You’re just repeating the nonsense written in your link. Sometime I think that guy doesn’t understand his own calculator.
Start by setting the blur diameter, or “circle of confusion,” to 10 microns. Yes, that’s a lot smaller than the “recommended” 30 micron circle of confusion for a 35 mm sensor. But that recommendation has its origin back in early 20th century film practice, while you’re using a Nikon D800, so just ignore it.
The CoC is a calculated figure. Based on the smallest point on a A4 print a human eye can see at an arm length distance. Calculating that point size back to the sensor size gives the CoC for that sensor size. That’s why CoC and sensor size are related.
Your tabel with airy disk diameters correspond with the figures given by PhotoPills. A little different maybe due different wavelength or roundings.
Again, your calculator is just a normal calculator with nothing new in it. Using it in the good way gives the same result as any other calculator since they use the same formulas. But it has a nice graphic interface, that’s true.
I can only repeat it because it works!!! In the real world, not some mathematical, theoretical world that few can understand.
Indeed. But it is calculated, based on the print viewing distance of an analogue print and discounting the extras resolution that hi-res digital sensors give.
Only if you stick to the limitations of film grain but not if you calculate based on pixel size, which is smaller than most grain. And, then, these calculators are designed for those who can’t be bothered to look up the pixel pitch, so they calculate it from size divided by pixel count to save effort.
Now that is where you are totally wrong. It was explicitly designed to take account of diffraction. The majority of calculators only concern themselves with basic focus blur.
Nope. they say around f/11 - and f/10 is near enough to qualify and what I use most of the time. Being a larger aperture, it also factors in an element of “safety margin”.
The only reason I mentioned f/5 is because that really is the “ultimate” diffraction limit for the ultra fussy, taking into account pixel pitch. But, as I have also said several times, I prefer to use f/10 (from a blur spot diameter of 20µm) because it gives a combination of a far more workable DoF with an acceptable sharpness up to any size print I am ever likely to need.
Here is a screen shot of TrueDoF-Pro using 20µm and the related diffraction limited aperture of f/10…
This is if you want to find a CoC for this purpose.
If you want for example to find a CoC for a film shown in cinemas you have to take into account a standard screen size and a standard viewer’s distance from the screen, the size of the film (35mm, 70 mm).
If you want to find a CoC for a 4m x 3m advertising poster viewed at 50 m you’ll get an other result.
CoC must be chosen according to the final medium and the viewing distance.
CoC determine how DOF (and sharpness) will be perceived, depending on the observation conditions.
So, how to get the right CoC for viewing an image at 100% zoom on a HD screen according to its pitch seen at a 60cm distance with a D850 sensor ?
“And this is where the question - with these dimensions, just how necessary is an LF camera when a hi-res digital can approach the same sizes?”
It isn’t necessary, it’s a pleasure to prepare (what film shall I use today? how much light will I get? how much wind on the beach or on the hill), to set up, tune (shift? tilt?) and focus; then the right moment to trigger exposure, plus the time spent for development to see the result.
Is this why some don’t talk of CoC, but of “Airy disk diameter”* or “blur spot diameter”* ? (I’m not comfortable with these english denominations and not sure of their precise meaning).
Does these take into account final medium and observation conditions ?
Or only what happens on sensors : how to get the sharpest possible image and where the sharpest zone is distributed (DOF) on the sensor.
This is what interest me : how to get the sharpest image with my sensor. Without taking into account final medium nor observation conditions nor taking into account limitations due to lenses optical quality, shooting conditions, etc …
EDIT : @Joanna we were writing about simultaneously and you responded that when I was writing on a post above.
@Joanna
To resume what I understand about this ebony/D850 comparison for landscape :
Resolution :
Ebony as a (very) little edge in resolution.
ISO :
D850 can get higher iso without noise (after post processing) but should be used at ISO64 for best result.
DOF :
This is a little less clear.
DOF without tilt-shift : D850 support f/10, ebony support f/32 but what comparing DOF ?
DOF with tilt-shift : how would D850 with tilt-shift lens compare ?
Focus stacking : D850 can take those shoot automatically and fast. Probably more tweaky with ebony (if possible : what about focus breathing ?).
Shutter speed :
In those conditions (iso - max usable aperture) D850 will allow faster shutter speed so less problem with moving objects (wind in trees or grass for example).
In matter of fact CoC is taking in account the viewing distance. A dot at an armlength distance, or 1 meter. If you print an image twice as big and view it at twice the distance you shouldn’t see any difference.
The combination of that circle of 0.3mm viewed at 1 meter distance is expressing the minimal angle the human eye can see.
But dof calculators don’t take that in account. And that is what we’re discussing, or trying to explain.
You still don’t get it. Step out off your circle. You don’t select a blur spot. You select a sensor size with its corresponding CoC.
The image you show is that from a crop camera, crop factor 1.5. It’s even mentioned in the app. It goes even further. Take PhotoPills and select a DX camera. You’ll see different categories: different sensor resolution. Thus different diffraction limitations.
What happens if you lower the aperture slider to 4? Show me that.
You are so right in some ways but I also take a lot of time, effort and care over my digital work. The main difference with digital being that I can shoot at whatever ISO I feel appropriate and I don’t have to mess around with chemistry and scanning.
And yet, there is something about the LF process that won’t let me get rid of the gear
I think you could be right. From my point of view, CoC is a crude approximation, based on an outdate’d concept.
Absolutely. And that is what I love about the precision and excellence that basing blur spot diameter on pixel pitch gives me.
Yes, using Topaz to finish off images for printing is what has been the biggest game changer for my attitude. Along with the 46Mpx sensor, I never dreamt I could get such astounding large prints.
The main problem with focus stacking is it doesn’t allow for moving subjects (even snails). Whereas, with the Ebony, I don’t have to spend out on an expensive shift lens, as the tilt on the Ebony works with any lens I want to use.
Indeed, but I do find I sometimes need to slow down the speed on the D850 to give it that lovely soft feel that gently waving grasses can give.
Ah! Now that is the biggy…
Fuji Acros 100 B&W - around 14 stops with appropriate over-exposure and under-development . Must expose for darkest shadows with details and extend/process for brightest highlights.
Fuji Velvia 100 colour transparency - limited to 5-6 stops if you are lucky and have to use graduated filters to tame highlights beyond that.
Nikon D850 - 14.6 stops at 100 ISO, less at higher ISOs, colour or B&W
I don’t know how to say this any other way - I couldn’t care less about ancient, approximate, CoC methods. But you must feel free to use whatever you want.
And you still don’t get it. Sensor size has no influence on blur spot diameter. Old school calculators like PhotoPills only use that as a shortcut for those who can’t be bothered to be precise.
Certainly TrueDoF-Pro does show different sensor/film sizes, but they are only for those who want or need to use them. As an advanced and skilled user, I just go for the blur spot diameter.
And yet once again, those are only there for the feeble minded who don’t truly understand diffraction with a hi-res sensor camera.
This is a regular dof calculator. Showing the dof for a 1.5 crop camera, 50mm lens and an aperture of f/4. Nothing special with it. Except the limit with diffraction at f/10. But for a 1.5 crop camera with a 14MB sensor. Do you have a 1.5 crop camera with a 24MB sensor then the diffraction limit will be f/8.
image 1
If you set distance to 3.92 m , you’ll get from 1.96 m to infinity “spots” less or equal to [20 µm + diffraction - so more than 20 µm] on your sensor.
image 2
If you set distance to 5.26 m , you’ll get sharper image to infinity (smaller “spots” since nearer to focal point) and so from 2,64 m to infinity you’ll get “spots” less or equal to [20 µm in total - so no more than 20 µm].
This is what I understand and it seems logical.
It seems obvious a simple DOF calculator wich does not take diffraction into account is not good and can lead to very uniformely soft images. This one seems to take it into account.
