Inno3D Home Page Inno3D Home Page

FAQ News Search Archive Forum Articles Tweaks Technology Files Prices SETI
Visit NVIDIA's home page.
3D Imagery Menu
  The Human Eye
  Stereo Vision
  Stereo Viewing
  3D Glasses
  Natural Techniques
  Theoretical Methods
Favorite Pics
Click to Enlarge
OCZ Tech Titan 3
1.0GHz Pentium III
eVGA MX Shootout
nForce Preview
VisionTek GeForce3
2001 Spring Lineup
GeForce3 Preview
eVGA TwinView Plus
VisionTek GF2 Ultra
OCZ GeForce2 Pro
Winfast GeForce2 MX
GeForce2 vs Quake3
Linksys Cable Router
GF2 FSAA Shootout
GeForce2 MX Preview
Benchmarking Guide
Quake 3 Tune-Up
Power DVD Review
Live! Experiences
Memory from

FastCounter by bCentral

 Visitors Are Online
Powered by
3D Chipset
Gamers Ammo
Reactor Critical
GeForce FAQ
Dayton's Misc.
G-Force X Sweden
Maximum Reboot
Media Xplosion
nV Italia
Riva Station
nV News Home Page

3D Imagery

Theoretical Methods

Why Don't We Have "Holo-decks"?

On TV and in the movies we see a whole different kind of 3D imaging. Programs like Star Trek often show a holographic image actually projected into the middle of the room, and the users walk around the object, viewing it from different angles.

Actually Star Trek takes things even further and allows the user to touch the holographic images, fight holographic ninjas, and bed holographic women. Of course what we see in Star Trek is just special effects, but why can't we generate holograms for real?

Note: A lot of this is just speculation and ideas.

Experts in the fields of holographs are often asked to create a holographic object in the center of the room that people can walk around and observe from all angles. To do such a thing is presently impossible because of one fundamental problem: How can you make light reflect off nothing? You see, light is only visible if the photons actually reach your eye. An object is only visible if some light strikes it, bounces off, and some of the reflected light reaches your retina.

To create a hologram in thin air is impossible because there is nothing there to bounce the light off. I have seen TV shows and commercials where they project the image by focusing several laser beams in the center of the room. This wouldn't work either. A laser beam is only visible if it reflects off something. This common misconception is obvious to everyone who plays with a laser pointer for the first time. They expect to see some Star Wars like laser beam shine across the room and instead see nothing but a small red dot projected on the wall.

Now, you may have seen laser beams at night clubs and alike. In these places the beams are visible. Why? because the rooms are filled with artificial fog, and not to mention tobacco smoke. The laser light bounces off all the little particles of fog and smoke and its the illuminated particles that you see as being the "laser beam". If you have a laser pointer try it outside on a foggy morning, it looks really cool.

Anyway if this were a movie we could generate a subspace field, or project the photons directly into the middle of the room over a transporter beam. But this is real life, and I intend to discuss this realistically. But lets humor the idea of the laser based system for the moment and look at some other limitations and considerations of it.

All with a dot

We can't create a holographic image with beams of light as it would look something like this...

Instead we need dots of light to construct and image. If one dot could be projected in three dimensional space then we have it made because it would be a relatively simply task to adapt that into rendering a complete image. A complete image could be made up millions of points of light. You monitor works by moving a single point across the screen like so..

The beam scans across the surface of your screen and is alters in strength as it does so. When this is done 75 times a second (a cheap monitor) the result is an apparently solid image. With a dot projected in 3D space, which we call a voxel, this process could work also. The voxel would move in 3D space, in a scanline fashion in three dimensions, something like this...

By varying the voxel in brightness as it scans we could render a 3D image. Directing the beams in a scanline pattern would be easy, you could just use a set of rotating mirrors. The laser tubes themselves don't need to move.

Making The Dot

Now that we have established that we can create a 3D image with a single dot in such a manner, how do we project that dot? Well as I said before light isn't visible unless it reflects off something. But lets say that we can fill a three dimensional space with particles of matter, e.g. fog. Then we could focus several laser beams onto one spot. Each laser beam would be too weak to be seen individually, but when they all meet at a particulary point in space, the particles in that area of space reflect sufficient light to be seen as a dot...

The problem with filling a room with fog or mist is that it would be impossible with current technology to keep the density of the fog consistent and uniform. Even if you could it would be distorted once someone in the fog filled room moved, talked, breathed, broke wind, anything to cause a little draft and upset the placement of all the little fog particles. The same would apply for a liquid too.

But what about a solid? It might be possible to construct a three dimensional cube of a glass like material with reflective pieces of matter uniformly placed throughout its structure. Furthermore the placement of the reflective pieces could be ordered in such a way that a laser beam could strike any piece without any other pieces getting in the way.

So if this were possible, what would you have? well you would have a cube like solid of glass like material, within it could be projected a three dimensional image. This is called a volumetric display. Using laser beams in this way is just my idea. There are probably other (and better) ideas for creating a volumetric display.


Our imaginary laser based 3D system has another major flaw. The image it projects is transparent. If we try to project a block it may look like this.

We can see that all sides of the image have been rendered, it is truly 3d, but the downside is that we can see far parts of the image that should be obscured by the closer parts. The image is just made of light, there is no real matter there to block the light from the far side of the image. So how could we fix this? Well if we only had one user, then he or she could wear a simple tracking device that allowed the computer to see where the viewer is and which parts of the image should be visible, and which parts should not. Such tracking devices are already used in virtual reality.

Real Life Volumetric Displays

Well, as we have seen we cannot project a hologram into the middle of the room like R2-D2. But its feasible at least to create a 3D image within a confined space, or volumetric display. No volumetric displays are currently available, but there is some research going on in this field. The details on how such systems work is usually kept secret so info is hard to come by. However, they usually fall under two types: Swept Volume, and Static Volume displays.

Swept Volume displays rely upon a moving surface that moves so fast that the eye perceives it as being one solid display. You can see simple examples of these displays in science or novelty stores. One in particular is a wand with a row of LEDs (small red lights) on it, when you wave the wand back and forth rapidly through the air you see a message.

More complex version of this technique can render multi-layered 3D images.

Static Volume displays are like our discussed laser based idea. They contain no moving parts. Here is a quote directly from this article.

"Whereas most volumetric systems under development at present are of the swept-volume type, static-volume displays, which do not employ a moving component to sweep out the display volume, are also under development. One means of generating isolated voxels in a static volume is to utilize a stepwise excitation of fluorescence processes at the intersection of two invisible (usually infrared) laser beams. This requires that the display volume contain atoms, ions or molecules that exhibit this behavior with suitable quantum conversion efficiencies, output fluorescence frequency and decay times. The most promising medium at present employs rare-earth ions doped into an infrared-transparent glass. An alternative static-volume technique has the display volume composed of a 3D array of individually addressable voxel elements. This method may thus achieve ultimate parallelism, with every voxel addressable each refresh."

Errm yeah, in English? Well basically they are using laser beams in a similar way that I discussed earlier, although their idea is better. The infrared laser beams are invisible, but the elements within the volume radiate visible light when struck with infrared laser. Since I have no idea what quantum conversion efficiences are, I will stop talking writing here before I get any further into a big pile of you know what :).

Next: Conclusion

Last Updated on February 24, 2000

All trademarks used are properties of their respective owners.