Exercise 3.3 – Timeframes

What do the timeframes of the camera actually look like? If you have a manual film camera, open the camera back (make sure there’s no film in the camera first!) and look through the shutter as you press the shutter release. What is the shortest duration in which your eyes can perceive a recognisable image in bright daylight? Describe the experiment in your learning log.

Because the lens elements can distort the view, I have done this exercise using ‘bare’ shutters. I was intrigued as to whether the result is the same with leaf shutters and focal-plane shutters.

Incidentally (a minor niggling point) I regard ‘shutter’ as a misleading name for a device that is really an ‘opener’.

For the leaf shutter, I took a large-format lensboard and removed the lens elements leaving a bare Synchro-Compur with speeds between 1 sec and 1/400s. I was able to perceive a sunlit scene at all available shutter speeds.

For the focal-plane shutter, I removed the lens of a Pentax Spotmatic SLR. This has a horizontally-travelling focal plane shutter with speeds between 1sec and 1/1000 sec and a flash-synchronisation speed of 1/60s. I was interested in whether I got different results below the synch speed (where the shutter is fully-open for a time) and above (where the second-curtain is released before the first is fully open). Again, I was able to perceive a sunlit scene at all shutter speeds.

I wonder if this counts as a ‘failure’ in the experiment, as I found no cut-off speed at which the image was lost. Perhaps removing the lens made it too easy. (Alternatively, perhaps attempting the exercise with a lens in place, as anticipated by the course notes, distorts the results)

However, there were two interesting observations to be made:

First, the focal-plane experiment is a demonstration of persistence of vision. I saw the whole scene and the whole of the circular lens-throat outline even at the higher speeds where the shutter is never fully open. Effectively, a continuous set of moving slit-images appeared as a whole.

Second, the eye takes time to focus. If I concentrated on the shutter, I could feel my eye muscles adjusting focus to the scene. At higher speeds, it is necessary to ‘pre-focus’ on the distant scene in order to see it properly through the shutter.

Exercise 3.2 – some stuff from my archives

Before leaving Exercise 3.2 and the representation of motion, I will post a few examples of some of my earlier (pre-OCA) experiments.

The first image is not, strictly, a multiple exposure but a sequence of images (1/1000s with the camera in ‘machine gun’ mode) montaged in Photoshop. The focus blur on the rearmost image is accidental but seems to ‘work’. My montaging technique is not perfect, so there is some uneven colour in the sky around the kite-lines.


The next set is of an Irish Coastguard helicopter that was doing training exercises with a car ferry that I was travelling on, which gave me the chance to experiment with the effect of shutter speed on the rotors are depicted. In the first image, at 1/1000s, the main rotor is ‘frozen’ which gives the uncomfortable feeling that the engine has failed and the helicopter is about to fall out of the sky


At 1/400 we see some blur but the rotor is still effectively static. At 1/15s we lose it almost completely. The intermediate images with shutter speeds of 1/100s to 1/50s work best for me.

Slow shutter speeds and moving people can produce results ranging from delightful to bizarre, with quite a lot of ‘interesting’ in between. The two dancer images have exposure times of 0.8 secs, resulting from the very low light in the hall, but which capture the movement of the veils (which is the object of this dance)

In crowd scenes with an exposure of about 1/2s as with the examples below, people walking will blur but not disappear. This sort of image emphasises the stationary people and is a way of emphasising stillness in the middle of bustle.

Shutter speed is critical as there is a very odd effect which occurs at exposure times of 1-2 seconds. No matter how fast a walking or running person is moving, the foot on the ground is stationary. At these shutter speeds, a moving crowd becomes a sort of fog full of disembodied feet, which is very disturbing.

The final image is made with a pinhole camera and an exposure of 5 minutes. This is Maidstone’s main shopping street on the weekend before Christmas. The grey ‘fog is a crowd of moving shoppers. Only the group taking a breather are recognisable.

Maidstone pinhole mk1-003

Exercise 3.2 – working with moving clouds

The images in this posting continue my Exercise 3.2 work, but viewing clouds in motion rather than water. I was fortunate enough to visit the Angel of the North on a breezy day with broken cloud. My viewpoint is downwind of the Angel, with the intention that the clouds would be advancing toward me.

The first set of images, effectively, repeat the water exercise with single exposures at shutter speeds between 1/800s and 40 secs.

Clouds are slower-moving than water and I saw no significant blur at shutter speeds faster than 1 second. The 20- and 40 second images give an impression of motion but introduce a second problem, clouds have rather softer definition than water and the motion blur tends to smudge into a featureless highlight.

In the final two images below, I have used the E-30’s multiple exposure facility to capture four images on a single frame in-camera. A function named ‘automatic gain control’ has handled the issue of total exposure, presumably by reducing the effective ISO sensitivity.


4 exp, 5 secs at 30 sec intervals


4 exp, 1/100s at 30 sec intervals

The second version, with multiple short exposures gives the best definition but a sense of staccato movement. My preference is for the first image, combining reasonable definition with some motion blur.

Exercise 3.2 – shutter speed and moving water

I have long held the belief that there is no ‘correct’ shutter speed for moving water or, to put it better, there is no such thing as an incorrect speed for moving water. Different shutter speeds simply tell different stories. In this post, I examine a wide range of shutter speeds and also two different types of motion.

The first two sequences are at the Lodore Falls in Borrowdale, running a little empty after a period of dry weather. I therefore worked in close to show a reasonable flow of water and to avoid glare as much as possible.

The first sequence was shot at shutter speeds between 1/640 second and 60 seconds.

At 1/640, my intention was to freeze the motion, for comparison with the slower speeds. Individual water droplets are frozen; there is confusion but I get no sense of motion. At 1/200 and 1/100 it is possible to make out individual droplets together with enough blur to show there is motion and some (admittedly small-scale) violence.

At the intermediate speeds, the movement is  shown by streaks rather than blurred droplets. The one that works best for me is 1/25 s.

The three slowest images were made using an ND10 ‘big stopper’ filter. As the exposure time increases, the details in the moving water smooth out. There is a sense of ‘flow’ in the falling water, the splashes have disappeared or form a light mist, and the froth below the stone takes on a milky appearance, with most of the detail disappearing between 5 and 20 seconds. This area is confused rather than flowing.

The second set of images takes a wider view, with an obvious direction of flow. Shutter speeds were between 1/250s and 60 seconds. A polariser was used to minimise glare by ‘killing’ some reflection from the rocks behind the water. The lower fall and the rock are those seen in the first sequence.

Again, the fastest shutter speed freezes the motion. In this case, there is enough detail to be interesting. However, my preference is for the images between 1/100s and 1/10s, where we see both detail and motion blur and get a feeling of flow and tumult. With the longer times, particularly at 60 secs, there is a feeling of ‘flow’ but the individual droplets are lost and the impression is much more peaceful. The appropriate speed depends on the message one wants to convey. My favourite from this set is the image made at 1/8s.

In the third set, I look at a water surface moving in the form of ripples but with no overall flow. Shutter speeds were between 1/1250s and 60 secs

Tonal differences between the top and bottom row may arise from the use of an ND10 filter for the bottom row. We see individual ripples in all the top row images, with blur starting to intrude at 1/13s. With the two faster images on the bottom row, we know that the surface is rippled but is being smoothed out; I get the best impression of motion a 1.6 secs. In the two longer exposures, the motion has been smoothed out to a ‘frosted glass’ texture. There has also been time for clouds to pass over and their shadows to average-out, so the overall tone is smoother.

I am aware that bigger waves have a longer period and would require correspondingly longer exposures to get a similar effect. The ‘surface’ would also appear more as a layer of mist.

Overall, there appear to be three ‘zones’ of effect arising from shutter speed. Very fast shutters ‘freeze’ the movement and show detail, but can look a bit artificial. Medium shutter speeds (say 1/100s to 1 sec) give the best impression of movement and tumult. The very slow shutter speeds, seconds or minutes, average-out the random fluctuations (thereby appearing more serene) and show the underlying flow or stillness of the body of water.

Exercise 3.2 – Camera movement

Most of the course paper examples show the camera moving, rather than the subject. Broadly, these movements fall into two categories, accidental and deliberate.

The Robert Capa image from Omaha Beach on D-Day is an example of accidental camera movement. Capa’s famous maxim, ‘If your photographs aren’t good enough, you’re not close enough’ (Magnum 2014) was, perhaps taken to extremes by joining the invading forces in the landing craft. Slideshow of images here. Several of his images have blur and/or tilted horizons. As viewers, we understand that the photographer was really there, dodging bullets and having nothing to shoot back with but a camera, and we share his experience vicariously. In this case, the imperfections contribute to the sense of authenticity and involvement.


FRANCE. Normandy. June 6th, 1944. US troops assault Omaha Beach during the D-Day landings.

The traffic image by an un-named OCA student (course notes p65) also shows accidental camera movements resulting from the circumstances in which it was taken. In this case, the relevant circumstance is that the photographer suffers from Parkinson’s disease and the image is at least as much about her involuntary hand movements as it is ostensibly about the traffic on Kings Road.

In most cases, however, accidental camera movements are caused by photographer error, setting a shutter speed too slow to hand-hold securely.

Deliberate camera movement is likely to be some form of panning. With a moving subject and a static background, the photographer moves the camera to match the subject movement, resulting in a sharp subject and blurred background. This technique is common in sports photography, particularly motor sport, but has been used in other contexts. In another of Robert Capa’s images a woman runs for shelter during an air raid on Barcelona in the Spanish Civil War. The background shows movement blur, as do her legs, giving us a sense of hurry, if not of panic.


SPAIN. Barcelona. January 1939. Running for shelter during the air raid alarm. The city was being heavily bombed by Fascist planes as General Franco’s troops rapidly approached the city.

The opening chase sequence of Chungking Express (YouTube clip here) uses a similar technique, particularly in the chase sequence between 1:25 and 1:45. The camera, with a slow shutter, follows the cop and the fugitive, leaving the background and extras blurred. Combined with frequent cuts (no clip is more than a couple of seconds long) this gives a sense of movement, confusion and involvement. This is similar to the ‘hand-held action’ sequences that have become a cliche in action movies such as the Bourne and later Bond films, but is handled well in our example.

We are also directed toward Chris Marker’s La Jeteé (opening sequence) (balance of movie) which is told as a sequence of still images. My own reaction is to view the film as an AV sequence; in some parts there is a sense of time passing, but not of movement.

Deliberate camera movements are used in other ways, for instance the camera club and magazine cliche of vertical panning in a forest, described by Steve Gubin (2013) among others. Although it does not convey movement (at least, not to me) it does give an impressionistic view and vertical emphasis to forest scenes.


Gubin, S. (2013) How to use the Vertical Pan Technique for Special Effects [online] at: http://stevegubin.com/blog/2013/12/how-to-use-a-vertical-pan-for-special-effects (accessed 13 May 2016)

Magnum Photos (2014) Robert Capa [online] at:http://www.magnumphotos.com/C.aspx?VP3=CMS3&VF=MAGO31_9_VForm&ERID=24KL535353 (accessed 13 May 2016)

Exercise 3.2 – Subject movement and shutter speed

An earlier posting, Slices of time, looks at the effect of freezing subject movement with a fast shutter. My conclusion is that the sense of movement is lost, even if the subject is in a statically unsustainable position such as mid-air.

This post looks at some examples of blur due to subject movement and a slow shutter.

The first category results from shutter speeds that are hand-holdable but too slow to freeze a fast moving subject. This includes one of my favourite images, by Sergio Larrain during his time in London in 1958-59. The girl and her surroundings in Trafalgar Square are still, but the pigeons have been recently disturbed and their blurred wings give a sense of movement,flurry and confusion.

Blur of this type can be seen in some sports and concert reportage as a result of high subject speed or low light respectively. It can also be used as a stylistic preference, for instance by Robert Frank in ‘Elevator – Miami Beach 1955’ . The lift (‘elevator’) attendant appears sharp as part of the fixed scenery while the hotel guests have somewhere to go, and are moving and blurred as a result. Taken with her bored expression and eyes directed upward, we feel something about the drudgery of her job.

Longer shutter times require the use of a tripod to hold the camera steady. Fixed objects are rendered sharply while moving objects are blurred.  The technique is used with water and clouds often enough to have become a cliche. Cloud example by Michael Kenna. When the moving elements have an one-way movement, as with clouds, waterfalls or rivers there is a feeling of flow. Where the movement is cyclic, such as waves on water, then long exposures tend to even it out and the sense of motion is lost. Example here.

Long exposures can reveal movement that is not obvious to the naked eye, for instance star trails.

The course notes refer to the images of Hiroshi Sugimoto, who photographs inside cinemas, opening the shutter at the beginning of the film and closing it at the end. Frankly, these images leave me cold. The screen appears as a blank white rectangle, which is the effective light source for a view of the cinema interior. There is no sense of motion or of time passing. Example here.

Michael Wesely uses even longer exposures, measured in weeks or months, which can capture the whole progress of a building project. Examples on this page. This gives a good impression of time passing, but not of motion – his subjects are essentially static, although growing. The bands of light in the sky represent the track of the sun, varying from day to day in the same way as a Justin Quinnell solargraph.

Exercise 3.1 – The frozen moment

This post represents a bit of monolithic dual avicide. First, there is the exercise itself, using fast shutter speeds to isolate a frozen moment of time in a moving subject. Second, it gives me a chance to discover something that I have long been curious about, just what does the beginning of a stream of water running from a tap look like?

This was my first attempt. The set-up appears simple, an outside tap and an Olympus E-30 on a tripod and set to ‘machine-gun mode’ (rapid sequence). I would start a camera sequence and turn the tap on and off while the camera is running. As always, the devil is in the details.


As noted in the brief, there is a trade-off between shutter speed and ISO. My first set of images were taken with the kit 14-42mm zoom at the 42mm end (84mm full-frame equivalent) and its widest aperture, f/5.6. The image above was taken at ISO800, 1/320 second, which is not fast enough to freeze the water emerging under pressure. I also tried 1/800 at ISO1600 and 1/1250 at ISO2500, which was too noisy. After that I switched to the 50-150 telephoto zoom which opens to f/2.8 at the short end and allowed me shutter speeds of 1/1600 and 1/2000.

The second issue is timing, which is largely a matter of luck. Although 1/1600 is fast enough to freeze the water flow, the 5FPS sequential shooting speed is not great for capturing the fast-moving leading edge. Here are a few images in full flow, and we see that it is not a simple symmetrical cylinder of water.

Another issue was getting the full stream in focus with a shallow depth of field. This is the reason for the later images being more side-on, but including the white pipe in the foreground.

With the tap turned off, the last dregs of water fall under gravity and rather slower.

And, yes, I did manage to freeze that leading edge.

This final image is my favourite, from 150 total shots.