October 23, 2010

Long Exposures

This blog appeared on the Singh-Ray site on Tuesday, October 19, 2010

Cole Thompson explains how he uses long exposures to create his mystical black and white images

Fine-art photographer Cole Thompson creates his images and visual essays in dramatic black and white. His images involve the use of very long time exposures of 30 seconds to several minutes. “I have always been intrigued by monoliths, first by the statues on Easter Island, then by the monolith in the film 2001: A Space Odyssey and most recently by Stonehenge in England. In each case, seeing these monoliths prompts the question; Who built them and for what purpose? I’ve always loved visiting Bandon Beach in Oregon because of the natural monoliths strewn along the coastline. Randomly placed, it is as though the earth were God’s chessboard and the monoliths the pieces from an unfinished game. I’ve included several images of these Bandon Beach monoliths to illustrate my use of long exposures for my various monochrome print portfolios.

“I have often created my monolith images using long exposures of 30 seconds, but recently I found myself experimenting with daytime exposures of up to 5 minutes. In each case, whether 30 seconds or up to 5 minutes, the Singh-Ray Vari-ND filter was essential to creating these images. If you’ve ever tried such exposures with a fixed ND filter, you know the challenges. First, to compose you have to remove the filter because the filters are so dark you cannot see through the viewfinder. Second, when removing and replacing these filters you can accidentally change the focus or the focal length of your lens. Often you’re not even aware of these lens movements until later, when it’s too late.

“When I mount the Vari-ND filter on my lens, I dial it to the minimum (Min) setting. At this point the filter has already added about 2 f-stops of density which is just enough to darken the image in the viewfinder a bit, but I can still see clearly enough to compose and focus accurately. I then rotate the front ring to increase the density by as much as 6 additional f-stops. When I need even more density, I stack the Singh-Ray Mor-Slo 5-stop ND Filter in front of the the Vari-ND to achieve up to 13 stops of density. This allows me to use very long exposures, even in full daylight.

“Here are some of the things I’ve learned about using very long exposures. First a tripod is a must and a remote shutter release is desirable. While it’s true that slight camera movements do not seriously affect the very long exposure, it’s best to eliminate vibrations as a “best practice” and because camera movement can ruin an image with a 5 second exposure or less. I shoot at ISO 50 at F22, put my camera into RAW/monochrome mode and set the exposure mode to manual.

“The most frequent question I am asked is how I determine my exposure. I use the in-camera meter for exposures up to 30 seconds, and beyond that I find I must extrapolate because the digital SLR’s I use only meter up to 30 seconds. With my camera set to F22 and 30 seconds, I dial my Vari-ND’s density ring until my exposure is correct. To extrapolate for exposures longer than 30 seconds, I do the same thing but set the exposure to 2 f-stops below ideal and then quadruple the exposure in my head. If 30 seconds is what I’m metering for, then quadrupling brings me to a 120-second exposure.

“But strange as it seems, if I expose at 120 seconds, the image will be very underexposed. So I’ll generally expose at 300 seconds for a good exposure. I’m not sure why the 120 seconds doesn’t produce the correct exposure, but it could be due to some kind of digital equivalent to ‘reciprocity failure,’ which is a phenomenon experienced when shooting extra long exposures with film. The reason for such reciprocity failure with film is that the longer the exposure, the less effective the film is in recording the light and so the exposure length needs to be increased. In the case of my digital camera, when the meter says I need a 120 second exposure I extrapolate that reading and give it 300 seconds. It works for me.

Something else that is very important when using the in-camera meter is that I must completely isolate stray light from coming into the viewfinder and affecting the meter reading. To do this I use a Hoodman eye-cup which allows me to seal the viewfinder with my eye. This is essential for a correct exposure.

“Shooting long exposures at the beach creates a couple of additional challenges. Waves hitting the tripod legs will cause them to sink into the sand, ruining the exposure. I can either move out of the water line or build wide feet for my tripod that give it a larger footprint. The beach also has strong winds and a lens makes a great wind sail. I combat this by using a heavy tripod that allows me to hang a weight onto the center column to steady it. I also position myself between the wind and the camera and often turn my jacket up above my head and use it as a wind shield.

“Lastly, let me repeat that camera movement is the primary enemy of long exposures. Even the slightest movements can ruin the image. Sometimes it’s just not easy to detect this movement by checking the camera’s display. For this reason, I check every image with a hooded loupe which enables me to see the image clearly even outdoors.

I have found extra-long exposures to be especially appealing to me, perhaps because they help convey nature as timeless. I’ve come to a point where the technique has become part of the message in my work. Choosing the length of the long exposure will control how that movement looks in both water and sky. Exposures from 2-30 seconds can give a completely different look in water while the longer exposures measured in minutes are usually needed to produce dramatic skies.

There is no better tool than the Vari-ND to produce these types of images. As I’ve said before in my other Singh-Ray posts, this filter not only makes it easier for me to create these images, in many cases I could not have produced them without the Vari-ND.”

To learn more about Cole’s work check out his previous stories on this blog and pay a visit to his website.

 

at 11:24 AM 0 comments

Labels: , ,

 

 

16 thoughts on “Long Exposures

  1. Hi Cole. Nice article. Thank you for showing us your insights on long exposure.
    As yourself, I had a really hard time figuring out exposure time with ND filters. I hate doing math. Until this FREE iPhone app came along.
    Look into the store for this app: LongTime Exposure.
    Best app EVER!

  2. Thanks everyone, I do enjoy the long exposure. Many express surprise that I would reveal my “secrets” but the techniques are not the secret, it’s the vision you bring to a project.

  3. Okay, the question is asked why you quadruple the exposure when you are two f-stops underexposed and you want a correct exposure.

    It’s really all about math! For every f-stop you open up on your camera, you are letting in twice as much light and for every f-stop you close down, you are reducing the light by half. The same is true with the shutter speed, if you change from 1/60 to 1/125 you are letting in half as much light and to go from 1/125 to 1/60 you are letting in twice as much light.

    That’s why f8 at 1/125 is the same as f16 at 1/60. The f-stop change reduced the light by one half and the shutter speed change increased the amount of light by double. They canceled each other out and both let the same amount of light in.

    Okay, so now for the extrapolation. My meter only reads to 30 seconds but I want a several minute exposure. So I set my camera at f22 and 30 seconds and my meter shows that I am underexposed by 2 f-stops. I need to let more light in to get a proper exposure, but how much more?

    To go from -2 f-stops to -1 f-stop, I need to double the amount of light I let in, so I go from 30 seconds to 60 seconds. Now I want to do that again to get from -1 f-stop to optimal exposure and so I need to double the amount of light again, going from 60 seconds to 120 seconds.

    So in total I’ve gone from 30 seconds to 120 seconds to compensate for those 2 f-stops I was short. Quadruple.

    And then as I point out in the article, when light is this faint, it becomes less effective and you have to let even more light in than your calculation says. This is called reciprocity and it doesn’t really affect exposures under 30 seconds. For over 30 seconds you have to experiment a bit to get the exposure right (in the film days we had reciprocity tables to refer to).

    Thank goodness for digital with it’s zero cost exposures and the immediate feedback!

    Cole

  4. Cole, great work as usual. I bought the ND110 filter recently but have not had time to play with it yet. Couldn’t get the Sing-Ray at the time, out of stock 🙁

    The following link has a helpful table for working out times for long exposures and like you recommends adding 2-3 stops if the time is getting over 30 seconds, actually thinks in terms of greater than 6 stops from the filter. http://www.bwvision.com/photography-techniques/the-ultimate-guide-for-long-exposures-part-3/

    Look forward to seeing more of your long exposure work. Thanks for sharing.

    Phil

  5. Thanks for the link Phil. What I’ve still yet do determine is if there is really such a thing as digital reciprocity, and if there is, do the film reciprocity tables apply?

    Cole

  6. Cole, here is my theory on why the math does not stand up for very long exposures with digital. My background is electronics engineering so I will take it from this angle. However I must state that I have no experience with camera electronics so this is just a theory.

    With all electronics there are compromises made in circuit design to get optimal operation at specific points for the product. I expect in cameras the design is for shutter speeds between 1/4000 sec through to 30 sec.

    The reason I believe that the time blows out is because the sensor is basically a lot of photovoltaic cells, each cell produces an electrical output based on the intensity of light hitting that cell. The voltage difference between bright light and dull light is probably only millivolts. This is then passed to the electronic circuit that converts this into a digital image.

    Now here is my theory – components used in the circuit that is collecting this light value do not work in a linear fashion, so the circuit designers work to optimize and make it as linear as possible over a time period that is most commonly used. (1/4000 to 30 seconds). When you exceed this you then start getting into the edges of the electronic components operation and outside the optimized design. The engineers would be taking into account noise and colour sensitivity as well so the design would get quite complex.

    I remember this very well back in the days when I built amplifiers for a hobby and worked in radio communications design in the late 70’s.

    As I mentioned this is just my theory but thought I would put it out there…

    The way to solve this is as you say, experiment and take lots of photo’s, easy to delete the ones that don’t work 🙂

    Phil

  7. Thanks Phil, now we have two possible explanations. The bottom line is that we may not know why, but thankfully we have the immediate feedback of digital and can certainly experiment and figure it out.

  8. I’ve been exploring for a bit for any high quality articles or blog posts on this kind of area . Exploring in Yahoo I at last stumbled upon this site. Reading this info So i am happy to convey that I have a very good uncanny feeling I discovered exactly what I needed. I most certainly will make sure to don’t forget this site and give it a look regularly.

  9. As far as I’m aware there is no reciprocity failure in digital capture. I have found metering without a filter, to get a favourable histogram and then doubling the exposure time for each stop of my filter which is 14.5 stops. I used to do it with a calculator but now the Long Time app is available, I use that. i’m thinking you may be a stop out in your calculations? Using the histogram you can also calibrate your filter to get the exact time needed for a perfect exposure. Thanks for the information you provide here Cole, very refreshing stuff!

  10. Lee, I just cannot explain it. When I meter and extrapolate, the exposure is always way off, too dark. I almost think it’s a failing of the light meter in very low light situations.

Leave a Reply

Your email address will not be published. Required fields are marked *