Hi,
I have written a square root finding algo which uses the fact that N^2=sum of first N odd integers,so I simply add odd integers until the sum is greater than N,which gives an error of no more than 1,but does anyone know of a better and more accurate way to do this in floating point?
Thanks