exposure of dark evergreen trees?

[John Fowler]

I'm interested in the pre-exposure solution to this conundrum (likely the best solution). However, I've misplaced my Adams book explaining the system. To what level and with what neutral subject matter would one pre-expose Velvia for such high contrast situations?

[james mickelson]

It can increase shadow detail by 1/2 stop if done correctly. I used to use a grey card held well out of focus. Now I use a small piece of translucent plexiglass.

[Brian C. Miller]

I have found that Kodak E100S pulled one stop produces great results. While I haven't yet tried it with evergreens, I have used it with architectural shots.

My Pentax Spotmeter V was worked over by Zone VI Studios. According to the old web site, they modify the Pentax and Soligor to have IR filters and other good stuff.

Besides pulling the film, try gradient filters like Cokin or Lee, and open up the lens. I remember someone saying that Galen Rowel got a total of a 14-stop range using judicious filter positions. Put the ND part of the filter over the bright part of the scene and go for it. I have one Tiffen split ND filter, and it's convinced me I need something like Cokin or Lee.

[Lyndon Gordon]

I must express credit for some of the following theory and paraphrasing, to Roger Hicks & Frances Schultz, the authors of the excellent book "Perfect Exposure." The book has helped refresh the knowledge I learnt from the training I received at the Royal Australian Air Force School of Photography.

Bill, some of the answers to your question have discussed a couple of the least understood factors affecting the procedure of obtaining "perfect exposure." Once the factors are understood you'll need to use logic and an exposure meter. A wise photographer will use both, much like an aircraft pilot trusts an aircraft's instruments, with the knowledge that they've been calibrated and set properly. However, just like an aircraft's altimeter doesn't always read the actual true altitude, an exposure meter doesn't always read the true exposure needed. For the novices out there, opening your mind to such facts is the beginning of your understanding.

So what are those factors I mentioned? The first one is CONTRAST (at the subject, focal plane, and how much is desired during any reproduction of the image). The second one is SPECTRAL RESPONSE (of the exposure meter, and the film). There is also another important parameter that wasn't really explained properly by anyone so far. It's the EXPOSURE METER CALIBRATION to a certain "average" scene reflectance.

I used the term "perfect exposure" because it's impossible to achieve "correct exposure" of everything in a scene. If you create an image that meets your requirements (or the end user's) then it could be said that it's perfect (this applies to composition, sharpness, and all aspects of a photograph).

N Dhananjay correctly discussed Metering Technique and Contrast Range. However, the words "contrast" and "range" in a sensitometric context mean the same thing. CONTRAST describes a range of brightness (or "density" in a film/print image). The "Subject Brightness Range" at the scene is first reduced (very little by good lenses, and sometimes a lot by "flary" lenses) in the camera. You need to take that into consideration if the camera & optics (including filters) cause much flare. Zeiss and Leica lens users will feel smug when they read that! Oh, there is such a thing as a lens that's too contrasty ... when it's "hot" at the image center and a bit dim at the edges. Then you need to understand how much of the "Image Brightness Range" (at the focal plane) your film can record before very dark or very light features will have their brightness range "compressed" on a film's characteristic curve (a graph of density versus log exposure) "toe" or "shoulder." This is when you can't distinguish subtle details in shadow or highlight areas. The textures of the evergreens are an example of shadow detail that can be lost in an image that's too contrasty. At each stage of analyzing contrast, it can be quantified objectively with ratios, logarithms, G- bar's or gammas, density ranges, and paper/scanner log exposure ranges; or subjectively with Zone System values. However, most of the time it isn't necessary to get into that photographer's science of "sensitometry." Basically, a "contrasty" film (like Velvia), or one that is given an amount of development (especially for B&W) that "expands" relatively small ranges of scene brightness so that similar shades from features are more discernible, cannot be expected to record the full range of reflectance from a "contrasty" scene. So those who suggested trying a less contrasty film had a partial solution for you.

The parameter SPECTRAL RESPONSE also needs to be understood. Exposure meters, and each type of film (or image sensor) have a certain fixed response to different wavelengths of ultraviolet, visible, and possibly infrared light. So, they may not (and generally will not) respond to all colors equally. To add to the confusion, usually the response of the exposure meter, film, and our own eyes are all different! But don't worry, perfect exposure is a combination of so many factors that some things cancel others out. A little knowledge and experience can easily compensate accurately enough for the others ... and after all, we still have "latitude." Alec was definitely on to something when he said "you're all wrong!" But as Lot mentioned, Alec was giving an answer to a question that you didn't really ask! That's because you were using an incident reading. I should mention though, what Alec was leading to is that for a reflected reading you need to apply a exposure correction value/factor (how much depends on the meter's spectral response to the metered color) for subjects that are predominantly one color, and then know (from studying a "spectral response graph" if published) how your film (or sensor) will respond to that color too (because a further small exposure correction may be needed). Other issues that Lot and others discussed really don't account for much of the underexposure problem though Bill.

So, if you match the type of film (and/or amount of development, especially for B&W) to appropriate image contrasts, and you've given exposure correction (for meter and film spectral response), but you're still getting underexposed evergreens ... it's probably because of the third and last (thankfully, some of you may be thinking!) misunderstood parameter I have to discuss ...

... It's the EXPOSURE METER CALIBRATION. This may upset some people whose minds aren't open enough to question some of the "rules" that they've learnt about photography: it's a myth that most exposure meters are calibrated to "18% gray" reflectance! 18% gray is a photographic mid-tone half way between black and white. That's true enough, but it's not the magical number that many think it is. Furthermore, many or maybe most scenes don't even reflect that almost legendary value of 18%! The amount of light reflected from so called average scenes may vary from as little as 10% to as high as 25% of the light that is incident upon it (from the sun and sky). More extreme values are common in many parts of the world (even if the solar angle was held constant for all examples). Bill, your subject of the evergreens probably only reflects as little as 3% of the incident light in winter, or as much as 8% in summer. This is logical because evergreens have dark colored (green) leaves, and they, combined with tree branches create many dark shadows. Many exposure meters are actually calibrated to a more modern "average" reflectance of around 12% which is a more representative figure of what's actually "out there." The 18% figure came from the old days when Eastman Kodak's researchers were trying to quantify sensitometric parameters. 18% reflectance may be fairly average for the geographic latitudes (and terrain types) like those around Rochester, New York, but the modern day manufacturers of the Gossens and Sekonics (etc.) we use know a little better now. They should too, because they've had more time to get it "right." So how can you Bill, get it right?

Well, simply put, you need to make correction(s) to what your meter says. But don't go overboard, after all, dark subjects like evergreens should look dark and light subjects should look light. One (often perfect) method is too average reflected and incident readings (duplex metering). The important thing is to get detail where you (or the end user) want's it without sacrificing too much of the rest of the picture's tones. Let's say your exposure meter is calibrated for 12% and you're taking the incident light reading of the sun and sky light falling upon the evergreens. The meter will indicate how much exposure is required to make a subject that reflects 12% of the light, actually look like it's reflecting 12% of the light. Of course that's assuming that the film speed set on the meter, actually works "perfectly" for the film (yes, that's another "rule" that you may have to throw out the window ... film speeds are just a guide, they can be "offset" or vary with lighting conditions or development amount). But the evergreens might only be reflecting 3% of the light, so you have to compensate, keeping in mind that they should still look somewhat dark in the final image. If the meter was calibrated for 3% reflectance, you wouldn't have to do that ... but such a meter wouldn't be good for bright subjects. The point here is that an exposure meter only gives "correct" readings when metering for subjects that reflect about the same percentage of light that the meter is calibrated for. The "phenomena" of needing an additional 1 to 1.5 stops of exposure for the evergreens is mostly because the subject is somewhere between only one quarter (3%/12% = 0.25) to two thirds (8%/12% = 0.67) as bright as what the meter "thinks" will be reflected back from the subject. The remaining difference to the exposure value that you know works is probably made up by the effects of the other factors. Partly CONTRAST (with Velvia, whether by intent or not, your original readings tended towards keeping highlight detail, and that's caused the shadow details to be compressed together on the toe of the "characteristic curve"), and SPECTRAL RESPONSE not being taken into consideration.

By choosing the APPROPRIATE FILM AND/OR AMOUNT OF DEVELOPMENT, to first get the CONTRAST right, and then taking SPECTRAL RESPONSE, and EXPOSURE METER CALIBRATION (or actual subject reflectance) into consideration when taking light readings, you should get perfect exposures almost all of the time! Good luck Bill.