FF vs. Crop - DOF changes?

I've heard that FF gives less DOF. This is counter intuitive to me. It should be the same. I went to DOFMaster and calculated the DOF for a 50mm f/1.4 at 5 feet.

30D - 0.16 feet
5D - 0.25 feet

That is more DOF on the FF. Why? Is it because the circle-of-confusion is smaller for the FF since it won't be magnified as much to produce a particular size print?

If I go to an 80mm f/1.4 on the FF I get:

5D - 0.1

Now we have the shallower DOF on a FF which everyone says will happen. So it appears that FF gives less DOF but only if you switch lenses to give the same field of view.

At the same distance the Crop camera will have a smaller DoF, but if you frame each the same with the same lens the crop will be further away, which will increase its DoF.

That was the part I was not able to model in DOFMaster. Shoot something with the 5D, 50mm lens at 5 feet. How far away do you need to be with the 30D to get the same framing and will the extra DOF added by backing up result in more DOF in the 2nd shot or the first?

wilvoeka wrote in post #5089533
At the same distance the Crop camera will have a smaller DoF, but if you frame each the same with the same lens the crop will be further away, which will increase its DoF.

also if you use a longer lens with FF to achieve the same framing as the cropper you will get shallower DOF.

ed rader

wilvoeka wrote in post #5089533
At the same distance the Crop camera will have a smaller DoF, but if you frame each the same with the same lens the crop will be further away, which will increase its DoF.

wrong... DOF is a function of ONLY DISTANCE to the subject and FOCAL LENGTH of the lens. a crop will not change either of these, so the DOF STAYS THE SAME. what will happen though, is that the photographer will need to move backwards to fill the same frame, so the distance increases. the only other variable is in circle of confusion, for which a more dense camera will have a smaller CoC than a less dense one. however, by that definition, a 1dsmkiii with it's 22.1mp in full frame will actually have less dof than a 10d with a 1.6 crop but only 6mp

basroil wrote in post #5089657
wrong... DOF is a function of ONLY DISTANCE to the subject and FOCAL LENGTH of the lens. a crop will not change either of these, so the DOF STAYS THE SAME. what will happen though, is that the photographer will need to move backwards to fill the same frame, so the distance increases. the only other variable is in circle of confusion, for which a more dense camera will have a smaller CoC than a less dense one. however, by that definition, a 1dsmkiii with it's 22.1mp in full frame will actually have less dof than a 10d with a 1.6 crop but only 6mp

Your comments on the circle of confusion are not correct and will cause, well, confusion. CoC is not a function of pixel density at all. CoC is affected by the total enlargement to the print, and thus for a given print size and viewing distance and larger sensor (or sheet of film) will have a larger CoC and larger DOF.

This sensor size effect on CoC is why the OP found the 5D to have greater DOF for equal focal length, distance and aperture. Realize of course that this will not have the same framing.

If the OP changes the original case to put an 80mm lens on the 5D for equal framing he will find the 5D has shallower DOF.

Aaah — no. DOF is perception. A poster-size print will have a shallower DOF than a 6x9.

If you assume equal-sized prints, then this statement is correct. Moving backwards to compensate for the longer focal length will result in the same depth of field as a closer shot using a shorter focal length.

The DOFMaster site provides the equations

http://www.dofmaster.c​om/equations.html

If you take a look at them you can easily see how CoC changes hyperfocal distance and DOF.

JeffreyG wrote in post #5089813
Your comments on the circle of confusion are not correct and will cause, well, confusion. CoC is not a function of pixel density at all. CoC is affected by the total enlargement to the print, and thus for a given print size and viewing distance and larger sensor (or sheet of film) will have a larger CoC and larger DOF.

This sensor size effect on CoC is why the OP found the 5D to have greater DOF for equal focal length, distance and aperture. Realize of course that this will not have the same framing.

If the OP changes the original case to put an 80mm lens on the 5D for equal framing he will find the 5D has shallower DOF.

sensor size is larger, but density is also smaller with 5d. CoC in the optics sense is different than in the printing sense. in optics, the more densely packed a sensor is, the less distance between the first nodes each point can have before being blended with another point. you know how diffraction is a bigger problem the more densely a chip is packed (f16 for ~7.2um chips, smaller ap for less dense). when you view the image, then your medium also matters. from a purely recording based perspective, given a certain density and f-stop, only distance and focal length will change the dof.

basroil wrote in post #5090080
sensor size is larger, but density is also smaller with 5d. CoC in the optics sense is different than in the printing sense. in optics, the more densely packed a sensor is, the less distance between the first nodes each point can have before being blended with another point. you know how diffraction is a bigger problem the more densely a chip is packed (f16 for ~7.2um chips, smaller ap for less dense). when you view the image, then your medium also matters. from a purely recording based perspective, given a certain density and f-stop, only distance and focal length will change the dof.

The higher density (finer pixel spacing) results in finer detail being potentially captured or greater "linear resolution" - no major disagreement with that. Trouble is, DoF does not exist until the image is viewed, so it's not possible to talk about DoF "from a purely recording based perspective". DoF is defined by the areas of the image that are perceived to be "as sharp as the sharpest part of the image" (and the corresponding distance of those objects from the plane of sharp focus).

What you are discussing relates to the degree of geometric enlargement possible (at a fixed viewing distance - ie. print appears bigger and bigger) before the print , as opposed to the human eye's acuity, becomes the limiting factor in perceived sharpness. Again, this is largely -true - but it is NOT what DoF is about!

Here's one more take on it, in more general "easier to use when out shooting" terms:

n1as wrote in post #5089456
I've heard that FF gives less DOF

The real short version is yes, you can think of it that way, *if* you are either changing your zoom length in the lens, or moving yourself with your feet, to get the same field of view with the crop camera as you are with the full frame camera.

On the other hand, let's say you took a shot with a full frame camera, and then swapped only the bodies (but kept the same lens, same aperture, etc.) to something with that of a crop camera. You then later cropped the image from the full frame in Photoshop to match the image taken by the crop camera. Print them both out to the same size and you'll see the same depth of field.

Ignoring the details of pixel densities and etc., in real usage this is basically what you'll get.

AJSJones wrote in post #5090202
The higher density (finer pixel spacing) results in finer detail being potentially captured or greater "linear resolution" - no major disagreement with that. Trouble is, DoF does not exist until the image is viewed, so it's not possible to talk about DoF "from a purely recording based perspective". DoF is defined by the areas of the image that are perceived to be "as sharp as the sharpest part of the image" (and the corresponding distance of those objects from the plane of sharp focus).

What you are discussing relates to the degree of geometric enlargement possible (at a fixed viewing distance - ie. print appears bigger and bigger) before the print , as opposed to the human eye's acuity, becomes the limiting factor in perceived sharpness. Again, this is largely -true - but it is NOT what DoF is about!

eh, i never read dictionaries... i hate those small clauses built into certain words

luckly, at 1:1 crop that doesn't make a difference, but does change def. a large bit when looking at 300dpi from 6in

AJSJones wrote in post #5090202
"as sharp as the sharpest part of the image"

Focus is in a plane, so - DOF would be a range more like "not unacceptably less sharp" than the focal plane, and not really "as sharp" as the focal plane.

Although that's being really picky. In general terms I like to remember, what you said is fine.

The reason a 5d will have a smalled depth of field than a crop sensor is this.

If I stand at 5 feet with a 50mm lens at 2.8 with a crop sensor and take a picture of a flower that fills the frame I will have a certain amount of DOF.

If I take the same lens at the same settings and distance I will have the same DOF but the flower will not fill the frame using the 5d.

So what the normal person does is step closer to fill the frame with the flower. Now the 5d has a narrower depth of field because you are closer to the flower.

That is why people say a FF sensor will create shallower DOF. Because you reframe to fill the image in the frame by getting closer. The 5d does not actually have a lesser depth of field because of mechanics or size of the sensor, but because most people will not waste pixel by cropping away the area around said flower, they get closer, thus the effect.

tracer bullet wrote in post #5090301
Focus is in a plane, so - DOF would be a range more like "not unacceptably less sharp" than the focal plane, and not really "as sharp" as the focal plane.

Although that's being really picky. In general terms I like to remember, what you said is fine.

Yup - nitpicky indeed
That's why my post had the qualifier areas of the image that are perceived to be"as sharp as the sharpest part of the image" - this DoF thing is all about perception