Whenever the term “processor architecture” is mentioned, what comes to mind is the design of the computer processor. Perhaps the number of pins and how big it is but, that is a wrong conclusion. Computer architecture reffers to either one of these two things; the abstract model of a computer or the implementation of a given instruction set. That definitely sound like unscathed knowledge and totally different from what you already know but, read on and become aware of the nitty gritties. This article shall cover one of the aspects; the abstract model of a computer. Discover as irrational beliefs on the subject are debunked.
Instruction Set Architecture(ISA), also referred to as computer architecture is the abstract model of a computer and how it is implemented. It is safe to state that the term ‘abstract‘ here means ‘theoretical’. Theory leads to practice. Once a theoretical architecture is realized, it is said to be implemented. With various theories and their implementation, there arises differences is size, performance, cost as well as a few other things. Important to note is that ISA serves as an interface between the software and the hardware.
But where exactly does 32 bit and 64 bit architectures arise from?
To answer that question, we have to understand a specific component of a CPU; the register. You can read more about the computer CPU here. The register is a temportary storgare area built into the CPU. There are two types of registers; user accessible and user inaccessible. The user inaccessible registers consist of the instructions register (IR), the memory buffer register (MBR), the memory data register (MDR) and the memory address register (MAR). The purpose of the IR is to retrieve the instructions from the program and hold these instructions while the processor acts on them while the memory registers pass data to and from RAM. All this happens in miliseconds.
The register as explained, stores memory addresses which is how the processor accesses data from RAM. In the register, bits are used to reference bytes in memory and one bit can reference one byte. In this sense, a 32 bit system can address a maximum of 4 GBs of RAM. This is why installing RAM that is more than 4GB in a 32 bit computer is a waste of time. In plain numbers, a 32 bit processor can reference 232 bits (values) translating to 4,294,967,296 bytes or 4 gigabytes. It is important to note though that it is not possible to store all those values in actuality because the system stores temporary files. The available amount of memory is normaly equal to or less than 3.8 GBs.
The 64 bit processor on the other hand has a much higher capability in terms of referencing. It can refernce a maximum of 264 or 18,446,744,073,709,551,616 values in memory. In terms of gigabytes, the 64 bit processor can handle a maximum of ~ 18,446,744,073 GB. I know 18 billion GBs of RAM sounds absurd but is true nonetheless. I doubt we going to need that kind of RAM any time soon on personal computers.
Why the big difference though?
We have seen from the numbers that a 64 bit processor can handle 4,294,967,296 time more values than the 32 bit processor. We are dealing with exponents. Considers a small number; 22 and 23. 2 squared is equal to 4 while 2 cubed is equal to 8. See the big difference already?
Next time you buy a computer and want more than 4GB of RAM, go for a 64 bit processor – a 32 bit processor just can’t handle the pressure.
Now you know.