Dynamic Range in Cameras: Video, Film, Photo Explained

Reading Time: 11 minutes

Published: December 18, 2025 | Last Updated: December 19, 2025

You notice dynamic range when the frame includes very bright areas plus areas that are close to black. A face in shade with a bright sky behind it is a classic test. Either the sky clips, or the face gets noisy once you lift it.

A night street with neon signs plus dark doorways is another test. Neon can clip while the doorways turn into chroma noise and blotchy texture if you push them too far in post.

If the scene has more contrast than your camera can hold in a single exposure, you must choose what detail to protect. You can protect highlights, protect shadows, or reduce contrast in the scene with fill light, negative fill, ND filters, flags, diffusion, or changes in blocking.

Human vision vs a camera sensor

People often say the human eye “has more dynamic range” than a camera. That can be true in practice, but the reason matters.

Your eye adapts as you look around. It changes sensitivity over time, and your brain combines what you see into a stable impression. In a single instant view, your eye’s usable range is much smaller than the full adapted range.

This explains a common frustration on set. You can look from a bright window to a darker face and feel like you see both, but the camera must capture both at the same time in one exposure, with no adaptation.

What dynamic range looks like in real footage

Dynamic range problems usually show up as two failures: highlight clipping and shadow noise after lifting.

Highlight clipping is lost texture

Highlight clipping happens when a bright area exceeds the sensor’s maximum recordable signal. Texture turns into a flat patch. Cloud structure disappears, window frames disappear, and white fabric turns into one solid tone.

Example: an interview next to a window. Expose for the face and the window can clip. Expose for the window, and the face lands too low, then gets noisy when lifted.

Shadow noise shows up when you lift dark areas

Shadow noise appears when dark tones sit near the camera’s noise floor. When you raise those shadows in post, you raise noise too. You can see chroma speckle, blotchy patches, macroblocking from compression, and smeared texture from heavy noise reduction.

Example: a dim room lit by one practical lamp. If you expose to keep the bulb from clipping, the corners of the room may show blotchy noise or macroblocking once you lift them in the grade.

Crushed shadows and blown highlights

Two common words for these failures are crushed shadows (dark detail collapsed to black) and blown highlights (bright detail collapsed to flat white or tinted clipping). Both are detail loss, just on opposite ends of the range.

Why this happens inside a sensor

Digital cameras have a top limit and a bottom limit:

• Top limit (saturation): the photosite is full and can’t record more light, so highlights clip.
• Bottom limit (noise floor): the signal is buried in read noise, pattern noise, or quantization noise, so shadows fall apart.

Clipping is not always neutral white. One color channel can hit saturation first, which can shift color in bright areas before the image looks fully blown out.

Stops are how we describe dynamic range

Dynamic range is usually measured in stops. One stop is a doubling or halving of light. More stops means more room between clipped highlights and noisy shadows in a single exposure.

Stops, explained fast

+1 stop means twice as much light. -1 stop means half as much light.

If your scene has more stops of contrast than your camera can hold, you will lose highlight texture, see shadow noise, or both.

Dynamic range vs exposure latitude

Dynamic range is the total brightness range the system can record with usable detail.

Exposure latitude is how far you can overexpose or underexpose, then push back in post, while still getting acceptable results for your delivery and workflow.

They’re related, but not identical. Latitude depends heavily on codec, noise reduction, debayering, highlight recovery tricks, and how ugly you consider the last stops to be.

Why published dynamic range numbers don’t always match

You will see different stop numbers for the same camera. That usually comes from three things:

• Different usable thresholds: highlights clip at a clear point, but shadows fade into noise gradually, so testers must pick a cutoff.
• Different recording pipelines: RAW vs log vs baked looks, plus noise reduction and sharpening.
• Different codecs: heavy compression can break shadows when lifted (macroblocking, banding). If you want a practical handle on that side of the problem, start with what video bitrate is.

One practical rule

Compare dynamic range numbers only when the testing method is similar. Then run a short test in your own codec and grade, so you know what is usable for your delivery.

Representative camera dynamic range table

How this table was made (important): I did not personally lab-test these cameras. I synthesized representative data from publicly available sources. I relied mainly on manufacturer pages (marketing and engineering claims) and CineD’s lab tests (IMATEST-based measurements). Where a camera has multiple modes, measured values can vary. The lab column is included as a practical comparison point, not an absolute truth for every mode and workflow.

How to read the two key columns:

• Manufacturer DR or latitude claim is the brand’s published messaging. This can be optimistic, measured differently, or describing latitude more than strict dynamic range.
• Lab-measured DR (stops, SNR=2) is a lab-style result using a specific noise threshold to decide where shadows become usable. Change the threshold (SNR=1 vs SNR=2), and the stop count changes.

Camera	Segment	Sensor	Manufacturer DR or latitude claim	Lab-measured DR (stops, SNR=2)	Notes (mode and workflow matter)
ARRI ALEXA 35	Cinema	Super 35	“17 stops” (latitude claim)	15.1	CineD ARRIRAW test at ISO 800 reports 15.1 (SNR=2), 16.3 (SNR=1).
ARRI ALEXA Mini LF	Cinema	Large format	Not stated	13.4	CineD reports 13.4 (SNR=2), 14.5 (SNR=1) in ARRIRAW.
Sony VENICE 2 (8.6K)	Cinema	Full-frame	“16 stops” (latitude claim)	14.7	CineD X-OCN XT test at ISO 800 reports 14.7 (SNR=2), 15.9 (SNR=1).
Sony BURANO	Cinema	Full-frame	Not stated	12.4	CineD shows multiple results by mode. One example is 6K FFc X-OCN LT at ISO 800 at 12.4 (SNR=2). Some downsampled or processed modes read higher.
RED V-RAPTOR [X]	Cinema	Full-frame	“17+ stops” (marketing claim)	12.8	CineD notes the IMATEST count can include a recovered stop. Treat the last stop or two with caution in real grading.
RED KOMODO-X	Compact cinema	APS-C	Not stated	12.9	CineD notes IMATEST includes highlight recovery. Usable DR depends on how you treat recovered highlights.
RED KOMODO	Compact cinema	APS-C	Not stated	12.5	CineD reports 12.5 stops (SNR=2).
Canon EOS C300 Mark III	Cinema	Super 35	“16+ stops” (DGO messaging)	12.8	CineD pre-production lab test reports 12.8 stops at SNR=2.
Canon EOS C70	Compact cinema	Super 35	“16+ stops” (DGO messaging)	12.3	CineD reports about 12.3 (SNR=2) for Cinema RAW Light LT at ISO 800. Internal XF-AVC modes can read higher depending on processing.
Canon EOS C400	Cinema	Full-frame	Not stated	10.5	CineD reports 10.5 stops at SNR=2. ISO and mode matter.
Sony FX9	Broadcast / cinema	Full-frame	“15+ stops” (marketing claim)	11.5	CineD internal XAVC test reports 11.5 (SNR=2) at ISO 800.
Sony FX6	Cinema	Full-frame	“15+ stops” (marketing claim)	11.7	CineD internal XAVC-I test reports 11.7 (SNR=2) at ISO 800.
Blackmagic URSA Cine 12K LF	Cinema	Full-frame	Not stated	13.0	CineD reports 13.0 (SNR=2), 14.5 (SNR=1) at ISO 800 in BRAW 3:1.
Blackmagic URSA Mini Pro 12K	Cinema	S35-ish	Not stated	11.8	CineD reports 11.8 (SNR=2) in 12K at ISO 800. Downscaling in post can improve practical results.
Blackmagic Pocket Cinema Camera 6K Pro	Compact cinema	Super 35	“13 stops” (marketing claim)	11.8	CineD reports BMCC6K at 11.6 (SNR=2) and notes BMPCC6K/6K Pro read slightly higher in their tests.
Sony a7S III	Hybrid	Full-frame	Not stated	12.4	CineD reports 12.4 (SNR=2) at ISO 640. Internal noise reduction behavior can affect results.
Sony a7 IV	Hybrid	Full-frame	Not stated	12.9	CineD reports 12.9 (SNR=2) in S-Log3 at ISO 800.
Canon EOS R5 Mark II	Hybrid	Full-frame	Not stated	13.3	CineD reports 13.3 (SNR=2) in a compressed 10-bit mode at ISO 800. Processing can influence measured DR.
Canon EOS R5 C	Hybrid	Full-frame	Not stated	12.0	CineD reports multiple results by mode. 4K XF-AVC CLog3 can read about 12 (SNR=2), while some RAW modes read lower.
Panasonic LUMIX S1H	Hybrid / video-centric	Full-frame	“14+ stops” (marketing claim)	12.7	CineD reports 12.7 (SNR=2) in 6K 3:2 V-Log at ISO 640.
Panasonic LUMIX S5 II	Hybrid	Full-frame	Not stated	12.3	CineD reports 12.3 (SNR=2) at ISO 640 in V-Log (5.9K open gate).
Panasonic LUMIX GH6	Hybrid	Micro Four Thirds	Not stated	11.0	CineD reports 11.0 (SNR=2). Dynamic range boost mode can help in practice.

Tip for mobile: Swipe or scroll the table horizontally if it doesn’t fit on screen.

Sensor size, pixel size, and why it can matter

People say bigger sensors have more dynamic range. Sometimes that’s true, but the real drivers are full well capacity and noise.

Full well capacity is how much signal a pixel can hold before saturation. If a pixel can hold more signal, highlights last longer before clipping. If read noise stays low, that increases dynamic range.

Sensor size can help because larger sensors often use larger photodiodes or larger pixels at the same resolution. But it’s not guaranteed. A smaller sensor with low read noise and smart processing can perform extremely well.

How exposure choices decide what detail you keep

Dynamic range matters most when you decide where to place highlights and shadows. Your exposure choice decides which tones keep texture and which tones become clipped or noisy.

Protect highlights when highlight detail matters

If the sky, window view, or practical bulbs matter, expose so those highlights stay below clipping on waveform or zebras. You accept darker shadows, then decide how much shadow lift (and noise) you can tolerate in post.

Raise exposure when shadow detail matters

If you must hold shadow detail, expose brighter as long as highlights don’t clip. Brighter exposure puts more signal above the noise floor, so shadows usually look cleaner once you grade.

Use monitoring tools that show exposure levels

LCD images can mislead you. External monitoring can help a lot here, especially if you rely on waveform, zebras, false color, and consistent viewing conditions. If you’re building a kit, start with what an external on-camera monitor is.

• Waveform: shows highlight headroom and where shadows sit.
• Zebras: warns when areas hit your chosen threshold.
• False color: helps place mid-gray and skin tones consistently.
• Histogram: quick overview of brightness distribution.

RAW, Log, and the low ISO trap

Rec.709-style looks can clip sooner than Log

A Rec.709-style contrast curve can clip highlights sooner than Log, even if the sensor captured more headroom. A Log profile usually stores highlight detail more gently and keeps midtones less contrasty, so you can add contrast later without crushing shadows or clipping highlights as fast. If you want the full workflow context, read Look vs Log vs LUT vs RAW.

Log does not create extra sensor dynamic range. It stores what the sensor captured in a way that’s easier to grade.

RAW keeps more recovery options

RAW stores data closer to the sensor output. That often lets you change exposure and white balance in post with fewer artifacts, and it can preserve more tonal steps when you push shadows or pull highlights hard. That same guide covers the tradeoffs between formats and workflow choices: Look vs Log vs LUT vs RAW.

Extended low ISO can reduce highlight headroom

Some cameras offer “Lo” or “ISO 50” as an extended setting. On many models, this is a digital pull rather than a true lower sensitivity mode. The practical result can be less highlight headroom at the same mid-gray exposure. If you want a clean baseline for ISO behavior, start with how base ISO works in video camera settings.

Bit depth is different from dynamic range

Dynamic range is how wide a brightness range you can capture. Bit depth is how smoothly the file describes tones inside that range. Higher bit depth helps reduce banding and keeps gradients smoother after grading, even if the sensor DR is unchanged. If you want a simple refresher tied to grading, FilmDaft’s color grading guide is a good next step.

Dynamic range on film

Film has dynamic range too. It often compresses highlights gradually, which can look more forgiving than harsh digital clipping. Scanning choices also matter. A scan can clip detail that still exists on the negative if settings crush shadows or cap highlights.

From capture to screen: SDR, HDR, and tone mapping

The viewer sees your final master, not your camera file. SDR and HDR displays have different limits, so your grade must fit the target. If you want the bigger pipeline view, see Post-Production in Film.

SDR has less brightness range than HDR

SDR delivery has a smaller brightness range. HDR delivery can show brighter highlights and more separation in bright tones if the display supports it.

Tone mapping compresses brightness to fit the screen

Tone mapping compresses brightness so highlights and shadows fit the screen. It can preserve highlight texture, but it can also flatten midtones or dim highlights depending on the mapping. That’s why an HDR master can look different across displays. For the standards side of this, including Rec.709 and Rec.2020, read what color space is.

How to fit a high-contrast scene into your camera’s range

If your scene is too contrasty, exposure alone cannot save everything. You must reduce the contrast in the scene:

• Add fill: bounce or a soft source can lift faces and dark wardrobe.
• Use negative fill: solids and flags cut spill and keep shape on faces.
• Use ND outdoors: ND filters help keep shutter and aperture where you want them while managing exposure.
• Change blocking: move the subject into softer light or away from harsh backlight.

For still photos, you can also use exposure bracketing and merge exposures into one HDR image. This can hold highlight and shadow detail that a single exposure cannot, but it can fail with fast motion.

How to test dynamic range yourself (video workflows)

Specs help, but your own test tells you what’s usable in your codec, workflow, and delivery. Keep it simple and repeatable.

Do a step exposure test

1. Film a controlled scene (or a Xyla-style chart if you have one).
2. Record clean one-stop exposure steps up and down.
3. Grade the clips the way you normally deliver.
4. Mark where highlights become textureless and where shadows become unusable.

Do a window test that matches real work

1. Place a subject next to a bright window.
2. Shoot one take exposed for the face.
3. Shoot one take exposed for the window.
4. Shoot one take with fill on the face.
5. Compare window detail plus face noise after a normal grade.

Check your grade like a colorist would

Dynamic range issues often appear during grading, not on set. When you lift shadows or pull highlights, watch for color shifts, macroblocking, banding, and plastic-looking noise reduction. If you want a quick reality check on what a colorist is solving in the grade, read what a colorist does.

How to test dynamic range for still photos (RAW workflows)

Still-photo dynamic range is easiest to evaluate in RAW, because JPEG tone curves, noise reduction, and sharpening can hide what the sensor actually captured.

1) Test ISO invariance (can you raise exposure in post?)

1. Set a controlled scene with deep shadows and some highlights (a window, a lamp, or a bright card).
2. Shoot RAW at base ISO, underexposing by 2 to 4 stops (without clipping highlights).
3. Shoot the same scene again at higher ISOs, keeping shutter and aperture the same but raising ISO to match brightness.
4. In your RAW editor, push the base-ISO file up by the same number of stops and compare shadow noise and color.

If the pushed base-ISO file looks similar to the higher-ISO file, the camera is closer to ISO invariant in that range. If the pushed file looks noticeably worse, the camera benefits from analog gain at higher ISO (or has a dual-gain switch you’re not engaging at base).

2) Check base ISO vs dual gain / dual conversion gain

Many modern sensors effectively have two sweet spots where read noise drops (often called dual gain or dual conversion gain). Practical test:

• Shoot the same scene at base ISO and at the camera’s known second base ISO (the exact value depends on the model and mode).
• Compare shadow cleanliness and color stability when you lift shadows by the same amount.

This helps you learn which ISO gives you the cleanest shadows for the way you shoot.

3) Compare RAW converters (they can change the result)

Different RAW processors can produce different shadow noise, color, and highlight roll-off because they use different demosaicing, noise reduction, and highlight recovery behavior.

• Process the same RAW in two apps you actually use (your default editor vs an alternative).
• Disable extra noise reduction if possible, match exposure, and compare shadow texture and color blotching.

Still-photo testing tip

When comparing files, match output size. Downsampling a high-megapixel image can make noise look dramatically better, which is real in practice, but it can mislead you if you judge sensor behavior only at 100% view. If you want a fast refresher on what megapixels actually mean, read What is a Megapixel?.

Summing up

Dynamic range is the usable distance between clipped highlights and noisy shadows in one exposure. We describe it in stops, but published numbers differ because usable depends on thresholds, processing, codecs, and test methods. In real work, you get better results by choosing what detail matters, exposing for that priority, watching waveform, zebras, and false color, recording in a format that survives grading, and reducing scene contrast with fill, negative fill, ND, and blocking. For still photos, test in RAW, check ISO invariance and dual-gain behavior, and remember your RAW converter can change what usable really means.

Related guides on FilmDaft

• Crushed Shadows and Blown Highlights Explained
• Look vs Log vs LUT vs RAW
• What Is Video Bitrate?
• What Is an ND Filter?
• What Is Color Space? Rec.709, DCI-P3, and Rec.2020