| const TInt | KMaxPrecision |
| const TInt | KPrecisionLimit |
This constant specifies the maximum number of significant digits available with floating point computations. Rounding and string formatting methods will not use more digits than this.
| const TInt | KIEEEDoubleInjectivePrecision |
Let D be the set of real numbers exactly representable by an IEEE-754 'double' For any positive integer n let X_n be the set of real numbers with an exact decimal representation using n significant digits. Let r_n : D -> X_n be defined by r_n(x)=y such that |y-x| = inf { |z-x| : z in X_n } and (in the case where two such y exist) that the last significant digit in the decimal representation of y is even. This constant is the least n such that r_n is injective.
| const TInt | KMantissaBits |
| const TInt | KMaxExponent |
| const TInt | KExponentBias |
| const TInt | KSpecialExponent |
| const TInt | KTReal32MaxExponent |
The maximum exponent for a 32-bit floating point number.
| const TInt | KTReal32MinExponent |
The minimum exponent for a 32-bit floating point number.
| const TInt | KTReal32ExponentBias |
| const TInt | KTReal32SpecialExponent |
| const TInt | KTReal32ZeroExponent |
A zero exponent value for a 32-bit floating point number.
| const TInt | KTReal64MaxExponent |
The maximum exponent for a 64-bit floating point number.
| const TInt | KTReal64MinExponent |
The minimum exponent for a 64-bit floating point number.