47mm Schneider not wide enough

[michael waldstein]

I recently rented a 47 mm Schneider (from Calumet) lens to try out. If 4x5 lenses are compared to 35mm at 3 to 1 then the 47mm should be equal to a 15 or 16mm 35mm lens. When I took a shot that I had previously taken with an 18mm Nikkor, I only got about 2/3 the coverage. Could someone please explain this to me? ( I couldn't get any help from the Calumet salesman because I wasn't 'buying a lens from them') Thanks, Michael Waldstein

[Lazybones]

Is it due to the differing aspect ratios of the formats? 4x5 is a "square" rectangle, having an aspect ratio of 2.5:2. 35mm is a more elongated rectangle with an aspect ratio of 3:2. As you can see, 4x5 "looses" some image area on the side(s) (in "landscape" orientation) that would be included in the 35mm ratio. Think of 35mm as being more "panoramic" as compared to 4x5.

[Roger Krueger]

That makes no sense whatsoever--the 47mm on 4x5 is a 12.3mm equivalent based on diagonal, 13.3mm equivalent based on long dimension only (ie cropped to a 3:2 format).

The only things I can think of are:

1) You were shooting on rollfilm, the 47 on 6x9 is only a 20mm equivalent, even longer on smaller rollfilm. 6x7 would give you about 2/3 of your 18mm-on-35 image.

2) Someone gave you the wrong lens and you didn't notice. A 58 or 65 would still be wider than your 18, given your 2/3 coverage a 90 would be the most likely suspect.

3) Maybe the back group was missing from the lens? I know at least some lenses will still work, albeit at modest quality, with an element missing, and they'll usually be longer than a complete lens. Seems extraordinarily unlikely, but I suppose not impossible.

4) There could be some discrepancy from an internal-focusing 35mm lens maintaining its field of view at close focus, while a unit-focusing LF lens will have its field-of-view shrink somewhat. But you'd have to be very near 1:1 to generate the magnitude of difference you report, and there's no way the Nikkor focuses that close by itself.

[Patrik Roseen]

This is surely something I have been thinking about and not really understood.

According to the Schneider web site the angle of view is 105 degrees, same as the 65mm/5.6 75mm/5.6 and 90mm/5.6. The difference seems to be in the image circle getting larger with longer focal length.
For the 47mm it is 123mm (really not covering 4x5" requiring at least 150mm) and for the 90mm it is 235mm (i.e. almost double that of the 47mm).
The image will be more spread out on the filmplane as the focal length becomes longer given the same angle of view.

Looking at the 18mm nikkor ...what is the angle of view for this lens...is it more than 105 degrees?

Someone please explain?

[Dan Fromm]

Patrik, a lens projects a cone of rays. The angle at the cone's apex is the lens' angle of view. The circle defined by the intersection of a plane one focal length behind the lens' apex with the cone is the circle the lens illuminates.

If one considers only the part of the circle illuminated that contains a "sharp enough" image, then the lens' angle of view with good image quality is 2 arctan (that circle's radius/focal length).

So for the same angle of view, be it for the circle illuminated or the smaller circle with good enough image quality, the longer the focal length the larger the circle.

Cheers,

Dan

[Dan Fromm]

I recently rented a 47 mm Schneider (from Calumet) lens to try out. If 4x5 lenses are compared to 35mm at 3 to 1 then the 47mm should be equal to a 15 or 16mm 35mm lens. When I took a shot that I had previously taken with an 18mm Nikkor, I only got about 2/3 the coverage. Could someone please explain this to me? ( I couldn't get any help from the Calumet salesman because I wasn't 'buying a lens from them') Thanks, Michael WaldsteinWhich 47 SA did you rent? SA or SA-XL?

[Ole Tjugen]

Which 47 SA did you rent? SA or SA-XL?

Also what Nikkor did you use? There's a big difference between a 18mm rectilinear WA lens and a 18mm fisheye, even if the focal length is the same.

[Leonard Evens]

I think Patrik and Dan may have it. The SA is designed for 6 x 9, not for 4 x 5 and only has an image circle of 123 mm. The circle of coverage is that in which one gets sufficiently high resolution. There is also the circle of illumination, outside which there is essentially no image. That is presumably also somewhat less than 153 mm, the diagonal of a 4 x 5 frame. So, if what Michael is observing is that the image doesn't fill the whole frame, that might explain it. On the other hand his remark that he is only getting 2/3 the coverage he expects from an 18 mm lens used with 35 mm format still doesn't make sense. If you compare short dimensions, the ratio of forrmat sizes is about 4, you compare diagonals, it is about 3.6, amd if you compare long dimensions it is about 3.3, In no case is it 3. The 4 x 5 focal lengths which might be considered comparable to 18 mm are thereby 72 mm, 65 mm, and 60 mm. Thus, no matter how you calculate it, a 47 mm with 4 x 5 has a larger angle of view than an 18 mm lens with 35 mm format. Even if the conjecture of coverage is correct, his report still doesn't make sense. Assuming he is using the SA, its image circle is 123 mm, which is already 80 percent of the diagonal of the 4 x 5 frame. The circle of illumination is presumably somewhat larger, certainly noticeably larger than 2/3 the frame.

The SA-Xl, on the other hand, is a 4 x 5 lens, and does cover the 4 x 5 frame, albeit with little to spare.

I don't understand Patrik's remark about the Schneider website. The angle of view, as distinguished from the angle of coverage, depends on both the focal length and the format. It is given by

2 x arctan(half-diagonal_of_format/focal_length).

For the same format size, the angle of view goes down as the focal length goes up. So a 47, 65, 75, and 90 mm lenses used with 4 x 5 format don't the same angle of view. The approximate angles of view are: 47 mm, 116 deg; 65 mm, 99 deg; 75 mm, 91 deg; 90 mm, 80 deg. On the other hand, it is possible Patrik is reporting the angle of coverage. It is quite possible that the angles of coverage are the same, particularly if the lenses have the same basic design, but only differ in focal length. Dan's explanation for why the longer focal length lens has a larger circle of coverage is correct.

It should be noted that all of this assumes the lens is focused at infinity. If the lens is focused at a finite distance, then you have to replace the focal length in the formulas by the lens to film distance, which is always greater than the focal length. That means the angle of view is smaller. For example at magnification 1:1, a 90 mm lens used with 4 x 5 format would have an angle of view of only about 46 degrees, compared to 80 degrees at infinity. On the other hand, the angle of coverage may be pretty much unaffected by where you focus, so the circle of coverage would be larger when focusing close-up. In most ordinary photography, when you are outside the close-up range, the lens to film distance is not significantly larger than the focal length, so for all practical purposes, the figures at infinity apply.

[Patrik Roseen]

...don't understand Patrik's remark about the Schneider website. The angle of view, as distinguished from the angle of coverage, depends on both the focal length and the format. It is given by

2 x arctan(half-diagonal_of_format/focal_length).

For the same format size, the angle of view goes down as the focal length goes up. So a 47, 65, 75, and 90 mm lenses used with 4 x 5 format don't the same angle of view. The approximate angles of view are: 47 mm, 116 deg; 65 mm, 99 deg; 75 mm, 91 deg; 90 mm, 80 deg. On the other hand, it is possible Patrik is reporting the angle of coverage. It is quite possible that the angles of coverage are the same, particularly if the lenses have the same basic design, but only differ in focal length. Dan's explanation for why the longer focal length lens has a larger circle of coverage is correct...

Leonard, I was looking into the page below and it talks about the angle of view...maybe it is talking about angle of coverage but in that case the schneider site is using the wrong terminology. Sorry if I am adding to the confusion but that is what I read.

http://www.schneideroptics.com/info/vintage_lens_data/large_format_lenses/super-angulon/data/5.6-47mm.html

[Oren Grad]

Leonard, I was looking into the page below and it talks about the angle of view...maybe it is talking about angle of coverage but in that case the schneider site is using the wrong terminology. Sorry if I am adding to the confusion but that is what I read.

Patrik - Schneider's use of the term "angle of view" is very unfortunate, and is a source of much confusion. Rodenstock uses the term "image field" instead, which is perhaps a bit clearer. But the 105 degree figure indeed refers to the cone projected by the lens inside the camera, not the breadth of the scene that is included in the picture.

[michael waldstein]

OK, I'll be more specific. I was actually using a Nikkor 12 to 24mm lens on a DX body. I shot at 12mm, so that's equivalent to 18mm on a 35mm body. I was set up across the street from a building that I needed to get all in the shot. No problem with this lens.
The Schneider was an XL 47mm on a 4x5 Cambo body. I shot from the same position and could get in only about 2/3's of the same scene.

[Leonard Evens]

I agree that it is clear that Schneider is using the term in a confusing matter. They are talking about the angle of coverage, not what is normally called, in English, the angle of view. My Horseman 980 film holder has an opening with diagonal just about 100 mm. Using that, the angle of view is about 94 degrees. With a 6 x 7 film holder, it would be even smaller. Perhaps the German term makes more sense, and it was just a matter of imperfect translation.

[GPS]

Michael, why do you think that your 12mm Nikon on DX is the equivalent of 18mm on a 35mm body? (I wouldn't take a seller's word for granted in this matter). I suspect there is the cause of your problem...

[Dan Fromm]

Hmm.

If I got the calculations right, a 47 mm lens will see 90.6 degrees vertically and 103.9 degrees horizontally on a piece of film 95 x 120 mm. On a chip 15.7 x 23.7 mm, a 12 mm lens will see 66.4 degrees v and 88.8 h.

Something ain't right here, and it may not be my arithmetic.

[Lazybones]

...So 4x5 and 35mm have the same aspect ratio? Well, I'll be!

[Ole Tjugen]

Geometric distortion is of the barrel type at 24 mm and is visible there, but not occurring to a detrimentral degree. As you zoom the lens to wider focal lengths, distortion is reduced and around 18-20 mm you have a lens suitable even for architecture work. Towards shorter settings, the geometric distortion, predictably, picks up and a barrel-type of distortion is plainy visible at 12 mm. There must be higher-order components in the geometrical rendition too, because the lines curve much more towards the extreme corners. You should not consider shooting architecture with this lens set at 12 mm.

Reference: http://www.naturfotograf.com/AFS12-24DX_rev00.html

All right - that lens has considerable barrel distortion at the wide end. That means that the field of view at 12mm is considerably wider than a rectilinear WA 12mm lens would give on the same size film/sensor.

The reason I asked is that two of my most used 35mm lenses are a 18mm Zenitar fisheye and a 21mm Color-Skopar. The fisheye gives 180 degrees from corner to corner thanks to the whopping great big barrel distortion that is a "design feature" of these lenses. The 21mm Color-Skopar is very very wide, but also very far from 180 degrees!

So in essence you are comparing the coverage of a lens which is going a long way towards a fisheye (which is really only the extreme limit of barrel distortion) with an absolutely rectilinear LF lens. A little like comparing apples and bananas - both fruit, so how do I peel this apple?

[GPS]

Hmm.

If I got the calculations right, a 47 mm lens will see 90.6 degrees vertically and 103.9 degrees horizontally on a piece of film 95 x 120 mm. On a chip 15.7 x 23.7 mm, a 12 mm lens will see 66.4 degrees v and 88.8 h.

Something ain't right here, and it may not be my arithmetic.

And 18 mm on 35 mm film gives 90° horizontally and 67° vertically - still less than the 47mm on 4x5...

[Mik Wenger]

Assuming that the sensor on a Nikon DX camera is 23.7 mm wide, then at 12 mm focal length the angle across the width is 89 degrees.

Assuming that the film on a 4x5 Cambo is 120 mm wide, then at 47 mm focal lenght the angle across the width is 104 degrees.

So Michael, I simply do not believe your story :-).

A 47 XL on 4x5 is the widest LF you can get btw.

[Ole Tjugen]

A 6cm Hypergon is wider, but only on 8x10" film. And no, there were no shorter Hypergons made.

[Gordon Moat]

Using my pCAM software on my PalmPilot, the only way I could get the 47mm to only be 2/3 the image from a 12mm DX Nikkor was when comparing a 6x7 or 6x9 film area, certainly not 4x5. What might be of some interest from Ole's comments and link to Bjørn Rørslett, concerning the extreme distortion at 12mm was the statement: You should not consider shooting architecture with this lens set at 12 mm. However, even with distortion, this lens does not behave like a fisheye 12mm.

Using pCAM software again, a 2/3 image would work out to mean about 9mm to 10mm on standard 35mm frame or closer to 7.5mm on a DX frame camera body. Assuming 10m distance across the street, the 47mm lens should show an image area encompassing 21.5m by 26.9m. At the same distance of 10m, an 18mm on 35mm would show an area of 13.3m by 20m, while a 12mm on DX would show an area of about 13.9m by 25.1m. You could almost do a vertical shot with the 47mm on 4x5 and compare it to a horizontal shot using 35mm or that DX body. Maybe the OP was shooting landscape mode with the smaller camera, and did not rotate the back on the 4x5 camera?

Just to add more information, at that 10m distance, the 47mm on 6x7 would show an area of 12.7m by 14.8m. Using 6x9 holder, the area shown would be 12.7m by 19.1m. Using a newer digital back on that Cambo with the same lens would show less than about 9m by 12m of the same scene. Putting a digital medium format back onto a 4x5 camera means it is no longer a 4x5 camera, and actually much closer to cropped 645 format (actually near double 35mm size).

So hopefully the OP will chime in here and solve this puzzle. Either he had the 4x5 back in a different orientation, was using a roll film back, was using a Polaroid pack film back, or was using a digital medium format back (Leaf, Sinar, PhaseOne, et al). The other possibility, as others mentioned, is that Calumet did not hand him a proper 47mm lens.

Ciao!

Gordon Moat
A G Studio (http://www.allgstudio.com)

[GPS]

Then - out of the 4x5 condition - even a 38mm Schneider is wider on 6x12 cm format. At least horizontally and diagonally...

[michael waldstein]

I think 'Lazybones' has the most likely answer. The aspect ratio is different.
Thanks