Hi, new forum member here. I'm a casual photographer who is looking to improve my skill and technical knowledge. Last week I overheard a couple of people in a heated discussion at the camera store. One was saying the histogram should be pushed to the right and the other said no, the histogram should be centered. They both seemed pretty insistent that they were right. Which is correct or are they both right?
the true answer is that there is no correct exposure. You have to look at the scene and expose it the way you want it.
There's another way to think about that. Unfortunately the term "correct exposure" can cause more arguments than ETTR. Describing it as you do here sounds like film photography language. Film had a tone curve response that had be factored in. A digital camera sensor however records data linearly. Save a
raw file (this only applies to working with raw files) and it's a blank slate -- you get to make it whatever you want. And in that case I can use this term: "best possible technical exposure." A best possible technical exposure exists and it's the exposure that fully utilizes the camera sensor's recording capacity.
(We can tack on to the end of that last sentence the qualifier "to the degree possible" because circumstance (motion capture, DOF, low light) may force us to expose less than the sensor's capacity. In that case the best possible technical exposure becomes: expose as much as possible.)
Any exposure less than the full recording capacity of the sensor records less data and lower quality data. I'm prepared to call less and lower quality incorrect. Any exposure more than the full recording capacity of the sensor records less data and is clipping highlights in the image. I'm prepared to call less and clipped highlights incorrect.
Here's some examples (two photos taken walking around the neighborhood):
Those are RawDigger histograms of the raw file data. Exposure info is included. Both photos are from the same Fuji X-T2. Both photos were taken at the camera's base ISO (200). The actual exposures are 1/600, f/5.6 for Mary and 1/600, f/8 for the heart house. However note that there's an EC .3 factor for Mary and an EC 1.7 factor for the house.
Look at the histograms and note the magenta bar. I added that. The magenta bar indicates the saturation limit (sensor capacity). Look at the stats to the right of each green histogram and note the Max value of 15360. The theoretical limit is 2^14 for a camera with a 14 bit ADC. Each camera will be close but slightly different (talk to the engineers). 15360 in the green channel is the saturation limit for that Fuji X-T2 camera and
here's the point: I just reached it. Each photo (taken in sunshine) contains some very small specular highlights (reflections) and those should make it to the limit of the sensor's full capacity.
Shape and distribution of the rest of the histograms? Doesn't matter as long as I have usable data -- in both photos I do. It's very rare that I wouldn't with a modern camera.
So the camera exposures are different but as far as I'm concerned they are also both the same exposure in that they are both a full capacity sensor exposure which is the best possible technical exposure. So every photo I take gets the same exposure: Place the brightest diffuse highlight at the sensor saturation limit -- click, and for me that's always the correct exposure.
If most of the photo is in shadow or darkness then your histogram will look mostly all on the left because that’s where all the information is. If it’s a very right scene then it’ll be mostly on the right. If it’s equal amounts it’ll be in the middle.