cshimokita
[Photo Forum Moderator]
10714
Big wheel keep on Rolling... a technical-ish look at shutter effect
Recently "cazalea" from the Seiko Forum was in Japan to visit (surprisingly) two of Seiko's facilities. In order to do that he had to take the Shinkansen to Morioka... along the way he snapped the following photo out the window of the 300 kph bullet train.
When I first saw the leaning utility pole I thought of the famous 1912 photo by Jacques Henri Lartigue depicting a racing car and people along side the road.
The 1912 photo is a result of a technology that is well understood and somewhat easy to explain... somewhat. In order to do that, we should first understand the operation of a camera's shutter. I will briefly cover three types of shutters, the diaphragm or leaf shutter, the focal plane shutter, and the rotary disc shutter.
Diaphragm or leaf shutter: is constructed by using a number of blades somewhat similar in appearance to the aperture arrangement used in modern lenses. Ideally this type of shutter opens and closes quickly, in a circular pattern to expose the film for the duration that it is open. For fast shutter speeds the opening and closing time can become longer relative to the exposure period and may cause vignetting. Examples are the Synchro-Compur and the Japanese Copal shutter.
Rotary disc shutter: used in motion picture cameras, a rotary disc with a pie shaped opening is located between the lens and the film. The shutter angle indicates the amount of the disc that is open. The shutter has a mirror surface that reflects the image to the cinematographer's eye piece when the film is covered... at the same time the film advances one frame. The primary reason that cinematographers adjust the shutter angle is to control the amount of motion blur that is recorded on each successive frame of film. A tight shutter angle will constrict motion blur, and a wide shutter angle will allow it. A 180° shutter angle is considered "normal".
Focal Plane Shutter:
The focal plane shutter sits behind the lens (when viewed from the front of the camera) and just before the film. Early use focal plane shutters used a single curtain with multiple, different size openings. The selected a shutter speed, would deturmine which opening size to used. The shutter moved at approximately the same rate for each opening and thus the size of the opening, together with the aperture selected, determined the exposure. As the slit opening moves across the film, the exposure is accomplished. Notice that the entire frame does not get exposed at the same instant.
multiple slits - one slit per shutter speed
to allow for a full range of shutter speeds
Whew, let's take another look at Lartigue's photo from 1912. The first point is that Lartigue was able to keep the driver almost distortion free while the race car passed by. This was because he panned the camera, albeit slightly slower than the moving car... The second point is that he was using a camera with a Focal Plane shutter and the exposure swept up from the bottom of the image to the top... thus the bottom of the image was exposed first and the top last having the following effect:
a) the feet of the people along side the road are fixed during the initial phase of the exposure... but as he paned the camera to his right, the upper part of the individuals moved progressively toward the left edge of the frame. Thus the people appear to lean left.
b) the bottom of the tire is exposed first... the tire (car, and contents) are moving a little faster than the camera is panning, thus the top of the tire is slightly right of where it would be if he matched the speed of the car and thus the tire appears oblong. The driver looks to be distortion free, but his head (like the top of the number "6" and the tire) is in fact right leaning.
Note: For this photo, Lartigue was thought to have used an ICA camera, possibly a Spiegel-Reflex-Klapp foldable plate camera that may have used a Rouleau (focal plane) shutter. The camera used was quite possibly a view camera where the image is inverted and thus the focal plane shutter moved from camera top to bottom / image bottom to top. The lens might have been 150mm f/4.5 or 120mm f/4.6.
Which brings us back to the 300 kph bullet train...
For that photo of the utility pole "cazalea" was using an iPhone 6 and the EXIF data indicated a shutter speed of 1/1883 second. In effect the photo is a panned shot of a pole at somewhere around 300 kph maximum-ish... and given that shutter speed the train would have moved about 12cm give or take, far less than the displacement of the pole top to bottom... someone is not telling the whole truth...
That something might be the "Rolling Shutter effect".
When the entire frame is captured at one time the camera is said to use a "global shutter". Rolling shutter is an image capture method (still or video / film or digital) that scans across the composition e.g. line-by-line vertically and horizontally within each line. Not all parts of the image are recorded at exactly the same time. The results can be similar to the focal plane distortion as seen in Lartigue's photo. The term Rolling shutter is more commonly associated with digital devices rather than film; digi-cams and dSLR still cameras with video seem to be prone to exhibit this type of distortion, somewhat mfg/camera model dependent.
Just for fun... (I know that it sounds a little strange at this point, but)... The video rate of the iPhone 6 is 60 fps. Let's assume that's the maximum sustainable cycle rate. At 300 kph the pole would displace 138cm top-to-bottom in 1/60 second, which seems to match the photo... Does that imply that each pixel or set of pixels are collecting for a duration equal to the shutter speed, but the capture method takes 1/60 second to move thru a cycle? ... fantasy time...
A big thank you to "cazalea" for sharing his photo and allowing me to use it in this write-up. BTW, Jacques-Henri Lartigue took more than one photo in his lifetime, some of them not bad 
.
Thanks for looking,
Casey