If A is greater than B then both numbers are exchanged to make sure that A is never greater than B. Expands the interval between ALo and AHi by subtracting ACoeff from ALo and adding it to AHi Checks whether A is equal to B in a way which does not crash when one of the arguments is NaN ("not a number"). Returns the minimum of the numbers A and B. Does not crash when one of the values is NaN (Not a Number). Returns true when AValue is within or at the bounds of the interval between ABound1 and ABound2 Returns true when AValue is within or at the bounds of the interval between ABound1 and ABound2 Whether AValue is at the bounds of the interval is allowed within the specified tolerance AEpsilon where the default AEpsilon=0 is interpreted as a 1E-12 fraction of the interval length. Returns a value of NaN ("not a number") with range checking and overflow checking turned off. Converts the double-precision value AX to an integer 2's complement representation.The difference between resulting integers can be interpreted as distance in ulps. Returns a value of the special number Infinity with range checking and overflow checking turned off. Returns true when AX is within or at the bounds of the interval between ALo and AHi The double-precision parameters are converted to an integer 2's complement representation. The comparison is made with a tolerance of AMaxUlps. Calculates the value of the cumulative normal distribution for the argument AX Calculates the value of the inverse cumulative normal distribution for the argument AX