The occulus rift will screw up your eyes, maybe this won't

Discussion in 'Physics & Math' started by stateofmind, Oct 12, 2014.

1. stateofmindseeker of liesValued Senior Member

Messages:
1,362
I woke up this morning with a random thought. I'm not sure if this has already been proven, but I've reasoned that the cause of reduced eyesight accuracy from extended periods of staring at a computer screen is mostly caused by the illusion of depth that the images on the screen display. Tricking the eye to see depth on a 2 dimensional plane trains the eye over time to focus incorrectly at the depth of 2-4 feet that you're sitting away from your computer screen. In other words, 20/20 vision is a habit.

Okay, so the occulus rift uses two samsung galaxy note screens for each eye, a centimeter or 2 from the eyes. Video games exploit depth illusions a lot and so this is going to fuck up people's eyes royally I think.

So back to my random thought from this morning. What if we rethought the computer screen. I noticed an interesting property with mirrors a couple weeks back. Did you ever realize that mirrors have actual depth? They do. You have to look into a mirror to see the objects that it reflects. You actually have to zero in on their correct depth from the mirror to be able to focus them. In other words, a mirror does not create a 2D image, it's 3D.

So here's the idea... what if each computer screen had several small LED emitters inside them. instead of one huge array of LEDs that we are now accustomed to It would have 1 emitter for say... a distance of ~1 foot, another for a distance of ~10 feet, another for ~20 feet, and so on. Each LED emitter would shoot its image onto a path of mirrors that reflect onto one another, like how a camera lens redirects the light entering it, totaling a specific distance (1 foot, 10 feet, 20 feet, etc) before reflecting onto a "master" mirror which all the mirror paths of each LED emitter project onto. The "master" mirror would replace the single LED array that we are all now looking at as we browse sciforums.

So the master mirror would be a composite image of all the LED emitters. What I think this would create is a monitor that literally has depth, not the illusion of it. The tricks we now use to create the illusion of depth would become obsolete on a monitor like this... instead you would just program each objects to be displayed at whatever depth you want it to be (obviously the depth gets bit crushed to an extent... for instance if you only have 3 emitters at 1, 10, and 20 feet, you'd only be able to use those depths and objects would be "crushed" into them. The solution is to add many emitters at many depths to increase the depth resolution.

This would largely fix the problem of vision impairment we now see in people who look at computer screens for extended periods of time. It would also give us a whole new realm of realism to play around with in the digital world.

Last edited: Oct 12, 2014

3. danshawenValued Senior Member

Messages:
3,950
Does the same argument apply to something like Google glass?

5. stateofmindseeker of liesValued Senior Member

Messages:
1,362
I've never personally tried google glass so I really can't say. If it uses artistic illusions to create depth then I would assume it would also train the eyes to develop bad focusing habits over time. If we look at a 2 dimensional image and an object within it appears small due to illusions like giving it some blue haze to simulate lots of atmosphere in between yourself and the object, as well as making t smaller in comparison to a "foreground" object (I use quotes here because every object in a 2 dimensional image is always literally in the foreground) then it plays a trick on your mind because your natural instinct is to cause your ciliary muscles to relax when these conditions appear (blue haze, smaller, anything else that hints at distance), decreasing the magnifying intensity of the eyes. But we have to override these instincts in order to see the 2D image that's right in front of us in focus. So if you look at these kinds of 2D images a lot, that use these tricks frequently (video games especially), then when you actually see an object that's literally in the distance in true 3D, your mind is so used to overriding its natural instincts that it no longer has the muscle memory to focus them with good accuracy.

7. stateofmindseeker of liesValued Senior Member

Messages:
1,362
In case I didn't explain this clearly enough, when I say "mirror paths" I mean something similar to this:

So basically you'd have different mirror paths of different distances, reflecting the focused light beam from each LED emitter, before finally shooting the beam onto the main mirror screen, the "master" mirror (this is the big mirror that replaces the standard "computer screen" (the thing that you're looking at right now), it acts essentially like the wall that a projector shoots its light onto, only in this scenario, there are basically multiple projectors at varying distances shooting their light beams onto the same wall.) The reason we use mirrors and not literally have light emitters 20 feet from the master mirror, is to conserve space. Like a human intestinal tract, how it fits literally miles of intestines into the stomach of every person, a similar concept could be applied to mirrors. We could have mirror paths inside this monitor that reflect the light beam from a specific LED emitter for miles before it terminates onto the master mirror screen, and in only the space of maybe 1-2 cubic inches. So this monitor does not have to be HUGE to actually create genuinely vast distances in 3D space.

8. StryderKeeper of "good" ideas.Valued Senior Member

Messages:
13,104
The main problem is that even if you use mirrors, your still going to have a surface in close proximity (Mirrors and glazing in general will always reflect a small amount of light)
I pondered various points about this back when dealing with using a CRT where the EM output was harmful. (caused my eyes to sting, headaches even nausea.) I found that TFT's/LCD's while not outputting the EM levels of a CRT can still be extremely bright (they don't cause any of the effects of prolonged CRT exposure).

I eventually considered using projectors on backdrops to use the diffusion of reflected light. (Indirect light through this manner is less energetic), the problem was that projector bulbs can cost a lot to replace.

You could project an image onto a diffusion backdrop and then reflect that diffusion as a secondary layer to your primary, this way it won't be an intense overlay however the quality is dependent on the quality of the backdrop and the projection method used.

9. stateofmindseeker of liesValued Senior Member

Messages:
1,362
As for brightness issues, I've experienced this myself though I don't think I'm as sensitive to bright lights as some people. There are various utilities you can check out that can help out a lot with this. One of them is called DimScreen, which allows you to set keyboard shortcuts to increase and decrease the brightness of your monitor on the fly. Another, more interesting one (in my opinion), is called NegativeScreen. This will actually remap all of the colors being displayed on your monitor to totally different palettes. The default ones provided with the utility are very good and are more catered to making light colors dark and vice versa, but you can also program your own quite easily to suit your needs. It literally makes the whole internet a completely different place... I definitely recommend checking it out if only for idle curiosity/amusement.

By "overlay" I'm assuming you mean when the LED emitters stack up on top of each other, adding their intensity to one another. I've considered this and I think this would be dealt with nicely by software. Algorithms would determine where each LED emitter would be projecting on the x/y axis of the master or "primary" (as you say) backdrop (in this case just a big square monitor-sized mirror). There would be a hierarchical relationship between each emitter layer, so that emitters on a closer layer would cause all pixels of all layers of greater distance (longer mirror paths) firing on the same coordinates it is currently firing on to be turned off - hence no overlay.

10. stateofmindseeker of liesValued Senior Member

Messages:
1,362
A thought just occurred to me. If I remember correctly you can actually emit light through tubes of certain types of glass that bend and have them come out the other end undistorted. I admit I don't know much about this technology but it seems like this would be a more distortion-free way to emit light long distances. What I wonder though, is that if you look into the tube, would the light on the other end appear to be far away, proportional to the distance of the tube? I assume it would and it would therefore be usable for something like this.

11. Neddy BateValued Senior Member

Messages:
2,365
Very interesting topic, stateofmind. I think you want something like this:

transparent screen #3 by wins72, on Flickr

(Click the thumbnail to see the larger image.)

This is a photoshop -- I don't think technology can give us this level of coolness, yet. Note how the app windows are opaque, not even the slightest bit transparent. But imagine if the that could be true, and also the transparent areas were truly transparent. Your eyes could focus on the distant background, or a 3D effect could be achieved with a few layers of screens.

12. stateofmindseeker of liesValued Senior Member

Messages:
1,362
That level of depth is definitely achievable but it would require many emitter "layers" to recreate. For instance, if you tried to recreate that "image" in a 3D space in the mirror setup I've been describing, but you only had, say, 3 emitters - one calibrated to 2 inches, a second to 8 inches and a 3rd to 6 feet - then that image would be what we call "bit crushed" into the 3D space. The book would be a flat image that's emitted on the master viewing mirror by the 2-inch-emitter, the bulky computer monitor in the picture would be a flat image on the 8-inch-emitter and the back wall would be emitted as a flat image on the 6-foot-emitter. You'd have depth but it would look more like a holographic pokemon card than a true 3D space. It would be a true 3D space for sure, but sort of a cartoonish version of it... But let's say you had 1000 of these emitters instead, with a max depth of 6 feet, the emitter distances divided evenly into roughly 1/8" increments. With a resolution like that you can begin splitting up each of the objects in your 3D space among the different emitters. So the computer monitor would be sliced up beginning at the 8-inch-emitter and incrementing in 1/8" slices up to the 20-inch-emitter where the other edge of the monitor is. At this resolution objects would begin to appear literally how we see them in the analog world.

There would be a problem with this scenario though. The main one being that in order to create a depth of a mile, you'd have to literally have a computer screen with transparent layers a mile deep. That would be quite expensive as well as difficult to fit on a person's desk. Also, there's no such thing as a perfectly transparent material, so the more layers you add, the more difficult it will become to see the layers further back.

13. stateofmindseeker of liesValued Senior Member

Messages:
1,362
I just found this interesting snippet from wikipedia about fiber optic cables:

So you can transmit images through them... but I have no idea if they would retain their depth through this method...

14. stateofmindseeker of liesValued Senior Member

Messages:
1,362
Just had another idea. Why even use mirrors at all? Instead there could just be a single projector. But inside the projector would be made up of many different individual emitters that combine into the main output beam, each transmitting their images through different lengths of fiber optic cable (assuming the light depth is retained in the cable - which I think it would be). Then you would be left with a single device without any fragile and delicate mirrors to have to adjust that you could project onto any surface and literally create a digital 3D world.

15. Neddy BateValued Senior Member

Messages:
2,365
For simplicity, let's just stick to three layers. Let's call the layer that is farthest away from the viewer the "rear layer." Let's call the layer that is closest to the viewer the "front layer," and we can call the other layer the "middle layer."

The rear layer does not have to be transparent, because there is nothing behind it. The other two layers must be transparent to allow the rear layer to be seen through them.

But in the picture I posted, note that the blue app window is just sitting in front of the bulky monitor in the background. That blue app window is not transparent at all, because it was Photoshopped. That is what I don't think we have the technology to achieve yet. Do you see the problem? The front layers would appear 'washed out' compared to the rear layer.

16. stateofmindseeker of liesValued Senior Member

Messages:
1,362
Right. I don't think this is an issue of the technology not being there yet. There's a logical contradiction inherent in the design. On the one hand you chose transparency for your first layer in order to see layers behind it, but then you also want it to be opaque at times as well. I'm sure a material with these properties could be achievable sometime in the future but then again, what are you trying to achieve with this design? Do you simply want a computer screen that can turn transparent? What's the practical benefit of such a device?

17. stateofmindseeker of liesValued Senior Member

Messages:
1,362
I organized my thoughts on this subject some more and refined the design a bit:

Have you ever noticed that you have to "look into" a mirror to be able to focus on the objects it's reflecting? For instance, you could choose to focus on your face or you could also focus on the smudge on the mirror, but not both at the same time. This is because a mirror creates a 3D image instead of a 2D one.

This means that when we look at objects, what we're seeing is the entire light path of all of the light beams that travel into our pupils. It only appears that non-reflective objects have short light paths (for instance, you can't see the entire light path of the light beam that travelled from the sun, through the atmosphere and onto your dog or cat, because they're non-reflective coats diffuse the light beam so much that there's no longer a single point of focus to be able to follow it all the way back to the sun.)

Now we can take this knowledge and apply it to our computer screens. Right now what you're looking at as you stare into your screen to read this message is an array of LED's, each emitting a light beam. You're probably sitting about 2-5 feet from your computer, so this means that the depth of all objects that appear on our computer screens will always be that same distance of 2-5 feet.
Fiber optic cable has what is called "total internal reflection" and so any light beams that travel its length will come out the other end the same way as they entered it, regardless of how bent and twisted the cable is. Recall that whenever we see light entering our eyes, we always see the entire light path (assuming it's not extremely diffused). So this means, an image emitted through a fiber optic cable of a short length will actually appear closer to us than the same image emitted through a fiber optic cable of a greater length.

Okay, now here's where things get interesting. Suppose we construct a projector, like the one you've seen in the classroom in school. But instead of having a single array of LED's, let's now create multiple arrays of LED's and shoot each of the images they create down fiber optic cables of differing length, say, 1 foot, 10 feet, 20 feet and 100 feet. Let these fiber optic cables all have their ends meet up at the same place, directed at the same spot to be projected onto the wall. Now when you see the projected images, they will all literally have different depths of 1 foot, 10 feet, 20 feet and 100 feet, instead of the fixed length of the project > to the wall > to your eyes.

Now let's add some software. We can write a program which tells the projector which images should be projected at which depth, and if any of the images overlap, the LED array of the shortest distance will take precedence and all other LED arrays of greater distance projecting on that same spot will have their LED's turned off in those corresponding areas (this way the images don't stack on top of each other, and closer objects can move on top of further objects without blending into them).
Right now this is fine to create a 3D space, but with a z-axis of only 4 values (1 foot, 10 feet, 20 feet, 100 feet), our 3D space is quite cartoonish.. like a holographic pokemon card or something. Well, all we have to do is increase the depth resolution in just the same way as we increase the x/y-axis resolution on an archaic 2D monitor... in the x/y-axis example we squeeze in more pixels... in the z-axis example, we add more LED arrays of varying depths.

So if we got the depth resolution to something like 1000 arrays, with varying distances of 1 inch to 1 mile (apportioned on a logorhythmic scale so you could have finer intervals for closer up objects where it's more important). We could then start programming for the z-axis in a literal way, not creating artificial 3D spaces onto a 2D plane. On this new kind of computer monitor, if a 3D object has a depth of 3 feet, and is 2 feet away from the 0 coordinate on the z-axis, and we have z-axis intervals of 1/8", then you would cut the depth of your computer-generated 3D object into 1/8" slices and apportion them to the corresponding LED array at that same depth in our projector. You can now LITERALLY look into your computer monitor. No more optical illusions that screw up your eyes!!

18. Neddy BateValued Senior Member

Messages:
2,365
I don't think that is correct. The light coming out the end of a fiber optic cable looks exactly the same no matter how long the fiber is. That is why we can use them to send data over long lengths, like Verizon FIOS.

I was trying to actualize your idea of having a display screen with a few different layers to provide a few different focal lengths for the viewer's eyes. If a display screen could be transparent, but also display opaque (non-transparaent) pixels as well, then layering a few of those display screens in front of one another would give us exactly that. However, with today's technology, the transparent displays can only display transparent pixels, (as far as I know). So the front layers will not be truly opaque, although the rear layer could be opaque.

Last edited: Oct 15, 2014
19. stateofmindseeker of liesValued Senior Member

Messages:
1,362
You're confused here. While I have not personally confirmed that light depth is retained in a fiber optic cable, this in no way affects or would hinder how fiber optic cables are used for data transmission. The receiver on the other end of the cable is only concerned with on/off values, much like morse code. I believe there are receivers that can also evaluate different frequencies of the binary light values as well... but I'm not sure so don't quote me on that.

Ah, I see. For the sake of argument, let's say the material does exist today and that it's also cheap enough to produce in large quantities. With this setup it would be quite impractical to create focal distances of anything greater than a foot. For instance, a focal distance of 1 mile would require your computer screen to be a mile long. Most people don't have desks that big.

20. Neddy BateValued Senior Member

Messages:
2,365
I was just trying to help. Good luck.

21. stateofmindseeker of liesValued Senior Member

Messages:
1,362
Pick your chin up. A scientist has to be critical of their ideas in order to create better, more valuable ones. Getting your ideas and beliefs taken apart is a process everyone goes through. Ultimately you have to ask yourself, which do you value more? Truth or pleasant fantasies?