Discover "Impossible Optics" from Mid '90s Invention TV Show

[SkepticJ]

Back in the mid '90s a show on the Discovery Channel called Invention showed an invention I shall call "impossible optics", since I don't know what else to.

It didn't really explain how it worked (other than a number of lenses were involved, no mention of optical gratings), and it's been so long that I've forgotten who made it. This has made it a nightmare trying to find it again.

What it did - if indeed it really was doing it - was to have everything in focus at the same time, no matter how far away or close they were. A watch face a few inches from the lens and a building across the street and another shot of a caterpillar on a plant in macro-view and a crop dusting plane hundreds of meters away, getting closer and then flying over head and the dust rains down are two of a few shots they had on the show demonstrating this invention. The whole bit on the invention - talking, video and interviews - was about ten min. long, then onto another invention, which is how this show went.

Fantastic show Invention was, if you ever get a chance to watch it in syndication you should. I've only seen one episode in the last ten years because nobody runs it anymore.

Anyway, I'm off track... I've always been good a spotting where special effects are used - and the computer special effects of the mid '90s are a special kind of fake - so I almost totally discount this as what was going on. Plus Discovery Channel was a fully reputable network at the time and didn't show the kinds of BS they sometimes do now, so I don't think it was that either.

But how did it work? This seems to break the rules of how lenses work, hence why I call it "impossible optics". If you know of any links to technical info or a link to the company that made (makes?) it that would be more than my years of searching have ever come up with.

 

[DaveC426913]

This is not the answer, but you can get a full focal range from 0 to infinity with a arbitrarily small aperature.

I have a picture of a model I built where the front and back of the model are entirely in focus (about 18" long). This isn't all that impressive until I point out that the camera-to-subject distance was ... 0. The front of the model was in physical contact with the lens of my camera.

I made a pinhole aperature with a piece of tinfoil poked with a straight pin. I had calculated the aperature to be approximately f300.

 

[SkepticJ]

Interesting. How long was the exposure time?

 

[berkeman]

I seem to remember many years ago seeing ads for dual-focus cameras. They could focus at two different depths at the same time, and produced sample images like the ones you mention.

I tried googling dual focus lens, and also dual focal plane lens, but I'm not sure if any of the hits apply. Ufortunately for the google search, a combination auto focus and manual focus lens is also called "dual focus". You might try that approach on google and see if you can filter your hit list a bit with some more terms. Let us know what you find!

 

[DaveC426913]

Interesting. How long was the exposure time?

That I don't remember. I'm sure I bracketed the exposure.


Let's see, I started with an assumption of f2 at 1/60th.

For every stop down in aperature, you halve the shutter speed.

So, f2 > f300 ([tex]2^{ \sqrt{2}*y}[/tex]=300) is 14 stops.

So, [tex]2^{14} * \frac{1}{60} [/tex] = 273s = 4.5 minutes.

I think I decided that was too long to be practical, and instead did, like, 30 seconds.

 

[Chronos]

You can replicate this effect by looking through the wrong end of a telescope.