And also for others like @Joanna and @JoPoV . https://www.photopills.com/calculators/dof-advanced
Here you can set another CoC based on the wanted print size and viewing distance. Changing that will produce another CoC for the same sensor size.
It seems that the standard settings are based on
Print size of 8’‘×10’’ (20cm×25cm).
Viewing distance of 10" (25cm).
Manufacturers standard visual acuity (the viewer can perceive details which size is roughly 0.01”).
Read the text below the calculator.
If diffraction becomes vissible it becomes vissible first in the sharp area.
Which is all well and good if you know those things ahead of shooting the image. But, if you don’t, then it makes sense to always shoot as if you wanted a large print - which is what I do by basing my blur circle diameter on 4x the pixel pitch my sensor (20µm)
When needed, I use a DOF calculator with the standard 30 µm COC. Then I adjust parameters taking into account diffraction and the intended viewing size and distance.
Anyway, the shot is perfectly sharp in the plane of focus, anywhere else sharpness is «acceptable», so you have to balance that with the effect of diffraction essentially in the plane of focus. @Joanna has her own process and she is happy with it. Others have their own way to proceed, if they get the results they want / need, what else ?
Once more
You don’t set the CoC in your tool. You choice another sensor size with a smaller CoC. There’s a diference between them. In PhotoPills calculator you set/select the CoC for that sensor and you just choice another sensor. Let me explain the difference.
D850, CoC 0.03mm, printed to the reference size 8x magnification. That CoC will become 0.24mm on print.
Nikon 1, CoC 0.01mm, printed to the reference size 24x magnification. That CoC will become 0.24mm. All the same as it should be.
Now your example.
D850 but you select the Nikon 1 with a CoC of 0.01. But printing to the reference size is still 8 times magnification. Your CoC will become 0.08 on your print. Your dof scale is total worthless.
I don’t know how to set the CoC directly in PhotoPills calculator but change the viewing distance to 0.08m gives me a Coc of 0.01, the one you choiced. And now with the right sensor size, so the right figures for dof etc.
That’s why I wonder how it affects the hyperfocal distance.
No. Don’t bother. I have a methodology that works and produces stunning prints. I know that increasing the size of the blur spot diameter beyond 20µm or reducing the aperture beyond f/10 gives me softer images.
Maybe you feel you need to limit yourself to the PhotoPills calculator but I have found that TrueDoF-Pro gives me reliable results at the sizes I print to and, what is more, I don’t have to factor in final print size before taking the shot.
There is no way you are ever going to convince me otherwise so, would you please stop dragging this thread in a direction I didn’t want it to go?
So the question is, if diffraction becomes visible :
Is diffraction a “blur” added to “dof blur” (which increase the size of the “blur diameter” whatever “dof size” is) ?
or does diffraction can be smaller than the size of the dof blur when the dof blur becomes very large in the image ?
In first case the response seems obvious :
What means asking for hyperfocal distance on the calculator ? :
You enter a blur circle diameter and ask what value to enter as focal distance to get this blur diameter at infinity, and what will be the nearest distance at which we will find this same blur diameter so you’re sure this diameter is never exceeded between those 2 extremes.
Problem is that this blur diameter is defined by 2 blur parameters : the one we classically call dof and the diffraction and second one can be bigger or smaller than the first one.
If a calcultor only take dof parameter into account and not diffraction, it can under evaluate the real blur diameter since diffraction can make it bigger if diffraction happens.
So say you want a 20µm blur circle.
case 1 : aperture produces no diffraction bigger than 20µm, no problem.
case 2 : aperture produces diffraction bigger than 20µm, say 25µm : here is a choice.
you overlook diffraction and compute an hyperfocal for 20µm wich gives some range but with visible diffraction.
you take diffraction into account and get an hyperfocal distance for 15µm (20 - (25-20)).
This gives is an other range without visible diffraction.
And this is of course a simplified vision, see below :
@George Maybe this link will help you improve your calculator ?
Here is some extract :
Most photographers are familiar with the classic method for calculating depth of field based on some acceptable COC (Circle Of Confusion). A typical formula for macro applications is :
TDOF_coc = 2 * C * f_lens * (m+1) / (m*m)
where
-TDOF_coc is the total DOF (front to back) such that the lens “blur circle” remains
smaller than C.
-C is the diameter of the COC, typically taken as 1/1000 to 1/1500 of the sensor width.
-f_lens * (m+1) is the effective f-number of the lens, estimated from the lens’s own f-
number and the magnification.
-m is the magnification
Formulas like this always ignore diffraction and instead are based on the simplifying assumptions of ray optics (also called geometrical optics)
Geometry limited: when the Airy disk is much smaller than the COC, the classic blur
circle model is accurate because there’s a good match between the envelope of the
diffraction pattern and the diameter of the ray optics blur circles.
Transition: when the Airy disk is only slightly smaller than the COC, the blur circle
model no longer accurately predicts the shape of the MTF curve, but it still does a pretty good job of predicting overall sharpness and thus overall DOF. In this regime, the actual diffracting lens will not resolve coarse detail at the same high contrast predicted by ray optics, but in exchange it resolves substantially finer detail at low contrast.
Diffraction limited: when the Airy disk is larger than the COC, the blur circle model becomes unusably inaccurate. In this regime the resulting images will not meet the sharpness criterion implied by the selected COC, but in exchange the DOF will be significantly greater if you’re willing to accept what sharpness there is. Another reasonable way of thinking about this regime is that if you’re working here, then you’ve just implicitly agreed to accept a larger COC, at which point you could repeat the classic calculation with the larger COC and get a decent prediction that way.
It’s not meant as shouting but accentuating. It was written on my cell phone.
No.That’s what I’m opposing against all the time. In her tool and all the others you select a sensor size and the corresponding CoC. See post 103. PhotoPhills Advanced dof calculator gives you the opportunity to overrule the standard CoC.
About diffraction, thats supposed to become visible when the airy disk is twice the sensel size. But please let that be another thread.
It’s now about her use of her Dof calculator. As soon as you select another sensor size as your camera’s you get wrong figures. You just can’t use them.
Total rubbish!!! it works for me as well as several of my photographic friends.
As for diffraction vs focus blur, the a look at my series of photos…
The focus doesn’t change, the DoF does change obviously and, as the aperture shrinks, beyond the diffraction limit, diffraction becomes more and more obvious.
As a silent observer of this discussion, it seems to be about a subject of which there are many possible solutions. It’s like asking:
— Which language is better, Spanish, English, Hindu, French……
For one of you to be “right” it doesn’t mean that the others are “wrong”.
The real, and final, answer, are the resulting photographs.
As an outside observer, and without a LF camera, I’ll never be more than that, those of you who have posted or published photos already have credibility from answering the questions.
It’s like “15 x 4 = 60”
Which does not mean “10 x 6 = 60” is any more right or wrong.
As for me, the only one of this group who is posting THEIR photos along with the information of how they did so is Joanna. So I’ll sit here and read lots of things, many of which I don’t understand, but to me, the bottom line being the resulting photo, she seems to be the only person posting here based on actual experience, not theory.
Please excuse my interruption, but unless/until I buy a LF camera and start processing and printing again, there seems to be an infinitely long list of factors that are far more important for me to struggle with, even those things I think I understand already.
You just refuse to understand how a dof calculator works. That is extremely disappointing of you.
Try comparing the results on your calculator between a D850 with 46MB pixels and set by your method to 0.01mm blur circle and a Nikon 1 camera with a standard CoC of 0.01. They’re the same! If this doesn’t make you thinking…
Rubbish!
A typical reaction of a person who just doesn’t want to learn. Sorry.
Look at post 94, your image with diffraction limiting. CoC 0.02,f/10. And then compare it with PhotoPills Diffraction Calculator and the D90. Maybe you’ve to correct the number of megapixels slightly. Even your calculator uses around twice the sensel size. But PhotoPhills also makes a difference between viewing 100% and a print. Look for yourself.
How can someone coming out of the world of computers even think of an answer like this!
As long you refuse to think rational and try to understand how a program works there is no reason to continue this discussion.
George
I’m sorry George but I find your attitude to be condescending and self-aggrandising.
You have obviously not studied the effects of diffraction in addition to those of defocus blur and I put far more faith in George Douvos than in you.
Now, take the time to read his article, in which he explains the rationale behind his apps.
He says that diffraction needs to be considered, mainly at smaller apertures but not necessarily at larger apertures and, thus, his app introduces corrections and limitations for those of us who wish to go further than the standard CoC methodology for critical use cases.
in other words, if you want to continue with your methodology, carry on and you will be satisfied in your own mind.
I, on the other hand, have found his calculator to be immensely useful for the type of work that I do with large prints, which goes beyond the “norms” and gives me a simple, fast and reliable way of calculating the combined effects of defocus and diffraction.
Because, as with my work as a consultant software engineer, I think things through, not taking anything at face value, and looking as much outside the box as in it. When I find something that may not be considered “normal” I evaluate it thoroughly, test it practically and either throw it out or keep it. In this case, after a full and thorough evaluation, I concluded years ago that George Douvos knows more about the subject than you claim to.
From your link
Question Through many years of experience, I know what value for CoC to use for my needs. Now you want me to work in terms of blur spot diameter. How do I do that? Is it a simple 1:1 translation? Or do I need to start all over again with experimentation to see what works?
Short answer Unless he has extraordinarily critical needs, a photographer should use as his target for a suitable blur spot diameter the same value as the CoC that he has been happy with.
But that’s not the problem. The problem is that in determining the dof, there is a relationship between the print and a CoC and therefore the magnification of the sensor. And you let go of this relationship.
Reading this did you look outside the box? Normal it’s 2.5 time the sensel size. Have a look at CambridgeinColor a widely appreciated web site Diffraction Limited Photography: Pixel Size, Aperture and Airy Disks.
Go to the diffraction calculator, try booth normal and advanced. Not much different as PhotoPills.
Simple. At the time of taking, I don’t know the finished print size and, since I also use Topaz Photo AI to resize and sharpen, sometimes long after the image was captured, how on earth am I meant to apply any formula that references print size and magnification?
Nonetheless, to placate you, here is PhotoPills, setup to find the area of sharp focus for a 28mm lens at f/10 for a hyperfocal shot and printing to A2 (59.4cm x 42cm)…
Note that the CoC is 16µm but that the DoF near limit is 2.56m and the hyperfocal near limit is 2.49m. Both of which are too close to avoid diffraction limiting sharpness at the near distance. Running the same numbers through TrueDoF-Pro gives me a diffraction-free hyperfocal distance of 5.28m and a near distance of 2.65m.
The only number that changes is the max CoC, which gets smaller when based on pixel size. But both resultant max CoCs end up at less than the 20µm that I normally use. And the pixel based one even approaches my ultimate blur circle size for working at f/5.
And neither set of results is diffraction limited.
Now, this calculator can’t work with f/10, so here are the calculations for f/11 instead…
This time it is the airy disk size that changes but, once again, all results are within the 20µm that I would normally use.
Now, the calculation based on pixel size shows as being diffraction limited at f/11 but, I’m willing to bet that, at the slightly larger aperture of f/10 that would not be the case.
Now, what you were saying about my calculator? At least it can handle f/10.
