Slashdot Mirror
Laser Powered Virtual Display
Tedger writes "The Feature has an article discussing an interesting portable display system developed by the University of Washington. Unlike your traditional mini displays mounted in glasses this system has no display, it is a 'virtual' display created by lasers and microscopic fast moving mirrors. The image is in fact printed onto the retina and has feasibly a infinite resolution. Can anyone say true VR?"
Puts hiro protagonist's display to shame (his required glasses I think) Wonder how long before someone tried to snowcrash a person through it :D
My Portfolio
remember to use that screen (retina?) saver!
Either something incredibly dangerous (Do Not Look Into New Monitor With Remaining Eye) or amazingly trippy (with Pink Floyd playing in the background).
1) power 2) speed. A CD-ROM laser could hardly hurt your eye. And if I take a laser pointer and quickly "sweep" it over your eyes, you won't feel a thing too. That's how "disco lasers" that are projected into crowd work - the beam power would be enough to damage retina of someone whose eye would accidentially enter it, but it sweeps displaying "shapes" so quickly that even if it hits someone's retina, it won't be harmful - the flash lasts too short to cause any damage.
(think photo camera flash, your eyes survive it easily, but if you were exposed to light of such brightness for a second or two, you'd go blind permanently.
Anagram("United States of America") == "Dine out, taste a Mac, fries"
After all the obstructive heads up type units we finall have one with the potential to co-exist with our normal field of vision. The "augmented reality" could give us new ways of seeing the world, with a 3-d overlay on reality. In the article they mention and automotvie expert system which will give the user a visual overlay of the system their looking at.
Also it should give you the ability to use PDA's in a private fashion while still having a large view. In fact, this could redefine the PDA format, instead of the little notepad style device. Just gotta get the production levels up, cost down, so it's more affordable than the $4000 price tag.
Bacardi + slashdot = negative karma.
Would be if, since they're already sticking us with a laser beam in the eye, was if they could track eye movements.
This way we coul play tetris (or by that time Grand Theft Auto on a cell phone) just by tiny eye movements.
It's all fun and games until someone burns an eye out.
Yeah, especially wtih one of these...
I ordered mine yesterday along with a co-worker, and we'll hopefully see them by early next week.
- "Nobody came out that night, not one was ever seen. But Old Man Stauf is waiting there, crazy sick and mean!"
I mean, if I remember my optics correctly, the way the cornea/lens assembly works is that all incoming light originating at the same point out there ends up in the same spot on the retina, regardless of which path they take through the lens. This is what enables us to see a clear image.
Although it has certain other intersting proerties, laser light obeys normal refraction.
Yet they talk about suåperimosing the image on the normal view. How can you project to any other part of the visual field than the area where you see the projector?
Anyone know what the trick is?
sudo ergo sum
Yes, and that's how crt's work too, by constantly sweeping electrons across your screen. When they repeatedly sweep across the same area, ie. when your screen is displaying the same image all the time, you get burn in. This display would be much different than a stray laser sweeping across your eye one time, this would be constant. If the disco laser heats up your receptors for a fraction of a second, they will cool right back down, but if they are being constantly heated, even by a low power laser, could there possibly be long term consequences? I think that's what the parent post is concerned about.
"the universal aptitude for ineptitude makes any human accomplishment an incredible miracle" - Stapp's Law
I'm not too worried about safety: if you limit the maximum power output of the laser, even in case of short-circuit, it shouldn't be a problem.
This is a technical problem, engineers have been good at solving those.
The human limitations may be much more difficult to overcome: show a 'static image' to a moving man and you have a problem: eye say static, inner ear say 'you're moving' --> conflict --> sea-sickness!
Well, it obviously can't have an infinite resolution, the best it could get is 1:1 mapping with the rods and cones in the back of your eye.
And of course this is old fashioned analog technology, just like in a CRT firing beams of electrons in the rough direction of dots in the phosphor, it's not accurate. What you need is a direct digital plug in the back of your optic nerve!
Jolyon
Please read my Canon EOS tech blog at http://www.everyothershot.com
Not really. This laser "sweeps" too, true it points at your eye all the time, but the beam hits only a small fraction of your retina a time, different groups of receptors get "heated" and despite the ray returning to the same point over and over while displaying sequence of frames of a still image, the delay between "frames" should be quie enough for receptors to "cool down". Also note laser is mostly about coherent, very narrow beam of light, not about power - you can make the beam as weak as you want, to and beyond point when it's absolutely safe for it to shine right at your retina without causing any damage. Except, if you move it fast enough then, it won't be visible at all...
Anagram("United States of America") == "Dine out, taste a Mac, fries"
So when the scanning mechanism (moving mirrors) stop functioning you get a burnt spot on your retina. Remember, if you're doing a Megapixel display, the laser is 1,000,000 times as much power as a single pixel requires. When the scanner breaks, how long do they have to detect the fault and shut off the laser before damage is done? Perhaps it can be done, but determining failure modes and implementing fast and effective diagnostics is tricky business (it's part of my job).
At least with DPL a broken mirror wouldn't hurt and you can use non-laser light which is safer.
6 years ago or so they where working on this type of system here at the university. I had the pleasure of trying it out (after signing a disclaimer of course :). At the time is was red only, but very very cool. They couldn't focus the beam depending on what distance you were focussing on. So the images they projected where sharp only at one fixed "focus distance" for your eyes.
They could produce a low resolution overlay image over what you were actually looking at. They could only produce very simple line drawings floating in the air. But still.. you had your own private (head ache inducing) lasershow.
I would think that this design would require the user to always look directly forward. Otherwise the laser wouldn't hit the same spot when the user looked slightly to the side. The visual distortion that this would cause would probably make you pass out. In order to really make it work you would probably need to track eye movement as well. Although this is possible, it seems like it would be error prone and would make the system too expensive for consumer use. The bottom line is that unless they place the laser emitter right on your cornea, any eye movement would cause distortion and make the user very dizzy. The further the distance between the emitter/mirror and your cornea, the bigger the impact of even tiny eye movement.
Actually, this is a well documented issue if the display uses IPv4. (Scroll down to item #3).
It's not that simple. You kids probably haven't seen this, with TFTs being so common these days, but CRTs move a high intensity electron beam across the screen to stimulate the phosphorous which then emits visible light. When you turn off an old television, you can sometimes see that the part that moves the beam across the screen is turned off before the beam ebbs off. This produces a bright white spot in the middle of the screen because all the energy which would previously be distributed across the surface area of a couple square feet now ends up in one spot.
It's the same with lasers. The laser has to be powerful enough to activate all the sensors of your retina during one sweep. Let's say the system of tiny moving mirrors fails and the laser gets reflected onto one spot on your retina for just a fraction of a second. In this short time, the spot on your retina has received more than 100000% of the energy which is sufficient to make you see bright white.
I will use these things when they have independent deflection systems so that the failure of one system will result in the beam being deflected out of my eye instantly.
One of the problems with these devices is that they tend to end up classified as medical devices due to the tight integration with the retina.
This stuff is cool, but I don't see it becoming available in the U.S. any time soon. I would worry about a bad capacitor or something that suddenly released an hour's worth of exposure in a microsecond and fried my retina. Somebody with more engineering knowledge of these systems may know whether that's impossible or not, but it will always represent a consumer concern, I imagine.
If your bitterest enemies are people who hack the heads off civilians, then I would say you're doing something right.
Unless there is an separately calculated image for both eyes, and a head tracking unit, it will not appear like an object is "virtually in front of you". Without these two things, you simply have a 2D overlay on your regular vision. The separate images are required to make your eyes focus at a particular distance, the head tracking so that when you walk rightward, the object goes leftward, etc.. Perhaps the technology is there, but not described in the article...
First of all, you'd have to stare into a CD laser for some time before there was damage. These lasers will be even lower-power than that. Second, you can use a simple timed driver circuit to control the scanning mirror, so that as long as the laser unit has power, the system is scanning, with a safety interlock circuit which disables the laser if it detects that it has stopped moving. This can all be done at a low level and frankly it doesn't sound very hard to me; it might be hard to make a system that doesn't detect false positives but I'm betting you can build the laser, the scanning circuit, and the safety circuit into a single chip using MEMS and have the cost be basically nothing (in terms of what the device will cost) - the chip will just return pulses for synchronization so the video solution can tell the RAMDAC what to do, and it will have a system to synchronize two of the devices together.
Now, I'm no EE so maybe there's problems with this, but it seems simple enough to implement. There's just not a lot going on; the laser scans across, and each time it hits the end, it jumps down a line. If you don't get the pulses occurring within a certain time, which can be based on filling a capacitor as I'm sure you well know, then you just shut it off. It's easiest to do with fixed-resolution displays, but all you have to do is use a different cap (or multiple caps) for different resolutions, or just accept that the laser might stay put for five or six lines' worth of scanning at some resolutions, which is highly unlikely to damage anyone's eyes.
I don't think that non-laser light is really any safer. With a laser, you can use a lower intensity of light because your results will be more accurate with less light. Either way you need to get the same amount of light to the user's eye; this is, quite simply, how you will be controlling intensity.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
"One can watch the Sun in a telescope exactly two times in a lifetime. First - with left eye, second - with the right."
3.243F6A8885A308D313
This entire problem is solved the same way the real-world saccade problem is solved.
:-)
Your visual processing system (more specifically, the transferral of visual cortex information into your internal "world-map" representation) is for the most part shut down during a saccade. Whatever comes in is assumed "irrelevant" by your attention system.
This is why you have to play focus games like the ones you describe in order to notice the effects of artifacts during saccades. You don't notice this stuff much unless you're actively looking for it, as the parent post instructed us to do.
Besides, it's not "zero persistance". The response of your receptor cells to incident light rays is neither instantaneous nor of zero duration. There's an implicit persistance in the time during which a rod or cone will fire action potentials after receiving a burst of light.
Scan your system significantly faster than that duration, and you can't see transient effects, even with a laser. You don't see flicker with an LED cycling at 1000Hz; even though an LED is just as instantaneous as a laser. Your neurons simply don't respond that fast. I believe the cutoff for absolutely flickerless images is around 120Hz.
Though, you can still get fun effects if you vibrate your head in frequencies near the refresh rate. If you happen to be a bass, you can sing notes near 60Hz for cool rippling effects with LED clocks and the like.
Of course, this shouldn't affect a head-mounted laser display very much.
I stole this sig from someone cleverer than me.