Nepedia
Back to Discover
Advanced Reader
💻

Graphics card

technology Maturity 11-13 war conflict Next article →

This article covers sensitive topics: war_conflict. Parents can manage visibility in Parental Controls.

A graphics card is a specialized computer expansion card designed to generate visual data. It sends a feed of graphics output to a display device like a monitor. These components are also known by several names. You might hear them called video cards, display adapters, or graphics accelerators.

AMD RX 6900XT
AMD RX 6900XT

In many professional settings, they are called graphics add-in boards, or AIBs. This name highlights their modular nature. Users add these boards to a motherboard to increase a computer's visual power. While most people use them for gaming, they are vital for many digital tasks.

The core of the card is the graphics processing unit, or GPU. The GPU is the primary component that performs complex mathematical computations.

SAPPHIRE Radeon HD 7970
SAPPHIRE Radeon HD 7970

By handling these tasks, the GPU reduces the workload on the central processing unit (CPU). This division of labor allows the computer to run more efficiently. Modern GPUs are not limited to just drawing pictures. They can be used for general-purpose computing through platforms like CUDA or OpenCL. This allows them to assist with massive tasks like AI training, cryptocurrency mining, and molecular simulation.

There are two main ways a computer handles graphics: discrete and integrated. Integrated graphics are built directly into the motherboard, the CPU, or a system-on-chip.

These are often called "on-board video." Integrated solutions are usually low-cost, compact, and use very little energy. However, they must share system resources and RAM with the CPU, which limits their speed. In contrast, a discrete graphics card is a separate unit. It features its own dedicated random access memory (RAM), power regulators, and cooling systems. This independence allows for much higher performance in 3D animation and video editing.

The history of graphics hardware shows a steady climb in capability. Early IBM PC compatible standards included MDA, CGA, and VGA. Each new standard allowed for more colors and higher resolutions. In the late 1980s, specialized cards for the Apple Macintosh II helped accelerate 2D graphics. A major shift occurred in the mid-1990s with the 3dfx Interactive Voodoo series. This was one of the first consumer GPUs to support 3D acceleration. It offloaded 3D rendering from the CPU to the GPU. Later, in 1997, the NVIDIA RIVA 128 integrated both 2D and 3D processing onto a single chip. This innovation made powerful graphics accessible to more people.

Modern high-performance cards require significant electrical power. As processing power increases, so does the demand for electricity. For example, the GeForce Titan RTX has a thermal design power (TDP) of 280 watts.

Gpu-connector
Gpu-connector

The GeForce RTX 2080 Ti has been tested to average 300 watts of consumption. Some extreme cards, like the custom "Hall of Fame" RTX 3090, can reach peak draws of 630 watts. Because the PCI-Express connection is limited to 75 watts, most powerful cards need extra power. They use six-pin or eight-pin sockets to connect directly to the power supply. This high energy use also creates intense heat that requires advanced cooling.

Because these cards are powerful, they are often very large and heavy. A high-end card like the RTX 4090 can exceed 300mm in length.

HIS Radeon HD 5850 ports
HIS Radeon HD 5850 ports

Some cards are designed as "low profile" to fit in smaller computer cases. However, the weight of large cards can cause a problem known as GPU sag. This happens when the card hangs downward after installation. If left unsupported, this sag can cause long-term damage to the hardware. Manufacturers often include support brackets to prevent this issue.

Some advanced users perform multicard scaling to increase processing power. This involves linking multiple graphics cards together to work as a single unit.

DVI Connector Pinout
DVI Connector Pinout

Users typically use proprietary methods like AMD's CrossFireX or NVIDIA's SLI. While consumer use of scaling is becoming less common, it remains important in other fields. Supercomputers, like the Summit system, use multiple GPUs to solve complex problems. Workstations also use many cards to accelerate 3D rendering and visual effects. This massive scale of computing helps scientists and creators handle the world's most difficult digital challenges.

672 words

Where will your curiosity go next?

Keep exploring at your chosen reading level.

🖼️ Images & Media (19)
File:AMD RX 6900XT .jpg
AMD RX 6900XT .jpg
File:SAPPHIRE Radeon HD 7970.jpg
SAPPHIRE Radeon HD 7970.jpg
File:S3 Graphics chrome 530 GT card (3044640155).jpg
S3 Graphics chrome 530 GT card (3044640155).jpg
File:HIS Radeon HD 5850 ports.jpg
HIS Radeon HD 5850 ports.jpg
File:Gpu-connector.jpg
Gpu-connector.jpg
File:DE-15 female.svg
DE-15 female.svg
File:DVI Connector Pinout.svg
DVI Connector Pinout.svg
File:MiniDIN-9 Diagram.svg
MiniDIN-9 Diagram.svg
File:HDMI Connector Pinout.svg
HDMI Connector Pinout.svg
File:DisplayPort Connector.svg
DisplayPort Connector.svg
File:Composite-video-cable.jpg
Composite-video-cable.jpg
File:S-video-connection.jpg
S-video-connection.jpg

+ 7 more

More to explore

🔬 Go deeper

More advanced topics to explore

🪜 Step back

Simpler topics to build understanding

What is Nepedia?

A free, ad-free encyclopedia for children. Every article is written at five reading levels, so the same page works for a five-year-old and a fifteen-year-old — use the level switcher above to see this one change. No account needed to read.