If you’re just learning about the world of computers and electronics, the terminology used to refer to different parts can be confusing. One component term you may have encountered is “CPU,” which is an acronym for central processing unit.
In short, CPUs reside in almost all devices you own, whether it’s a smartwatch, a computer, or a thermostat. They are responsible for processing and executing instructions — the brains of your devices. Here we explain how CPUs interact with other parts of your devices and what makes them so integral to the computing process.
What makes a CPU a CPU?
The CPU is the core component of what defines a computing device, but it isn’t the sole component — again, it’s just the brains. It’s a chip that sits in a special seat (socket) located on the main circuit board (motherboard or mainboard) inside the device. It’s distinctly separate from the memory, which is where it temporarily stores information. It’s also separate from the graphics card or graphics chip, which renders all video and 3D graphics displayed on your screen.
CPUs are built by placing billions of microscopic transistors onto a single computer chip. Those transistors allow it to make the calculations it needs to run programs that are stored on your system’s memory. They’re effectively minute gates that switch on or off, thereby conveying the ones or zeros that translate into everything you do with the device, be it watching videos, or writing an email.
One of the most common advancements of CPU technology is in making those transistors smaller and smaller. That’s resulted in the improvement to CPU speed over the decades, often referred to as Moore’s Law.
In the context of modern devices, a desktop or laptop has a dedicated CPU that performs many processing functions for the system. Mobile devices and some tablets instead utilize a System on Chip (SoC) which is a chip that contains its CPU alongside other components. Intel and AMD both offer CPUs with graphics chips and memory stored on them too, meaning they can do more than just standard CPU functions.
What does a CPU actually do?
At its core, a CPU takes instructions from a program or application and performs a calculation. This process breaks down into three key stages: Fetch, decode, and execute. A CPU fetches the instruction from RAM, decodes what the instruction actually is, and then executes the instruction using relevant parts of the CPU.
The executed instruction, or calculation, can involve basic arithmetic, comparing certain numbers together, or moving them around in memory. Since everything in a computing device is represented by numbers, those simple tasks equate to what a CPU does. It’s what facilitates everything from starting up Windows, to watching a YouTube video.
In modern systems, the CPU doesn’t do everything, but it still must feed to specialized hardware the numbers they need to function. It needs to tell the graphics card to show an explosion because you clicked on a fuel barrel (whoops), or tell the hard drive to transfer an Office document to the system’s RAM for quicker access.
Cores, clocks, and costs
Originally, CPUs had a single processing core. Today’s modern CPU consists of multiple cores that allow it to perform multiple instructions at once, effectively cramming several CPUs on a single chip. Almost all CPUs sold today are at least dual-core. Four cores are currently considered “mainstream,” while there are more expensive chips ranging from six to a massive 32 cores.
Some processors also employ a technology called multithreading. Imagine a single physical CPU core that can perform two lines of execution (threads) at once, thereby appearing as two “logical” cores on the operating system end. These virtual cores aren’t as powerful as physical cores, but they do share the same resources. Overall, they can help improve the CPU’s multitasking performance when running compatible software.
Clock speed is another number that’s highly advertised with CPUs — the “gigahertz” (GHz) figure quoted on product listings. It effectively denotes how many instructions a CPU can handle per second, but that’s not the whole picture regarding performance. Clock speed mostly comes into play when comparing CPUs from the same product family or generation. When all else is the same, a faster clock speed means a faster processor, but a 3GHz processor from 2010 isn’t going to be as fast as a 2GHz processor from 2018.
So, how much should you pay for a CPU? We have several guides to give you some suggestions for the best CPUs you can buy. For a general outline, however, unless you’re a hardcore gamer or someone looking to edit photos or videos, you don’t need to spend more than $200. Stick to a recent generation.
For Intel CPUs that means eighth, ninth, or 10th-generation chips. You can determine their generation by the product name. For instance, the Core i7-6820HK is an older sixth-generation chip while the Core i5-10210U is a newer tenth-generation chip.
AMD does something similar with its Ryzen CPUs: The Ryzen 5 2500X is a second-generation chip based on its new “Zen” core design while the Ryzen 9 3950X is a third-generation CPU. We made the indicating numbers bold so you can see for future reference.
How important is the CPU?
Although the CPU isn’t as important for overall system performance as it once was, it still plays a major role in running a device. Since it is solely responsible for executing commands within programs, the faster your CPU, the faster many applications run.
That said, a fast CPU isn’t everything. A processor, no matter how powerful, can’t easily render the latest 3D games, nor can it store information. That’s where other components, like graphics cards and memory, come into play.
In short, the CPU isn’t everything, but it’s highly important. In general, a faster CPU will mean that your system or device will run faster. At the very least it won’t be a bottleneck in its own right. Multiple cores and threads can help you do more things at once.
Want more help in buying your next CPU? Here is our guide to the best chips from AMD and Intel.