RADIX returns a FIXED BINARY(31,0) value that is the model-base
used to represent the floating-point expression x.
>>-RADIX(x)----------------------------------------------------><
- x
- REAL FLOAT expression.
RADIX(x) depends on the floating-point format used to represent x. It is:
- 2 if x is held in IEEE binary floating point format
- 10 if x is held in IEEE decimal floating point format
- 16 if x is held in z/OS hexadecimal format
RADIX(x) can be used in restricted expressions.