digital-signal-processing

in the Sterling approximation of factorial(n), we use an oracle based on the sw::universal::decimal type, which is an adaptive data type, to generate a reference value. However, the direct approximation of the factorial through a native floating point type yields different approximations.

factorial(40) calculated with double and decimal oracle rounded to double
815915283247897734345611269

The currently supported radices are 2, 3, 4, and 8. Small primes 5 and 7 should be added for parity with other FFT libraries. Larger powers of two might be desirable for performance.