Magic Leap One Creator Edition Costs More Than $2,000

Status
Not open for further replies.

This isn't VR, but rather AR. As in, being able to view your surroundings with overlaid 3D objects drawn into them.

Even so, I get the impression that the Magic Leap One is going to be a failure, at least in this initial offering. From what I've seen, the overall design seems pretty bad. To start, just look at the thing. AR is intended to be worn for extended periods throughout the day, but you're not going to want to wear this. It's not like a pair of eyeglasses, or even something like Google Glass that can kind of blend into the background. The Magic Leap One is bulky, and looks like the most awkward pair of coke-bottle glasses ever.

Even beyond how silly it looks, and the fact that almost no one will want to wear it in public, it's also likely to be uncomfortable to wear for extended lengths of time. In addition to being a bulky piece of plastic strapped around one's head, there's also a thick cable running down to the processing/battery box strapped to the wearer's waist. And perhaps most importantly, it's going to have a narrow field of view. Not just the limited area where overlaid objects can appear, but even one's real world surroundings will be significantly obstructed around the edges due to the depth of the thing. It will be like wearing a pair of blinders, blocking out your view of everything that is not directly in front of you. Apparently, the lenses will block a majority of passing light as well, making them like wearing a pair of sunglasses indoors.

VR headsets can get away with being a bit bulky, since they are intended to be worn for relatively short periods of time in a confined space, and their entire point is to completely replace your view with that of a virtual world. AR, on the other hand, is intended to be enhancing your existing view of the real world, and should not be significantly impeding it. Magic Leap One appears to completely fail at delivering AR that people might be willing to wear for more than short bursts, so it won't be getting used throughout the day to augment one's daily life. And for entertainment purposes, a proper VR headset would likely provide a far better experience, especially considering how weak the onboard graphics appear to be in this device. Maybe it will find some use in certain business/industrial settings, but as a consumer device, this initial product looks like a flop. The price is also pretty absurd, and is unlikely to attract many developers. Perhaps they know that the device has major issues, and are just trying to cut their losses, knowing that it will never take off in its current form.

As for "sending VR to the grave one expensive device at a time", VR already has a number of reasonably priced devices available. There are plenty of Windows Mixed Reality headsets with controllers available in the $200-$300 price range, and the Oculus Rift with touch controllers has been $400 since last year. I do think some devices like the Vive Pro are overpriced though, at least as consumer products, and it would be nice to see a new generation of VR headsets priced in the sub-$500 range.
 

bit_user

Polypheme
Ambassador

Those guys have probably burned through most of their funding and can't afford to eat such a big subsidy. They probably figure the pricing is in line with what MS charged for Hololens dev kits, so what's the big deal?

That said, they'd be smart to setup a research grant program, or something, to get subsidized units in the hands of top university researchers and maybe a handful of entrepreneurs.
 

bit_user

Polypheme
Ambassador

That's where AR has been making its niche.

I'm willing to wait for the reviews. I agree that this isn't really going to catch on among the general public, but I don't know if that's even their plan.
 

The thing is, they seem to have been pushing this as a consumer device, or something you would wear throughout the day, from the start. Just look at some of their early CG concept trailers, which were made by special effects companies. Then compare those to what the actual experience was like in some pre-recorded clips they showed off last month...

Expectation? https://www.youtube.com/watch?v=kPMHcanq0xM

Reality! https://www.youtube.com/watch?v=bT5LFe-muR8

They posted a trailer yesterday showing off more of these smartphone-quality AR gaming experiences...

https://www.youtube.com/watch?v=K5246156rcQ

Along with this thing, which at least manages to look somewhat interesting...

https://www.youtube.com/watch?v=5iTLN3AuBws

But what these videos don't show is the actual device's very limited field of view. Aside from blocking off the peripheral view of your actual surroundings, the area where overlayed graphics appear covers a rectangular region just 30 degrees vertical by 40 degrees horizontal, according to a leak. That's a little larger than what the two and a half year old Hololens offers, but still only a tiny fraction of the field of view found in current VR headsets...

https://venturebeat.com/2018/07/31/magic-leap-ones-field-of-view-leak-signals-another-ar-disappointment/

And without a reasonable field of view, most of these entertainment experiences are going to suffer. That interactive audio-visual experience actual looks kind of cool in the trailer, until you realize that objects are going to be cut off at the edges of that rectangular region. And ultimately, what benefit is there to any of these experiences being for AR, rather than VR? It might initially be kind of novel having little characters running around on your real-world coffee table, but in what way is that better than having them on a virtual coffee table, or in some far more interesting environment?
 

bit_user

Polypheme
Ambassador
You guys seem awfully eager to hate something you haven't tried. What about at least holding back and waiting for reviews?


I think they're just SDK examples and demos. It's not like those are the only things you can do with AR.


If you don't believe in the value proposition of AR, then just stop. If you want to see what's possible only with AR, check out some of MS' more recent Hololens demo videos.

You might be ready to litigate AR, never having tried it (?), but you'll have to go there without me.
 

bit_user

Polypheme
Ambassador

The same could be said of the Hololens dev kit. Apparently, it didn't fail epically enough, because they're about to launch a successor.

In fact, it's a good thing you weren't calling the shots in the early PC industry, or else we might not have one. A lot of new technology is clunky, at first.
 

I would be more willing to cut Magic Leap some slack if they weren't constantly exaggerating their device's capabilities. Just look at this recent interview with their founder and CEO, which contains some "interesting" claims...

https://venturebeat.com/2018/08/08/magic-leap-ceo-interview-for-2295-start-living-10-years-ahead-of-everyone-else/

Our Lightpack, that little computer that sits in your pocket, has the power of a notebook computer that people might buy for more money than our entire system.
To start, I'm not sure what kind of pocket they're expecting that to fit into. It appears to be several inches (around 10cm) in diameter, and at least an inch and a half (4 cm) thick including its battery. Maybe some kind of coat pocket? I suspect that the device would overheat if you actually managed to stow it inside a pocket while it was running though.

But more importantly, the claim that it offers the power of a notebook computer costing more than $2,300 is obviously not going to be true. At that price level, you would be looking at systems with mobile i7 and GTX 1070/1080-level hardware, and that's certainly not happening in their device's "pocketable" enclosure. So they're obviously being deceptive about their product's actual capabilities. Judging by the demonstrations they've shown so far, it seems to be more around the power of a tablet, or a low-end laptop at a fraction of the cost.

Then we have another computer in the Lightwear, the thing you wear on your head, that has computer vision and a little baby Jarvis AI and sensing that you can program. You find those things in satellites and self-driving cars that cost tens of thousands of dollars.
And it's also probably fairly similar to the inside-out tracking found in a $250 Windows Mixed Reality headset. : D

People are so used to their notebook computer screen. You pay more than a Magic Leap One for a 15-inch screen. You pay tons of money for a phone with a 6-inch screen. We could fit hundreds of phones into our volume. We could fit lots of televisions and full-size dinosaurs and full-size cars. It’s an entire mental phase shift. What they end up comparing it to — your brain and eye get the entire real world at natural eye resolution. When you start chasing that, you enter a completely different paradigm.
That is an impressive way of skirting around a question about limited field of view. And a 15 inch screen costs more than a Magic Leap One? Where does this guy buy his computers? >_>

In a few months, you’ll be able to download the full Dr. G’s Invaders. That’ll be a full triple-A style game. We’ve been showcasing that for people here. The Weta team will have a launch date for that. It’s completely amazing.
This is actually somewhat interesting, in that it looks as though the movie special-effects company that did that CG concept trailer with the robots has actually recently expanded into developing actual video games, or at least one game based on that concept for Magic Leap...

https://variety.com/2018/gaming/features/weta-gameshop-dr-grordborts-game-1202870315/

From the sound of it, the game will likely be more of a wave-shooter though, perhaps with some story elements and other features. It probably won't look on-par with the concept trailer, but might look better than the demos for other software shown so far, seeing as it's apparently been in development for years by a company known for producing impressive visual effects in films, though it sounds like it will be their first game release. And my guess is that calling it a "triple-A style" game is probably also going to be something of an exaggeration. There are only so many ways you can shoot at robots and things inside your house. Perhaps it will show off other potential uses of the platform though, a bit like Valve's "The Lab" demo, but maybe a little more fleshed out.

They did specify here that they were focusing on selling the Magic Leap One to "creators and enterprises", but I'm not sure that goes along with much of what they've been showing. Perhaps they're simply trying to set in place the framework for a more consumer-oriented product a few years down the line. I'm not so sure that many "creators" will be interested in a product where very few people will be able to experience their creations though.

If Magic Leap felt their device was able to stand on its own, they wouldn't need to constantly exaggerate what it can do. I could get more behind it if it seemed less like they were trying to con people into thinking their device is something that it's not. In general, they have a trend of describing Magic Leap as if it's the most amazing thing ever, and something completely new and innovative, when in reality it seems to be little more than another Hololens with somewhat better screens and processing capability in its current form.

I do think AR has potential for consumers, but in order to realize that potential, it has to be in a form that people are willing to wear for extended periods of time in public. Magic Leap one doesn't seem to be that at all. The same goes for Hololens, but I would argue that device at least looks a little less awkward, and doesn't impede one's view of their surroundings as much. The Project North Star design from Leap Motion (the similarly named hand-tracking company) at least offers the potential for much lower prices than any current AR headset, while offering a much wider field of view. It will be interesting to see if any manufacturers make use of that reference design.
 

bit_user

Polypheme
Ambassador

Agreed. The TX2 is roughly on par with a GTX 1030 or maybe Intel's Iris graphics.

CPU-wise, it's only dual-core, although they're Nvidia's custom Denver cores. Still not on par with Intel's U-series, however.

These are embellishments that, while inaccurate, I can see past as they're not selling a notebook. IMO, its compute and graphics horsepower only need to be sufficient to power the software and experiences the device provides. IMO, comparing it to a laptop is just unnecessary.



No, it definitely solves a harder problem. Windows MR is used within something like a 8' square. Inside-out tracking, when you can walk an arbitrarily complex path and then return (possibly by another route) poses additional issues, such as loop closure.


https://www.apple.com/mac/compare/
MacBook Pro 15" starting at $2399.


I can believe its development cost as much as a AAA game. Having a movie VFX studio produce games certainly seems like a good way to burn through a good chunk of their funding.


They're trying to build a platform. They don't have the time or expertise to build medical visualization apps or other things that might compete with what they ultimately want their partners to build for it. I think these demos you're seeing are mainly are just to test, stress, and demonstrate the APIs for the benefit of the initial app developers. I would take them with a grain of salt, with perhaps the exception of Weta's game which, like first-party launch titles from big console makers, should give a pretty clear idea of the platform's full capabilities.


They're just trying to keep people interested and engaged, over its super-long development cycle. Part of the CEO's job is to get out there and generate interest. Sometimes, they can overdo it.

Where I'm at is: hey, it's finally launching! Just wait until we can get some reviews or actually try it!

You're forgetting a pretty major point: light fields! This sets it apart from every other AR device on the market. Hopefully, light fields will even make it into VR, before long. I care a lot more about light field displays than about Magic Leap or any other single player in the VR or AR market. I just want to see/experience the tech and have it gain traction.

The thing is, if he just goes out there and start waxing about light fields, people's eyes will glaze over and they'll tune out. It's something you have to experience, in order to appreciate.


It doesn't need to be. It just needs to be good enough to either carve a niche for them and sustain them until either v2 or v3... or until they can get bought up by Apple or somebody that can continue funding them, while they refine it into a more consumer-friendly form.
 

But comparing the price to satellites and experimental cars? And while the Magic Leap One's inside-out tracking might very well be more advanced than that in a Windows Mixed Reality Headset, it's still likely of limited use in this initial device. They have already said that it is intended for indoor use, and will have problems tracking outdoors. Even indoors, I don't get the impression that many will find it comfortable enough to wear throughout the day, so traveling from room to room with it might not be that common.

Also, while the software is not really designed for larger-than-room-scale tracking with a Windows Mixed Reality headset (since that would not be practical for a tethered VR device) the hardware uses similar tracking to Hololens, and should at least be able to handle larger volumes than an 8 x 8 ft square. Here's a video where someone set up a 30 x 30 ft area in a conference room...
https://www.youtube.com/watch?v=eKd8-ng2qVE

There are some examples of people testing larger outdoor areas too, albeit with mixed results, since things like direct sunlight and moving foliage seem to confuse the tracking system. That's apparently the case for the Magic Leap One as well though. Here's a video showing a WMR headset being used on an athletic field, and at least the tracking works pretty well in this instance (There are English subtitles)...
https://www.youtube.com/watch?v=6brHtMGkHPQ

The tested games, however, don't always behave correctly outside an area much larger than the previous room, which seems to mainly come down to the developers not designing their software with such large tracked volumes in mind.


He said "You pay more than a Magic Leap One for a 15-inch screen", not that you can pay more for a particularly high-end laptop that happens to have a 15 inch screen. Its a bit like the satellite comparison, in that while examples may technically exist, they don't necessarily make for relevant comparisons when discussing a product's cost. And it's not even really important to have a laptop or phone screen filling your field of view, since they are used in a different way. You don't have large chunks of virtual objects in your environment getting cut off until you center them in view and stand back far enough. If you're playing a game on a laptop screen, that game's field of view doesn't need to match the field of view of the physical screen, so it's not a particularly good comparison.


I'm not really forgetting about them, but from what I've heard the capabilities of the light-field tech in the Magic Leap One have been greatly scaled back compared to the non-portable prototypes they were showing investors and media outlets years ago. So, while it could be a big improvement over what the competition offers, for now I will only consider them as having "somewhat better displays" than a 2.5 year old device.


Sure, a future version could be great, but they're selling this version now, and for a relatively steep price, for what is essentially a devkit that will likely have very little software available. They should be trying to get the thing into developers hands at a relatively low cost of entry, not charging a big premium for hardware that will probably be obsolete within a couple years. And again, that comes back to the CEO's recent claims about the device being priced "like a high-end tablet". If they can't be up-front about simple details that are easy to verify or disprove, then where else are they stretching the truth? Maybe the device will be great, but the way they talk about it doesn't instill much confidence.
 

bit_user

Polypheme
Ambassador

You compared it to Windows MR, so I just pointed out something that's different. The issue isn't one of room size, so much as you use it in an area and leave. Then, you return (possibly by a different route) and it must still recognize the room and accurately calibrate to it, so that your virtual objects are exactly where you left them.

Yes, this is something Hololens should do, but that's priced similarly to this kit.


I noticed you cut out my quote where you asked:
And a 15 inch screen costs more than a Magic Leap One? Where does this guy buy his computers? >_>
I was literally just answering your question. Mac users probably think that's just how much a 15" laptop costs.


How do you even know? Did you try it? Did you find a review of it? If not, why are you trying to say how good it is or isn't?


Yes, it's a dev kit, with little software available. Just like Hololens, which is priced similarly (in spite of much lower hardware specs). Did you have the same complaints about it?



You're just nit-picking. Why are you trying to litigate against it, when reviews are finally just around the corner? Why not wait until they actually reach people's hands?

You don't have to like the CEO. I don't like Nvidia's Jensen Huang, who is also no stranger to spin and misleading statements, but I don't deny they make good products.

All I'm saying is that instead of trying to litigate how he spins their product, let's just see about the actual results. If the lightfield implementation doesn't provide any practical benefits over Hololens, then we can consider this generation to have missed the mark. And not on the basis of how it was spun, but on the actual result.
 
Status
Not open for further replies.