Computer Systems

Every computer system is made of two fundamental parts: hardware and software. Hardware refers to the physical components you can touch, like the CPU, memory, and monitor. Software refers to the programs and instructions that run on the hardware. Software is broadly divided into two categories: systems software and application software. Systems software manages the computer itself. It acts as the intermediary between the hardware and the application programs. The most important piece of systems software is the operating system (OS), which controls everything from memory management to user interfaces. Other examples include utility programs (like disk defragmenters) and device drivers. Application software is designed to perform specific tasks for the user. If you are writing an essay, browsing the internet, or playing a game, you are using application software. These programs rely on the systems software to function.

At its most fundamental level, a computer is made of billions of tiny electronic switches called transistors. Each switch can be in one of two states: ON or OFF. To represent these two states, computers use the binary number system, which has only two digits: 1 (ON) and 0 (OFF). Every piece of data in a computer—whether it's text, images, sound, or instructions—is stored as a sequence of binary digits, or 'bits'. While we use the denary (base-10) system with digits 0-9, computers use binary (base-2). Binary place values are powers of 2, increasing from right to left: ..., 64, 32, 16, 8, 4, 2, 1. An 8-bit binary number is called a 'byte'.

The world around us is analogue, meaning data is continuous and has an infinite number of possible values (e.g., the smoothness of a sound wave, the infinite shades in a photograph). Computers, however, are digital and can only handle discrete data (finite values represented by 0s and 1s). To process real-world information, computers must convert analogue data into a digital format. For an image, this process is called sampling. The image is overlaid with a grid, and each square in the grid becomes a pixel (picture element). The colour and brightness of each pixel are measured and assigned a specific binary value. The number of pixels (width x height) determines the image's resolution. The number of bits used to store the colour of each pixel is called the colour depth. A higher colour depth allows for more colours but results in a larger file size.

Logic gates are the fundamental building blocks of all digital circuits, including CPUs. They perform a logical operation on one or more binary inputs to produce a single binary output. These operations are based on Boolean logic, which deals with TRUE (represented by 1) and FALSE (represented by 0) values. The three most basic logic gates are AND, OR, and NOT.

• AND Gate: Outputs 1 only if ALL of its inputs are 1. Think of it as 'this AND that must be true'.
• OR Gate: Outputs 1 if AT LEAST ONE of its inputs is 1. Think of it as 'this OR that must be true'.
• NOT Gate: Has only one input and inverts it. A 1 becomes a 0, and a 0 becomes a 1. It is often called an inverter.

The behaviour of a logic gate or circuit is shown in a truth table, which lists all possible input combinations and the corresponding output.

Artificial Intelligence (AI) is a branch of computer science focused on creating machines capable of performing tasks that typically require human intelligence. These tasks include learning from experience, solving problems, understanding human language, and recognizing objects. The core principle of an AI system is a simple cycle: input, processing, and output. The system takes in data from its environment (input), uses complex algorithms to analyze that data and make a decision (processing), and then performs an action or provides information (output). For example, a digital assistant like Siri or Alexa takes your voice command as input, processes the language to understand your request, and provides an answer or performs an action (like playing music) as the output.

AI is no longer science fiction; it is a technology used in many industries. Two key applications are automation and image recognition. In industry, AI helps automate repetitive or dangerous tasks. For example, in a car factory, AI-controlled robotic arms perform welding and painting with a level of precision and speed humans cannot match. In logistics, robots navigate large warehouses to find, pick, and pack items for shipping, a process guided by AI. Image recognition is a powerful AI capability that allows computers to 'see' and interpret visual information. An AI is 'trained' by being shown millions of labelled images until it can identify objects, people, text, and patterns on its own. This is used in self-driving cars to identify pedestrians and traffic signs, in medicine to detect tumours in X-rays, and on your phone to unlock it with your face.

Evaluating an AI system is a complex process that goes beyond simply checking if it produces an output. A thorough evaluation must consider several key criteria:

1. Accuracy: How often does the AI make the correct decision? For a medical diagnosis AI, this could be a matter of life and death.
2. Efficiency: How quickly and with how much computational resource does the AI operate? A self-driving car's AI must be efficient enough to react to hazards in milliseconds.
3. Bias: Is the AI system fair? Bias occurs when an AI produces prejudiced results because it was trained on skewed or unrepresentative data. For example, a facial recognition system trained mostly on one demographic may perform poorly for others.
4. Ethics: Does the AI operate in a morally responsible way? This involves complex questions, such as who is at fault if a self-driving car has an accident, or whether it's right for an AI to make hiring decisions. A truly well-designed AI system must be accurate, efficient, unbiased, and ethical.