Slashdot Mirror
18-Inch 3D LCD Screens
Rob Polyn sent in a story about a new 18" LCD screen using DTI to simulate 3D. An excerpt describes the technique:
"The second approach to true 3D animation is known as autostereoscopy (which DTI monitors utilize). In this method, two solid and unyielding images are produced for the user to view. These images are merged together, and if viewed by one eye, will appear to be two overlapping images, which don?t quite merge together correctly. However, when viewed with two eyes, autostereoscopy can produce vivid lifelike 3D images."
yay
Can someone tell me how the heck this thing works? The concept doesn't make any sense to me.<P>
Thanks.
Using your sig line to advertise for friends is lame.
Sounds to me like a Magic Eye. Very uh.. odd. I can't imagine people being that enthusiastic about "looking past the screen" to get the correct image.
of pornography awaits! Chalk another one up for science! Whoo-hoo!
Imagine a beowolf cluster of these...
I read about this when a site I check for news a lot had a link to an article about it a week or so ago. I understand the concept, and it seems like a good idea and it might work really well, but I thought the whole point of spending money on technology was so you wouldn't have to deal with the real world, 3D and all :)
This kind of display effectively displays a stereo pair without the need of special glasses etc. Every example I have ever seen requires the viewer to be in a relatively constrained angle to the monitor to receive a true effect. The best 3d I have experienced was a pair of polarized projectors (one horizontal, one vertical) with a passive set of polarized glasses, a camera would track the viewers motion and redraw the scene appropriately, the affect was very much like looking through a window.
Does any one know of any other 3d visualization system being developed, any links would be most appreciated.
If they get cheap and popular, what are we gonna do with them?
The no-brainer applications are gaming and 3d modelling, and someone will certainly come up with a new form of Pr0n.
What would you use a flat stereo screen for?
-- veni vidi nuclei deceri --- I came, I saw, I dumped core.
Unfortunately, this concept has proven to create problems for individuals prone to epilepsy or similar medical conditions. The chance of exciting or aggravating a condition such as this increases if the images are of an autostereoscopy, but are also flashing. Another area of concern is that it seems to cause headaches in many individuals, also.
The goal would be creating a system capable of delivering images to multiple viewers at their respective locations. Current technologies allow a single viewer with a 30-degree viewzone. This may or may not be practical with the current design. Also, another hurdle is producing full color 3-D and proper occlusion (depth cue allowing an object in the foreground to block the ones behind it).
Thanks.
Domenic R. Merenda
Director of Strategic Business Development
BeOpen.com
Something else to make me nauseous while surfing. :o)
$11,000? Yikes!
LCD is gonna be cool. My dream is for a hardware standard puts 12" LCD displays in the stores for cheap, like $100 or less. Each of these LCD displays could function as an independant monitor, but the coolness would be that you could take the plastic edges off and expose the LCD going all the way to the edge, and there would be an androgynous connector running down each side that could plug into another identical LCD. Take four of these and plug them together in a square, and you have a 24"x24" monitor. You could go out and buy a couple panels every paycheck until eventually you were satisfied with the size or had a monitor-wall to run Quake on.
This would work for TVs as well, and could really make it easier to get big TVs without needing to spend so much money at once.
Just an idea...
I wonder how these work for people who are cross-eyed?
thank you.
thank you.
i took a bitchslapping for natalie portman!!
I read that somewhere about three weeks ago... They are cool none the less, but they will be too expensive for the average user until they mass produce them. Sony has apparently opened a factory to mass produce lcd monitors. that might affect the price - I hope.
EOT
"Patience is a virtue, afforded those with nothing better to do." - I don't remember
I'd think (though I may be wrong) that the only sure way to ensure perfect 3D vision would be to use separate screen for each eye, but there doesn't seem to happen much in that area. Are the glasses just too heavy? Or is it damaging the eyes?
Oh well, maybe we'll have to wait for the day when we can just connect our brain directly to the video card.
You can find a list of similar products here. Another interesting link is the Spatial Imaging Group at the MIT Media Lab.
Given one hour to live, the student replied: "I'd spend it with professor FP who can make an hour seem like a lifetime."
That's strange. The testers explicitely say that they tested the screen with non-3d-capable game (quake 3)... No wonder that absolutely no 3d effect is to be expected.
As for the screen itself, the dimness of the image shows that either the technology is not mature, or the test pattern was unappropriate... Or else, the technology simply doesn't work : "psychosomatic 3d"...
-- javaDragon is an instance of JavaDragon.
Personally I find those magic eye pictures to be very disorienting and I have very little sucess with them. It seems to me that this display will end up with many people like me who have problems seeing a magic eye being frustrated. I'd much rather have something based on alternate technology, that wouldn't require looking beyond the screen.
-----------------
It's not really funny, unless someone doesn't get it
In this method, two solid and unyielding images are produced for the user to view. These images are merged together, and if viewed by one eye, will appear to be two overlapping images, which don?t quite merge together correctly. However, when viewed with two eyes, autostereoscopy can produce vivid lifelike 3D images.
That's just not an explanation. But, I figure, it's just a review by some graphics fans. So I checked the company's website. (Which barely works. A peek at the image directory got me this. I guess we know they're hosting on a Mac, huh?) Their FAQ, in response to "Q: I am wondering how your display works?" links to http://www.dti3d.com/dev/, which is not especially useful. I downloaded the developer's package. The readme says:
dti_vw libray diretory has source files for our driver.
dti_vw app directory has sample file for how to use our libray in a application program.
Our library is so simple and easy to use.
There for this sample is good enough to know how our library works.
Our library make a application can communicate between a computer and our unit.
If we change our the communication method and way, we will update immediately.
I gotta be honest: This all looks pretty sketchy. Has anybody seen/used one of these? I'm not convinced that this thing is legit. I don't have the skills to be able to read the code to figure out how all of this works. But "view with two eyes" just ain't gonna cut it for this crowd.
-Waldo
The monitor works by providing 2 images, one for each eye. As the article mentioned, if your program supports it, each eye will get a slightly different image, which provides the 3d effect. Otherwise, each eye will get a copy of the same image. I would guess this is done by angling alternate pixels slightly towards each eye.
You can actually try this effect on your own monitors. Just open 2 copies of any picture, and put them next to each other. Now, look at the pictures thru the monitor, as if they were far away. Eventually, you will be able to merge the 2 pictures together, while everything else goes a bit blurry. When that happens, you are looking at one picture thru each eye. You should be able to get a slightly 3d effect depending on what kind of picture you chose.
You can get 3d cameras that take a picture from 2 angles, and use a special viewer that forces each eye to look at its corresponding picture to view them. The innovation with this monitor is just allowing you to view it without using any special viewers. Of course, the review said that there were some problems with dimness and vertical lines in 3d mode, but these should be easy to fix.
---
One of the links I mentioned refers to the Richmond Holographic Studios. One very nice feature of this technology is that a wider viewing angle is supported allowing multiple people to see the same image. This might not be so important for games, but for architectural rendering or other 'real work' (that is unless you work for a gaming company ;-), this may be vital.
I suspect that there are some strong negatives and would love to read comments from anyone who knows more.
Given one hour to live, the student replied: "I'd spend it with professor FP who can make an hour seem like a lifetime."
.- CitizenC (User Info)
after a little more research i came up with this Philips research paper. (be sure to look at the nice diagrams in the slides).
the gist of it is this: much like 3d postcards, they use a grid of cylindrical lens over the LCD panel. each lens covers a specified number of real LCD pixels, 4 being a common number. since the lens is constructed to have the LCD pixel be at the focal point, when you look at the screen through the lens your eye will be directed towards one of the 4 pixels and not the others. thus the lens has turned 4 real pixels into one 3d pixel. (and dropped your resolution to 1/4th!) if you shift your viewing angle then you will look at a different one. if, like many people, you have two eyeballs which are separated by a few inches, then each eye will see a different image.
another way of thinking about it is to imagine that four zones of images are being projected out from each pixel to your eyes. as long as your eyes are in separate zones then you are okay. this is the case if you are sitting at normal reading distance. but if you get too far away (or have a head the size of a mouse) then your eyes will end up in the same zone and you lose the 3d effect.
philips has also done some innovative work to even out the resolution loss and improve the viewing angle.
- joshy
after reading how it works i now understand why it's so dim. if there is a 4:1 ratio of real pixels to 3d pixels, then each eye is only getting 1/4 the light it used to. guess they are going to have to beef up that backlight. then you can switch back to 2d and have a blinding image reflect of your face, just like in the movies. :)
Prop me up beside the jukebox if I die.
.. as I am blind in one eye. As well as fearing wearable computers for the same reason. How many other "disabled" people out there wonder how we will adapt to the toys of the "perfects" in the future of computing???
I am Slashdot. Are you Slashdot as well?
Thank you osm!!!! Choose your own identity!!! Woo hoo!!!!!!
These exact monitors were slashdotted back in February. http://slashdot.org/articles/00/02/19/1217254.shtm l Come on guys, this is really getting lame.
And it explains the bright/dim pattern. Unfortunately the original review fails to explain how this system works. Thanks for the insight.
On a side not, CmdrTaco posted both of these stories about the same exact LCD screen.
Trouble is: when looking at realistic scenes, motion parallax (i.e., what happens to the image when you move your head slightly), not stereo, is probably the primary motion cue. Stereo cues in the kinds of scenes you get from 3D games are likely more confusing than immersive, since they often simply reinforce the impression of looking at a tiny, toy-like scene. If you want that kind of appearance, you can already get simple LCD shutter glasses for relatively little money, but they probably haven't caught on for a reason.
The best solution for immersive 3D games is head mounted displays, which give you excellent head tracking and motion parallax. The next most important cue is likely peripheral vision, which is a bigger engineering challenge. Once you have a head-mounted display, adding stereo is technically easy (but increases the cost somewhat since you need two displays).
Though full stereoscopy is not widely supported by many recent games, DTI's 3D mode is still useable in games such as Quake 3, and produces results very similar to games that fully support stereoscopy.
Okay, how can this possibly work? How can the driver or anything else possibly guess the distances at which we're supposed to perceive different objects?
I could see how they could produce some uniform 3D effect, such as making the top of the screen appear farther away than the bottom, but how can they do anything which relates to the contents of the image?
However, when comparing Doom, a game which has stereoscopy, with Quake 3, a game which doesn't, the differences were negligible.
Right. This really makes me think any benefits are largely imagined.
/* The beatings will continue until morale improves. */
Considering the other current applications of 3D images - mostly in movies at iMax or theme parks, for its novelty value - I don't think it's likely to take off.
3D movies at the cinema were incredibly popular at one point during the 1940s/50s, and then everyone realised the triviality of being able to see in 3D. Now all we have are the cliched "oh no look out, the dinosaur's coming towards me!" type we see at theme parks. Is this also the future for 3D monitors?
Put on top of this the prohibitive cost, and it sounds like you've just got another side-show for the Epcott centre.
However, I imagine it is a fairly restrictive veiwing angle and thusnot that great if you want to show something to others.
It's not all THAT bad. (If this is what I think it is) there are a SET of narrow angles from the screen where the stereo effect works correctly.
They're bisected by another set of angles where the depth is reversed, and the space between the clean images (normal or reversed depth) has regions where the two images wash into each other.
So a person can sit closely beside you (distance from your right eye to his left is one, three, five, etc. times the distance between your eyes) and simultaneously see the same image.
The main problems are...
- You have to be at distance from the screen equal to a constant times the spacing between your eyes (plus or minus maybe 20%) to get the effect. At the wrong distance the images for each eye also bleed into the other eye, giving you a triple image - the one you want, plus two single-eye ghosts.
- Images TOO far ahead of or behind the screen will give you eyestrain - because your eyes have to focus at the distance to the screen, but the paralax depth cue says the object is far from the screen. So your eye muscles hunt and get tired.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
Additionally, we were shown the herpes virus in 3D, which brought up another major use for this monitor.
...rather, we were thinking, "I can't wait to start my 3D porn collection, this is gonna be so cool!"
And that would not be that...
The medical community could easily implement this into various aspects of training and detection of diseases in patients.
Eh...
...a monitor screen covered with tiny projectors (projexels?) instead of pixels. What do I mean? Each projexel projects a complete image. If you darkened the room, turned off all the projexels except one and held a piece of white paper up to the screen, you would see a complete image of the scene projected on the white paper.
In other words, active holography. Now, this would require a lot of bandwidth if you did it the stupid way. OTOH, it seems you could exploit coherency in the image to a great deal in order to avoid having to retransmit data that doesn't change too often from projexel to projexel. Possibly, something as simple as run-length encoding could do this.
I'm glossing over a lot of details here. This is an idea I've had for quite a while. Also, if anybody tries to patent active holography, they can bugger off. You saw it here first.
BTW, Theirs is $11,000. I'll make you one of mine for $11,000,000.
For all intensive purposes, "whom" is no longer a word. That begs the question, "who cares"?
Stereo games (such as Doom) already generate left- and right-eye data. Non-stereo games (such as Quake 3) could have depth = 1/luminance; dim things are farther away.
Will I retire or break 10K?
Images TOO far ahead of or behind the screen will give you eyestrain - because your eyes have to focus at the distance to the screen, but the paralax depth cue says the object is far from the screen. So your eye muscles hunt and get tired.
wow, I must say your post was very interesting.. I sure didn't know all of that stuff, etc, etc.
But what I'm wondering is, what do you mean by images too far ahead of or behind the screen?
I'm sorry, I'm just a bit slow at this.. do you mean images that are being attempted in 3-D that are too far in a depth beyond the physical location of the screen? (well, I guess so.) That's interesting.. so, that would mean the image effectively "hits" the screen and can't go any farther back, so wouldn't that kill some of the depth of the image and put things on the same plane that aren't mean to be?
Insert mind here.
Think of the LCD screen as being divided into lots of tiny vertical stripes one pixel wide. Now you send the even stripes to your left eye and the odd stripes to your right eye (or the other way around). Naturally the image sent to the screen would not be a normal image but one where the left eye view is sent to the aforementioned even stripes and the right eye view to the odd stripes (this is done through the 3D API like DX7 or a special version of Open GL --though it might have to be a special version).
Of course, like normal photographic paper, normal computer displays (LCD or otherwise) do not usually do this. So now you have to modify them so they will. Both in the cases of 3D photographs and LCD screens this is done by fitting the surface with a v ertical lens over each right-left pair of image stripes which directs the appropriate stripe into the appropriate eye. This (plus the quirks of LCD technology) is why the display has to be viewed from a very narrow angle since from other angles the lenses would not work quite right
At least this is the way it's done sometimes (with the 3D photographs and the Phillips display) and I'm pretty sure this is the way this particular display must work as well.
From the excellent book "The Visionary Position":
By experimenting with the display -- moving, by degrees, from a 20-degree field of view to a 30-degree field of view and so on up to 120 degrees, the team discovered that at the "60- to 80-degree point, it was like a switch went off in your head. Instead of looking at a picture, all of a sudden you thought you were in a place. You had a different way of interacting with the display. You brought in a different set of innate capabilities. ... And we realized more and more that we were onto something really big. We found that you couldn't forget it, because it was like this world was a place. And we found that people learned really quickly when they were inside of it, that there was a remarkable acceleration of the ability to learn these things, to interact with them..."
Currently the best field of view is around 30% with VR goggles. I am excited about 3D, but it seems funny that no one seems to be working on increasing field of view.
Magic Eye hard to look at? Nahh... I loved to stare at them as a kid, twisting the tuning knob back and forth, watching the gap open and close, finding stronger stations to make the gap close tighter.. but I guess I'm showing my age...
satire, n: 1) witty language used to convey insults or scorn; 2) a form of humor lost on most slashdot moderators.
Since the monitor's refresh rate is, as they said, 60 Hz, shouldn't this be attributed to the tearing effects that can happen when vertical sync is disabled? That doesn't sound like the monitor's fault...sounds more like they let the vid card get out of sync with it.
I was wondering if anyone would post that little problem....
I have a bit more than just a "lazy eye", I'm flat out "wall eyed". Each of my eyes looks seperately in it's own direction, whichever one I'm not 'using' at any given point in time tends to wander off in a different direction (loads of trouble with members of the opposite gender) although there is some corelation to the movement (i've watched many hours of tape of my own eyes). Anyway I was born this way, so my eyes have never both focused on any given point - and as a result my brain simply doesn't grok the concept of putting the two signals together into a "3D" world. This was painfully proven when I participated in a reasearch project where they used head mounted displays that were independently positioned for each eye, so that in a non-moving "rest" state each looked directly into the display. The result was a lot of major headaches to say the least.
The bottom line is that "magic-eye" posters, red/blue 3d glasses, vr gogles, and this technology are bloody useless to me and anyone else with monocular vision; but then who cares? I don't. Go for it folks, if you can make this level of technology then just imagine what else you'll be able to do with what you learned trying to make it work.
I tried to RTFM, but it looks like their site has fallen prey to the slashdoteffect.
I agree that field of view is very important for many applications, such as pharmaceutical design.
But rest assured, people are working on it...
This is a biased response -- our firm has developed (and is finishing up the next generation model) volumetric 3-D display with a 360-degree field of view: Actuality Systems
That is, the imagery takes up a real volume, and multiple people can walk around the display to see it from anywhere in the room. We even have a demo set up that lets you pick up a joystick and fly a helicopter over a moving terrain.
Anyhow, if you are interested in background information on 3-D displays in general, let me suggest: SIGGRAPH overview
and (by a team of students in Germany): Survey Article
-Gregg Favalora
3D Porn (not strictly porn, just someone in a bikini).
Personally I find those magic eye pictures to be very disorienting and I have very little sucess with them.
Generally, people have problems with them due to the unnatural eye position needed to view them, you have to be very cross-eyed. This sort of display doesn't require the unnatural eye position.
What is potentially problematic with this sort of display, however, is how precisely you need to have your head position. To view 3-D for long periods, it needs to be *perfect* -- no times when the frames aren't quite in sync (left and and right eye not seeing a frame rendered with the same geometry), that sort of thing. An alternative technology is LCD shutter glasses, although the problem with them is that you effectively halve your screen refresh rate, and you have exactly the same potential problem with out of sync images.
Ooh, a sarcasm detector. Oh, that's a real useful invention.