5Diii vs 7Dii (FF vs APS-C)

[Loswr]

Both aperture AND relation of the subject to camera will determine DOF.

...AND focal length. Also, for the sake of completeness and hopefully not to confuse the issue, the fourth factor is the circle of confusion (CoC, which is related to sensor size, and somewhat arbitrarily defined for various sensor formats based on a specific print size and viewing distance).

So...if you hold the other listed factors constant, you get shallower DoF with:

• Wider aperture
• Closer focus (subject) distance
• Longer focal length
• Smaller circle of confusion

The last one confuses many people (pun intended), because CoC decreases with decreasing sensor size, i.e. DoF is actually shallower with a smaller sensor. But remember - the above holds true when you change one factor only. So, if you keep everything but the body constant - same focal length, aperture, and distance, the 7D will give a shallower DoF than the 5DII (try it - plug any numbers you like into DoFMaster, and change only the camera popup menu). Of course, your framing will be quite different. It's not just esoteric, though - it has real-world implications when doing an AFMA with a commercial tool (e.g. LensAlign), where you shoot at the same distance relative to focal length, independent of sensor format. In that setting, you can easily see the shallower DoF with the 7D (compared to the 5DII), and while the standard ruler is fine for 400mm f/5.6 on APS-C, the same lens on FF is marginal, and easier to calibrate with the longer ruler.

 

[mreco99]

that reminds me (T n L) comment on dust... ive never cleaned my sensor, rarely cleans my lenses tbh, since purchase in 2008. Will i still go to heaven?

 

[Meh]

Basically, aperture determines DoF.

Correct as usual but I'll add, just in case a few people don't know, it is the actual physical size (the diameter) of the aperture that affects DoF not the f-number (although the f-ratio typically falls out of optical formulae when you end up with an f/d). If it wasn't for diffraction, a tiny aperture would render everything in focus with no need for a lens which is how a pinhole camera works. Yes, in the DoF calculators you must plug in the f-ratio but that is just used (along with focal length) to calculate the diameter (in mm) of the aperture. My turn to wait for a correction from neuro

Question to neuro (because I just tried but quickly gave up playing with the formulas): changes in subject distance and focal have an opposite effect on DoF but is it equally offsetting for the same framing/subject size? There is an article on Luminous Landscape that demonstrates that if camera position changed to keep subject size the same within the frame, DoF remained visually equal regardless of focal length. This makes sense when one understands that the physical size of the aperture is smaller for shorter focal length lenses for the same f-number. But is it mathematically equal at least first order?

 

[Loswr]

@Meh - Exactly, it's the iris diaphragm that matters. Practically, since you have to actually do simple math (eek!) to determine that from the focal length and f/number, it's more practical to use those values to determine DoF.

LL's experiment with the gremlin is approximately correct. In fact, at the distances involved it's actually not the best approximation. I don't agree that the degree of unsharpness is identical - but it's close, good enough for approximation. I've tried an equivalent experiment, as a post hoc test, with my AFMA testing. Since I use 25x the focal length for all lenses, and the LensAlign fills the same proportion of the frame (i.e. distance and focal length are equal and opposite). With an actual distance scale on the ruler, it's apparent that f/2.8 at 16mm (16-35/2.8L II) and at 200mm f/2.8 (70-200mm f/2.8L IS II), the measured DoF is the same, when distance and focal length are reciprocal.

Actually, the reduction of DoF to dependence only on magnification holds at macro distances (and for microscopy, although we usually call it axial resolution not DoF, and use numerical apertures vs. f/numbers).

 

[mreco99]

Whoosh....

 

[Hillsilly]

So, if you keep everything but the body constant - same focal length, aperture, and distance, the 7D will give a shallower DoF than the 5DII (try it - plug any numbers you like into DoFMaster, and change only the camera popup menu). Of course, your framing will be quite different.

I thought I was noticing shallower DOF and more background blur with an APS-C sensor (compared with film) and just assumed it was my imagination. Instead, it is probably because I'm tending to using my lenses from a similar distance and changing the framing.

Therefore, if I make the switch to FF, I'm not going to get "more" background blur with the same lens at the same aperture unless I get in closer. I'll mark this in favour of a 7D - I like my space.

But I also read the macro discussion earlier. I actually thought I was getting more magnification. But I'm not. I'm just getting a cropped image. So I'll mark this in favour of a 5D.

As far as image quality goes, at higher ISOs, it is generally agreed that the 5D is the clear winner.

For people with both cameras, comments seem to be equally divided about which camera they prefer if they're just going out to shoot general photos. While there's a slight leaning towards the 5D, the 7D fans seem more vocally devoted. From what I understand, the 7D is a better camera to use than a 5Dii, which is probably why its users are more loyal. It will be interesting to see if the feature set of the 5Diii changes this.

Clearly, there's no definitive winner and probably never will be. But leaning towards the 5Diii (if Canon ever release such a thing) as my next camera. Thanks for your comments so far.

 

[mreco99]

ok, as i was talking about my images earlier, ive included my store link in my signature, does it work?


thanks

 

[K-amps]

@Meh - Exactly, it's the iris diaphragm that matters. Practically, since you have to actually do simple math (eek!) to determine that from the focal length and f/number, it's more practical to use those values to determine DoF.

LL's experiment with the gremlin is approximately correct. In fact, at the distances involved it's actually not the best approximation. I don't agree that the degree of unsharpness is identical - but it's close, good enough for approximation. I've tried an equivalent experiment, as a post hoc test, with my AFMA testing. Since I use 25x the focal length for all lenses, and the LensAlign fills the same proportion of the frame (i.e. distance and focal length are equal and opposite). With an actual distance scale on the ruler, it's apparent that f/2.8 at 16mm (16-35/2.8L II) and at 200mm f/2.8 (70-200mm f/2.8L IS II), the measured DoF is the same, when distance and focal length are reciprocal.

Actually, the reduction of DoF to dependence only on magnification holds at macro distances (and for microscopy, although we usually call it axial resolution not DoF, and use numerical apertures vs. f/numbers).

I thought I had this figured out till I read HillSilly's post...

Neuro or any other Savior : Please dumb this down for me... (you have done this many times ) : Bokeh notwithstanding.... do I get more or less OOF blur (Quantity) is APC-S or with FF.

Ok so let say i use a 50mm f1.8 on both a 5d and 7d, subject is 10 feet away from both cameras. After capture and 100% crop (lets say we frame 1 feet on all sides of the head of the subject (to get some background) in the crop )... basically like this frame for a passport picture

With image will show more blurring of the background?

 

[Meh]

@Meh - Exactly, it's the iris diaphragm that matters. Practically, since you have to actually do simple math (eek!) to determine that from the focal length and f/number, it's more practical to use those values to determine DoF.

LL's experiment with the gremlin is approximately correct. In fact, at the distances involved it's actually not the best approximation. I don't agree that the degree of unsharpness is identical - but it's close, good enough for approximation. I've tried an equivalent experiment, as a post hoc test, with my AFMA testing. Since I use 25x the focal length for all lenses, and the LensAlign fills the same proportion of the frame (i.e. distance and focal length are equal and opposite). With an actual distance scale on the ruler, it's apparent that f/2.8 at 16mm (16-35/2.8L II) and at 200mm f/2.8 (70-200mm f/2.8L IS II), the measured DoF is the same, when distance and focal length are reciprocal.

Actually, the reduction of DoF to dependence only on magnification holds at macro distances (and for microscopy, although we usually call it axial resolution not DoF, and use numerical apertures vs. f/numbers).

Neuro: Please dumb this down for me... (you have done this many times ) : Bokeh notwithstanding.... do i get more or less OOF blur (Quantity) is APC-S or with FF. Give me some numbers in feet etc if you can... cause "framing and similar terms are relative :-\

I'm no neuro but I can take a shot to rephrase it into less technical terms.... we'll see if I have this right!

If you use the same lens, set to the same aperture, and place the camera the same distance from the subject, you will get more OOF blur with an APS-C sensor compared to a FF sensor.

If you use the same lens, set to the same aperture, and place the APS-C camera 1.6X further away from the subject than the FF camera (to get the same framing) you will get more OOF blur with the FF camera.

If you use the same lens, set to the same aperture, and place the APS-C camera 1.26X further away from the subject than the FF camera you will get the same OOF blur. (1.26 is the square root of the ratio of CoC of FF and APS-C)

If you use different lenses (two different focal lengths) on the same camera but set to the same aperture, and place the camera at the same distance from the subject, the shorter focal length will give less OOF blur.

If you use different lenses (two different focal lengths) on the same camera but set to the same aperture, and place each camera at 25X (or any multiple) of the focal length, both lenses will give the same OOF blur. (This is the gremlin example on LL that keeps the subject the same size)

 

[Loswr]

Neuro: Please dumb this down for me... (you have done this many times ) : Bokeh notwithstanding.... do i get more or less OOF blur (Quantity) is APC-S or with FF. Give me some numbers in feet etc if you can... cause "framing and similar terms are relative :-\

Practically speaking, shallower DoF with FF. Meh nailed it. But to give an example:

Head/shoulders portrait shot with an EF 135m f/2L where I want a strong background blur. With the 5DII, I'd stand about 8 feet from the subject, and since faces aren't completely flat, I stop down to f/3.2 to get a 2.3" thick DoF. If I put the same lens on a 7D, to take that same shot of the subject's head and shoulders, I need to stand 13 feet away. Now, with the 135L lens set to f/3.2 my DoF is close to 4" thick - deeper because of the greater distance, so I don't get as much OOF blur. I'd need to set the aperture to f/2 to get the 2.3" DoF.

Now, if I just switched to the 7D, and stayed 8 feet from the subject at f/3.2, I'd actually get a thinner DoF than with FF - a little under 1.5" thick, and thus more OOF blur. But then the image would cut off the subjects shoulders, hairline and possibly ears - not a very flattering portrait. So, in a contrived situation (like that example, or like AFMA testing), the APS-C has shallower DoF. But in a real-world application, DoF is shallower with FF (because you have to move further from the subject to compensate for the narrower angle of view with a crop sensor).

Hope that makes sense.

 

[K-amps]

Practically speaking, shallower DoF with FF. Meh nailed it. But to give an example: ...Hope that makes sense.

Sure does, thanks,

 

[K-amps]

I'm no neuro but I can take a shot to rephrase it into less technical terms.... we'll see if I have this right!
........

Perfect response Meh .... really dumbed down for me!

You get a gold star!

 

[Meh]

But in a real-world application, DoF is shallower with FF (because you have to move further from the subject to compensate for the narrower angle of view with a crop sensor).

I like the way you phrased "real-world application"... unless something is preventing the photographer from moving, he/she will move forward or back to compose the shot and to get the desired framing one would naturally (without thinking about the technical aspects) move a little further away with a crop sensor camera according to what is seen in the viewfinder. And because we don't walk around with DoF calculators we might tend to choose the same aperture setting and when we look at the images we do in fact observe that the FF gave more background blur.

 

[Meh]

I'm no neuro but I can take a shot to rephrase it into less technical terms.... we'll see if I have this right!
........

Perfect response Meh .... really dumbed down for me!

You get a gold star!

Thanks!

 

[awinphoto]

ok, as i was talking about my images earlier, ive included my store link in my signature, does it work?


thanks

Store link works... nice photos...

 

[mreco99]

ok, as i was talking about my images earlier, ive included my store link in my signature, does it work?
thanks

Store link works... nice photos...

Thanks, appreciate the feedback
also, Neuro, thanks alot, thats really helped, and after 6 months of obsessive reading (i would have anyway) im going to pull the trigger on the 5dm2 and 2 or 3 lenses, (once ive nailed down the lenses lol)