Where and how does software meet hardware? [closed]

Answer 1

There is actually one more layer below machine language that I have found out recently from a friend. It's called Microcode.

See this Wikipedia article for details: http://en.wikipedia.org/wiki/Microcode

Answer 2

The premise of your question is incorrect. There is no divide. Software is entered and saved electronically. It is just that software is displayed to you in an english like language so that humans can follow it easily.

So for example, when you enter "Change my desktop screen to Green," via keyboard, it is going directly on metal electronically (read how keyboard/mouse/touchscreen) work. It is just that you see it as english text on your screen. Take one step further, the text you enter is in high level language, it get converted into machine code so that CPU can perform logic on it. The machine code is in electronic form also. For example, the text will get converted into opcodes etc, but the opcodes are in electronic form and CPU can process those.

The bottom line is that everything inside your computer is in electronic form from the moment it is entered. It is either on your hard disk or memory (transistors, etc). It is just that you see it in text on your screen.

Again, there is no divide. Software saved in memory (transistors etc) or hard disk is already there electronically. Hardware (hard logic - gates) is needed to run software (what you want hardware to do with your software). You CANNOT enter anything in computer in non electronic form.

I hope this makes sense.

Answer 3

Well, if you know how the processor accesses memory then you already know the answer. Memory and other hardware are accessed in (almost) the same way, depending on wether the processor uses "memory mapped IO" or "IO mapped IO". In the first case, the processor just tries to read and write to a memory address - but there are no memory there - instead it's some other hardware device, but the processor really can't tell the difference. The latter case is very similar.

On the other hand, if you have no clue on how the processor accesses the memory, you should look up what an "address bus" and a "data bus" is to get you started.

Answer 4

I'd very much suggest that you read the book Code, it gives a detailed history of how the computer evolved from different systems.

It's very engaging and will explain to you how things evolved from morse code through to a simple adding machine and then on to some assembly. It should give you a picture of exactly how machine instructions are interpretted and used by the hardware of a CPU, memory, etc.

Answer 5

Lets assume that the most real is current and voltage (if you go deeper, you may find that even an electron is an abstraction). The first step of abstraction is to consider that a +5V voltage represent a bit set to '1' and a 0V voltage represents a bit set to '0'. With a switch, you can decide the value of your wire.

With a second wire and a second switch, you get two binary values. The most interesting to do with them is to combine their binary values with operations such AND, OR and NOT. Two concepts are involved here: transistor for the real and logic for the abstraction. With that, you can perform addition, substraction and many other operations on your two binary inputs. You can add more wires to represent values other than '0' and '1'. Going this way, you obtain an ALU.

Now it's time to include time. If you want several results from your ALU, you have to provide input data one at a time, thus needing a clock and a control unit for sequencing operations. You also need memory for storing input values and results, and input and output devices for commanding and observing. Going this way, you obtain a Von Neuman Architecture, which is the basis of almost all computer architectures.

Next abstraction level is when you decide, with appropriate electronics, that this particular bit in memory will stimulate that particular pixel of your LCD display.

Answer 6

On the road to Changtse.