Is Optimal Waterfall Shutter Speed a Function of Flow Rate?

Discussion in 'Nature' started by larry h., Jul 31, 2003.

  1. I know that the shutter speed used on waterfalls is a matter of personal taste and the only thing I can do is experiment. I have also read other threads on this topic, including this good one on the use of ND filters (particularly Stephen Hinch's comments): http://www.photo.net/bboard/q-and-a-fetch-msg? msg_id=004vjy&unified_p=1 But I am looking for a general guideline of where to start my exposures. Last weekend, I took some photos of my favorite stream and its waterfalls. The photos varied from 17-35mm taken from f/8 @ 1/8 sec to f/22 @ 1 sec using Fuji Superia Reala and a polarizer (I will use Provia 100F in the future, but with its reduced latitude, will everything wash out?) Having read other threads, I was thinking 1/8 sec would be way too short and wanted to get longer than 1.0 sec, but couldn't, even under full shade conditions. After getting my prints back, I found that I liked the shorter exposures better, and maybe would've liked shorter than 1/8 sec even better. This is because all detail was washed out of the longer exposure. I am going to try to attach a dual photo showing the same scene at 1/8 and 1 sec to show what I mean. (This is just a simple flatbed scan of the prints -- my purpose here was not to produce a quality scan.) If it doesn't succeed, please just bear with my description. With the longer exposure, there was so much turbulence, all I got in large areas of the photo was pure white, no texture at all. With the shorter exposure, you could see the little dips and eddies in the waterflow. There's still plenty of blur, but it's less washed out. My main question is: Have others of you experienced the same thing? Is there a rule of thumb for shutter speeds based on waterflow or other parameter? Does it matter how close you are to the waterfall? My goal is not to have to bracket all the way from 1/60 sec to 1 sec on each shot to see which I like best. Besides, 1/60 at this light level and 100 speed film would have required f/2.8, which would not have been as sharp or have adequate DOF (I'd rather not use faster, grainier film). In short, I was just shocked that I preferred the shorter exposures. My other question is related. I have read of a technique where you underexpose the shot by 2-3 stops and take 4 or 8 multiple exposures of the scene. Has anybody tried this and do you like the results?
    005dPt-13836984.JPG
     
  2. Sorry for the long question and photos, but here's another example.
    005dQR-13837184.JPG
     
  3. Short answer: no.

    Long answer: the quality of motion blur is dependent on the "passing by speed" [angular speed] of the object relative to the motion of the film. (which you might pan ...)
    With a long lens, this angular speed increases, with a wide lens it decreases, i.e., less water moves less of a distance on the film or sensor. Just think of taking a picture of a speeding train: head on, you can possibly get by with 1/125 sec, while at right angles the same train will require 1/2000 sec to freeze it, since it moves so much more rel. to the film at right angles.

    And for that dreamy, creamy waterfall pic, you could easily go up to or need 10 or more secs to get it right.
     
  4. I have shot from 6 minutes to fractions and this is what I usually do- always a polarizor, Velvia, not Provia and never print film, and try to stop down to get about 2-4 seconds worth of exposure. If you have to go longer switch to E100VS for reciprocity failure. Now most of my falls are shot in the tight SOuthern Appalachian Forests, so if your shooting in the wide west this may change. Also pick a very bright overcast day for best results. Good luck, and here is a link to my waterfall page. http://www.kendunnphoto.com/waterfalls.html
    005dYt-13840484.jpg
     
  5. I have become so tired of the blurred look in waterfalls that I shoot at 1/30 or faster. Much more interesting IMO. People have been shooting waterfalls at slow shutter speeds for 150 years now. Time for something different.


    Kent in SD
     
  6. I like the look of both the blurred effect as well as a more "frozen in time" effect. If presented with a waterfall, I would probably take some images with varying effects.

    Sorry I can't help with the technicals. I have to take a lot of waterfall shots before I get one I like and haven't really gotten it down to a science yet.
     
  7. Yes, Kent, they have. But people have been drinking beer for longer than that, and frankly, I'm not tired of either one yet. A little variety is usually a good thing, however.
    005ddH-13844684.jpg
     
  8. 19 exposures, each of 1/4 second. 121mm Schneider Super Angulon f:8 at approximately f:80, Ilford FP4+ 5x7". Yes, the multi-trick works.
    005dnJ-13847184.jpg
     
  9. Ole, I can't see your image all that clearly on my screen, but I can't see that you've accomplished anything a single shot could not have gotten. Can you explain? Would you mind e-mailing me the photo? Thanks.
     
  10. Interesting answers, a question where hydrology and photography meet. In general, I would say once you reach a certain (slow) shutter speed, the image of the flow will blur almost all streamflow, not including waterfalls. But yes, the velocity of the water will determine the fastest shutter speed you want to freeze or the slowest to get the flowing image you want.

    The velocity of water is generally over a narrow range, such as 2 to 6 feet per second, well within a few shutter speeds, so the difference appears insignficant, but some streams and waterfalls do have high velocities requiring really slow shutter speeds. Try stopping the water in a long, vertical waterfall compared to a small stream over rocks, your example. Or trying getting the smooth flow from the same water fall compared to a really slow moving river.

    Everyone has different techniques for capturing the flow, such as bracketing over several to many shutter speeds. I like to take the full range, from about 1/125th to as much as 8 or 16 seconds, which often includes using ND filters to get the light down for the long exposures. I've found the range produces interesting images to really see small differences.

    Good luck.

    --Scott--

    Scott M. Knowles
    Hydrologist, MS-Geography

    "All things merge into one, and a river runs through it."
    - Norman MacLean
     
  11. This is sort of a non-technical answer to this tech question, but one key difference between your original offerings, Larry, and some of these other images is the relative amount of the frame taken up by the bright water areas. In composing to get the really silky water, that water becomes a textural element, and if the bright white, detail-free areas take up most of the frame, it may be worth shooting shorter exposures to retain more detail. Just a thought...
     
  12. Congratulations, Ole, you've demonstrated that calculus applies to photography.
     
  13. It's not velocity and it's not just angular velocity that determines blurring on the film plane. It's the (vector) velocity component projected on the film plane. Suppose V is the object velocity perpendicular to the optical axis and f1 = the object distance from the lens (from the lens front nodal point), f2 = image distance from the lens (from the rear nodal point, but just think of the nodal points as being in the center of the lens), and f =focal length. Then:

    Velocity on film plane = V*(f1/f2) where,

    1/f1 + 1/f2 = 1/f

    The blurred distance on the film is then V*time = V* shutter speed

    This gives you what you need to calculate blurring. Actually, I have ever calculated blurring myself. I prefer to keep it a right brain activity and think about the composition and intuitively choose a shutter speed. It would be a good exercise though to choose a typical water velocity (say 5 ft/s), a typical focal length, and then make a table of blurring versus shutter speed and object distance. This could then easily be scaled to calculate the blurring on a typical print size. I think that I will do that. It might help fine tune my intuition.
     
  14. Whoops. The velocity equation should read:

    Velocity on film plane = V*(f2/f1)
     
  15. I don't mean to bore y'all (yes Ken, my shots are from the Southern Appalachians, too, N. GA) with more of my shots. But I posted these because I think that Marshall may be on to something about how the white parts of my pictures may be taking up too much of the photo. So here are two of the same shot as my second pair, but at 17mm. I like the turbulence of the 1/8 sec exposure because it shows the power of the rapid flow of the water. But do more people like the featureless white of the 1 sec exposure? I welcome other comments on the artistic merit of my pictures, too, not only technical comments. Ken, your Ramsey Creek shot is an example of why I asked about flow rate, not just stream/angular velocity. That shot looks like there was very little flow over the rocks, and it took a long exposure to get a significant amount of wisps. (Rich's also looks like a very low flow rate.) Is that correct? This is what led me to wonder whether my favorite stream just has too much flow to be good photographically. I agree with Kent that the blurred effect has gotten tiresome. But I'm trying it while I am learning what I like. That's why I also plan to try the multiple exposure idea. So, Ole, do you have other examples of multi shots? Maybe some where each exposure is less than 1/4 sec (my 35mm SLR won't go to f/80 ;-) and/or where you are closer to the water? Finally, Scott, you may be the best qualified to answer my technical questions (and I like your Minolta expertise as well). But I admit I don't quite understand what you are saying, despite a little background in hydrology and fluid mechanics myself. Can you clarify a little for me? I do like your suggestion to try the whole range of shutter speeds. Perhaps I will vary the speeds by two stops at a time (a factor of four) as I cannot tell much difference when comparing two shots varying by only 1 stop, say 1/4 and 1/2 sec. Please keep the comments coming. I am learning a lot! Larry
    005dwI-13851684.JPG
     
  16. pvp

    pvp

    The correct answer (to your original question) is "no." Different shutter speeds will give different effects, and it's the photographer's job to previsualize the final shot and choose accordingly. What works for you today might not be what you want tomorrow.
    005dxT-13852684.jpg
     
  17. Larry - I actually like the feel of the bottom frame better, though I understand your point about seeing the turbulence in the top frame. Even in the bottom, I think it would be stronger with more of a shape around the highlights of the flowing water (the right side feels a little dark and empty), but then again, I'm only seeing the small jpeg. In some of the other images here, it feels like it's a picture of the place, with flowing water as a strong element, which I like. But it's all just my opinion; when you want to show the turbulence and the shape of the waterfall, you now have the technique to do so. Enjoy.
     
  18. I now carry a high end digital camera where I can control shutter speed. I can do test shots on the digital camera and see the results. That shows me the shutter speed that yields the result that I want. Then I can shoot my "real" images on my film cameras and can get the results i want with little or no bracketing. I should say that I have used the "intuition" approach successfully but sometimes walk away from a waterfall second guessing myself. Now I can walk away from the shot knowing that I have on film what I wanted.
     
  19. If you don't like the lacy/silky effect of long exposures and still wish to avoid the frozen-in-time result, try a series of short exposures on the same frame. Perhaps a dozen or more of 1/250 sec shots. Most persons will wonder how you got the picture.
     
  20. This is getting to be one of the better discussions in recent times. As to the stream velocity and blurring, the comment about the perceived stream velocity, on the film plane, is correct. This is something I forgot to think about. The blurring depends on the position of the film plane to the stream or waterfall, hence, as stated, it's the movement on the film from the streamflow and the shutter speeds that controls the blurring effect.

    I'm not sure it's a simple equation because stream velocity varys at any moment, across the stream and in angles going downstream. All this translate to small distances on the film and compared to the shutter speed. An extreme example is looking directly at the top of a waterfall (no vertical component). The flow will effectively be still. Then looking at a waterfall at some distance parallel to the falls will produce the maximum possibilities.

    An additional factor is the position of the film plane to the surface of the stream or waterfall flow, very similar to the position and tilt of a large format camera. The movement then calculates from there on the film plane. You could calculate it out, but then the flow is continuously varying which changes everything. How many photographers take scientific calculator with them?

    As to shutter speeds, I guess I take more shots than many. I usually start at 1/125th and increment to 8 or 16 second exposures. The fastest speeds usually stops almost all flow and turbulence (individual drops frozen in time). Once you get to seconds, I've noticed you get to the mist effect, which is why I take longer exposures, to see the difference, and who knows what you'll get. Sometimes this takes using ND filters to drop the shutter speed below what the film, lens and camera will allow.

    This often take 1/3-1/2 a roll of film, and I have to admit, this is one place a digital camera would be a benefit where you can immediately see the results, and adjust from there. This is one discussion I'll print, view the photos, read the comments, and ponder more, next to a waterfall of course. And maybe consider taking velocity measurements next time too to compare.

    --Scott--
     
  21. ShunCheung

    ShunCheung Administrator

    Instant feedback with digital has many advantages. However, even though you think you have figured out what the "optimal" shutter speed at a particular situation, I would still take a few more shots at (very) different shutter speeds just to get a few variety in case you change your mind later on and prefer a different effect.
     
  22. Ultimately it is going to depend on what you want to communicate with the image. Though some may be tired of seeing the silky waterfalls, I still think they portray a more serene environment than freezing the water in place, which usually shows 'power.'

    If you have some fine details within the stream, exposing for too long tends to wash them out (ie. >1 sec). I tend to like the 1/8 - 1/2 second range. Just like with exposure, you will want to bracket a bit (as you have already done.) So the 'optimal' shutter speed is this - the one you like best!
     
  23. Kent Staubus is my hero! I think it was the great Paul Caponigro who popularised the longer exposure time for flowing water, and then Pop Photo and others began presenting it as a brilliant technique that"shows motion", and now it's become virtually mandatory. Sometimes it works, but only sometimes. I think if the shape of the blurred water adds something to the shot, it works. The same people who practive this technique religiously probably freak at a blur in any other image. Be still, troubled waters.
     
  24. Ya'll are blinding me with science.

    This may be simplistic and non-scientific but one of the factors that I look at is the delicateness and repeatability of small feathered flows .... as opposed to a forcefull non-repeating flow. The latter is hard to capture at slow shutter speeds without milking out.

    Rich's 2nd image is a great example of smaller feathered flow.
     
  25. It depends on the waterfall... each fall/cascade has a different character and personality. The problem is it's hard to pre-visualize the results of different shutter speeds. It also depends on the light at the time... harsh or soft. For slower shutter speeds I use slow film instead of filter, or if I want really long exposure times I can use moonlight at night. A tripod is obviously needed, but sometimes I like using faster film in full sunlight so I can handhold the camera and freeze motion. A few falls look good with limited blur... like I said, they are all different. Here's one I did in the shade, setting the camera at just the right angle so the blue sky is reflected off the small waterfall.
    005f49-13890884.jpg
     
  26. Wynn Bullock used long exposures in the fourties and fifties to effectively blur ocean scenes. Some of his images make the surf look more like fog than breaking waves and are among my favorite photographs of all time. This was long before Paul Caponegro, who made and continues to make (with the help of Photoshop) great images. I doubt that Bullock was the first person to blur water motion, considering the slow film speeds used in the old days. But perhaps he was one of the first to do so intentionally for artistic effect.
     
  27. Thank you to all who have answered. I just got back from my week of photographing waterfalls. When I get my slides developed & scanned, I will show you my best results. But this is what I gleaned from earlier posts. Despite my effort to scientifically analyze waterfall shutter speeds, the most enduring comment in my brain last week was the guy (I can't scoll back to find your name, sorry) who said that perhaps my problem was that I had too much water (white space) in my photos. So I backed up and got more background in the photos. I also tried the multiple exposure technique on the most forceful waterfalls, including my favorite waterfall where I used 50mm, 85mm and 85mm + 1.4x converter. And yes, I still think that flow rate has something to do with it, so I found some waterfalls where the flow rate was smaller to get less milked-out areas.
     
  28. I liked the answers that said that the optimal shutter time varies.
    It varies with the velocity of the water, the turbulence of the water, the effect desired by the photographer. It depends on the obstructions and eddies in the water, standing waves, and any other details which might be desireable or undesirable to the photographer's vision of the subject. That said, I also liked the answers where they said Digital. If you can't come back often, I think that a wide time-bracket with a digital camera can show various effects better. Perhaps an additional problem is that I cannot possibly see details on the digital camera that I can see when I download to the PC, so bring a laptop to preview a serious shot! That's my summary of my current opinion, which is, of course, always subject to change.
     

Share This Page

1111