Tuesday 16 August 2011

Exercise 28: Measuring Exposure

Light meters measure the intensity of light My Canon 5DMkII has manual exposure adjustment so I am able to complete this exercise.  In relation to photography they produce an exposure value (EV), which is used to determine an exposure setting given in terms of a lens aperture/shutter speed combination.  My 5D MkII has a (TTL) through-the-lens meter this measure the intensity of reflected light entering the lens.  Reflected light meters are calibrated to 'see' everything as 18% grey.

The first part requires a number of images which have been deliberately over or under exposed.

Photograph 1: Over Exposed
Canon EF24-105mm f/4L IS USM Lens, 65mm, f/14, 1/80s, ISO 640
The image above is of one of my daughters stuffed toys.  This shot has been deliberately over exposed by 0.67ev.  The reason for this was to capture the texture of the teddy's 'fur'.  The dark brown fur hides a lot of detail if exposed 'correctly' or as metered.  The camera doesn't have the range of the human eye and the 'fur' doesn't render well.  The over exposure puts the detail back with the detriment of the white label and ribbon being blown out a little.

Photograph 2: Over Exposed
Canon EF24-105mm f/4L IS USM Lens, 93mm, f/14, 1/30s, ISO 3200
The above picture of my daughter was shot into a very light window.  To achieve this image a decision had to be made on what to expose correctly my daughters face or the background.  As the subject is my daughter I exposed her face correctly and this in turn required me to over-expose the background.  I stupidly left the cameras ISO mode on auto so therefore it compensated a little with a high ISO and therefore this image has a little noise.

Photograph 3: Under-exposed
Canon EF24-105mm f/4L IS USM Lens, 35mm, f/16, 1/40s, ISO 160
I wanted to emphasise the slightly dull sky and silhouette the outline of the tree for this shot.  I therefore took this image at -0.67ev with the detriment of the detail in the shadows under the tree.

The second part of the exercise involved bracketing a number of shots.  Bracketing is the process of taking two or more near-identical images at exposure values above and below the metered value.

The course notes suggested taking making five exposures around the metered value the first one stop darker the second half a stop, the third average, the fourth half a stop lighter and the fifth one stop brighter.

I shot all of the images using the evaluative metering (multi-segment metering) mode of my camera.  This works by measuring the levels of brightness in different sections of of the viewfinder to assess the brightness range and calculate an appropriate exposure value.

1 stop darker
Half stop darker
The cameras meter has worked well for this image of washing on a line.  half a stop either side of the exposure are still usable however.
Metered

Half stop brighter

1 stop brighter




1 stop darker
Half stop darker
Metered
Half stop brighter














1 stop brighter

This image of an statue in a garden works best at half a stop brighter than the meter suggested.  I think this brings out more detail in the figure rather than the green foliage which is not of interest.  I think the camera metered for an even exposure over the image but this image was to great a contrast.  I think the metered image could be used along with the full stop brighter image.






Again this shot of a chair in a garden works best at half a stop brighter than metered.  I think this is due to the complicated contrast in the image.




Another difficult image for the camera meter.  Again I think the half stop brighter image is the best and most accurate photograph.

This exercise has taught me that multi-segment metering is usually ideal for occasions when you want to spend little time worrying about exposure and are happy to allow the camera to make assessment of the lighting conditions.  Canon's 5DMkII multi-segment metering system is highly accurate even in quite complex lighting conditions, as is proven by the above images.  However, the  image is still based on what the camera is 'seeing' and not what I am trying to achieve with exposure.