Slide 80 of 97
Notes:
In cycle seven, we seethe fifth and last add.
By pipelining the code, instead of taking 15 clock cycles to do this loop, it took only seven clock cycles or less than half the time.
We saw different iterations of the loop, so if we look at step four, or in cycle four, we saw the second add, the third multiply, and the fourth load. We were working on different iterations of the loop and that's why when we saw the compiler output, we saw a several loads before the first multiply.