   |
David --
I didn't really intend to make flare testing all that big a deal. The procedure I outlined in the book is a way to become aware of the general effects of flare on image quality, and the results are interesting, but it's far from a definitive calibration procedure. I certainly don't recommend testing individual lenses for flare effect because I'd consider that an exercise in busy-work. There are far too many variables affecting flare to make any such calibration reliable, and as individual users (not trained in optics and lacking proper measurement equipment) we have no way to identify the causes of flare or measure their effects.
I suppose this sounds contradictory, given the apparent emphasis I've placed of "objective testing" procedures, but realistically we have to recognize that there are some things we can't measure and can't control. Rather than pretending that we can, and probably coming up with faulty data, I think it's better to look for trends of behavior and try to provide average compensation for effects we know are present. We do know that every lens-formed image will be affected by SOME flare and we know that modern coated lenses keep internal reflections (and potential flare effect) to a minimum. We also know that subject and light conditions can influence flare, but we can only guess how serious the effects will be. We do know that the general effect of flare is to increase B+F density and reduce image contrast (it also affects image gradation but we can't do anything about that during film exposure).
What we CAN do is decide on compensation that will tend to restore image contrast, and try to relate that compensation to subject range, because, as a rule, contrasty subjects are more flare-prone than are flat subjects. This is what we've tried to do in the ExpoDev program by providing an optional value of "flare compensation." Basically all this does is increase development in an attempt to restore contrast. We've settled on "0.02" as a suitable number for general use, but left it in an edit field so the user can change it. Increasing that factor will increase image contrast; decreasing it lowers contrast. Setting it to zero gives you exposure/development data that include no compensation at all and would be applicable for contact printing.
I have no idea what the maximum flare effect might be in pictorial photography, but I suppose it might result from fogging the lens and pointing it directly at the sun. Under more realistic conditions I know of no better way to cope with flare than to take pains to keep the lens surfaces clean, use an effective lens shade, try to avoid including light sources in the picture, and provide some development increase for subjects that seem likely to be unusually conducive to flare.
Practically speaking we have to keep things in perspective. As BTZSers are probably tired of hearing me saying, "some things are worth worrying about and some are not; the trick is to figure out which are which." I'd put individual lens flare differences in that last category.
|
|
|
|
|
|
|
