Magic Leap’s CEO Rony Abovitz Responds KGOnTech And Dodges Questions

Introduction

Rony Abovitz, CEO of Magic Leap, gave me a couple of shout-outs on his Twitter feed. I have cut and pasted images from the feed at the end of this article. I wanted to make a quick to respond to his response (he uses Twitter, I use my blog).

Dodging the Question and a Logical Fallacy

Jan Ciger wrote one thread, “Very interesting series of blog posts by Karl Guttag on Magic Leap One based on the patents and openly available information. Let’s hope this analysis is wrong or this will be one major pile of hype biting the dust (and a huge disappointment).

To which Rony, CEO of Magic Leap, replied by dodging the question with, “Karl has never seen or used a Magic Leap system – many others have.”

Rony employed question dodging trick of “Answering a completely different question.” I have laid out my sources of information, how I have drawn my conclusions and gone as far as providing the 3-D models so people can see for themselves.

Rony is also employing a logical fallacy. You don’t have to physically do something to know how it works. For example, you don’t have to walk on the moon to know how gravity behaves there; it is a matter of science. You might get a more precise answer by going to the moon and making measurements, but you can get very accurate answers without having to make the trip.

In another thread, Roope Rainisto‏ asked a different question and got the same dodge (although Rony used my full name this time)

It’s good marketing speak, but from what Karl Guttag and others claim, all I read about the reality of One is diffractive waveguides (like in Holo). What does a “natural light field” actually mean?

Rony Abovitz responded with,”Karl Guttag has never seen or used a Magic Leap system.

Rony could have said he wouldn’t answer that question or not responded at all. I need to point out that there is a mountain of evidence in patents and papers that indeed Magic Leap are using diffractive waveguides. In fact, the evidence points to them using two sets of 3 (red, green, blue) diffractive waveguides, essentially two Hololens diffractive waveguides stacked on top of each other.

Closing Random Comments

I do appreciate that Rony used my name this time and didn’t weasel around it as he did a year ago with “grumpy mouse blogger.”

The fact that he felt he had to respond to me by name suggests that this blog is getting to him and he is getting questions from others.

I’m hoping somebody in the media will follow up and not let him dodge the questions.

I’m happy to debate and defend my work on a public forum (podcast, video, or live). Who wants to take the side that the view of the real world looks different than what I presented and that they are not using diffractive waveguides?

BTW, I don’t usually follow Rony Abovitz’s twitter feed. I would like to acknowledge that I found out about these tweets from Noah_A_S on the Magic Leap subgroup on Reddit.

Snapshots directly from the Tweets with my annotations in red:

15 comments

  1. “You don’t have to physically do something to know how it works”, but you have to physically do something to know how it feels.

    That is what Rony meant.

    1. I’ve tried headsets with similar designs. I know how it feels. Also, you should note that I built a model so I know how the real world feels looking through the headset. No amount of images on the display can fix the garbage in your FOV from the headset.

    2. Well true – we learn as kids not to touch fire, regardless how many times our parents tell us not to.
      But once it was established that fire burns human hands, it became pointless to argue.

      The same stands for ML HDM, through the looking glass the gimmicky aspect might masquerade a lot of the short comings. Regardless, from a engineering point we know without trying where those short comings will be.

      There is large disparity of what is hyped and what will be reality. Nevertheless the first moments with the novelty will cloud a less critical viewers judgement.

    1. Thanks. It would be good to find out if you can how long they got to wear the unit and how bright light was in the room was where they used it. Were they showed a canned demo or was it open-ended?

      If you are in a dark room then the problems will be harder to notice. For example, the reporter for Rolling Stone said how solid everything looked but later on a Reddit AMA, he said that the reason might be because the room was so dark.

      If you can control what is shown and the environment in which is it shown you can impress people with a totally worthless product.

      1. Yes, definitely. Well, you can try to contact James directly (https://twitter.com/jamesashley). However, I suspect he will not want to talk on the record because he likely had to sign an NDA in order to try the device.

        I don’t have more contact than that Twitter handle – frankly, I was surprised to see the hullabaloo my tweets caused, including a direct reaction from ML’s CEO, given that I am a very small fish in this pond.

        I completely agree with that issue of controlled environment/demos – I am working in VR for some 20 years now and you can literally hide an elephant in the room if you properly stage it.

          1. I have sent the link to the comments and this article to Brian Crecente (the journalist from the Rolling Stone/Glixel article) since he commented on the thread too. Hopefully he can be more open with what he can say.

            BTW, I have realized one thing – is it possible that the people who have tried the device didn’t see the eye tunnel “ghosts” simply because the room was so dark in those demos, effectively making the problem invisible against the background? That could well explain the different experience vs. your analysis.

      2. In this AR robot demo, you can see clearly that it’s merely additive like the hololens. Things are not “solid”. I took two frames and overlaid them in gimp. First I used a filter to take the brightest pixel from both images. The I used a filter to take the darkest pixel from both images. There are no areas left behind, there is no magic light blocking technology as mentioned in prior patents.

        https://imgur.com/a/Y22mC

        1. This is no surprise, but I appreciate your work to overlay the images. BTW, the robot appears to have a lot of image issues including double images.

          Hard edge occlusion (pixel resolution level) may be physically impossible for the “general case” of variable distance objects in the real world. It is perhaps hard to understand, but with massive impact, is to have the mask in-focus.

          Magic Leap licensed some technology for hard edge occlusion from the U of Arizona, but that technology/optics is not only big and bulky, it only works for the special case of all the objects in the real world being about the same distance from the viewer (so they all focus the same).

          I wondering why (not really) they have not released a vergence-accommodation demonstration.

          1. One of the many “hard” problems of AR that people vastly overestimate. I’d just cheat, but even then it would not be easy. Use an occlusion LCD, but don’t bother trying to get it in the right focal plane. Just calculate based on your front facing camera what extra light will be missing (magic) and fill it in….assuming you can project it in the right depth of field. Or again, cheat and adjust a single focal plane (magic) to the depth the user is focused at (magic) and use rendering to make things off that plane be out of focus.

    2. There is a slight difference, if somebody who makes a living in R&D, to try any device, not just a HMD. We look much more critical on the areas where we are not supposed to look like. Hence we mentally dissect any invention, before we can fully appreciate it.
      Its the same a chef, a movie director, a composer will look anything within their genre much more critical.
      Is this the wrong way to critique a product ? Not at all, it yields a condensed version of what the masses will ultimately experience them self.

      There is much more to a HMD then a novelty factor, that is what Karl or myself is questioning. The downstream viability beyond a demo.

      1. Preaching to the choir here, I am for in the AR/VR field for some 20 years now myself, having started just as the first VR hype wave in the 90’s was dying down. I still remember meeting Myron Krueger in 1998 or so showing us what he was doing with video in the 70’s that was miles ahead in usability compared to the goggles & gloves “monkey suit” (his words) stuff that was norm of the day (and is being reinvented again today – not to mention that the best dataglove on the market is still the 20-something years old CyberGlove …)

        The worst part is that while the hardware made quite huge steps forward (e.g. original iGlasses vs Hololens), the applications and most problems associated with it are still the same as they were – the gear is bulky, fragile, often works only in ideal conditions, both the hardware and applications have terrible usability, often times it is more a gadget than something really useful, etc.

        I have seen a lot of overhyped products which either did not deliver or have gone bust outright. So one learns to be both cynical and to look at these things with a critical eye, comparing to what else exists, what the tradeoffs are and what it can be actually useful for. I think I am long past the stage of being easily impressed by flashy staged demos and marketing BS.

  2. you can see all that “other people” are famous like Beyoncé, Andre Iguodala, Shaquille O’Neal, Ricky Gervais,and even Saudi Arabia, Prince Mohammed bin Salman. what they are understand in technology. rony just do it for PR.

Leave a Reply

Your email address will not be published. Required fields are marked *

%d bloggers like this: