Large format lenses: generalities, classic vs modern

By John Sparks for the Large Format Page

Basic lens questions

What is the difference between symetrical and asymetrical?

A symetrical lens has two identical lens groups on either side of the shutter facing in opposite directions. Several lens abberations are automatically eliminated by symmetrical design (only at 1:1) so it's a popular lens design, especially before the days of computers when all the lens caculations were done by hand. The symmetrical designs require fewer computations to design. If truely symmetrical, the lens will perform best at 1:1. Asymetrical designs are all designs that are not symmetrical. Today, almost all lenses are at least somewhat asymetrical (the only exceptions I know of are lenses like the G-Claron, Apo-Ronar and other lenses optimized for use at 1:1). Any lens that is truely designed for distant subjects must be at least somewhat asymetrical.

There is nothing fundamental that makes symetrical or asymetrical designs better than the other.

What is the difference between coated and non-coated?

Coatings are applied to all air/glass interfaces to reduce the reflections off these interfaces. This reduces flare and increases contrast. There are actually three types, non-coated, single coated and multicoated. Single coating was first developed in 1936 and most lenses made since WWII have been at least single coated. Multicoating was developed later (late 60's or early 70's??) and most current lenses are multicoated.

The biggest advantage to coatings is to allow the lens designer to make more complex lens designs with more air/glass interfaces. More complex designs can decrease abberations giving a sharper lens or a lens with more coverage. For the fairly simple designs used by most large format lenses, the difference between single and multi-coatings is relatively small but the difference between uncoated and single-coated is significant.

The best lens designs before coatings were only 2 or 3 lens groups (the number of groups indicates the numbers of air spaces, each group is a set of lens elements that are cemented together with no air spaces). Modern large format lenses are mostly 4 groups and some (particularly wide angles which need better corrections to get their wider coverage) have 5 or 6 groups or more.

Multicoating's real contribution was to allow the 12 to 15 group zoom lenses possible. Without multicoating, these lenses would be unusably low contrast.

What is the significance of elements and groups?

Not very much just from the numbers. More elements and groups allow more degrees of freedom for the lens designer to make a better corrected lens, but with care and some limitations (like small maximum aperature or small coverage), very good lenses can be made with few elements and groups.

As I wrote above, more groups means more air-spaces and requires better coatings. The best of the large format lenses priar to coatings had only 2 groups (like the Dagor and Protar) or maybe 3 groups like the Tessar. The Dagor and Protars have good coverage, but are only really sharp at f/16 or small aperatures while the Tessars are good at relatively wide aperatures but with limited coverage.

After coatings were developed, most lenses were 4 groups (with a few 3 group tessars still around and some 5 or 6 group lenses) as this gives a better corrected design for a faster lens with good coverage like the current plasmats with 70-80 degree coverage that are very sharp even wide open.

Classic vs Modern lenses

I think "modern" means: maximizing coverage, multicoating, computer design, modern glass choices, good sharpness at all aperatures rather than requiring the f/22 to f/45 needed for good sharpness with older lenses. They will be sharper lenses, especially when you can avoid diffraction effects from stopping way down. Any wide angle by Nikkon or Fuji, Rodenstock Grandagons and Schneider Super Angulons are the "modern" wide angle lenses.

There's some stuff written about older lenses in the view camera book by Steve Simmons. There's also been a lot in various back issues of View Camera and Darkroom Techniques magazines (especially articles by Ron Wisner who knows a great deal about older lenses). Anyway for a brief overview, your best bet is to get used lenses of current or almost current designs. There was a big revolution in the 50's in 60's of computer designed multicoated lenses that really perform much better than the older lenses. This is especially true for 4x5. Larger formats require less enlargment, smaller aperatures with their greater diffraction so differences between lenses matters less. With really large format and contact printing only you can use almost any lens you can find that covers the format. Any Fuji or Nikkor lens can be considered modern as can most by Schneider or Rodenstock.

For more specific answers, I'll divide the lenses into three groups, wideangles, normals and long lenses.

All the lenses I currently use are modern lenses. I have one older 135mm lens from a press camera that actually works pretty well for B&W. It has very limited coverage on 4x5 and much lower contrast that my other lenses, but with B&W I can increase contast pretty easily to compensate for it. 16x20 prints from 4x5 negatives don't look less sharp than from my other lenses, but I'd guess I have to print on 1-2 grades higher contrast paper. I have made a few transparencies with it but the shutter is erratic enough that getting good exposures was difficult. The colors are quite a bit less saturated than from modern lenses.

For 4x5, I generally prefer modern lenses, though the best of the older lenses still perform well. With many older lenses you will get much sharper results with a yellow filter (many were designed when films were ortho or blue sensitive). Color requires only the best of old lenses or modern ones (older lenses will give softer more pastel renderings compared to modern saturated ones). With larger formats, older lenses may be the best or only choice and the smaller or no enlargment means less resolution is required. I have several older lenses I use on 8x10 though I can tell the difference even here compared to modern ones.

Have you ever made the same photograph with a modern lens and an older good uncoated lens of similar photographs. I have (in B&W). From seeing the results, I don't think I could pick a given print and say which lens made that photograph, but given two prints made of the same subject with modern and older lenses, it's quite clear which one is which. In color, I think I could even pick which lens was which without the side by side comparison (though the older lenses may yield pleasing images in color). The biggest problem with using older lenses with color (particularly color transparencies) is the questionable accuracy and repeatabiliy of old shutters. The ones I've owned vary enough that getting accurately exposed transparencies, even with wide bracketing, is very hit or miss.

But, I'm talking here about old uncoated lenses. With last generation modern multicoated non-apo lenses compared to the current apo lenses, I'm not sure I could pick which is which in side by side comparisions. There is very little difference between a Apo-Symmar and Symmar-S for example. The Apo lens is very slightly sharper, but you probably get more sharpness variations from film position in the holder than the difference between these two lenses (or similar lenses from other manufacturers).

I a lot of cases, older shutters have more problems than older lenses. Shutters can be replaced, but often cost more than an older lens is worth. A shutter who's speeds are off (almost all older shutters) is not a problem as long as the speeds are consistant (you just need to have the speeds checked and use the actual speeds instead of the marked speeds). An eratic shutter is useless.

Small scratches or "cleaning marks" have no real effect on image quality and mean much lower prices, but cloudy areas or other marks covering a fairly large area of the lens will reduce contrast and resolution (try to think of the mark as a percentage of the surface of the lens, a scratch covering 1% or less of the lens will have almost unnoticable effect while one covering 1/3 of the lens is much worse).

Importance of corrections by Richard Knoppow

There are two issues here, aberrations and coating. Coating first.

Coating prevents flare. Flare can be either an overall haze or ghost images. If we consider the haze only, its effect is to lower the contrast of the shadow areas of the image. In B&W this can be compensated for satisfactorily (if there is not too much flare) by using short toe film (high contrast in the shadows) and simply developing or printing for a little more contrast. In color, the haze, because it is mixture of all the light recieved by the lens, tends to reduce the saturation of all colors. There is less flexibility with color film for correcting this sort of fault. None, really with reversal materials. Also, since single coatings are more effective at some color than others, it may result in a slight color cast, which, because it is due to a colored haze, isn't easy to correct in any simple way. Multiple coatings have much more even reduction of reflected light so improve the color rendition as well as saturation.

The kind of flare that is manifested as ghose images or hot-spots is much more damaging to images whether B&W or color and are more due to the lens design than coating although coating will either reduce them or get rid of them.

Ghost images or images of the diaphragm have much more to do with the design of the lens than with coatings, although coating will reduce the effect when it is present. I have several old, uncoated lenses which do not produce ghost images. The main effect of lens flare is to generate an overall glow, which tends to fill in shadow areas and reduce the saturation and purity of colors in color photography, and to produce halos or flary areas around bright objects. While the advise of often given to use lens shades to reduce flare of uncoated lenses, a lens shade has no effect on light going into the lens to produce an image. They are mainly useful in cutting off non-image light which causes flare by bouncing around inside the camera. Multiple layer coating extends the range of colors over which the coating is effective. A single layer is effective at a single color and its effect falls off on either side of that paticular wave length. Because the range of colors used in normal photography or visually is only about an octave, a single coating is practical. Multiple coatings have a broader band of effectiveness. By using several coatings of the right materials and thicknesses it is possible to make the lens surfaces nearly completely non-reflective over pretty much the entire bandwidth used for photography. For those with backgrounds in electronics coating is related to filter theory and transmission line theory. Essentially, a coating forms an impedance matching section between the air and glass. The fewer glass-air interfaces there are in the lens the less the effect of coating will have on its performance.

A lens with residual aberrations may actually be _less_ noticeable with color than with B&W! This is because of the phychology of seeing. The brain interprets color images as having more detail in them than they do (this is why color television is possible). Also, even when an obvious color fringing is present, the fringes, in the case of lateral color, will be sharp, and the eye will see the image as sharp, whereas the same image in B&W will have all the colors integrated in the image which will look blurred. Even box cameras will take surprizingly good looking color pictures. Where panchormatic film is used, as is the almost universal practice today, a lens with superior color correction will make superior images in B&W as well as color. Many older lenses, (I mean the 70+ year ones) were designed with the assumption that ortho film would be used, and the acromatic correction was made toward the blue end because that's what the film recorded. More modern acormat design moved the colors more toward the yellow for panchormatic film. Also, modern glass types (post 1940) permit much smaller overall chromatic deviations regardless of the wavelengths chosen.

The argument about the definition of apochromatic and whether certain lenses are really apo's is somewhat silly. The real issue is the maximum deviation of the corrections at any color. Where a lens is to be used only with very sharp selective filters, the correction at specific wavelengths that correspond to the filters may be enough for perfect images but that is not a very common condition. The generic definition of color correction is that an acromatic lens is corrected for two colors, an apochromatic one three colors, and a superchromatic one for four or more. Modern lens design permits the designer to correct for any number of colors although there is, of course, a point of diminishing returns in cost vs: performance.

Warren Smith demonstrates in his book on design that a properly designed acromat may have better overall chromatic correction than an apochromat. He also demonstrates the design of superchromats.

Rodenstock and other manufacturers now use an ISO definition of apochromatic that is based on the lens having very small deviations from ideal. This may nott be in agreement with the rigorous definition used for microscope objectives but nonetheless, the performance, even for color separation purposes is likely to be superior to a "true apochromat" which is what matters. I suspect these lense are actually superchromats. The ISO standard, or rather, family of standards, which specifies this definition runs into a huge number of documents to which I haven't got access.

View or add comments