Lens, Focus, Light and Planes

[Tjack]

Ok I'm hoping someone can give me some good technical resources to help me out.

I'm confused by this Depth of Field

Ok so a point of light spreads out to the lens like a cone, and is then refracted back and hopefully converges to a point on the film.

But if that cone converges in front of the film, it starts another cone...

What is the nature of this cone? Is it upside down or right side up?
Why aren't some out of focus objects upside down and others right side up?

 

[Big Mike]

Welcome to the forum.

I'm not sure what your level of knowledge on this subject is...buy you might be digging too deep here. I don't think it's really necessary to fully understand the physics...but if you want to, more power to you.

To really understand DOF, you need to do some reading on the aptly named Circle of Confusion.

 

[Stosh]

Let's start with your point of light example. The point of light doesn't spread out to a cone. It emits light in all directions and the lens captures a small portion of this light. Think of it like the sun. It emits light out in all directions. The earth only gets a tiny portion of it's entire light. They just use a cone to show only the light that we're interested in because it's striking the lens.

Now we're to the lens. The lens grabs a small portion of light (let's say a circle of a couple of inches). At the moment the light hits the lens, it's just light. It's not an image, so therefore it can't be in any orientation. The lens bends the light (from a point source) to form a point at the focal plane. Since this point of light is just a point, it has no orientation on it's own. But when you combine many points of light they will be arranged upside down and reversed left to right.

Now move on to a "real" image. It is made up of infinitely many points of light all combined together to form the image. This too of course is upside down and reversed on the sensor, but the hardware or software correctly orients it for us.

Depth of field is much more complicated. For any object there is only one true focal point (or more accurately one true focal plane), but depending on the focal length and aperture, there are varying levels of "acceptably close" to in-focus. To greatly simplify it, the steeper the light cone from the lens, the more critical focus is. The shallower the light cone (or closer to parallel), the less critical focus is. To quantify the steepness of the light cone we use f-ratios or f-numbers or f-stops which all mean the same thing. If you care to know how the f-ratio is calculated, it's the focal length divided by the aperture (effective physical size of the inside of the lens). A lens with a focal length of 50mm and aperture of 18mm would have an f-ratio of f/2.8

Search the net. There are many sites with diagrams that can visually show you what's happening.

 

[Tjack]

ok but if the image is upside down on the film, how come it isn't flipped right side up if the focal plane where the light converges is in front of the film?

Like a magnifying glass that flips the image if you move it back and forth

Someone told me it was because the lens in a camera is made so it cannot be moved far enough from the film for an image to be flipped a gain.

what do you think :study:

 

[Stosh]

When you compare the film plane or sensor plane to your eye it gets very confusing because you eye has its own optics. A film plane does not have optics.

If you move the film plane in front of the convergence point or behind the convergence point it's just out of focus. No image flip happens at the convergence.

The image flip happens with your eye and magnifying glass because your eye is capable of refocusing the light coming through the magnifying glass. If your eye is beyond the convergence point, then the image is flipped. And if your eye is in front of the convergence point, then the image you see is correct. That is of course as long as the magnifying glass is close enough to the object so it's inside the convergence point on the other side lol.

If you wanted to simulate the magnifying glass experiment with a camera just put a magnifying glass (or any lens) in front of a camera with lens already installed and now you'll get the same thing. A point and shoot camera would be the easiest.