tk3dmaker3

Why anaglyph (red/cyan) 3-D works best on a computer screen.

There are a lot of 3-D movies available in 3-D anaglyph, usually they come with a pair or two of red/green glasses - actually red/cyan.   For comic books these work fine.   Good publishers adjust the colors of the printing inks to match the colors in the glasses. 

When 3-D movies hit big, in the 50's, they used 2 projectors with polarized images on a silver screen.  The viewer wore polarized glasses.  This worked quite well for separating right and left eye images which were projected on the screen at the same time.  It was complex and expensive undertaking.  Since Hollywood seemed to view 3D as a fad, there were a lot of poor movies made very quickly; and within a year 3D had mostly vanished.  The best films were made late in the run and got little or no 3-D release (Dial M for Murder, Kiss Me Kate, Hondo, Inferno, The Glass Web).

Later, looking for a cheaper way of doing 3-D moives, producers turned to the anaglyph option.  It used one projector, a regular screen, and cheap red/cyan glasses.   This worked marginally well but the system and did not see widespread use.  

So why doesn't it work at least as well on color TV?  The answer lies in the way color TV was invented.  Originally, TV was only black & white.  When the networks wanted to bring in color, the FCC said the broadcasts had to be compatible with the black & white.  They couldn't broadcast a picture that could only be seen on a color TV.  The system they came up with was "compatible color". 

What they broadcast are two signals.  Not a black & white picture and a color picture, per se, but a black and white picture and a chroma signal.  The black and white signal is called the lumenence signal and it is, indeed, the whole and complete picture rendered in shades of black, white and gray.  The chroma signal is a kind of overlay signal.  It's not the complete image, it's the colors in that image.  It is analogous to a coloring book.  The picture in the book is complete, all the lines are drawn, there's just no color.  When you color it in, the color picture is now complete.  But in order to broadcast this second (chroma) signal over the same channel they used for old black & white, they had to make serious compromises.  One of the big ones was that the color signal was incomplete.   The whole spectrum of color light was not present.  There were color frequencies that were just not there.   It would be like looking at a rainbow, with lots of gaps where there was no color in it. 

To make 3-D anaglyph work well, you need two distinct and separate images, each viewed only through the proper color filter.   If the colors are not quite right, you see ghosting of the second image.  Since TV is a black and white image with a color overlay, creating a full, rich, detailed second image is all but impossible.  The luminance image (cyan) is full and complete, all the lines are drawn and the color filled in where it should be.  The chroma is left to make up the second image by itself.  The result is an image that is not as sharp, and much of the black & white of the luminance image shows through.  Looking through the glasses, what you see through the red lens is usually a complete image with no ghosting.  The red lens sees the limenance image; the cyan (blue/green) lens sees the chroma image.  It is the opposite of what you might think, but if you look at a red and green dot on a piece of paper, the red dot disappears through the red lens, the green dot disappears through the green lens.

Why does a computer work better?  Computers have a different color space, a full spectrum of color, and the image sent to the screen is an R G B image.  That is, the screen displays the combination of a Red, a Green, and a Blue image, avoiding a lot of the issues of an NTSC TV image.    On top of that, the signal is digital video, which also has very good color latitude.   It is still not absolutely perfect, but it is far superior to NTSC color television.