Magic Leap House of Cards – FSD, Waveguides, and Focus Planes
kguttag.com
kguttag.com
I don't know whether Magic Leap actually will produce anything of merit, but this blog won't help anyone find out.
It is a fact that what ML started out saying they were going to do is not what they are currently doing. And what they are currently doing is incredibly similar now to the Hololens.
> I don't know whether Magic Leap actually will produce anything of merit, but this blog won't help anyone find out.
I think the author of this blog post isn't wrong.
But ML has pivoted away from nearly all of their major technology advances at this point, so they are much more likely to deliver now what is likely to be an underwhelming product that is little differentiated from Hololens.
I thought they were always going to be a Hololens-like AR product?
Either they engaged in highly illegal false advertising about their current product or their old product was too magic and futuristic so they cancelled it. Not sure which.
But they pivoted from a non-goggles technology to a googles-technology somewhere along the road.
The first time I heard of ML was the Wired piece, and I think that one was already referring to glasses.
edit: Like the other comment here says, I just went back and checked a few places (Wikipedia and other articles) and it seems they've always been about glasses..
The hardware is yet to be revealed (but it'll presumably use a headset)
they waved away the fact that reducing it down to a wearable pair of glasses and an affordable computer was far from inevitable.
The truth is that Magic Leap cannot get its huge prototype working in a much, much smaller version.
Nobody who knows anything about AR ever thought it was anything but some form of glasses.
Edit: Given the seeming impossibility of an FSD like the one they’re still selling to investors, it may be ML’s investors are not part of the “know much about...” side.
That someone would make an assumption about something that they have no knowledge about, that isn't even available to them, means what exactly?
You guys are shouting at clouds.
No official marketing, patent or legal documents or anyone familiar with the space said anything about a form factor that was anything other than glasses ever.
Here is one of their investor documents from late-2013:
http://www.kguttag.com/wp-content/uploads/2017/12/2013-07-12...
On page 37, it states "Enabled by Proprietary Scanning Fiber Projector"
Page 148 "Enabled by Proprietary Ultralightweight protection and true lightfield technology."
Also look at how light the googles are in the fundraising documents, and compare to the huge v1 release they have just released pictures of. Night and day difference.
Nothing in interviews, talks etc... have they said anything about what they would deliver, on what timeline or in what form factor. Not in the wired interview the Information article, anywhere. Their patents have a wide range of designs and cover a lot of territory, which means basically nothing aside from optionality.
The "promises" about a ML product come from many "technical" people speculating based on no first hand knowledge. It's really astonishing and possibly really impressive depending on the outcome, how ML has really let the technoverse drive the narrative about them as a company and their product. It's like fan-fiction almost.
Actually all of their official marketing promised this in the earliest years. Here is an example, I remember there are more as well:
https://www.reddit.com/r/woahdude/comments/3q5bxc/magic_leap...
http://www.kguttag.com/2018/01/06/magic-leap-fiber-scanning-...
Key point...
Thus the fiber needs go 10.17 meters X 240/sec = 2,400 meters/sec. For reference, the speed of sound is a mere ~343 meters per second (varies with temp and air density) so the fiber has to average about 7 times the speed of sound and because it is accelerating and decelerating back to zero at each end, it has to have a peak velocity of greater than 14 times the speed of sound!!!!!
Frequently in these situations, it turns out that the mass consumer market is still satisfied with the product, since their expectations of the technology are more modest than those of a tech expert.
I guess only time will tell whether Magic Leap will end up being a "con" or, instead, whether they are able to put in place some crude mitigations for these technical challenges that allows them to still release interesting and usable technology. In the latter case, the technology would likely still have flaws that Karl would find wholly unbearable, but may nevertheless allow for compelling products for ordinary consumers.
The Fiber Scanning Display is in a whole different category. This looks like pure lying on Magic Leap's part.
In this particular case I have tried to lay out the math that proves that Fiber Scanning Displays are never going to support high resolution. No amount of money or time is going to make Fiber Scanning Displays work. I picked this case because the math was simple to follow and easy to prove that it was not possible. Yet it has been a key point in Magic Leaps Presentations and patent applications through to the latest ones printed.
I think it is a very valid point to asked how a company could have been presenting something that was not possible, why people invested in a company saying things that were not possible, and why Magic Leap is holding to the fiction that it might be possible.
It seems to me to be a pretty simple question. Were they lying or did they not know what they are doing?
There are other areas of optics and image quality where you can argue about whether the image quality will be "good enough" but Fiber Scanning Displays is a black and white issue.
All of the possibilities could be true. I find it hard to dismiss the personal anecdotes of people going into the ML offices and being blown away though. I am not betting on anything, yet, just that someone somewhere will solve this problem, as it seems insanely profitable to do so. Whether it's ML or someone else I don't know.
Sure, ML might have discovered/invented some new tech or method that disproves all the previous established knowledge, and they are sitting on some revolutionary new product..
But is that more likely than the idea that they thought they could pull off this technological "leap" in theory, and in practice eventually realized they could not?
I've heard them referred to as the "Theranos of VR" more than once. And people are still throwing money at Theranos ( https://www.bizjournals.com/sanjose/news/2017/12/26/theranos... ).
About the only conclusion that I've been able to draw is that a lot of people simply have too much money.
Wouldn't "uBeam of VR" be a more apt description? There's no outright fraud yet, just (apparently) tons and tons of bullshit and impossible promises.
> About the only conclusion that I've been able to draw is that a lot of people simply have too much money.
In case someone with an itchy wallet reads this:
I'm working on quantum flux capacitance technology that can revolutionize cold fusion research schedules. I need only a modest $10M and zero oversight to bring it to a market. It's a safer and better investment (AAA+ five stars!) than Theranos - I'm not a kind of person that could turn it into a massive media scandal. Contact me, and I'll tell you where to wire the money.
This (less biased) article[0] seems to suggest your analysis is a bit off.
Isn't there some old saying about how an engineer can mathematically prove a bumblebee can't fly?
[0]https://gpuofthebrain.com/blog/2016/7/22/how-magic-leap-will...
"By using a piezoelectric actuator to achieve this scanning, one can maintain scan rates on the order of 10s of kHz"
10 kHz at a framerate of more than 100 Hz gives you less than 100 vibrations per frame. If each vibration covers a single row of pixels, you need at least a dozen fiber displays to produce a high-resolution display. There will likely be artifacts at the boundaries, but it might be possible to compensate.
After reading this, I think that the technology might be possible, but it likely won't be as amazing as claimed by Magic Leap's PR.
The parallel fiber idea is "optically silly" but it takes some understanding of light to prove it is impossible too. It is yet another example of trying to fool people. I did try and explain this before on my blog over a year ago. See http://www.kguttag.com/2016/11/20/magic-leap-separating-magi... and scroll down to the Appendix at the bottom.
Basically you have multiple fibers going in a circle each with their center of origin it becomes impossible to get them to act like a single image for use in near eye optic.
What the layman would not understand is that this is very different from image stitching on a projection screen that in part relies on the light being diffused/randomized by the screen. In the case of near eye optics, there is no way to get the multiple projector image collimated AND seamlessly put together.
It is yet another example of a good con has to seem believable.
Why would they go in a circle when all they need to do is provide a single wavelength on a single axis? A single axis movement seems to be enough, no?
> In the case of near eye optics, there is no way to get the multiple projector image collimated AND seamlessly put together.
Unless they did something completely crazy and invented a method to do this. And maybe with this "invention" they built a demo and showed it to a few folks who were like "hey, this really works, I think I'll write an article about it and get published in Rolling Stone" or "Dude, take my money, please."
Reminds me of another story (definitely true and not about bumblebees). Henry Ford wanted a V8 engine block made out of a single casting and all the engineers said it wasn't possible. So what to do? Turns out throwing a shitton of money at a secret project lead to the answer and the flathead ford V8 engine was the result.
Going in a circle (really a trochoid) is actually a pretty smart way to scan a 2D surface with a single light ray, you basically just rotate the vibrating fiber's plane of vibration around an axis to produce a cone of light. If you wanted to have only a single axis movement, you'd need many more fibers, one for each row of pixels.
>> In the case of near eye optics, there is no way to get the multiple projector image collimated AND seamlessly put together.
> Unless they did something completely crazy and invented a method to do this.
Did you read kguttag's argument on the matter? The reason this doesn't work is because the optics for collimation depend on the incoming angle of the light, so it must happen close enough to the light sources that the different light cones do not overlap. Then the light still expands within the optics before it gets straightened out, so there needs to be some gap for tolerance. So either they end up with some stray light rays that go into undesirable directions, or there is a gap at the border between the different regions.
Now maybe they showed a demo of this with visible artifacts to people and they were still blown away. Most likely they also told them that there was still room for improvement, while leaving open where exactly the fundamental limits are. Maybe they will release the imperfect version of their product than can actually be built, and people will still love it despite not living up to the marketing.
The problem is not that Magic Leap's ideas are completely useless or impossible to implement, just that they are significantly overhyping the expected capabilities of the finished product.
For proof of concept, one need only look at the fiber scan endoscope demos:
This is essentially the same technology, but operated in reverse (i.e. using the scanning fiber as a camera vice an image projector).
This is Especially applicable here in photonics and sight.
Most engineering works in approximations that get taught as rules. One needs to be very careful of the assumptions they're making about the physical world and eyesight when discussing AR.
Evaluates to false. Engineering is applied science, and thus all engineers are in fact scientists.
Most optics experts don’t dispute that you can deliver the experience they are promising - they dispute that you can do it in a product with cost and size that is compatible with any kind of volume manufacturing.
Basically you have multiple fibers going in a circle each with their center of origin it becomes impossible to get them to act like a single image for use in near eye optic.
What the layman would not understand is that this is very different from image stitching on a projection screen that in part relies on the light being diffused/randomized by the screen. In the case of near eye optics, there is no way to get the multiple projector image collimated AND seamlessly put together.
It is yet another example of a good con has to seem believable.
People today are all the time waving their hands and invoking "Moore's Law" and saying "Steve Job did this" and applying it to everything (I see it all the time in AR and Optics) and therefore anything is possible.
I can't say who I am going to see as a lot of it is private. I will be spending a LOT of time in the Main Convention center South Hall with the AR and VR companies. The special section for AR and VR has moved from the back of South Hall to near the front this year. Then you have to look for the more established companies that booths elsewhere.
The CES tools for finding booths and vendors have improved dramatically form where they were a few years back. You can search based on what you are trying to find. They also have a smartphone ap that comes in handy at the show.
So what? What are we supposed to take away from this?
Even if true, its clearly not stopping them, and its not clear if that will or will not make a successful consumer device.
If its rubbish, a few of people will lose a lot of money and we’ll all move on with our lives a bit disappointed, but otherwise not really caring.
The sooner they ship whatever it actually is, the better.
This kind of idle speculation is just gossip.