Medium format look – a topic debated at nauseam but now with a visual twist

What follows is a pre-amble to a visual test about of how to demonstrate what “the medium format look” means to me (YMMV).

A while ago I expressed my view about my view of what creates the "medium format look" or what makes the MF format different from samller sensor formats which got lost in the ensuing shuffle and one-upmanship on this board.

Here goes again:

“The MF look is in the interplay between sensor size (or film size), focal length, FOV and DOF. For a given focal length, a larger sensor delivers a wider FOV which allows you to move in closer to your subject (for the same framing as an FF would give you), which gives you a narrower DOF, but above all stronger focus fall-off or background blur. The bigger the sensor (or film), the stronger the effect. The Fuji sensor size cannot deliver the same degree of effect as say a 6x7 film camera for that reason. It is physics. If you compare to FF, the effect is there, but might be too small to notice.”

Let’s break this statement in pieces:

• What is my definition of the MF look?. I shoot mostly portraits, and what matters to me is the way I can play with focus and focus fall-off on the subject and how it is isolated against the background. In my experience, there is an observable difference between sensor/film formats. A visually easy reference point in this regards is the work of Martin Schoeller (there are others of course). As far as I understand it from the BTS videos that I have watched, he shoots his close-in portraits on a Mamiya 6x7 (which doesn’t prove that this is the only way you can get this look).

• From a AOV / DOF perspective, it is easy to calculate the equivalent focal lengths and f/stops required to obtain the same values between different sensor / film formats and lenses. No argument there, it’s just simple math. In my experience, in particular for flat-ish subjects at medium or farther distances, it is hard to distinguish between camera formats as many people on this forum have argued and shown.

• Besides AOV and DOF, there is a third dimension, which is the focus fall-off, which was the point I was trying to make as another dimension that distinguishes larger sensors from smaller ones, which got lost. The interplay between AOV / DOF and focus fall off is different between formats because the subject distance is different and the focal length is different. So on the “in-focus” area of the subject it makes no difference, but you can tell there is a difference by looking outside of the area in-focus. This effect is particularly strong at shorter subject distances, and far away backgrounds. This point can also be made mathematically by calculating blur at a distance from the subject in focus. Again, it’s math, not much to argue.

• My statement does not consider bit-depth, pixel size etc, which admittedly also play a role, as pointed out by a number of others

Link to the websites I used for the calculations that accompany my test:

Depth of Field (DoF), Angle of View, and Equivalent Lens Calculator • Points in Focus Photography

depth of field - Does amount of background blur change with focal length given equal framing? - Photography Stack Exchange

Rather than diving into the debate again with verbal arguments, it seemed easier to illustrate the difference with a simple experiment. Simply put; if there is such a thing as the MF look as I expressed it, you should be able to see it in the experiment. If it is so subtle that it isn’t visible, it is not worth debating. The image at the bottom of this post is one of the image comparisons of the test described below.

The test has been designed to show that there are circumstances in which it is visible. Whether these circumstances are relevant to the readers’ photography is another matter.

The comparison I made is between an X-Pro 2 (APS-C) and an GFX 50s (Medium-ish format). I chose to use a Canon 40mm STM and a Canon 70-300L on adapters for the experiment. These budget-ish lenses are not stellar performers by any means, nor do they have extreme wide-open aperture values. It eliminates a discussion about lens manufacturers, lens designs, coatings etc. It is also happens to be the gear that i can mount on both cameras that I have at-hand.

I am very aware that there are a number of shortcomings in this experiment, it does not stand-up to scientific scrutiny and is not intended to do so. For instance, the test does not consider Mpix count, bit depth, edge performance and other aspects that have been advanced as advantages of MF. It shows only what I would call the geometric aspects of comparing the formats. Since the cameras and sensors are dated, it does not consider the progress in technology since made in particular with FF cameras. It therefore does not answer the question whether for instance an R5 is a "better" camera than a GFX50s even for this particular use-case. I would be interested to seeing the comparison in a similar set-up.

The results

Since there are more images to look at than is practical to post in this thread, I have put the results in a pdf (with the Mod's permission). The document also contains the calculations for each comparison.

https://1drv.ms/b/s!AtJxl9Z1DHNsjNER3tW4f6Z90IWudA?e=oU4Skk

Predictably, the test showed:

1. The larger GFX sensor allows you to get closer to the subject (at the same focal length and same FOV), which leads to a more blurred background (test 3 and 4), and by extension what I call faster focus fall-off
2. At the same AOV and DOF settings (meaning the perspective on the subject and the DOF on the subject are the same), the background blur is stronger with the larger sensor, and by extension the focus fall-off. (test 5 and 6)

Whether this is enough to justify the additional cost and inconvenience of owning a MF system is an entirely different matter

Looking forward to your comments.