More 3D Displays to Come

Anonymous Writer writes "The first laptop using an autostereo display to show images in 3D without special glasses was the Sharp Mebius PC-RD3D in Japan, later released in the US as the Sharp Actius RD3D. NEC has a line of computers with autostereo displays as well. They are the NEC Valuestar T VT900/8D desktop, the LaVie S LS900/9E laptop, and LaVie RX LR700/8E laptop. The line uses NEC's SoundVu technology that uses the display as a speaker! Autostereo displays are becoming more popular according to Martyn Williams and Tom Krazit from the IDG News Service. In their article in PC World, they claim laptops are just the start of it. A new satellite service by Mobile Broadcasting will be broadcasting 3D content to handheld devices in Japan some time soon. Another player in this market is Dynamic Digital Depth (mentioned in a previous post of mine), whose content services convert 2D video to 3D for display in this medium. Sanyo may be releasing 50-inch Plasma Displays that can display 3D. MIT's Media Laboratory is developing a more advanced 3D display, calling it a full resolution autostereoscopic display, that would allow a viewer to walk around and not lose the 3D effect, which current autostereo displays can't do."

Comments

3D RasMol?

[clustercrasher]

Anyone planning to hook this into RasMol or PyMol? I would love to be able to look at my protein structures in 3D.

http://pymol.sourceforge.net/

3D Displays

[AnomalyConcept]

This would be extremely useful, especially in the CAD community. While I only know a little about the area of CAD and manufacturing, this combined with the inkjet plastics printing (I forget the term for it) or rapid prototyping machines would be really neat. Imagine designing something, and being able to view it in 3D from all angles (instead of a render), and then sending it to be printed off. I've never seen one of these 3D displays before; how are the objects rendered? How much processing power is needed to create such a display, especially from a 3D model? I'm sure it needs to be rendered first, but what about a flat-shading 3D program like Autodesk Inventor? 3D displays would be neat for new GUIs. Instead of having a flat 3D desktop, you could have a true 3D desktop. That would be interesting to see...

Re:3d displays

[ikkonoishi]

Traditionally the 3D demo object of choice is a teapot.

3D autostereoscopic displays from MIT

[cgenman]

I'm trying to wrap my mind around what exactly that convoluted mess of an MIT press release is trying to say. If I understand correctly, and someone please correct me if I'm wrong, the system tracks the heads of the people surrounding the display, then projects left-eye right-eye information through an adjustable polarized filter and lense system so that the viewable angle only includes the intended eye. The reason they need such a high refresh-rate is because they want a system that would work with 4 people... 4 people = 8 eyes = 8 times the updates.

In essence, that's very cool. Why couldn't they just say that?

And I'm assuming...

[TheMadPenguin]

This doesn't work with Linux? From what I read, they are assuming I run Windows ;)

Small viewing angle

[Saeger]

I imagine that the effective viewing angle on these planar 3D displays is very restrictive; move a couple inches to either side and all you'll end up seeing is the half-resolution image meant for one eye.

The "3D displays to come" that hold the most promise, however, will require that you wear (non-dorky) viewing glasses. These normal looking glasses will use a safe Retinal Scanning laser to directly overlay 3D imagery onto your field of view. Of course, we won't see this tech in BestBuy until the Law of Accelerating Returns has run the course of a few more years.

It's not too hard to think of several killer apps for augmented vision that make all other conventional displays pale in comparison. Even a wall-sized OLED display would take 2nd.

--

Re:Small viewing angle

[cgenman]

Actually, the LCD versions have several viewing angles... move an inch to the side and you're out of phase, move another inch and you're back in phase. It's not perfect, but it works.

Still, the reason that 3D displays are not currently popular is simply that people won't wear glasses. The SEGA Mastersystem had an excellent 3D effect from a simple pair of shuttered glasses. These are cheap and affordable, the type Kasperov used in his last (completely gimmocky) match against a computer. The 3D effect they produce is also darned good, with the only complaint being that you have to put on glasses. Now, if the only problem for the past 20 years is that you have to put on a pair of glasses, yet 3D has never taken off, why do you think there are so many people working so intently on glasses-free 3D?

If you look at the MIT site, they're working on what is essentially an eye-specific projection system from a centralized viewing pannel that uses head tracking and approximation instead of glasses. Now this technology has promise.

Does anything actaully use this?

[bcmm]

Does/will any software actaully use this?

It would be very cool for CAD, but this is going to take up to much processor for real-time gaming rendering, isn't it?

Re:Does anything actaully use this?

[jackbird]

CAD usually end up hitting system resources much harder than games, since games are designed to only render datasets that will provide acceptable performance on the target hardware, while CAD programs, by their very nature, can attempt to render models of arbitrary complexity. Furthermore, with an arbitrary CAD model, you can't get performance boosts from precalculated optimizations like BSP trees.

Also, speaking as someone who spends much of the workday turning 2D CAD files into 3D models, I don't think a 3D display would really be that useful in CAD, except maybe for client presentations. For starters, leveraging a 3D display to full usefulness would require a good 3D input device, and those just haven't arrived.

Furthermore, given the limited number of scanlines, It would seem you'd be restricted to a fairly low number of pixels (depth-xels?) of Z resolution, which could quickly become a problem with fine detail.

Reminds me of Looking Glass

[XyborX]

Wow.. Imagine this combined with Suns Project Looking Glass..

1. Perfect the technology
2. Lower the cost so consumers can afford it
3. "My laptop is more awesome than yours!"
4. Profit!

Anyone ever seen these?

[Titchener]

I would really love to hear a first-hand account from slashdotters who have actually seen these in person, at trade shows or whatnot. Popular media/press releases rob me of my soul.

Build your own rig

[MacFury]

Two of my friends have Canon Digital Rebels. We built a small rig using wood and machine screws that allows us to position the cameras side by side, lenses pointing slightly inward.

It gives us some pretty cool stereo graphic images. The only way I've found to get a true color 3D image is to put both images side by side, then look at their center cross eyed. Is there a better way?

As a professional 3D Photographer...

[callipygian-showsyst]

...I am very encouraged with the new LCD autostereo displays.

3D photo imaging never seems to become mainstream, and not having to wear viewing glasses may help its acceptance, at least in some areas (visualization, gaming).

And there's nothing like the natural appearance of a good 3D Photo.

Re:3D Displays

[NanoGator]

"This would be extremely useful, especially in the CAD community. While I only know a little about the area of CAD and manufacturing, this combined with the inkjet plastics printing (I forget the term for it) or rapid prototyping machines would be really neat. Imagine designing something, and being able to view it in 3D from all angles (instead of a render), and then sending it to be printed off."

Figured since I'm a 3D artist, you wouldn't mind if I chimed in. Would a stereoscopic display help me? If the display is convincing enough, yes! Right now, while I'm modelling, I'm constantly rotating the model around, sometimes just slightly, just to get a sense of the parallax. This gives me a clue as to what vertices are where. A stereo display could potentially relieve me from needing to rotate it as much. If that's true, I could get more detail on the screen without worrying about the vid card not being powerful enough for what I'm doing.

I wish I could tell you for a fact that it would or wouldn't work, but I've yet to experience stereoscopic work-flow. I am rather curious, though.

I wonder about people without stereoscopic vision

[kaladorn]

There are a number of us out there (yes me...), I think around 1-3%, who have effectively no 3D (stereoscopic) vision. In my case, I can detect a profound shift from eye to eye. When I tested on the fancy opthomalogical(sp?) machine where you try to line up 4 lines into a + sign (roughly), I could only ever see two at a time, which two depending on which eye I 'looked' through. In university geology courses, I could never use a stereoscope to examine stereoscopic pictures (trying to estimate a slide-mass was really fun....).

So, I wonder which, if any, of these 3D technologies will work for people with these kinds of problems? Or, will we just become another group of 'informationally handicapped' people? (Which would suck, since I'm a programmer!)

stereo + haptics

[rexguo]

That is great and all, but I believe the problem with all that, is that you -still- can't intereact with the 3D object you're seeing, at the place where it -appears-. That is, you're seeing the object in front of you, but your hand is like 30cm away on the mouse (or whatever 3D input device) trying to manipulate it. That's one thing we solved at ReachIn (a company where I used to work for) by projecting the stereo image onto a mirror, and have a 3-DOF force-feedback device installed under the mirror, so that the hand can be -at the same place as the object-!