We all know that a Graphics Card is used to play games and produce visual output, but I want to dive deeper into the working of a graphics card.

A Processing unit, be it a CPU (Central Processing Unit) or a GPU (Graphical Processing Unit) both perform mathematical operations on binary (0 and 1).

You mave have heard of cores? The cores are basically ALUs (Arithmetic Logic Unit) and perform arithmetic operations.

So, a CPU has small number of cores (as compared to a GPU) usually 2 or 4 cores at present. (AMD has manufactured CPUs with 64 cores, 3990X) These cores are very powerful on their own and each of them are assigned different tasks to perform (called multitasking). The main emphasis in CPU is low latency, i.e. user can interact with different processes in the real time.

On the other hand, a GPU has a large number of cores. A GTX 1050Ti has 768 cores (CUDA Cores, not diving into what is CUDA for simplicity), and it is a fairly low range GPU. Now the cores in a GPU are not as powerful as the cores of a CPU, but due to such high number of cores a mathematical calcualtion can be done very quickly as a GPU uses all of it’s cores to work on the one process, unlike a CPU which focuses on multitasking. The main focus of a GPU is to increase thoroughput (giving more output per unit time).

An image is made up of millions of pixels. The CPU sends information about every pixel to the GPU which in turn performs complex mathematical and geometrical calculations on it. As the GPU creates images, it needs somewhere to hold information and completed pictures. It uses the card's RAM (called VRAM) for this purpose, storing data about each pixel, its color and its location on the screen. Part of the VRAM can also act as a frame buffer, meaning that it holds completed images until it is time to display them. Typically, video RAM operates at very high speeds and is dual ported, meaning that the system can read from it and write to it at the same time. When the time comes to display the image, the GPU outputs it and is displayed on the screen.

So, to sum it up, to process an image a lot of mathematical and geometrical calculations are required, and a GPU is a specialized hardware which is designed to do just that very quickly. GPUs are also used in scientific research, data analysis and CAD modelling because these activities require mathematical calculations as well.

You can game on CPU but it simply won’t be able to render images quickly enough and you will encounter what is called lag, stutter, low FPS etc.

Also the actual chip is called a GPU, whereas the chip assembled on a motherboard, along with VRAM, heatsink and I/O is called a graphics card.

A TU102 GPU used in an RTX 2080Ti Graphics Card. Surrounderd by VRAM chips.

An RTX 2080Ti Founder’s Edition Graphics Card.

Hope it helps.