A perfect power, and then?

[Torbjorn Granlund]

nisse at lysator.liu.se (Niels Möller) writes:

  My bsqrt uses an iteration converging to a^{-1/2}, and broot uses an
  iteration converging to a^{1/n - 1}. Both division free.
  
So binv_sqroot (from mpn/generic/perfpow.c and your bsqrt seem to
compute the same function.

But the root functions seem somewhat different.

  They should use mulmod_bnm1 rather than mullo for larger sizes, but they
  currently don't. I haven't done any serious benchmarking.
  
That's just like the perfpow.c functions, then.

-- 
Torbjörn