

Remember, they are playing with this in the office. This isn't the 8-to-5 interface that they use to check e-mail. 15 minutes of physical activity every two hours is good for you, and slightly above that is better.

I'm sure that I would enjoy this, if I were developing it, meaning I did hardware or software most of the day, or if I were testing, meaning I did a lot of documenting in addition to the physical interface.

In 8 hour doses, this is going to get tiring, and they will refine it before release, or it won't be successful. Few people want AR/VR, and doing it wrong will kill the whole thing. When they get to alpha testing, the gestures or whatever will have to be refined.

I expect some sort of neural supplement, so that a modified keyboard plus neural reading plus large gestures for the big things, along with head tracking and eye tracking, is going to kill every other everything.

Re:Fake, not practical

[citizenr]

oh, so the magic AR system is making prop gun recoil with ... alpha waves?

Re:Fake, not practical

[epiphani]

Actually, I can believe it. It's a demo, probably very carefully planned - but I can believe the tech.

First, the practical applications of the interface are not the primary concern. Granted, waving your arms around isn't a good interface - but that's not the point. The point is the overlay - positionally aligning 3D objects in the field of view. Having a way to interact with them is also useful, but not in a day to day sense. Keyboards and mice will still win for the standard type of interactions.

Example: if you're walking in the airport on the way to your flight, with handy personalized directions floating in three dimensions guiding you along the way and you get a phone call and choose to ignore it? A quick wave in front of your face to clear the notification is nonintrusive and simple - and you don't look like a putz doing it.

Think about what Oculus is capable of. Then add in infrared mapping a la kinect or a similar technology. And make the overlay transparent, rather than a straight LCD.

Also, that gun was sitting on the desk the entire time. It's a prop, and the system recognizes it. I'd bet it's also an input method, with a trigger if not other inputs.

Re:Fake, not practical

[aliquis]

Yeah.

It's almost as horrible as standing desks.

No-one would use those.

Re:Fake, not practical

[null etc.]

This type of technology will be the reason why geeks finally develop a genetic predisposition to be physically fit.

Re:Fake, not practical

[recharged95]

Also having Xerox doing real $$$ R&D on UIs/HMI/GUIs on physical objects... makes sense that the current mouse, keyboard, screen are pretty much 98% efficient for 98% of the population. Considering that R&D was done nearly 40yrs ago.

All this VR stuff is maybe 5% efficient for 5% of the population. It's a start, but guys, don't sell it now as the solution to everything.

virtual money buys augmented clicks

[turkeydance]

profit!

"might look like" vs "We're doing this, now"

[QuasiSteve]

I'm sure Magic Leap's tech is wonderful. But I can't help but be unimpressed by any "What it might look like" videos.

Compare the video in the article with that from the March 18 Project Tango story's video which doesn't show "what it might look like", but what they've got working right now.
http://hardware.slashdot.org/s...

That's a lot more impressive to me than any video with a bunch of VFX.

Great Idea!

[Irate Engineer]

Let's have someone running around the office wearing goggles and brandishing a weapon at invisible beasties falling through the ceiling.

Don't see a problem at all with this.

You can tell it's fake because...

[david.given]

...nobody in real life could afford four of Dr Grordbort's Infallible Aether Oscillators. (Although the one they're playing with in the video looks like the Righteous Bison, which is the budget model at a mere $100. Probably because when they were filming it they were afraid it'd get dropped.) http://drgrordborts.com/produc...

Take The 10 E. to I-95 S. to Exit 26

[Anonymous Coward]

The bigger message here is that one can, indeed, live in a place that has a 0% personal income tax and no winter, and still do cool stuff.

Magic Leap is in the Design Center of the Americas, just south of Fort Lauderdale/Hollywood International Airport, and Miami—20 miles to the south—has a growing tech startup scene. If you can program, and you are tired of living in a place that is 67% male, South Florida beckons!

Weta

[pieisgood]

A video showing how it "could be" created by weta workshop. Yeah it's going to look nice, weta is a top notch shop, but i'll wait till something real is shown before being impressed.

Re:Fake, not practical Lawnmower Man

[asjk]

Really, I'm going to lift my arm up to head level, and make a huge gesture like I'm pulling laundry out of the washing machine, to look at a message?

Sounds like something that was on Community recently

AR has to be lower quality than VR

[abies]

If you can create AR device with better quality than dedicated, no-see-through VR, then just slap black box behind the AR googles and you have created better VR. So from very definition, AR is at most as good as VR and in real life (object detection, depth detection, variable lighting etc) is expected to be at least a bit worse.

This is why when I see concept demos which are way ahead of current Oculus/etc offerings and nobody mentions taking over VR device space, but instead touts AR gimmics, I just don't believe it.

If you have killer retina projector, put it into VR and start selling. Then spend another few years on getting all extra AR complexities solved.

Re:Oh dear ...

[gl4ss]

Let me guess, you're either a frail little twig or a barely mobile lard ass.

Try exercise and diet, then you'll be able to move more comfortably and without losing your breath.

feck off.

the problem with majority of these virtual reality companies and tech is that they try to be too much.

what they need to break into big market is VR gaming.

you know what's the best way to actually play for 3 hours with an oculus rift? a fucking space mouse? kinect? what the fuck. of course not. best way is to sit on sofa with either controller or sofa compatible mouse and kb. essentially you're just going to replace your monitor with the thing - and no it's not the same still as using a hmz-1 or something like that. the immersion is vastly superior and the head movements should work - BUT the head movement should be just something optional. like, if you need to use your head movement even for up/down looking it gets really tiring and aiming becomes a chore. good example about this is how hacked in vr support in hl2 was better than actual "virtual reality support" in tf2, because they tried to make too many modes in the official vr support. just having mouselook functions perfectly - and the head movement then just mapped another mouse. does this disorientate people? probably those who can't play fps's anyways - as a feeling it's not that much different than looking at a screen and using mouselook. the point however is that it doesn't tire you so much.

full movement virtual reality for home games is a wii motion control like gimmick - you just can't make it work and the disconnect is so huge anyways, you can't have tactile feedback anyhow.

just make a high res 90 fov+ head unit and it will sell like hotcakes. console and pc gamers will buy it in troves and just use their old desk or sofa and controls, trying to turn the room into a minority report joke ui is a bad idea(virtual high resolution monitors, 360 degree desktop and such not so bad idea, but would be interacted with kb and mouse finely.

if waving the hands was a good ui for day to day work, then we would be using something quite different from the old typewriter style we have.. the typewriter is better than the pen.

Re:Oh dear ...

[Anonymous Coward]

I've played games such as Lazer Tag and paintball for decades. If the idea of holding a gun while running around and shooting at other players is too tiring for you, then you are in bad shape and gaming should be the last of your priorities.