Medium Format Magic

[Roberto Lequeux]

I am looking closer at medium format photographs and I am trying to make sure I am analyzing what I see, and feel, in an accurate way.

I am just goggling around and I trying to make sure I don't look at digital photos which may have a smaller sensor and I am noticing this completely different perspective. This is no surprise since the focal lengths most likely to be used are going to be longer than I am used to. However the FOV is wider... so...

Is this realistic feel from the longer focal lengths? From a wider FOV which approximates human FOV much more than 35mm? Could any of this be linked to the average distance between our eyes being similar to the sensor with?

I would love to get a good breakdown of the elements to consider and their effects, and a comparison with human vision where relevant.

Thank you.

 

[XiaoSu Han]

I love taking medium format photographs since for me it's a combination of very shallow DOF due to the increased focal length, but also due to the increase of negative density the color/gradients/sharpness is perceived differently, this also has to be taken into account I think.

6x4.5 - 6x9 http://www.flickr.com/search/?q=120&w=23333447@N00&s=int

The funny is that already by extending the negativ by a factor of two (hasselblad xpan) I am seeing a very apparent increase in definition and "depth"

http://www.flickr.com/search/?s=int&w=23333447@N00&q=xpan&m=text

to be investigated further?

 

[joakim_hansson]

The negative sizes 6x7 and up have a kind of 3d effect that you can´t get with 35mm photography.
I can´t put my finger on what it is, but it´s there. Surely some techfreak will step in now and contradict this. But I would recommend that you go to a bookstore and have a look at medium format and larger photography.
It has a kind of 3d effect that is truly appealing.

as an example - this is a portrait I shot in 2006 with even larger equipment. Large format 8inch by 10inch. It has not been digitally altered. Except for scan and © mark.
http://www.modelmayhem.com/pic.php?pic_id=44df18ef0a75c&date=2006-08-13%2007:19:31&id=2486&pid=1145590&the_count=&group_id=&ua=

 

[David Rasberry]

For me at least lenses in the 85mm-100mm fl range render depth perspective most naturally regardless of format. For a wide FOV at these lengths you need medium format.

 

[Matthias Hutter]

I haven't shot MF/LF myself, but judging from photos the most important difference to 35mm is the increased lens microcontrast that can be seen even on relatively small print sizes (A4).

 

[David Rasberry]

I haven't shot MF/LF myself, but judging from photos the most important difference to 35mm is the increased lens microcontrast that can be seen even on relatively small print sizes (A4).

Yes, if you shoot smaller formats, it takes proportionally higher resolution higher quality lenses to get comparable results.

 

[Daniel Browning]

...I am noticing this completely different perspective.

All format sizes are capable of the *exact same* perspective. Just keep subject distance the same and scale the focal length with format width. If the example images you found had different perspective, it was because that's how the photographer chose to shoot it, not because of some inherent advantage or limitation of the format.

This is no surprise since the focal lengths most likely to be used are going to be longer than I am used to. However the FOV is wider... so...

...so they cancel eachother out and there is no difference in perspective.

Is this realistic feel from the longer focal lengths?

You equate perspective with "realistic feel". Since 35mm can do any perspective that MF can (and more), they both have the same "realistic feel".

From a wider FOV which approximates human FOV much more than 35mm?

No, MF is not capable of wider FOV.

Could any of this be linked to the average distance between our eyes being similar to the sensor with?

No.

I would love to get a good breakdown of the elements to consider and their effects, and a comparison with human vision where relevant.

MF sometimes has higher contrast (MTF). It often has higher resolution. MFDB aren't doing quite as well in dynamic range (e.g. smashed by D3X). Of course MF gets blown away in low light (e.g. 5D2).

...very shallow DOF due to the increased focal length...

For any given angle of view, half of the time it's 35mm that has thinner DOF than MF. If compared to the "cropped" MFDB (i.e. not full 56 × 41.5mm), it's not even close.

The widely available f/1.2 and f/1.4 lenses from wide angle to short tele on 35mm is the reason. Even with typical bellows factors. For example, the widest-aperture 35mm lenses for 35mm:

15mm f/2.8
20mm f/1.8
24mm f/1.4
35mm f/1.4
50mm f/1.2
80mm f/1.2
135mm f/2.0
200mm f/2.0

Given the same horizontal angle of view, the 645 lenses have the same angle of view at 1.56X longer focal length. Here's the lenses you'd need on 645 to get the same horizontal angle of view and DOF:

23mm f/4.4
31mm f/2.8
37mm f/2.2
55mm f/2.2
78mm f/1.9
125mm f/1.9
210mm f/3.1
310mm f/3.1

At most of the focal lengths, MF does not have an f-number wide enough to get much thinner DOF than 35mm. Depending on the specific manufacturers under comparison, there may be some focal lengths where MF may have lenses with slightly thinner DOF. At the wider portrait focal lengths (wide, normal, and short tele), the DOF is noticably deeper on MF and thinner of 35mm. Generally, for any given field of view, the fastest MF lenses usually have only about the same DOF as 35mm. Here's how Mamiya stacks up, for example:

24mm f/4 (thinner)
35mm f/3.5 (deeper)
45mm f/2.8 (deeper)
55mm f/2.8 (deeper)
80mm f/1.9 (same)
120mm f/2.8 (deeper)
150mm f/2.8 (thinner)
200mm f/2.8 (thinner)
300mm f/5.6 (deeper)

If you crop the sides off the 35mm so it matches the 4:3 of MF, DOF gets a little closer, but the 35-50-80 focal lengths on 35mm still have thinner DOF. Another factor is lens price/aberrations/MTF; it's possible that a f/2.8 MF lens may provide fewer aberrations and higher MTF than a similarly priced f/1.8 35mm lens, even though both will have the same DOF; however, I would guess that economies of scale will favor 35mm.

So while there are many performance differences between MF and 35mm, I don't think thinner DOF is a big one.

The negative sizes 6x7 and up have a kind of 3d effect that you can´t get with 35mm photography.

It's just thin depth of field and contrast. (People laud the same "3d effect" when comparing 4/3rds to 35mm).

as an example - this is a portrait I shot in 2006

Looks like thin DOF (helped by a tilt perhaps) and high contrast.

For me at least lenses in the 85mm-100mm fl range render depth perspective most naturally regardless of format. For a wide FOV at these lengths you need medium format.

You are mistaken. Digicam with 7mm lens has the exact same perspective as medium format with 85mm.

I haven't shot MF/LF myself, but judging from photos the most important difference to 35mm is the increased lens microcontrast that can be seen even on relatively small print sizes (A4).

Agreed.

 

[joakim_hansson]

Yup the techstuff is probably true Daniel.
Strange though that it´s seldom seen in practice.
I think it´s because of the "magic"...

 

[Roberto Lequeux]

OK, so I am totally buying this micro contrast thing... I can see how it could make something seem more "real". It makes perfect sense and all of the "magic" may lie there.

Joakim, do you see it the same way? If there was a more accurate representation of contrast the way we can see it then perhaps it isn't perspective, but this form of "higher resolution" by means of higher accuracy.

Did I grasp the concept correctly Daniel? And will we see higher micro contrast if we shoot with a bigger sensor area? I.E.: Same lens shooting with Epic-X in 5k vs. 4k?

 

[Daniel Browning]

Any sufficiently advanced MTF is indistinguishable from magic. --Arthur C. Clarke [paraphrased ]

Did I grasp the concept correctly Daniel?

Yes.

And will we see higher micro contrast if we shoot with a bigger sensor area? I.E.: Same lens shooting with Epic-X in 5k vs. 4k?

Generally, yes. Sometimes the microcontrast will be the same (e.g. both will have 80% MTF), but usually the MTF drops as sensor size shrinks. Usually this is because of aberrations, but even a perfect lens will lose MTF due to diffraction.

 

[D Fuller]

OK, I'm going to go out on a limb here, but I've been thinking about this for a while now, and I agree with Robert that there is a difference in the way a medium format lens draws an image that is not just about the difference in focal lengths for a given FOV. And I don't think it is only about depth of field.

My Contax 645 seems to draw images differently than my 35mm Nikon, or my D300. I've thought it might be partially the fact that the light path hitting the film plane is physically wider on the 645 than it is on the Nikon sensor, and that there seems to be a subtle difference that comes from that dimensional difference.

I can get the same microcontrast with Zeiss lenses on my Nikon, but I think there may be something to the idea that a wider light path draws a slightly more dimensional-feeling image.

 

[Roberto Lequeux]

OK, I'm going to go out on a limb here, but I've been thinking about this for a while now, and I agree with Robert that there is a difference in the way a medium format lens draws an image that is not just about the difference in focal lengths for a given FOV. And I don't think it is only about depth of field.

My Contax 645 seems to draw images differently than my 35mm Nikon, or my D300. I've thought it might be partially the fact that the light path hitting the film plane is physically wider on the 645 than it is on the Nikon sensor, and that there seems to be a subtle difference that comes from that dimensional difference.

I can get the same microcontrast with Zeiss lenses on my Nikon, but I think there may be something to the idea that a wider light path draws a slightly more dimensional-feeling image.

Yes, that is what I was thinking about pretty much. It also clarifies one thing for me which is the idea that I had that you are in fact capturing a larger "chuck" of light. Hhehe... feel free to put it in better terms. And it would very much go in line with the idea of a "3D feel" that does describe quite well what I get from them. The way I was thinking about it was that the bigger the sensor the more you should be able to see "around" an object. And it would make a lot of sense if this were closer to what we see naturally than one single 35mm sensor/negative.

Does that make any sense to you Daniel?

 

[Daniel Browning]

...there is a difference in the way a medium format lens draws an image...And I don't think it is only about depth of field.

Do you have an examples? I've seen 35mm and MFDB comparisons (can't find them now) that look very similar.

Are you sure it's not thinner DOF? Or higher contrast? More resolution? Better tonal gradations? Less grain?

I think DOF and contrast account for most of it. (On MFDB, the contrast is even higher due to the missing OLPF.)

 

[David Doko]

Circles of confusion for a given focal length do change as format size changes. Circles of confusion affect how one perceives whats out of focus and in focus. I'm not an optical physicist, and perhaps Daniel is, but my hunch is that what distinguishes medium format and its longer lenses has to do with that perception of "creaminess" or transitions as our eyes process those different areas of an image. Is that the magic we're talking about?

Regardless, everyone one reading this should go and get a medium format camera. The film ones are so cheap, it makes me ill that I spent so much on mine.

 

[Roberto Lequeux]

Circles of confusion for a given focal length do change as format size changes. Circles of confusion affect how one perceives whats out of focus and in focus. I'm not an optical physicist, and perhaps Daniel is, but my hunch is that what distinguishes medium format and its longer lenses has to do with that perception of "creaminess" or transitions as our eyes process those different areas of an image. Is that the magic we're talking about?

OK, so the transition from in focus to out of focus would be different? One Medium Format negative/sensor would be much closer to the total active area inside a pair of eyes... however if the DoF is far shallower or deeper than the limits of our eyes then I guess that "magic" may become less pronounced over a 35mm photo with the same DoF.

I would love to see a 35mm MF comparison with matching perspective that showcases the transition. Does anyone know where to find one?

XiaoSu Han, if I may, here are two of your beautiful pictures, which should give a good gauge for the in/out of focus transition:

 

[Daniel Browning]

Circles of confusion for a given focal length do change as format size changes.

Agreed.

Circles of confusion affect how one perceives whats out of focus and in focus.

Agreed.

But it's important to note that circles of confusion also change with f-number. Any two formats that have the same composition (angle of view, subject distance) and the same iris diameter (narrower f-number on larger formats), they will have the exact same circle of confusion at any part of the image. That means they have the same DOF and same background blur. (There is an exception for macro focus distances due to bellows factor.) They also have the same diffraction, FWIW.

So if you shoot MF with thinner DOF, you will get a different CoC and different transitions between in and out of focus. But if you stop down so that DOF is the same, then CoC and transitions, too, are the same.

Of course, every lens has different design, bokeh, aberrations, etc. So OOF transitions vary greatly for that reason, but that's true among lenses of a single format as well as between lenses of different formats.

For example, it could be that MF lenses tend to have purposefully undercorrected spherical aberration more often, which would give them nicer bokeh, and therefore nicer transitions in OOF, but that's not because of the format size, since someone could do the same for 35mm lenses.

Given the many experienced photographers that love their MF lenses and even use them on 35mm cameras, I'd guess that at least part of the magic is from the design of the lens themselves (separate from format size).

I'm not an optical physicist, and perhaps Daniel is,

No, I'm a software engineer by day; motion and still photographer by night.

...my hunch is that what distinguishes medium format and its longer lenses has to do with that perception of "creaminess" or transitions as our eyes process those different areas of an image. Is that the magic we're talking about?

I'd count those differences as DOF, since they go away when you stop down to the same DOF.

Keep in mind that all this assumes you adjust focal length to get the same field of view. If instead you keep focal length the same and physically move the camera, then all bets are off. Changing subject distance would result in a different perspective, and, depending on how close the focus distance is to hyperfocal, a difference in DOF and background blur.

 

[joakim_hansson]

Joakim, do you see it the same way? If there was a more accurate representation of contrast the way we can see it then perhaps it isn't perspective, but this form of "higher resolution" by means of higher accuracy.

Yeah I guess it´s because of that, but magic sounds more pleasing

 

[christopher witzke]

Larger format vs. smaller format is like fake tits vs. real..... they will both look the same if you present them in the same way.

I could shoot an identical scene with both my 8X10 and FF35... shot hyperfocal... post them in this thread... and you would not be able to pick which is which.

Let me remind you that a focal point is where the point at which initially collimated rays of light meet after passing through a convex lens.... it's at this point that format size becomes irrelevant.

The only advantage of larger formats is to get higher resolutions of film/sensor and optics....

 

[D Fuller]

Agreed.



Of course, every lens has different design, bokeh, aberrations, etc. So OOF transitions vary greatly for that reason, but that's true among lenses of a single format as well as between lenses of different formats.

For example, it could be that MF lenses tend to have purposefully undercorrected spherical aberration more often, which would give them nicer bokeh, and therefore nicer transitions in OOF, but that's not because of the format size, since someone could do the same for 35mm lenses.

Given the many experienced photographers that love their MF lenses and even use them on 35mm cameras, I'd guess that at least part of the magic is from the design of the lens themselves (separate from format size).


Keep in mind that all this assumes you adjust focal length to get the same field of view. If instead you keep focal length the same and physically move the camera, then all bets are off. Changing subject distance would result in a different perspective, and, depending on how close the focus distance is to hyperfocal, a difference in DOF and background blur.

This is what makes these things so hard to test, of course. I don't have the time or the inclination, really to try to figure out an objective test between MF and 35mm (I'd have to do it on film, as I don't have a digi back for the Contax) but even if I did, the differences in the lenses you refer to would overwhelm any possibility of objectivity. Too many variables. It may just be that medium format lenses are better because people are willing to pay more for them. There's not one of my Contax lenses that didn't cost twice as much as a Nikon equivalent in the Zeiss ZF line.

In an extreme example, do you think Ansel Adams' work could have had the same look if done on a D3X? Or is the particular look of his images in part the result of the large format optics? He tended to work at fairly small f-stops, so DOF played a pretty small role.

I tend to think that there is more to the optical characteristics of different formats' lenses than just DOF, but i really have only the sketchiest optical knowledge to explain it.

Maybe it's just magic.

 

[Matthias Hutter]

I think DOF and contrast account for most of it. (On MFDB, the contrast is even higher due to the missing OLPF.)

Agreed.
A similar (though smaller) difference can be seen when comparing SS35 with FF35. I haven't found real world side by side comparisons yet,
but I think the DPReview Lens Widget may help. (For the same DOF, the FF should have 1.6x the F-number, like F4/F2.5)

http://www.dpreview.com/lensreviews...sreviews/widget/LensReviewConfiguration.xml?4

Edit: EOS 5D@ 130mm F11 vs 20D@ 80mm F11

Source: http://www.bobatkins.com/photography/digital/full_frame_vs_aps-c.html
http://www.bobatkins.com/photography/digital/size_matters.html