Summary

  • Intel’s now relatively ancient Pentium processor from 1993 had a particularly complex circuit for speeding up the performance of a specific mathematical operation.
  • The operation in question was multiplying a number by three, and it turned out that this was an incredibly efficient piece of circuitry that enabled the Pentium to perform this task far faster than competing products.
  • The reason for this complexity was the mathematical optimization techniques used by the processor for number multiplication, which ultimately meant that multiplying a value by three was not straightforward.
  • This required a specialized, and very large, circuit just for this operation alone.
  • The discovery of this feature has given Intel engineers of the era a well-deserved pat on the back for their innovative solution, found within the Floating Point Unit (FPU).

By Donald Papp

Original Article