I know, I agree, Rick. It's an enternal problem that excites engineers, Qualcomm CEO included. On the other hand, I wonder if Steve Jobs would have been so interested in the smartwatch...I am just not so sure.
As I said, mirasol fascinates me. The implementation in the TOQ appears to be a standard approach: the tri-colored 3-pixel system used for color TV. The way the pixels get their color and are controlled is the wonderful part. The nearest analog I can think of right now is the Japanese Matryonim (created by Masami Takeuchi using phenomina discovered by Lev Termin and incorporated in his invention, the Theremin.)
The Matryonim uses body capacitance to tune one of two radio-frequency oscillators. The other oscillator is fixed, and the two signals are mixed (a form of interference) and filtered to produce the difference (beat) frequency, which falls in the audio range. In the Theremin, a second circuit (also using body capacitance) controls the volume output of the device, but the Matryonim have no volume control. Instead, the player simply removes their hand (which provides one plate of the tuning capacitor) to a distance that causes the pitch output to fall below the audible range. Within the range where the beat frequency is an audible pitch, the player chooses the pitch by adjusting hand distance and controls the pitch accuracy via hand-ear coordination.
The elements of the Mirasol are similar, except instead of using radio frequencies and a mixer with filters, the Mirasol element is an etalon with movable plates. Optically, the etalon passes (or reflects) light related to the distance between the plates. To get red, green and blue pixels, adjacent Mirasol elements have different "on" distances. To turn the pixel on, the two 'plates' are driven apart to mechanical stops which provide the proper distance to pass the desired color. To turn them off, the plates are 'closed' so that the light they reflect is outside the visiable range, similar (but not 'like') the Matryonim. (The black off-color comes from a black coating on the plate further from the eye.)
If the plate separation could be driven to a desired distance and left, without having to maintain drive to keep them there, then it might be possible to produce the complete color range from only one cell. You can't get white from a single cell, though: white isn't in the spectrum, it is a combinaton of lots of different wavelengths (or at least three!). But in cases where white was needed, devoting multiple neighboring pixels to provide the needed multiple colors (just as they do now) is possible. That means that a multiple-position plate pixel would triple the resolution over the current RGB approach. Color accuracy could be very good at preset distances, which is good, because individual chroma detectors for feedback would really complicate the assembly!
I don't believe we get that from any other display technology, at this point, and I don't know that we could. LEDs and similar emitters tend to be limited in wavelength emission, or quite underlimited.
And while that would be great, we have to keep in mind that we would get only hue variation, but not luminance: on the CIE chart, we'd get a thin-line U shaped choice of colors from a pixel. Some brightness control in local areas could be obtained with dithering (at the cost, once again, of resolution.) Coupling a black-and-white array of emitters which can be individually controlled for brightness with an array of multi-distance or continuously-variable-distance Mirasol elements would be... well, pretty awesome!
I tried to email you, but the address (with and without the leading 'b') bounced.
I don't claim expertise, but it's something I've spent a lot of time researching and considering. The big question: How can you interface a human with a computer in a useful way that doesn't involve distracting, distorting or really objectionably noticeable action on the part of the human?
The Computer->Human side is getting easier and easier. The Google Glass, now there's a great approach: beam the visual output directly into the eye with proper optical techniques (and power limits) to avoid distortion (or eye damage), and provide fairly inconspicuous earphone/s. But going the other way? [Cell phone gaffe: go to the bathroom. A familiar voice from the next stall says, "Hi! I haven't seen you in a while--what have you been up to?" You start to answer, and the owner of the voice continues the conversation he's having on his cell phone...]
There have been a few approaches. Voice recog is fine...but you're perceived as talking to yourself, or worse yet, someone in the vicinity. (Even worse yet, you're perceived as hebephrenic, and they send for the nice young men in the white coats...)
I've considered morse code via a simple 2-contact (ring and fingertip, two fingertips, etc) approach. Bite switches are possible, but interfere with normal talking and wires and bulges are unfortunate.
Basically, the problem is wearables without earning Neal Stephenson's "Gargoyle" appelation.
@Junko: Good point. The concept of the smart watch is like the perpetual motion machine. A facinating Rubrik's Cube no one has solved yet but many engineers would like to. How to make something so small and intimate really compelling???
Marc Andreessen in an interview here said the potential of the earpiece in Google's Project Glass is overlooked for its significance: voice reco and synthesis with an Internet connection will be significant, he said. And I think he's right, especially if you add the stereo audio stuff Toq has and you discuss in your comment.
I'd love to follow up with you about your views on wearables and displays, etc. You seem to have some deep expertise there. I'm at firstname.lastname@example.org
@Bill: Note Qualcomm is carewful to say Toq is NOT replaceing a smartphone. It's a second screen for a smartphone--a quick way to check notifications from the smartphoe and ndecide if you want to go into the phoe and do something.
I doubt SmartWatches will replace our current SmartPhones. Display size might be one reason.
Where I do see an interesting market for SmartWatches is in monitoring health (diabetes, heart attack, etc) back to the wearer. If the consumer's health needs continual monitoring by a doctor, I could see these watches also sending information to their doctor or a nearby hospital (yes, watch would also send GPS location). Plenty of hurdles to leap before this can be enabled.
To answer your "what do you think" question, I personally like have mechanically complex watches on my wrist! Much cooler IMHO!Of course, these have little microp's too. But a solar powered, 5 motor, world time chronograph seems to satisfy the engineer nerd in me better than a digital display. Perhaps of the smart watch had a 5 inch screen...
It's interesting that all of the sudden smart watches are getting so much attention but Motorola's Action watch series pretty much got almost no attention. In fact, I believe that they've discontinued it. It was really designed to work in conjunction with your phone (which is good from a service standpoint) but it was pretty much a full Android platform on you wrist. It was target at the exercise market but these other products really seem to be "me too" offerings.
A Book For All Reasons Bernard Cole1 Comment Robert Oshana's recent book "Software Engineering for Embedded Systems (Newnes/Elsevier)," written and edited with Mark Kraeling, is a 'book for all reasons.' At almost 1,200 pages, it ...