丹妮的short short ground: August 2006

(extracted from http://medfmt.8k.com/mf/lenstest.html)

Quick and Dirty Lens Test

You are in a camera shop or a flea market, how can you test a lens? First, look at it - really look. Shine a light through the lens and look for fogging (from volatile lubricants in hot weather). Is there fungus growth on the lens elements? Look at the screws holding the lens together. Are there any signs of scratches on them, suggesting amateur repair efforts?

Exercise the aperture stop-down ring by moving it from slowest to fastest apertures and back. Is the action quick? Are the springs tight and strong? Or does the lens stop down slooowly? Look at the aperture blades, especially for oil or other signs of friction wear or damage. Now exercise the zoom function (is there is one) and focus the lens slowly twisting the barrel from side to side at each setting. You are looking for jumps or binding in the focusing mechanics which could indicate internal damage, wear, or poor repairs.

Another easy tell-tale is the front of the lens. Is the filter ring dinged? If so, the lens probably has a history of hard impacts, which could easily dislodge and decenter lens elements, reducing lens performance. Scratches on the lens tell a similar tale. Look at the rear lens element in particular. Our lens faults pages give you a hierarchy of lens problems.

Probably 90% of the lenses out there are okay (maybe 70% on EBAY, versus say 30% of the bodies!). Without using film for a test, you can't be really sure how well it will perform, or if there is not some subtle problem with it. But the above tests will identify most of the bad and abused lenses out there, and greatly improve your odds of getting a good one!

Distortion

You can't tell the distortion of a lens by looking through the viewfinder. Most camera viewfinders are not optically designed to minimize distortion, and are much more distorted than any lens you are likely to use. So you have to use film for this test.

Find a subject with known straight lines, preferably vertical or horizontal or both. I usually pick houses with aluminum sidings that are lighted so it is easy to see the lines. Other folks use venetian blinds in windows. Another favorite source for testing distortion are those downtown skyscrapers with lots of windows and lines in their facade. Be very careful in leveling your tripod, using a bubble level if available to keep it level.

Don't make the mistake of projecting your slides and trying to measure distortion on the big screen with a ruler. Sorry, but the enlarger lens and tilted projection setup will again be more distorted than the lens you are trying to test.

Instead, look at the slide or negative under a high powered loupe. You need a true, thin straight-edge to evaluate distortion. The easiest one is a clean and unused safety razor blade or something similar. Carefully line up the straight-edge with the lines in the slide. Usually the worst distortion will be at the outermost lines. Now check and see if you have either barrel (bulging out) or pincushion (bulging in) distortion present. You can use this technique to compare two lenses for distortion in a general way.

In theory, you can use a straight-edge metal ruler with tiny distance markings on it to estimate how large the bulge is over a measured distance. The ratio of the amount of distortion could be expressed as some percentage. In practice, most fixed lenses have distortion below 1%, which is hard to see even with a straight-edge, let alone measure accurately. Even zoom lenses average around 1.5% or less distortion, but a few low cost ones might be as high as 3% or so (readily visible even on enlargements).

I should also add that for zoom lenses in particular, the amount of distortion can vary a great deal depending on the focal length setting. Some zooms even go from moderate pincushion distortion to modest barrel distortion as you zoom the lens through its settings (i.e., a compromise).

Stopping Down Shots

A number of lens aberrations are reduced or improved by stopping down the lens to a smaller aperture. Vignetting of the edges of the image is reduced or eliminated. Coma, astigmatism, and flare are also improved by stopping down. So for most lenses, you need to have a series of shots at several distances (infinity, portrait distance, close focus) to document these improvements.

Some lenses (e.g., wide angle zooms) may vignette if used wide open. This problem will show up best at the widest angle setting on the zoom with the lens open to its maximum aperture. You will see some darkening or light falloff in the corners. Stopping down will generally improve the image until the vignetting is eliminated, usually one or two stops down.

Another common cause of vignetting is using a thick filter on an ultrawide angle zoom. For example, I have a 17-28mm zoom which vignettes at the 17 to 18mm settings when I put on a 72mm red filter (for sunsets). The filter is too thick, and cuts into the photo, producing curved edges on the film. Special thinner filters are made to prevent such image vignetting with wide angle lenses. This issue is yet another reason to test your lens without a protective filter in place, and then again with the protective filter (if you use one). You will know if the problems with vignetting are in the lens, or the use of the too-thick filter.

To avoid these problems, you would have to test the lens at the widest focal length settings. You might also run a series of tests at different f/stops to pick the sharpest settings or "sweet spots" of your lens. These "sweet spots" will be where the lens will work best in terms of resolution and overall performance. A few pro lenses such as fast telephotos may be optimized for wide open performance. Here again, you would want to know this when using the lens!

Lens Coloration

You don't really need any film to check the lens for a color tinge. Many older lenses tend to have yellowish tinges from the aging of older formula lens adhesives. Other lenses were made with a deliberate cool ("blue") or warm ("red") coloration, sometimes closely matched with other lenses in the line. By simply holding up the lens to a white sheet of paper at an angle (circa 45 degrees) you should be able to see any lens color effects. On a rangefinder where the lens won't come off, you may have to open the back and adjust things so you can look through the lens (if possible).

The lens coatings may be colored magenta or blue or other colors to the eye, but they don't impact the lens color in the above test. We don't use film for this test, as films have color prejudices of their own. Be sure that the light source looks bright white, as from a daylight fluorescent or white light bulb or mid-day sunlight source. The paper should appear bright white for best results.

Flare Tests

Flare comes in two flavors, veiling glare and visible flare. Veiling glare is the loss of contrast seen in darker shadows from light bouncing around off the lens elements and film. To evaluate veiling glare, you want a thin dark subject against a bright white background. The easiest real world subjects I have found are thin tree branches against the sky in winter. In summer, the leaves get in the way, so I may opt to use a powerline against a bright cloudy sky. Ideally, the branches or wires should be dark black and transition abruptly to the white sky. The less solid black the branches or wires look, the more you are seeing the effects of flare. If you have a downloaded lens test chart, you can also do this test easily with the black and white lines on the chart.

To evaluate visible flare, you can take a photo of a bright point light source such as the sun. You may see a hexagonal or other pattern from reflections off the surface of the lens diaphragm blades. You may also see bright blobs or streaks of light reflected from lens elements. Try a shot without your UV filter, if you use one, to see why some of us don't use UV filters (they add to flare). Some lenses such as my 20mm f/3.5 AI nikkor have surprisingly little flare, even in direct sun in the photos shots. Zoom lenses tend to have much higher flare than other lenses due partly to the larger number of elements having more reflective surfaces. Again, this is one major reason many of us prefer fixed or non-zoom lenses.

The question of subject distance when testing a lens is important. Photographic lenses are normally optimized for middle distances, say, six to ten metres, the range in which they may mostly be used. The wider the aperture, the narrower the optimal range, a factor important from f/1.4. Source: Lens Test Charts by Geoffrey Crawley in British Journal of Photography, January 31, 2001, p. 26.

Normal Lens Test Surprises

I started out shooting Nikons because I was doing underwater photography, and Canon had yet to come out with their F1 model with interchangeable finders. So my pro camera choice is the result of Nikon having the only interchangeable action viewfinder for underwater housing use for their Nikon F camera. Naturally, I believed the hype of Nikon ads and other Nikon users about Nikon's great superiority to every other camera brand. As I needed cameras, I accumulated a bunch of cameras from Shutterbug ads, simply because I had to buy the camera to get the camera housings. Every camera housing was another potential student for each class, so they were also tax deductible!

Then I made the mistake of trying to show how my pricey Nikon kit was superior to my cheapy Minolta SRT101 setup. Bad mistake. I shot each camera with the same slide film, side by side on the same scene and identical lighting (one shot and different cameras and lenses right after the other), on the same tripod, with the same lab doing the processing. I could see some minor differences, but that Minolta lens was surprisingly good. I picked up a $17 used normal lens for my Spotmatic. Surely my Nikkor would blow it away. Ooops! These normal lenses were all so good that you could hardly tell the slides shot side by side apart. Maybe it was just me? So I asked some friends and students to try and tell them apart. They were equally surprised by how good the other lenses were.

Some years later, I found myself on a tight graduate student budget, and unable to afford most of the exotic OEM (nikkor) lenses I wanted. So I decided to settle for some third party lenses. I "knew" they'd be awful, based on what everybody had told me. Unfortunately, I decided to see just how terrible they really were. To my surprise, these lenses did pretty well, often as well as my nikkors at mid-range (diffraction limited) f/stops.

I reached the nadir of lens prejudice when I decided to find a "bad" lens to destroy it in some tests to see how bad scratches and even cracked glass would impair lens performance. But first I needed to test the lenses to see how bad they were before I scratched and bashed them up, right?

Yes, testing these $10 to $15 US lenses was another bad mistake. I bought a 28mm f/2.8 AI lens made by Osawa. Surely this junker would be awful. Well, it was - awful good that is. It turned out the dang lens was very nearly as good as my much more costly Nikkor 28mm. Hmmm?