The sensor size, DOF, crop factor, and effective focal length undergarments have been hanging out there too long. This issue has been discussed to death. Let's get onto something different.
Perhaps the definition of DOF is just too narrow and there should be more than one definition. You are taking DOF from a three-dimensional world and putting it onto a two dimensional piece of paper. Sounds to me like you just lost all your DOF. Unless you are using three-dimensional paper.
DOF didn't really exist until there was photography (capturing a 3 dimensional world on a 2 dimensional medium). Try to find a 16th Century painting that displays DOF. DOF is an optical phenomenon that can be described mathematically, but it is hardly a "law of physics" as some have claimed.
Why try to teach the whole story of DOF or just tell part of the story? Students have an inquiring mind and want to know the intimate details of how things work. It also saves many of the students time in the studio. They conceptualize the shot and with a complete understanding of DOF they can choose the lens, frame the shot, set the aperture and take the shot in much less time. That sure is impressive to an AD, as opposed to photographers that move the camera stand back and forth and switch lenses etc. trying to get the shot right. A through understanding of the techniques and processes goes a long way to solving complex compositional issues.
I just got of the phone with Canon Tech Support and asked about DOF and the calculations. The Tech, Jeff, stated that there would be NO difference in DOF between the 1Ds Mark II and the 1D Mark II when using the same lens at the same settings.
These are the guys that make the things.
He stated that because the rear element of the lens (and here is where the whole medium format argument gets thrown out the window) remains the same distance from the focal plane (sensor) the phyical attributes of the lens do not change and therefore the DOF scales remain a constant while mounted on either body.
The whole print size argument is moot because if you are printing an image taken with a 1Ds Mark II you are much more likly to print larger (I am) then if you had taken the same image with the 1D Mark II. Because you can!
Hear is the number (US) for anyone who wants to confirm this: 800-828-4040.
For all of you who keep quoting all of your varied internet sources, I will say that I just read on the internet this morning that Mars on the 27th of this month will appear as large as the moon. And as for quoting recognized sources, I beleive I could argue that slavery and genocide are the will of God by quoting a sentence here and a sentence there directly from the Bible. It's been done before.
I just got of the phone with Canon Tech Support and asked about DOF and the calculations. The Tech, Jeff, stated that there would be NO difference in DOF between the 1Ds Mark II and the 1D Mark II when using the same lens at the same settings.
These are the guys that make the things.
Sorry the guy or gal on the Tech Support line did not make the camera. The people that make the cameras are engineers and have Ph.D. after their name. They fully understand the nature of DOF and all it nuances and ramifications. The Tech Support guy is just that. He does not have Ph.D. after his name is ignorant about the factors of DOF. Did you ask him his qualifications for answering the question? My wife is a nurse and can answer medical questions. Would you like her to do your next surgery?
nsbca wrote:
He stated that because the rear element of the lens (and here is where the whole medium format argument gets thrown out the window) remains the same distance from the focal plane (sensor) the phyical attributes of the lens do not change and therefore the DOF scales remain a constant while mounted on either body.
And you believed him?
nsbca wrote:
I could argue that slavery and genocide are the will of God by quoting a sentence here and a sentence there directly from the Bible.
Jeff Donald wrote:
The Tech Support guy is just that. He does not have Ph.D. after his name is ignorant about the factors of DOF.
And your doctorate was in what? Optical engineering? I'll take the word of someone trained by Canon on the characteristics of a Canon lens over some link to a self proclaimed photography guru's web page.
This what I've kown all along, and the Canon Tech just confirmed that for me.
nsbca wrote:
And your doctorate was in what? Optical engineering? I'll take the word of someone trained by Canon on the characteristics of a Canon lens over some link to a self proclaimed photography guru's web page.
This what I've kown all along, and the Canon Tech just confirmed that for me.
Hmm, option (a) trust a photographer and lecturer in the subject when he tells you something about how cameras work or (b) believe what some low paid drone on a tech support line thinks is true but isn't in his database of answers.
There are just some people who, if they don't know, you can't tell them
I do wish, though, that we could get nsbca to shoot a view camera at f/4 or f/8 without tilt and then we'll see what he thinks about DOF and format size.
There are just some people who, if they don't know, you can't tell them
I do wish, though, that we could get nsbca to shoot a view camera at f/4 or f/8 without tilt and then we'll see what he thinks about DOF and format size.
How far is the focal plane from the rear element of the lens of the large format camera as opposed to that of a 35mm? Are we using the same (85mm) lens or ar the charateristics different? There is more to the equation you just put forth then merely the size of the sensor or film.
If I were to take my 1Ds sensor and put it directly in the middle of the same large format camera, all other things being the same, I would have the same exact DOF with an extremely cropped image.
I guess those that do not understand the concept of "circle of confusion" are doomed to be trapped within it (whether they work for Canon or Nikon or whomever).
I'm totally OK if someone wants to redefine how 1) to define and then 2) calculate DoF in a world of variable pixel pitch and variable format and variable screen resolutions and prints of different sizes from different formats at different ppi.
So far no-one who does not like/believe/understand the current (60+ year old) concept/definiton and calculation of DoF has proposed anything sensible as an alternative in this thread. Heck, they haven't even defined what they mean by DoF.
nsbca wrote:
If I were to take my 1Ds sensor and put it directly in the middle of the same large format camera, all other things being the same, I would have the same exact DOF with an extremely cropped image.
Yes, the same DoF at the sensor plane. But not the same DoF in an 8x10 print made from the sensor image and an 8x10 print made from the large format negative (say 8x10). I think you would prefer to look at an 8x10 print from 15 inches away than to look at a 22.5 x 15mm (7/8 x 5/8 inch sensor size) print from the same 15 inch distance.
You wrote "all other things being the same". That is not enough. You have to be precise. You might mean that the print size is the same (8x10) or you might mean the magnification to print size is the same (1:1) yielding 8x10 and 7/8 x 5/8, respectively.
As has been said, you have to account for all of the variables: 1) lens focal length 2) sensor or negative size, 3) print size, 4) viewing distance, 5) accepted size of circle of confusion in the print for that viewing distance.
N.B.: The distance to the rear element is not a factor and doesn't appear in DoF formulae or interactive calculators. A standard design 21mm lens for a rangefinder has a different distance to the rear element than a retrofocus 21mm such as an SLR might use, to make room for the mirror. But both lenses will have the same depth of field on 35mm film.
A picture, or in this case two, will be worth the thousands of words thrown around so far. Can someone with two different format cameras do this? I can simulate this with my single DSLR, but I suspect someone will cry foul.
Sam Bennett wrote:
The reason the DoF ends up being different is because you have to use different focal lengths to accomplish the same goal. So, say you want to take a headshot of someone, and you want the FoV of an 85mm lens. With a FF camera, you'd use an 85mm lens, with a 1.6 cropper you'd most likely use a 50mm lens. If you look at the math (I'm going to use a mythical 53mm lens for the 1.6 cropper, just to prevent hair splitting), you'll see that from the same distance you get:
85mm from 10ft @ f/2.8: 8.36 inches DoF
53mm from 10ft @ f/2.8: 13.8 inches DoF
So for all practical purposes (the ones that matter), then yes - FF cameras will produce narrower DoF. If you look at the difference between the two cameras, you'll see that 8.36 * 1.6 = 13.376, so the DoF difference calculation is pretty easy. Use the crop factor as a multiplier to get the DoF equivalent as well.
This also illustrates why most Medium Format lenses don't open up past f/3.5 in general. The DoF at f/3.5 is very narrow already, an 1.4 lens on a Medium Format body would be practically useless.
This also illustrates why FF cameras will in general have a sharpness advantage for certain types of photography. If you're shooting at f/1.8 on a 1.6 crop camera to achieve a certain Depth of Field effect, the same image can be obtained at f/3.6 on a FF camera, which with most lenses is much closer to the "sweet spot" in terms of sharpness....Show more →
Sam said it the best on the very first page. Read it again above. It's the simplist and the most practical explanation for 99% of us here. Getting bogged down in the science isn't helping the vast majority. Understanding what Sam wrote above is.
And your doctorate was in what? Optical engineering? I'll take the word of someone trained by Canon on the characteristics of a Canon lens over some link to a self proclaimed photography guru's web page.
This what I've kown all along, and the Canon Tech just confirmed that for me.
I have a Masters of Science in Photography and a Masters of Science in Education. I have completed approximately 25 hours towards by doctorate in education.
I've reviewed approximately 12 college text books I teach from and The Optical Society of America Handbook of Optics, second edition volume 1. Nearly every source referred to Applied Photographic Optics, third edition, by Sidney Ray as the source for there information on Depth of Field and Circle of Confusion. Both books are available from Amazon, for those interested.
I reproduced the pertinent page, 217, from Ray's book. You can also view the page if you log into you Amazon account. I quote "Depth of field is therefore determined by the geometry of the taking, enlarging and viewing conditions related to the circle of confusion standard adopted (Figure 22.3)."
If this isn't a good enough authority to conclude that enlarging and viewing are essential to determining DOF then I'm at a loss. nsbca and others are wasting their time and should be writing the next college text on the subject as it will revolutionize optics and photography.
slin100 wrote:
A picture, or in this case two, will be worth the thousands of words thrown around so far. Can someone with two different format cameras do this? I can simulate this with my single DSLR, but I suspect someone will cry foul.
When my 1Ds Mark II comes in next week I will be glad to do this.
Jeff Donald wrote:
I have a Masters of Science in Photography and a Masters of Science in Education. I have completed approximately 25 hours towards by doctorate in education.
I've reviewed approximately 12 college text books I teach from and The Optical Society of America Handbook of Optics, second edition volume 1. Nearly every source referred to Applied Photographic Optics, third edition, by Sidney Ray as the source for there information on Depth of Field and Circle of Confusion. Both books are available from Amazon, for those interested.
I reproduced the pertinent page, 217, from Ray's book. You can also view the page if you log into you Amazon account. I quote "Depth of field is therefore determined by the geometry of the taking, enlarging and viewing conditions related to the circle of confusion standard adopted (Figure 22.3)."
If this isn't a good enough authority to conclude that enlarging and viewing are essential to determining DOF then I'm at a loss. nsbca and others are wasting their time and should be writing the next college text on the subject as it will revolutionize optics and photography.
The above calculations are 100% correct if all thing given - focal length, aperture diameter, distance from subject and (your 8x10) size of print and viewing distance from print remain the same. If the sensor size changes under these conditions the image will indeed change and the percieved DOF will also change. I have no problem with that whatsoever.
In the practical world the 8x10 standard is non-existant. Use all the forementioned conditions, but change the 8x10 to 100% enlargement. What happens?
In the real world how many people buy a 1Ds Mark II to print at 8x10? Few. You can do that quite well and for a few thousand less with a 300D.
Let's go through a real life scenario. I have a FF 1Ds. I shoot a picture crop and enlarge the image to say 54x24. Nothing hard about that is there? Can I do that with my 300D? Not very well. If I wanted the same aspect ratio I would print the image from the 300D at 34x15. That would give me a very close DOF (+/-) and the print would be close to the same in clarity and sharpness.
The truth is that you will not print as large an image from a 300D as you can from a 1Ds.
On paper, with the 8x10 standard in place (all other variables staying the same) the DOF will change if the sensor size changes. In the practical world it remains a constant....Show more →
