Google Glass Teardown
catwig.com"Although functional, the experience was subpar, because the head proximity sensor was very unreliable when next to a lens."
I turned off my proximity sensor (that's an option in settings -- "On-Head Detection") because the false positives (like turning on or off because my finger covered the sensor briefly) were more annoying than having to turn it on and off with the power button.
So unless there's more use for the proximity sensor I'd say this is a small price to pay if you want to attach google glass to your regular glasses.
One possible use for the proximity sensor, though, is wink detection: https://github.com/kaze0/winky
On the other hand, there's a dedicated button for taking a picture so winking as an input method I think would only be useful in situations where you don't want to use either your hands or your voice. And there may well be such situations.
people haven't learned from microsoft office clipper!
Everyone rather have control than implied ominous prediction.
i hate that i can't press a button on my phone and switch from landscape/portrait modes. rotation sensor is silly most of the time.
It uses an old OMAP4430, also in the Samsung Galaxy 2: http://en.wikipedia.org/wiki/OMAP#OMAP_4
As chips get faster, wearable computing will benefit more than smartphones, because (eg) an iPhone 5 is more powerful than a smartphone can utilize (and more powerful than even a desktop needs to be, for most mainstream tasks.)
For most use cases, current hardware is sufficient. This will have a profound effect on wearable devices as we use exponential improvement to make devices smaller and more battery efficient rather than faster. This is exactly what Intel and Apple did with the new Macbook Air. They made it slightly faster, but much more power efficient.
If you look into the future and imagine that every 1.5 years Google Glass will half in size instead of getting faster, it doesn't take the long before it's the size of a dime (~4.5 years).
That's why Glass matters: it's just the beginning.
Short blog on Medium about this: https://medium.com/adventures-in-consumer-technology/74c201b...
I don't know if I would classify the OMAP4 as "old", unless you're Apple and spinning Cortex-A cores on your own silicon.
The OMAP4 is really no different than the Sitara AM335 being used in stuff like Beaglebone Black, with the exception of the video DSP. And I can see why Google would want that DSP on board to do things like hardware video encoding.
The chip is taken from the specs, you cannot see it in the photos.
I think the memory is mounted on top of the SoC as a package on package.
Great, now put it back together. I am the worst at reassembly of electronics I would have wasted +-$1,500.
I put it back together and it works fine! Only, it has suffered some irreversible cosmetic damage..
Ideal for those who always dreamed of becoming a spy (or Stasi informant). $1500 gets you a stealthy camera with NSA uplink.
Stealthy? Have you actually seen somebody wearing glass? It's the most noticeable thing about them. During IO there were a bunch of Glass wearers wandering around Moscone and people couldn't stop staring (myself included).
You'd be much better off with something like this: http://www.amazon.com/Waterproof-Camera-Watch-Video-Recorder....
Sure, everyone can see they’re wearing glasses. However, outside of our little tech circle, most people don’t know that those glasses contain a camera, mic, and location tracker. And even if a Glass-wearer gets called out, they can deny that they were recording or taking pictures (unlike if you point a camera watch at someone).
You could do much better by getting rid of the HUD entirely and embedding just a camera lens/sensor, which can be really small, into some normal glasses ... for spy pictures you don't need perfect composition ... 😓
So based on the teardown, anyone want to predict what the MSRP will be?
499$ Glass is the new iPad.
All kidding aside, Glass will go head to head with iWatch or whatever else wearable devices that the various companies come out with.
A watch computer != a HUD.
A watch could never do augmented reality overlays -- heck, even a phone can do those better than a watch ever could. IMO that's like saying the iPad will go head to head with laptops -- it's an apples-to-oranges comparison.
Glass can't do AR overlays either, unless you only wanted to overlay a small area in the upper right of your vision.
It's still a step up from holding your smartphone at arms length in front of your face. You have to start somewhere =)
And yet the iPad _is_ going head to head with laptops and seems to be destroying them, despite their obvious differences.
You're right: a watch != a HUD. However, in the current iteration, people would be using a watch for pretty much the same purpose as Glass (which can't really do full on augmented reality either) so, despite their differences, they would definitely be in direct competition with each other.
Is it? Maybe the iPad is supplanting laptops in some niches, but I'd argue that those are ones were maybe a laptop wasn't the proper device in the first place.
AIUI Glass can't do augmented reality only because the API is currently too restrictive, which is really too bad. Using a HUD as a glorified head-mounted camera seems like a huge waste of opportunity. Even if they never open up the API, a HUD can still do things that a watch is going to struggle to replicate.
This[1] is a good overview of what it takes to make an AR device, and it examines the Glass hardware in respect to that. The verdict? The hardware is there (although it's not stereoscopic) but the software is behind. You have to start somewhere, though.
iGlasses.
Very well presented and designed too. Thanks for sharing!
I see a chip marked SIRF - does it have GPS?
It looks that way. The SIRF chip is marked as GSD4e.
http://www.csr.com/products/35/sirfstariv-gsd4e
You can see more detail of that here:
http://www.catwig.com/google-glass-teardown/teardown/mainboa...
I hope so, as this will enable Google Maps-like functionality.
Yes it already has inbuilt GPS unit. Source: I own one
Then why does it need the MyGlass app running for GPS?
Probably for AGPS and a better positioning using both GPS chips and other positioning sensors (cell towers, wifi). The Glass GPS antenna is tiny, it can take ages to synchronize without aid.
"Ages" must mean more than half an hour inside a car, then. If I don't have the MyGlass app running on my phone, the location accuracy is so limited that I can only assume it's based on WiFi.
But what do you mean "without aid"? Even without MyGlass running, Glass still has WiFi-based location and an Internet connection (tethering is separate from MyGlass) -- isn't that sufficient for AGPS?
Battery life I imagine.
What are the QR codes used for?
I parsed two of them and came up with "|224010712|01294" and "0200014672"
As a complete guess: Much electronics manufacturing requires items to have individual (or batch) numbers for quality control.
When items fail test you can investigate the batch to find and eliminate problems of manufacture.
Sometimes this is good and effective and it works. You see that batch #3762387 failed, and that they all used a component from a delivery, so you look at the other batches using that component and they fail too.
Sometimes, however, it's just a paperwork exercise. It's frustrating for everyone trying to do the work, and the results are hopeless for anything because people are just faking the paperwork, or the paperwork is garbage. The company is only doing all of this to get a logo for their letterhead, and they must have that logo to do business with some other companies.
In the UK some of these systems (ISO900x; BS5750; etc) are sometimes seen as expensive makework nonsense.
Let me know if you ever need to find flaws in someone's paperwork because there are a few things where it's trivially easy to trip them up once you know what to look for.
The quality control and tracking reasons makes sense. It would allow for faster and more automated identifying of parts.
"Let me know if you ever need to find flaws in someone's paperwork because there are a few things where it's trivially easy to trip them up once you know what to look for."
I'm interested in this, can you give some more details?
Electronic companies use ElectroStatic Discharge (ESD) protection. Often this is the form of a conductive wrist strap connected (via a safety resistor) to ground. Operators are usually required to check that this strap works using a test station. They then have to sign a sheet.
The real function of this: You want to know if Bob has a faulty strap, so that if a batch of product is failing, and you see that Bob is the common element, you can check Bob's paperwork. When you see that Bob had a faulty wrist strap on the 14th of June you have found the possible cause of the faults, and you have some acton you can take.
Several things actually hapen: The wrist strap tests are offered as "Pass" or "fail". No-one likes saying they have failed a test. People think the purpose is just to get a working strap, so if it fails they'll do some fiddling, wiggling the wire until it passes, or licking their wrist. These temporary measures fail when they're back at the bench, but the paperwork doesn't reflect that. And people often forget to do the test, so they'll just sign off a bunch of days when they do remember.
Thus, when you visit a factory you can gauge the understanding of QA procedures by looking at these kinds of paperwork. You'll see a sheet full of signatures. That looks great, until you realise that this factory has almost zero absenteeism, and people rarely take holidays, and sometimes people have tested their wrist-straps on public holidays when the factory was closed.
When you have a barcode printed on a device it's hard to lose track of the numbers, but sometimes route-cards are just bits of paper attached to a box. Ask people how they can be assured that the devices in that box belong to that route-card.
Paperwork is often designed poorly, and is onerous to do. It's often kludged in from above, rather than reflecting the actual job. Give people extra work, while pressuring them to get product out, and they will take shortcuts. That's often the paperwork. Often just asking people about the paperwork, in a sympathetic voice ("Oh wow, all these forms, eh? Which are the useful ones, and which ones are a bit annoying?") will get remarkable answers.
Framing language is handy. Ask people using solder paste (screen printing circuit boards with paste prior to surface mount pick and place) about "waste" - they'll say they don't waste anything. Because no-one likes to waste stuff. But actually, paste has a life, and waste is part of production. A good answer would be "We like to keep the paste clean for production, so once it gets old we move it through to jobs with less-fine pitch components, or we use it for rework, or we use it for training. I guess you could call that waste. At the end of life we carefully dispose of it to recyclers, along with our other lead dross and tin / copper snippings" but people will tend to say "we don't waste anything" and you can ask them "so, you re-use that paste? It goes back into the pot, and you use it again?" and they'll say "yep".
Some of these are trick questions and don't really tell you much about actual production. You just get to know that the QA logo is just paperwork exercise.
If it's any reassurance the QA I experienced for ground-side aviation stuff was much better - it was rigorous and detailed and investigative and we did it all properly, and the auditing was very very good.
Most likely for tracking parts during production.
Well this just makes me want Glass even more - I wouldn't be able to resist taking it apart!
In public surveillance terms, this is basically the equivalent of gluing your camera to your temple. The Snowcrash quote is pertinent. Glassholes will be excluded from free spaces.
I don't get why glass is so different from ubiquitous smart phone cameras or (in the UK) CCTV.
Parents already have problems taking photographs of their children at school events[1] and people are regularly harassed and arrested because they are taking photographs in public. Most people seem to think this is too extreme, but I get the impression that they wouldn't mind if it was happening to glass users.
As far as public surveillance goes there are much worse situations than actual people wearing (currently) expensive hardware wandering around. It's cheaper and better to mount CCTV everywhere; and it's good for the operators to have hidden[2] low[3] level cameras.
[1] Although I agree with Louis CK about this - don't video the kids, watch them. The resolution on the kid is much better than on that little phone screen.
[2] To avoid vandalism.
[3] Face height, not on a tall pole.
One stays in your pocket.
The difference between Google Glass and smartphones is that with Google Glass it is much more difficult to know if someone is recording you.
Well, yes and no. They'll have to basically be staring at you for the duration of the video - which is somewhat normal in one-on-one conversations, but sticks out much more in group settings.
You'd think Star would be a little bit gunshy about wearable electronics. I kid, Star if you are reading this, I'm kidding!
Down for me due to a Cloudflare error: "DNS Points to Prohibited IP".
Well done. Nice site design and beautiful high res images.
Can anyone spot where the TI Omap 4 processor is in this?
Yep, it's right here: http://www.catwig.com/google-glass-teardown/teardown/mainboa...
Of the three largest chips, it's the furthest to the right (in case the large Texas shape didn't tip you off ;) ).
That's a TWL6030 power management chip, used for charging and powering the board from the battery. There are no side photos, but sometimes there is a trick you can do with the OMAP: put it under the RAM. It can provide pins for the RAM to be soldered on.
Yup, the BeagleBoards do this.
Yeah it should be under the Elpida memory module. We did not desolder that for this teardown.
You just open sourced Google's IP related to Hardware Design of Glass. Now just waiting for cheap knockoff's from China flooding the Walmart's of the world or flooding eBay catalogs.
Hardware is going to be easy to copy.
Google services (search, maps, etc)? Good luck to the imitators.
They could just use the software that is on Glass now. If the hardware is exactly the same it could work.