I don't follow either. When sin(0.01rad) = 0,0099998333... and you do need the 0,00000026 precision you just choose representation with enough bits, same as wi

I don't follow either. When sin(0.01rad) = 0,0099998333... and you do need the 0,00000026 precision you just choose representation with enough bits, same as with floating point.