You may have heard the word “primitive” before in context of computing, but what is a primitive? A primitive is a built-in type that can be used on its own or as part of a larger, more complex user-defined object (a class).

So what is an example of a primitive? Well, an int is a primitive, as is a double, float, char, bool. Here is a list of primitives for C++:


What’s important to note is that everything in a computer is bytes - everything. This means that each primitive is just a different interpretation of a series of bytes. As an example let’s take a look at this C++ code:

std::cout << sizeof(int) * CHAR_BIT << std::endl;

This code will output how many bits comprise the primitive int. In this case the output is 32. But how did we arrive at this number? Let’s break the code down: The two keywords to note here are sizeof and CHAR_BIT; cout and endl are just used for writing to the console. The sizeof operator is a keyword that when used on a datatype will return the byte size of that object. In the case of an integer this bit of code: sizeof(int) Will return the number 4. An integer is made up of 4 bytes, and we can see that a byte is 8 bits, so the integer must be 32 bits! :D

This is all looking good, but what if we are working on a system where a byte is not 8 bits? The best way to be sure is to use the numeric limit CHAR_BIT, which provides the number of bits in a byte.

Let’s look at the size of a float:

std::cout << sizeof(float) * CHAR_BIT << std::endl;

This code again says 32 bits. So does that mean a float is the same as an integer? In memory they do share the same size, however a float is interpreted differently so as to provide a fractional value. Floats are calculated using the IEEE 754 standard.