The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Influence of format size on a photograph

AreBee

Member
Folks,

I have read many times of the visual difference that a larger sensor returns when compared to a smaller sensor. Many photographers attribute "the Medium Format look" at least in part to the physical size of the sensor, be it film or digital.

All else being equal, I do not understand why the physical size of a sensor should make the slightest bit of difference to what is recorded. As far as I am aware a sensor crops from the image circle of the lens mounted in front of it. No more, no less.

In order for a comparison between sensors of different physical size to be made on a like for like basis, depth of field would require to be matched. In addition, the same lens should be used. It is probably not possible to normalise the difference in colour from a range of sensors, though I dare say that comparing cameras from the same manufacturer would help to minimise any difference that existed. However, as far as I am aware "the Medium Format look" has never been claimed to relate to colour - velvia on a Small Format camera returns the same colour characteristics as it does on a Medium Format camera, does it not?

The thought occurred to me that it should be relatively simple to test for the difference in look, if any, insofar as it relates to sensor physical size for a range of formats. Has anyone ever carried out such a comparison? Is there any reason that such a comparison would not be valid?
 

kzphoto

New member
The quick explanation:
Larger sensors / film sizes mean that your lenses must provide a larger image circle. Larger image circles mean less DoF, more tonality, and higher possible sharpness captured in a single frame. You're basically shoving more information into a larger capture.
 

jlm

Workshop Member
for the same field of view, the larger sensor will record more data and with finer resolution
 

AreBee

Member
kzphoto,

jzphoto said:
Larger sensors / film sizes mean that your lenses must provide a larger image circle. Larger image circles mean less DoF, more tonality...
Those who see a difference between format sizes often associate it with a difference in tonality, as you have noted. However, you refer to tonality as a function of lens image circle. Consequently, sensor size and tonality are independent from each other because a sensor is nothing more than a crop from a lens image circle.

Apparently there is a dilemma because if the same lens is used in a comparison of sensor size, I would expect the result from the larger format to be identical to that from the smaller format over the area covered by both. Hence, tonality must indeed derive from the lens...except that, as noted above, many photographers associate tonality with the size of format.

Which is correct, or am I missing something?

kzphoto/jlm,

kzphoto said:
...and higher possible sharpness captured in a single frame. You're basically shoving more information into a larger capture.
jlm said:
for the same field of view, the larger sensor will record more data and with finer resolution
That the surface area of a larger format can capture greater detail than a smaller is not in dispute, but neither can it be relevant because:

1. "the Medium Format look" is never mentioned in the context of differences that may be apparent when pixel peeping.

2. Greater resolution does not a Medium Format look make. Will cramming 80MP into a Small Format DSLR return "the Medium Format look"? No, not to those that see a difference.
 

jerome_m

Member
There have been various comparisons made, which a google search would find out. In substance: the so called "MF look" is elusive. There are differences, but they are not big enough for MF users to recognize which is which at first sight in 100% of the cases.
 

malmac

Member
Does God exist?

Those who believe can't prove to those who don't believe that God exists.
Those who don't believe don't seem able to prove that God does not exist either.

Seems much the same re MF image quality.

When they bring out a 500MP iphone with proven sharper image quality than MF, I won't be buying one cause iphones just swipe (rub) me the wrong way.

The market place may then decide that the images I make are irrelevant and like a blacksmith of old I may have to change my profession but I will live with that when the time comes.

So if you use MF and like the overall package, then you buy in, if you don't then you buy into what you like - seems as simple as that to me.

No objective comparisons, just if you believe you believe, if you don't then you don't.


Mal
 

gerald.d

Well-known member
Does God exist?

Those who believe can't prove to those who don't believe that God exists.
Those who don't believe don't seem able to prove that God does not exist either.

Seems much the same re MF image quality.
It shouldn't be the same at all.

You don't need to prove the non-existence of something. Replace "God" with "Santa Claus" or "The Tooth Fairy" or "The Flying Spaghetti Monster".

Conversely, the difference between how a lens projects an image on varying sensor sizes is a simple matter of physics.

Kind regards,

A leprechaun.
 

Geoff

Well-known member
Folks,
All else being equal, I do not understand why the physical size of a sensor should make the slightest bit of difference to what is recorded. As far as I am aware a sensor crops from the image circle of the lens mounted in front of it. No more, no less.
Why are some books 500 pgs long? Seems they could say it all in a few pages. The themes are more or less the same, everything else being equal.
 

Shashin

Well-known member
All else being equal, I do not understand why the physical size of a sensor should make the slightest bit of difference to what is recorded. As far as I am aware a sensor crops from the image circle of the lens mounted in front of it. No more, no less.
Light is a frequency. Its amplitude (contrast) varies with frequency. If you can image something at lower frequency, it can be better--larger formats use lower frequency. This is what MTF is about.

BTW, cropping does other things like change DoF.
 

RVB

Member
The larger sensor increases spatial resolution,it is better at recording high frequency details and has better tonal graduation..
 

dougpeterson

Workshop Member
An image system is just that: a system.

Discussing any single element (like sensor size) is entirely fruitless.

Here is a partial list of factors influencing the technical rendering of an image:
Lens coating > Lens elements (number/design/spacing/composition) > Aperture (blade design, blade quantity) > internal body coating > microlens design > Anti aliasing filter (found in 35mm but not in medium format) > IR filter > sensor photo well > bayer pattern (spectral overlap of R, G, and B filters) > sensor read-out (heat-sinking and/or active cooling very important here) > cables to A/D converter > A/D converter > (read-out of black calibration file from sensor recorded as adjunct to the image) > debayering algorithm, deconvolution / detail finding algorithm, noise reduction based on black calibration file > noise reduction based on image data > sharpening.

The sensor size, in and of itself, is a factor in the final outcome of the image rendering. But the "look" of MF is always going to be a combination of all of the above factors.

In general medium format engineers are less constrained by how their design decisions increase the cost, slow down the operation, or increase the physical size, of the resulting product. They are also often targeting a market which cares deeply about the look/feel of the image, and not just it's technical specifications or price. I don't think it's any surprise that many people thereby find the results of their design work (high-end cameras) to produce a special looking image.

Apart from all that (technical) stuff you have the non-technical element - the human element - the impact on the photographer of using a system which is slower and requires more thought and for which they likely spent more time researching and saving up $$ for. In theory one can spend as much time considering composition/color/subject-selection/aperture/exposure etc etc with a point and shoot as with an 8x10 camera. In theory it should not matter if you spend $1 on a camera or $10,000 - you should put equal thought behind learning to use each to it's greatest capability. But people are not machines and the tool often influences the man. So the work produced by medium format shooters (or, for instance, Leica rangefinders) often feels deeper in subject matter, better thought out, and more impactful.

In my opinion anyone who tries to isolate any photograph, or one camera system, down to one specific technical element is missing the bigger picture - pun intended.
 

RVB

Member
"In order for a comparison between sensors of different physical size to be made on a like for like basis, depth of field would require to be matched. In addition, the same lens should be used."

I don't think it's really possible to make such an exact comparison between sensor's,A sensor is only part of an imaging chain.. or system,and the best comparison may not be a sensor to sensor but system to system comparison..
 

AreBee

Member
Jerome,

There are differences...
Would you care to mention them?

Mal,

Does God exist?

Those who believe can't prove to those who don't believe that God exists.
Those who don't believe don't seem able to prove that God does not exist either.

Seems much the same re MF image quality.

So if you use MF and like the overall package, then you buy in, if you don't then you buy into what you like - seems as simple as that to me.

No objective comparisons, just if you believe you believe, if you don't then you don't.
With all due respect I think an analogy between God and MF is a poor one.

Belief or non-belief in God does not carry with it a financial cost; with MFD it does, assuming one purchased on the basis of belief in a look. Nor do I consider it sensible to approach the purchase of MFD, as I am, from the point of view of "...if you believe you believe, if you don't then you don't...". I would be astonished if any person shooting MFD had not made an objective assessment of it prior to purchase in order to make an informed decision.

In starting this thread my hope was to determine if "the Medium Format look" exists in reality simply because it could eliminate one of the options I am considering - a Small Format camera. It was/is not to judge those who see a difference.

Georgio,

Think of Richard Avedon's use of a Deardorf to capture his American West images.
I am sorry to say that I do not understand the point you make. Please can you elaborate?
 

dougpeterson

Workshop Member
I don't think it's really possible to make such an exact comparison between sensor's,A sensor is only part of an imaging chain.. or system,and the best comparison may not be a sensor to sensor but system to system comparison..
Bingo. You don't take pictures with a sensor. You don't take pictures with a format size. You take pictures with a camera system.

It only makes sense to go take real world pictures with the various camera systems you are comparing and see which you prefer.
 

Shashin

Well-known member
If we setup a camera and lens and then just swapped the sensor and took a picture. We then took those images and made prints of the same size and put them next to each other, what would be different:

1. Different angle of view

2. Different DoF

3. Different apparent perspective

4. Different apparent contrast

Unfortunately, that does not actually show what the MF "look" is, if there is a clear look. All those show is what happens when you fix a bunch of variables, but that is not the way we use cameras.
 

Floyd Davidson

New member
If we setup a camera and lens and then just swapped the sensor and took a picture. We then took those images and made prints of the same size and put them next to each other, what would be different:

1. Different angle of view
True. But by selecting a lens focal length appropriate to the sensor size the exact same angle of view can be obtained. Therefore that cannot cause any character distinctive to MF.

2. Different DoF
Same as above. The exact same DoF can be obtained, therefore it is not significant to the character of MF.

3. Different apparent perspective
That is not actually true. Perspective depends on location, not focal length or sensor size.

4. Different apparent contrast
That is not true either. Two different sensors can have the same contrast even with different sizes. And one can have higher contrast than another that is larger or higher contrast than another that is smaller.

Unfortunately, that does not actually show what the MF "look" is, if there is a clear look. All those show is what happens when you fix a bunch of variables, but that is not the way we use cameras.
That is a valid analysis.
 

AreBee

Member
Geoff,

Why are some books 500 pgs long? Seems they could say it all in a few pages. The themes are more or less the same, everything else being equal.
Please can you provide me with an example where one reader reads content that another cannot see, or where one reader infers a meaning that cannot be resolved by the rules that govern language?

Will,

...cropping does other things like change DoF
I had assumed that DOF was matched between sensors in order to eliminate the former as a variable. Is this not possible because a change in aperture will change MTF, thereby negating a like for like comparison?

Doug,

Here is a partial list of factors influencing the technical rendering of an image:
Lens coating > Lens elements (number/design/spacing/composition) > Aperture (blade design, blade quantity) > internal body coating...
I concede that aperture would have to change in order to match DOF. The other parameters you mention would be constant in a comparison and therefore could not be attributed to a difference in look.

microlens design > Anti aliasing filter (found in 35mm but not in medium format) > IR filter > sensor photo well > bayer pattern (spectral overlap of R, G, and B filters) > sensor read-out (heat-sinking and/or active cooling very important here) > cables to A/D converter > A/D converter > (read-out of black calibration file from sensor recorded as adjunct to the image) > debayering algorithm, deconvolution / detail finding algorithm, noise reduction based on black calibration file > noise reduction based on image data > sharpening.
The above can all be dismissed given that "the Medium Format look" existed prior to digital cameras.

In general medium format engineers are less constrained by how their design decisions increase the cost, slow down the operation, or increase the physical size, of the resulting product. They are also often targeting a market which cares deeply about the look/feel of the image, and not just it's technical specifications or price.
I fully agree.

Apart from all that (technical) stuff you have the non-technical element - the human element...
Any comparison will necessarily require elimination of "the human element", given how notoriously unreliable humans are when it comes to objective assessment.

I infer from the same quote above that "the Medium Format look" is related in part to the time and care with which the photographer captures an image? This surely cannot be true.

Please can you clarify?

Rob

I don't think it's really possible to make such an exact comparison between sensor's,A sensor is only part of an imaging chain.. or system,and the best comparison may not be a sensor to sensor but system to system comparison...
I agree that a comparison made system to system would be less credible, and quite possibly meaningless.
 

Shashin

Well-known member
Will,



I had assumed that DOF was matched between sensors in order to eliminate the former as a variable. Is this not possible because a change in aperture will change MTF, thereby negating a like for like comparison?
So, if you change the f-number to match DoF, then you are changing diffraction at the image plane which also changes the numeric aperture (entrance pupil) in the object space. So then you don't have the same optical playing field.

And this really is the problem, no matter how many variables you line up, there are going to be some that get further apart.
 

AreBee

Member
Will,

...if you change the f-number to match DoF, then you are changing diffraction at the image plane which also changes the numeric aperture (entrance pupil) in the object space. So then you don't have the same optical playing field.

And this really is the problem, no matter how many variables you line up, there are going to be some that get further apart.
Damn. :D

Oh well. It was worth the discussion to me. Thank you all. :)
 
Top