Thursday, April 26, 2012

iGlasses: Apple's New TV Set Isn't a TV Set


Some recent news items and a host of Apple rumors are converging into a surprising picture; a picture that is very small, close to the eye, three-dimensional, and retina-display quality. A plethora of sources has reported on Apple's supposed forthcoming foray into the realm of television sets. Imagine that instead of a giant flat panel display with built-in computer intelligence, Cupertino releases a small, lightweight, relatively inexpensive headset that acts as a personal viewing device. I'll call the product the iGlasses. It may seem far-fetched, but recent news and some basic mathematics show that it may in fact be possible if not probable.

What Are Apple iGlasses?

Head-mounted display

Dimensions: lens area: 3-4"w x 1.5"h x 2.5-3"deep
Displays: Two stereo 1.5" x 1.1" Retina Displays capable of semi-transparency
- Resolution: 2048px x 1536px, 1390ppi
Operating System: iOS 6
Announcement: June 11, 2012
Availability: September 2012?
Price: $349-$399

Interface: 

iGlasses will require an iOS device to function. The app downloading and selection and most of the processing power will be handled by an iPad, iPhone, or iPod touch. The display elements will be streamed to the iGlasses via AirPlay. Since iGlasses' screens will be capable of turning semi-transparent, the iDevices will be able to be operated while wearing the glasses.

Additional functionality may be handled by Siri. iGlasses could contain a microphone and iPhone 4S's extra processing chip required to improve speech parsing. This will allow for voice commands like changing channels if television streaming is enabled or choosing movies through an Apple TV-type interface.

Evidence

Recent News

Google recently made a splash in the tech world with rumors and a concept video of Project Glass, an eyeglasses type Heads Up Display. The glasses would provide realtime information such as directions, translation, restaurant reviews, and other things, based on what the they "see" in the environment. Initially rumored to be available in 2012, Google has more recently denied that they will sell this year.
In another tidbit, news from last fall's IFA trade show included the announcement of Sony's HMZ-T1 Personal 3D Viewer. While this flew under the radar for almost everyone, it was an important marker that at least one major electronics company sees the field of personal headset viewing devices as worthy of pursuit. While the original i-glasses by i-O Display Systems date all the way back to 1995, only recently have technological advances made the idea of a high resolution, light, untethered 3D head-mounted display feasible. Sony's implementation doesn't hit the mark, but it could be the vanguard in a nascent market that Apple could jump into and dominate just as they did with digital music players and tablets.

In other news, early last year a company called QD Vision announced a micro-display system with a resolution of 800x600 pixels in only a 15.6mm (0.6") display. Using "quantum dot light emitting diodes" (QLEDs) allowed a super high resolution, bright, and color-accurate display to be produced inexpensively and on thin, flexible, or semi-transparent materials. While full-color QLED displays were not yet available as of January 2011, they were in the pipeline at QD Vision.

The Numbers

With these building blocks in place, let's turn now to the mathematical support for the feasibility of iGlasses. What type of display would make sense for this product? What resolution would it be? What size? Obviously, there would be two of them, a twin panel for each eye in order to produce a stereo, three-dimensional effect. Each panel would be small but have to have high enough resolution to qualify as Retina Display in Apple-speak.

So as to the question of resolution, Pilky.me published a good article on calculating the pixel density required to create a retina display at a given viewing distance. If we start with the guess that each screen would sit about 2.5" in front of the corresponding eye, the necessary density would be a whopping 1,375 ppi. At first this seems untenable, but the QLED display just mentioned has a 0.6" diagonal display at 800x600. This equates to 1,628 ppi!

As far as size, if we begin with the assumption that any new iDevice will fit into the existing display resolution brackets, we can assume a third-gen-iPad-esque 2048x1536 resolution. Given this resolution, the aforementioned viewing distance, and a retina display, we can calculate the necessary screen size. Each screen would be 1.8" diagonally; 1.5" wide by 1.1" tall. This yields 1,390 ppi. At a distance of 2.5" this is 101% of retina quality. This screen size also happens to be exactly three times the size of QD Vision's existing display.

But wait, there's more. For a head-mounted display to be comfortable and to avoid eye strain, the screens should be placed directly in front of each pupil.  Pupillary distance is the amount of space between the pupils of the eyes. Average pupillary distance ranges from 41mm (1.6") in children to 70mm (2.8") in large adults. Since the proposed iGlasses screen is 1.5" wide, the closest pupillary distance would be just over 1.5" (to accommodate the housing). Thus, the device would be exactly the right size to adjust to fit almost any size face.

Price

Aside from industry-leading specs, another major factor leading to massive sales will be the price. As mentioned, I believe iGlasses will require an iOS device to function. This key factor enables the new visors to be small, run wirelessly, have excellent battery life, and perhaps most importantly, sell for a reasonable price. Making many assumptions based on the pricing of third generation iPad and iPhone 4S components here is a rough breakdown: DRAM $10, displays $80-$100, processor $15, cameras $5, sensors and wireless $15, power and battery $15, body $25, manufacturing $15. The total Bill of Materials would be around $200. Given the standard Apple markup of about 80% (before other expenses which bring profit margins down to around 35%), iGlasses will cost about $350. Most Apple products are priced at increments of $100 ending in $99, so perhaps the final price would be $299, with Apple taking a slightly smaller profit in order to expand the new market. (Apple's recent guidance for Q3 2012 is even lower than usual.) Considering the Sony product sells for $800, this would be a very reasonable offering.

Timing and Operating System

Given all of the above specifications, many new APIs will be required for iGlasses. While graphics will fit the current iPad resolution, the user interfaces will need to be redesigned. Adding 3D will also increase the new burdon on developers. For these reasons, iGlasses will require a new major update of iOS, to version 6. The logical time to announce these sweeping changes and a whole new product category will be Apple's World Wide Developer Conference, WWDC. The keynote on June 11, 2012 is the obvious date.
These changes will require time to implement but Apple would naturally want to be poised to sell iGlasses for the holiday season. They also would want to bring the product to market as quickly after announcement as possible in order to minimize the Android-esque mimicking of the product by other companies. Since many new products are released to coincide with the beginning of the school year in September, I suspect this time frame for iGlasses too. This gives three months to ramp up production, finalize any necessary certifications, and allow developers to update their apps. Apple would be well positioned to sell many new devices this winter.

Apple Advantages

How would Apple create this new market where none has previously existed? As mentioned, head-mounted displays have existed for almost a decade but have never caught on. Previous devices have suffered from several fatal flaws which Apple is in a position to overcome.

First, all older devices were tethered, requiring a direct connection to a video source and often a power supply. Apple already has AirPlay technology allowing wireless streaming of content from one device to another. Secondly, while the screens of prior implementations were indeed immersive, they were not very high resolution. Seeing pixels and the space between is a sure way to snap someone out of the suspension of disbelief. Apple pioneered the Retina Display and as demonstrated, they could extend the same resolution to the iGlasses using existing technology. In addition, motion tracking in past head-mounted displays has been a hit-or-miss affair. The latest Sony headset has not even included the hardware. With the years of experience and software behind accelerometers and motion detection in iOS devices, Apple would likely provide a seamless implemetation in iGlasses. A final problem with older visor displays was the overall design. The products have ranged from close to a pair of sunglasses to the complete Robocop meets Geordi La Forge wraparound. I'm confident that Sir Jonny Ive and his team could come up with an elegant, fashionable design that people would actually want to wear.

There are other factors that could distinguish Apple's offering.  Siri could naturally extend its current functionality to serve as  a convenient primary interface. Whereas devices of the past required manual adjustment and calibration, Apple engineers could conceivably use a small camera above each eye to automatically set the focus and pupillary distance. These cameras could also add eye movement detection that would make 3D games truly amazing. Also, imagine reissued 3D versions of all the classic Pixar movies, specially formatted for the higher-than HD high-res display. Given the tight bonds between Apple and Pixar/Disney, this would certainly be possible. These possible additions would make iGlasses a popular choice.

Example

Will Apple really produce a head-mounted display? Rather than speculate further, judge for yourself. You can print out my template for the proposed iGlasses display (though based on printer limitations it oviously will be only 600 dpi). Cut and fold according to the directions. Hold the paper display touching just below your eyes. The image will be out of focus, but you can still get a feel for the screen size. (This may be affected by the focal optics in a functioning device, probably making it appear bigger.) Next, set your vision so that the two images of the ball or player align (though they will still be blurry). Then move the paper slowly backwards until you are able to focus. This will give you a sense of the type of 3D possible. This example (based on Isaac Cheung of Isaac Cheung of ViCGI) uses only four parallax-type layers, whereas 10s of layers of depth will be available to developers.