The sky is a white blank and the trees under it are a black mass. Nothing is wrong with your exposure: the scene held more light than the sensor could record in one frame, so the camera clipped both ends to make it fit.
Dynamic range is the distance between those two failures. Learn to estimate it before you raise the camera and you stop taking photos that were never going to work.
What is dynamic range in photography?
Dynamic range in photography is the span between the brightest tone a sensor records before it clips to pure white and the darkest tone it records before detail vanishes into noise. It is measured in stops, also written as EV, and each stop is a doubling of light. A scene spanning 10 stops has roughly 1,000 times more light in its brightest area than in its deepest shadow.
Two dynamic ranges are in play on every shoot, and confusing them causes most of the frustration:
- Scene dynamic range is how much contrast is actually in front of you, from sun on a white wall down to the shade under a parked car.
- Camera dynamic range is how much of that contrast your sensor records in a single exposure.
When the scene fits inside the camera, one frame holds everything. When it doesn’t, something gets sacrificed, and the only question is whether you choose what that is or the camera chooses for you.
How many stops of dynamic range does a camera have?
Modern full-frame cameras measure roughly 14 to 15 stops at base ISO. DxOMark’s Landscape score puts the Nikon D850 at 14.8 EV at its ISO 64 base and the Sony a7R V at 14.8 EV at ISO 100, the top of their database. Good APS-C sensors come close behind, and Micro Four Thirds sits a stop or two further back.
Those numbers are optimistic. Bill Claff’s Photographic Dynamic Range measurements at Photons to Photos rate the same cameras about three stops lower, because DxOMark counts shadows down to where the signal equals the noise while PDR counts only shadows that still hold a usable signal-to-noise ratio at normal viewing size. Plan around the PDR figure: the bottom two or three stops of any camera’s rated range are recorded in the technical sense and too noisy to publish.
What is the dynamic range of a typical scene?
Scene contrast decides whether you have a problem, and it varies far more than camera dynamic range does. These are rough working figures, close enough to judge a scene by eye:
| Scene | Approximate range | In a single exposure |
|---|---|---|
| Studio with controlled lighting | 5 to 7 stops | Fits easily, the ratios are yours to set |
| Overcast day, flat light | 6 to 8 stops | Fits with room to spare |
| Golden hour landscape | 8 to 10 stops | Fits on almost any camera |
| Midday sun, subject in open shade | 10 to 12 stops | Tight, highlights need watching |
| Backlit portrait at midday | 12 to 15 stops | Face or sky, pick one |
| Interior with a bright window in frame | 15 to 20 stops | Beyond any single frame |
| Sun inside the frame | 20+ stops | Beyond any single frame |
Your eye takes in roughly 10 to 14 stops at any single instant, which is not far off a camera. Its advantage is adaptation: as your gaze moves, pupil and retina re-expose locally, pushing the effective range past 20 stops. That is why a room with a sunlit window looks normal while you stand in it and turns into a white rectangle in the photograph.
Base ISO gives you the widest dynamic range
Camera dynamic range peaks at base ISO and falls by roughly one stop for every stop of ISO you add above it. Base ISO is 100 on most bodies, 64 on the Nikon D850 and the Z7 series, and 125 on recent Fujifilm X-Trans sensors. Shoot a 14-stop camera at ISO 3200 and you are working with something closer to 9.
One useful exception: many recent sensors use dual conversion gain, a second readout mode that switches in partway up the ISO range, commonly between ISO 400 and 800 depending on the body. Dynamic range recovers at that switch point instead of continuing to fall, which is why ISO 500 can be measurably cleaner than ISO 250 on Nikon’s Z-series bodies. Knowing where your own sensor switches changes which high ISO settings are genuinely safe.
RAW keeps the dynamic range that JPEG discards
A JPEG is 8-bit with a contrast curve already baked in, so it holds roughly 8 stops no matter what the sensor captured. A 14-bit RAW file keeps everything the sensor recorded, which is why a sky that looks solid white on the back of the camera often still has cloud detail sitting in the file.
One point trips up a lot of people: bit depth is not dynamic range. A 14-bit file has 16,384 tonal steps per channel, which describes how finely the range is divided, not how wide it is. A 14-bit file off a 12-stop sensor still holds 12 stops, just in smaller increments. The argument for shooting RAW vs JPEG in contrasty light is recovery, not step count.
The histogram is caught up in this too. It and the flashing highlight warning are generated from the embedded JPEG preview rather than the RAW data, so they clip about a stop before the RAW does. A flat picture profile with contrast turned down makes the preview track the file, so you stop surrendering exposure you never needed to lose.
How do you protect highlights while shooting?
Expose for the highlights, because clipped highlights are gone permanently while dark shadows are merely noisy. Four habits do most of the work:
- Turn on the highlight warning. The blinking areas are a clipping alarm, and reading them costs half a second per frame.
- Spot meter the brightest thing you care about and place it about two stops above middle grey. Anything brighter, like the sun itself or a glint off chrome, is supposed to clip.
- Expose to the right when the scene allows it. Pushing exposure as far right as the histogram permits without clipping puts more signal in the shadows and less noise in the file. On an ISO-invariant sensor the gain is small, so never risk a blown highlight for it.
- Use the two settings that reach the RAW file. Canon’s Highlight Tone Priority (marked D+) buys about a stop of highlight headroom and raises the minimum ISO to 200. Fujifilm’s DR200 and DR400 underexpose one or two stops and lift the tone curve back up, which is why they require at least twice and four times base ISO: on an X-T5 based at ISO 125, that means ISO 250 and ISO 500.
What do you do when one exposure cannot hold the scene?
When a scene runs three or four stops past what the sensor can record, no exposure choice saves it and you switch to one of three fixes.
Bracket and merge. Auto exposure bracketing shoots a burst at different exposures, typically 3 to 9 frames spaced 1 or 2 EV apart. Merging them in Lightroom or Photoshop produces a DNG that holds the combined range and still edits like an ordinary RAW file. Most high dynamic range photography is done this way now, and it looks nothing like the crunchy HDR of ten years ago.
Filter the bright half. A 2 or 3 stop soft graduated ND filter darkens the sky and leaves the ground alone, solving the problem in-camera. It works beautifully on a clean horizon and badly on a jagged one, where the transition rides straight over mountains and rooftops.
Add light to the shadows. A reflector or a fill flash at around quarter power raises the dark end rather than taming the bright end, and it is the only fix that survives a subject who moves between frames. For a backlit portrait it beats bracketing outright.
Sometimes the honest answer is a different hour. A landscape at golden hour runs several stops flatter than the same view at noon, which is a technical advantage as much as a stylistic one.
Judge the scene before you meter
Look at the brightest thing you care about and the darkest thing you care about, and ask whether they are more than about 10 stops apart. Most of the time they are not, and you can simply shoot. When they are, you have caught it while a tripod, a filter, or a flash can still fix it, instead of finding out at the computer with a white sky and no second frame to merge.
