Magic Leap is a Tragic Heap | Hacker News Reader

Magic Leap is a Tragic Heap

palmerluckey.com

114 pointspain_perdu39 comments

For anyone who didn't happen to notice, it should be noted that this review was written by Palmer Luckey, the founder of Oculus VR and designer of the Oculus Rift.

https://en.wikipedia.org/wiki/Palmer_Luckey

True. But his assertions are mostly fact based and support his thesis.

Who really thinks Magic Leap holds up to the hype and is really a good use of its investment capital?

Nothing could hold up to the amount of hype Magic Leap got, but based on what I've seen it's been an excellent use of its investment capital. They've been experimenting with new technologies for years now and they finally have a development kit that speaks to the path they plan to follow in the future.

I also wouldn't trust Palmer Lucky to have anything like an objective opinion about it.

> They've been experimenting with new technologies for years now

True.

> and they finally have a development kit that speaks to the path they plan to follow in the future.

...which is following the same path everyone else is, except further behind, because all those new technologies failed to work out?

I've been in AR for almost a decade now and I have always been in the "we really need Magic Leap to work" camp, as well as recognizing just how hard it really is to do AR right. I'm also not a VR fan, but appreciate how its moved the needle on these technologies broadly. That said, Palmer makes his best points with these quotes:

sucked all the air out of the room in the AR space...allowed them to monopolize funding in the AR investment community

That is unquestionably what happened, as, if you talked with any major VC in the last 5 years they would admit that they were waiting on Magic Leap to make a move before they would really think about seriously funding anything else. So in that sense, it's been a constant frustration.

I'm still hopeful that Apple, Microsoft, FB or Google can make something compelling, and I think ML has been generally positively received - however it's always caveated as "these are the early days." Which is true, but the hype for ML has been unfortunate.

Again, consumer AR - especially HMD AR - is exceptionally hard to do right, so my hat is off to ML and the amazing people they have. However even Rony Abovitz (ML CEO) agrees that they were arrogant in their approach, and I think the end result is that it's hurt the AR industry in the larger sense.

This blog is going right into my special bookmarks folder along with "Microsoft Surface is on Slippery Ground" by Tim Cook and "Ford? More like BORED!" by Elon Musk.

Like Michael Dell saying that Apple should be shut down and all the money returned to the investors?

Like that?

Links?

I can't find anything relevant by googling those titles myself.

I'm guessing that was sarcasm, meaning something along the lines of "would you believe an article about Microsoft by Apple's CEO?"

Been anticipating this teardown ever since this hackernews post [1] comparing the promises with the patents and nailing the prediction that ML1 would be yet-another LCOS headset.

Frankly I've been cheerleading their implosion ever since this farce of a TED talk [2] in 2012. An opportunity to present their technology and they jump around in monster costumes wasting everyone's time. Thwaxo's Strangely Demented Space Fudge ? This was months after securing half a billion in funding. Just left a really bad taste in my mouth.

[1] https://hackernoon.com/what-is-magic-leap-really-642e1660fcc...

[2] https://www.youtube.com/watch?v=w8J5BWL8oJY

I was blown away by Magic Leap when I tried it a couple weeks ago. I have no affiliation with the company and was skeptical beforehand. Maybe there are other things that are better, but it was really, really cool. It felt like a new kind of thing entirely.

What specifically blew you way compared to other AR headsets like HoloLens for example?

That’s the thing. I haven’t tried the others (except Google Glass 4-5 years ago). They might be better.

Let's assume for the sake of a conversation that this article is right, and Magic Leap (ML) is lackluster. What is it exactly that makes VR tech so difficult, then? ML had all the money in the world, and still (presumably) couldn't hit the mark. What exactly is missing - is it the material sciences for eye-projection systems? Ability to downsize the graphics and processing tech into a small package?

What critical milestones are still missing that, when available, will allow truly immeresive VR? And anyone dare predict when they will be available?

So, AR is harder than VR because of things like surface detection and occlusion. That being said, the problems with closed-headset VR are twofold: the price and the arid software landscape.

A Vive Pro headset, the most advanced consumer-ready VR headset, costs $800, plus a few hundred $ more for each of the base stations and controllers. On top of that, you need a good PC to run any VR software. That's a lot of money to invest.

And if you do, you're met with a bunch of weird experiments, a few weird experiments with some polish, and a handful of games that offer a truly novel experience in VR.

But because the barrier to getting into VR is so high, it doesn't make financial sense for most companies to try to make software for the space. And because there are too few games, there isn't a good reason to buy a headset, so the whole ecosystem goes full ouroboros.

So, I guess the thing that is missing is the price falling low enough that the VR ecosystem can become self-sustaining. Good VR systems totally exist; they just cost $4000 and you can't do anything with them, lol.

+1 for using the phrase "goes full ouroboros"

I don't think it's fair to use the Vive Pro as your benchmark for the price of VR. The Rift is equally consumer-ready and half the price, with all base stations and controllers included.

Michael Abrash (of Oculus research) has given several good talks that discuss this. Here's one example: https://www.oculus.com/blog/vrs-grand-challenge-michael-abra...

Being able to make 2 stacked displays per eye that you can see through and made out of lightweight materials at a reasonable cost is quite challenging. Magic leap could have made the entire waveguide structure out of glass and probably made more progress quicker with that. Making waveguides is hard, especially 6 of them per eye..

I think the having multiple focal planes likely is over indexing on one thing that likely is not the most important thing..

What do you think is "truly immersive VR"? The technical challenges for VR depend on what you think makes it "truly" work or makes you dismiss previous solutions. I think I experienced immersive VR over 20 years ago in the CAVE system, which had shutter glasses, head tracking with stereo projection on either multiple walls or a thing they called an ImmersaDesk, and a user-input device called "the wand." It used a multi-rack SGI Origin, which I am sure is outclassed by a modern gamer's PC. There were some artistic demos and many science/visualization application demos.

I saw these in multiple research labs, but I'm not sure they ever really served a real application base. They were in constant use by grad students and postdocs writing new experimental apps and demos. Unfortunately, I think real technical users often find that immersion is a gimmick rather than a real tool. You'd rather spend your money or compute resources on higher resolution and high dynamic range images than stereo rendering. And, when you have challenging visualization problems you are often struggling to sustain any interactive rendering and exploration at all, much less the high framerates needed for immersion.

You could have immersion at any time, if you are willing to adjust your problem size and expectations. Stereo rendering always means rendering twice as much as for a regular screen, so you are also going to require a larger hardware investment for any given application and quality metric. Natural, immersive navigation always means limiting your model such that the working set remains well cached and you don't have massive drops in rendering rate as the user moves. People get aggravated when ripped out of immersion by application stalls.

Also, it is hard to make an immersive display system low friction/accessible to casual use. But, if you have to design your application to easily degrade to normal screens and keyboard+mouse, then you cannot really focus on and exploit the compelling parts of immersive rendering and immersive user inputs. Can you ever find enough users to agree on this set of compromises to make a product successful?

Just a hypothesis here: they failed because of, not in spite of, the amount of money they had. VR/AR will succeed by making just-barely-good-enough products for specific, limited environments (surgery, repair, control of robots in remote environments, education). Then, it will get slightly better. This requires that a product with a limited market is acceptable, at first, as it was for the very early computers. Big funding encourages one to strive for too much, too soon, too all-in-one-project. Incremental change is required, and that requires that an incremental improvement in market size is good enough.

I have no particular inside knowledge of this, it's just an idea/hypothesis.

Also, another factor -

Magic Leap decided, perplexingly, to locate itself in a suburb west of Fort Lauderdale. Not the most far-flung location, but not exactly a tech hub, either.

I was repeatedly contacted by their recruiters despite living in a completely different part of the country and having limited experience at the time. The impression it left me with is that they were really hurting for talent. Maybe the location is one reason why.

About 20 years ago, Motorola had a large plant in Boynton. Employed a few thousand engineers working on paging systems and eventually cellular telephony. Lots of it moved up to Chicago later on, but not all of it.

What’s left there are the echoes of the employees that hung around. IIRC, XM Satellite Radio has their engineering office in Lauderdale. Wonder where they found all the RF people needed to spin that one up.

> Boynton

Which one? https://en.wikipedia.org/wiki/Boynton

I'm guessing the Detroit one since you said "lots of it moved up to Chicago later on".

Boynton Beach.

http://articles.sun-sentinel.com/2004-01-23/business/0401221...

With the caveat that I also have no particular inside knowledge of this, I think you're on to something. I keep seeing these demos which show virtual video game characters or hyper-realistic objects and I can't help but think that it's complete overkill for a first generation consumer product. From a visual perspective, all I need out of my first gen AR glasses is to project some text annotations next to a real object, or show an arrow where I need to turn left. You don't need to fool me into thinking a virtual object is real. I'm looking for utility, not immersion.

> What is it exactly that makes VR tech so difficult, then?

Content

Grand Theft Auto 5 cost $265 million to create.

VR content is going to be much harder and expensive to create to just equal it.

It also has to work sitting down. No one wants to come home after a hard days work and relax standing.

(And don't even bother with the it will be useful in work environments, high end users often don't even use gui's and prefer text. Even the cliche of buying a house even with a cookie cutter plan the cost to make it 3D and look good from pre existing CAD drawings is probably too much)

Sadly Magic Leap I think had to ship something so they could start up a new hype cycle and get more funding.

Give it 3 months and they will be talking about how they didn’t expect to sell many of this first product anyway and now they are working on the “future for sure this time”..

I'm surprised they shipped anything. The hype was at perpetual motion machine level.

Ever notice the similarity between Bitcoin and VR blogs and comments? Always, "it's early days," even after years of waiting for something useful.

Well, if the technology isn't impressive at least they still have Neal Stephenson. He may be able to create experiences that no other VR platform offers.

https://magic-leap.reality.news/news/magic-leaps-neal-stephe...

FWIW Neal Stephenson also supported "Clang!" [0], a notorious early (2013) Kickstarter project (also dealing with "virtual in the real") that failed pretty miserably.

0: https://www.kickstarter.com/projects/260688528/clang/descrip...

Didn’t VR have the same hype train surrounding it that led to Facebook dropping a cool billion on it to acquire Oculus? To date what do we really have to show in VR?

How is Oculus/Palmer any less guilty than Magic Leap? Give it 5 more years and I think we’ll call both initiatives a failure when compared against their promises.

I think you are severely underestimating the current state of VR. It's still an enthusiasts market, to be sure, but it's well past the gimmick point, there are real applications, real content, and real consumer grade hardware out there. VR after the hype is real, just not selling like gangbusters.

I don't think your assessment of the VR industry is particularly sharp.

Why is that? As an outsider it seems like it launched as a neat gadget, but lacked staying power due to sparse content and high cost to entry.

I got as far as "I want what is best for VR" because in my opinion is a load of crap coming from a sell-out.

Does anyone know why he left Facebook?

He didn't leave. He was fired for unknown reasons.

But according to Zuck not due to "controversial political views" (Palmer is a passionate libertarian who supported Donald Trump, several Republican senators and the Republican party both in California and Washington.), but that's quite interesting because after word got out he supported Trump he was silently stripped of management responsibilities at Oculus ("In the wake of the revelation, Luckey’s role at Facebook was quietly reduced." -- https://www.theguardian.com/technology/2017/mar/31/palmer-lu... )

Also was a big disappointment for me personally, I got interested into Oculus due to his work and that of Carmack.

Reply on news.ycombinator.com