depth of field question

Messages
2,145
Name
James
Edit My Images
No
Ok, I have been thinking too much recently, and its showing because my brain is starting to overheat reading conflicting arguments on different websites.

I am 99.9% sure I know the answer to this, but I just want to check ;)

If you have a FF and a DX body, do they provide the same DOF at the same effective focal lengths?

For example.

FF camera with a 24-70 2.8
DX camera with a 16-55 2.8 (same effective focal length as a 24-70)

So, if you take a photo with the FF @ 70mm and the DX at 55 (effectively 70) both using the same aperture, would the depth of field be the same?

I really need to stop reading dpreview as I am sure that I am right, and then you read a load of tech nonsense that makes you question your own knowledge...

So, anyone care to pass their thoughts?
 
Last edited:
No, you'd need to be at f/2 on the DX lens to get the same DOF at equivalent focal lengths.
 
Last edited:
I think Nawty is right. Sensor and film format size do come into the maths. Smaller sensors give more DoF, thats why small cameras get away with having fixed lenses.
 
The reason sensor size matters is DoF is ONLY relevant when you print.

Briefly: a lens only has a single plane of focus. All items closer or further away will be out of focus - it just depends whether you can see it as out of focus. All DoF calculations assume a print of a certain size, viewed from a certain distance. The smaller the sensor, the more you have to magnify the resulting image to get it to that size, so the more you "see" into the captured image.

Also, if you print bigger than the "standard" size, or view it from a different distance, the DoF changes.

Have a further read here:

http://www.talkphotography.co.uk/forums/showthread.php?t=352736

and

http://www.talkphotography.co.uk/forums/showthread.php?t=372129

I'm sure Alan (woof woof) will be along to disagree with me soon, but he's wrong ;)
 
Ok, I have been thinking too much recently, and its showing because my brain is starting to overheat reading conflicting arguments on different websites.

I am 99.9% sure I know the answer to this, but I just want to check ;)

If you have a FF and a DX body, do they provide the same DOF at the same effective focal lengths?

For example.

FF camera with a 24-70 2.8
DX camera with a 16-55 2.8 (same effective focal length as a 24-70)

So, if you take a photo with the FF @ 70mm and the DX at 55 (effectively 70) both using the same aperture, would the depth of field be the same?

I really need to stop reading dpreview as I am sure that I am right, and then you read a load of tech nonsense that makes you question your own knowledge...

So, anyone care to pass their thoughts?

If you're stood in the same place, Yes. (But you'll have to step back to get the framing the same on DX so really it's a No)
 
Quick and easy way of working it out is multiplying the aperture by the crop factor.

So f/2.8 at 50mm on DX I's roughly equivalent to f/4.2 at 75mm on FX in DoF terms.


I'll now sit back and wait for woof woof to arrive, threads like this often get interesting :D ;)
 
Ed Sutton said:
Does 'interesting' mean the same as 'anal and boring'?

Depends on your point of view ;). I have an interest in the maths and physics of photography so I find it very interesting sometimes. If you're only interested in actually taking photos, well. . .

I also find it quite interesting that there is still disagreement on here and in general about how the facts of how DoF behaves effect the final photo, there seems to be a circle of confusion around the whole subject :D . . . Sorry, I'll get my coat.
 
Last edited:
If you're stood in the same place, Yes. (But you'll have to step back to get the framing the same on DX so really it's a No)
only if you are using the same lens...

Thanks for the info all, I am currently sat in the airport at the philippines so I have plenty of stuff to read to pass the time now :D
 
I'd love to get a clear idea of exactly why the dof is related to camera sensor and lens focal length aperture.

My understanding is that an equal sized aperture gives equal dof.

Eg
A 50mm f1.4 has an aperture 36mm. Gives the following dof on a nikon aps-c sized sensor:-

Online%20Depth%20of%20Field%20Calculator-2.jpg


A 70mm lens at f2 has an aperture of 35mm. Gives the following dof on a "full frame" sensor (almost identical to the above):-

Online%20Depth%20of%20Field%20Calculator-4.jpg


As the focal length is increased by a factor of x1.5 between sensors, widening the aperture by a stop (x1.4) more or less equals things. Or at least it appears to to me...
 
Last edited by a moderator:
I'd love to get a clear idea of exactly why the dof is related to camera sensor and lens focal length aperture.
The sensor size is simply there as a measure of how much you have to magnify the image formed at the sensor to get a printed picture. If you were using an 8x10 large format camera and made a contact print at 8x10 your magnification factor is 1. Printing from a DSLR (although the aspect ratio isn't the same) is somewhere around 8 (as a ff DSLR is around 1"x1.5"). The more you enlarge the image formed at the sensor, the less depth of field it has (as you are seeing more of the out of focus areas).
 
Hi andy, wouldnt that mean a camera with a tiny sensor would have less depth of field, as the image needs to be magnified a lot. But this doesn't happen in practice.

:thinking:
 
Hi andy, wouldnt that mean a camera with a tiny sensor would have less depth of field, as the image needs to be magnified a lot. But this doesn't happen in practice.

:thinking:
But you are forgetting the focal length and aperture that needs to be used with the smaller camera to get the same field of view. My 10x compact travel zoom has a 4.6-46mm lens on it and even then it is limited to F5 at the long end. And if you look at the pictures at any reasonable size, they will be OOF, just the quality isn't very nice (which is why I've just bought a u4/3 for travel). My FF DSLR is just starting to get going with DoF at 46mm... And most peoples thoughts of shallow DoF is portrait style shots which are typically shot with fast (F1.x/2.x) lenses and at 75+mm in length.
 
Hi Andy, i was aware if the importance of focal length and aperture - take a look at my post at 1117!

What i am trying to (and failing to ) understand, is all the principles at play. I (think) i understand how the aperture / focal length / subject distance work together..But still need to clarify the effect of sensor size. Going to dig out Langfords advanced and read up, but maybe this evening, too nice a day hear to be reading up on optical theory!!
 
Last edited by a moderator:
James, to answer your original qustion from an expert (ie not me..):-

"The effective depth of field (DOF) on a DSLR is greater than on a 35 mm camera when a lens of equivalent focal length is used at the same aperture. For example, if the 50 mm lens on the Nikon was set at f/5.6 then the DOF on the Nikon would be greater than the DOF on the 35 mm camera when a 75 mm lens at f/5.6 was used even though the field of view was equivalent."

From Langfords advanced photography, p64
 
But still need to clarify the effect of sensor size.
Other than contributing to your choice of lens length (which does have a significant effect on DoF ;)), sensor size contributes only to the size you print out at. It is in the formula as CoC (Circle of Confusion) and if you look http://en.wikipedia.org/wiki/Circle_of_confusion and in particular the section titled: "Circle of confusion diameter limit in photography" you'll see that it assumes a certain print size and viewing from a certain distance. In fact, CoC is defined as:

CoC (mm) = viewing distance (cm) / desired final-image resolution (lp/mm) for a 25 cm viewing distance / enlargement / 25

If those assumptions are changed (for example pixel peeping at 100% on a 24" screen from 1m) you can plug the figures into that formula and see how the CoC changes. The enlargement is the direct link to sensor size in the formula above.
 
<snip>
If you have a FF and a DX body, do they provide the same DOF at the same effective focal lengths?

<snip>

Where have you been all these years James?! The answer is no. Listen to Andy. In the hope that this might help... :D

DoF is all about magnification (including the f/number, but let's not go there) and can only be properly measured and compared when all the parameters are adhered to.

The starting point, and this is the internationally agreed standard, is a print 10in wide viewed at a distance equal to the diagonal, ie 12-ish inches. That is the basis for all calculations, though it also works precisely pro-rata for different size prints because everything stays the same when you adjust the viewing distance accordingly.

When you look at that 10in print from 12in, the standard says the human eye cannot detect anything smaller than 0.2mm wide, and everything is worked back from that assumption.

That is the final circle of confusion size, so since a full frame image has to be enlarged about 7x to make a 10in print, 0.2/7=0.03mm. That's the circle of confusion size that goes into the formula, and that's what changes in calculators like www.dofmaster.com when you change the camera - all it does is change the CoC according to the sensor size.

As mentioned previously, an easy way of getting equivalence of DoF between formats is to use the crop factor, say 1.5x for Nikon/Sony. Therefore f/4 on Nikon DX would be 4 x 1.5 = f/6 on full frame FX. About 1.2 stops difference. Canon 1.6x almost identical. Micro 4/3rd has 2x crop factor, and delivers exactly two stops more DoF at the same aperture.

Compacts with their very small sensors usually have a crop factor of around 5x, which explains why it's often quite hard to avoid getting everything sharp - you can only get shallow DoF effects when shooting very close, or use one of the few models with a very low f/number, like f/2-ish.

BTW, the only relevant way to compare DoF is when you shoot the same subject from the same distance (to maintain perspective), and adjust focal length (by the crop factor) to maintain equivalent framing. Images will then be identical in every respect when the aperture is asdjusted. If you don't do that, you're not comparing the same photo and the numbers go all over the place. Woof :D
 
Last edited:
:lol: No discussion on DoF is complete without an alternative view to argue the relative insignificance of CoF :D

These days many people own different format cameras and it's almost laughably easy to prove to yourself what makes a difference and what looks the same to you. The difference that differences in the CoC between slightly different cameras makes is pretty insignificant to the vast majority of people and the vast majority or screen or printed images unless you take thing to ridiculous extremes of reduction or increase in image size, IMVHO. The things that make the real world differences to DoF are aperture, camera to subject distance and focal length, IMVHO.

Shots at 70 and 55mm from different formats will look different even if taken from the same distance at the same aperture. To get the DoF to look the same you'd have to change the camera to subject distance because you're using a wider angle lens on one of the cameras and that makes the DoF look deeper but even then the DoF will look different because of the different rendering of the foreground / background that you get when using lenses at different focal lengths. Gotta remember that when shooting at 55mm on APS-C what you see when you view the results is a crop cut from the middle of a 55mm image. You'd have to magnify individual image elements to equalise the pictures so it's a non starter. If you used the 70mm on two different formats at the same aperture and camera to subject distance you'd get different FoV but the DoF would be essentially the same, if you looked closely at the section of the image that both images contain.

So to get the look of a 70mm lens at f2 you have to use a 70mm lens at f2. A 55mm at f2 will never give you exactly the same look even if you decrease the camera to subject distance or use a wider aperture as the rendering and the spacial relationships and size within the image will be slightly or even wildly different.

I could of course post images that prove all I've said here :D

However, the best thing is to try different lenses on different formats at different apertures and camera to subject distances and view the results with an open mind.
 
:lol: No discussion on DoF is complete without an alternative view to argue the relative insignificance of CoF :D
You know, the funny thing is that you are agreeing with both Richard and I, it's just that you can't see it from our way of explaining it.....
 
You know, the funny thing is that you are agreeing with both Richard and I, it's just that you can't see it from our way of explaining it.....

I can see your way of looking at it but it's not a real world view for me and probably not for most other people.

What matters to most people isn't the difference in CoC between camera A and camera B, that's insignificant most of the time (IMVHO.) What matters is aperture, focal length and camera to subject distance. Even when you change formats those basic things remain the main factors.

That's such a biggie it's worth saying again... What matters is aperture, focal length and camera to subject distance. Even when you change formats those basic things remain the main factors.

"Compacts with their very small sensors usually have a crop factor of around 5x, which explains why it's often quite hard to avoid getting everything sharp - you can only get shallow DoF effects when shooting very close, or use one of the few models with a very low f/number, like f/2-ish."

The reason a smaller format gives greater DoF is it's got a wider lens and a smaller aperture. Stick a smaller format cameras 12mm f4 lens on a 5D and you'd get a lot of DoF. Surprised? You shouldn't be :lol: The only way of getting shallow DoF would be to focus on something very close to the camera. Sound familiar? :lol:

Stick a 50mm f1.4 on a 5D and you'd get shallow DoF, fit the same lens to a smaller format camera you'd get shallow DoF. Surprised? You shouldn't be. Try it. :D

When people go on about CoC and DoF tables to three decimal places and the difficulty of getting DoF from smaller formats I begin to think that they haven't stopped looking at the screen long enough to take cameras with different formats out and see what works for them.

Anyway, as you know, I only posted as I'd been baited to... :D
 
I can see your way of looking at it but it's not a real world view for me and probably not for most other people.
Your "mistake", if I can call it that, is to think in terms of the same lens on different format cameras.

If you do put the same lens on different format cameras and print the images to the same size (without cropping) you will get different DoF. The reason you get different DoF from different cameras is solely due to the sensor size. It is very predictable and the smaller format camera will give less depth of field as you are taking less of the image area and magnifying it more to get to a defined image size. Same aperture, same focal length, same lens. The difference is the sensor and the printing.

I'd also say the other "mistake" I've seen you make is to compare out of focus areas. That isn't the definition of DoF. DoF is defined as the point at which the image looks out of focus. That means to compare DoF, you need to look at the area of image that is at the point of going out of focus. The examples you've posted are about how the out of focus areas are rendered, not showing the areas that are critically sharp. Talking about different sensors for the same picture means putting different length lenses on and the difference between short and long focal lengths is how quickly they blur the background. A short focal length lens on a small sensor will blur the background less quickly than a longer focal length lens on a larger sensor. However, this is about the quality of the out of focus areas, not whether it is in or out of focus which is what DoF is about.

What I don't fully understand is that you seem to totally dismiss what the maths says. All both Richard and I are doing is reiterating what the maths says. If you disagree, and you seem to be doing so, you are rewriting the fundamental understanding (from a variety of respected sources) on how the maths DoF works. That's all well and good, but you need to provide something a little more mathematical than "I've tried it and it doesn't look right to me"...
 
Alan, I agree with you when you say format size doesn't, for most of us, make a huge difference. Switching between full frame and DX crop is only a smidge above one stop shift in terms of DoF, hardly more than, for example, the difference between f/4 and f/5.6. In practice, a DoF change of half a stop is barely noticeable.

When I want to make a noticeable difference to DoF, I shift things one stop or two stops at least. These things are always expressed in small fractions, because that's how the maths works and you have to draw the line somewhere, but in actual picture taking it's rarely anything like that critical and DoF fades gradually, sometimes imperceptively.

But that's as far as agreement goes. When you compare any DSLR to a compact, the change in DoF is huge.

Yes, to make a proper comparison between formats you have to adjust the focal length to get the same picture (merely changing distance alters perspective, so invalid) but that's a chicken and egg argument - why does focal length have to change? Obviously because the sensor size has changed, so that is the underlying driver - the sensor is fixed and we change focal length to suit the format, not the other way around.

Talking about putting a 12mm lens on a full frame 5D and then on a compact is completely irrelevant, because the resulting image is utterly different. On FF 12mm is a mega-super-wide with an angle of view over 120 degrees, whereas it's a modest telephoto on a compact. As I said above, the only way to make a meaningful comparison is to adjust focal length and aperture to suit the sensor size, and then you have an identical image - same framing, from the same distance, same perspective, same DoF, same everything. When you've done that, the camera with the smaller sensor delivers greater depth of field at a given f/number.
 
Andy, not sure where you're going with your third paragraph. Are you implying that even if you change formats and adjust focal length accordingly, the out of focus areas look different?

That is not true. If you maintain equivalence by adjusting focal length to maintain the same framing, then shoot from the same distance to maintain the same persepctive, and adjust the aperture to maintain the same DoF, then all aspects of the image will be identical - the in focus areas, the out of focus areas, the hyperfocal distance, everything.

For example, from the same distance, Canon 7D with 50mm lens at f/4, exactly the same as 5D with 80mm lens at f/6.4.

Sorry if I've misunderstood you.
 
Andy, not sure where you're going with your third paragraph. Are you implying that even if you change formats and adjust focal length accordingly, the out of focus areas look different?
No.

What I'm saying is that the pictures I've seen Alan post are the out of focus areas. This has nothing to do with the DoF (other than to say they are out of focus). DoF is when the image goes from sharp to unsharp and putting up pictures of out of focus areas - no matter how in or out of focus they are - proves nothing.
 
Wow, thanks for all the information and its certainly a lot clearer now (even if some of it is rather hard to understand for a simpleton like me).

But, it certainly goes to show why I am getting some rather inconsistent results and not seeing things that I expected to see.

Thanks for the help everyone :D
 
Back
Top