A common question: "how does the depth-of-field on 1.5X crop factor (APS size) compare to full-frame film?"
I don't want to go into the theoretical and mathematical details of depth-of-field. So I will use a Palm program to solve this. The circle-of-confusion used by the program is 30um for full-frame, and 25um for APS-DSLR.
In short, the DOF is affected by: focal length, shooting distance, aperture, and film (sensor) format. So the above question is a tricky question because it has 2 different answer. The DOF on 1.5X DSLR is bigger then 135 full-frame, and it's also smaller. It all depends on how you look at the question.
First, if you are using 'equivalent' lens on your DSLR, shooting at the same position, using the same aperture.... the DOF on DSLR is greater than full-frame film. eg:
full-frame: 75mm lens, f/2, distance=2m
APS-DSLR: 50mm lens, f/2, distance=2m
You will get the same framing and perspective with the above 2 setting. But the DOF with APS-DSLR will be bigger than full-frame 135:
DOF of full-frame: 1.96m to 2.04m, ie. 8cm
DOF of APS-DSLR: 1.92m to 2.08m, ie. 16cm
So, many people say full-frame has more 3D feeling than APS-DSLR. The main reason it the difference in DOF: a shallower DOF gives more 3D feeling.
On the other hand, if you are using SAME LENS, shooting at SAME DISTANCE, than the DOF of APS-DSLR will be shallower than full-frame 135, eg:
full-frame: 50mm, f/2, distance=2m, DOF=1.91m to 2.10m, 19cm
APS-DSLR: 50mm, f/2, distance=2m, DOF=1.92m to 2.08m, 16cm
So if you are going to use the DOF scale on your lens, the DOF on APS-DSLR will be approx. 1/4 stop shallower.
4 comments:
You've missed one important factor - magnification.
In your first set using equivalent lenses, the difference in DoF is only so when one views the 2 digital files on the same screen at full resolution, where the magnification for both the APS size image and the full 135 size image are at 1:1. Yet everyone seems to ignore the fact that the APS size image is by default smaller than that of the full 135 size image.
However, when one takes those two same images and printed them to the same print size, say 20"x30", then the magnification from the APS size image to that same 20"x30" print would be considerably larger than that from the full 135 size file to printed to the same size. In that case, the differentiable DoF on both printed images would be roughly the same.
The circle-of-confusion actually taken care about magnification.
However, one major problem in my calculation is it took 30um for 135 full-frame, and 25um for APS size. Given same final-image size, the figure for APS size should be 20um. I will change the parameters in the Palm program when I have time, and re-do the calculation.
One question. Do those set of your CoC numbers already take into consideration of the absolute resolution of the 2 different size sensor/film being constant or vary differently to known extents? As different sensor/film resolutions also affect the perceivable DoF.
The original idea of CoC doesn't consider the resolution of the film / sensor. Indeed, it doesn't need to since it only concerns about the 'normal viewing condition', which means a not-too-big photo being viewed at a not-too-close distance.
However, if you want to compare 100% view DoF on screen, you will need to work out your own CoC. Chances are, such calculation will be meaningless cause such DoF calculated will be very small.
Post a Comment