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

Magic Leap House of Cards – FSD, Waveguides, and Focus Planes

kguttag.com

Post reply on HN
Print view View original

21–30 of 60 posts

Re: Magic Leap House of Cards – FSD, Waveguides, and Focus Planes

#21

Post by jfaat »

UncleEntity wrote: post #20
kguttag wrote: post #7

Earlier quoted context omitted.

Objective is often in the eye of the beholder. I do express opinion, but I try to back them up with evidence. 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…

> Were they lying or did they not know what they are doing? 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...

Had to look that saying up [0], but it seems like it’s just a made-up thing to help people who aren’t scientists cope with that fact.

[0] https://www.snopes.com/science/bumblebees.asp

Re: Magic Leap House of Cards – FSD, Waveguides, and Focus Planes

#22

Post by sprafa »

kguttag wrote: post #7
drcode wrote: post #5

Earlier quoted context omitted.

I'm definitely a Magic Leap skeptic, but these Karl Guttag posts do not read like posts from someone who is trying to do an objective analysis of the tech, they read like posts from someone who has an axe to grind. Often, when I see posts from very smart and technically savvy people that call things "cons" or "scams" it's because they have very high personal standards of technology and are disappointed that others do…

Objective is often in the eye of the beholder. I do express opinion, but I try to back them up with evidence. 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…

Your articles are mildly helpful. However it's also possible that ML has new insights into this technology or that they have developed new tech whose patent is still under consideration? What if you're right and these particular patents are just FUD designed to mislead competitors and they're going with something else?

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.

Re: Magic Leap House of Cards – FSD, Waveguides, and Focus Planes

#23

Post by yorwba »

UncleEntity wrote: post #20
kguttag wrote: post #7

Earlier quoted context omitted.

Objective is often in the eye of the beholder. I do express opinion, but I try to back them up with evidence. 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…

> Were they lying or did they not know what they are doing? 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...

That article doesn't really address the issue of the fiber speed except for

"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.

Re: Magic Leap House of Cards – FSD, Waveguides, and Focus Planes

#24

Post by bhouston »

ghusbands wrote: post #16

You're not going to find the technical truth on this blog. Look at the history of posts; the author clearly believes that Magic Leap is fraudulent at its core and wants to prove it. Everything Magic Leap does/says is attacked from that inherently biased position. Any facts or data presented on the blog will be with that goal in mind, and any facts or data that support the opposite will be downplayed or omitted. I don…

A lot of experienced people in the computer graphics industry doubt Magic Leap and their claims -- myself included.

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.

Re: Magic Leap House of Cards – FSD, Waveguides, and Focus Planes

#25

Post by bhouston »

Steeeve wrote: post #10

I thought Magic Leap had already waived the white flag and decided to become yet another goggle vendor.

They have practically done that, none of their unique technology works, but not everyone is aware that this is what has happened.

Re: Magic Leap House of Cards – FSD, Waveguides, and Focus Planes

#26

Post by rubidium »

jfaat wrote: post #21
UncleEntity wrote: post #20

Earlier quoted context omitted.

> Were they lying or did they not know what they are doing? 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...

Had to look that saying up [0], but it seems like it’s just a made-up thing to help people who aren’t scientists cope with that fact. [0] https://www.snopes.com/science/bumblebees.asp

Not disagreeing that the saying may be made up, but Scientist != engineer.

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.

Re: Magic Leap House of Cards – FSD, Waveguides, and Focus Planes

#27

Post by URSpider94 »

sytelus wrote: post #18
kguttag wrote: post #7

Earlier quoted context omitted.

Objective is often in the eye of the beholder. I do express opinion, but I try to back them up with evidence. 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…

But... they had done demos to many investors and at least one journalist who were convinced to give them money and write article. They have already announced that there will be product in 2018. What do you think they demoed and what the actual product will be?

I have trusted acquaintances who have seen their demos. They are reportedly beautiful. But, they are also reportedly tied to extremely large and complex optical benches.

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.

Re: Magic Leap House of Cards – FSD, Waveguides, and Focus Planes

#28

Post by kguttag »

yorwba wrote: post #23
UncleEntity wrote: post #20

Earlier quoted context omitted.

> Were they lying or did they not know what they are doing? 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...

That article doesn't really address the issue of the fiber speed except for "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…

The GPU of The Brain author seems to be well meaning and nice guy but he don't not understand optics.

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.

Re: Magic Leap House of Cards – FSD, Waveguides, and Focus Planes

#29

Post by kguttag »

onion2k wrote: post #15
QAPereo wrote: post #2

Another post from today, really laying out the math behind hi-res FSD being a bad joke. 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 acceler…

I don't know anything about FSD so I'm asking as a layman: why only have one fiber? If you could focus, say, 16 fibers on the same plane (as spirals making overlapping circles) then each one would only need to cover 120x68 pixels of the HD display so they could move much more slowly.

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.

Re: Magic Leap House of Cards – FSD, Waveguides, and Focus Planes

#30

Post by kguttag »

alakin wrote: post #17
QAPereo wrote: post #2

Another post from today, really laying out the math behind hi-res FSD being a bad joke. 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 acceler…

Couldn't they just use multiple fibers per display to get around that problem?

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.

Post reply on HN