AVgecko
HomeLearn AV
Toolssoon
← All chapters

2.Video fundamentals

  1. 1Pixels & resolution
  2. 2Frame rate & refresh
  3. 3Color depth & chroma subsampling
  4. 4Video bandwidth
Chapter0/4
← 2. Video fundamentals

Video bandwidth

Putting it together: calculate the data rate of a video signal and check it against a link.

Chapter 210 minBandwidth calculator

You now know the four ingredients of a video signal: resolution, frame rate, bit depth and chroma. Multiply them together and you get its bandwidth — the data rate every cable, switcher, extender and network link in the path must carry. This one number decides more AV product choices than any other.

The formula

data rate = width × height × frames per second × bits per pixel

Bits per pixel comes from bit depth and chroma. At 4:4:4, each pixel carries three full channels, so 8-bit colour means 24 bits per pixel. At 4:2:0 it's half that.

Take 1080p at 60 fps, 8-bit, 4:4:4:

1920 × 1080 × 60 × 24 = 2.99 Gbps

That's about three billion bits every second, for one ordinary laptop signal. Now try the same maths for 4K: four times the pixels, so four times the data — almost 12 Gbps before anything else is added.

The hidden extra: blanking

Digital video interfaces inherited a habit from analog CRTs: after each line and each frame there's a pause called the blanking interval. Today that time carries audio and auxiliary data, but it still occupies the link. Standard timings for 1080p and 4K add about 19% on top of the visible picture. That's why 1080p60 is usually quoted as about 3.6 Gbps rather than 3.0, and 4K60 at 8-bit 4:4:4 as about 14.3 Gbps.

Bandwidth calculator
Bits per pixel

24

Active picture

11.94 Gbps

With blanking

14.26 Gbps

Does it fit?
HDMI 1.4 / HDBaseT class · 10.2 Gbps raw✕ too much · 8.16 usable
HDMI 2.0 · 18 Gbps raw✓ fits · 14.4 usable
DisplayPort 1.4 · 32.4 Gbps raw✓ fits · 25.92 usable
HDMI 2.1 · 48 Gbps raw✓ fits · 42.67 usable

Uncompressed estimate using an HDMI-style model and 60 Hz CTA-861 blanking ratios — real timings vary slightly by format. HDMI carries 4:2:2 in a fixed 24-bit container. Compression (such as DSC) and vendor-specific extender technologies are not included.

Start at 4K, 60 fps, 8-bit, 4:4:4 — it just squeezes into HDMI 2.0. Now switch to 10-bit and watch it overflow. Then drop to 4:2:0 and see how subsampling rescues it.

Raw rate vs. usable rate

Interface specs quote a raw signalling rate, but some of those bits are spent on line coding — extra bits that keep the receiver locked and detect errors. What's left for video is the usable (payload) rate.

LinkRaw rateCodingUsable for video
HDMI 1.410.2 Gbps8b/10b8.16 Gbps
HDMI 2.018 Gbps8b/10b14.4 Gbps
DisplayPort 1.432.4 Gbps8b/10b25.92 Gbps
HDMI 2.148 Gbps16b/18b≈ 42.7 Gbps
HDMI 1.4
Raw rate10.2 Gbps
Coding8b/10b
Usable for video8.16 Gbps
HDMI 2.0
Raw rate18 Gbps
Coding8b/10b
Usable for video14.4 Gbps
DisplayPort 1.4
Raw rate32.4 Gbps
Coding8b/10b
Usable for video25.92 Gbps
HDMI 2.1
Raw rate48 Gbps
Coding16b/18b
Usable for video≈ 42.7 Gbps

8b/10b coding, for example, sends 10 bits on the wire for every 8 bits of data, so only 80% of the raw rate is available. Always compare a signal against the usable figure.

Key idea

When a signal doesn't fit, something must give. You can lower the frame rate, lower the bit depth, subsample the chroma, reduce the resolution — or compress it. Every AV product that moves video makes one of these trade-offs, and good designers know which one.

Why this shapes product choices

  • Switchers are rated by the bandwidth each input and output can handle. An 18 Gbps switcher can carry 4K60 4:4:4 at 8-bit, but not at 10-bit.
  • HDBaseT extenders are generally designed around the same 10.2 Gbps-class signals as HDMI 1.4, so 4K60 typically needs 4:2:0 or some form of compression.
  • AV-over-IP runs on 1 Gbps or 10 Gbps network links. Even 1080p60 uncompressed exceeds 1 Gbps, which is why most AVoIP systems compress — a trade-off you'll study in Chapter 9.
  • Cables are certified for bandwidth: a cable rated for 18 Gbps may fail at 48 Gbps.
In the field

Before specifying any switching or extension product, write down the most demanding signal the system must carry — resolution, frame rate, bit depth, chroma — and check it against every link. It takes two minutes and prevents the most common reason a 4K system “doesn't work” on handover day.

Key takeaways

  • 01Video bandwidth = width × height × fps × bits per pixel, plus blanking overhead.
  • 02Standard 1080p and 4K timings add about 19% blanking on top of the visible picture.
  • 03Compare against a link's usable rate, not its raw rate — line coding takes a share.
  • 044K60 8-bit 4:4:4 just fits HDMI 2.0's 14.4 Gbps; 10-bit does not.
  • 05When a signal doesn't fit, you trade frame rate, bit depth, chroma, resolution — or compress.

Check your understanding

Q1Roughly how much uncompressed data does 1080p60 8-bit 4:4:4 carry, including blanking?

Q2HDMI 2.0 is quoted as 18 Gbps. How much of that is usable for video?

Q3A 4K60 10-bit 4:4:4 signal won't fit through an HDMI 2.0 switcher. Which change makes it fit while keeping 4K60?

← Previous

Color depth & chroma subsampling

Next →

HDMI, DisplayPort & USB-C