May 1, 2025
Do you want to learn about computers but you have no idea how or where to start? Cool, then this is a quick guide which explains all the basics you'll need.
(This topic is incredibly complex and it would take thousands upon thousands of pages to just explain how the CPU works, this is a VERY oversimplified explanation, so keep that in mind, I won't cover topics such as multithreading, PAE, CPU cycles, etc.)
What is the computer made out of?
First of all, let's talk about the central processing unit, the CPU. It's the "brain" of a computer, basically it does all the calculations. Its main task is to execute instructions which are given to it by programs. Nowadays, the computers contain more of these CPUs on a single chip, they're called cores. So when someone has a dual core CPU in a computer, it means there are two little processors which cooperate together in a bigger unit called the main CPU, each of them can work on different tasks and this is quite effective.
However, there is another measurement which is important, the clock speed. The CPU does many tasks, cycles or repeats in one second. This is measured in megahertz or gigahertz. One gigahertz is equal to 1 billion cycles per second. So, a CPU which has 3GHz will be faster than a CPU which has 2GHz but this also depends on the CPU generations and vendors.
One more thing which will concern you is the integer range. Most modern CPUs are binary based, meaning that they use 0s and 1s for representing numbers. The CPU's integer range determines the range of values it can operate on. A 64–bit CPU can work with numbers which are in a range of 2^64. You can still find old computers which have 32–bit CPUs and those are unable to run 64–bit programs.
For your computer to be able to render pixels on your screen, there is a graphics card. The graphics processing unit is one of the components in it, similar to a CPU, it contains cores but there are thousands of them and they do simpler calculations together with repetitive tasks. In order to run videos or games smoothly at 60 frames per second, they transform 3–dimensional data to an image on your screen. If the resolution is 1920x1080 pixels and your monitor has 60Hz, that's 2073600 pixels pushed through the chip 60 times per second.
There are integrated and discrete graphics. Integrated graphics often have less performance than a graphics card because the GPU inside integrated graphics needs to share system resources with the CPU. Dedicated graphics cards have their own separate RAM and a cooling system, they are also usually more expensive and they require different drivers.
For communicating with the GPU there is an application programming interface, API for short. Some notable APIs are Opengl, DirectX and Vulkan.
This is the computer's short–term memory, called the random–access memory. When a program wants to open files or move them around, it uses RAM for storing information which needs to be used quickly. Then the CPU loads these instructions and executes what the program wants. When you turn off the computer, all information is wiped from the RAM and it gets lost.
The memory management unit, MMU, which is a part of the CPU, points at addresses which are stored in the memory, then the CPU can retrieve them, these addresses are mapped all over the RAM and this prevents different programs from overwriting each other's instructions or from abusing resources. There's a difference between physical and virtual memory. Physical memory is the actual size of RAM, as in the physical component. It's measured in KiB, MiB or GiB to the power of 2, so 1024 MiB is 1GiB.
Your computer can run out of RAM easily under some circumstances, in order to prevent the system from crashing, the data ends up being written to virtual memory on the disk. This process is called swapping.
Floppy disks have been gone for a long time now, hard disk drives and solid state drives are common in computers, even though HDDs are slowly diminishing. The technology of how they work is complex, what you need to know is that magnetism is at play here. Solar storms and strong electromagnetic fields can wipe data from the disk, even your microwave can.
Hard disks are mechanical, the disk inside of them spins and wears out over time. Solid state drives on the other hand are chips which don't have moving parts, they are faster and they also last longer. However, SSDs can degrade if they're not used for prolonger periods of time and this leads to data losses.
All of these disks are measured in bytes – MB or GB, these are not the same as MiB or GiB. 1000 MB make 1GB but not 1GiB, because it's less bytes than 1024 MiB. So, when you buy a disk which is advertised as 512GB, you weren't scammed when your operating system recognises it as 500GiB.
But how do they work when it comes to storing data?
They are the computer's long–term memory, every disk is divided into a few parts called partitions. Each partition is isolated and handles either the bootloader, an operating system or other data. The size of the disk itself is divided into sectors where it writes data, it converts it again to 1s and 0s.
Essentially, everything stored on the disk is just a stream of 0s and 1s, these numbers get converted back into the files you interact with. These files are scattered all over the disk, in order to find them quickly, your operating system point a little pointer to that data, although it's still present on the disk until it gets overwritten by new information.
What handles the files on the disk?
Operating systems use filesystems for managing data. There are a couple of them and they have different functions. For example, one of the most used filesystems on Linux is EXT4, if you put it on a USB stick, you won't be able to read the contents of it on Windows, since Windows uses NTFS and it doesn't support EXT4.
Formatting a USB stick or a disk might sound scary, but all you're doing is rewriting the first few bytes of it, you can still recover the data under specific circumstances. When you format the USB stick to FAT32, it can be recognised on Windows and on Linux too. The downside is that you won't be able to move files bigger than 4GiB to it, since FAT32 is a filesystem which doesn't permit it.
With all that said, let's power on the computer.
By pressing the power button, the computer makes a few sounds, the screen lights up and you get greeted by the operating system. What exactly happened and how? Is this magic?
Well, close enough, so let's go over these steps.
As soon as you pressed the button, the power supply sent electricity to the motherboard where the components such as the CPU, the GPU, the RAM and so on are. The fans spin a bit, the disk clicks, all of these steps are a part of POST, the Power–On Self–Test. One more component wakes up from slumber, the BIOS chip.
On this tiny chip is the first stage bootloader – BIOS or UEFI, these are not mutually exclusive but they handle the boot process a bit differently. What they have in common is they look for a bootable device – the disk. Older computers use legacy BIOS but all modern computers are shipped with UEFI. Legacy BIOS and UEFI are firmware, which is software embedded into hardware.
The look for a second stage bootloader which sits on a boot partition on the disk. In Linux it's usually GRUB or another bootloader such as Syslinux or LILO. During this complex process, many things get initialised and hardware parts reserve resources for themselves in order to operate properly. When everything goes well, the 2nd stage bootloader boots into the operating system.
It's a software layer between hardware and programs which your run. Programs run as processes and the operating system makes call to the kernel in order to allocate appropriate resources for them. The kernel is also a piece of software, it communicates with hardware and prevents the abuse of resources. It has unrestricted power, if a program behaves in a malicious way, it's powerless against the kernel.
Your operating system does pretty much everything, it handles the RAM, the input with output, it writes information to the disk thanks to a filesystem and it provides the programs with shared CPU time. It keeps an eye on the programs by isolating them from each other, that way they can't steal each other's resources or come into contact with one another unless they're managed.
On top of that, it registers and logs most things which have happened, such as processes and it registers the keys on your keyboard as input, external devices too. In order to terminate programs, the operating system sends them a software interrupt in a form of a signal. A software interrupt is a message to a process that an event has occured. For example, by pressing CTRL–C in a terminal, you can modify the behaviour of a currently running process – by terminating it.
Apart from that, the operating system provides an interface for programmers and users.