DOF theory question

smelni

smelni

Veteran
I ran a test to help understand the math of DOF conceptually and got a interesting result but cant really explain it.

Test setup as follows:
camera -> 3 feet -> chart-> 5 feet -> wall

We set the lens to 15 mm and marked on the monitor where the chart height landed. We then put marks on the wall on the edge of frame.

Then we moved the camera back to the following setup
camera-> 6 feet -> chart -> 5 feet -> wall
The camera is now double the distance to the chart - we set the lens to 30mm. this put the chart at the same image size - now however you see less of the wall in the frame.

This is a demonstration of the old trick of backing the camera up to make the DOF seem shallower - but if you consult charts for DOF - the DOF for the chart in this example is the same at both distances becuase the distance doubles and the focal length doubles.

The result of having less of the wall in the frame is that it is "stretched" to fill the frame and therefore looks more out of focus. This makes sense since the chart is double the distance (6 feet) but the wall is less then double the distance so you will see less of the wall.

SO here is the question (if you managed to follow this up to now) - If the DOF around the chart is identical - why is the wall less in focus - it seems that if you extend beyond the DOF the properties of the images are different in the 2 examples - but why is it the same within the DOF but different outside in terms of what is in focus?

a mouthful I know - and hard to explain - but any insight would be helpful
 
beacause its essenatially "Zooming in on the background" so the DOF is the same, but think of blowing up footage in post, it gets blurrier.
 
If thats the case then by that logic the depth of field should also be narrower - but it is not
 
its because its a log curve, the dof realtive to focal length. ITs strange but consistent, and predicatble.
 
Yes, just remember, like Timur said, the DOF is always relevant to the focal length, which will always be the same at a specific area on a lens, not taking into account aperture. Not sure if this helps, but hopefully you'll understand it.
 
This is why I love to use a telephoto for most close-ups or medium close-ups. Though the DOF with a telephoto (with equal subject framing) is the same that it would be with a lens of a shorter focal length, it still appears to be much more shallow. With a telephoto, you can close down the aperture to lengthen the DOF, and yet still have a nice shallow looking DOF. This can give the actor a little more room to move around without worrying about slipping in and out of focus; and you will still get a nice shallow looking DOF to focus the viewer's attention on the actor.

If you consider the angle of view that different focal length lenses provide, it makes a little more sense. Also keep in mind that DOF is a measurement that has to do with the total area that is considered to be in focus. It does not account for how quickly or slowly the focus falls off outside, or inside, the DOF (the curve that Timur mentioned).

This page has some info (about 3/4 the way down:
http://www.cambridgeincolour.com/tutorials/depth-of-field.htm

My DOF tutorial touches on it as well (part 5):
http://snodart.com/tutorials.php
 
Last edited:
you see i liek to shoot with my 35 lenses at like, F8.0. The shalow look is still there, but the backgrounds are definitly softer than stock lens.
 
TimurCivan said:
you see i liek to shoot with my 35 lenses at like, F8.0. The shalow look is still there, but the backgrounds are definitly softer than stock lens.
Click to expand...

Yea, you have to love adapters... so much more freedom.
 
yea definitly. Forget the shallow DOF, im actually more pleased with the increase in lattitude.
 
TimurCivan said:
you see i liek to shoot with my 35 lenses at like, F8.0. The shalow look is still there, but the backgrounds are definitly softer than stock lens.
Click to expand...


I got a chance to shoot @ F11 on a 135mm Lens on a shoot that had ample light. It brought tears to my eyes.

I could actually see the background details, but they were "defocused" enough to not draw attention to themselves.
 
I found it_ I knew I had it around here some where- here is the math behind what you are experiencing:

ND = H x S / H + (S-F)

FD = H x S / H - (S-F)

ND = Near distance
FD = Far distance
H = Hyperfocal distance
S = Distance fomr camera to object
F = focal legnth of the lens

TimurCivan & Sondart did a good job explaining it. It all comes down to physics. To put it yet another way, as said by cinematographer Blain Brown: "If you frame a 6' tall person from head to toe with a 50mm lens and then you put on a 24mm lens and reframe accordingly - moving the camera significantly forward, you are decreasing the subject distance and increasing the image magnification, both of which decrease the depth of field. This will then leave you with the same DoF as you had with the 50mm lens.
 
You end up with the same DOF, yes. But not the *appearance* of the same DOF.

It's somewhat tricky, but the way it works is -- the background is equivalently optically focused. But it's magnified. Any picture, blown up enough, is going to look less focused than it actually is. If you took Ryan's example above, and did the 50mm lens shot, and then the 24mm lens shot, you'd end up with a similar-sized subject, but a very different-sized background. There'd be much more background in the 24mm lens shot. So the 50mm lens is going to have a smaller portion of background, but it will be optically magnified to fill the screen. That optical magnification is going to make it look less focused.

If you took the 24mm lens shot, plucked out an equivalent amount of background and magnified it in PhotoShop to fill the shot, you'll find it looks exactly the same as the 50mm lens' background.

Or, to drag out this picture once again, check this out:
DOF-Combined.jpg


That's a real-world example of the situation Blain Brown is describing. If you look at the stucco wall, the telephoto shot looks like the background is much softer, right? But if you shrunk it down to the same basic size as the wide-angle-shot's background detail, you'd find that it's equivalently focused. Or if you took the wide-ange shot's background and blew it up so the leaves were exactly the same size, you'd find they're just as out-of-focus as the tele shot's leaves are.

So -- technically the DOF doesn't change. But as a practical matter, it sure looks like it does. The illusion of shallower DOF applies when you choose a more-telephoto lens.
 
The example Barry has shown is also why closer shots on your talent lend themselves to longer lenses compressing space between actors and making their relative size appear more equal while shorter lenses lend themselves to wider shots.
 
I dont think i have done a good job really getting at the crux of the question. I understand all of this about DOF - let me try to clarify

dOF is based on a circle of confusion - being a small point of light and at what point does it cross into being out of focus based on the circle of confusion you choose. So with a 50mm lens at 20 feet and a 25mm lens at 10 feet - the DOF is the same so a point of light the same distance from the subject within the DOF will be sufficiently in focus -

now here is the crux - why are the optical characteristics of the area inside of the DOF the same but different outside. It would seem to me that the stretching should also also effect the space inside the DOF and make is shallower - but it only effects the area ouside the DOF - I understand the curves have different slopes but how is it possible to be exactly the same inside the DOF and drastically different outside.
 
I'm not so sure that the "optical characteristics of the area inside of the DOF are the same but different outside". The focus fall-off is a curve. The difference might be more drastic outside the DOF, but there is still a difference inside as well. The difference inside doesn't really matter though because it starts out being as small as a single point of light and then curves its way up until it reaches the size of the permissible circle of confusion (all of which is seen as in focus). The DOF formula does not calculate the curve; It only calculates the total area that will be in focus. To understand the reason for the change in the curve when the FL is changed, you might need to dig pretty deep into optics.

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

http://en.wikipedia.org/wiki/Optics
 
Last edited:
Also keep in mind that the DOF is not always exactly the same in these cases. It will vary slightly (especially when the FL grows smaller). It won't be a very big difference though.

Another point is that even though the DOF is the same as the FL changes, the DOF's distribution on either side of the focal plane is not the same. So, with a short FL lens, a higher percentage of the DOF will be in front of the focal plane. It might be something like 65% in front of the focal plane and 40% behind the focal plane. As the FL increases, the percentage get closer to 50/50 (the rear is alway larger though, even if it is by a very small amount).
 
ok - then now it makes sense to me - that was the clincher - if the distribution is different then it makes sense that the curve is different and therefore further away is softer - thanks much snodart
 
You must log in or register to reply here.
Back
Top