You keep hearing the GPU all the times, especially when you are reading a review of a gaming laptop or a desktop. Do you know what all the terms mean when it comes to reading a GPU overview? Let us explain!
A graphics processing unit (GPU) is commonly described as:
An special electronic circuit that relentlessly manipulates to alter memory which speeds up the creation of images and graphics.
Unlike general-purpose CPUs, GPUs manipulates graphics and image processing—exactly what you need to smoothly play all the latest games and do computationally intense 3D work.
The first commercially available GPU for a desktop computer was introduced on August 31, 1999, called the GeForce 256. Prior 3D accelerators relied on the CPU to perform software transform and lighting, significantly limiting their computational ability. We have come a long way since 1999, and modern consumer-grade GPUs efficiently deliver buttery-smooth 1080p or even 2440p gameplay at 60 frames per second and enable graphics professionals to quickly render realistic 3D screens and create stunning works of art.
As the technology evolved, the number of terms one should be familiar with to correctly decide which graphics cards is the most suitable for his or her particular needs increased exponentially. This concise and well-organize article is here to help you make sense of all the technical terms you are likely to come across when shopping for your next graphics card. No prior technical knowledge is required so that you that you can dive right in and expand your GPU knowledge.
To attach a GPU to a motherboard, both have to have a compatible interface. Unless you’re running an ancient machine, you shouldn’t run into any issues.
A Walk Down the Memory
Just like most add-on cards during the early 90s, the first graphics cards connected to Peripheral Component Interconnect (PCI) hardware bus. Due to its severe limitations, it was superseded in 1996 by the Accelerated Graphics Port (AGP), a high-speed point-to-point channel for attaching a video card to a computer system, primarily to assist in the acceleration of 3D computer graphics, explains Traian Teglet for Softpedia. Some of the most popular graphics cards that connected to the AGP slot on the motherboard included the Radeon HD 3850, the Radeon HD 2600 XT, and the venerable GeForce 7600 GS.
A List of Modern Interfaces
The two PCI Express standards that you will come across today are PCI Express 2.0 and PCI Express 3.0. There are also PCI Express 2.1 and PCI Express 3.1, but these bring just minor improvements regarding power efficiency and functionality.
Most modern motherboards feature several PCI Express 3.0 x16 slots, theoretically allowing for 1000% increase in computing power, even though no current GPU can take advantage of the maximum throughput of the latest PCI standard. In other words, no matter what motherboard you purchase, the chance that you run into any bottlenecks is virtually nonexistent.
GPU Engine Specs
GPU engines specs are tied to the base clock speed and the boost clock speed. Similar to your CPU, the base clock speed is a guaranteed clock speed for typical full utilization. Clock speed is the rate at which a microprocessor executes and synchronize instructions using an internal clock that regulates the execution and synchronization rates, explains Vangie Beal for Webopedia.
Generally speaking, greater clock speeds mean faster computing. When things get though, it’s time for the boost clock speed to kick a give you a boost of a few extra FPS, unfortunately often at the expense of higher and louder fan speeds.
GPU memory is not only about the amount of VRAM at your disposal. Sure, modern video games often require more than 4 GB of GPU memory, but not all memory is made equal. There are three other important factors to consider: memory speed, memory interface with, and memory bandwidth.
Memory speed is exactly what it sounds like: it indicates the maximum amount of data it can transfer each second. Many modern graphics cards have memory speeds around 8 Gbps.
Memory interface is like a pipe through which memory data can travel. The wider the pipe, the more data can travel through it. These days, we typically see memory interfaces in the form of 128-bit, 256-bit, or 384-bit.
Memory bandwidth is then the result of the maximum memory speed combined with the memory interface of the graphics card. Current-gen graphics card have around 192 GB/sec memory bandwidth.
Software Interface and Technology Support
To display graphics on the screen, GPUs rely on a range of software technologies that make it easier for developers to quickly turn their projects into reality without accessing the hardware itself. The latest versions of these application programming interfaces (APIs) for handling tasks related to multimedia, especially game programming and video include Microsoft DirectX 12, Vulkan API, and OpenGL 4.5.
All the latest graphics cards from AMD and NVIDIA support all of these technologies, allowing you to enjoy the latest games and run state-of-the-art software with no problems.
Dimensions and Power Specs
Last but not least, it’s also worth considering how much power a particular graphics card consumes and how long it is. High-end graphics cards often require 1000+ watt power supplies—not something the average PC users have in his or her case. What’s more, many cards are so long that they may not even fit inside your case. Keep these things in mind before you make a purchase and you’ll end up with a graphics card that will serve you for many years to come.