Video Overview

Introduction

The Magic Leap One’s mixed-reality tech has been so much pie in the sky for so long, we can hardly believe we have it on our teardown table. Based on the amount of money raised for this project, we’re hoping it’s powered by pixie dust—but only a teardown will tell.

Want to see more hardware secrets? Then plug in with us on Facebook, Instagram, or Twitter for up-to-the-minute teardown news.

This teardown is not a repair guide. To repair your Magic Leap One, use our service manual.

  1. There's a lot to unravel here. Let's start with some specs ... err, specifications:
    • There's a lot to unravel here. Let's start with some specs ... err, specifications:

      • Nvidia Tegra X2 (Parker) SoC with two Denver 2.0 64-bit cores and four ARM Cortex A57 64-bit cores

      • Integrated Pascal-based GPU with 256 CUDA cores

      • 8 GB RAM

      • 128 GB onboard storage

      • Bluetooth 4.2, Wi-Fi 802.11ac/b/g/n, USB-C, 3.5 mm headphone jack

    • That Nvidia SoC is designed for automotive applications, and is prominently used in autonomous vehicles, including Teslas. This seems like off-label use, until you consider the Magic Leap's multiple arrays of external sensors for mapping and understanding its environment—much like a self-driving car.

    The CUDA cores are a fixed feature of the Parker SoC - not a separate semiconductor. They don’t deserve an individual bullet in this listing.

    alexander.stohr - Reply

    • After Magic Leap's claims about this headset's otherworldly experience, we had to try it ourselves.

    • Thanks to our trusty infrared camera, we can see a strobing IR projector for depth sensing just above the nose bridge—similar in principle to what we found in the iPhone X, and before it, the Kinect.

    • If you look closely, you can also spot four additional IR LEDs in each lens, "invisibly" lighting up your eyeballs for tracking. (We'll dig up the trackers shortly.)

    That is not how you wear the headset…

    Shaaheen Peyvan - Reply

    • Before we get started, let's orient ourselves with a high-level tour:

      • Content creation begins in the Lightpack. It provides power and handles the processing, sending image and sound data to the headset.

      • Meanwhile, the Lightwear headset tracks the controller's position and orientation, and maps your surroundings to help insert the virtual elements.

      • How those virtual elements are generated is a whole other can of worms.

    Add Comment

    • “Mixed reality” is hard. It’s one thing to augment what you see on a screen (like a smartphone or a VR display with a feed from exterior cameras).

    • It’s much harder to augment actual, unfiltered reality coming directly into your eye. To pull off this illusion, the Magic Leap One uses a couple neat pieces of tech:

      • Waveguide display - Essentially a transparent screen that’s lit invisibly from the side. The waveguide (what Magic Leap calls a "photonic lightfield chip") guides light—in this case, an image—across a thin layer of glass, magnifying it and angling it into your eye.

      • Focus planes - On a VR display, everything is in focus all the time. Reality is different—some things look crisp while others can look blurry, depending on where your eye is focused. Magic Leap mimics this effect by stacking multiple waveguides to create focus planes—slicing up the image into crisp and blurry areas.

    Add Comment

    • Let's get to the optical treasure in this thing! A quick test rules out polarized lenses—we'll have to dig deeper to make any discoveries.

    • The insides of the lenses are surprisingly ugly, with prominent IR LEDs, a visibly striated waveguide "display" area, and some odd glue application.

    • The waveguide consists of six not-so-prettily laminated layers, each with a small air gap.

      • The edges look to be hand-painted black, likely to minimize internal reflections and interference.

    I would more tend to think that the waveguides are optically bonded and there are no air gaps. Usually optical bonding minimizes partial reflections at interfaces in order to increase overall contrast. But the waveguides must also rely on total internal reflection. That sounds like conflicting goals.

    Jerome - Reply

    I was surprised by this as well, but I think if it were bonded all the way across with OCA you wouldn’t be able to see the obvious glue layers around the edges.

    Jeff Suovanen -

    The waveguides work by total internal reflection (TIR) cause by the index of refraction mismatch between glass and air. If the layers of glass were bonded together with no gap (or other indexed mismatched material in between) then the light would exit the waveguide to the next layer an you don’t have a waveguide/TIR.

    Karl Guttag -

    • Inside the headband, we note a Class 1 Laser label. It might seem like a scary thing to find on your eyewear, but it's safe for all normal use conditions and likely no more dangerous than a CD player.

    • Twirling away the standard Torx screws and removing the panel reveals the first of two speakers, wired via spring contacts, and protected by color-coded gaskets—great repairability so far!

    • Also hiding under those panels: the two upper ends of the device's single built-in cable—and a few of the magnetic bits that help adjust the fit.

    • But what's that odd little black box protruding from the right side of the headband?

    CD players use Class 3B lasers which are dangerous to the eye. Class 1 is much, much, much safer, and it is eye-safe at all distances.

    Daniel Lu - Reply

    • Survey says: a six-degrees-of-freedom magnetic sensor coil for tracking the position of the controller.

      • The intensity of the three perpendicular magnetic fields is measured to determine the position and orientation of the controller relative to the headset.

    • Cracking open the controller, we find the (much bigger) emitting half of the tracker, and an 8.4 Wh battery to boot.

    • The copper shielding sprayed into the coil housings likely protects from RF interference, while letting the magnetic field through.

      • Interference could explain the tracker's odd placement, and this may be a temporary solution. It's "old" tech, and will probably be worse for left-handed use.

    • Not pictured: we also dug up what looks like a custom-designed trackpad ringed with LEDs (for future light-tracking hardware, perhaps?).

    The magnetic coil 6 dof sensor system is almost certainly adapted from either Polhemus’ AC magnetic system or Ascension Technologies’ DC magnetic system. Patents on both are long since expired.

    Tony Asch - Reply

    The LED light ring is just for aesthetics, animations etc.

    Justin Alvey - Reply

    How do we know it’s AC or DC type magnetic tracking from this tear-down image?

    James Di-Bao Wang - Reply

    • After detaching the headband and inner faceplate, we have a better view of the eye-tracking IR emitters. We note they're all wired in series, rather than individually controlled.

    • And finally, at our fingertips, the heart of the Magic Leap: the optics and display assemblies.

    • We're going all the way here, so strap in.

    Could you please disassemble the strap as well? We haven`t seen a mechanism like this in any HMD yet. Packed in a small space, supposed to last years with enough force to keep it on various head sizes. I really want to know what kind of spring they used for this.

    Jan - Reply

    @tancho I only got a glimpse of it when we had it apart, but it looked like a single constant-force spring packed into the center plastic segment at the back.

    Jeff Suovanen -

    • Lifting up one of the external sensor arrays, we find beneath: the optical system for injecting images into the waveguides.

    • These bright colors come from ambient light reflecting off the diffraction gratings, and don't represent specific color channels.

      • Each spot lives at a different depth—corresponding to a single layer of the waveguide.

    • At the back, we find the actual display device: an OmniVision OP02222 field-sequential color (FSC) LCOS device. It is likely a customized variation of the OmniVision OP02220.

    • The KGOnTech blog correctly guessed this was exactly what Magic Leap was doing, based on patent filings back in 2016.

    Add Comment

    • Let's take a deeper dive into the projector and waveguide optics.

    • So what's all this about six layers? There's a separate waveguide for each color channel (red, green, and blue) on two distinct focal planes.

    • "Figure 6" from the Magic Leap patent application 2016/0327789 shines some light on the optics' inner workings.

    • For your edification and delight, we've included our own "tl;dr diagram" for this system, complete with cats.

    The “tl;dr diagram” has the positions of the LED light source and LCOS grid swapped: Light from the LCOS will not be reflected into the waveguide. The figure from the Magic Leap patent is correct.

    Jack Boyce - Reply

    Well, yes and no. The position of the components shown in our diagram is correct—unlike in the patent drawing, the physical Leap One does have the light source at the back, firing forward (toward the waveguides). The LCOS grid is on the top, facing down. However, our graphic is incomplete and does not show every optical component—so if you’re trying to follow the light path through the entire system, you’re going to get tripped up and the system won’t “work.” Our graphic was originally meant to be a high-level conceptual overview and probably got too detailed for its own good. It certainly won’t satisfy any optics experts. But if you’re trying to understand the whole system, here’s a hint: there’s a mirror on the other side of the beam splitter, directly opposite the LCOS panel.

    Jeff Suovanen -

    I wished there were more detailed images of complete breakdown of the projection system instead of trying explain the process. Some of us already have a good understanding of the process and have relied on IFIXIT to show us the components to verify our assumptions. With only few images we are still relying on vague block diagrams.

    All Things 3D - Reply

    • A cast magnesium block holds all the optics and sensors, and is surprisingly hefty for an HMD. The VR headsets we've taken apart all used lightweight plastic.

    • But metal makes a better heat sink, and the electronics and IR illumination (likely VCSEL devices) all produce heat.

      • That pink stuff is thermal paste to help dissipate the IR rangefinder's heat.

    • Metal also offers a stiffer mounting place to keep the optics stable and focused after their rigorous calibration.

      • But stiff isn't always best—some of these components are mounted with foam adhesive, which will be more forgiving as things flex when heated.

    Add Comment

    • No longer perfectly placed, we may as well pop the covers off the sensors for a closer look.

    • These twin sensor arrays sit at each of your temples, with the strobing IR depth sensor perched right in the middle.

    • A closer look at the nose-bridge depth sensor gives us the room-reading hardware in the form of:

      • IR sensing camera

      • IR dot projector

    • There's no need to set up receiver stations for this device—it does the projection and reading all on its own!

    Add Comment

    • Connecting all that sensory equipment to the headband, we have pricey, layered flex cables hosting:

      • Movidius MA2450 Myriad 2 vision processing unit

      • SlimPort ANX7530 4K DisplayPort receiver

      • OmniVision OV680 sensor bridge for processing simultaneous image streams from multiple cameras (like we found in the Amazon Fire Phone)

      • Altera/Intel 10M08V81G - 8000 logic cell FPGA, possibly for glue logic, or managing MV part or camera bridge data

      • Parade Technologies 8713A bidirectional USB 3.0 redriver

      • NXP Semiconductors TFA9891 audio amplifier

      • Texas Instruments TPS65912 PMIC

    This page is more useful to tell what’s inside the Altera/Intel 10M08V81G (the linked datasheet only lists operating conditions and specs that are of no interest to most people)

    https://www.intel.com/content/www/us/en/...

    Family table

    https://www.intel.com/content/dam/www/pr...

    juliov - Reply

    The Myriad 2 is a vision processor, not a display processor. It’s going to be used here as a front-end processor for the cameras.

    Scott Morrison - Reply

    Fixed! Thanks for the tip. No idea how we managed to bungle that one—had a link to the right datasheet but somehow not the description. :p

    Jeff Suovanen -

    “Possible camera combiner chip” is more specifically the OmniVision OV00680-B64G Sensor Bridge

    Trevor Yancey - Reply

    Good find—thanks for sharing! We’ve updated the description.

    Jeff Suovanen -

    With the way TI marks packages, “78CS9SI” is not the part number; use the second number instead. It’s tough to tell exactly from the image, but it looks like it might be “65912,” which would make it the TPS65912 power management unit chip.

    Trevor Yancey - Reply

    Confirmed (again—2 for 2!). Thanks a ton!

    Jeff Suovanen -

    • Popping off one of the two IR emitter rings, we find the elusive eye-tracking IR camera hiding behind a dark filter.

    • Placing a single camera below each eye may limit the accuracy and range of eye tracking—the camera can better view the eye/pupil when the user is looking downward than upward.

    Add Comment

    • Now things get a bit destructive, but it's worth it to get a peek at the optics chain.

      • A tiny ring of six LEDs starts the process—red, green and blue, times two for two focus planes.

      • The LEDs then shine on the LCOS microdisplay to generate an image. It's mounted to the black plastic housing next door.

      • From inside that housing, a collimating lens aligns the raw light output from the LEDs, and is mounted to a polarizing beam splitter.

      • The polarized beams then pass through a series of lenses to focus the image into the entrance gratings on the waveguides.

      • The entrance gratings themselves look like tiny dots embedded in the six (now slightly shattered) waveguides.

    • We pick up the "injection" unit for a closer look, revealing the colors associated with each entrance grating: two red, two green, and two blue.

    Thanks for this. It also explains the process they are using to derive the six planes, which incidentally corresponds to the six color fields the OmniVision LCOS is capable of generating at I assume 1/360th of second per field. However this does not explain how they can sustain a two complete RGB frames in 1/60th of second unless the image resolution is 1280x720 (allowing 120 fps) instead of 1920x1080 (60 fps). Unless they multiplex the six LEDs at 1/360th of second with two frames of image information. Essentially if the resolution is 1080P (unlikely) the NVIDIA TX2 is processing two (1920 x 2) x 1080 images (one for each eye) every 1/60th of second. Pretty impressive, but I think it is actually just 720P as confirmed in my Unity SDK the image is 1280 x 960.

    All Things 3D - Reply

    • Now that we've had our tasty optical treat, it's time to turn our attention to the brains of this operation, the Lightpack!

    • It's hard to miss those prominent cooling vents. Does this little pocket PC have an active cooling system? We'll soon see.

    • These FCC markings don't yield up much, other than it's designed by Magic Leap and assembled in Mexico. The identity of the actual hardware manufacturer is said to be a closely guarded secret.

    Add Comment

    • Cracking the Lightpack open takes a lot of work, but heat and careful prying does the trick eventually.

    • Most VR headsets seem to take after PCs with lots of cables, but this is a single permanent cord—trapped under a status LED strip, some screws, and some copper tape.

      • A single cable to the headset makes for some elegant ergonomics, but your cat better not chew this life line or your device is kaput.

    • Another solid bit of cast magnesium out of the way, and we behold the motherboard!

    Add Comment

    • We ignore the modular headphone jack and button board, in favor of that shield-filled field of silicon.

    • A PC favorite Cooler Master fan graces this PCB, explaining those vents we saw earlier.

    • Twirling away screws isn't enough to free the heat sink, which is very stubbornly glued in place. After a good ten minutes of heating and prying, it finally loosens its conductive grip.

    • This is a lot of cooling for a small, wearable device, but it makes sense given the job it has to do. There's a lot of heat-generating silicon in there—and in this case hot pockets are a bad thing.

    Add Comment

    • A few shields later and its time to look at the chips that make the magic happen:

      • NVIDIA Tegra X2 "Parker" SoC, with NVIDIA Pascal GPU

      • 2x Samsung K3RG5G50MM-FGCJ 32 Gb LPDDR4 DRAM (64 Gb or 8 GB total)

      • Murata 1KL (likely Wi-Fi/Bluetooth module)

      • Nordic Semiconductor N52832 RF SoC

      • Renesas Electronics 9237HRZ buck-boost battery charger

      • Altera (owned by Intel) 10M08 MAX 10 field programmable gate array

      • Maxim Semiconductor MAX77620M power management IC and Parade Technologies 8713A bidirectional USB 3.0 redriver

    do check, if those RAMs are POP (Package On Package), and hide some more processing power under them…

    felixlazarev - Reply

    Yellow box is not Parade Technologies 8713A  , that is WIFI AC, BT MODULE from Murrata

    Dashnamoorthy - Reply

    Thanks for pointing that out, I’d mixed up two chips. It’s correct now.

    Adam O'Camb -

    • And a little more magic on the back side:

      • Toshiba THGAF4T0N8LBAIR 128 GB NAND universal flash storage

      • Spansion (now Cypress) FS128S 128 Mb quad SPI NOR flash memory

      • Texas Instruments TPS65982 USB Type-C and USB power delivery controller

      • uPI Semiconductor uP1666Q 2 phase buck controller

      • Texas Instruments INA3221 bi-directional voltage monitor

    Add Comment

    • Next the saucer section lifts off, leaving the battery housing vulnerable to our prying hands.

    • Getting to the battery was so tough it's almost ironic to find pull-to-remove tabs, but they're better than nothing!

      • All these layers and adhesives probably help with shock-resistance and durability. But when the battery inevitably dies, you're looking at a full device replacement or a tough repair—one likely to be a major headache for recyclers.

    • Magic Leap packs this twin-cell battery sandwich with 36.77 Wh, running at 3.83 V. That's in the same range as some popular tablets.

    Add Comment

    • The Magic Leap One is clearly an expensive, short-run piece of hardware. Every bit of construction is intended to maintain the precise calibration for the life of the device. Our guess would be that this was pushed out at full speed, regardless of the price, to get something on the market.

      • Let’s hope for a consumer edition that maintains the thoughtful design and dedication to durability, while also avoiding the short-sightedness of this device.

    • Special thanks to Karl Guttag of KGonTech, who contributed his invaluable time and expertise to help us navigate this teardown.

    • VR expert, and our sometimes-nemesis, Palmer Luckey also contributed some great content, as well as hardware access. You can read his full take on the Magic Leap One here.

    • Any mistakes in this teardown are almost certainly ours, and not those of our contributors.

    • Lastly, it's time to assign a repairability score.

    Palmer got fired for lying and stealing. He broke signed contracts and without community involvement or their consideration is now a mod at Reddit in violation of several moddiqette rules. He is a joke and you disgrace yourself accepting anything from him.

    Lou Wallace - Reply

  2. Final Thoughts
    • The speakers are easy to remove and replace with just a single screwdriver.
    • The threaded fasteners are all of the standard Torx and Phillips variety.
    • Disassembly is mostly nondestructive—on paper, anyway. With this much glue on this many fragile components, you’d better have buckets of patience and a very steady hand.
    • The battery is only replaceable if you’re willing to remove the motherboard and tiptoe your way past several intense glue barriers.
    • There is no upgrade path for any of the optics or processor—slightly disappointing on a $2,300 piece of gear.
    Repairability Score
    3
    Repairability 3 out of 10
    (10 is easiest to repair)

13 Comments

In Step 10, figure 2 (the figure from the patent) the ‘optional’ polarizing filter (2072) is not very optional. And you skipped it in your summary image.

It’s likely a three-wavelength, 1/4-wave retarder used to rotate linearly-polarized light to circularly-polarized light before it bounces off the LCOS. The reflected light goes through another 1/4 wave so it’s now 1/2 a wave from the input light. That is - it’s linearly polarized 90 degrees from the original light. That’s why the polarizing beam splitter cube is able to separate the input field (100% coverage) from the LCOS-reflected fields. They’re 90 deg. different polarization.

Also, that means the RGB emitters aren’t likely LED’s. They’d have to be laser diodes (or VCSEL’s) to have the narrow spectrum necessary for the 1/4 wave retarding film.

Wick - Reply

Wick, you seem to know something about this subject, but you are wrong about the Figure from the patent. The beam splitter will polarize the light but having two might work better.

In the final configuration and as diagrams by iFixit (the 3rd figure with respect to Step 10) they absolutely need a polarizer (as show) right after the LEDs or else half the light will go strait into the injection optics without modulation. This 3rd figure does leave out a quarter-wave plate and polarizing mirror on the right side of the beam splitter cube that is necessary to bounce the light back to the beam splitter so it can be directed out to the injection logic. As drawn, the light from the LCOS would go through the beam splitter and out the side of the cube.

Karl Guttag -

I want to add to my prior comment that I have only seen the teardown pictures and had to mentally reconstruct what was going on from picture. I did not see the components as they were taken off. Based off the pictures, I was trying to make it work based on what I was seen from static pictures.

Something that concerns me in this design is that the LED light is shot straight at the beam splitter and if the polarization is not great with the first/LED polarizer and the beam splitter, there will be a spot of uncontrolled light (about the size of the LEDs) in the center of the image. I’m assuming they are using wire grid polarizers in both locations.

Another issue with the design is that the image has to pass through and reflect off the beam splitter (if based on what I was shown and was told is correct). Usually with a wire grid polarizer, you don’t want to pass the “image” of the LCOS through the polarizer as it will do damage to the image. The FIG 6 from the patents would be the “normal” configuration.

Karl Guttag -

Ye-gods, thats a lot of impressive technology squeezed into a small package! No wonder it has taken so long to develop. I’m skeptical that it will ever be a commercially viable product. They are surely shipping significant dollars along with each developer unit. My prediction is that, at best, it will be a high-end business-use product, never a consumer product. Most probably MagicLeap will be acquired by a major before too long, and a year later the major will kill it, having found the skeletons in the closet and determined it will never be profitable — especially with consumers now embracing lower-tech, lower AR quality devices. But hey, congrats to ML for being aggressive with technology, and I wish them the best of luck. Thanks for the superb tear-down, finally revealing the long-standing mystery of ML, IFIXIT!

lannierose - Reply

The device is reminiscent of a camcorder, with a lot of optics and electronics compressed into a small volume and not a lot of consideration to repairability. There’s a lot more engineering here than a VR headset and I’m impressed how far Magic Leap has iterated the design for a developer release. That said, having a non-replaceable battery would be a disaster for a consumer release IMHO. Great teardown!

Jack Boyce - Reply

Every bit of the teardown proves my thought that ML was in full “use whatever it takes to get something awesome and tangible into the light of day and hands of tech journalists”.

The idea of ML’s defining headlines being stuff that had words like “failure, $500M+, defunct, etc” was something they were going to break everything to avoid.

This iteration of the tech is a proof of concept device purely for the sake of quieting naysayers and settling the growing number of very worried investors and future buyers.

If this operates at even 75% “WOW Factor” it may earn them the time they need to finish designing their actual consumer level device. I have a feeling their mass production model will barely resemble this. The battery placement and total absense of maintenance/repair/replace indicators kinda proves that.

I am skeptical that the mass-produced model will ever be a reality, but I have a bit more hope than normal now that I've seen the Nissan GTR level of obsessive care in the build quality of this device.

Steven - Reply

Also, I didn't have enough space to write how absolutely blown away I am at the quality of this teardown. Beyond just the physical work, the extra research put into explaining it is brilliant. The confusing bits of tech are easy to understand and including the links tobsources and part data right with the test instead of just a pile of footnotes at the bottom helps let me decide just how deep down the tech rabbit hole I want to go!

I've read and loved many of these from ifixit, but you guys/girls/robots have reset the bar (again) with this one. Well done!

Steven - Reply

I agree, this is absolutely awesome work by ifixit!

Callum Vincent -

Confused with one camera can magic leap track each eye independently?

Jeffrey Cooper - Reply

One eye-tracking camera per eye, you can see the corresponding black square in the lens opposite the one we opened up!

Sam Lionheart -

Has iFixit torn down the Microsoft HoloLens from 3 years ago?

How does Magic Leap One compare to the HoloLens??

John Nguyen - Reply

@johnwnguyen We have not. My understanding is HoloLens uses a waveguide display as well, albeit with a narrower field of view. Magic Leap One also has the benefit of an additional focus plane, although how much that actually contributes to the experience I couldn’t say. If you want the details, I can’t recommend Karl’s blog highly enough—he sniffed out Magic Leap’s trail early and correctly predicted that they’d use display technology more similar to HoloLens than not. (He’s a bit hard on Magic Leap at times, but IMO does a great job cutting through the hype and presenting all the known facts.) Palmer’s writeup also makes numerous comparisons to the HoloLens, summing up Magic Leap One as “slightly better than Hololens in some ways, slightly worse in others, and generally a small step past what was state of the art three years ago – this is more Hololens 1.1 than Consumer AR 1.0.” Worth a read.

Jeff Suovanen -

I’m curious about the microphones, I’ve seen that there may be four of them. Are these in the “sensor arrays”? Where are they aimed? Are they being processed by a chip and then input into the processor, or are they all processed in realtime by the processor unit?

Ian Page-Echols - Reply

Add Comment

View Statistics:

Past 24 Hours: 106

Past 7 Days: 519

Past 30 Days: 3,366

All Time: 97,907