3D "Crystal Ball" Monitors

glesga_kiss writes "Actuality Systems have issued a press release announcing sales of their 3D display technology, as reported by Yahoo Finance. The system works similar to an old spining disk optical illusion, except that the 21st century version produces an image that can change through the use of digital projection. In this case the screen is a rotating disk that is capable of producing light at any point that it passes through. The upshot is that you get a real 3D representation of your object, that can be viewed from 360 degrees around the display, without the need for any special goggles. Not quite ready for Hollywood, but the scientific and engineering communities have some obvious uses for it already..."

Comments

photos

[sstory]

Why no one linked to the photos is beyond me, but slashdot posts are well-known for poorly-placed/defined links. Anyway, here it is.

Deja vu

[caouchouc]

Hmm... This seems all too familiar.

More Information and schematic

[LiquidCoooled]

here have fun...

I disagree; occlusion is possible

[Anonymous Coward]

Actually, spinning-screen displays are capable of viewer position-dependent effects, such as occlusion. The spinning screen isn't the point - it's the screen. In order to make an arbitrary light field (through piecewise approximation), you need to be able to control both the amplitude AND the trajectory of each "ray bundle." If you use a screen that is not a diffuser, but something with beam-steering capabilities, you can do occlusion. For instance, see US Pat. 6,487,020.

patent link

-gregg

Re:Not Ready for Hollywood

[Gilmoure]

...And have a kid!

Re:Comment and mirror

[Mac Degger]

Uh---did you look at the sugar molecule? Or any of the other pics? This thing runs on voxels...it does do solids.

As for your explanation as to how this thing works...it's woefully lacking and even misleading. The thing displays a full slice every degree or so. It creates the illusion of solidness the exact same way moving pictures are faked: the slices change for every angle of rotation and with an rpm of 760, you get multiple slices per angle per minute.

A quick view of the sugar molecule movie shows how this does work for solids.

(btw, I saw the movies a couple of years back [2001 I beleive], so maybe they're not there anymore).

Re:Comment and mirror

[IPFreely]

Um, No. Actually, you missed the point.

Whether the image is solid, wireframe or just points, you will be able to see through it. The way you solve this in 3D projection to 2D surface is to use hidden surface removel methods to not draw the obscured surfaces, Z-buffer being the most common for 3D accelerated cards on PCs.

In true 3D like this, you do not necessarily know what direction the user is viewing from, so you do not know which surfaces should be obscured. When it draws the backside, you WILL be able to see it through the front side. There is nothing solid about the front side, it's just a light hanging in space.

If the viewing direction IS known in advance (as in a prepared movie) then you could use hidden surface removel methods to alter the displayed image and remove the backside, but just from that one angle. But in general, the spherical nature of this display makes no rules about the viewing angle.