3D LED printer makes a contact lens display possible | Hacker News Reader

3D LED printer makes a contact lens display possible

geek.com

60 pointsmmastrac21 comments

Jesus, these comments are so pessimistic. As of now, the three top-level comments are "Wouldn't this cause headaches?", "Wouldn't these have a uselessly small viewing area?", and "Wouldn't these be too blurry?". Come on, people; have some imagination. These aren't even show-stopping physical limitations (which humanity has sidestepped thousands of times before); these are relatively minor engineering problems.

Even as someone with essentially no experience in this area, I can think of a few potential solutions to the complaints thus far. The eye strain and focus issues are solved by using pre-collimated laser light instead of LEDs. Our fabrication isn't quite there, but again, it's an engineering challenge. The viewing area size is expanded by eye tracking (possibly via sensors in the contact lens).

It's good to question these sorts of things, but at least start with interesting questions!

Pedantry is a Hacker News tradition. Negative comments are much more plentiful because they're easy. Better not discuss this for too long. The moderation team is extremely averse to meta discussion and introspection.

There is a lot of hype out there so many stories are welcome with skepticism. The advantage is that you can post here more technical details.

I still don't understand how you are going to focus an image that is jut on the top of the cornea. The lens of the eye can't adjust to this distance. For example, Google glass has some external lens so the virtual monitor image is far enough. (I n't find good digrm now, but this is an interesting discussion http://www.reddit.com/r/askscience/comments/19plgu/how_can_g... )

Do you have a diagram that shows how the image will be formed in the retina?

(I also think that the saccades are a problem, but I don't know enough about that subject to weight the technical information.)

The article talks about power requirements and the manufacturing process, but seems to leave out that any image projected into your eye from 0cm away will be blurry beyond recognition. The eye won't be able to focus on it. Has research overcome this limitation? It's very interesting to think that it's possible to project a sensible image into an eye from point blank.

That's only the case if the LEDs transmit light in a broad cone or hemisphere. If the cones are very tight (like at laser in the most extreme incarnation), you can point the light sources at different angles and force it to map onto the retina properly. The important point is to reduce the circles of confusion acceptably. You don't necessarily need them to be zero radius.

Microlenses!

The article Geek.com links to located: http://spectrum.ieee.org/biomedical/bionics/augmented-realit...

is probably a better read.

Or wait until contact lens displays with Retina resolution (the Apple iEye, perhaps)? :-)

The Retina(tm) retina!

'This rabbit got to wear an “active contact lens,” and showed no observable ill effects.' sounds so much better than 'animal testing has so far suggested no observable ill effects'

Unless I missed something when we were covering eye physiology at uni, contact lens screens are useless. The area we can actually attend to and process data is tiny, which means the screen has to be directly in the center of the contact.

For example: A physically separate screen means you can swivel your eyeball and look at a different bit of text on a book page, or a different character on a movie screen. With a contact lens display the content you are attending to is always on the center and swiveling your eyeball to change targets will do nothing.

Is it feasible to move the content instead?

If the lens knows its own orientation, then its display can change with the orientation. The lenses themselves don't need to embed gyros. I can imagine wearing an upside-down U over the bridge of your nose which has embedded sensors that measure where you're looking.

>swiveling your eyeball to change targets will do nothing.

Unless you put a MEMS accelerometer/gyro on the contact lens and change content as the user moves their eye.

These are engineering problems, not insurmountable physics problems. Saying stuff like "contact lens screens are useless." is unrealistically pessimistic :)

Pair it with eye tracking. Main problem I see with it is LEDs on a contact lens wouldn't be in focus, at least without some other optics to make them more directional.

That problem seems reasonably solvable.

The main problem I see is that saccades are ridiculously fast, so not only does the eye tracking have to be extremely precise, but the latency for the track-update-display pipeline has to be tiny.

This problem is already difficult for head motion. Doing it at the eyeball level is way more ambitious.

I'd imagine applications even with a low res, blurry display might include directional indicators and notification lights in the periphery of the users vision. This would only require a ring of a few LEDs round the edge of the device and would be more discreet than e.g. glass.

Wouldn't something like this cause extreme eye-strain and/or headaches? Even if you could make sense of the display...

You don't have to focus on a screen that's that close to your eye. The pixels are so small and so close to your eye that the light doesn't disperse much before it hits your retina. That means the dots of light will seem to be in focus no matter how you focus your eyes.

That is not true; please be careful to do research before posting so you don't spread misinformation.

If a light source is diffuse/omnidirectional, then it will be blurrier the farther it is from the distance your eye is focusing on.

Only if the light is a ray (a collimated beam with very little spread and very small diameter), such as the ray produced by a laser, then it will be in focus regardless of the distance at which it is produced.

Somehow the first thing that came to my mind is "contact lens low-light amplifiers" (e.g. for military use or driving vehicles at night). I'm not really sure this would be at all possible, but I find the idea much more intriguing than contact lens HUDs.

I'm imagining the Blue Screen of Death filling your area of vision.

Reply on news.ycombinator.com