Graduated ND filter exposure for film.

Discussion in 'Modern Film Cameras' started by orourke, Feb 6, 2011.

  1. I'm not sure if this belongs in this forum or not. I will be using an F4, Fe2, or FE for my exposures so if nothing else, that's a reason for placing it here.
    I want to experiment with Graduated ND filters. Cokin to be specific. (I am of the understanding that they are actually grey and not true ND filters) My question is this.... In order to darken clouds, (both color and B&W) making them appear more ominous for example, do I meter the scene without the graduated filter in place and then drop it in after I determine my exposure info? Will there be a difference if I use the matrix meter on the F4 as opposed to the FE2 or FE ? I will be bracketing but I'd kind of like to know what to expect, or if there are any basic rules of thumb to be considered.
    I'd like to add that I probably (99.9%) will not be manipulating the image after the fact (ie photoshop) so I'd like to get it right, right out of the camera. Thanks.
  2. The graduated ND filter is used to overcome extremes in contrast, such as bright a bright sky with main subject in shadow. I use them when shooting in a city with tall buildings to even out the lighting, or with landscapes just before sundown for the same reason. They will not really have the effect you are looking for.
    The filter rotates like a polarizing filter, so just look through the viewfinder and rotate until you get the effect you want.
  3. If you meter without the filter first, then when you drop in the grad ND, the sky and clouds will darken. Depending on the filter, you can reduce the stops in the sky by 1, 2 or more stops. I don't know if that will give you the effect you want. Why don't you try it with a digital camera first to experiment with various aperture settings, bracketing etc..
    You might try a polarizer as well as that will darken the sky against the clouds creating more contrast and maybe more drama. If you're shooting B/W film add a red filter for really dark (black) skies. You won't need the polarizer in that case.
  4. As Alan said, meter the scene first, and try to keep as much sun/sky light out of the lens as possible. What you want is to meter the scene below the horizon. Then, when you drop the filter in, you can open up the lens by about 1/2 stop to compensate for light reduction in the bottom part of the filter. I have two ND filters, filter factors 1.2 - 4 and 1.2 - 8, which equate to ranges of about 1/2 stop to 2 stops and 1/2 stop to 3 stops, respectively. Don't confuse the filter factors with actual stops. Each change in filter factor by a factor of 2 is a 1 stop difference in lens setting, so:
    Filter Factor> f/stop
    and so on.
    By the way, all of this assumes you are shooting from a tripod and in manual mode, because there will be some slow shutter speeds involved.
    Polarizer filters have somewhat the same effect, but they also increase the contrast (which you may or may not want), and have maximum effectiveness at right angles to the direction of the sun.
  5. I don't see the point of metering the scene with a different configuration (without the filter) than you intend to shoot it. When I've used GNDs, I put the filter on, frame the scene, adjust the filter position, then meter and shoot. Works fine. I don't meter before adding a CPL, ND, or contrast filter; why should a GND be any different?
    I doubt the difference between center-weighted and matrix metering really matters much here. The results may be slightly different (just as when you shoot without filters), but fundamentally it's still the same question of the meter trying to figure out a sensible exposure based on the scene presented to it.
  6. Craig. Actualy you raise an interesting point. Why meter on the foreground first? Classic instructions say you should. But using your way, the sky gets pulled down to the light level of the ground (more or less), so metering the whole scene afterwards makes sense too. Certainly if it works for you, why not others? The only time I see a problem is if the grad stops are too low so the light in the sky causes the foreground to get too light in the meter and over exposing. Thanks for the creative concept. I've got to try it. Alan.
  7. Actually, having thought about it some more, there is one objection I can see to doing TTL metering with a GND in place (though it's easy to work around). If your camera does open-aperture metering, as most cameras have for the past 40 years, then the effect of the GND will be somewhat different when metering than when actually shooting with the aperture stopped down. Of course, the solution to this is to use stop-down metering, which is no big deal since you have to stop down before shooting anyway to be sure you have the filter positioned just right.
  8. Wow, thank you all for your input. I will use your tips and see what develops, so to speak. I used a Cokin Grad (tobacco) filter about 25 years ago with color film. I don't recall how I metered the scene but I remember it came out awful and the filter went bag in the bag and was never used again. A few years past and I sold all my film gear in favor of a point and shoot. Since that time I've seen a number of images where grad filters were used with excellent results, many of them in the archives of this site. Recently I have been assembling a new film 'kit' and thought I'd revisit my failed grad filter attempt from years past. I want to try and avoid a contrived look which seems to be a common, easily achieved result. (I've got a shot of my own to prove it) Although I'll try it with color film, my first mission is to shoot black and white.
    I have a polarizer and I'm familiar with the use and results of that filter. That's not quite what I'm after. I suppose I'll apply some of your tips here, experiment a bit and report back. Wish me luck and,thanks again, I really appreciate it.
    P.S. Perhaps if my experiment doesn't pan out I'll start a new thread titled "Worst photographs I ever took"
  9. ‘…why should a GND be any different?’ Unlike a pol, ND or contrast filter, all of which have a uniform effect over the entire frame, a gradND affects only a portion of the frame.
    Mounting a gradND, metering and shooting is a haphazard approach; you may or may not get lucky. Consistently pleasing results is the objective. Better to make a conscientious decision re: the luminance of the prioritized portion of the scene and expose accordingly. Do so by metering (exposing for) the priority area and let the remainder of the scene fall where it may or, for example, sacrifice a half stop reduction of the shadows (e.g. in foreground) in order to retain detail in the high values (e.g. bright clouds) or the inverse may be preferred.
    Craig, please clarify/explain: ‘…the effect of the GND will be somewhat different when metering than when actually shooting with the aperture stopped down.’ Not sure what you’re saying here.
  10. For what it's worth, this is similar to the effect I am looking to achieve. I just ran across this image posted by another member on this site.
    It would seem that this would be a good situation to apply a graduated filter as the diminishing gradient would disappear into the distant hills...........yes?
  11. david_henderson


    This isn't as simple as it sounds because there are several reasons for wanting to use a ND grad , different films with huge differences in the contrast they can cope with and several ways of metering the scene, some of which will easily help you understand the difference in brightness between sky and foreground and some of which will do so only with more difficulty and less precision.
    At its simplest, use a spot meter to read from the average foreground and from the average sky and observe how much difference in brightness you have. Try and use a grad that will create a one stop difference between a blue sky and the foreground for the most natural look. If you don't want the most natural look you can tailor the grad accordingly. Note that this is a simplification.
    You can get soft or hard edge grads to deal with different "horizons"
  12. Thank you David. I wasn't aware you coud get hard edge grads. I suppose if one weren't going to manipulate the image after processing they'd have an even more limited application. I'm waiting for the right weather to get out and experiment, preferably something north of zero. I appreciating your tip about spot metering the foreground and sky seperately to observe what the actual difference in brightness is and will try that. Thanks again.

Share This Page