How Important Is Optimizing Your Exposure In-Camera?

Edited Fujifilm X-T4 JPEG.  +1 sharpening applied in-camera.  (click to enlarge)

A couple of weeks ago, I was wondering something that had crossed my mind many times in the past—that is, how critical is exposure with today's digital interchangeable lens cameras?  In other words, optimizing exposure and, as is often said, getting it right in-camera?  Could I put my camera on aperture priority, zero out the exposure compensation dial, just shoot away and get not only usable files but good image files?  

I've always extensively used the exposure compensation function in my cameras and paid close attention to the histogram (always displayed in the EVF) to ensure I maximized my image files' quality.  But does it really make a significant difference if I didn't?  Not knowing for sure, but suspecting today's sensors, exposure metering systems and image processing algorithms make optimum exposure less critical than it has been in the past, I set out to do some short tests to find out for myself.

The question I sought to answer was, "How important is optimum exposure?  Is it a big concern, moderate concern or no concern at all?"  Additionally, I was curious as to how each camera handled exposures that had no adjustments.  Would the cameras tend to overexpose, exposure pretty accurately or underexpose?  To judge dynamic range and "correct" exposure, I decided that JPEG files would give me the most accurate results.  In my experience, different RAW converters will show the same file in a variety of ways depending upon who is writing the code.  Lightroom may show a RAW file one way but Capture One may interpret it another way.  But a JPEG file is created in the camera and will be displayed as made by a particular camera.

So I thought I would conduct an experiment.  My experiment was non-scientific, not extensive, not covering all situations and can't fully tell me everything.  (I wasn't willing to put in that much time and effort)  But it would tell me enough to satisfy my curiosity.

I took all three of my cameras, the Nikon Z7 (14-bit files), a full frame camera with close to the best available dynamic range, a Fujifilm X-T4 (14-bit files) with dynamic range about one stop less than the Nikon and the Olympus E-M1 Mark III (12-bit files), again, with a stop less dynamic range less than the X-T4, and made a number of exposures of a variety of subjects in both bright daylight with harsh shadow as well as bright overcast light.  I did not photograph indoors. 

The top of this old inboard boat motor was in direct sunlight and the lower parts in deep shadow.  The
Nikon Z7 JPEG file captured all of it with no issue.  This file was edited to raise the shadows and tone
down the silver highlights to show the detail was not lost. (click to enlarge)

I went out on three different days.  To really push the sensors on the first two days, I went out at near midday (most contrast/harshest shadows) of bright sunny, no clouds, solid blue sky days and found some bright direct sunlit white painted subjects which also had dark shadows in the same scene.  I photographed with the sun behind me, at 90 degrees and with my subjects backlit.  In other words, really pushing the dynamic range of any sensor.

On the third day, I chose a day that was bright but overcast and found additional subjects with bright white objects coupled with dark green foliage.

I set all the cameras to their base ISOs, which optimized dynamic range.  I set the cameras to record RAW plus JPEG so I can see the histogram differences between the two file types and if exposure made a significant difference with either file type.  I also set exposure compensation to zero.  All three cameras had their meters set for Matrix or Multi, etc.  I set the Nikon to f/11, the Fujifilm to f/8 and the Olympus at f/5.6.  I know that matching depth of field really doesn't make a difference in this test, but I wanted all images to have a similar look.  Additionally, white balance was set on Auto as I wanted to see how each camera, in the exact same circumstance, handled white balance.

So what did I find?  Without going into a lot of technical stuff, the answer to my first question is that getting right in camera is of  'moderate to minor concern.'  That was for all three cameras.  With only a couple of exceptions the sensors in all three cameras captured all highlights and shadows in JPEG.  All was captured with the RAW format.  In the instances when a bit of highlights were lost in the JPEGs, it was very little an in non-critical areas.  That was only with the Olympus, which has two disadvantages—the smallest sensor and only recording with 12-bits.  The Nikon and Fujifilm didn't lose any shadow or highlight detail.  

One area where a bit of highlights was lost was some specular highlights on chrome (which don't really count), another the very edge of a white painted surface in direct sun and the third the flat silvery top of an inboard boat motor where the bottom of the motor was in deep shadow.  There was no textured highlights lost.  No shadows were lost.  That said, I'm sure there are situations one could encounter that are different and 'worse' photographically than this which may impact the highlights and shadows more.  Again, my goal wasn't to try to set up a comprehensive test covering tens of different situations but find scenes that would normally be a good test of a camera's dynamic range and see how my three cameras fared.

To move on, if you are shooting RAW, you probably won't lose any highlights or shadows as the camera's meters (at least mine) put you right in the middle of the histogram.  If you like shooting JPEGs, you might want to pay a little attention to your in-camera histogram when shooting really high contrast subjects.  In fact, in Lightroom Classic all JPEGs and all RAW files showed all data within the boundaries of the histogram, with the exception of specular highlights.  The RAW histograms were generally more compact and the JPEGs often went right to both left and right edges, but still without clipping.  

This Olympus JPEG held the cloud detail in the bright overcast sky as well as kept the detail in the
dark foliage and shadows on the right.  (click to enlarge)

So, how do the cameras generally expose?  One of the first things I noticed when I looked at the files in Lightroom Classic was that each camera consistently underexposed the different subjects between 1/3 and 1 full stop from what I would call optimum brightness.  However, it depended upon the individual subject and subjects with larger areas of bright white received less exposure as you would expect.  This is the result of the tendency for meters to try to make everything 18% reflectance, I would surmise.  Matrix metering is good but not foolproof.  Also, by optimum exposure, I mean when I looked at the files in Lightroom, I needed to brighten them by that much to look correct to my eye.  

I used the JPEGs to judge this as Lightroom's interpretation of RAW files can vary with manufacturer.  I suspect that my cameras' metering is programmed that way in order to 'protect' the highlights.  More specifically, the Fuji and Olympus mostly underexposed by about 1/3-2/3 stop and the Nikon about 1/2-1  stop.  However, I found these exposures acceptable, in my opinion.  The slight underexposure allowed me to recover really bright highlights when dynamic range almost exceeded the sensor's capability.  Overall, the meters did a very good job of selecting an exposure that kept the highlights intact even when there was deep shadow.   All needed editing.  None would look good straight out of the camera.

As for auto white balance, all three cameras were very close in their white balance selections.  Still, I had to adjust each camera's white balance in Lightroom as they all were off just a bit.  Auto white balance gets you close, but to me, not close enough.

In summary, for 95% of the time (at least with my cameras and photographing the kinds of things I photograph), I feel comfortable that I could just set my aperture, not worry about exposure compensation, and shoot away.  I think the cameras give you a pretty good 'middle of the road' exposure so your shadows and highlights are pretty well protected.  There is enough leeway, even in the JPEGs, to be able to fully edit your files to make them pleasing to the eye.  All files looked 'dingy gray' when first opened in Lightroom and all had to be brightened.  All files looked very good after editing and, with the leeway today's sensors provide, you won't lose much in quality, if anything.  That said, I recommend you still pay attention and not be lazy about exposure.

Now, don't take this post as me endorsing not paying attention to your histogram, not optimizing your files by exposing-to-the-right (ETTR) or being sloppy in your technique.  Not at all.  After all, 50% of all the information contained in a digital file is in the brightest one f/stop of light.  Twenty-five percent more is in the second brightest f/stop.  For the best quality images, with the least amount of noise in your shadows and mid-tones, it is wise to ETTR.  However, I'm satisfied that if a rapidly evolving situation occurred and I didn't have time to optimize my exposure, my cameras would do a very good job of getting it pretty close.  And, that is what I wanted to know.

Join me over at my website, https://www.dennismook.com.  

Thanks for looking. Enjoy!  

Dennis A. Mook  

All content on this blog is © 2013-2021 Dennis A. Mook. All Rights Reserved. Feel free to point to this blog from your website with full attribution. Permission may be granted for commercial use. Please contact Mr. Mook to discuss permission to reproduce the blog posts and/or images.