A technique for implementing the instruction set of a processor as a sequence of microcode instructions ("microinstructions"), each of which typically consists of a (large) number of bit fields and the address of the next microinstruction to execute. Each bit field controls some specific part of the processor's operation, such as a gate which allows some functional unit to drive a value onto the bus or the operation to be performed by the ALU. Several microinstructions will usually be required to fetch, decode and execute each machine code instruction ("macroinstruction"). The microcode may also be responsible for polling for hardware interrupts between each macroinstruction. Writing microcode is known as "microprogramming".
Microcode may be classified as "horizontally encoded" or "vertically encoded". Horizontal microcode is as described above where there is a fairly direct correspondence between the bit fields in a microinstruction and the control signals sent to the various parts of the CPU. Not all combinations of bits will be valid (e.g. two units driving the bus at once). Vertical microcode is closer to machine code because a bit field value may pass through some intermediate combinatory logic which generates the actual control signals. This allows a few bits of a microinstruction to determine several control signals and ensure that only valid combinations of those signals are generated (e.g. a field may be decoded to determine which unit drives the bus). The disadvantage with vertical encoding is that the encoding is usually fixed and takes extra time compared with horizontal encoding which allows any combination of signals to be generated and takes no time to decode.
The alternative to a microcoded processor is a hard-wired one where the control signals are generated directly from the bits of the machine code instruction. This is more common in modern RISC architectures because it is faster.