Intel’s x86 architecture is one of the most well-known architectures in computer technology. Developed back in the early 1990s, it has since then become the de facto standard for personal computers (PCs).

You probably know this from context — software made for earlier generations of PCs often won’t run on newer machines without some tweaking or even no longer working at all. That’s because older programs were designed with an 8088 or 80386 processor in mind!

Here are some examples:

Most Windows apps were built to take advantage of Microsoft’s new API layer called DirectX which only works on processors that use the x86 instruction set. This includes both 32-bit and 64-bit versions.

So if you want to run those apps, you need either a dual core 1GHz+ CPU or better yet, an AMD Athlon XP chip or higher!

Many games also make heavy use of graphics hardware which likewise requires an x86 chipset and GPU (graphics processing unit) – typically Nvidia GeForce or Radeon.

And don’t forget about media players and other types of software that depend on audio APIs such as DirectSound or OpenAL. These too require legacy x86 code.

That’s not to say there aren’t alternatives out there! But they’re usually much less popular than their competition.

Instruction set

What is an instruction?

An instruction is a piece of computer code that a processor executes to perform a specific task. Most processors have several hundred different instructions they can execute in any given time period.

The most common type of instruction used in x86 (the architecture we use for almost every computer you’ve ever owned) are called micro-ops or opcodes. There are about 15 unique opcode types, each performing a distinct function.

Some examples of micro-op codes include ADD, SUBTRACT, AND, OR, EQUAL, MULTIPLY, DIVIDE, RSHIFT, LSHIFT, LOAD, STORE, CALL, RETURN, JMP, IF, WHILE, GOTO, EXIT. You get the idea!

There are two more important things to know about micro-operations. One is that they are one way to do something, so there is only one right answer when using them. For example, if your program contains an IF statement that calls FORMAT then it would be incorrect to add a RETURN at the end instead because THAT returns nothing!

The second thing is that all computers work by executing these instructions in sequence. This means that once ONE micro-operation has been completed, the next one in line will begin running!

This sequential execution is how CPUs manage speed – the faster an operation takes, the later it gets run until technology catches up.

Memory map

What is memory mapping?

Memory mapping is when your computer has no “real” memory, but it can recognize how much RAM you have and what each individual section of RAM stores. Computers do this by looking at the chips that make up the RAM to see if there are any electrical connections (i.e., wires) coming out. If so, then they assume that area of memory is part of the main memory region because it has an internal wiring system!

With x86 processors, one common type is the von Neumann architecture. This means every byte in the computer has its own separate chip that it is stored on. These individual storage devices are referred to as bits or locations, with 1s and 0s being placed here and there to create different information and memories.

The reason why we talk about these chips as having either a 1 or a 0 is because some places require more concentrated amounts of electricity to activate them, making them represent those symbols. That’s how binary works!

This article will focus mostly on the differences between using physical DRAM (dynamic random access memory), which is already pre-programmed and working, and using FLASH memory, which must be programmed before use. It also goes into detail about the CPU registers and how they work for both types of memory.

Register file

The x86 architecture uses three different types of registers to store information. General use registers, which are typically named after what domain they belong to (e.g., %ebx), special purpose registers, such as %ecx for counting, and register files, which are a collection of regular general use registers organized into sets that are accessible quickly.

The first group is called general use because they can be used for many things; usually this means storing or retrieving small pieces of data. An example of a general use register would be saving the current value of %esp so it can be restored later. Special use registers are similar to general use registers but only usable in specific domains. For instance, the %edx register needs to contain an integer number when performing addition and subtraction calculations, so it cannot be overwritten randomly like most general use registers.

A register file is a way to make accessing all of the registers more efficient. Rather than having separate registers for each function, a register file will have one set per domain. This cuts down on overhead since you don’t need dedicated circuitry and software to access every single register.

Think about how we described register files before. These will do the same thing, except now there are no longer individual registers, just one register per domain! Because there are not separate registers, hardware will automatically know which register belongs to what domain. This reduces latency slightly and makes using registers much faster.

CPU registers

Computer processors work by changing the state of groups of special pieces of hardware called registers. These registers contain information about your computer’s memory, and they are used to perform arithmetic calculations, store values, and communicate with other components in the computer.

The most important register for understanding how computers work is the one known as the program counter (PC). The PC contains an address or location in memory where the processor is currently looking. By using the PC register, you can tell the processor which memory locations it should look at next.

Another very important register is the stack pointer (SP), which is similar to the way a cup has a handle that you put things into and a top part that holds material. The SP register works similarly; it contains an address or reference to the place where the processor can find stored data and memories. Using this stored value, the processor can add more information onto the existing information or remove old information from the memory.

There are several different types of registers in x86 architecture CPUs. Some do only logical operations such as adding numbers, while others do mathematical computations like multiplication and division. There are also general-purpose registers that do not have any assigned functions but instead can be used for anything the processor needs to know about.

Because there are so many different uses for these registers, microprocessors usually have a large number of them.

Instruction decoding

When an x86 processor executes a program, it first needs to know what instructions you want the computer to do. The most basic element of an instruction is called the opcode (short for operation code). An x86 processor has several different sets of opcodes it can use in various situations.

Some opcodes are universal across all programs, like the POPCOUNT instruction we looked at earlier. Others are more specific, such as LOAD BALANCE with its own set of parameters that differ depending on whether the instruction is used in a game or a productivity app.

The second part of every x86 instruction is one or more operands. Operand types include numbers, strings, registers, memory locations, and other CPU registers. For example, the ADD instruction adds two numbers together and stores the result into a third number.

So how does the x86 processor figure out which parts of an instruction apply to a particular situation? That’s where their internal logic comes in. Most modern processors have very sophisticated decoders that take care of this task for you.

They test each piece of the instruction against the settings in the processor, and determine if there’t he right one for it to run. If so, the engine takes control and runs it using these pre-defined values!

This process is known as “decoding” the instruction.

Steps in executing an instruction

The first step of execution for most x64 instructions is to check if the destination register is empty. This is done by taking a look at the register’s byte 5 (the low bit). If this flag is set, then the next step will be to perform another operation on either a value or a operand.

On the other hand, if this flag is not set, then we can skip directly to the second stage of execution. This is because there are no more changes that need to be made to the register, so it can be left as is!

The second stage of x64 processing begins when all of the registers have been checked and processed. At this point, the next thing to do is determine what kind of instruction this is. For example, if the instruction is a add, then the source and destination registers must contain two values that being summed together.

Execution time

As we saw in our earlier discussion of computer architecture, one of the most important parts of any computer is its execution engine or processor. This is what actually does the work of a computer program by performing math, executing software programs (or instructions), and doing all sorts of other tricks like looking up information online or storing new data in memory.

The more powerful your CPU is, the faster it can perform these functions and thus compute things quickly!

A common misconception about CPUs is that they are always working as fast as possible to make computers as quick as possible. That’s not true at all! In fact, making a really hard effort to run as fast as possible would probably slow down an average computer significantly because there are limits to how fast modern processors can operate.

These speed limitations come from two main factors: the electrical properties of transistors and the physical structure of the chip itself.

Advantages of the x86 architecture

One of the biggest strengths of the x64 architecture is its use of what’s called extended memory or eMEM. What most people refer to as RAM is made up of discrete chunks that are organized into banks, with each bank able to hold a specific amount of data. More advanced computers have more than one bank per chip, which allows for even higher levels of organization depending on your needs at any given time.

The way this works is by using what’t known as virtual addresses. A computer has a very basic concept of location, space, and something you put in that space and get back out again (like a library or room). Virtual address spaces work similarly, but instead of referring to real locations, they reference other parts of the same address space.

This is where things can get tricky. As we know, there is no such thing as infinite space, so every item in an address space has a pointer next to it indicating where the rest of that information lives.