Million floating point operations per second  shortened to MFLOP as an acronym  is a computer science measurement that describes supercomputer processing speeds. The measurement normalizes the time it takes for a computer to execute an instruction. Computer instruction time can vary between 29 and 600 computer cycles. The MFLOP measure serves as a good approximation of the speed of a supercomputer as long as you use common baselines when comparing the speeds of two competing supercomputers.

Normalize the number of computer cycles assumed to execute per computer instruction. To demonstrate in a general formula, this value is represented by Y cycles / instruction.

Obtain the microprocessor cycles per second execution speed from the documentation provided with your processor, or from benchmarking programs. For the Macintosh IIx processor, the speed is 15.6672 m cycles/sec.

Obtain the speed of processor in MFLOPS by inverting the cycles per computer instruction value and multiplying by the number of cycles per second of the processor:
1 instruction / Y cycles * 15.6772m cycles / second.

Compare the result to other computers conducting simular benchmark testing. In this example, the processor speed is 15.6772 MFLOPS.