Large Format equivalent to hasselblad SWC lens

[M. Axel Wikstrom]

Wide angle lenses for large format should be rectilinear, and should make objects at the edges of the image circle appear larger (actually all focal lengths do this, but with longer lenses it's not as noticeable).

This is a feature. This is how perspective control works. The cause for the difference in size is projection distance between the center of the image circle and the edges. Distance at the center is less than distance away from center. The classic tall building photo with front standard rise puts the top of the building closer to the edge of the image circle - magnifying it's size. It works because the lens is rectilinear - keeping straight lines straight, and eventually parallel with the proper amount of rise. It also works because the lenses are designed to project a flat plane of focus. Perspective control!

 

[DREW WILEY]

Yeah, a slip there, Dan. Thank you. I was contemplating my own lens choices this afternoon, planning for a trip which potentially includes a 90SW, and apparently crossed wires.

 

[DREW WILEY]

OK. Mr. Wikstom. Now please take that observation and extrapolate it to the fact that the wider the angle relative to focal length, the greater the "stretch" distortion will be. That's why the Hassie SW lens was designed to "just fit" the 6x6 format, unlike a view camera with perspective control. Being quite short for its given format, it still looks like a wide angle shot, but
does not warp the scene to the same degree. But for the same reason, if I can employ the smaller images circle of, say, my Fuji 125 W on 4x5 film, with its modest amount of excess coverage, I know there will be dramatically less stretch toward the corners than a 120 Super-Angulon with its huge image circle. In photogrammetric applications, the lens is fixed, so such distortions can be minimized. There are of course other lens design factors. But there's no generic formula via merely focal length. Not all wides are the same by any means. I'm not an optical engineer, but I do know darn well from actual applications that a "W" of a given focal length stretch-distorts lens than an "SW", and certainly less than an "XX-W", as if we were dealing with clothing sizes. And this is a completely different issue from pincushion vs barrel distortion. The example typically given of a building with its straight sides doesn't really illustrate the problem at hand very well. Include some round street lights in the corners of those same images and all of a sudden they're stretched into ovals. Peoples heads are warped, as is getting sucked into space toward the corners, like all those wide-angle snapshots you see in National Geographic and so forth. That's a kind of distortion.

 

[M. Axel Wikstrom]

OK. Mr. Wikstom. Now please take that observation and extrapolate it to the fact that the wider the angle relative to focal length, the greater the "stretch" distortion will be. That's why the Hassie SW lens was designed to "just fit" the 6x6 format, unlike a view camera with perspective control. Being quite short for its given format, it still looks like a wide angle shot, but
does not warp the scene to the same degree. But for the same reason, if I can employ the smaller images circle of, say, my Fuji 125 W on 4x5 film, with its modest amount of excess coverage, I know there will be dramatically less stretch toward the corners than a 120 Super-Angulon with its huge image circle. In photogrammetric applications, the lens is fixed, so such distortions can be minimized. There are of course other lens design factors. But there's no generic formula via merely focal length. Not all wides are the same by any means. I'm not an optical engineer, but I do know darn well from actual applications that a "W" of a given focal length stretch-distorts lens than an "SW", and certainly less than an "XX-W", as if we were dealing with clothing sizes. And this is a completely different issue from pincushion vs barrel distortion.

It's more about the size of the image circle compared to focal length. A 38mm lens can be engineered to project a much larger image circle than the biogon does on the SWC (making it more of a W, SW, or XL lens in large format parlance), which would allow movements, or as you state cover a larger format with a wider relative perspective. All things being equal, that large image circle lens would project the same exact image, centered on a 6x6 frame.

Regardless, I've used an SWC, and objects toward the edges of the frame do appear larger. How else does one fit a large angle of view into that 6x6cm frame while keeping straight lines straight without the film or sensor being the same shape as the inside of an eyeball? I don't see it as an unwelcome distortion - it's simply the nature of wide angle lenses projected onto a flat plane of film. As long as it's rectilinear (little to no barrel or pincushion distortion) it's all good. The Biogon on the SWC is rectilinear.

 

[DREW WILEY]

Well, it's a matter of degree. That's what I was implying. I've had to do architectural shoots where the super-wide effect was the only option to squeeze the whole given area onto the film (prior to today's stitching techniques). But I hated the stretching effect. Generally, I don't like the look wide angle lenses give at all. But if I can select a regular lens of the same focal length, and it still covers, I'll do it every time rather than choose a true wide. But when ya need em, ya need em.

 

[M. Axel Wikstrom]

Well, it's a matter of degree. That's what I was implying. I've had to do architectural shoots where the super-wide effect was the only option to squeeze the whole given area onto the film (prior to today's stitching techniques). But I hated the stretching effect. Generally, I don't like the look wide angle lenses give at all. But if I can select a regular lens of the same focal length, and it still covers, I'll do it every time rather than choose a true wide. But when ya need em, ya need em.

Angle of view is angle of view. To take a picture of something large and/or close, one has to have a lens that has adequate angle of view. These are wide angle lenses. A well-corrected 150mm lens that covers 8x10, when centered over a piece of 4x5 film will project a normal looking image - same as a 150mm lens that just covers 4x5. If you put an 8x10 piece of film behind that wide angle 8x10 lens it would appear extremely wide, but take that 8x10 negative and cut a 4x5 piece right from the center and it will be identical to the image formed by the (normal) 150mm lens that only covers 4x5.

 

[DREW WILEY]

Don't tell me something I have already known for the past half century. But did you get what I just said? Apparently not.

 

[M. Axel Wikstrom]

Drew, I guess I don't understand it when you say "But if I can select a regular lens of the same focal length, and it still covers,..."

With a well corrected lens the image will look identical. If however, you're comparing using movements with a lens that has a large image circle, to not using movements with a lens that barely covers (of the same focal length), of course the photos will look different. Apples to oranges... you're not taking the same picture.

 

[DREW WILEY]

I'm absolutely certain that lens designs can differ in this respect because I often shoot with different types in the same focal length. Just don't have time at the moment to dig out some relevant literature as to why, design-wise. Oh well. It wouldn't help the original poster anyway. But I was speaking of using view camera movements in either case.

 

[jim10219]

I like my SA 65/8 lens a lot. I've never had issues with distortion. Lines appear straight. And with the lens stopped down to f:16 or more, which is required to get the full 4x5 negative exposed, you'll not have to worry about the curved field of focus.

What you will have to worry about is vignetting. Wide angle lenses like that tend to vingnette pretty seriously in the corners. To counter that, you'll need a center filter designed for that lens. Center filters are a lot harder to come by than the lenses and can cost as much, or more than the actual lens. They'll also slow down the speed of the lens. So that is something to consider when going down this road. If you can, find a lens that comes with the center filter. It'll be more expensive, but likely cost you less than purchasing them separately, and save you the hassle of finding one.

 

[wiltw]

Drew, I guess I don't understand it when you say "But if I can select a regular lens of the same focal length, and it still covers,..."

With a well corrected lens the image will look identical. If however, you're comparing using movements with a lens that has a large image circle, to not using movements with a lens that barely covers (of the same focal length), of course the photos will look different. Apples to oranges... you're not taking the same picture.

I may be interpreting your posts wrongly, but it seems that you may be confusing two different specifications of a large format lens ...

• Angle of View...what the frame will see, which is IDENTICAL for lenses of the same FL
• Angle of Coverage...how big the image circle is, to permit shift and rise/fall movements

You can have two 90mm lenses, both of which would cover same Angle of View, but one might have an image circle that covers 4x5 with enough image left to permit shift and rise/fall but inadequat image circle for 5x7 coverage with shift and rise/fall, while the other has a large enough image circle to cover and provide shift and rise/fall for both 4x5 and 5x7 formats...the Angle of Coverage would differ between the two lenses, while the Angle of View (for 4x5) is identical. And contrary to your statement, the images would NOT 'look different' if both lenses are well corrected to have little pincushion or barrel distortion, both would produce identical images within the 4x5 frame area.

Similarly, you can have two very different FL lenses, both of which possess the same Angle of Coverage.

"The word 'angle of coverage' is the term used to express the section of the picture area which the lens has the capacity to cover. 'Angle of coverage then expresses the parallel angle in relation to the area of light which the lens has the capacity to photograph. On the other hand, the picture angle is the parallel angle in relation to the film size from the center of the lens. "Angle of coverage" then, is the term for describing the lens performance which does not vary in relation to the size of the picture area available. It is calculated as Image Circle = tan w x f (focal length)".​

 

[locutus]

Rectilinear has a formal definition.

Which formal definition are you referencing here?

 

[Ian Grant]

I'm unsure what Drew argues about in most threads in terms of his perspectives and there's usually never anything that meets general consensus.

A Rectilinear lens by definition renders straight line correctly i.e. remaining straight in an image particularly towards the margins of that image. WA LF lenses will do that, there may be slight deviations which can be measured scientifically on an optical bench that can't be seen in practice.

There's a tendency to be pedantic and when points can't be backed up by example images from the writer one becomes skeptical. One reason many architectural photographers shoot LF in the well corrected WA lenses.

Ian

 

[M. Axel Wikstrom]

I may be interpreting your posts wrongly, but it seems that you may be confusing two different specifications of a large format lens ...

• Angle of View...what the frame will see, which is IDENTICAL for lenses of the same FL
• Angle of Coverage...how big the image circle is, to permit shift and rise/fall movements

You can have two 90mm lenses, both of which would cover same Angle of View, but one might have an image circle that covers 4x5 with enough image left to permit shift and rise/fall but inadequat image circle for 5x7 coverage with shift and rise/fall, while the other has a large enough image circle to cover and provide shift and rise/fall for both 4x5 and 5x7 formats...the Angle of Coverage would differ between the two lenses, while the Angle of View (for 4x5) is identical. And contrary to your statement, the images would NOT 'look different' if both lenses are well corrected to have little pincushion or barrel distortion, both would produce identical images within the 4x5 frame area.

Similarly, you can have two very different FL lenses, both of which possess the same Angle of Coverage.

"The word 'angle of coverage' is the term used to express the section of the picture area which the lens has the capacity to cover. 'Angle of coverage then expresses the parallel angle in relation to the area of light which the lens has the capacity to photograph. On the other hand, the picture angle is the parallel angle in relation to the film size from the center of the lens. "Angle of coverage" then, is the term for describing the lens performance which does not vary in relation to the size of the picture area available. It is calculated as Image Circle = tan w x f (focal length)".​

That's exactly what I've been trying to say. (Please read my responses to Drew's posts.) Assuming that the image circle is centered on the film, the same focal length will give the same exact view for the same format regardless of the size of the image circle.

 

[Dan Fromm]

Axel, Ian, mapplng lenses are often described as orthographic. When used with reference to lenses, orthographic means "very low distortion." I suspect that when Drew says rectilinear he means orthographic.

There's no point disagreeing with Drew.

Drew, please direct me to documentation that shows the existence of 90 mm Biogons. I've already looked at Arne Croell's Zeiss Oberkochen piece, which mentions a 90/2.8 "prototype for 6.5 x 9." That doesn't count.

 

[Ian Grant]

Axel, Ian, mapplng lenses are often described as orthographic. When used with reference to lenses, orthographic means "very low distortion." I suspect that when Drew says rectilinear he means orthographic.

Dan, I suspect you are right, the other term is Telecentric.

Ian

 

[Dan Fromm]

When in doubt, go to the documentation. I looked at distortion curves for the several vintages of 38/4.5 Biogons and 40/4 Distagons for Hasselblad. Look here: http://www.hasselbladhistorical.eu/HW/HWLds.aspx

Short summary, 38/4.5 Biogons all have a little distortion, around 0.2 mm (0.25% of 80 mm) of barrel distortion around 30 mm off-axis. The Distagons are all much worse. For comparison, good mapping lenses' distortion is at most a few microns.

Ian, Wikipedia, the source of all truth and all falsehood, gives this definition of telecentric:

An entrance pupil at infinity makes the lens object-space telecentric. Such lenses are used in machine vision systems because image magnification is independent of the object's distance or position in the field of view. An exit pupil at infinity makes the lens image-space telecentric

Telecentric is also a Ross name for perfectly ordinary telephoto lenses that aren't telecentric.

 

[Dan Fromm]

Oh, dear, I'm losing it too. Orthoscopic, not orthographic. For an example of a lens claimed to be orthoscopic, see USP 2,845,855. It isn't clear that this lens was ever commercialized.

 

[wiltw]

That's exactly what I've been trying to say. (Please read my responses to Drew's posts.) Assuming that the image circle is centered on the film, the same focal length will give the same exact view for the same format regardless of the size of the image circle.

It was an opportunity to state some factual things that contradict what some others have posted, without confronting them!

 

[DREW WILEY]

Hi, Dan. I already corrected that 90 thing with the 75 Biogon, which was once in vogue for wide open press use with 4x5 Technikas, but had little surplus coverage. It was late at night and I was multitasking, including planning for a trip that needs my 90 SW along, so I subconsciously crossed wires. Regardless of the nuances of terminology, if you take a lens of, say, 110 degrees of coverage at a certain f/stop and compare that to another of the same focal length which allows a maximum of 80 degrees at that same stop, it should be obvious that the much wider version is going to give you greater stretching if you approach the boundaries of that bigger image circle using greater movements. That can be proven from W.'s own diagram. Both lenses might cover the same size film up to a certain point. I'll leave it at that for now. But there are additional factors in how lenses are designed to begin with.

 

[DREW WILEY]

The post-War Biogon series in question had a similar configuration but better corrected complement in the second-generation Wild Aviogon. Super Angulons started coming out slightly thereafter. But the quite different pre-War Biogon had it's complement in the Wild Aviotar. It would be unthinkable to substitute one of our view camera Super Angulons for any aerial photogrammetry usage, not to mention all the improvements 60 yrs hence! Of course, we're limited by cost and wt constraints. But here you have sets of very similar optical configurations which differ in terms of distortion expectations. Same goes for the distinction between ordinary LF taking lenses and high-end graphic lenses of analogous optical construction and focal lengths, but where even the shape of dots in the corners of the field are expected to overlap perfectly round (or square in some versions), with no elliptical distortion. Try that with any of your Super Angulons!