amd_fastpowf
float amd_fastpowf(float x, float y)
Computes x raised to the power y (x^y).
Maximum ULP: 1.99
Note: Does not handle special cases like NaNs or INFs
Implementation: Uses the identity x^y = e^(y*ln(x)) for general cases, with optimizations for special values like y=0.5 (square root), x=0, and y=0
amd_fastpow
double amd_fastpow(double x, double y)
Computes x raised to the power y (x^y).
Maximum ULP: 1.99
Note: Does not handle special cases like NaNs or INFs
Implementation: Uses the identity x^y = e^(y*ln(x)) for general cases, with optimizations for special values like y=0.5 (square root), x=0, and y=0