DOF and cropping/enlargement

Started Nov 13, 2014 | Discussions
Lee Jay Forum Pro • Posts: 54,273
DOF and cropping/enlargement
3

Let's put this to rest.

Cropping doesn't change DOF, but the enlargement that you must do to make the now-cropped image match the original size does.

Here's some dispassionate proof.

These DOF evaluations were done with a computer, not by eye. I thought I'd use a Canon sample image from the new 100-400L II since it was just announced.

Original image

DOF evaluation zoomed out for the full image, and fully enlarged to 1:1 (which is a change in enlargement of about a factor of 6.5 for this sample). As you can see, the DOF is much narrower when fully enlarged (make sure you click on original):

I posted this image before, to demonstrate this fact. This is all the same image, just cropped and enlarged differently (make sure you click on original):

Incredibly, some people claimed that all of those have the same DOF. So, I ran the same computer analysis on those samples to make the point crystal clear (make sure you click on original).

Still don't believe it?

DOFMaster uses the exact DOF equations. Results are below, and you can verify them here:

http://dofmaster.com/dofjs.html

  • Canon 5D (full frame), 50mm, f/2.8, 10 feet, DOF = 2.06 feet
  • Canon 7D (1.6-crop), 50mm, f/2.8, 10 feet, DOF = 1.29 feet
  • Olympus E-PL3 (2.0 crop), 50mm, f/2.8, 10 feet, DOF = 1.02 feet
  • Nikon 1 J1 (2.7 crop), 50mm, f/2.8, 10 feet, DOF = 0.75 feet
  • Canon Pro 1 (4.0 crop), 50mm, f/2.8, 10 feet, DOF = 0.54 feet
  • Canon S95 (4.7 crop), 50mm, f/2.8, 10 feet, DOF = 0.41 feet
  • Canon SX30 (5.6 crop), 50mm, f2.8, 10 feet, DOF = 0.34 feet
-- hide signature --

Lee Jay

 Lee Jay's gear list:Lee Jay's gear list
Canon IXUS 310 HS Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon EF 8-15mm f/4L Fisheye USM +23 more
Canon EF 100-400mm F4.5-5.6L IS II Canon EOS 5D Canon EOS 7D Canon PowerShot S95 Nikon 1 J1 Olympus PEN E-PL3
If you believe there are incorrect tags, please send us this post using our feedback form.
DaveOl
DaveOl Veteran Member • Posts: 3,313
Re: DOF and cropping/enlargement

Your list of cameras and depth of field are going the wrong direction.  The FF should have the smallest depth of field.

It may be that the crop factor is involved in these measurements and if you were to shoot the cameras so that the exact same image was taken, then the dof's would be correct.

 DaveOl's gear list:DaveOl's gear list
Nikon D300S Fujifilm X-E2S
OP Lee Jay Forum Pro • Posts: 54,273
Re: DOF and cropping/enlargement

DaveOl wrote:

Your list of cameras and depth of field are going the wrong direction. The FF should have the smallest depth of field.

Nope. It's correct as written. Check it yourself.

It may be that the crop factor is involved in these measurements and if you were to shoot the cameras so that the exact same image was taken, then the dof's would be correct.

That's right. Smaller sensor = more enlargement = smaller DOF. The reason everyone believes the opposite is that they are correct when the framing is held constant. That means the larger sensors need longer focal lengths and that more than makes up for the enlargement difference, making larger sensors have shallower DOF with the same framing and the same f-stop.

  • Canon 5D (full frame), 50mm, f/2.8, 10 feet, DOF = 2.06 feet
  • Olympus E-PL3 (2.0 crop), 25mm, f/2.8, 10 feet, DOF = 4.28 feet
  • Olympus E-PL3 (2.0 crop), 50mm, f/2.8, 10 feet, DOF = 1.02 feet
-- hide signature --

Lee Jay

 Lee Jay's gear list:Lee Jay's gear list
Canon IXUS 310 HS Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon EF 8-15mm f/4L Fisheye USM +23 more
dsjtecserv Veteran Member • Posts: 3,761
Re: DOF and cropping/enlargement

DaveOl wrote:

Your list of cameras and depth of field are going the wrong direction. The FF should have the smallest depth of field.

It may be that the crop factor is involved in these measurements and if you were to shoot the cameras so that the exact same image was taken, then the dof's would be correct.

You may not be familiar with the discussions that gave rise to this post. those discussion began with a question about using the same focal length, subject distance, and aperture on different formats. In other words, everything except the sensor size is kept the same. That, needless to say, results in different framing. So we aren't discussing the situation where framing is kept the same on both formats (by changing focal length, subject distance, or a combination of both. Under the given conditions, an image created by a smaller sensor will have less depth of field than the one from the larger sensor, because the smaller image must be enlarged more to create final images (print or screen) that are the same size. Lee Jay's comparisons illustrate that.

Dave

 dsjtecserv's gear list:dsjtecserv's gear list
Canon EOS 5DS R Canon EF 17-40mm f/4.0L USM Canon EF 100mm F2.8L Macro IS USM Canon TS-E 24mm f/3.5L II Canon EF 24-105mm f/4L IS USM +4 more
enemjii Senior Member • Posts: 1,889
Re: DOF and cropping/enlargement

dsjtecserv wrote:

DaveOl wrote:

Your list of cameras and depth of field are going the wrong direction. The FF should have the smallest depth of field.

It may be that the crop factor is involved in these measurements and if you were to shoot the cameras so that the exact same image was taken, then the dof's would be correct.

You may not be familiar with the discussions that gave rise to this post. those discussion began with a question about using the same focal length, subject distance, and aperture on different formats. In other words, everything except the sensor size is kept the same. That, needless to say, results in different framing. So we aren't discussing the situation where framing is kept the same on both formats (by changing focal length, subject distance, or a combination of both. Under the given conditions, an image created by a smaller sensor will have less depth of field than the one from the larger sensor, because the smaller image must be enlarged more to create final images (print or screen) that are the same size. Lee Jay's comparisons illustrate that.

Dave

Actually, the discussion started was about using the same lens with the same subject distance and aperture on FF and APSC.

http://www.dpreview.com/forums/thread/3749926

 enemjii's gear list:enemjii's gear list
Sony SLT-A65 Sony a77 II Sony DT 55-300mm F4.5-5.6 SAM Sony DT 35mm F1.8 SAM Sigma 85mm F1.4 EX DG HSM +6 more
dsjtecserv Veteran Member • Posts: 3,761
Re: DOF and cropping/enlargement

enemjii wrote:

dsjtecserv wrote:

DaveOl wrote:

Your list of cameras and depth of field are going the wrong direction. The FF should have the smallest depth of field.

It may be that the crop factor is involved in these measurements and if you were to shoot the cameras so that the exact same image was taken, then the dof's would be correct.

You may not be familiar with the discussions that gave rise to this post. those discussion began with a question about using the same focal length, subject distance, and aperture on different formats. In other words, everything except the sensor size is kept the same. That, needless to say, results in different framing. So we aren't discussing the situation where framing is kept the same on both formats (by changing focal length, subject distance, or a combination of both. Under the given conditions, an image created by a smaller sensor will have less depth of field than the one from the larger sensor, because the smaller image must be enlarged more to create final images (print or screen) that are the same size. Lee Jay's comparisons illustrate that.

Dave

Actually, the discussion started was about using the same lens with the same subject distance and aperture on FF and APSC.

Would you disagree that the same lens has the same focal length on either format?

Dave

 dsjtecserv's gear list:dsjtecserv's gear list
Canon EOS 5DS R Canon EF 17-40mm f/4.0L USM Canon EF 100mm F2.8L Macro IS USM Canon TS-E 24mm f/3.5L II Canon EF 24-105mm f/4L IS USM +4 more
Aberaeron Veteran Member • Posts: 7,188
Re: DOF and cropping/enlargement

dsjtecserv wrote:

enemjii wrote:

dsjtecserv wrote:

DaveOl wrote:

Your list of cameras and depth of field are going the wrong direction. The FF should have the smallest depth of field.

It may be that the crop factor is involved in these measurements and if you were to shoot the cameras so that the exact same image was taken, then the dof's would be correct.

You may not be familiar with the discussions that gave rise to this post. those discussion began with a question about using the same focal length, subject distance, and aperture on different formats. In other words, everything except the sensor size is kept the same. That, needless to say, results in different framing. So we aren't discussing the situation where framing is kept the same on both formats (by changing focal length, subject distance, or a combination of both. Under the given conditions, an image created by a smaller sensor will have less depth of field than the one from the larger sensor, because the smaller image must be enlarged more to create final images (print or screen) that are the same size. Lee Jay's comparisons illustrate that.

Dave

Actually, the discussion started was about using the same lens with the same subject distance and aperture on FF and APSC.

Would you disagree that the same lens has the same focal length on either format?

Dave

I've not followed the discussion but doesn't that depend on the respective camera's flange distance? Maybe I'm missing something here?

 Aberaeron's gear list:Aberaeron's gear list
Fujifilm X20 Sony SLT-A57 Olympus OM-D E-M10 Panasonic Lumix DMC-G7 Olympus 7-14mm F2.8 Pro +20 more
OP Lee Jay Forum Pro • Posts: 54,273
Re: DOF and cropping/enlargement

Aberaeron wrote:

dsjtecserv wrote:

enemjii wrote:

dsjtecserv wrote:

DaveOl wrote:

Your list of cameras and depth of field are going the wrong direction. The FF should have the smallest depth of field.

It may be that the crop factor is involved in these measurements and if you were to shoot the cameras so that the exact same image was taken, then the dof's would be correct.

You may not be familiar with the discussions that gave rise to this post. those discussion began with a question about using the same focal length, subject distance, and aperture on different formats. In other words, everything except the sensor size is kept the same. That, needless to say, results in different framing. So we aren't discussing the situation where framing is kept the same on both formats (by changing focal length, subject distance, or a combination of both. Under the given conditions, an image created by a smaller sensor will have less depth of field than the one from the larger sensor, because the smaller image must be enlarged more to create final images (print or screen) that are the same size. Lee Jay's comparisons illustrate that.

Dave

Actually, the discussion started was about using the same lens with the same subject distance and aperture on FF and APSC.

Would you disagree that the same lens has the same focal length on either format?

Dave

I've not followed the discussion but doesn't that depend on the respective camera's flange distance?

No, it doesn't.  If the image is in-focus with the same lens at the same settings, then the flange distance is the same.  If it's out-of-focus, then none of this matters.

-- hide signature --

Lee Jay

 Lee Jay's gear list:Lee Jay's gear list
Canon IXUS 310 HS Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon EF 8-15mm f/4L Fisheye USM +23 more
Aberaeron Veteran Member • Posts: 7,188
Re: DOF and cropping/enlargement

Lee Jay wrote:

Aberaeron wrote:

dsjtecserv wrote:

enemjii wrote:

dsjtecserv wrote:

DaveOl wrote:

Your list of cameras and depth of field are going the wrong direction. The FF should have the smallest depth of field.

It may be that the crop factor is involved in these measurements and if you were to shoot the cameras so that the exact same image was taken, then the dof's would be correct.

You may not be familiar with the discussions that gave rise to this post. those discussion began with a question about using the same focal length, subject distance, and aperture on different formats. In other words, everything except the sensor size is kept the same. That, needless to say, results in different framing. So we aren't discussing the situation where framing is kept the same on both formats (by changing focal length, subject distance, or a combination of both. Under the given conditions, an image created by a smaller sensor will have less depth of field than the one from the larger sensor, because the smaller image must be enlarged more to create final images (print or screen) that are the same size. Lee Jay's comparisons illustrate that.

Dave

Actually, the discussion started was about using the same lens with the same subject distance and aperture on FF and APSC.

Would you disagree that the same lens has the same focal length on either format?

Dave

I've not followed the discussion but doesn't that depend on the respective camera's flange distance?

No, it doesn't. If the image is in-focus with the same lens at the same settings, then the flange distance is the same. If it's out-of-focus, then none of this matters.

But if one is,say, 8mm further from the sensor, surely the focal length is different? Different to that stated on the lens.

 Aberaeron's gear list:Aberaeron's gear list
Fujifilm X20 Sony SLT-A57 Olympus OM-D E-M10 Panasonic Lumix DMC-G7 Olympus 7-14mm F2.8 Pro +20 more
OP Lee Jay Forum Pro • Posts: 54,273
Re: DOF and cropping/enlargement

Aberaeron wrote:

Lee Jay wrote:

But if one is,say, 8mm further from the sensor, surely the focal length is different?

No, if you move the lens further away, you make the shot out of focus.  You've moved the focal plane much closer to the lens.  That's what an extension tube does.

-- hide signature --

Lee Jay

 Lee Jay's gear list:Lee Jay's gear list
Canon IXUS 310 HS Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon EF 8-15mm f/4L Fisheye USM +23 more
dsjtecserv Veteran Member • Posts: 3,761
Re: DOF and cropping/enlargement

Aberaeron wrote:

Lee Jay wrote:

Aberaeron wrote:

dsjtecserv wrote:

enemjii wrote:

dsjtecserv wrote:

DaveOl wrote:

Your list of cameras and depth of field are going the wrong direction. The FF should have the smallest depth of field.

It may be that the crop factor is involved in these measurements and if you were to shoot the cameras so that the exact same image was taken, then the dof's would be correct.

You may not be familiar with the discussions that gave rise to this post. those discussion began with a question about using the same focal length, subject distance, and aperture on different formats. In other words, everything except the sensor size is kept the same. That, needless to say, results in different framing. So we aren't discussing the situation where framing is kept the same on both formats (by changing focal length, subject distance, or a combination of both. Under the given conditions, an image created by a smaller sensor will have less depth of field than the one from the larger sensor, because the smaller image must be enlarged more to create final images (print or screen) that are the same size. Lee Jay's comparisons illustrate that.

Dave

Actually, the discussion started was about using the same lens with the same subject distance and aperture on FF and APSC.

Would you disagree that the same lens has the same focal length on either format?

Dave

I've not followed the discussion but doesn't that depend on the respective camera's flange distance?

No, it doesn't. If the image is in-focus with the same lens at the same settings, then the flange distance is the same. If it's out-of-focus, then none of this matters.

But if one is,say, 8mm further from the sensor, surely the focal length is different? Different to that stated on the lens.

I think you are overthinking this. Since the focal length of a lens is a property of the lens, it will always have the same focal length, regardless of what camera it is used with, or whether it is mounted on a camera at all. The question (and all of the responses) presume that the same lens can be used successfully on either camera, and that means that it will be mounted at a flange distance that allows focusing with both cameras.

My  questions to enemjii was partially facetious, as he seems to questioning whether the lens changes its focal length when it moves to a different camera. It doesn't, of course, and we hope he recognizes that.

Dave

 dsjtecserv's gear list:dsjtecserv's gear list
Canon EOS 5DS R Canon EF 17-40mm f/4.0L USM Canon EF 100mm F2.8L Macro IS USM Canon TS-E 24mm f/3.5L II Canon EF 24-105mm f/4L IS USM +4 more
tex Veteran Member • Posts: 7,810
Here's my problem with all of this
3

Let's try this: Take a photograph with x camera at x aperture. Print that photograph. Now Take scissors and cut off x amount around the sides.

Someone please explain to me how the DOF has magically changed because you used scissors to trim it. Is this some kind of "spooky action at a distance"?  It's exactly the same photograph, now physically trimmed down. Perhaps we can say that the relative or perceptual dof in the "new" image makes the dof seem different, but we know it is exactly the same image, shot with the same camera, lens and aperture---we just took scissors to the print.

So, with the above conditions, I am remaining unconvinced that we have an actual change in dof.

-- hide signature --

tex_andrews, co-founder and webmaster of The LightZone Project, an all-volunteer group providing the free and open source LightZone photo editing software. Personal website: www.texandrewsart.com
"Photography is the product of complete alienation" Marcel Proust
"I would like to see photography make people despise painting until something else will make photography unbearable." Marcel Duchamp

 tex's gear list:tex's gear list
Pentax 645Z Pentax K-1 Pentax K-1 II Ricoh Theta V Pentax smc FA 77mm 1.8 Limited +29 more
dsjtecserv Veteran Member • Posts: 3,761
Re: Here's my problem with all of this
1

tex wrote:

Let's try this: Take a photograph with x camera at x aperture. Print that photograph. Now Take scissors and cut off x amount around the sides.

Someone please explain to me how the DOF has magically changed because you used scissors to trim it. Is this some kind of "spooky action at a distance"? It's exactly the same photograph, now physically trimmed down. Perhaps we can say that the relative or perceptual dof in the "new" image makes the dof seem different, but we know it is exactly the same image, shot with the same camera, lens and aperture---we just took scissors to the print.

So, with the above conditions, I am remaining unconvinced that we have an actual change in dof.

Because the two pictures aren't comparable. They are different sizes. People don't make prints of different sizes just because their film or sensor is smaller, do they? In order to properly compare prints, they have to be viewed under the same conditions, which includes the same print size and viewing distance. To accomplish that, you need to enlarge the cropped picture (whether it was cropped in camera or after capture) to a greater degree than the larger one. Enlargement increases blur. The point at which blur becomes visible is what defines the limits of depth of field. Since the blur throughout the print from the smaller sensor is enlarged more, things that had the same amount of blur on the sensor will be blurrier on the print, and the blur threshold will be exceeded by a narrower range of depth of field.

Lee Jays pictures demonstrate that empirically.

Dave

-- hide signature --
 dsjtecserv's gear list:dsjtecserv's gear list
Canon EOS 5DS R Canon EF 17-40mm f/4.0L USM Canon EF 100mm F2.8L Macro IS USM Canon TS-E 24mm f/3.5L II Canon EF 24-105mm f/4L IS USM +4 more
moving_comfort
moving_comfort Veteran Member • Posts: 8,242
You left out the most important step
2

tex wrote:

Let's try this: Take a photograph with x camera at x aperture. Print that photograph. Now Take scissors and cut off x amount around the sides.

Someone please explain to me how the DOF has magically changed because you used scissors to trim it.

I'd suggest you re-read the OP - you've missed the point completely.

If you don't enlarge your snipped image to the original size, nothing changes. If you enlarge, the blur becomes more apparent, your DOF changes.

See below for another demonstration, with a further resize to (hopefully) drive the point home - and remember the concept of "acceptable sharpness" is part of the definition of DOF.

Please click 'original size' and expand to full size with '+'. In my browser (chrome) this results in separate tabs for each image, a good way to compare.

#1:

#2:

.

Now, in the first image, to my eye, the knob on the back of the cannon is fully 'acceptably sharp'.

Do you judge it to be so in the second image?

Has your perception of the level of acceptable sharpness of the knob changed in the second image?

.

-- hide signature --

Here are a few of my favorite things...
---> http://www.flickr.com/photos/95095968@N00/sets/72157626171532197/

 moving_comfort's gear list:moving_comfort's gear list
Pentax K20D Nikon D800 Nikon AF-S DX Nikkor 35mm F1.8G Nikon AF-S Nikkor 300mm f/4D ED-IF Nikon AF Nikkor 20mm f/2.8D +10 more
tex Veteran Member • Posts: 7,810
Sorry, still don't get it /nt
-- hide signature --

tex_andrews, co-founder and webmaster of The LightZone Project, an all-volunteer group providing the free and open source LightZone photo editing software. Personal website: www.texandrewsart.com
"Photography is the product of complete alienation" Marcel Proust
"I would like to see photography make people despise painting until something else will make photography unbearable." Marcel Duchamp

 tex's gear list:tex's gear list
Pentax 645Z Pentax K-1 Pentax K-1 II Ricoh Theta V Pentax smc FA 77mm 1.8 Limited +29 more
dsjtecserv Veteran Member • Posts: 3,761
Re: Sorry, still don't get it /nt

tex wrote:

[nothing]

Well, I'd try to explain it better if you'd give me an idea of what doesn't register with you.

Dave

-- hide signature --

tex_andrews, co-founder and webmaster of The LightZone Project, an all-volunteer group providing the free and open source LightZone photo editing software. Personal website: www.texandrewsart.com
"Photography is the product of complete alienation" Marcel Proust
"I would like to see photography make people despise painting until something else will make photography unbearable." Marcel Duchamp

 dsjtecserv's gear list:dsjtecserv's gear list
Canon EOS 5DS R Canon EF 17-40mm f/4.0L USM Canon EF 100mm F2.8L Macro IS USM Canon TS-E 24mm f/3.5L II Canon EF 24-105mm f/4L IS USM +4 more
OP Lee Jay Forum Pro • Posts: 54,273
Re: Here's my problem with all of this
1

tex wrote:

Let's try this: Take a photograph with x camera at x aperture. Print that photograph. Now Take scissors and cut off x amount around the sides.

Someone please explain to me how the DOF has magically changed because you used scissors to trim it.

Because, the standard viewing distance is equal to the image diagonal, or some multiple thereof.  If you keep the viewing distance proportional to the image diagonal, you'll have to move that print closer to you.  And that means you'll be able to see blur circles that are smaller than you could when it was farther away, which means the DOF is shallower.

Perhaps we can say that the relative or perceptual dof in the "new" image makes the dof seem different, but we know it is exactly the same image, shot with the same camera, lens and aperture---we just took scissors to the print.

Perceptual DOF is the only DOF.  DOF is a 100% perceptual phenomenon.

-- hide signature --

Lee Jay

 Lee Jay's gear list:Lee Jay's gear list
Canon IXUS 310 HS Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon EF 8-15mm f/4L Fisheye USM +23 more
enemjii Senior Member • Posts: 1,889
Re: Here's my problem with all of this

tex wrote:

Let's try this: Take a photograph with x camera at x aperture. Print that photograph. Now Take scissors and cut off x amount around the sides.

Someone please explain to me how the DOF has magically changed because you used scissors to trim it.

Because, the standard viewing distance is equal to the image diagonal, or some multiple thereof.  If you keep the viewing distance proportional to the image diagonal, you'll have to move that print closer to you.  And that means you'll be able to see blur circles that are smaller than you could when it was farther away, which means the DOF is shallower.

Perhaps we can say that the relative or perceptual dof in the "new" image makes the dof seem different, but we know it is exactly the same image, shot with the same camera, lens and aperture---we just took scissors to the print.

Perceptual DOF is the only DOF.  DOF is a 100% perceptual phenomenon.

>>>Perceptual DOF is the only DOF. DOF is a 100% perceptual phenomenon.
 enemjii's gear list:enemjii's gear list
Sony SLT-A65 Sony a77 II Sony DT 55-300mm F4.5-5.6 SAM Sony DT 35mm F1.8 SAM Sigma 85mm F1.4 EX DG HSM +6 more
enemjii Senior Member • Posts: 1,889
Re: Here's my problem with all of this

tex wrote:

Let's try this: Take a photograph with x camera at x aperture. Print that photograph. Now Take scissors and cut off x amount around the sides.

Someone please explain to me how the DOF has magically changed because you used scissors to trim it.

Because, the standard viewing distance is equal to the image diagonal, or some multiple thereof.  If you keep the viewing distance proportional to the image diagonal, you'll have to move that print closer to you.  And that means you'll be able to see blur circles that are smaller than you could when it was farther away, which means the DOF is shallower.

Perhaps we can say that the relative or perceptual dof in the "new" image makes the dof seem different, but we know it is exactly the same image, shot with the same camera, lens and aperture---we just took scissors to the print.

Perceptual DOF is the only DOF.  DOF is a 100% perceptual phenomenon.

>>>Perceptual DOF is the only DOF. DOF is a 100% perceptual phenomenon.
>>>Perceptual DOF is the only DOF. DOF is a 100% perceptual phenomenon.
 enemjii's gear list:enemjii's gear list
Sony SLT-A65 Sony a77 II Sony DT 55-300mm F4.5-5.6 SAM Sony DT 35mm F1.8 SAM Sigma 85mm F1.4 EX DG HSM +6 more
enemjii Senior Member • Posts: 1,889
Re: Here's my problem with all of this

Really?

 enemjii's gear list:enemjii's gear list
Sony SLT-A65 Sony a77 II Sony DT 55-300mm F4.5-5.6 SAM Sony DT 35mm F1.8 SAM Sigma 85mm F1.4 EX DG HSM +6 more
Keyboard shortcuts:
FForum MMy threads