17
\$\begingroup\$

When a rolling shutter activates it exposes the sensor through a slit (vertical/horizontal) - I can see why every bit of the sensor is exposed for the same amount of time. The "exposure slit" starts from the top and ends at the bottom.

A leaf shutter on the other hand starts exposing the image from the center which would expose the center of the sensor first; it then opens up all the way which still includes the center of the image; and while it's closing down the center of the sensor will still be exposed. It seems that the center of the sensor receives the most amount of light and that is reduced in a radial gradient way across the sensor, correct? (well, it's not even a perfect circle due to the geometry of the leaves)

Do digital cameras have to compensate for this? How did they compensate in the film era?

\$\endgroup\$
3
  • 1
    \$\begingroup\$ So you think f/32 vignettes more than f/8? Better recheck; I've got a lens for my Speed Graphic that vignettes when wide open, but not when smaller than f/16. \$\endgroup\$
    – Zeiss Ikon
    Dec 16, 2021 at 12:11
  • 3
    \$\begingroup\$ A slit shutter is typically a focal plane shutter, i.e., it sits right in front of the image sensor. A leaf shutter is inside the lens, so it doesn't operate on the focused image, but on the as yet unfocused light. \$\endgroup\$ Dec 17, 2021 at 14:00
  • 1
    \$\begingroup\$ Re, what @PeteBecker said, Specifically, the leaf shutter is not anywhere near a real image of the subject within the lens. It is deliberately placed in a plane that is extremely out-of-focus. \$\endgroup\$ Dec 20, 2021 at 0:42

6 Answers 6

16
\$\begingroup\$

There are actually a variety of leaf shutter designs, so a single answer might not be entirely appropriate... but it seems apparent that you are talking about the in-lens aperture diaphragm type.

There is an error in your understanding. The central light path does not expose the central portion of the sensor/negative... e.g. using a small f-stop does not result in a small circular image in the middle of the image area. Instead, the image circle remains the same size and is limited by the lens barrel.

The first thing to understand is that there is all of the light required to create a complete image at every point on the objective lens; i.e. the light from every point in a scene spreads out and hits every point on the lens.

And aperture adjustment is really just changing the exposure by the stacking of multiple images upon the image plane (fewer/more). This type of leaf shutter does not cause mechanical vignetting... the only mechanical vignetting will occur at max apertures where images taken from the edge of the objective element are included; and those images include the edge of the lens barrel (which causes the shading; and is cured by stopping down).

Some other types of leaf shutters can cause mechanical/optical vignetting if they are outside of the lens assembly (more like a focal plane shutter or apodization filter).

Some lens designs can suffer from natural vignetting due to the inverse square law (cosine fourth law) where the light from far off-center portions of the scene have to converge at the center of a small diameter lens and then spread back out widely over a short distance in order to cover the image area... Basically, wide angle lenses with short flange distances are most susceptible. But it is not caused by, nor cured by, stopping down an aperture restriction.

\$\endgroup\$
10
\$\begingroup\$

The best way of approaching this might be by thinking of the aperture. A stopped down aperture does not only project light on the centre of film plane. Instead, the entire film plane is exposed more or less evenly (ignoring vignetting for the sake of the argument).

A leaf shutter acts in a similar way, in that it is very similarly shaped to an aperture and therefore does not obstruct light from reaching the edges of the film plane when the shutter is in a more closed state.

enter image description here

The shutter action is very quick, so for for moderate to slow shutter speeds any changes in DOF due to the shutter are negligible. However, for faster shutter speeds, DOF can be affected by the design of the leaf shutter [1, 2].


  1. https://shuby.de/blog/post/journey-photography/
  2. https://en.wikipedia.org/wiki/Shutter_%28photography%29#Diaphragm_shutter
\$\endgroup\$
3
  • 1
    \$\begingroup\$ "ignoring vignetting for the sake of the argument" - but that's the problem; the leaf shutter causes its own vignetting and this is accumulated over the entire time ~of exposure~ that the leaves are travelling (at different "diameters") \$\endgroup\$
    – adrianton3
    Dec 16, 2021 at 10:43
  • 1
    \$\begingroup\$ @adrianton3 I've made an amendment to my answer \$\endgroup\$
    – timvrhn
    Dec 16, 2021 at 11:03
  • 4
    \$\begingroup\$ A leaf shutter does not cause vignetting, any more than an aperture stop does. All of the lens illuminates all of the film/sensor, unless geometry (either within the lens or inside the camera body) blocks light from part of the open aperture. Vignetting in the above image is probably due to a very wide angle lens, where distance reduces illumination at the edges and corners (superwide lenses for large format often have center neutral density filters matched to them to combat this). \$\endgroup\$
    – Zeiss Ikon
    Dec 16, 2021 at 12:09
4
\$\begingroup\$

I am writing a totally ignorant answer, but let me just make some theoretical assumptions.

As normally a leaf shutter is located inside the lens, it has some distance to the focal plane. So, in the period on which it opens and closes, it works as a diaphragm. It still let the light fall into all the area of the focal plane.

Think about this. An f22 aperture still lets all the light fall into all the frame, the same as an f2.8. I think something similar is happening on a leaf shutter.

The difference with a curtain shutter is that this is located next to the focal plane, so, yes, this produces a very hard shadow when the contains move across the plane. When this movement is "tracked" when you are photographing a horizontally moving object, the result is a rolling shutter effect. But the shutter itself is not rolling, it is moving from top to bottom.

A real rolling shutter is the one used on a film cinema camera, where you need the shutter opening and closing, so you just put a rolling disc as a shutter.

On a digital camera, the rolling shutter effect is not due to the shutter, but how the lines of the sensor are scanned and transmitted. The camera can not process all the sensors at the same time, so it does line by line.

A raw processing software has some vignetting compensation. Some programs have specific information about lens-aperture combinations, so in the case that some vignetting is produced by the leaf shutter, can be compensated the same as any other type of vingetting.

\$\endgroup\$
3
\$\begingroup\$

Ideally, a leaf shutter would open instantaneously, expose the film/sensor, and then close again instantaneously.

For longer exposure times, this instantaneous approximation holds true, as the shutter open/close time is much shorter than the exposure. However, for short exposures, the open/close time is comparable to the exposure time, meaning you can indeed suffer from vignetting and an increase in depth of field associated with smaller apertures.

\$\endgroup\$
4
  • 2
    \$\begingroup\$ The aperture type leaf shutter doesn't/can't cause vignetting... \$\endgroup\$ Dec 16, 2021 at 19:30
  • \$\begingroup\$ @StevenKersting The editor of Wikipedia seems to disagree (en.wikipedia.org/wiki/…). So if you are correct, perhaps that needs updating with the correct information. \$\endgroup\$
    – Matt Dunn
    Dec 17, 2021 at 9:25
  • \$\begingroup\$ I would agree that it can affect the DoF, but not vignetting (stopping down corrects for vignetting). \$\endgroup\$ Dec 17, 2021 at 13:49
  • 1
    \$\begingroup\$ If a leaf shutter is outside of the lens (like a focal plane shutter), then it may cause vignetting \$\endgroup\$ Dec 17, 2021 at 14:11
1
\$\begingroup\$

The vignetting you are thinking about because of different exposure times at the periphery and the center of the leaf shutter when using very short exposure is there, but you are looking for it at the wrong place.

Because of the positioning of the leaf shutter, it acts as an aperture, so you get to see its effect where you can also see the shape of the aperture: in the bokeh and in the diffraction patterns usually determining sunburst shape.

Bokeh circles will not have quite the high-contrast edge focal plane shutters leave in place. As opposed to the sunburst stars from the aperture, however, the diffraction patterns from the leaf shutter get smeared out since the shutter edges change angle throughout its action. This smearing out from the leaf diffraction pattern is also responsible for another artifact at high leaf shutter speeds: a softer image (lower local contrast and softened edges).

\$\endgroup\$
0
\$\begingroup\$

The leaf shutter is in the spatial Fourier plane of the lens. The shape of the aperture with time as it opens and closes will expose different spatial frequencies for different times. The low frequency component is in the centre of the spatial Fourier plane and would get the greatest exposure than the high frequency component. This would result in a slightly less exposure of small details in the image and not usually be noticed.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.