The floating point expressions needed to represent the math library functions with decent precision and performance becomes significantly more weird and complex with turns.
Please stick to radians.
Posit arithmetic requires not only sin(x), but also sin(2πx), correctly rounded that is. I wish IEEE floats had that as well.
https://posithub.org/docs/posit_standard-2.pdf
Posit arithmetic requires not only sin(x), but also sin(2πx), correctly rounded that is. I wish IEEE floats had that as well.
https://posithub.org/docs/posit_standard-2.pdf