# Depth of field for various formats and lenses

Started May 30, 2012 | Discussions
 Like?
 Depth of field for various formats and lenses May 30, 2012

It is common to discuss DOF for various formats and lenses. There are two statements commonly made:

1) Small format means more DOF
2) Longer lens means shallower DOF

I wanted to verify this by doing actual DOF calculations. I used http://www.dofmaster.com/dofjs.html to do these calculations.

Methodology:

I set the standard to compare with as using a full frame camera with an 85mm f/1.4 lens, standing 2.5 metres from the subject - an acceptable distance for head and shoulder portrait I guess. Then, for a given format and lens spec, I calculated the 35mm equivalent focal length and adjusted the subject distance. For e.g., if I am using a 50mm f/1.4 lens on Nikon D7000, the subject distance would be 75/85 * 2.5m = 2.2m.

The overall assumption is that the framing of the subject is the single most important factor and has to remain constant.

Following are the results for some common format/portrait lens combos.

(focal length, aperture setting, subject distance - DOF in metres).

Full frame:
85mm, f/1.2, 2.5m - 0.06m
85mm, f/1.4, 2.5m - 0.07m
50mm, f/1.4, 1.47m - 0.07m
50mm, f/1.0, 1.47m - 0.05m
100mm, f/2, 2.94m - 0.1m

APS-C (Nikon size):
50mm f/1.4, 2.2m - 0.11m
60mm f/2.2, 2.65m - 0.17m
85mm f/1.4, 3.75m - 0.11m
100m f/2, 4.4m - 0.15m

MFT:
45mm, f/1.8, 2.65m - 0.18m
75mm, f/1.8, 4.4m - 0.18m
25mm, f/1.4, 1.47m - 0.14m

(I chose 60mm f/2.2 with APS-C because that is the combo I have right now using the Tamron 60mm f/2 SP Macro lens with Nikon D40)

What I found:

1) Small format means more DOF - TRUE. Mainly because for the same lens focal length, subject distance increases, increasing the DOF.

2) Longer lens means shallower DOF - FALSE. As the focal length increases, so does the subject distance and the eventual DOF remains constant.

I hope I've not made a mistake above. What do you think?

amangupta's gear list:amangupta's gear list
Nikon D40
Nikon D40 Nikon D7000
If you believe there are incorrect tags, please send us this post using our feedback form.
Complain
 Like?
 Re: Depth of field for various formats and lenses In reply to amangupta, May 30, 2012

Yes you have made a mistake. For a valid comparison, for a given format, you need to use the same aperture on all lenses. For example on any format, a 100mm lens at f/1.0 (if you could find one) will have less DOF than the 50mm at f/1.0. More realistically the 100mm lens at f/2.0 will have less DOF than a 50mm lens at f/2.0.

By also changing the subject distance you are introducing a further error, but the aperture issue is the main problem.

Complain
 Like?
 Re: Depth of field for various formats and lenses In reply to amangupta, May 30, 2012

amangupta wrote:
[snip]

2) Longer lens means shallower DOF - FALSE. As the focal length increases, so does the subject distance and the eventual DOF remains constant.

DoF Master does indeed confirm your statement provided you keep the field of view constant by moving backwards.

However, if you are standing in a fixed spot and zoom in on something, the depth of field does decrease.
--
Chris R

Chris R-UK's gear list:Chris R-UK's gear list
Canon EOS 7D Panasonic Lumix DMC-GF1 Panasonic Lumix DMC-GH2 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Canon EF 50mm f/1.8 II +6 more
Complain
 Like?
 Re: Depth of field for various formats and lenses In reply to ajscullard, May 30, 2012

Not sure what you meant about using same apertures - in the list of lenses for a format, I don't say they have the same DOF, I explicitly list the DOF value. And, as you can see, when the apertures are same, DOF values are also same.

amangupta's gear list:amangupta's gear list
Nikon D40
Complain
 Like?
 Both of your conclusions are correct In reply to amangupta, May 30, 2012

amangupta wrote:

1) Small format means more DOF - TRUE. Mainly because for the same lens focal length, subject distance increases, increasing the DOF.

Or, equivalently, from the same distance focal length has to get shorter for the smaller format, increasing the DOF.

2) Longer lens means shallower DOF - FALSE. As the focal length increases, so does the subject distance and the eventual DOF remains constant.

That's also true, given the same framing (which was your premise).

-- hide signature --

Lee Jay
(see profile for equipment)

Lee Jay's gear list:Lee Jay's gear list
Canon ELPH 500 HS Canon PowerShot SX260 HS Canon EOS 5D Canon EOS 20D Canon EOS 550D +23 more
Complain
 Like?
 It all depends on what you hold constant. In reply to amangupta, May 30, 2012

First, we usually assume constant subject framing. For most cases this is the logical/fair thing to do. But there's an important exception—DoF scales, where distance is the constant, not framing—smaller formats have LESS DoF at a given aperture.

But lets say you're keeping constant framing.

Now it's constant f/stop vs. equivalence. At the same f-stop, yes, the smaller format has more DoF.

But, if the sensors are equally advanced the larger sensor should have enough better noise performance you can stop down to equal DoF, compensate with a higher ISO, and end up with both noise performance and DoF being identical.

Max DoF is equal, because the limiting factor—diffraction—precisely counteracts the DoF effects as format scales. The smaller format hits whatever threshold you have for diffraction blurring at a numerically smaller f-stop. Given the same diffraction-limited-at-display-size-x goals, all formats have the same maximum DoF, it just comes at different apertures.

Equivalence breaks down when the smaller format runs out of apertures—the equivalent of a 24/f1.4 wide open on FF is approx. 16/1.0 on APS-C, the equivalent of an FF 50/1.0 on m43 is 25/f0.5

Oh, and you can't keep subject magnification completely constant with lenses having different angles of view. You can keep only keep it constant at one plane.

That is, if you have a 25 and a 200 that have the same framing/magnification at the plane of focus, your framing/magnification 20 feet behind the plane of focus will be very different.

For narrow DoF and not-radically-different angles of view this is often trivial to the point of non-existence. But f16 with the wider lens approaching hyperfocal it's huge.

Complain
 Like?
 Re: Depth of field for various formats and lenses In reply to ajscullard, May 30, 2012

ajscullard wrote:

For example on any format, a 100mm lens at f/1.0 (if you could find one) will have less DOF than the 50mm at f/1.0. More realistically the 100mm lens at f/2.0 will have less DOF than a 50mm lens at f/2.0.

Yes if you keep the distance constant, no if you keep the framing constant.

By also changing the subject distance you are introducing a further error, but the aperture issue is the main problem.

No, that's a perfectly valid thing to do—really more reasonable than comparing different framing for most purposes.

Complain
 Like?
 Re: Depth of field for various formats and lenses In reply to amangupta, May 30, 2012

amangupta wrote:

It is common to discuss DOF for various formats and lenses. There are two statements commonly made:

1) Small format means more DOF
2) Longer lens means shallower DOF

Statement #1 is false; statement #2 is true.

.

amangupta wrote:

What I found:

1) Small format means more DOF—true. Mainly because for the same lens focal length, subject distance increases, increasing the DOF.

2) Longer lens means shallower DOF—false. As the focal length increases, so does the subject distance and the eventual DOF remains constant.

I hope I've not made a mistake above. What do you think?

I think you're confusing a lot of things and hence, deceive yourself.

Depth-of-field depends on many factors. You want to know how factor A affects depth-of-field but you're looking at how a very particular combination of factors A, B, and C affects it. You see the effect of A, B, and C combined and falsely attribute the effect to factor A, ignoring the presence of factors B and C. That's why your conclusions about the original two statements are entirely wrong.

Complain
 Like?
 Re: Depth of field for various formats and lenses In reply to Olaf Ulrich, May 30, 2012

Olaf Ulrich wrote:

Depth-of-field depends on many factors. You want to know how factor A affects depth-of-field but you're looking at how a very particular combination of factors A, B, and C affects it. You see the effect of A, B, and C combined and falsely attribute the effect to factor A, ignoring the presence of factors B and C. That's why your conclusions about the original two statements are entirely wrong.

Agree he's wrong and clearly wrong

Focal length is part of the calculation for DOF..it's cast in stone anyone who denies this does not understand the formula for DOF

But it is a variable as are other areas, thus distance to subject is also a variable too

The luminous landscape did a pretty awful article which said that the DOF was the same for all lenses if the subject was the same size (focal length variable v distance to subject variable)

That is not entirely correct, in many (but not all) cases the overall DOF is the same or very near, there are examples where this is not correct too. The distribution of DOF is not the same either with short or long focal lengths. Lots of bad DOF articles out there people get confused.

Complain
 Like?
 Re: Depth of field for various formats and lenses In reply to Barry Fitzgerald, May 31, 2012

Barry Fitzgerald wrote:

Olaf Ulrich wrote:

Depth-of-field depends on many factors. You want to know how factor A affects depth-of-field but you're looking at how a very particular combination of factors A, B, and C affects it. You see the effect of A, B, and C combined and falsely attribute the effect to factor A, ignoring the presence of factors B and C. That's why your conclusions about the original two statements are entirely wrong.

Agree he's wrong and clearly wrong

No, he's not.

Focal length is part of the calculation for DOF..it's cast in stone anyone who denies this does not understand the formula for DOF

You both need to read his post. I'll quote the relevant portion for you:

"The overall assumption is that the framing of the subject is the single most important factor and has to remain constant. "

Focal length and subject distance cancel each other out for DOF calculations:

7D, 50mm, f4, 5 feet, DOF = 0.45 feet
7D, 500mm, f4, 50 feet, DOF = 0.45 feet

-- hide signature --

Lee Jay
(see profile for equipment)

Lee Jay's gear list:Lee Jay's gear list
Canon ELPH 500 HS Canon PowerShot SX260 HS Canon EOS 5D Canon EOS 20D Canon EOS 550D +23 more
Complain
 Like?
 Re: Depth of field for various formats and lenses In reply to Olaf Ulrich, May 31, 2012

Olaf Ulrich wrote:

amangupta wrote:

It is common to discuss DOF for various formats and lenses. There are two statements commonly made:

1) Small format means more DOF
2) Longer lens means shallower DOF

Statement #1 is false; statement #2 is true.

You got it backwards, given his assumptions of constant framing and constant f-stop.

I think you're confusing a lot of things and hence, deceive yourself.

He got it exactly right, given his assumption ("The overall assumption is that the framing of the subject is the single most important factor and has to remain constant.") and showed it with DOFmaster calculations.

From his original post:

Number 1 ("Small format means more DOF - TRUE."):
MFT - 25mm, f/1.4, 1.47m - 0.14m
FF - 50mm, f/1.4, 1.47m - 0.07m

Number 2 ("Longer lens means shallower DOF - FALSE."):
MFT - 45mm, f/1.8, 2.65m - 0.18m
MFT - 75mm, f/1.8, 4.4m - 0.18m

-- hide signature --

Lee Jay
(see profile for equipment)

Lee Jay's gear list:Lee Jay's gear list
Canon ELPH 500 HS Canon PowerShot SX260 HS Canon EOS 5D Canon EOS 20D Canon EOS 550D +23 more
Complain
 Like?
 Almost In reply to Lee Jay, May 31, 2012

ljfinger wrote:

He got it exactly right, given his assumption ("The overall assumption is that the framing of the subject is the single most important factor and has to remain constant.") and showed it with DOFmaster calculations.

From his original post:

Number 1 ("Small format means more DOF - TRUE."):
MFT - 25mm, f/1.4, 1.47m - 0.14m
FF - 50mm, f/1.4, 1.47m - 0.07m

Except that this assumes constant aperture rather than constant noise performance. Make that the same by adding two stops of ISO and aperture to the FF and bingo, DoF is the same.

Number 2 ("Longer lens means shallower DOF - FALSE."):
MFT - 45mm, f/1.8, 2.65m - 0.18m
MFT - 75mm, f/1.8, 4.4m - 0.18m

Again, this only works for situations where the far limit is nowhere near infinity. As the far limit approaches infinity it goes down in flames:

MFT - 45mm, f11, 10m = 54.9m
MFT - 75mm, f11, 16.67m = 22.3m

(also with DOFMaster)

Complain
 Like?
 Re: Depth of field for various formats and lenses In reply to Lee Jay, May 31, 2012

ljfinger wrote:

You both need to read his post. I'll quote the relevant portion for you:

"The overall assumption is that the framing of the subject is the single most important factor and has to remain constant. "

Sure, if you make that assumption. But why would you? Under what crazy set of circumstances is a 500mm lens at 15m an alternative to a 50mm lens at 1.5m?

What does making that assumption do for you? Worse than nothing: in the OP's case, you learn that focal length does not affect DoF, which is wrong: DoF is inversely proportional to the square of focal length.

And dofmaster is inadequate: it does not even allow the degree of scrutiny to be plugged into the formula.

Complain
 Like?
 Crazy?? In reply to Sante Patate, May 31, 2012

Sante Patate wrote:

ljfinger wrote:

You both need to read his post. I'll quote the relevant portion for you:

"The overall assumption is that the framing of the subject is the single most important factor and has to remain constant. "

Sure, if you make that assumption.

It is a perfectly reasonable assumption, given the nature of the question.

But why would you? Under what crazy set of circumstances is a 500mm lens at 15m an alternative to a 50mm lens at 1.5m?

Nothing crazy about a long lens at a distance being used to get similar framing to a shorter lens close... in fact that is the principle use of a long lens!!!
--
Regards,
Baz

"Ahh... But the thing is, they were not just ORDINARY time travellers!"

Complain
 Like?
 Not Crazy?? In reply to Barrie Davis, May 31, 2012

Easy to say airily ... but, please, specify a circumstance where a 500mm lens at 15m is a realistic alternative to a 50mm lens at 1.5m. A head and shoulders portrait? Conceivably, but hardly usual. A bird? No. Football? No. Lions feeding? No.

Complain
 Like?
 Re: Depth of field for various formats and lenses In reply to amangupta, May 31, 2012

amangupta wrote:

What I found:

1) Small format means more DOF - TRUE. Mainly because for the same lens focal length, subject distance increases, increasing the DOF.

True, but mainly because for the same framing, focal length decreases, and shorter focal lengths increase depth of field.

The overall assumption is that the framing of the subject is the single most important factor and has to remain constant.

I think this is true, but if you're comparing formats, framing is more than just keeping one object within the scene at a constant size; it also needs to include the context that object is in with respect to foreground and background elements, and how all of those elements are arranged into your composition. To get the same framing with different formats, you need to stand in the same place with a lens giving the same angle of view, and smaller formats will always have a shorter lens.

2) Longer lens means shallower DOF - FALSE. As the focal length increases, so does the subject distance and the eventual DOF remains constant.

Actually, this is also true, and the very condition you state proves it. If the longer focal length didn't decrease the depth of field, it wouldn't be balanced by the increased depth of field given by the greater shooting distance, and thus keep depth of field constant. If a longer lens didn't decrease depth of field, then your depth of field would always increase with greater shooting distances, regardless of any change in the focal length used.

In using the depth of field calculator you cited above, at least three facts are constant, or else the calculator wouldn't work: 1) DOF increases with shorter focal lengths and decreases with longer focal lengths; 2) DOF increases with smaller apertures and decreases with larger apertures; 3) DOF increases with greater shooting distances and decreases with shorter shooting distances. Saying that #1 is false just because it can be offset by a change in #3 doesn't make any sense.

Complain
 Like?
 Re: Depth of field for various formats and lenses In reply to Sante Patate, May 31, 2012

Sante Patate wrote:

ljfinger wrote:

You both need to read his post. I'll quote the relevant portion for you:

"The overall assumption is that the framing of the subject is the single most important factor and has to remain constant. "

Personally, I interpret that requirement in two different ways.

For normal photography, it means we keep the camera to subject distance constant, and depending on the sensor size, choose a suitable focal length to give the same framing.

For macro photography, it means adjusting both subject distance and focal length to give the same framing. Often with a compact camera, the macro capability works at just one fixed focal length, so camera-to-subject distance must be varied. With a DSLR, different focal length macro lenses are available, though an individual user may have a limited selection, so again subject distance may have to be varied to give the required framing.

And dofmaster is inadequate: it does not even allow the degree of scrutiny to be plugged into the formula.

"Degree of scrutiny" - I think this is covered by the ability to select a required circle of confusion. Dofmaster gives two ways to set this:

• One, by choosing a specific camera from the drop-down list. For example the Canon S95 gives a circle of confusion of 0.006mm.

• Two, the circle of confusion may be selected directly from the same drop-down list. The smallest figure available is 0.001 mm, which corresponds to about 40 to 160 MP on the S95, depending on whether or not the AA-filter is taken into account.

Regards,
Peter

Complain
 Like?
 Re: Not Crazy?? In reply to Sante Patate, May 31, 2012

Sante Patate wrote:

Easy to say airily ... but, please, specify a circumstance where a 500mm lens at 15m is a realistic alternative to a 50mm lens at 1.5m. A head and shoulders portrait? Conceivably, but hardly usual. A bird? No. Football? No. Lions feeding? No.

It's an object case. It isn't a specific photographic solution.

However, situations like that do arise.

For instance, I once had to photograph greeting cards (Hallmark) standing up on backgrounds supplied by the AD. Unfortunately the b/g were too small for the cards standing up, having originally been bought for them lying flat and in contact.

Long lens to the rescue...

Shooting with a much longer lens than usual increased the effective size of the backgrounds... and we sneaked under the wire into success territory. The client also preferred the lack of variation in size between the two flaps of the cards, resulting from the distant perspective....

(Note: this was pre-digi.. no cheap way to grow the backgrounds available at the time.)
--
Regards,
Baz

"Ahh... But the thing is, they were not just ORDINARY time travellers!"

Complain
 Like?
 Re: Almost In reply to Roger Krueger, May 31, 2012

Roger Krueger wrote:

ljfinger wrote:

He got it exactly right, given his assumption ("The overall assumption is that the framing of the subject is the single most important factor and has to remain constant.") and showed it with DOFmaster calculations.

From his original post:

Number 1 ("Small format means more DOF - TRUE."):
MFT - 25mm, f/1.4, 1.47m - 0.14m
FF - 50mm, f/1.4, 1.47m - 0.07m

Except that this assumes constant aperture rather than constant noise performance.

Constant f-stop, yes, but that's what the OP implied as his assumption.

Make that the same by adding two stops of ISO and aperture to the FF and bingo, DoF is the same.

Right.

Number 2 ("Longer lens means shallower DOF - FALSE."):
MFT - 45mm, f/1.8, 2.65m - 0.18m
MFT - 75mm, f/1.8, 4.4m - 0.18m

Again, this only works for situations where the far limit is nowhere near infinity. As the far limit approaches infinity it goes down in flames:

MFT - 45mm, f11, 10m = 54.9m
MFT - 75mm, f11, 16.67m = 22.3m

Right - by examples he was assuming SD

-- hide signature --

Lee Jay
(see profile for equipment)

Lee Jay's gear list:Lee Jay's gear list
Canon ELPH 500 HS Canon PowerShot SX260 HS Canon EOS 5D Canon EOS 20D Canon EOS 550D +23 more
Complain
 Like?
 Re: Depth of field for various formats and lenses In reply to Sante Patate, May 31, 2012

Sante Patate wrote:

ljfinger wrote:

You both need to read his post. I'll quote the relevant portion for you:

"The overall assumption is that the framing of the subject is the single most important factor and has to remain constant. "

Sure, if you make that assumption. But why would you?

Because I select my focal length to get the framing I want, regardless of where I might be standing.

What does making that assumption do for you?

It means you get a picture framed the way you want it framed.

And dofmaster is inadequate: it does not even allow the degree of scrutiny to be plugged into the formula.

Yes, it does, it allows you to select any CoC you want.

-- hide signature --

Lee Jay
(see profile for equipment)

Lee Jay's gear list:Lee Jay's gear list
Canon ELPH 500 HS Canon PowerShot SX260 HS Canon EOS 5D Canon EOS 20D Canon EOS 550D +23 more
Complain