How (and why) does sensor size change DOF?

[Pi]

The truth is, two photographers standing next to each other, one crop sensor, the other ff sensor, with exactly the same camera settings and the same framed image (meaning the crop owner has a wider lens), if they both take an image then the crop sensor image has more dof.

We are going in circles. Read my first post in this thread. "Same settings" does not make much sense. Do you consider 50/2 and 80/2 to be "the same settings"? Yes or No? What about 50/2 and 80/3.2? Are they "the same"? How do you exactly choose "the same settings" if the FL's must be different? You are confused by "f/2" without understanding that this is just a ratio. The meaningful thing to do is to shot not with the "same settings" but with the settings which would give you the same picture in terms of AOV, DOF, shutter speed and shot noise. Those are the equivalents settings. So in your scenario, the crop camera owner would choose, say, 50/2 to match 80/3.2, if that is possible. The problem is, sometimes it is not.

Or, consider the corollary of your statement "But the fact remains: smaller formats do not allow you to get deeper DOF.", that would be, larger formats do not allow you to shoot narrower dof.

Oh, no, this is not a corollary of my statement.

 

[privatebydesign]

More semantics.

Reread the second answer in this thread, on page one, and its two links, they contain the points you have made, even, in the equivalence link, the superfluous "some lenses don't exist".

With regards equivalence, yet again, crop cameras are far more limiting with the lenses we actually have available than ff cameras are.

 

[Pi]

Reread the second answer in this thread, on page one, and its two links, they contain the points you have made, even, in the equivalence link, the superfluous "some lenses don't exist".

That is the whole point, actually. The next question would be: why don't they exist, when they do not.

This is not semantics, it is an attempt to honestly answer the OP question, especially the (and why) part.

 

[Meh]

This is only an illusion of sharpness. The truth is what really matters (the information). Looking at the print from far away only proves that human vision is very limited. At close-up you can see all the information captured by your camera, both sharp and blurry parts. So, sharpness = information. Then from the distance you see much much less information despite that it looks sharper. This kind of sharpness ≠ information. This trick is about the CoC of your eyes, DoF has nothing to do with it.

As I pointed out previously, you appear to be confused about what DoF is. As neuro just pointed out... DoF is not an objective paramater. It is entirely subjective by definition. It is the distance in front of and behind the plane of focus that appears sharp to a human observer. There are implicit assumptions in the DoF derivation about viewing size, viewing distance, and visual acuity of the observer.

The only truth is that only the plane of focus is sharp. And it's only maximally sharp not perfectly sharp. Every plane in front of and behind is less sharp. How much less sharp depends on a few things. Whether you can see that it's less sharp depends on how big the print is, how far you are from it, and how good your eye sight is.

For you to say at this point in the thread "that only proves that human vision is very limited" suggests you are still fundamentally missing the point that DoF is in fact a function of the limitations in human vision.

 

[CarlTN]

Hey guys,
I've read it multiple times now and i really dont understand it. Why does the sensor size affect the DOF and why do FF cameras have a smaller DOF? Seems like a pretty basic question but i really couldn't wrap my head around this concept.

Greetings from Germany,
Knut Skywalker

I have to say, that is one hell of a name!! Did your father fight in the clone wars, and also have rage issues?

This thread has drawn out all the optics nerds, I see. Jolly good show! It's a shame with all your collective knowledge, you can't go out and collectively build me a custom lens.

 

[Wild]

Yes, the same area, as a percentage of the total image area. That is where your enlargement and viewing distance come in. Enlarge a point and it gets bigger and more obvious until it is a circle, but it stays the same size in relation to the image as a whole. CoC is just about setting a standard about the distance at which a sharp point becomes a blurred circle to a humans eye, you cannot take the human eye out of the equation because the very definition of DoF contains the words "acceptably sharp" or "beyond the resolution of the human eye". That is a subjective element that is generalised to 0.2mm at 25cm in the final output. Obviously if we stand back our eyesight becomes the limiting factor so we can increase the CoC.

Maintain a reproduction size and viewing distance ratio such that the CoC (point at which you can't see the difference between a point and a circle) and you can go as big, or as small, as you'd like.

Wow that makes soooo much more sense now. I don't know why it was so hard for me to grasp that DOF is a subjective quality of a photo, which is highly dependent upon the human eye. Thanks for taking the time to provide the great explanations and descriptions!

 

[Marsu42]

Wow that makes soooo much more sense now. I don't know why it was so hard for me to grasp that DOF is a subjective quality of a photo, which is highly dependent upon the human eye.

Welcome to the club :-> ... my guess is that often we talk about "hard" nerd/tech stuff which is a nice distraction and sometimes very helpful if you need specific advice. But over all that, the elephant in the room gets overlooked, meaning that so much about photography is "soft", subjective and relative. The only time this regularly surfaces is if someone is cornered in an argument and states something beyond the original point like "Well, but a good photog can shoot great images with a 10d and tech details don't matter anyway."

 

[CarlTN]

Wow that makes soooo much more sense now. I don't know why it was so hard for me to grasp that DOF is a subjective quality of a photo, which is highly dependent upon the human eye.

Welcome to the club :-> ... my guess is that often we talk about "hard" nerd/tech stuff which is a nice distraction and sometimes very helpful if you need specific advice. But over all that, the elephant in the room gets overlooked, meaning that so much about photography is "soft", subjective and relative. The only time this regularly surfaces is if someone is cornered in an argument and states something beyond the original point like "Well, but a good photog can shoot great images with a 10d and tech details don't matter anyway."

Good point. And I assume somewhere in this thread, someone has mentioned how the depth of field always looks deeper via your eye through the viewfinder, than what the image sensor sees? There's probably a name for this phenomenon. (It's at least partly due to the fact that your eye has its own iris, along with the eye's "imager" not being the same size as the camera's...and probably a host of other factors.) I didn't notice this as much before I started using sensors larger than aps-c. Oddly enough the first time I really noticed it was with my 58mm f/1.4 on the 1D4, a 1.3x crop camera. This is why it's almost impossible to manually focus a fast aperture lens accurately...especially if you're just using the standard focusing screen, and the distances are closer than 10 feet or so.

However, none of this brilliance you all are imparting, is getting the lens I want, designed. 60-110mm f/0.7 zoom w/4 stop multi-axis IS, painted metallic pearl white...with 14,000 months/ no interest...one day only sale! (The day they're only open half the day...and no web orders taken during closed hours...etc...)

 

[bdunbar79]

DOF is subjective? Hmm. If my DOF is 8 feet in a photo, that is, 8 real-life feet out in the field, how in the world does that ever change after I take the photo?? 8 feet is 8 feet isn't it?

Actually, I wouldn't even need to take the photo. The DOF is still 8 feet.

Are you suggesting that by being subjective, it could be 8 feet, or 6 feet, or 10 feet, or 7.23838383 feet? How silly.

 

[privatebydesign]

Are you suggesting that by being subjective, it could be 8 feet, or 6 feet, or 10 feet, or 7.23838383 feet? How silly.

Look at my two images on the previous page. On the top picture the zip is sharp, it comfortably falls inside DoF calculators range. In the bottom image it clearly is not in focus. The DoF has changed due to the reproduction size. Increase your viewing distance to make the big zip the same size in your fov that the little zip was and it becomes sharp again. From across the room the 46" print is razor sharp, put your face up in it and it still is, just less of it.

DoF is a figure based on aperture and subject magnification. If you magnify your subject more (increase your output size) you reduce DoF, just like in my two images.

 

[bdunbar79]

Are you suggesting that by being subjective, it could be 8 feet, or 6 feet, or 10 feet, or 7.23838383 feet? How silly.

Look at my two images on the previous page. On the top picture the zip is sharp, it comfortably falls inside DoF calculators range. In the bottom image it clearly is not in focus. The DoF has changed due to the reproduction size. Increase your viewing distance to make the big zip the same size in your fov that the little zip was and it becomes sharp again. From across the room the 46" print is razor sharp, put your face up in it and it still is, just less of it.

DoF is a figure based on aperture and subject magnification. If you magnify your subject more (increase your output size) you reduce DoF, just like in my two images.

Ok, I get it. Thanks.

 

[neuroanatomist]

DOF is subjective? Hmm. If my DOF is 8 feet in a photo, that is, 8 real-life feet out in the field, how in the world does that ever change after I take the photo?? 8 feet is 8 feet isn't it?

Actually, I wouldn't even need to take the photo. The DOF is still 8 feet.

Are you suggesting that by being subjective, it could be 8 feet, or 6 feet, or 10 feet, or 7.23838383 feet? How silly.

It seems you think that based on your equipment, there's a 'slice' of the photo that's in perfect focus, say 3.8 feet in front of where you focused, and 4.2 feet behind it, then WHAM like magic at 4.3 feet behind the focal plane, everything gets blurry. That's not how it works.

Light from the plane of focus (which is best approximated by a plane in the geometric sense - 2D and infinitely thin) is focused on the image sensor (we're ignoring field curvature, of course). Everything outside that plane, even a few millimeters, is blurry...and the further from the focal plane, the blurrier it gets. That's optical physics. Whether it looks blurry to you depends on viewing size and distance and your visual acuity.

Tell me - how do you know your hypothetical shot has that 'real' 8 foot DoF? Did you use a DoF calculator? That calculator determines the 8 foot DoF based on an assumed specific print size and viewing distance (commonly 8x10" viewed at 1 foot). Change those assumptions, you change the calculated DoF.

 

[CarlTN]

Are you suggesting that by being subjective, it could be 8 feet, or 6 feet, or 10 feet, or 7.23838383 feet? How silly.

Look at my two images on the previous page. On the top picture the zip is sharp, it comfortably falls inside DoF calculators range. In the bottom image it clearly is not in focus. The DoF has changed due to the reproduction size. Increase your viewing distance to make the big zip the same size in your fov that the little zip was and it becomes sharp again. From across the room the 46" print is razor sharp, put your face up in it and it still is, just less of it.

DoF is a figure based on aperture and subject magnification. If you magnify your subject more (increase your output size) you reduce DoF, just like in my two images.

For a 46 inch print to look sharp with your face against it, is that 300 dpi? What prints so large at 300 dpi?

 

[privatebydesign]

For a 46 inch print to look sharp with your face against it, is that 300 dpi? What prints so large at 300 dpi?

I print at 240dpi, but don't get me started on peoples preconceptions on print resolution and how printer software dithers dots

I shot it with a 1Ds MkIII and a 100mm L Macro at f7.1. Native resolution is 3744 x 5616, I then re-sampled it in one go (I did all the tests in PS years ago and one step re-size works best for me) to 7488 x 11232, that prints at 240 dpi to 31.2" x 46.8", the tiff is 84.1MB, exactly four times the size of the .CR2.

Here is a 700 x 700 px crop from the eye and eyebrow, to get a better idea of the print in person zoom your browser so it is just under 3" square, or print it out at 240dpi. Obviously at this size and viewing distance dof becomes minute, you can see it fall off even in the hairs less than an inch off the eye.

That is what we can do with 21mp cameras, and that is why I, personally, don't need or want more MP when they inevitably come around.

 

[CarlTN]

For a 46 inch print to look sharp with your face against it, is that 300 dpi? What prints so large at 300 dpi?

I print at 240dpi, but don't get me started on peoples preconceptions on print resolution and how printer software dithers dots

I shot it with a 1Ds MkIII and a 100mm L Macro at f7.1. Native resolution is 3744 x 5616, I then re-sampled it in one go (I did all the tests in PS years ago and one step re-size works best for me) to 7488 x 11232, that prints at 240 dpi to 31.2" x 46.8", the tiff is 84.1MB, exactly four times the size of the .CR2.

Here is a 700 x 700 px crop from the eye and eyebrow, to get a better idea of the print in person zoom your browser so it is just under 3" square, or print it out at 240dpi. Obviously at this size and viewing distance dof becomes minute, you can see it fall off even in the hairs less than an inch off the eye.

That is what we can do with 21mp cameras, and that is why I, personally, don't need or want more MP when they inevitably come around.

That's a nice picture and illustration of your point. However, I recently did a 20x30 print scaled to 9000x6000 for 300dpi, shot with my 6D, with the 40mm pancake lens closed to f/9, ISO 100. (Native resolution of the file before resizing, was slightly below the 6D's native due to minor cropping after correcting some lens distortion and adding some vertical angle correction of the shot.) I was rushed at the last minute so I didn't perform a final noise reduction after some subtle sharpening done to the enlarged image. After the print was done, I couldn't see any of the noise visible on the 9000x6000 file when I viewed it at 100%. I concluded that the print could very likely have been done at 240dpi, rather than 300 (good thing this was only done on the cheap). I can certainly still see the difference between 240 and 300 when I view from less than 1 foot away...and obviously it makes more difference at the smaller 20x30 size, than it would with a 46 inch wide print...because it has been scaled less of a percent above the native.

Did you use the standard "bicubic" in photoshop when you scaled it up? I've found it works best, where "bicubic smoother" (claimed to work best for enlargement)...only adds softness to everything. Even better is Perfect Resize 7, but I didn't even bother using it on this most recent 20x30.

 

[privatebydesign]

For smaller prints I'll print at higher dpi, indeed I never downsample, but for the big stuff I found 240 dpi to be perfectly fine, people don't literally put their nose on a 46" print, I found, like you, much more than 12"-20" viewing distance and the difference between 240 and 300 (I actually never use 300, but 320 after getting recommendations direct from Epson pro printers support re the printers native sampling figures of 720) becomes moot, though it can depend heavily on the printer paper and the subject matter.

Also, printing takes some detail out, so I won't remove so much noise for a print and will sharpen more when compared to an electronic output.

I don't remember which resample mode I used, I normally try a couple, I watched a great PS tutorial by Deke on Lynda.com a long time ago and he very nicely demonstrated files where the recommended method just broke down, so going on his suggestion I'll try two or three and see which works better for that particular image.

Hope this all helps.

 

[bdunbar79]

DOF is subjective? Hmm. If my DOF is 8 feet in a photo, that is, 8 real-life feet out in the field, how in the world does that ever change after I take the photo?? 8 feet is 8 feet isn't it?

Actually, I wouldn't even need to take the photo. The DOF is still 8 feet.

Are you suggesting that by being subjective, it could be 8 feet, or 6 feet, or 10 feet, or 7.23838383 feet? How silly.

It seems you think that based on your equipment, there's a 'slice' of the photo that's in perfect focus, say 3.8 feet in front of where you focused, and 4.2 feet behind it, then WHAM like magic at 4.3 feet behind the focal plane, everything gets blurry. That's not how it works.

Light from the plane of focus (which is best approximated by a plane in the geometric sense - 2D and infinitely thin) is focused on the image sensor (we're ignoring field curvature, of course). Everything outside that plane, even a few millimeters, is blurry...and the further from the focal plane, the blurrier it gets. That's optical physics. Whether it looks blurry to you depends on viewing size and distance and your visual acuity.

Tell me - how do you know your hypothetical shot has that 'real' 8 foot DoF? Did you use a DoF calculator? That calculator determines the 8 foot DoF based on an assumed specific print size and viewing distance (commonly 8x10" viewed at 1 foot). Change those assumptions, you change the calculated DoF.

Remember when I said this:

"Ok, I get it. Thanks."

Sorry neuro, I couldn't resist ;D

 

[ecka]

DOF is subjective? Hmm. If my DOF is 8 feet in a photo, that is, 8 real-life feet out in the field, how in the world does that ever change after I take the photo?? 8 feet is 8 feet isn't it?

Actually, I wouldn't even need to take the photo. The DOF is still 8 feet.

Are you suggesting that by being subjective, it could be 8 feet, or 6 feet, or 10 feet, or 7.23838383 feet? How silly.

It seems you think that based on your equipment, there's a 'slice' of the photo that's in perfect focus, say 3.8 feet in front of where you focused, and 4.2 feet behind it, then WHAM like magic at 4.3 feet behind the focal plane, everything gets blurry. That's not how it works.

Light from the plane of focus (which is best approximated by a plane in the geometric sense - 2D and infinitely thin) is focused on the image sensor (we're ignoring field curvature, of course). Everything outside that plane, even a few millimeters, is blurry...and the further from the focal plane, the blurrier it gets. That's optical physics. Whether it looks blurry to you depends on viewing size and distance and your visual acuity.

Tell me - how do you know your hypothetical shot has that 'real' 8 foot DoF? Did you use a DoF calculator? That calculator determines the 8 foot DoF based on an assumed specific print size and viewing distance (commonly 8x10" viewed at 1 foot). Change those assumptions, you change the calculated DoF.

So here we go. Once that optical physics hits the sensor it's no longer optical, it's information. The sensor cannot capture the infinitely thin plane of focus made of sharp points. Instead, it captures everything between the two distances where "a circle" has the size of a pixel or smaller, so everything in that range is same sharp, because there can't be anything sharper than a pixel. At that level, enlarging the image isn't going to decrease the DoF, only soften it, because there is no hidden information.
How do we call THAT THING? The sharpest area between the two distances where "a circle" meets the pixel?

 

[Wild]

So here we go. Once that optical physics hits the sensor it's no longer optical, it's information. The sensor cannot capture the infinitely thin plane of focus made of sharp points. Instead, it captures everything between the two distances where "a circle" has the size of a pixel or smaller, so everything in that range is same sharp, because there can't be anything sharper than a pixel. At that level, enlarging the image isn't going to decrease the DoF, only soften it, because there is no hidden information.
How do we call THAT THING? The sharpest area between the two distances where "a circle" meets the pixel?

Hey Ecka, I was in a similar state of skepticism here for a while so I just want to help get you where I am now using some common sense.

Forget all the technical stuff about aperture and sensor size for a second and let's just think about this from a simpler perspective. Imagine we just took a picture of a friend. You look at the picture later at 100% on your 30" computer monitor and realize that you accidentally focused on their nose, so their eyes are slightly out of focus. Wouldn't you agree that the plane of focus is on the nose, but doesn't extend to the eyes? Yes of course.

Now let's imagine the friend we took the photo of wants to post that photo to Facebook, and so you post it and it becomes their new profile photo. The whole time you're thinking man that photo wasn't even in focus, but then you go onto their Facebook page, and voila, nobody can even tell if their eyes are in focus because the photo's so small. Basically, the photo looks great. Now I think we can agree that as far as we can tell, their profile picture is in focus. Which would mean that either:

A. You uploaded the wrong picture
B. Magic
C. Our DOF changed because we're now looking at a much smaller picture

And just in case you're not sure. It's obviously B. ;D

 

[studio1972]

DOF is subjective? Hmm. If my DOF is 8 feet in a photo, that is, 8 real-life feet out in the field, how in the world does that ever change after I take the photo?? 8 feet is 8 feet isn't it?

Actually, I wouldn't even need to take the photo. The DOF is still 8 feet.

Are you suggesting that by being subjective, it could be 8 feet, or 6 feet, or 10 feet, or 7.23838383 feet? How silly.

It seems you think that based on your equipment, there's a 'slice' of the photo that's in perfect focus, say 3.8 feet in front of where you focused, and 4.2 feet behind it, then WHAM like magic at 4.3 feet behind the focal plane, everything gets blurry. That's not how it works.

Light from the plane of focus (which is best approximated by a plane in the geometric sense - 2D and infinitely thin) is focused on the image sensor (we're ignoring field curvature, of course). Everything outside that plane, even a few millimeters, is blurry...and the further from the focal plane, the blurrier it gets. That's optical physics. Whether it looks blurry to you depends on viewing size and distance and your visual acuity.

Tell me - how do you know your hypothetical shot has that 'real' 8 foot DoF? Did you use a DoF calculator? That calculator determines the 8 foot DoF based on an assumed specific print size and viewing distance (commonly 8x10" viewed at 1 foot). Change those assumptions, you change the calculated DoF.

This is very correct. Before one can calculate DoF, one must first define what "in focus' actually means. A 300 pixel wide web image will have much more DoF than a 20MP version of the same original shot.

In answer to the original question, full frame sensors do generally allow more control over DoF than crop sensors, this is really to do with the lenses, as a 50mm 1.4 lens will have less DoF than a 35mm 1.4 lens (which is designed to do a similar field of view on a crop sensor based camera, all other things being equal.