http://www.youtube.com/watch?v=Aj3_v7xCyJ0&feature=related
http://www.youtube.com/watch?v=f1YNyQqbiF0
something I used to be passionate about, holographic displays. how to light a pixel inside a volume without it and the wires, being so big. The best plan so far seems to be making a matrix of points in a space where two laser intersecting at that point, providing enough energy, makes the point light up. problem is building on that tiny scale, enough pixels to make a real picture worth looking at. Like you need billions of points and so you need tons of laser lighting up all those points fast enough to make a 3D picture.
I wonder though,
http://www.youtube.com/watch?v=8G3wWmtkN88
Those are unwatchable. There's too much crap (the ads stuff, or whatever once you own one and want to use it yourself), behind the picture to pay attention to the picture (the clear parts of the picture. I mean the point is for the picture to be transparent.. but why?)
The saving grace is that the rendered picture becomes it's own volume inside the 'picture' frame meaning unlike that transparent display where the picture is a drawing on a transparent sheet that has no light of it's own (depending on your POV the picture is brighter, or dimmer), a proper volume display renders a picture in 3D but also lights it up so that you can't see through the rendered object from any POV (like the cube in the first link).
Also the only way to appreciate 3D is if the other sides (from your POV) is black (for contrast). So that your bright rendered 3D objects or scenes aren't competing with whatever is behind the display for your attention in the frame of the display.
It should be easier to make a volume display like those 3D cubes if you're only able to view it from one side. I mean who walks around their TV? The last think Samsung or Sony, Fox and NBC are going to do is get me to walk around my TV while I watch. POV is static for a defined viewing angle.
Otherwise I mean if you had a virtual receptionist or something in a tube of pixels them yeah maybe you'd want to walk around it, but the virtual receptionist would have to turn and face the person interacting with it while the others would see whatever side is being rendering from their POV. Pointless unless it was a real hologram, aware of each viewer and they could see from their POV a unique rendering. For each person looking at the tube, they would see the virtual 3D secretary/or tour guide, looking at them. Same applying for TV. How would you have a news broadcast in 3D unless each viewer felt like the newscaster being rendered was looking at them?
I'm not personally happy about the concept of 3D (on a screen) or volume displays that need glasses either, because they rob some of the light and have a limited frame rate. Just the same as LEDs and wires in those cubes block light. If they are made small enough to be almost invisible, and for it to be hard or impossible to see the space between each point of light, like the transparent laptop screen, and still be suspended in a cube, would you get the same resolution? or would brightness fade deeper into the cube from your POV? how small would each pixel have to be for the cube to be crystal clear.
IF instead the cubes (as displayed above) render as fast as they can, for each pixel, then there are no blank frames. It would be like an analogue display. Which panel makers should be able to accomplish now with digital displays given the high resolution and tiny dot pitch, in a flow-frame method where pixels that don't change don't go black, they just color shift as needed, instead of displaying full pixel frames and shuttering them, but that's another issue.
Still.. at 256x256x256 a 3D volume display would be kick ass, especially for gaming, as long as it was crystal clear. That 16x16x16 cube is already very impressive but each pixel point of light is so big it becomes like muddy water half way through.
That and I wonder if you started with something like water, added bio-luminescent cells (which can be grown in a lab), you could possibly grow the framework, with electric links between bundles of cells (for each color of RGB and CYM, (how bright can each one be while still being transparent enough to work as a high definition display, and growing it in water, can you leave the water in there without it distorting the picture or would it have to be drained). Would artificial bio-luminscent cells be small enough to work as pixels in a cube?
And using water to grow the display means you can 'repair' it by refilling it with water and either clearing out bad sections or dumping the whole volume and starting from scratch. Plus using cells, they might even be able to go opaque (providing a bright background sky to a scene, or dark murkiness).. problem being how fast they could refresh or colour-shift. Again, I think if you aren't dealing with shuttering frames in the whole volume, that having only the pixels that need to shift, do so, that you can trick the eye into seeing movement more naturally by shifting groups of pixels less quickly than whole frames more quickly. I wonder if it would cause ghosting issues though..
For TV would people mind dropping resolution and picture quality to go from 3DTV to holographic/volume displays like those LED light cubes? Some people are talking about going directly from 1080p (for 3D you'd double the number of frames), 1920x1080 pixels per frame something like 24 or 60 frames per second, to 1920x1080x1080. All well and good, but would having something like 16 billion pixels even be necessary (nevermind possible)(16 billion from 4000x2000x2000), colour shifting, or the whole volume of 16 billion pixels per frame refreshing at least 60 times a second.
Just one frame in a volume like that would take, from 1 megabyte to 16 gigabytes (or more), even with an average from 3-8 gigs, time 60 frames in a second, something like 50 gigs per second, modern technology can handle it, but you'd need a beast of a computer to power that picture.
Considering doing it for movies first, like if IMAX wanted to support SuperHiVision, something like this:
http://www.youtube.com/watch?v=am_azZxTFOI
Where they say you can't see (appreciate) the picture on a screen smaller than 100".. I wonder what lunatic programming (some 10 to 15 years from now) would even be worth the effort of a 3D version of SuperHiVision... 100" diagonal and what four or five feet deep? @.@ 150 billion pixels per frame at 60-120 frames per second. Totally impossible with modern technology...
BUT if you think about 3D gaming (on much smaller screens), and from that large LED cube (the second link) the real point is that you can't see the whole seen using only 17 million pixels of a normal game screen at the same time no matter what. You need z buffering to turn a 3D scene into a reasonable 2D picture. If anything you'd only need 3-6 pixels depth to add richer more vibrant colours and to help fake true depth.
Meaning those 16 billion available pixels in a reasonable LED cube display, running 3D games which can render infinity inside the computer, only need to show you a 2D picture, even in a 3D volume. Nothing really changes (for a single person looking at the display), you only need the 4000x2000 x 3 or 6 pixels for RGB CYM colour and brightness for the picture to kick ass. Anyone else looking into the volume from the side might not even see the picture because only small sections would light up all over the cube.. only the person looking at it from the 'front' needs to see the picture.
And if you're faking 3D anyway, with glasses, you don't need the full 2000 pixels deep, it's hard to even notice that something is farther away without a frame of reference... which I suppose is the hardest to imagine about a 4Kx2Kx2K display without seeing one in real life.
That chinese model running 256^3 pixels when displaying a fish near the end, is hard to see as a 3D object.
The picture is about the same size as a desktop icon blown up to the size of a room. There's no detail to see a tiny pixelized fish floating around at 256^3.
At 2K^3 or 4K^3 you'd be able to tell the fish was the clown fish Nemo and maybe even with a high enough refresh rate it would be eerily lifelike, giving children nightmares for the rest of their lives. But would you be able to tell it was 3D without shading and blending.. stuff that again, is best understood if you're a fan of 3D gaming.
It is even hard to tell that the actors in Avatar are 3D because they seem so much flatter versus the 3D CGI stuff (plus the low frame rate of that movie).
It is exciting to think about it though. anything rendered in a 3D volume at a good frame rate would appear to be a real object floating in space. If that 256^3 light cube is scaled up even further so that you can see stuff with a good amount of detail..
Hmmmm....
http://www.youtube.com/watch?v=f1YNyQqbiF0
something I used to be passionate about, holographic displays. how to light a pixel inside a volume without it and the wires, being so big. The best plan so far seems to be making a matrix of points in a space where two laser intersecting at that point, providing enough energy, makes the point light up. problem is building on that tiny scale, enough pixels to make a real picture worth looking at. Like you need billions of points and so you need tons of laser lighting up all those points fast enough to make a 3D picture.
I wonder though,
http://www.youtube.com/watch?v=8G3wWmtkN88
Those are unwatchable. There's too much crap (the ads stuff, or whatever once you own one and want to use it yourself), behind the picture to pay attention to the picture (the clear parts of the picture. I mean the point is for the picture to be transparent.. but why?)
The saving grace is that the rendered picture becomes it's own volume inside the 'picture' frame meaning unlike that transparent display where the picture is a drawing on a transparent sheet that has no light of it's own (depending on your POV the picture is brighter, or dimmer), a proper volume display renders a picture in 3D but also lights it up so that you can't see through the rendered object from any POV (like the cube in the first link).
Also the only way to appreciate 3D is if the other sides (from your POV) is black (for contrast). So that your bright rendered 3D objects or scenes aren't competing with whatever is behind the display for your attention in the frame of the display.
It should be easier to make a volume display like those 3D cubes if you're only able to view it from one side. I mean who walks around their TV? The last think Samsung or Sony, Fox and NBC are going to do is get me to walk around my TV while I watch. POV is static for a defined viewing angle.
Otherwise I mean if you had a virtual receptionist or something in a tube of pixels them yeah maybe you'd want to walk around it, but the virtual receptionist would have to turn and face the person interacting with it while the others would see whatever side is being rendering from their POV. Pointless unless it was a real hologram, aware of each viewer and they could see from their POV a unique rendering. For each person looking at the tube, they would see the virtual 3D secretary/or tour guide, looking at them. Same applying for TV. How would you have a news broadcast in 3D unless each viewer felt like the newscaster being rendered was looking at them?
I'm not personally happy about the concept of 3D (on a screen) or volume displays that need glasses either, because they rob some of the light and have a limited frame rate. Just the same as LEDs and wires in those cubes block light. If they are made small enough to be almost invisible, and for it to be hard or impossible to see the space between each point of light, like the transparent laptop screen, and still be suspended in a cube, would you get the same resolution? or would brightness fade deeper into the cube from your POV? how small would each pixel have to be for the cube to be crystal clear.
IF instead the cubes (as displayed above) render as fast as they can, for each pixel, then there are no blank frames. It would be like an analogue display. Which panel makers should be able to accomplish now with digital displays given the high resolution and tiny dot pitch, in a flow-frame method where pixels that don't change don't go black, they just color shift as needed, instead of displaying full pixel frames and shuttering them, but that's another issue.
Still.. at 256x256x256 a 3D volume display would be kick ass, especially for gaming, as long as it was crystal clear. That 16x16x16 cube is already very impressive but each pixel point of light is so big it becomes like muddy water half way through.
That and I wonder if you started with something like water, added bio-luminescent cells (which can be grown in a lab), you could possibly grow the framework, with electric links between bundles of cells (for each color of RGB and CYM, (how bright can each one be while still being transparent enough to work as a high definition display, and growing it in water, can you leave the water in there without it distorting the picture or would it have to be drained). Would artificial bio-luminscent cells be small enough to work as pixels in a cube?
And using water to grow the display means you can 'repair' it by refilling it with water and either clearing out bad sections or dumping the whole volume and starting from scratch. Plus using cells, they might even be able to go opaque (providing a bright background sky to a scene, or dark murkiness).. problem being how fast they could refresh or colour-shift. Again, I think if you aren't dealing with shuttering frames in the whole volume, that having only the pixels that need to shift, do so, that you can trick the eye into seeing movement more naturally by shifting groups of pixels less quickly than whole frames more quickly. I wonder if it would cause ghosting issues though..
For TV would people mind dropping resolution and picture quality to go from 3DTV to holographic/volume displays like those LED light cubes? Some people are talking about going directly from 1080p (for 3D you'd double the number of frames), 1920x1080 pixels per frame something like 24 or 60 frames per second, to 1920x1080x1080. All well and good, but would having something like 16 billion pixels even be necessary (nevermind possible)(16 billion from 4000x2000x2000), colour shifting, or the whole volume of 16 billion pixels per frame refreshing at least 60 times a second.
Just one frame in a volume like that would take, from 1 megabyte to 16 gigabytes (or more), even with an average from 3-8 gigs, time 60 frames in a second, something like 50 gigs per second, modern technology can handle it, but you'd need a beast of a computer to power that picture.
Considering doing it for movies first, like if IMAX wanted to support SuperHiVision, something like this:
http://www.youtube.com/watch?v=am_azZxTFOI
Where they say you can't see (appreciate) the picture on a screen smaller than 100".. I wonder what lunatic programming (some 10 to 15 years from now) would even be worth the effort of a 3D version of SuperHiVision... 100" diagonal and what four or five feet deep? @.@ 150 billion pixels per frame at 60-120 frames per second. Totally impossible with modern technology...
BUT if you think about 3D gaming (on much smaller screens), and from that large LED cube (the second link) the real point is that you can't see the whole seen using only 17 million pixels of a normal game screen at the same time no matter what. You need z buffering to turn a 3D scene into a reasonable 2D picture. If anything you'd only need 3-6 pixels depth to add richer more vibrant colours and to help fake true depth.
Meaning those 16 billion available pixels in a reasonable LED cube display, running 3D games which can render infinity inside the computer, only need to show you a 2D picture, even in a 3D volume. Nothing really changes (for a single person looking at the display), you only need the 4000x2000 x 3 or 6 pixels for RGB CYM colour and brightness for the picture to kick ass. Anyone else looking into the volume from the side might not even see the picture because only small sections would light up all over the cube.. only the person looking at it from the 'front' needs to see the picture.
And if you're faking 3D anyway, with glasses, you don't need the full 2000 pixels deep, it's hard to even notice that something is farther away without a frame of reference... which I suppose is the hardest to imagine about a 4Kx2Kx2K display without seeing one in real life.
That chinese model running 256^3 pixels when displaying a fish near the end, is hard to see as a 3D object.
The picture is about the same size as a desktop icon blown up to the size of a room. There's no detail to see a tiny pixelized fish floating around at 256^3.
At 2K^3 or 4K^3 you'd be able to tell the fish was the clown fish Nemo and maybe even with a high enough refresh rate it would be eerily lifelike, giving children nightmares for the rest of their lives. But would you be able to tell it was 3D without shading and blending.. stuff that again, is best understood if you're a fan of 3D gaming.
It is even hard to tell that the actors in Avatar are 3D because they seem so much flatter versus the 3D CGI stuff (plus the low frame rate of that movie).
It is exciting to think about it though. anything rendered in a 3D volume at a good frame rate would appear to be a real object floating in space. If that 256^3 light cube is scaled up even further so that you can see stuff with a good amount of detail..
Hmmmm....
pia:
thanks for the comment in fluor love 