I just made a comparison test, using four different film SLRs, to test six different lenses: 35 and 50mm. A mix of AF and mostly manual glass. All were shot on the same roll of portra 160.
All of the lenses tested comparably, except for one highly-acclaimed MF 50mm, which gave me super soft results. Blurry, really. Everything was tested at close range, and between f1.4 and 2.8. This blurry 50 was the only lens tested on this particular body.
What might be the cause of the problem? A lemon lens? I've had that experience once before, with a 50 Lux-ASPH. But this is an SLR lens, and the view through the viewfinder was the largest, clearest and most easily focusable of all the camera+lens combos I tested. Wouldn't I SEE a problem? If it's not a wonky lens, is there something in the body that could do that? Doesn't look like mirror vibration. It just looks like a bad or miso used lens, but all of the shots - 1.4, 1.8, 2, and 2.8 were equally bad compared to the others. And there were no vibration effects or camera movement problems seen on the entire roll.
Hi,
Because you only made images with one lens fitted to that body, you can disregard any comparisons between other combinations. There's no direct benchmark. So: this is simply a matter of doing some basic focus checks.
I don't know how much you know, sorry, so bear with me. Except for scale focus cameras and their ilk, generally speaking, any camera that features a calibrated focus system enabling the photographer to precisely set the focus of the taking lens to a particular focal point, is actually comprised of a couple of sub systems.
Eg: a rangefinder has its lens focus set by, the focus stop of the lens (or body in the case of, say, the Contax). But it also has a rangefinder focus patch, and the mirror/pivot/prism/ is adjusted to calibrate this
to the lens.
A TLR has two lenses: the taking lens is set correctly to infinity by (usually) the stop on the focus lens. The viewing lens is turned on its threads so that it
agrees with the taking lens.
In both the above examples, the cameras employ optical components that are independant of the primary lens to enable accurate focus to be set. They must however always be calibrated to match the primary lens and are always secondary to the primary lens focus adjustment.
An SLR configuration is a little different. As it's name (single lens reflex) implies, there is only one lens, the primary lens which records the image. The system used to view the focus screen and set your desired focus diverges from the primary system at the reflex mirror that is dedicated (in the case of non-pellicle designs) to reflecting the image into the focus screen. So, what you need to work out is whether the primary focus of your lens on the film plane is good, first. Knowing that you have a good sharp view through the finder doesn't tell you much, in the first instance, until you've worked out whether the film plane is actually sharp when the lens is set to infinity. Then, and only then, can you think about inspecting and adjusting the sharpness of your viewfinder image. The finder must match the film plane. Not
vice-versa!
To check the film plane, you can use a ground glass and a loupe to examine the formed image, view camera-fashion, through the lens. Obviously the shutter must be set to "B" or "T" (a lockable cable release in the socket makes life that much easier if you have one).
Alternatively, you can fashion a usable target to be used in conjunction with a second, collimating SLR out of a spare piece of unexposed, developed film. Eg. the first couple of frames of a processed roll of black and white often has some unfogged portion you can snip off, and very gently scribe some cross hairs into with a sharp knife. With this target film placed across the
film rails, (emulsion side to the lens, naturally) and sandwiched by a clear piece of glass resting on the
pressure plate rails, with the lenses of both camera set to infinity, you may visually inspect the focus of the camera under test at the actual film plane once you've set the shutter to Bulb. You'll want some natural light or another light source behind the camera to prpvide some illumination, of course.
Both techniques have their merits and I have used both on numerous occasions with very good results. Actually, I've done precisely that, today, to calibrate a Voigtlander Vitessa that has needed a lot of work. Many times, in fact, because of the need to pull the top cover off that particular camera and put it back on, every time a adjustment is made! But that's another story...
If using another SLR as a collimator, there are two important points worth noting. Firstly, the camera used must be definitively trustworthy. If you're not sure of its accuracy, using it to set another camera is likely to be a frustrating exercise. So think about the kit you have, and select the piece you have the most confidence in as to its accuracy.
Secondly, accuracy is improved greatly by mounting a lens to the collimating camera that has a substantially longer focal length than the lens fitted to the one being tested. If you're examining an SLR fitted with a 50mm lens, you can inspect the focal plane using a 35mm lens, but, don't expect a lot of accuracy if you do. You should really be using a lens with a focal length of at least 100mm (2x), ideally more, as the additional magnification provided will afford you a much more precise view of the patient cameras film plane, and even the smallest focus error in your target markings will be clearly visable. When setting the Vitessa mentioned above, which has a fixed 50mm Ultron lens, I used my Hasselblad with its 250mm f/5.6 Sonnar (more magnification) and its 150mm Sonnar f/4 (brighter view) to inspect the film plane focus and adjust it accordingly. Given a choice, use a long, fast, telephoto as the most accurate and easiest lens with which to collimate.
Assuming you have used either of the above techniques to inspect the primary lens focus at your film plane, you will now know, definitively, whether or not the focus of the lens is correctly set. If it's not, you'll need to adjust this before proceeding further, and the procedure for this varies a lot, depending on which camera/lens you're discussing.
If the film plane focus is really sharp and crisp, you're over halfway there. You've worked out where your problems
aren't, and that's good. So if you're not getting a good, accurate focus image through the viewfinder of the camera, you know it's not the lens at fault. Any discrepancy between what you, the photographer, is seeing, and what the lens (film plane) sees, can only result from either the position of the reflex mirror or the focus screen itself. (A non-standard viewfinder eyepiece dioptre is a third possibility I suppose, however if this does not match your eyesight your view of the focus screen should be unsharp at any distance with any lens, which ought to be fairly obvious, so we can probably discount this one).
To sum up:
Check the film plane focus at infinity. Adjust if necessary.
Check the viewfinder focus at infinity. Adjust if necessary. Usually this will be via an adjuster that changes the angle of the reflex mirror, however you might, in some cases, need to remove the top cover and adjust focus screen mounting clips or shims.
Once set, it is always worth using a ground glass to cross check the film plane focus and the viewfinder focus at close distance. If you're working with a quality camera, when both optical subsystems have their infinity focus accurately calibrated, at any close distance, they should also be absolutely bang on. If so, you can be very confident that your cameras focus adjustment is excellent. On the other hand, if the infinity looks good, but the minimum distance cross check is off, your infinity adjustment likely needs to be reviewed, and fine tuned.
Cheers,
Brett
