Bit Depth Explained: 8-bit vs 16-bit vs 24-bit Color

When you save an image, you might see “8-bit” or “16-bit” options. When you check a file’s properties, it might say “24-bit color depth.” HDR displays talk about 10-bit color. What do these numbers mean?

What “bit depth” measures

Bit depth is the number of bits used to store each color channel value per pixel. The number determines how many distinct values that channel can have:

• 8-bit per channel: 256 possible values (0 to 255)
• 10-bit per channel: 1,024 possible values
• 12-bit per channel: 4,096 values
• 16-bit per channel: 65,536 values

Since color images have 3 channels (R, G, B), the “total” bit depth is 3× the per-channel value:

• 8-bit per channel = 24-bit total = 16.8 million colors
• 10-bit per channel = 30-bit total = ~1 billion colors
• 12-bit per channel = 36-bit total = ~68 billion colors
• 16-bit per channel = 48-bit total = 280+ trillion colors

People sometimes say “8-bit image” meaning 8-bit per channel (24-bit total) or 8-bit total (256 colors total — the original GIF format). Context usually clarifies which.

What 8-bit/channel actually looks like

256 values per channel gives 16.8 million possible colors per pixel. This sounds like a lot. For most everyday images, it is.

Standard JPGs are 8-bit/channel. PNGs are usually 8-bit/channel too (16-bit PNG is rare). Most monitors display 8-bit per channel. Most screen content you’ve ever seen is 8-bit.

For a typical photograph, 8-bit is enough. The eye can’t distinguish 16.8 million colors anyway.

Where 8-bit shows its limits

8-bit causes problems in specific situations:

Smooth gradients (sunset skies, foggy landscapes, smooth backgrounds): with only 256 brightness steps per channel, you sometimes see visible “banding” — bands of slightly different color where the gradient should be smooth.

Heavy editing: when you push exposure, contrast, or color significantly during editing, you’re working with the 256 values per channel. Pushing them around creates rougher transitions; details get lost.

Deep shadows: in very dark areas, 8-bit has limited steps to express subtle variations. Pulling up exposure on shadows reveals noise and banding.

HDR content: high dynamic range scenes (very bright + very dark in the same image) need more bits to capture the range.

Where 16-bit/channel matters

16-bit gives you 65,536 values per channel, or 280+ trillion colors total. The eye can’t tell the difference looking at the final image. But during processing, the extra bits provide headroom.

Use 16-bit when:

Editing photos in RAW: cameras shoot 12-14 bit per channel. Working in 16-bit during editing preserves this information.

Creating images with smooth gradients: 16-bit avoids banding even with subtle gradient editing.

Heavy color correction or grading: pushing colors around in 16-bit doesn’t compound quantization errors like 8-bit does.

Master copies for archival: 16-bit masters give future editors maximum flexibility.

HDR workflows: HDR images need 10+ bits to capture full dynamic range.

The output (the final JPG you share) is usually 8-bit. The 16-bit is for the editing pipeline.

10-bit and HDR displays

Modern HDR (High Dynamic Range) content uses 10-bit per channel. This is part of standards like:

• HDR10: 10-bit, the most common HDR format for streaming and video
• Dolby Vision: dynamic metadata, but the base format is 10-bit or 12-bit
• HDR10+: 10-bit with dynamic metadata

For HDR to look right, every step needs to support it: source content, file format, software, display. Mixing 8-bit and 10-bit pipelines loses the HDR benefit.

Most ordinary monitors and TVs are still 8-bit. “HDR-capable” displays in 2026 are usually 8-bit panels with software that approximates 10-bit (called “FRC” or “dithering”). True 10-bit panels are getting more common but aren’t universal.

What the standard image formats support

• JPG: always 8-bit per channel. No 16-bit option.
• PNG: supports 8-bit and 16-bit per channel. Most PNGs you’ll see are 8-bit.
• WebP: 8-bit per channel.
• HEIF / HEIC: supports 10-bit, 12-bit, sometimes higher. (Why iPhone photos can support HDR.)
• TIFF: very flexible — 8, 16, 32-bit per channel options.
• RAW formats: 12-14 bit per channel (varies by camera).
• AVIF: 8, 10, or 12-bit options.

So for files you’ll share:

• JPG: 8-bit, simple
• PNG: 8-bit or 16-bit
• HEIC: can be higher bit depth, supports HDR

For master files:

• RAW: highest bit depth, native camera capture
• TIFF: full control over bit depth
• PSD (Photoshop): 8, 16, or 32-bit working

Bit depth and file size

Higher bit depth = more data = larger files:

• 8-bit RGBA image: 4 bytes per pixel
• 16-bit RGBA image: 8 bytes per pixel
• 32-bit RGBA image (floating point, for HDR): 16 bytes per pixel

For a 4000×3000 image:

• 8-bit RGBA: ~48 MB uncompressed
• 16-bit RGBA: ~96 MB uncompressed

After compression (PNG, etc.), the difference is smaller but still meaningful. 16-bit PNGs are typically 50-100% larger than 8-bit PNGs of the same image.

Color depth vs color gamut

These are related but different:

Color depth: how many distinct values per channel. More bits = more distinct steps in the color range.

Color gamut: how wide the color range is. sRGB is the standard but limited gamut. Display P3 (used by recent Apple devices) is wider. Rec.2020 is very wide. ProPhoto RGB is extremely wide.

You can have a wide gamut with 8-bit (more colors per “step” but coarser steps within the range) or narrow gamut with 16-bit (fewer total colors but finer transitions within the range).

For most “I want my photos to look right” purposes:

• sRGB at 8-bit: standard web display
• Display P3 at 8-bit: modern Apple devices, better color in supported tools
• HDR Rec.2020 at 10-bit+: HDR pipelines specifically

Practical recommendations

For everyday photos and sharing: 8-bit JPG is fine.

For photos you’ll edit heavily: shoot in RAW (camera’s native bit depth, often 14-bit) → process in 16-bit → export to 8-bit JPG for sharing.

For graphics with smooth gradients: save your working files as 16-bit (PNG or TIFF). Export to 8-bit for delivery.

For HDR work: needs every step of the pipeline to be HDR-capable. Probably not relevant unless you’re specifically doing HDR.

For ordinary work: 8-bit per channel is the norm. Don’t overthink it.

How to tell what bit depth you have

In Photoshop: Image → Mode → shows current bit depth (8, 16, 32 bit).

In macOS Finder: select image → Get Info → look for bit depth in the More Info section.

In Windows Properties: right-click image → Properties → Details tab.

Command-line (with ImageMagick):

identify -format "%[depth]\n" image.jpg

Most files you’ll encounter are 8-bit. 16-bit and higher are mostly intentionally created for specific purposes.

TL;DR

• Bit depth = bits per color channel = how many distinct values per channel
• 8-bit/channel = 24-bit color = 16.8 million colors. Standard for screens and JPG.
• 16-bit/channel = 48-bit color = 280 trillion colors. For editing pipelines, RAW workflows.
• 10-bit/channel = HDR standard.
• For finished delivery: 8-bit is the norm. Don’t overthink it.
• For editing: 16-bit gives processing headroom without compounding errors.
• JPG is always 8-bit. PNG supports up to 16-bit. HEIC supports higher.
• Higher bit depth means bigger files; only use 16-bit when there’s a reason to.