[aliquis]

Not Pentax lens specific, not pentax Q&A, so maybe it belongs here even though it's regarding lenses and not specifically other brands than Pentax? =P

Anyway, I learned that DOF changes against aperture as crop-factor changes on sensor sizes. That is a f/2.8 lens on 2x crop factor will behave as a f/5.6 lens on 1x crop and so on.

That's good and all but is that only valid for the area the lens "draw" so to speak? I can't understand why a crop from a bigger image would have a different DOF?

So the DOF on a fullformat lens even on an APS-C-sensor is the same as it would be on a fullformat sensor? Or not?

If not why not?

There's no different space between the sensor and lens between them right? Doesn't the APS-C sensor just pick up a smaller part of the image?

I can't get it =P

Or is it just that an f/2.8 lens at say 100mm will work as 160mm on 1.6 crop and it won't give the same DOF as a 160mm/2.8 on FF? I can understand that. But if the lens is designed for FF shouldn't it have just the same DOF as 100mm/2.8 on FF? But cropped ..

If focus length affect DOF at all but I find it way more likely than crop of image?

I just wonder if a fast say 50mm lens on an APS-C camera will blur just as good as it would on an FF camera, except the crop, but maybe that's a stupid comparison because one will back further away and hence it won't blur as much.


WTB camera, but not invest in the wrong lenses..

[SpecialK]

So the DOF on a fullformat lens even on an APS-C-sensor is the same as it would be on a fullformat sensor? Or not?

Not.

If not why not?

The APS-C sensor is smaller.

I just wonder if a fast say 50mm lens on an APS-C camera will blur just as good as it would on an FF camera

No. See above.

[aliquis]

Not.
The APS-C sensor is smaller.
No. See above.

K, thanks.

Though I find it hard to "imagine" why.

Like if someone took a portrait of some person and then cut away all the sides of the photo, the DOF would stay the same =P. Can't understand the difference unless it's distance related to the subject because one want to have the same coverage of the sensor/area.

[stevebrot]

There are complex formulae related to DOF calculation, but when properly cooked down, there are only two variables*:

• Absolute aperture size (not f/stop)
• Reproduction ratio at the film/sensor (size on media:actual size)

*assuming display/print size and viewing distance are constant.

In practice, we control the aperture through changing the f/stop and control the reproduction ratio by either changing the camera-to-subject distance or by choice of focal length. Cropping (as in cropped sensor) changes neither variable, though to get the same relative image size on the sensor/film, you need either a greater camera-to-subject distance or a shorter focal length (smaller reproduction ratio).

The assumptions (* text) are also very important. The larger your final image and the closer the viewing distance, the narrower the DOF. A good example would be my avatar picture. As a thumbnail, the entire image appears acceptably sharp. Here is the same image at 800 x 536 px:


K10D, LZOS MC Jupiter-9 85/2

At full resolution for the original 10 megapixel image, DOF is practically nonexistent.

Steve

P.S. This thread really should be in the general photography section.

---

Last edited by stevebrot; 05-10-2010 at 03:15 PM.

[Adam]

Moved to general photography.

[GingerBeer]

Quite some time ago I looked up the depth of field formulas (for calculating the hyperfocal distance, near limit and far limit) and created a spreadsheet that replicates the DOF calculators that can be found online. I found it beneficial for better understanding what influences DOF.

I'm not an expert but the different depth of field between APS-C and 35mm (or other formats) stems from the circle of confusion for the format - the larger the format the larger the CoC. I couldn't really tell you all that much about circle of confusion other than regurgitating what I can find on the internet. Wikipedia and/or google searches for circle of confusion might be a good place to start.

I hope this helps.

[tuco]

I just wonder if a fast say 50mm lens on an APS-C camera will blur just as good as it would on an FF camera, except the crop, but maybe that's a stupid comparison because one will back further away and hence it won't blur as much.

As noted already, no. But, yes, you will get shallow DOF with a fast 50mm on a APS-C. Shallow enough, I should say. I have a SMC Pentax-A 50mm f1.2 and tried it on my APS-C wide open and f1.4. At close range, I couldn't imagine wanting it much shallower. YMMV.

[Class A]

Though I find it hard to "imagine" why.

Everything stevebrot wrote is correct.

Note that there are two effects working in the opposite direction.

1. In order to achieve the same output size, you have to enlarge the image from the smaller sensor more than that of the larger sensor. As a result, the image from the smaller sensor will lose DOF.
   
2. In order to achieve the same AOV, you have to move the camera with the smaller sensor back or use a lens with a smaller focal length. As a result the DOF increases.

The second effect is higher than the first and hence the net result is that DOF increases for the smaller sensor.

There is a good article about Depth of Field by Bob Atkins.

[Marc Sabatella]

Here's a simple way of wrapping your head around the idea of how cropping affects DOF:

Take a picture, then print it 4x6". Now crop out a section with a pair of scissors. Obviously, no change to DOF.

But then go back to the original picture and do the crop *before* printing. Still print it 4x6". Now, suddenly, the image is *larger* in the print. And *that's* what makes DOF seem less. Everything looks in focus if you view it small enough, but the bigger you view something, the easier it is to tell if something is slightly out of focus.

So it's not the crop that changes DOF - it's blowing the image up bigger to view the cropped image the same size as the you would have viewed the original that changes DOF.

[stevebrot]

Everything stevebrot wrote is correct.

Note that there are two effects working in the opposite direction.

1. In order to achieve the same output size, you have to enlarge the image from the smaller sensor more than that of the larger sensor. As a result, the image from the smaller sensor will lose DOF.
   
2. In order to achieve the same AOV, you have to move the camera with the smaller sensor back or use a lens with a smaller focal length. As a result the DOF increases.

The second effect is higher than the first and hence the net result is that DOF increases for the smaller sensor.

There is a good article about Depth of Field by Bob Atkins.

Thank you @Class A for the affirmation and for filling in the blanks...

Steve

[wed7]

Another Question about DoF. They say DoF or openning of lenses varies as per focal length. Now
which has more DoF? For example, F/8 of a 50mm lens or F/8 of 200mm?

Thank you.

[Lowell Goudge]

Not Pentax lens specific, not pentax Q&A, so maybe it belongs here even though it's regarding lenses and not specifically other brands than Pentax? =P

Anyway, I learned that DOF changes against aperture as crop-factor changes on sensor sizes. That is a f/2.8 lens on 2x crop factor will behave as a f/5.6 lens on 1x crop and so on.

That's good and all but is that only valid for the area the lens "draw" so to speak? I can't understand why a crop from a bigger image would have a different DOF?

So the DOF on a fullformat lens even on an APS-C-sensor is the same as it would be on a fullformat sensor? Or not?

If not why not?

There's no different space between the sensor and lens between them right? Doesn't the APS-C sensor just pick up a smaller part of the image?

I can't get it =P

Or is it just that an f/2.8 lens at say 100mm will work as 160mm on 1.6 crop and it won't give the same DOF as a 160mm/2.8 on FF? I can understand that. But if the lens is designed for FF shouldn't it have just the same DOF as 100mm/2.8 on FF? But cropped ..

If focus length affect DOF at all but I find it way more likely than crop of image?

I just wonder if a fast say 50mm lens on an APS-C camera will blur just as good as it would on an FF camera, except the crop, but maybe that's a stupid comparison because one will back further away and hence it won't blur as much.


WTB camera, but not invest in the wrong lenses..

I have had a similar argument in the beginner's forum, where people say that DOF is a function of sensor size, and the problem is, it is not, but under certain constraints people impose it can be, well sort of, maybe.,...

Here goes.

You are correct, take an image with any lens, and cut the center 60% portion out of it, did the DOF change in any way biy cropping. NO WAY

BUT now enlarge that 60% crop, up to reach the dimensions of the origonal ucropped size. Now DOF changes. The whole point is that DOF for any lens is simply a function of the lens, the aperture, the origonal shooting distance (or magnification ratio), and the final image size (the image to print magnification ratio). the problem is, when you go to a cropped sensor, to get the same field of view, not only do you change focal length, but you change the magnification of the print, and the initial image magnification, so you really are comparing oranges and apples.

What is worse, is that in today's environment where every one views an image on a different size monitor, DOF is almost irrelevant.

[RioRico]

Another Question about DoF. They say DoF or openning of lenses varies as per focal length. Now which has more DoF? For example, F/8 of a 50mm lens or F/8 of 200mm?

Remember that an f-stop is a ratio, the ratio of the maximum aperture opening to the lens' focal length. And DOF is a complex function of photography, presentation and perception. If we look just at the photographic elements, just at lenses, we get this: At the same f-stop, a shorter/wider lens will have thicker DOF than will a longer lens. And at the same focal length, a lens with a tighter f-stop (smaller iris opening) will have thicker DOF than a lens with a wider f-stop. Those are the basics.

There is a simple way to compare DOF of various lenses and settings, with all else being equal. (Of course, all else is NOT equal, but we'll ignore that right now.) This simple metric is the DOF INDEX and it goes like this: divide focal length by f-stop. So a 50/2 lens has an index of 25. So does a 100/4 lens. This index doesn't say what the DOF is, just that 50/2 and 100/4 and 200/8 and 400/16 will have similar DOF, and that will be thinner than 50/3.5 or 100/5.6 or 200/11. The larger the index number, the thinner the DOF. Which has thinner DOF, a very expensive 58/1.2 or a very cheap 135/2.8? Do the math and see.

What is worse, is that in today's environment where every one views an image on a different size monitor, DOF is almost irrelevant.

Amen. At least, DOF as an objective calcuable factor. We can still judge *relative* DOFs, that one image may seem to have thicker or thinner DOF than another, but it can't be pinned down. Too many variables. So when shooting, just go with the basics:

* For thinner DOF, use longer focal lengths and/or wider apertures and/or larger frames and/or closer working distances.
* For thicker DOF, use shorter focal lengths and/or tighter apertures and/or smaller frames and/or further distances.

Those are the DOF ROTs (rules of thumb). They handily ignore the presentation and perception areas, and all the minutiae of diffraction and Circles of Confusion and all that. They ignore what can be done with light, or with split-dioptre lenses or massively-tilted lensboards, or the various other tricks that can affect DOF. In a word, they over-simplify. But those are the factors a photographer can most easily control.