Machine method to perform newton iterations for reciprocals
AssigneeMicron Technology, Inc.
InventorJames H. Hesson
Life of this patent
Estimating the term…
Loading patent…
AssigneeMicron Technology, Inc.
InventorJames H. Hesson
Estimating the term…
The machine method of the present embodiment relates to iterative numerical techniques adapted for use in digital circuitry, such as floating point multipliers and floating point adder-subtractor units. Using the Newton method of reciprocal computation of a value, several computational steps can be merged and performed with a single floating point multiplier unit. The preferred embodiment of the invention provides an improved method for computing the reciprocal of an unknown value C using Newton's method. The improved method forms the reciprocal Newton iteration in the following two steps: 1. cx.sub.m form the product of x.sub.m and c. 2. x.sub.m {2-cx.sub.m } form the product and difference simultaneously. Each step requires only the use of a multiplier-accumulator, and is more efficient and hence computationally faster than prior methods.
A method for performing a reciprocal computation of a value C using iterative numerical techniques in a multiplier, the multiplier having an SX input and an XR input respectively, the XS and XR inputs having variable values dependent upon which iteration the multiplier is processing, said variable values including the value C, x.sub.m, and p.sub.m, wherein x.sub.m is an mth iteration value of C, and wherein the p.sub.m is equal to a product of the value C and the x.sub.m, the method comprising:
Dependencies are inferred from the claim text.