DOF! Now there’s a topic! Getting a clear formula for calculation is a challenge because different people want to see/do different things. Bokeh is a bit like that too, there are different definitions floating around and even those teaching at the college/university level don’t always have a clear understanding or explanation.
Some time ago, on another forum, someone asked what they could have done to improve their image and I replied with a post of several paragraphs dealing with different aspects of the subject photo. One of the things I said was: “The further back you are, and the shorter the lens, the more depth of field you have available. If you want them both sharp in this position, you need about 9 feet or 3 meters DOF, easy at the 18 mm end, harder at the 55 mm end of the lens.” The photo was taken with a Rebel T3i and kit 18-55 mm lens. “This position” was a man pushing a woman on a swing, with the swing at the far end of the arc, from him. So, another contributor (who in some other post said he had a PhD in optics, and has consistently good advice) replied with: ”
Just because this keeps coming up …
IF YOU KEEP THE SUBJECT THE SAME SIZE IN THE FRAME DEPTH OF FIELD DOES NOT DEPEND ON FOCAL LENGTH. .
If you leave the camera in the same place and change focal length lots of things change, including DOF around the subject, and subject’s size wrt the frame. .
Another site that has a calculator and discussion of the topic is http://www.dofmaster.com/ . They have an on-line calculator here: http://www.dofmaster.com/dofjs.html . The interesting thing about the calculator is that lots of people say a full frame sensor provides a shallower DOF than a crop (APS-C) sensor. So, you use the calculator and plug in some random numbers: Say, 10 feet, f/4 and a 50 mm lens. Choose a crop body, then repeat the calculation with a full frame body.
A 7D, 30D, Rebel T3i and Rebel T3, all result in exactly the same set of results. A different result is provided by choosing a full frame body, but a 1Ds (Mk II or Mk III), 5D (Mk II or Mk III), and 1Dx, all result in the same result.
Near Far Total Hyperfocal Circle of Confusion
APS-C group: 9.16 11 1.84 108.1 0.019
Full Frame group: 8.74 11.7 2.94 68.5 0.03
The calculator at http://www.cambridgeincolour.com/tutorials/dof-calculator.htm gives slightly different, but similar results.
Doesn’t look like this editor likes spaces, so the results didn’t form into a table. Anyway, the DOF Master calculator says the full frame DOF is 2.94 feet and the APS-C DOF is only 1.84 feet, so APS-C has less DOF! Then why do people say it has more? I think it gets back to what people see in the viewfinder, and where they stand to see it. With more real estate in full frame, you can take a step or two closer with a short telephoto to get a portrait. Or, you can stand in the same place and use different focal lengths to get the same framing with each body.
Last time I looked at this I took these photos:
Full frame at 10 feet
APS-C at 10 feet
And, full frame at 6 feet
I don’t know if that helps, or just muddies things more. I like the concept of keeping the subject the same size and getting the same DOF, but it’s not the way I think in the field, so more experimentation is needed to solidify that concept in my mind. I don’t think sensor size matters, at least not the dimensions of the chip, pixel size and density may have an effect. How much you have to enlarge the photo after it is taken may make a difference. That the computer tends to expand or shrink the frame to fit the monitor just adds confusion.
As a practical matter, to blur the background, a longer lens, closer subject and more distant background, works pretty well. More so, if you have a wide aperture lens.
So much for a quick answer! I’m out of time but there are a couple of other things like a video about sensor size and bokeh that may mess you up even more, but I think shows the state of confusion pretty well. I will post a link when I have a chance.