[Enkhbayar]

to take the same subject with same lens same aperture, APS-C camera need to be further from subject than FF's.
if take picture from same distance, same aperture, same lens then obviously APS-C's background will look more blurry, but it looks cropped from FF's.

the 2 charts below shows blurriness with APS-C and full frame sensor camera.
first chart is area of CoC and area of sensor ratio, I think this is more close with viewer's feeling of blurriness.
second chart is diameter of CoC and sensor height(16mm or 24mm) ratio.

in both charts
F=85mm
f=F/1.4
Size of view(m):2.0*1.3
APS-C shooting distance 7.0m
Full Frame shooting distance 4.7m




so what do you think?

---

Last edited by Enkhbayar; 01-31-2012 at 01:14 AM.

[Adam]

Same as with the 645D- the bigger the sensor, the less DOF at the same aperture.

[Anvh]

Same as with the 645D- the bigger the sensor, the less DOF at the same aperture.

You mean the 645D has more DOF because the COC is larger.


The problem with these graphs are that you don't use the 85mm the same way on APS-C as you do on FF because the angle of view is different.
So you need to compare the 85mm on FF with 55mm on APS-C and then you get a different score.

[mysticcowboy]

Another variable is aperture. The APS sensor will have a larger depth of field than a FF at any given aperture. You are also dealing with the sophistication of the camera's image processor.

Anyway, depth of field (apparent sharpness) can be an advantage in either direction. Some photos look better with shallower focus, some with greater. Practice and familiarity with your equipment makes the details just something you learn to adjust for.

[RioRico]

DOF is a complex function of photography, presentation, and perception. Toggers likely can't control the mindsets of viewers and may have limited control over how an image is shown, so we must concentrate on the photographic aspects: frame size, focal length, aperture, and camera-to-subject distance.

With all else equal, DOF differs between frame sizes IFF (if and only if) we try to fill the frame with the same image. So if I use 6x9, 645, 135/FF, APS-C, and m4/3 cameras with lenses of the same focal length and aperture, at the same pixel density, and we fill the frame of each with a subject, DOF (and perspective) will differ substantially because we'll shoot from different distances. But if we shoot that subject from the same distance with each, DOF (and perspective) will be EXACTLY THE SAME.

Frame size alone does not determine DOF. It's how you fill the frame that matters.

[vonBaloney]

When people say the DOF is different depending on the sensor, is it to be taken as a given that they are also changing position to fill the frame the same way with the same subject? Or are they saying what falls under "acceptable sharpness" changes with the sensor because of resolving power or something? Because if you don't change position and don't change lenses and don't change apertures and don't change focus points, what was 10ft away is still 10ft away and is being rendered the same no matter what the sensor. I find all these discussions very confusing -- can someone spell out in English just what exactly is being asserted regarding the DOF difference between sensors of different sizes? Please assume ALL else is equal -- just swapping sensors.

[Anvh]

Another variable is aperture. The APS sensor will have a larger depth of field than a FF at any given aperture. You are also dealing with the sophistication of the camera's image processor.

Depends on what you leave the same, if you use the same lens, same aperture and the same distance then the DOF will be smaller on the APS-C then on FF

[Anvh]

Because if you don't change position and don't change lenses and don't change apertures and don't change focus points, what was 10ft away is still 10ft away and is being rendered the same no matter what the sensor.

The rendition is indeed the same so the only differnce is the sensor size.
To explain it very simple, if you have two sensors one is 1,5 x 2 cm and the other is 3 x 4 cm and you want to make a 60 x 80 print you need to magnify the 1,5x2 sensor 40 times and the 3x4 sensor 20 times, so the details of the smaller sensor are rendered 2 times as large.
With that the point where things appear in focus or out of focus will shift because the smaller details are magnified more of the image.

It's a bit how sharp you 800px photo might look on the web but can be fuzy when loooking at it close up, that's the same effect.


this a good read
toothwalker.org/optics/dof.html

[RioRico]

Depends on what you leave the same, if you use the same lens, same aperture and the same distance then the DOF will be smaller on the APS-C then on FF

Totally wrong. If all else remains the same (lens and distance and presentation) and only frame size varies, then DOF is exactly the same. This is shown by the old example: Cut a picture from a magazine. Draw a rectangle on the picture. Draw a smaller rectangle inside that. Those rectangles represent the different-size frames. DOF in the picture doesn't change, no matter the rectangle size.

Now, if we're filling the frame with the same subject, DOF *does* change -- but DOF for the smaller frame is THICKER, not thinner as you said. It's thicker because with the smaller frame, we need a greater distance to capture the subject. Remember the basics of DOF:

* For thicker DOF use a smaller frame and/or tighter aperture and/or further subject distance

* For thinner DOF use a larger frame and/or wider aperture and/or closer subject distance

[SteveM]

I think the two of you are both right, it just depends upon *why* you are asking the question.

That's a very good analogy RioRico and it certainly make perfect sense when allowing the end print size to be the variable, which is technically the correct way to calculate this.

As Anvh points out, many people will use the crop factor in calculating DOF as they use composition as the constant. While one can argue the correctness of this technically, I believe that this is the piece of information that is most typically request when people speak of DOF. People will want to know what the effect is in what they see in the viewfinder, much like increasing the focal length if you stuck with the same format sensor.

I'm not sure however that many people factor in what happens to the picture when you adjust the angle (get closer or further from the subject to set composition in the viewfinder) , how this rounds out faces, changes the size of foreground objects etc.

[Kerrowdown]

Wow this thread is very techy, I'm gonna go back to the beginning again, cos I'm sure I will learn from it, I just missed it on the first pass.

[vonBaloney]

And still we can't agree -- some say you get more DOF with a smaller sensor (which is what I've heard generally, but I think they are assuming a changed position and an "equivalent" lens), others say the opposite. Can we get the story straight?

[Anvh]

Totally wrong. If all else remains the same (lens and distance and presentation) and only frame size varies, then DOF is exactly the same. This is shown by the old example: Cut a picture from a magazine. Draw a rectangle on the picture. Draw a smaller rectangle inside that. Those rectangles represent the different-size frames. DOF in the picture doesn't change, no matter the rectangle size.

Now, if we're filling the frame with the same subject, DOF *does* change -- but DOF for the smaller frame is THICKER, not thinner as you said. It's thicker because with the smaller frame, we need a greater distance to capture the subject. Remember the basics of DOF:

* For thicker DOF use a smaller frame and/or tighter aperture and/or further subject distance

* For thinner DOF use a larger frame and/or wider aperture and/or closer subject distance

Circle of Confusion (CoC), look it up.
Becuse ASP-C is smaller you need to magnify it more and that's where the differnces comes from.

And you can make your rule simplere.
thicker DOF = less magnification and/or tighter aperture
thinner DOF = more magnification and/or wider aperture
But this not 100% accurate though so you shouldn't use it here in this technical talk.
Pupil magnification is for example missing and I think you should factor CoC in it as well

As Anvh points out, many people will use the crop factor in calculating DOF as they use composition as the constant. While one can argue the correctness of this technically, I believe that this is the piece of information that is most typically request when people speak of DOF. People will want to know what the effect is in what they see in the viewfinder, much like increasing the focal length if you stuck with the same format sensor.

Not only there but it also comes back when your view your photos on the same size screen or print for example, therefore you should factor it in because it does make a differnce in the photo.

[Anvh]

And still we can't agree -- some say you get more DOF with a smaller sensor (which is what I've heard generally, but I think they are assuming a changed position and an "equivalent" lens), others say the opposite. Can we get the story straight?

Now you're saying equivalent lens and that changes things.

If you keep the lenses and his settings the same then it makes the same image on the image plate so that part stays the same no matter how much you crop from it.
However if you factor in the differnce of the Circle of Confusion then smaller crops will have smaller DOF because of this.

If you change the lens to keep the angle of view the same between formats then because of the lesser magnification of the lens you end up with more DOF.
You need lesser magnification because if something fills the frame on a large format it needs to be smaller on the image plane for a smaller format to fill it as well.

[RioRico]

Circle of Confusion (CoC), look it up.
Becuse ASP-C is smaller you need to magnify it more and that's where the differnces comes from.

And you can make your rule simplere.
thicker DOF = less magnification and/or tighter aperture
thinner DOF = more magnification and/or wider aperture
But this not 100% accurate though so you shouldn't use it here in this technical talk.
Pupil magnification is for example missing and I think you should factor CoC in it as well

I am well aware of CoC. And I specifically said ALL ELSE BEING EQUAL -- enlarging images by different factors violates that.

Also, except on dedicated macro lenses, magnification is not a lens|camera setting. Your rules are not field-usable.