Personally, I'd like a high pixel fill-factor and an OLPF with a null at about 0.5 cycles/pixel. 🙂
Personally, I'd like a high pixel fill-factor and an OLPF with a null at about 0.5 cycles/pixel. 🙂
I know I'm somewhat reviving a slow thread by now - sorry.
Math is good. I like math. But in this case, I don't find it very helpful.
My depth of field rule of thumb -- most SLR/DSLR cameras have a DoF preview toggle -- use it, and use your eyes to determine whether the DoF is what you want. Get in the habit of looking at it even when it doesn't matter much for what you're shooting, because you'll get to know about what it's going to be in different conditions.
Please ignore.
Just as important as the aperture etc. while shooting, which you can put into a DoF calculator, is the size of the print and the viewing distance or, combined, the angular size of the print as seen by the viewer.
As an example, take this image which was IIRC part of the launch campaign of the Canon 1Ds Mark III. All three examples are the same image at different sizes of enlargement and if you stay behind your computer at the same distance the only variable will be the 'print size'.
<><
The third is at 100% (one pixel of the image is displayed by one pixel of the screen) and now you can see that the ear is extremely blurry.
If you have the room, now stand up and move backward while looking at the third image and see the DoF getting deeper while the image gets smaller for you as the viewer.
If you're serious about DoF you'll have to know in advance how the image will be viewed and even what the eyesight of your viewers will be.
Thanks for the 'move back' demo. As a 3 or 4 diopter myopic, I found it most instructive.
@JohnVickers has written: @SrMi has written: @JohnVickers has written: @SrMi has written: @JohnVickers has written: @SrMi has written:Is anyone using the object-field technique?
You mean something like:
DOF ~= (size of subject)^2 * (f/ number) / 10
?
[for FF, in SI units]Object-field technique:
Set your aperture equal to the size of the smallest near objects you’d like resolved, then focus on infinity.
(aperture size: focal length divided by the aperture f-number).Oh, I see what you mean. Not really. I think you would end up being diffraction-limited at standard-to-long focal lengths. At Ultra-Wide, the aperture can be a similar size to the area covered by a pixel for quite nearby objects.
Which approach do you use for focusing?
If there's no foreground - like shooting off a cliff, or a peak, or a tall building, then shoot at the lens best f/number, and focus on the main interest - which is likely in the hyperfocal-infinity range.
If there's something important in the foreground, focus on that, and stop down for distant stuff.
Which is sub-optimal, but in the second case, I'm almost always shooting fairly wide, so my "foreground" is "hyperfocal" - or close - already, and I figure I'm not missing out so much.
I hadn't seen JIm K's notes on f-stops and focus distance before this thread. I need to think more about that.
The interesting question is where to focus if you want "infinity"/far-objects and as much of the foreground to be sharp and/or what CoC to use for hyperfocal when you do not know in advance what kind of output and viewing distance you should count on.
I am also working through Jim's post, looking for more insights into the topic.
Y'all might find this of interest after reading it through a time or two ...
www.trenholm.org/hmmerk/TIAOOFe.pdf
The only 'rithmetic involved is figuring the aperture diameter (a long-forgotten fundamental factor in everyday photography - now obscured by the f/number paradigm).
Everything else is just point-and-shoot 😉
Y'all might find this of interest after reading it through a time or two ...
www.trenholm.org/hmmerk/TIAOOFe.pdf
The only 'rithmetic involved is figuring the aperture diameter (a long-forgotten fundamental factor in everyday photography - now obscured by the f/number paradigm).
Everything else is just point-and-shoot 😉
His “arithmetic” might be right but it is irrelevant. What matters for the perception of detail is the angular resolution, not the absolute resolution on the scene.
@JohnVickers has written: @SrMi has written:Is anyone using the object-field technique?
You mean something like:
DOF ~= (size of subject)^2 * (f/ number) / 10
?
[for FF, in SI units]Object-field technique:
Set your aperture equal to the size of the smallest near objects you’d like resolved, then focus on infinity.
(aperture size: focal length divided by the aperture f-number).
a la Merklinger: