DJO wrote:
Thank you Alan. Someone gets it. I think you've said it much better than I. Yes I've used a light meter in a camera. Someone thought I never had. But no matter how fancy your camera is or how much it cost, it is still REFLECTED light. A reflected light reading is an assessment, a complex calculation in some instances but an assessment nonetheless, of the actual amount of light. An incident light reading IS the actual amount of light. There seems to be some oversight of your using the word "fooled" in quotation marks. Of course the light meter isn't fooled; it's a machine. It's just not always giving accurate information. This is why, as many have shared, photographers devise methods of
compensating for the inaccuracy. Some are experts at doing this and I truly admire their skill. To each their own. Sometimes I think the best method I ever heard of is "f8 and be there". I do, however, find it hard to believe that a sophisticated and experienced photographer does not know the meaning of subject tonality.
Thank you Alan. Someone gets it. I think you've ... (
show quote)
I think in this point of view, you are considering only one or possibly two modes that in-camera meters work, and therein lies the confusion and misunderstanding.
If you set your meter to a "calculating" mode - in Nikonspeak that would be 3D Dynamic or Dynamic - yes, it is a complex calculation, which in some cases even adds facial recognition, etc to the mix. It can result in good exposures but it can also result in bad exposures.
When you set the meter to spot or average or center weighted, there is no calculation, other than the case of center weighted, where the center is given a priority over the rest of the scene. This was quite useful with film to get in the "ballpark" but not always reliable.
Using the in camera meter to simply record the intensity of reflected light or the difference between highlights and shadows is immensely valuable in high contrast settings.
I have given numerous examples but I will try again.
Take a subject, like a black bird with a white head, and the sun is shining. I shoot 3,000 images of eagles between January and March that fit this description. I also shoot another 2,000 images of Common Terns, Skimmers and Oystercatchers in their breeding grounds. White bird, nearly black head, or vice versa.
I have a choice. I have used an incident meter to establish exposure. Sometimes I can't because I am under the shade of a tree and the bird is over open water with no shade. My results with incident reading the light was accurate for the scene - however in measuring only the light and not the contrast between the middle tone value and the highlights, I got results like the first image.
The second image was done using my typical method of reading the highlights using the in camera's spot meter function and adding 1 stop to the camera's reading.
The difference is that I know the camera's limit for bright tonal capture, and could read the highlight I needed detail in and use my judgement to add in the correct amount of extra exposure to take the tone of the eagle's white plumage from gray to white with detail, without loss of highlight information.
The first image, exposed according to an incident meter, totally blew the highlights.
How is this possible? By reading only the light, and not taking into consideration the reflectance of the subject, you can in high contrast scenes not introduce compensation to ensure that the highlights, which are more than a couple of stops brighter than the reading generated by the incident meter.
The third is another case, where the subject is in two types of lighting, half in shade and the other half in bright light. I read the highlight and added in 1 stop more to properly record the highlights, which are in sunlight, while the dark body of the bird is in shadow - a very wide contrast range, and nearly impossible to get accurate with an incident meter.
Ok, you may argue that this is not landscape. You are right. Here is a landscape example. Waterfall is in sunlight, I am standing under trees with the light filtered by the canopy, and I can't easily get to a place to measure the light falling on the waterfall.
These are fairly commonplace situations, in addition to all the others I posted, where it is important to understand light, reflectance, your gear's capabilities, and how to put it all together to make a good exposure. Blindly relying on an incident meter is not good photography. Knowing when to use the correct metering approach and how to use it correctly, be it incident, reflected, spot, or using the camera set to average and placing an ExpoDisc on the lens and using it as an incident meter - all are valid approaches in the right situation - that is good photography.