A1.1.5 — The GPU

← Back to A1.1 overview

What is a GPU?

A Graphics Processing Unit (GPU) is a specialised processor with thousands of small cores designed for parallel tasks. Unlike a CPU which has a few powerful cores for sequential work, a GPU has thousands of smaller cores optimised to handle many operations simultaneously.

GPUs can be integrated (built into the CPU chip) or discrete (a separate card). They communicate with software via APIs such as DirectX and OpenGL.

Features

FeatureDescription
Parallel processingThousands of cores handle subtasks simultaneously..
High throughputProcesses large volumes of data simultaneously(good for rendering)
VRAMDedicated high-speed memory for storing textures and rendering data. Enables real-time high-resolution output without buffering.

Real-World Applications

  • Graphics rendering: Video games and 3D software (Blender, Maya) — complex scenes with lighting and shadows in real-time.
  • Machine learning: Training neural networks — matrix multiplications map directly onto GPU parallel architecture.
  • Scientific computing: Physics simulations, climate modelling, genome sequencing.
  • Video editing: Real-time rendering of high-resolution footage with effects.