Circle of the confused

Nawty

Nawty

Messages
6,473
Name
Ned
Edit My Images
Yes
I've been thinking (always a dangerous thing!) a bit about the differences between FX and DX lenses and I know that in order to fill the FX frame you need physically bigger (I assume wider?) glass.

That then got me thinking about the circle of confusion, of which I know little so I did a little reading and it all concerns focussing to a point.

But, clearly the sensor is not a point.

So how do you get a whole frame in focus? is it just that each pixel is a point and the lens can focus evenly over each of these points (and that's what is REALLY clever about lenses), or is it that the true focus point of the lens is behind the sensor and what you are actually doing is miss-focussing the convergence point so the sensor is in focus, or something completely different?

Can anyone enlighten me? :)
 
The idea is that the circle of confusion ( and it can be confusing) is the size a point can be still be seen by the human eye and still seen as a sharp point rather than a fuzzy circle.

This relates to an image on a print, and from memory it's an 8 x 10 print held at a specific distance ( I think its 10 inches but I could be wrong). It relates to film technology, rather than digital, but it tends to still hold true.

What it means is the eye is fooled into thinking an image is sharp when it isn't ( up to a point) The human eyes resolution is well below that of the camera so the brain is fooled into thinking the image is still sharp. Look closely at a photo in a newspaper , you'll see it's made up of dots of various sizes. Now at normal reading distance you don't see the dots, you simply a continuous tone image. The Coc works on the same principle.

Now of course having a bigger print and viewing it from the same distance as an 8 x 10 would result in a having to have a smaller CoC due to the magnification. However view it a a "normal " viewing distance, with the original CoC value. then the image will appear sharper due to the eyes inability to resolve extremely fine detail.

It doesn't matter if the recording material is film or a digital sensor, it's the visual effect on the final image, be it on screen or on paper.

A cropped frame or APS camera would normally have a smaller CoC than a full frame one, due to its need for a greater magnification of the image to match the full frame camera

A typical CoC for an APS sensor is around 0.018mm but for a full frame model it's 0.029mm ( Yes I looked it up ). If you want to you could work out how many pixels that would cover on your sensor, but I suspect you've better things to do
 
Thanks for that :)

I understand what you've said (what with all the recent arguments about DoF) but what I still don't understand is how the whole plane of the image is captured (in focus), rather than just a point.
 
Basically you can't get a whole frame in focus unless it's a flat plane (like a brick wall) and your lens is sharp from corner to corner.

Once you start to phograph normal subject at different distances from your camera some parts will be in focus and some won't (Depth of Field) and the longer the focal length of your lens the shallower this DOF becomes (assuming the aperture remains the same.)

And the converse is also true which is why you don't need to focus when taking pics on your phone - the FL of the lens is so short that the DOF is extremely large.

.
 
On a landscape shot for example, with the foreground and the distance in focus
We choose the CoC size by using the variable aperture restriction to force the lens into only collecting narrow light rays that will form this acceptable CoC at the sensor.

In this way they also appear as sharp as those focused at the focal plane…hey presto, all of the scene looks in focus ..even those parts we didn’t focus on.
 
Thanks guys, I wasn't actually talking about depth of field, which I understand. I was asking about WIDTH of focus across the sensor/film.

Having been reminded of my A-Level physics (it was a long while ago now!) I remember that the whole point of optics is that the angle out is relative to the angle in and therefore the subject position in relation to the glass affects where that image is projected.
 
Nawty said:
Thanks guys, I wasn't actually talking about depth of field, which I understand. I was asking about WIDTH of focus across the sensor/film.

Having been reminded of my A-Level physics (it was a long while ago now!) I remember that the whole point of optics is that the angle out is relative to the angle in and therefore the subject position in relation to the glass affects where that image is projected.
Click to expand...

I think you're talking about DEPTH OF FOCUS, which is like DOF but applied to the film plane or sensor plane.

The main thing to remember about depth of focus is that it is TINY - measured in thousands of an inch.

More here:

http://en.wikipedia.org/wiki/Depth_of_focus

Hope that helps a bit.

But if you're asking how a lens can give a sharp picture across the width of a flat sensor and not require a circular bowl shaped sensor, well that is down to the lens manufacturers and it also helps to explain why many lenses are not as sharp in the corners as the centre.

In the end it's all down to compromise and the visual acuity of our own eyes.

.
 
Last edited:
You must log in or register to reply here.
Back
Top