abs()

which is faster, /prog/?

const int SIZE_OF_DOUBLE = sizeof(double) * 8;
double absoluteValue = (double)(bitset<SIZE_OF_DOUBLE>((unsigned long)value).set(SIZE_OF_DOUBLE - 1, 0).to_ulong());

VS.

double absoluteValue = abs(value);

OR

is there an easier way to unset a double-precision value's sign bit?

Hey guys, do you think maybe the folks who wrote your absolute value implementation thought about this already and either (a) implemented it efficiently or (b) realised it was never going to matter?