Main page
List of articles

The Charms and Myths of the Medium Format
How one should compare different formats?
___________________________________________________________

Now that digital technologies are getting more and more widespread and the good old 35mm photography is going to disappear, some old-fashioned photographers and/or beginners turn to the medium format. Some of them claim that it outperforms 35mm film even if the frames are not significantly enlarged. But is it really so? To answer the question, we have to learn how to compare different formats with each other correctly.

The conditions of fair comparison

Firstly, one must compare photo systems with equal angles of view. This means we should deal with larger focal distances when we switch from 35mm photography to the medium format. For example, a normal 50mm lens on a 35mm camera corresponds to an 80mm lens on a medium format camera.

The exact figure is not important here. For the purposes of this article, we may for example assume that the normal lens for a medium format camera should have a focal length of 100mm.

Secondly, the comparable photographs must have the same depth of field (DOF). In other words, the degree of fuzziness should be equal for the corresponding elements of the photos. Let us discuss this requirement in detail.

The formula that describes fuzziness in the picture can be found in my article that explains physical meaning of sharpness. That formula shows that the degree of fuzziness is proportional to f ² / N, where N = 1.4; 2; 2.8; 4, 5.6; 8; …

When we speak about final prints, we must allow for the enlargement. Thus, we may say that the fuzziness in the photograph is proportional to C_*:

C_* = (K f ²) / N = (K f) (f / N),

where
K stands for the enlargement factor;
f is the focal length of a lens;
N is f-number (1.4; 2; 2.8; 4, 5.6; 8; … ).

The unchanged depth of field means C_* is costant in our prints:

C_* = const.

This condition can be split up into the following two conditions (which must be simultaneously met):

A.   (K f) = const
B. (f / N) = const

Condition A is met almost automatically. On the one hand, when we switch from the 35mm photography to the medium format, we have to use lenses with larger focal distances to preserve angles of view. On the other hand, medium format negatives do not have to be enlarged that much to produce the final picture of the same size. As a result, (K f) is maintained constant.

Condition B means we must stop down our medium format lens to maintain the same degree of fuzziness in our prints.

Thus, if both conditions are met (A and B), the resulting fuzziness is maintained constant. If our lenses and films were ideal, the final photographic prints produced by different formats would be absolutely identical. But unfortunately there are no ideal lenses and films. When our prints become larger, the imperfections of any given combination of a lens plus film become more evident.

Both gurus and novices like to compare lenses and to talk about specific peculiarities of optical devices. However, mention should be made that all such peculiarities result from imperfections. Ideal lenses do not have any peculiarities at all.

CONCLUSION
Thus, if we switch from the 35mm photography to the medium format and:
— increase focal distances of our lenses to maintain constant angles of view; and
— stop down our lenses to maintain DOF constant (f/N = const); and
— print the final pictures of the same size (K f = const); and
— our lenses and film are ideal;
then the final pictures will be absolutely identical.
However, whereas the first three conditions are easy to observe, the last condition is always broken. All the differences between various optical devices are due to the absence of ideal lenses and films. The larger the final prints, the more obvious imperfections and peculiarities are.

BY THE WAY
When we talk about imperfections and peculiarities, it is more correct to talk about combinations “lens + f-stop + film” rather than simply about “lens + film”. For example, the conclusion above says that a medium format lens with f = 80mm @ f/6.7 should be compared to a lens with f = 50mm @ f/4 in the 35mm photography. However, almost any lens at f/6.7 outperforms another lens at f/4 in terms of sharpness and resolution. This fact may serve as yet another explanation why the medium format pictures may be better than their 35mm counterparts.

From theory to practice

Now let us check our conclusions with real life images.

First of all, let us have a look at two pictures. In this case magnification is not large, i.e. the images account for a large part of the film frame. Both pictures were made in compliance with the above mentioned requirements.

Fig. 1 a. 35mm photograph; f = 50mm @ f/8; b. medium format; f = 80mm @ f/13

On the whole, the theory works well. Both pictures (Fig. 1) are basically the same.

Of course, when saying that, we have to ignore some subtle differences. But they are not difficult to detect.

First, there are some experimental faults. The focal distances of the lenses were not exactly equal to 50mm and 80mm. Moreover, 50/8 = 6.25, while 80/13 = 6.15. All such factors should be taken into account, when comparing one picture to another.

Second, there is some additional fuzziness in Fig. 1b. It happened because the wind stirred some elements of the picture during exposure. The results can be seen only locally (mainly in the left lower corner). Although this effect is easy to detect, it does not prevent us from judging the picture as a whole. On the other hand, this extra fuzziness shows that the medium format has its own specific drawbacks. In this case, I had to increase exposure to 1/2 seconds, after the lens was stopped down to f/13. The camera was on a tripod, and I waited for the moment with minimal wind. Still, some twigs and leaves got stirred.

Now lets us look at another pair of pictures. In this case, magnification is large. Both images account for a small part of the initial frames (Fig. 2).

Fig. 2 a. 35mm photograph; f = 50mm @ f/8; b. medium format; f = 80mm @ f/13

Again we can see that the pictures are basically the same. However, some subtle differences became more obvious now.

CRT monitors make them more distinct than TFT monitors.

Well, what about more evident differences? To see them, we will need to analyze greater enlargements (Fig. 3).

Fig. 3. Enlargements of the central parts of the images shown in Fig. 2. a. 35mm photograph; f = 50mm @ f/8; b. medium format; f = 80mm @ f/13.

At such a high magnification, it becomes evident that the 35mm photography produces pictures with larger grain and lower resolution compared to the medium format photography.

Does the medium format justify our hopes? Is it substantially better? It depends. There is no simple and definite answer to this question.

Of course, the formats differ from each other in many aspects. Sometimes, such differences are important. Sometimes we may neglect them. However, to detect and analyze any differences we must stick to the rules mentioned in this article.

Conclusions

If our lenses and films were ideal, the different formats would produce the same result.

Thus, the advantages of the medium format (compared to the 35mm photography) are mainly as follows:

1. Imperfection of film is not that important.
2. The medium format allows photographers to get larger prints of a given quality.
3. The medium format frame is larger; therefore its potential for cropping is larger.
4. For any picture of an equal size, grain is less noticeable
5. Finally, medium format lenses may be characterized by their specific set of imperfections. Of course, all those imperfections may be aesthetically attractive in some cases.

In its turn, the 35mm photography also has an advantage. We will not have to stop down our lens that much for any given degree of fuzziness. As a result, our exposures will be shorter.

Links

1. Igor Yefremov. Understanding physical meaning of sharpness. Is Harold Merklinger's theory correct? (In this article, you may find a formula that describes fuzziness in photographs.)
2. Lars Kjellberg. 35 mm, medium format, or large format? (This article treats the subject matter in the same way. Actually, I wrote my text in order to supplement Lars▓ ideas with some math and real pictures.)

* * *

P.S.
If you are not happy with the criteria of this article, send me your arguments. Any illustrations are welcome. I will be glad to place them on my web site. Thank you.

Main page
List of articles

You must obtain a written permission from me to use any materials of this site for any commercial or non-commercial purposes, unless there is an explicit statement to the contrary.