You know that Kindle you're so in love with? Man, that's so 2009. Mirasol showed up at CES with a display technology that's been in the works for around a decade, and it has finally reached the point where ODMs and OEMs are talking shop about integrating said panel into all manners of future devices. In the past, the company has stuck to making screens for relatively minuscule devices, but its latest display is a 5.7-inch XGA color screen that would fit just perfectly within your next e-reader. We spoke at length with executives, and they're currently predicting an ebook reader running Mirasol in Fall of this year. What we've heard "around" is that the device will actually be new Kindle running a Mirasol display -- we can't confirm it, but it sounds logical. What's interesting, though, is that this panel is not only color, but it can play back video when used with the right processor. In other words, a powerful enough Mirasol reader could double as a small tablet, and we imagine you could actually browse the web with some level of enjoyment. The demo we were shown (vid's past the break!) was nothing short of stunning, and we were told that it could scale up to fit within 9 to 10-inch slate / tablet PCs. It requires no backlighting whatsoever, and instead relies on ambient light to show its stuff. We were told that future devices would be priced competitively, even to LCD / e-ink rivals, and we can safely say we'll be bummed if we hear the word "delay" when autumn rolls around. Don't make us frown, Mirasol.
@scratchfury but they don't keep the picture without power going to the display. with this you don't need power while reading an ebook, only when turning pages
@Xstream actually mirasol does need power to sustain its image according to wikipedia, however it a relatively low amount of power. E-ink, needs no power.
@Xstream I lied, after further reading of a qualcomm pdf, it seems mirasol is bi-stable and can retain an image in the absence of power. (which is a good thing)
@R E Hmm, the official site claims that the Mirasol technology is bistable. Though the exact description suggests that pixels do indeed draw a minute amount of power in their retracted state.
I have some technical questions to pose to whomever.
1. What happens to non-reflected ambient photons? I suppose they pass straight through the device ala an etalon, else they're absorbed by the lower conductive plate. Either way, doesn't this mean one could adapt this into a solar panel, accumulating energy from all the ambient light that isn't being used to display? A combined display/solar panel would be a great boon no?
2. How does/does this device vary reflected light intensity? It appears the device can only control the reflected colour and not the rate of photons being reflected. How would one then achieve greyscales? Would one be forced to use dithering? In which case would we have to sacrifice resolution for intensity precision? Else would it be achieved by rapid pixel switching ala DLPs? Wouldn't that put a great strain on power usage?
3. Can this reasonably be used to produce colour samples as a continuum of wavelengths for a truer human response (presumably again at the expense of resolution) or will each imod device be fixed to a single (or small set of) predefined constructive wavelength(s)?
1. This device is like a "folded etalon" and I believe the unreturned photons are absorbed. I have heard rumors that the QMT guys have considered solar panel applications... 2. With RGB subpixels that are either "off" or "on," there are two ways to achieve grayscale. The first is to divide up the subpixels and address each section separately. This is painful for layout and driver design of course. I know that demos shown at SID in previous years did this with 3 or 4 sub-subpixels. The other way is through temporal dithering. Driving at 120Hz for a 30Hz video would give you 4bpp for example, but air damping of the floppy membrane makes this tough at higher frequencies. 3. My understanding is that current designs are RGB pixels that are either off or on, but an "analog" version is the ultimate goal.
@R E et. al. This is not truely a bistable device. A voltage is required to hold a subpixel in either its off or on state. Additionally, to avoid charging which causes stiction, the polarity of the device must be constantly switched. however, to hold a static image, the polarity only needs to be switched at low frequency (like 1Hz). Since the device is like a big parallel plate capacitor, leakage currents and polarity switching cause power consumption...
@displaysRus A slight correction: driving @ 120 Hz for a 30 Hz video gives you 5 distinct levels (0,1,2,3,4), assuming each period is equal length. This is a little over 2 bpp, not 4, which implies 16 levels. To get 4 bpp, you need periods which are of relative size (8,4,2,1), though this really implies driving at 15x the base frequency.
I'd imagine that bi-level-based systems would tend to use both temporal and spacial dithering together to get the best cost/benefit trade-off.
What they need is a small not too expensive trade-show&tear-down camera that can focus closer than 2 feet, now that would be a novel and bombshell device, one day, in the far future.
@Wwhat I agree, and maybe try some different angles to show glare and viewing angle. I know the presenters want to show it at it's best, but a gadget blog should try to cut through that presentation to see the shortfalls in tech too.
From what I understand of the Mirasol technology it cannot be BACK-lit. EInk has the same limitation. It is however possible with Electrowetting (and obviously LCD).
Booklights work pretty well with eink though. And I imagine with this too.
That could be done. Still the Sony PRS-700 demonstrates the problems with having a built-in front-light. The diffuser necessary to light a screen that size more evenly also causes some blurring.
How about a standard issue book light? Like the kind you clip onto the cover of a book and it pops up? Would that pose the same problems the "integrated" book lights Sony has? Because if it doesn't one way of providing light is to have the external light pull out from the casing of the book and hover over it. And when you don't need it, it stores away out of sight.
@(Unverified) You obviously haven't read on e-ink then. Or you don't read anything longer than the first 5 comments on an engadget article. Have people been bothered with books that aren't "nice bright"? Some lighting tech is a good idea, but except for a few situations (planes, next to a sleeping partner, outside during a meteor shower) there us usually plenty of ambient light.
I knew if I waited for CES to buy an e-reader I'd end up delaying even more. Now it looks like instead of waiting till spring I'll be waiting until Fall and then some. E-reader overload of course who knows maybe Apple will be releasing the" tablet/slate/whatever" at the end of the month with a screen in it just like this one. I wouldn't put it past them.
@(Unverified) Haha...they should really get a white balance reference when they go and hold it in the frame so they can correct white balance as they go. Much of the real photography of Mirasol displays has been done with terrible white balancing in terrible lighting. Wonder if there'll be better shots during the show, under some better light.
@BubbaJ I can attest that this technology currently has limited color gamut and depth. You can see in the pictures that the colors are pretty washed out.
So this makes me wonder how well it can integrate with touchscreen methods.
Obviously you could put a resistive layer on top, but that leads to crappy contrast and glare as Sony has so graciously provided example of.
Plastic Logic has a flexible substrate so you can put resistive touch underneath, but I am guessing Mirasol is not going to be flexible.
Wacom? Even if it didn't interfere, the iRex DR800sg reviews have shown us folks don't like a stylus only interface.
That leaves Capacitive. Could a capacitive interface work without interfering with the screen draw?
Ultimately I can't imagine this thing succeeding in a tablet format without touch.
Oh and why is that when someone is filming a new display tech at a trade show do they not ever change the camera angle so folks can get a feel for the viewing angles and glare. Seems amateurish.
As far as why they don't change the camera angle, it's often because there are too many people crowded around the product, and the camera guy is actually several feet back with the camera zoomed in.
@ DisplaysRus Ta very much. On a parallel note, do you happen to know if there exists such an 'analogue' varient for DLPs? Something able to bend the mirrors to arbitrary angles rather than juat on or off the lens? I figure if I got a knowledgable guy on the line, I gotta ask everything!
@elmer Not as an array of pixels, to my knowledge. I can't think of any purpose for that. The DLP is designed to point the mirrors at a specific angle. However, there are single mirror devices that can be driven to many angles. A single axis mirror is used in some laser bar code readers, for example. A two axis mirror is used for rastering a laser to create an image in Microvisions pico projector. Mirrorcle is one company that also produces MEMs mirror modules. Perhaps there are other like devices used in optical network switching..
Now that we've thrown 'em off the trail, use the form below to get in touch with the people at Engadget. Please fill in all of the required fields because they're required.
posted January 7th 2010 3:46AM
Always room for innovation in the e-reader market.
How is this technology different than transflective LCD? My Blackberry 7250 and Game Boy Advance were sunlight readable with video and no backlight.
@scratchfury but they don't keep the picture without power going to the display. with this you don't need power while reading an ebook, only when turning pages
@Xstream
actually mirasol does need power to sustain its image according to wikipedia, however it a relatively low amount of power. E-ink, needs no power.
Besides what Xstream already mentioned, Mirasol and most other e-paper technologies also offer a better viewing angle and contrast.
@Xstream I lied, after further reading of a qualcomm pdf, it seems mirasol is bi-stable and can retain an image in the absence of power. (which is a good thing)
@R E
Hmm, the official site claims that the Mirasol technology is bistable. Though the exact description suggests that pixels do indeed draw a minute amount of power in their retracted state.
http://www.mirasoldisplays.com/mobile-display-imod-technology.php?p=2&techID=2#
@Plothole Yay for accuracy!
@R E and others
I have some technical questions to pose to whomever.
1. What happens to non-reflected ambient photons? I suppose they pass straight through the device ala an etalon, else they're absorbed by the lower conductive plate. Either way, doesn't this mean one could adapt this into a solar panel, accumulating energy from all the ambient light that isn't being used to display? A combined display/solar panel would be a great boon no?
2. How does/does this device vary reflected light intensity? It appears the device can only control the reflected colour and not the rate of photons being reflected. How would one then achieve greyscales? Would one be forced to use dithering? In which case would we have to sacrifice resolution for intensity precision? Else would it be achieved by rapid pixel switching ala DLPs? Wouldn't that put a great strain on power usage?
3. Can this reasonably be used to produce colour samples as a continuum of wavelengths for a truer human response (presumably again at the expense of resolution) or will each imod device be fixed to a single (or small set of) predefined constructive wavelength(s)?
@elmer
1. This device is like a "folded etalon" and I believe the unreturned photons are absorbed. I have heard rumors that the QMT guys have considered solar panel applications...
2. With RGB subpixels that are either "off" or "on," there are two ways to achieve grayscale. The first is to divide up the subpixels and address each section separately. This is painful for layout and driver design of course. I know that demos shown at SID in previous years did this with 3 or 4 sub-subpixels. The other way is through temporal dithering. Driving at 120Hz for a 30Hz video would give you 4bpp for example, but air damping of the floppy membrane makes this tough at higher frequencies.
3. My understanding is that current designs are RGB pixels that are either off or on, but an "analog" version is the ultimate goal.
@R E et. al.
This is not truely a bistable device. A voltage is required to hold a subpixel in either its off or on state. Additionally, to avoid charging which causes stiction, the polarity of the device must be constantly switched. however, to hold a static image, the polarity only needs to be switched at low frequency (like 1Hz). Since the device is like a big parallel plate capacitor, leakage currents and polarity switching cause power consumption...
@displaysRus A slight correction: driving @ 120 Hz for a 30 Hz video gives you 5 distinct levels (0,1,2,3,4), assuming each period is equal length. This is a little over 2 bpp, not 4, which implies 16 levels. To get 4 bpp, you need periods which are of relative size (8,4,2,1), though this really implies driving at 15x the base frequency.
I'd imagine that bi-level-based systems would tend to use both temporal and spacial dithering together to get the best cost/benefit trade-off.
What they need is a small not too expensive trade-show&tear-down camera that can focus closer than 2 feet, now that would be a novel and bombshell device, one day, in the far future.
@Wwhat I agree, and maybe try some different angles to show glare and viewing angle. I know the presenters want to show it at it's best, but a gadget blog should try to cut through that presentation to see the shortfalls in tech too.
this looks really promissing, can't wait to see what devices make use of it and how it works. The ui looks spiffy though.
i have nothing to say but, awesome!
I really hope they come up with some sort of backlighting tech , coz i like to read my books in the dark too !
From what I understand of the Mirasol technology it cannot be BACK-lit. EInk has the same limitation. It is however possible with Electrowetting (and obviously LCD).
Booklights work pretty well with eink though. And I imagine with this too.
@Plothole
Front lighting?
That could be done. Still the Sony PRS-700 demonstrates the problems with having a built-in front-light. The diffuser necessary to light a screen that size more evenly also causes some blurring.
@Plothole
How about a standard issue book light? Like the kind you clip onto the cover of a book and it pops up? Would that pose the same problems the "integrated" book lights Sony has? Because if it doesn't one way of providing light is to have the external light pull out from the casing of the book and hover over it. And when you don't need it, it stores away out of sight.
@Wolfticket and Plothole,
Front lighting is possible and QMT has developed this. Presumably any real product would include front lighting.
Finally, a screen that will match the awesome visuals of Duke Nukem Forever. I'm sure both will be bundled together.
I'll take less battery life and a nice bright screen any day...
@(Unverified)
Along with copious amounts of eyestrain...
@(Unverified) You obviously haven't read on e-ink then. Or you don't read anything longer than the first 5 comments on an engadget article. Have people been bothered with books that aren't "nice bright"? Some lighting tech is a good idea, but except for a few situations (planes, next to a sleeping partner, outside during a meteor shower) there us usually plenty of ambient light.
I knew if I waited for CES to buy an e-reader I'd end up delaying even more. Now it looks like instead of waiting till spring I'll be waiting until Fall and then some. E-reader overload of course who knows maybe Apple will be releasing the" tablet/slate/whatever" at the end of the month with a screen in it just like this one. I wouldn't put it past them.
Looks more like Sepia.
@(Unverified) Haha...they should really get a white balance reference when they go and hold it in the frame so they can correct white balance as they go. Much of the real photography of Mirasol displays has been done with terrible white balancing in terrible lighting. Wonder if there'll be better shots during the show, under some better light.
@BubbaJ I can attest that this technology currently has limited color gamut and depth. You can see in the pictures that the colors are pretty washed out.
Looks like they've got good blue, so-so green, and weak red.
Much better than monochrome, but not as good as Liquavista's tech.
Gameboy color forever........ You kids dont know nothin with your DSi's.......
@(Unverified)
Word. Gameboy Color FTW!!!
Whats with all the Apple stuff ont the videos on their site? Like Safari etc...
So this makes me wonder how well it can integrate with touchscreen methods.
Obviously you could put a resistive layer on top, but that leads to crappy contrast and glare as Sony has so graciously provided example of.
Plastic Logic has a flexible substrate so you can put resistive touch underneath, but I am guessing Mirasol is not going to be flexible.
Wacom? Even if it didn't interfere, the iRex DR800sg reviews have shown us folks don't like a stylus only interface.
That leaves Capacitive. Could a capacitive interface work without interfering with the screen draw?
Ultimately I can't imagine this thing succeeding in a tablet format without touch.
Oh and why is that when someone is filming a new display tech at a trade show do they not ever change the camera angle so folks can get a feel for the viewing angles and glare. Seems amateurish.
@RandomGuy, there are many other touchscreen technologies that may be used. See: http://en.wikipedia.org/wiki/Touchscreen
As far as why they don't change the camera angle, it's often because there are too many people crowded around the product, and the camera guy is actually several feet back with the camera zoomed in.
@ DisplaysRus
Ta very much. On a parallel note, do you happen to know if there exists such an 'analogue' varient for DLPs? Something able to bend the mirrors to arbitrary angles rather than juat on or off the lens? I figure if I got a knowledgable guy on the line, I gotta ask everything!
@elmer
Not as an array of pixels, to my knowledge. I can't think of any purpose for that. The DLP is designed to point the mirrors at a specific angle. However, there are single mirror devices that can be driven to many angles. A single axis mirror is used in some laser bar code readers, for example. A two axis mirror is used for rastering a laser to create an image in Microvisions pico projector. Mirrorcle is one company that also produces MEMs mirror modules. Perhaps there are other like devices used in optical network switching..