6x6 and FF equivalent look

There was an old thread (about how expensive some medium format lenses were) that got me thinking about this exercise, but I just never got around to it. Till now.

 

From an equivalence point of view, a medium format 6x6 film image (56mm x 56mm) shot with a standard 80mm f/2.8 lens, is equivalent to a 35mm f/1.2 on a FF camera (same crop = 24mm x 24mm)

 

 

The film image was shot on a Carl Zeiss Biometar 80mm lens @ f/2.8 on expired Agfa Optima 400Y film

 

The FF was shot with a Rokinon 35mm lens @ f/1.2 on a FF digital, which had to be tweaked a little to match Agfa's lower saturation Optima film emulsion.

 

There are some colour differences that I did not bother to tweak out, and I did not bother with film speed equivalence - I was interested in the DoF & the look of Medium Format. There is also a different field curvature on the Rokinon, and there is contrast differences between the 50 year old single coated lens and the modern multi-coated one with many more elements. The Biometar also has more spherical aberrations wide open (again, expected).

 

The only other difference that you see are the bokeh, the Biometar has stronger outlining while the Rokinon is quite smooth. But that is a difference between the lens character.

 

You can see that the field-of-view and depth-of-field is equivalent.

 

So basically, the big fad for expensive fast lenses and shallow depth of field is to replicate the look of Medium Format on the current FF cameras. They just don't want to admit that.

 

(and then there is large format equivalence.... ;)

So basically, the big fad for expensive fast lenses and shallow depth of field is to replicate the look of Medium Format on the current FF cameras. They just don't want to admit that.

Who are 'they' Tom?

I don't think there's any big conspiracy here. Because even when film was the only choice, there were many, many photographers that chose to use 35mm over rollfilm. Despite, IMHO, its inability to deliver anything other than moderate image quality. But now digital sensors have changed all that.

 

On a pure depth-of-field based comparison: If we tidy up your square-format composition above, and lose the obtrusive and unnecessary bush to the right of the frame. Then we can turn our 24x36mm format upright and use it to better effect. The focal length needed then becomes a 50mm 'standard' lens, and we only need an aperture of f/2.2 or thereabouts to mimic the medium format shot.

 

The square format was only invented to allow constant waist-level viewing with a TLR or SLR camera; without the need for a rotating back. Not for any aesthetic reason. You only have to look at historic aspect-ratios to see that. All common plate sizes were rectangular, as are common printing paper sizes.

 

Admittedly the 24x36mm format is oddly over-long, and you can blame Oskar Barnack for that!

Edited August 18, 2020 by rodeo_joe|1

Who are 'they' Tom?

I don't think there's any big conspiracy here. Because even when film was the only choice, there were many, many photographers that chose to use 35mm over rollfilm. Despite, IMHO, its inability to deliver anything other than moderate image quality. But now digital sensors have changed all that.

No Conspiracy. My personal experience is, there's a lot of talk about fast lenses for that shallow depth of field look - with never any mention of the medium format look (which gets discussed on the MF boards). But that's probably because the current crops of photographers have never experience Medium Format.

 

The choice of 35mm was portability and cost, good enough pictures. Kind of like how smart phones cameras are good enough.

 

Back in the film days, there was talked about 35mm film catching up to MF, which was falsely self fulling as MF used the same emulsions. But digital FF has certainly caught up, and more.

 

If you crop the 6x6, then you are comparing 645 (or 6x4) with FF, which is a 1.5 crop. We use to call 645 the faux-medium format (kind of like APS-C sensors), but emulsions got better, the look did not.

 

If you want to look at equivalence, you really should use the same aspect ratio. I have no issues composing on square, it's my preference right now. (The square format was invented for Aerial Photography, it maxes out the sharp circle of the lens). If it's rectangles, one really should use 6x7 or 6x9 (thus maxing out both 120 and 135 film), the 100mm f/2.8 Planar lens on a 6x9 Tecknica 70 would be closer to a 45mm f/1.2 on FF.

 

The choice of 35mm was portability and cost, good enough pictures. Kind of like how smart phones cameras are good enough.

 

 

Don't forget versatility. There were soooo many more accessories available for 35mm systems. Not just lenses but fast motor drives, high capacity film backs, telescope and microscope mounts, etc.

 

And maybe it's doesn't have to be this way, but working with medium format in my opinion is a much slower and deliberate process. It takes me more time to focus for example. I'm not saying that's bad, but it does mean that there's some situations where I don't want to use a medium format camera.

 

Quality isn't just about the size of the film. An out of focus or missed shot is a bad outcome.

 

I say this as a fan of medium format film cameras. But I shoot about twice as much more 35mm film and 10 or 20 times as many digital images.

Edited August 18, 2020 by tomspielman

Absolutely agree with tomspielman:

 

"And maybe it's doesn't have to be this way, but working with medium format in my opinion is a much slower and deliberate process. It takes me more time to focus for example. I'm not saying that's bad, but it does mean that there's some situations where I don't want to use a medium format camera."

 

For me, MF is more enjoyable. I love the big view through the WLF (as much as I hate the tiny view through my first Nikon DX camera), and I love using the big negatives, If I did this for a living, I probably wouldn't use MF.

I think the slower and deliberate process had a lot to do with the cost ($ and time) per image, which was much less on 135 film. I shot the Pentax 645 (v1) back then, and it was as quick as my 35mm at that time (Nikon FM/FA - without winder). And the 67 was no slouch either.

 

Correct focus is always an issue - it was for the above example on both 6x6 and FF. Once you are in the shallow DoF range, format becomes irrelevant. Auto-focus/live view has really made a difference there.

 

I shoot MF film because I like the process. Certainly not for better quality - as shown here, FF digital is probably better.

We use to call 645 the faux-medium format....

Who's this 'we'?

56mm is 56mm, whether it's the long side of a 645 negative, or whichever side of a so-called 6x6 negative.

 

I used to shoot square format 120 cameras, until I realised that almost every print or composition was cropped rectangular. And customers expected 10x8 prints or thereabouts. Therefore I stopped wasting film and switched to 645, with absolutely no drop in quality - naturally.

 

The square format was invented for Aerial Photography, it maxes out the sharp circle of the lens

But that was using 6" x 6" plates or 5" film. Absolutely irrelevant to the ground-based pictorial photographer, who never even considered shooting, or composing into, a square format until Rollei and other TLRs came along, with their impossible-to-use-sideways viewfinders.

 

Same as nobody considered the oblong 2:3 aspect ratio until Barnack lashed together a quick prototype to use cine film.

 

Now that we have digital sensors that don't have sprocket-hole limitations, there's absolutely no need to stick with 24 x 36mm. A 27mm x 36mm (4:3 aspect ratio) sensor would make far more sense, both aesthetically and economically.

Who's this 'we'?

We'll, I'm not Royal, so it's the dictionary definition of "we". I recall an article in Modern or Popular Photography that contributed to the discussion among us.

 

For most of us, MF was about maximizing the negative size. 645 was about minimizing the cost per frame, which worked well in business when the emulsions got better. When we shot 645 for clients, it sometimes cost more, as you had to shoot a vertical and a horizontal.

 

Let's not go into where the 4:3 ratio originated, but the world would be very boring if everything had to be 4:3. I also like 6x17.

 

Cine started with 4:3, with the long side across the film, so that would be equivalent to 645 on 120 - and my 75mm f/2.8 on 645 would be equivalent to a 32mm f1.2 in cine (24mmx18mm). Still a fast lens, even for cine.

....but the world would be very boring if everything had to be 4:3.

I agree, but I would say that much of the medium format 'look' comes from using aspect ratios that aren't 3:2, and that 'Barnack's rectangle' immediately marks out a shot as coming from a 35mm camera or being digital.

my 75mm f/2.8 on 645 would be equivalent to a 32mm f1.2 in cine (24mmx18mm)

I think you're forgetting to reduce the C-o-C in proportion to the format. I make the depth of field much shallower @ f/1.2 on the cine format.

Edited August 20, 2020 by rodeo_joe|1

(snip)

 

The square format was only invented to allow constant waist-level viewing with a TLR or SLR camera; without the need for a rotating back. Not for any aesthetic reason. You only have to look at historic aspect-ratios to see that. All common plate sizes were rectangular, as are common printing paper sizes.

(snip)

 

I suppose so, but at least for amateur use, if you send the film out, you got back

square prints. If you print yourself, you can crop any way you want.

 

And 126 is also square, and labs made square prints from it.

I never knew any TLR or waist level finder for 126.

(snip)

 

Back in the film days, there was talked about 35mm film catching up to MF, which was falsely self fulling as MF used the same emulsions. But digital FF has certainly caught up, and more.

 

 

(snip)

 

If you consider shutter speed (camera and subject motion) and aperture for depth of field,

as is the subject of this discussion, then you need faster film with MF cameras,

with the same subject lighting.

 

It always seemed to me that 35mm came along when lens quality increased enough,

at the smaller sizes needed, to take equivalent pictures. (Compared to the simpler

cameras at the time, or even TLRs.)

 

MF, with the higher quality lenses came later, but then you need fast enough

film to use it in common situations. I suppose for studio work, you just get

bigger flash units.

(snip)

I think you're forgetting to reduce the C-o-C in proportion to the format. I make the depth of field much shallower @ f/1.2 on the cine format.

 

Not counting the complications of aspect ratio, it seems to agree with:

 

Depth of field - Wikipedia

 

I was once trying to figure this out, to figure out what film speed you needed,

keeping shutter speed and depth of field constant. You need faster film with MF

to match the change in aperture.

 

On the other hand, if you want shallow depth of field, and have fast film

in a 35mm camera, you might need a ND filter.

(snip)

I think you're forgetting to reduce the C-o-C in proportion to the format. I make the depth of field much shallower @ f/1.2 on the cine format.

Not counting the complications of aspect ratio, it seems to agree with:

 

Depth of field - Wikipedia

 

I was once trying to figure this out, to figure out what film speed you needed,

keeping shutter speed and depth of field constant. You need faster film with MF

to match the change in aperture.

 

On the other hand, if you want shallow depth of field, and have fast film

in a 35mm camera, you might need a ND filter.

 

Isn't the Circle of Confusion taken into account by the difference in magnification? Or did I get that wrong? Emulsion thickness / digital well depth are not scaled, so that would play a roll in the sharpness, but not the DoF significantly.

 

This is part if the reason for doing the exercise, the pictured examples could be used to compare 645 to cine 24x18mm, and the DoF appears to match.

 

I did not try to match film speed and exposure equivalence this exercise.

OT, but a while back there was a link here to images where "impossible" large aperture lenses were simulated by shooting a scene with a longer lens, taking it as a grid and stitching the images together. The result was shallow DOF that would be expensive or impossible to duplicate with a single fast lens.

I suspect that could be done with Lytro, without any stitching.

 

I believe with Lytro, you are supposed to be able to increase depth of field with its processing,

so you can probably decrease it, too.

 

I suspect that other types of processing could also be used.

OT, but a while back there was a link here to images where "impossible" large aperture lenses were simulated by shooting a scene with a longer lens, taking it as a grid and stitching the images together. The result was shallow DOF that would be expensive or impossible to duplicate with a single fast lens.

I think you may be referring to the Brenizer method. Not something I’ve ever dabbled with.

Yes, that was it. Surprisingly hard to look something up if you don't know the name!

Isn't the Circle of Confusion taken into account by the difference in magnification?

Sorry Tom. I slightly underestimated the cine C-o-C and overestimated the MF C-o-C when I plugged the numbers into my DoF spreadsheet. Your numbers do give pretty much the same DoF.

 

However, the C-o-C with projected cine film is a bit difficult to define, since the magnification and viewing distance can be so varied. It's totally unlike holding a small(ish) print at a comfortable distance.

 

Also, many DoF calculators use the format diagonal to compute the CoC, which is OK as long as the aspect ratio stays reasonably regular, but square and panoramic formats make the diagonal a poor comparator. For example: If you have a 10" x 10" print, and then crop 1" off each side to get a 10" x 8", does the print become less sharp in the process, and the DoF change? I think not. Or if you have a 10" x 25" panorama that you scan from end to end; what's the 'normal' viewing distance for that?

OT, but a while back there was a link here to images where "impossible" large aperture lenses were simulated by shooting a scene with a longer lens, taking it as a grid and stitching the images together. The result was shallow DOF that would be expensive or impossible to duplicate with a single fast lens.

The depth of field (DOF) is identical between two images with the same absolute magnification of the subject and relative aperture (f/stop), regardless of the frame size or focal length. This principle would apply to stitched images, except the "print" would be very large to maintain the magnification. In practice, that stitched image would appear to have a much larger DOF, if presented at a convenient size.

OT, but a while back there was a link here to images where "impossible" large aperture lenses were simulated by shooting a scene with a longer lens, taking it as a grid and stitching the images together.

That would be no different from using a larger format and a longer lens with a smaller aperture.

 

For example: My DoF calculator shows that using a 300mm lens @ f/5.6 on 10" x 8" gives roughly the same DoF as a 50mm f/0.95 lens on the 24x36mm format. At the same subject distance of 3 metres.

The depth of field (DOF) is identical between two images with the same absolute magnification of the subject and relative aperture (f/stop), regardless of the frame size or focal length.

I think that needs further clarification Ed, since my DoF calculator and personal experience would appear to differ.

Edited August 23, 2020 by rodeo_joe|1

FWIW- The Brenizer Effect With Fantastic Examples

I think that needs further clarification Ed, since my DoF calculator and personal experience would appear to differ.

A quick perusal of "The Camera," by Ansel Adams should clarify my statement. Remember the absolute size of the subject (magnification) depends only on the focal length and the distance. It might occupy more or less of the available film or sensor. In order to keep the subject size the same you must be further away the longer the lens. That cancels the effect of FL on DOF.

With the Bernizer photos, you can take something like a 135mm f1.8 lens, and shoot a 2-D panoramic so your 24x36mm sensor scans out an area the same as a 4x5 large format film. You then get a 4x5 image shot with a 135mm f1.8, which is something you cannot get on large format. Of course you could shoot at f/8 or f/16, and mimic a 4x5 image.

 

 

Re: Ed's comment about absolute magnification and relative aperture:

 

the OP sample has the same absolute magnification, and the aperture diameter is 28.6mm for the 80mm f/2.8 lens, and 29.2mm for the 35mm f/1.2 lens.

 

However, the DoF is fixed by the lens/aperture, printing big or small does not change it. If the DoF is from 2m to 3.5m, it stays that way regardless of how big you print. However, the apparent DoF, or look, changes if you cover a larger area, or crop to a small area - which is the same as smaller formats appearing to have a larger DoF for the same FoV. That's my understanding anyways.

Remember the absolute size of the subject (magnification) depends only on the focal length and the distance.

Ah, OK. You mean in-camera magnification, not final viewing magnification. Right.

 

Final viewing magnification makes it completely different.

In order to keep the subject size the same you must be further away the longer the lens. That cancels the effect of FL on DOF.

On closer scrutiny of my DoF calculator that's still not exactly true.

 

With f/2.8 set on both lenses, a 24mm lens @ 1metre distance gives almost 5mm more depth-of-field than a 240mm lens @ 10 metres. Admittedly not a huge difference, but a difference nonetheless.

 

The reason being that DoF depends on the effective aperture, and a unit-focussing lens changes effective aperture with distance. The closer you focus, the smaller the effective aperture becomes. So we'd have to open the marked aperture on the shorter lens to compensate. Something that's rarely done in practise.

 

Regardless of that; I fail to see the relevance of changing the subject distance, perspective and angle-of-view when discussing comparative DoF between two different sized formats. When the whole point is to keep the end result as similar as possible.

Edited August 24, 2020 by rodeo_joe|1