Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

My first year at Magic Leap and the opportunity ahead

magicleap.com

Post reply on HN
Print view View original

91–100 of 174 posts

Re: My first year at Magic Leap and the opportunity ahead

#91

Post by spaetzleesser »

nikanj wrote: post #30
raesene9 wrote: post #21

I must admit I'd forgotten about Magic Leap, amazed to see they're still going and raising even more money. In my head, I always put them in a bucket with Theranos and uBeam, Startups who made product claims which experts in their fields said were not possible. I guess slick marketing wins over those doubts in some VC circles but you wonder if anyone at the early stage Magic Leap investors is getting asked tough ques…

> I guess slick marketing wins over those doubts in some VC circles but you wonder if anyone at the early stage Magic Leap investors is getting asked tough questions about why they approved it, or if they've all just moved on to other roles. I would guess this depends on whether or not they managed to hoist the stock to someone else in subsequent rounds. Investing in a scam can still be a good deal, as long as you ge…

That seems to be the playbook for today's highly valued unicorns. Squeeze out all the value then dump the stock on the public before things start going down. It's not exactly a scam but it's the same kind of thinking: Doing irrational things is fine as long as you are confident there will be a greater fool later on.

Re: My first year at Magic Leap and the opportunity ahead

#92

Post by Matrixik »

idontwantthis wrote: post #41

Does anyone here work with super expensive "enterprise" VR/AR? My gut instinct is that they try to make something good, realize the tech to fulfill their dream is incredibly expensive, then pivot to "enterprise" to get more funding. Am I being cynical?

Here example of enterprise Varjo XR3 (full version cost 6000$ plus 1500$ subscription) from someone only testing consumer VR headsets till now. In short: it's way better than best consumer headsets.

https://www.youtube.com/watch?v=NOk_M1Ib5F0

Re: My first year at Magic Leap and the opportunity ahead

#93

Post by moron4hire »

throw_m239339 wrote: post #3

I think HN is pretty set on the fact that Magic Leap is vaporware, isn't it? do they even have a product?

They do have a product. It didn't live up to Abovitz' hype, but that's also why he's not CEO anymore.

The ML1 sits somewhere in the vicinity of the Hololens 1 in capability. Some things it does better, some things it does worse.

For example, having a real controller was very useful as the input latency on the Hololens' hand tracking is a big impediment and not having a variety of physical buttons makes designing games with lots of different types of interactions really hard. I think Waltz of the Wizard on the Quest 2 is the only game I've seen with hand tracked input that isn't just "point and click".

Wider field of view than the Hololens was nice, but it came at the price of lower pixel density.

Being based on Android was a drawback for me, though. Hololens has a workspace metaphor, akin to a 3D desktop, that Magic Leap never really put much effort into. Hololens also has speech recognition as a first-class citizen, whereas Magic Leap left it up to developers to roll their own. It's really, really hard to get good performance out of Android, too.

A bunch of the other stuff was just too wonky to consider. The surface meshing was unusable in real life scenarios. The hand tracking as well. A lot of people don't realize that the Hololens could do full, positional hand tracking, it just wasn't a great experience, so everyone designed around nose-pointers and just using the hand as a clicker. The hand tracking on Magic Leap was even worse, yet they went with trying to showcase it.

I think, if Microsoft were to release a controller to get over the latency issues of hand tracking on the Hololens, it'd be strictly better all around.

I'm rather bearish on AR devices making a big splash in the consumer market, especially ones using waveguide displays. Waveguides are a technological dead end. There isn't really any room for them to grow by even 2x, when they really need to grow by at least 20x. Good AR application design is also excruciatingly hard. What you typically end up with is a VR app made shitty by having to run on an AR headset. Oculus' approach of starting with a VR headset and adding AR-passthrough makes a lot more sense.

Re: My first year at Magic Leap and the opportunity ahead

#94

Post by jazzyjackson »

breezeTrowel wrote: post #63
raesene9 wrote: post #21

I must admit I'd forgotten about Magic Leap, amazed to see they're still going and raising even more money. In my head, I always put them in a bucket with Theranos and uBeam, Startups who made product claims which experts in their fields said were not possible. I guess slick marketing wins over those doubts in some VC circles but you wonder if anyone at the early stage Magic Leap investors is getting asked tough ques…

> Startups who made product claims which experts in their fields said were not possible. Can you clarify what you mean by this? What aspect is considered impossible? Augmented Reality has been around for a while. There's the Microsoft HoloLens. Not to mention all the AR demos Apple does whenever they announce a new iPhone. Check out this page, for example: https://www.apple.com/augmented-reality/ especially the Snapc…

????

the hype was around their fiber scanning display patents, they were showing investors a totally new technology that projects images into your eyeballs. There are numerous articles describing the patented pie-in-the-sky vs the state-of-the-art, just one example here: [0]

> They don’t even have an decent brightness control of the pixels and didn’t even attempt to show color reproduction (requiring extremely precise laser control). Yes the images are old, but there are a series of extremely hard problems outlined above that are likely not solvable which is likely why we have not seen any better pictures of an FSD from ANYONE (ML or others) in the last 7 years.

Magic Leap was unable to improve and miniaturize this technology - its a dead end - so they ended up using the same tech as everyone else

[0] https://kguttag.com/2016/11/28/magic-leap-no-fiber-scan-disp...

Re: My first year at Magic Leap and the opportunity ahead

#95

Post by baybal2 »

knownjorbist wrote: post #89
baybal2 wrote: post #43

Earlier quoted context omitted.

One on my list of mind boggling "Tech 2.0" scams were: 1. Teranos 2. Magic Leap 3. Andelur Ghost Only the last one still didn't get enough media attention. People who scored Andelur handshakes with the military & intelligence are the types who did it for Advanced Tactical Security & Communications Ltd ( https://en.wikipedia.org/wiki/ADE_651 ) Andelur is basically trying to sell the governments around the world a Chin…

Would you care to reply to the people who attest that the magic leap is actually underwhelming given the current state of AR?

I will. I work in an engineering company which shipped a number of AR glasses projects (for clients I can't name,) and I can say that whomever did engineering for Magic Leap had no idea whatsoever about basics of optics, and such.

And their overall engineering level is totally amateurish — you cannot have such shoddy job being done for money, let alone such amount of money, but they have everything to tick the checkbox in the buzzword department: waveguides, holographics, structural colour, photonic chips...

Want talk more, leave a way to contact you on your profile page

One main point:

Holographic waveguides, and such are a dead end development with a substandard image quality which you cannot do anything about, as well as its inefficiency (these waveguides waste 70-90%+ of light.)

You can make them more opaque to preserve more light, but then you get VR sunglasses, instead of what the promise is.

The fundamental problem is that the amount of image light transferred is inversely proportional to the amount of light obscured. — they are extremely inefficient, and there is physically no way around this.

For this reason this solution really had to be sent to the wastebin on day one, and not be presented as the next best thing after sliced bread. This is obvious to anybody knowing optics on above a high school level, and it's inconceivable how anybody paying real engineers can opt for this solution.

Re: My first year at Magic Leap and the opportunity ahead

#97

Post by tomas789 »

cmeacham98 wrote: post #54
krageon wrote: post #25

Earlier quoted context omitted.

I have tried the leap in practice and the experience was significantly above what I expected. The tracking worked most of the time (how???), with occasional wonkyness. The accuracy was very good (outside of the wonks). Where it is very limited is number of applications.

The problem, to me, looks like getting 90% of the way there is possible. However, getting from 90% to 99.9% (to get rid of that "wonkyness"), seems to be way way way harder. Magic Leap has made little to no progress in working out the kinks in their product despite raising literally billions of dollars.

I wrote a camera/IMU tracking software (VIO) before. I can confirm this statement. Get it working is difficult and requires getting all the algos right. Getting it work reliably is a lot of rigorous testing, collecting datasets, evaluating slightest changes, tuning numerical stability, experimenting with different techniques which might work better in low light and optimizing for the hardware (VIO is intensive but the best VIO is useless if there is no computation power left for the app itself).

Neither of this is required for a slick demo in well lit room and perhaps on better HW.

Re: My first year at Magic Leap and the opportunity ahead

#98

Post by leetrout »

sydthrowaway wrote: post #81
leetrout wrote: post #36

If you have not experienced the magic leap videos do not do it justice. Heres my video flying inside my house: https://youtu.be/Grlk03MdScQ The jaw dropping aspects: - it correctly knows when to mask for the column - it does lighting effects from the planes headlights - it does particle collisions with my furniture when it crashes - it crashes by detecting i hit the wall It really is amazing tech but it is very unpol…

Are you serious? This looks like an undergraduate project hacked up with OpenCV.. I'd still give it an A though.

As I said, video does not do it justice. If you walk around while flying everything stays very well locked in place.

It probably _is_ using OpenCV under the hood.

Re: My first year at Magic Leap and the opportunity ahead

#99

Post by tootie »

breezeTrowel wrote: post #63
raesene9 wrote: post #21

I must admit I'd forgotten about Magic Leap, amazed to see they're still going and raising even more money. In my head, I always put them in a bucket with Theranos and uBeam, Startups who made product claims which experts in their fields said were not possible. I guess slick marketing wins over those doubts in some VC circles but you wonder if anyone at the early stage Magic Leap investors is getting asked tough ques…

> Startups who made product claims which experts in their fields said were not possible. Can you clarify what you mean by this? What aspect is considered impossible? Augmented Reality has been around for a while. There's the Microsoft HoloLens. Not to mention all the AR demos Apple does whenever they announce a new iPhone. Check out this page, for example: https://www.apple.com/augmented-reality/ especially the Snapc…

What they lied about was the fidelity. They made demo video in 2015 that was a fully prerendered video of what they expected their AR experience to look like. Years later, the top of line headsets are fuzzy and very limited in field of view and have not seen any adoption outside of very niche applications.

https://youtu.be/ctUx3BzuQgQ

Re: My first year at Magic Leap and the opportunity ahead

#100

Post by snek_case »

sydthrowaway wrote: post #81
leetrout wrote: post #36

If you have not experienced the magic leap videos do not do it justice. Heres my video flying inside my house: https://youtu.be/Grlk03MdScQ The jaw dropping aspects: - it correctly knows when to mask for the column - it does lighting effects from the planes headlights - it does particle collisions with my furniture when it crashes - it crashes by detecting i hit the wall It really is amazing tech but it is very unpol…

Are you serious? This looks like an undergraduate project hacked up with OpenCV.. I'd still give it an A though.

Yeah. I'm kind of left feeling like I'd rather fly the plane in a better looking simulated environment, with a first-person perspective. It's more flexible and fun than being tied to flying it in the environment around you.

I still haven't seen a compelling use case for AR.

Post reply on HN