Phil Steitz a écrit : > Committed in r731822. I noticed one more thing that I just commented > for now in the code. It might be faster and/or more accurate to use a > solver to compute the nth root of the modulus. Should we consider doing > this and exposing the associated real function in MathUtils? If yes, > what solver with what parameters?
Is Math.pow so slow ? I would be interested in some benchmarks about this. I generally trust modern processors to provide wired implementations of most mathematical functions. Luc > > Phil --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org For additional commands, e-mail: dev-h...@commons.apache.org
