What Is Bit Depth? Definition, 8-bit vs 10-bit vs 12-bit

Reading Time: 8 minutes

Published: December 19, 2025

Bit depth is often confused with other specs. Here’s the difference:

• Resolution is how many pixels you have (see megapixels and resolution explained).
• Bitrate is how much data per second the codec writes (how compressed the file is). See video bitrate explained.
• Bit depth is how many steps each stored value can describe for tone and color.

Bit depth in plain numbers

In video, bit depth is usually listed as 8-bit, 10-bit, or 12-bit. These numbers almost always mean bits per channel, not the total for the whole pixel.

• 8-bit stores 256 steps per channel (2⁸).
• 10-bit stores 1,024 steps per channel (2¹⁰).
• 12-bit stores 4,096 steps per channel (2¹²).

If you picture an RGB image, those steps multiply across channels, which is why you’ll often hear that 8-bit can represent “millions of colors.” That’s a useful mental model for step counts.

Real-world video is commonly stored as Y’CbCr (luma + chroma), often with chroma subsampling (like 4:2:0 or 4:2:2) and sometimes with legal/limited ranges depending on the signal. If you want a quick refresher on these terms, the FilmDaft cinematography glossary is a good place to start.

The practical takeaway stays the same: more bits per channel = finer tone steps and smoother gradients, especially once you start grading. Bit depth also connects directly to color spaces like Rec.709 and Rec.2020, because wider gamuts and HDR workflows make gradient problems easier to spot.

Where bit depth exists in the pipeline

Bit depth can describe more than one stage in the camera and post pipeline. The stage that matters most is the one that reaches your recorded file.

• Sensor and analog-to-digital conversion define how precisely the camera turns light into numbers.
• Internal processing defines how much precision the camera keeps during debayer, color transforms, sharpening, and noise reduction.
• Recorded bit depth defines what you actually keep in the codec or RAW file.
• Output and display bit depth defines what your monitoring chain can show (see external monitors and recorders).

This is why a camera can process internally at high precision but still record an 8-bit file in certain modes.

What bit depth looks like on screen

The easiest place to see bit depth limits is a smooth gradient.

A classic example is a bright sky that fades from white near the sun into deeper blue overhead. If the file does not have enough steps, you may see visible stripes or rings instead of a smooth fade. That artifact is called banding.

Bit depth often becomes more noticeable when you grade (see the Color Grading hub):

• Increase contrast
• Push saturation
• Lift shadows
• Shift white balance or tint

If you start with 8-bit, banding and color breaks tend to appear sooner as you push the image. With 10-bit or 12-bit, the same adjustments usually hold together better.

Why more steps equals more grading headroom

Color correction works by remapping values. It stretches some regions and compresses others. If a smooth part of the image (like a wall, a sky, or soft shadows) only has a small number of stored steps, stretching those steps makes gaps visible as banding or posterization.

Higher bit depth gives you more stored values to redistribute, so gradients usually remain smooth longer under the same grade. This is also why many professional finishing pipelines lean on dedicated colorists (see what a colorist does) and measurement tools like a vectorscope.

Quick reference table

Here’s a quick reference table for you:

What should you shoot? (fast decision guide)

Choose a recording mode based on how much you plan to do in post:

• Minimal grade, SDR delivery, fast turnaround: 8-bit can be fine if exposure and white balance are accurate. If you need a refresher on exposure basics, see the exposure triangle.
• LOG recording and a real grade: aim for 10-bit (often paired with 4:2:2) to reduce banding and protect skin tones. See Look vs LOG vs LUT vs RAW.
• Heavy exposure recovery, VFX pulls, green screen: prioritize 12-bit / RAW (or the highest-bit-depth intraframe option available). For VFX basics, a good companion read is what a VFX plate is.
• HDR mastering: treat 10-bit as a practical minimum. 12-bit/RAW can help with complex highlight roll-offs and difficult gradients. If you’re working across standards, see Rec.709 vs DCI-P3 vs Rec.2020.

Why 10-bit is a common baseline for grading

10-bit doesn’t guarantee clean results, but it reduces the odds of visible banding when you apply meaningful contrast and saturation changes. It’s a widely used baseline because it usually gives you more room before problems appear, especially with LOG workflows.

• If you plan light correction and your exposure is solid, 8-bit can work.
• If you plan strong contrast moves, LOG recording, or careful skin tone work, 10-bit is a safer baseline.
• If you plan heavy exposure recovery, VFX pulls, or RAW workflows, 12-bit and higher can be useful.

A night exterior with neon signs and deep shadows shows this clearly. When you lift shadows and control saturation, low bit depth can break dark gradients into chunky steps. Higher bit depth gives you more steps to reshape those tones.

Why banding can appear even at higher bit depth

Bit depth is a major factor, but it’s not the only factor.

• Heavy compression can introduce banding and blocking. Bitrate choices matter here (see bitrate explained).
• Aggressive noise reduction can remove texture that hides tone steps.
• Very clean gradients (skies, painted walls, studio backdrops) make steps easier to see.
• Display and monitoring limits can reveal or exaggerate banding depending on the pipeline.
• Fine noise or dither can hide banding by breaking up smooth steps (see film grain explained for how “texture” changes what you notice).

How to reduce banding (practical fixes)

If you’re seeing banding in skies, walls, or soft shadow falloffs, these fixes usually help:

• Record a less-compressed option (higher bitrate, intraframe codec, or a better codec family) if available.
• Avoid over-smoothing in-camera (strong noise reduction or excessive sharpening) that turns gradients into flat zones.
• Expose cleanly. Avoid underexposing LOG and then lifting shadows aggressively. If you need a practical “what went wrong” reference, see crushed shadows and blown highlights.
• In post, add a small amount of grain/dither before heavy gradient areas. This often hides banding better than blur (see film grain).

If the banding shows up mainly after export, it can be an encoding problem. This guide on exporting high-quality video with small file sizes is a useful checklist for codec and bitrate choices.

Bit depth with LOG, RAW, and HDR

Bit depth matters most when you use formats designed for post work.

LOG video

LOG stores tones in a flatter way to preserve highlight detail. You then add contrast back in post. That contrast rebuild is where low bit depth can break gradients, which is why LOG modes are often paired with 10-bit recording. For a deeper walkthrough, see Look vs LOG vs LUT vs RAW.

RAW video

RAW workflows often keep higher precision and allow larger exposure and white balance adjustments. Higher bit depth helps when you need strong corrections, especially in shadows and saturated areas. RAW also changes how you plan post, so it pairs well with a broader view of what post-production is.

HDR delivery

HDR makes gradients and highlight roll-offs more visible. Higher bit depth helps keep skies, practical lights, and smooth highlight fades from breaking apart in HDR masters. HDR also lives inside specific standards and gamuts, so it helps to understand color space and where Rec.709 ends and wider systems begin.

How bit depth relates to other specs

How to read a camera spec line

A spec like “10-bit 4:2:2, 400 Mbps” is describing three different things:

• 10-bit: tonal precision per channel
• 4:2:2: how much color detail is sampled compared to brightness detail
• 400 Mbps: how compressed the file is at that resolution and frame rate (see bitrate explained)

You generally want enough of all three for your job, not just one.

Bit depth vs dynamic range

Bit depth and dynamic range are related, but they are not the same thing.

• Dynamic range is how much highlight and shadow detail the sensor captures in one exposure.
• Bit depth is how finely the recorded file describes tones inside that range.

A camera can capture wide dynamic range, but if you record it in low bit depth, that range is stored with coarser steps. When you stretch contrast in grading, banding can appear in skies, walls, and soft shadows. Recording the same scene in higher bit depth gives you more steps across that same range, which tends to keep gradients smoother when you reshape the image.

If you want the full “stops” picture (including noise floor and why shadows fall apart), see dynamic range in cameras.

Bit depth vs chroma subsampling

Bit depth is not the same as chroma subsampling, but the two often appear together in camera specs.

• Chroma subsampling controls how much color detail is stored compared to brightness detail, such as 4:2:0, 4:2:2, or 4:4:4.
• Bit depth controls how many steps each stored value can use.

For green screen, fine edges like hair need clean color separation. A format like 10-bit 4:2:2 usually keys more cleanly than 8-bit 4:2:0, because color values are stored with more precision and with more samples. If you want a practical on-set checklist for cleaner keys, see how to shoot and key green screen footage.

Bit depth vs bitrate

Bitrate controls compression strength. Bit depth controls precision. You need both.

• 10-bit at a very low bitrate can still look blocky due to compression.
• 8-bit at a high bitrate can look sharp but still show banding in smooth gradients.

When you compare recording modes, evaluate bit depth and codec settings together. If your main pain point is delivery compression (YouTube, social, fast web playback), start with bitrate basics and then check your export settings.

When bit depth matters most

Bit depth is more important in footage where color is of high importance and when you want to record smooth gradients. Here are some of the most common culprits for when too few bits tend to cause issues:

• Skies and soft walls with smooth gradients.
• Heavy grades that reshape contrast and color (see color grading workflows).
• Mixed lighting with difficult skin tones.
• Green screen and VFX work (see VFX plates and green screen keying).
• HDR delivery with visible highlight transitions (see color space standards).

Sources and standards (so you can verify this)

This guide follows how bit depth is defined in digital imaging (bits per channel, steps = 2^bits) and how modern video formats signal precision in practice. If you want to go deeper, these references are useful:

• ITU-R BT.709 (HD color system context). For a plain-language overview, see color space explained.
• ITU-R BT.2020 (UHD wide-gamut color system context).
• ITU-R BT.2100 (HDR system reference).
• SMPTE ST 2084 (PQ transfer function used in HDR).
• Your camera and codec documentation (recording mode bit depth, chroma subsampling, bitrate, and HDR/LOG specifics).

Summing Up

Bit depth defines how many tone and color steps your recorded file can store per channel, and higher bit depth usually means smoother gradients and more grading headroom.

8-bit can work for light correction and fast SDR delivery when exposure is solid. 10-bit is a practical baseline for LOG and serious color work. 12-bit is common in RAW workflows and demanding pipelines where you want maximum flexibility in post.

That’s a lot to remember, but if I were to give a few key takeaways, they would be:

• Bit depth is steps per channel (8-bit=256, 10-bit=1,024, 12-bit=4,096).
• Banding shows up most in smooth gradients and after heavy grading.
• For LOG and serious color work, 10-bit is a safer baseline.
• Bit depth doesn’t replace bitrate. Compression can still ruin gradients.