search

data61 / cuda-fixnum

Extended-precision modular arithmetic library that targets CUDA.
Other
40 stars 28 forks source link

Implement faster Newton-Raphson #63

Open unzvfu opened 5 years ago

unzvfu commented 5 years ago

The initial estimate can be improved with a degree two approximation; also the iteration can be modified to cube the error, rather than simply square it. See here.

unzvfu commented 4 years ago

Follow up at https://github.com/unzvfu/cuda-fixnum/issues/36.