Z80 CPU - A New Project

Inspired by Eric Lippert's series on creating a Z-Machine interpreter, I decided to create a Z80 CPU Emulator and write about it.

This will very much be a 'I'm just going to do it' and then learn from it, so expect me to make some very silly stupid mistakes as well as going back on decisions that I have made in the past.

I am in the enviable position of having no deadline and doing this just for the fun of it, so I will probably spend a lot of time re factoring; to make the code neater and easier to understand.

A CPU performs a set of instructions that have been supplied to it. Each instruction is specified by an instruction code specified in a byte; this is known as an opcode.

An opcode may require a number of operands, which supply extra information to the opcode. In the case of the Z80 CPU, an opcode can have 0, 1 or 2 operands each of which are also a byte.

There is a lot more to emulating a CPU than just performing operations; things such as timing, interrupts and pin usage needs to be configured but I'm going to start with operations and work my way up.

To begin with, I need to define some classes.

Since all of the opcodes and operations are definable as bytes, I could use the built in byte class but I don't really want to do that. So instead, I am going to define my own ZByte class which is a collection of 8 bits.

So I need a Bit which I am going to define as an enum.

public enum Bit
    {      
        Zero,
        One
    }

public class ZByte
{
      private Bit[] _bits = new Bit[8];

 public Bit this[ByteBitLocation location]
        {
            get
            {
                return _bits[(int)location];
            }
            set
            {
                _bits[(int)location] = value;
            }
        }

      /* Other methods */
}

With ByteBitLocation being defined as:

public enum ByteBitLocation : int
   {
       Zero = 0,
       One = 1,
       Two = 2,
       Three = 3,
       Four = 4,
       Five = 5,
       Six = 6,
       Seven = 7
   }

There is an argument that this is more verbose than required and that using integers would be simpler and easier, but I want to take advantage of compile time checking as much as possible, so that means not using primitive types where possible.

Next week, we will start on our first operation.CodeProject