|
| Author | Thread |
|
|
07/06/2008 09:33:10 PM · #1 |
I'm hoping to get this answer from someone wiser than I. It's been my understanding that a "full frame" DLSR like the Canon 5D would yield thinner DOF that my APS-C sensor Rebel XT.
So, I was visiting the DOFmaster website (here) and using their calculator. When I input the same focal length and aperture, but change the camera from the Rebel XT to the 5D it indicates that the 5D actually creates a deeper depth of field.
What gives? |
|
|
|
07/06/2008 09:38:49 PM · #2 |
DOF is independent of the camera body in many ways. It is largely lens-dependent. A 100mm lens at f/5.6 has the same DOF, basically, whether it is mounted on APS-C or FF sensor. HOWEVER>>>>
DOF is a matter of perception, so a finer-grained image will have the appearance of more DOF than a coarser one, even if shot with same lens and exposure parameters. The 5D, accordingly, delivers a tad "more" DOF because it is creating a smoother image than the rebel.
R.
|
|
|
|
07/06/2008 09:49:50 PM · #3 |
Originally posted by Bear_Music:
DOF is independent of the camera body in many ways. It is largely lens-dependent. A 100mm lens at f/5.6 has the same DOF, basically, whether it is mounted on APS-C or FF sensor. HOWEVER>>>>
DOF is a matter of perception, so a finer-grained image will have the appearance of more DOF than a coarser one, even if shot with same lens and exposure parameters. The 5D, accordingly, delivers a tad "more" DOF because it is creating a smoother image than the rebel.
R. |
Then why do I hear that the full frame sensor will give a shallower DoF? Or is it that it is a crisper DoF and then fades OoF faster?
Edited to quote Robert the Great.
Message edited by author 2008-07-06 21:51:04. |
|
|
|
07/06/2008 09:54:59 PM · #4 |
| Just a WILD guess here but DoF decreases as you get closer to your subject. You need to get closer to your subject with a full-frame sensor. Therefore for the same shot the full-frame camera will have less DoF. |
|
|
|
07/06/2008 10:05:51 PM · #5 |
There's mass confusion over this topic. The DOFMaster calculator, excellent as it is, doesn't help resolve any of it either. Let's do this by example:
If you take a lens and shoot a subject at f/5.6 with a 50mm lens on a 40D, then switch to a 5D and shoot the same subject with the same lens, still at f/5.6, and still at the same distance, the DoF will be the same, if you use the same assumption for a parameter aptly named the "circle of Confusion" or CoC for short.
The problem arises since the CoC for images from the two cameras is not the same, if the images are displayed at the same size. The 40D shot is framed much tighter than the 5D shot (remember, same distance). The DoF for the 40D shot will actually appear shallower, in prints of the same size.
Now another exercise. Shoot the same subject with the 40D, again at f/5.6, again with that 50mm lens. Now switch to the 5D, but shoot the subject at a distance that produces the same framing as the 40D. Now print both files at the same size, and the 5D will have less DoF.
In general, the larger the sensor, the shallower the DoF can be, given that longer focal lengths or shorter distances are required for the same framing.
Back to the topic of CoC... normally, CoC is calculated based on a specific print size and viewing distance. There is another, print-independent way to look at CoC though. It's based on the smallest detail that a camera/lens system can resolve. If we assume that the lens out-resolves the sensor, which is true for many situations, then the limit of resolution is defined by the sensor and the anti-alias filter. the smallest resolvable detail is given by the "Nyquist limit" which is two times the pixel pitch. Most cameras closely approach this limit in resolving luminosity detail. So one can set the CoC manually in DoFMaster to twice the pixel pitch for a more accurate indication of what will appear sharp, sharp being defined as "sharp at 100%."
|
|
|
|
07/06/2008 10:09:05 PM · #6 |
| Oh god, my eyes are bleeding, this isn't normal... |
|
|
|
07/07/2008 01:12:09 AM · #7 |
Originally posted by kirbic:
Now another exercise. Shoot the same subject with the 40D, again at f/5.6, again with that 50mm lens. Now switch to the 5D, but shoot the subject at a distance that produces the same framing as the 40D. Now print both files at the same size, and the 5D will have less DoF.
In general, the larger the sensor, the shallower the DoF can be, given that longer focal lengths or shorter distances are required for the same framing.
|
This is the part I forgot about, equivalent framing. Thank you for yet another free photo nugget o' knowlwdge Mr Byle. :) |
|
|
|
07/07/2008 01:05:41 PM · #8 |
Now another exercise. Shoot the same subject with the 40D, again at f/5.6, again with that 50mm lens. Now switch to the 5D, but shoot the subject at a distance that produces the same framing as the 40D. Now print both files at the same size, and the 5D will have less DoF.
In general, the larger the sensor, the shallower the DoF can be, given that longer focal lengths or shorter distances are required for the same framing.
Ok -this makes sense. Thank You!
Back to the topic of CoC... normally, CoC is calculated based on a specific print size and viewing distance. There is another, print-independent way to look at CoC though. It's based on the smallest detail that a camera/lens system can resolve. If we assume that the lens out-resolves the sensor, which is true for many situations, then the limit of resolution is defined by the sensor and the anti-alias filter. the smallest resolvable detail is given by the "Nyquist limit" which is two times the pixel pitch. Most cameras closely approach this limit in resolving luminosity detail. So one can set the CoC manually in DoFMaster to twice the pixel pitch for a more accurate indication of what will appear sharp, sharp being defined as "sharp at 100%." [/quote]
This is a bit beyond my depth - no need to explain further, though... I'm happy with the first explanation :)
|
|
|
|
07/07/2008 01:12:34 PM · #9 |
Originally posted by kirbic:
Now another exercise. Shoot the same subject with the 40D, again at f/5.6, again with that 50mm lens. Now switch to the 5D, but shoot the subject at a distance that produces the same framing as the 40D. Now print both files at the same size, and the 5D will have less DoF.
In general, the larger the sensor, the shallower the DoF can be, given that longer focal lengths or shorter distances are required for the same framing.
|
Except that DOF is independent of focal length. It is a function of the physical size of the aperture. What confuses people is that a 100mm lens at f/4 has less DOF than a 50mm lens at f/4; but if you stopped the 100mm down to f/8 then both lenses would have a 12.5mm aperture and DOF (at the same reproduction ratio) would be the same.
IN other words, shooting macro at 1:1, a 50mm macro lens doesn't have inherently "more" DOF than a 100mm macro lens, assuming identical physical apertures. I found this out much to my surprise, actually :-)
R.
Message edited by author 2008-07-07 13:13:01.
|
|
|
|
07/07/2008 04:22:51 PM · #10 |
| Robert, your example is true for Macro, but the relationship does not hold at low magnification. At "normal" focusing distances, there *is* a focal length effect. |
|
|
|
07/07/2008 04:30:11 PM · #11 |
Originally posted by kirbic:
Robert, your example is true for Macro, but the relationship does not hold at low magnification. At "normal" focusing distances, there *is* a focal length effect. |
I don't believe that's true. Hypothetically, if you frame a scene with the 100mm lens at f/8, say and then move in closer and frame the same thing with the 50mm at f/4 (both physical apertures 12.5mm) then the DOF is the same. If you SHOOT the 50mm from the same spot as the 100mm then crop to framing, THEN you'll see a difference...
I could be wrong, but that's how I've understood it. As long as the actual subject magnification is the same, DOF is the same between identical physical apertures regardless of focal length, in summary. You've agreed this is correct for 1:1, right? So tell me why it would change at other magnifications? I'm sincere in this request, I'd love to understand if I'm wrong...
R.
|
|
|
|
07/07/2008 04:50:24 PM · #12 |
Robert, go here, select whatever camera you like, and try the exercise. I used the setting for Canon 5D:
100mm, 10 ft, f/8 = 1.42 ft DoF (46% in front, 64% behind)
50mm, 5 ft (same framing), f/4 = 0.71 ft DoF (46% in front, 64% behind)
Now, if the *f ratio* is held constant:
100mm, 10 ft, f/4 = 0.71 ft DoF, same as 50mm.
Edited to actually include the link I meant to :-P
Message edited by author 2008-07-07 18:35:08. |
|
|
|
07/07/2008 06:30:33 PM · #13 |
Originally posted by kirbic:
Robert, go here, select whatever camera you like, and try the exercise. I used the setting for Canon 5D:
100mm, 10 ft, f/8 = 1.42 ft DoF (46% in front, 64% behind)
50mm, 5 ft (same framing), f/4 = 0.71 ft DoF (46% in front, 64% behind)
Now, if the *f ratio* is held constant:
100mm, 10 ft, f/4 = 0.71 ft DoF, same as 50mm. |
Very cool info! I have had a "practical" understanding of DOF for some time but never found this example to verbalize it. Thanks Scott |
|
|
|
07/07/2008 06:53:09 PM · #14 |
Here's an explanation i made for a friend a couple of weeks ago regarding bokeh, for Full Frame vs 1.5crop sensors.
Bokeh - D3 vs D300]bokeh
Message edited by author 2008-07-08 12:01:42. |
|
|
|
07/07/2008 07:03:04 PM · #15 |
Originally posted by kirbic:
Robert, go here, select whatever camera you like, and try the exercise. I used the setting for Canon 5D:
100mm, 10 ft, f/8 = 1.42 ft DoF (46% in front, 64% behind)
50mm, 5 ft (same framing), f/4 = 0.71 ft DoF (46% in front, 64% behind)
Now, if the *f ratio* is held constant:
100mm, 10 ft, f/4 = 0.71 ft DoF, same as 50mm.
Edited to actually include the link I meant to :-P |
I think that the confusion really stems from the fact that people only experience the effects of "effective bokeh" and not "physcial bokeh" (if anyone has better suggestions for these terms, just let me know).
Basically, when working with 2.8 optics @ 2.8, if you want a blurrier background, you'll move away from the subject and zoom in. The framing of the subject will be the same, but the background will essentially be cropped.
Now, the actual "physical bokeh" will remain the same. For example, a leaf on a tree behind the subject will still be blurred by, say, 4cm. The difference now is that the leaf is bigger because you zoomed in, so 4cm appears like more bokeh. This is what a mean by effective bokeh: the amount of blur measured in pixels.
Just to reiterate. Going wider and getting closer to the subject makes the background tree so small, that the 4cm blur becomes negligible. The tree then looks sharp.
I find DOF is a misleading measure because of all this. |
|
|
|
Current Server Time: 08/18/2025 06:27:53 PM  |
Home -
Challenges -
Community -
League -
Photos -
Cameras -
Lenses -
Learn -
Help -
Terms of Use -
Privacy -
Top ^
DPChallenge, and website content and design, Copyright © 2001-2025 Challenging Technologies, LLC.
All digital photo copyrights belong to the photographers and may not be used without permission.
Current Server Time: 08/18/2025 06:27:53 PM EDT.
|
