Going into this exercise, I was particularly curious about Michael Wesely‘s work and wanted to do something that would capture the essence of movement over time. His technique of opening the shutter for several days at the time is not easy to reproduce with a digital camera, so I decided to experiment instead with multiple exposures over several hours.
I tried several sequences at different lengths. The camera was set on a tripod with a 90mm lens pre-focused on my cat, who would take long naps every morning. The first sequence was for a total of 20 frames, at a rate of 1 frame per 10 minutes (3 hours 10 minutes in total). Shutter speed was set at 1/200s for each frame and the aperture fluctuated in mid values of f/5.6 ~ f/7.1 for most of the frames. ISO was fixed at 800. The camera was set to add every picture and average the exposure, so that movement is built progressively. In that set-up, the final photograph, shown below, is an average of itself and all the previous 19 photographs:
The camera is able to store all the intermediate photographs, so you can see the build-up process leading to the final image. In this case, “build-up” is not perhaps the most appropriate term, as the cat starts perfectly defined in the first image and is progressively blurred until it becomes unrecognisable. I have prepared a short video to show the construction of the sequence, which can be found here.
My next sequence was for approximately the same length of time, but at a faster frame rate: 100 pictures were taken at a rate of one image every 2 minutes, for a total of 3:18 minutes. The camera was set just like in the first sequence. In the first sequence, the cat continued present in the frame for the entire shot-out, but he got up and left about 1/3 into the second sequence, and as a result the final frame, averaged with the previous 99, shows very faint traces of the cat, reminiscent of the more subtle photographs from Wesely, where there is almost no movement.
The build-up video, which shows the cat vanishing through time, can be found here.
——————— ooo ———————
I was also intrigued by the work of Maarten Vanvolsem and how he tries to capture movement, rhythm and acceleration in his shots. My initial attempts at this were rather crude and primarily consisted in opening the shutter for a specified period of time while walking on an escalator, going through the ticket barriers of the Underground, or simply standing on a carriage while traveling between two stations. The results are attractive studies on light and motion, but they do not convey the rhythm that you can see in some of Vanvolsem dance shots.
I then experimented with several of the line scan apps that are available for mobile phones. Some of these are set just to give a distorted image and not really to capture rhythm or acceleration over time. The ones that worked the best for me were “Slit-Scan Maker” (link) and “Poloska” (link). Both offer a still slit-scan mode – where the line of pixels that is captured is fixed and the picture is made by moving the phone – but the duration of the clips in Poloska can be extended for longer periods of time, resulting in long panoramic shots:
The Slit-Scan Maker app offers and additional mode, which is that of the moving slit. This operates like a regular document scanner, in as much as the row of pixels that is captured travels through the frame from top to bottom. When combined with movement from the person holding the phone, this mode has the effect of compressing movement vertically, like if the long panoramas from Poloska (see above) were squeezed down, as can be seen in the pictures below.
All the above images were taken while walking to work early in the morning. The images taken with the Slit-Scan Maker delineate rhythm quite dramatically. In “Walking over the yellow like”, I was holding the phone just above the yellow line on a busy train terminal platform while I walked towards the exit alongside hundred other commuters. The pendular movement of walking is nicely captured as a zig-zag of the line. A similar, but perhaps more subtle effect can be seen in “Crossing the road”, where the different colours of the pavement tiles swirl in line with the movement of the photographer.
——————— ooo ———————
While researching the work of Dr Vanvolsem, I came across the so-called Marey-effect and from there I found about the work of Harold E. Edgerton with stroboscopes to capture the rapid movement of subjects in sequences within the same frame (see link to some of these images here). I set to try to replicate this at home with the aid of three flashes mounted on three separate cameras set next to each other, all pointing directly at the subject. The flashes were set to manual mode and adjusted so that they all had the same intensity. The central camera was set to run in bulb and to open the shutter for 10 seconds. With the aid of three remote timers, I programmed the cameras to shoot one after the other with about 2 seconds difference between them. The end result, when all was in sync, was three short flashes within a 9 second exposure. The subject (myself) was placed about 1 meter in front of the camera and the lens was pre-focused and set to manual mode. ISO was set to 100 and the aperture to f/16 to ensure the subject was in sharp focus and as the background was darkened. To make sure that only the flash light was illuminating the scene the shots were taken in a dark room with no ambient light. The subject moved his head through the 9 second exposure and such movements were frozen by each of the three flashes. Here are some of the resulting images:
While the results were close to what I expected, there were some issues that I would like to improve in future experiments if possible. The first one is that the room where I took the pictures was relatively small and I was constrained by the lowest manual setting of 1/4 power in one of the flashes I had on hand. Ideally, I would have chosen less flash power and a longer room, with more distance between the flash and the end wall so that the background wall was less illuminated or pitch black if possible. As it was, because the background was iluminated by the first flash, the image of the head moving sideways appears less defined (because is a mixture of the background and the head). One additional aspect I would want to try is more flashes in a longer shot. This could perhaps be achieved by using a small strobe light like the ones used in disco parties. The problem with this approach is that the intensity of these strobes is probably too strong and cannot be regulated like that of a manual flash.