Numeric comparisons (=, <>, <, >, <=, >=, and the fixed-format
COMP / CABxx / CASxx operations) compare algebraic values aligned at the
decimal point, matching IBM i. Neither operand's declared length nor its
number of decimal positions affects the outcome, and the two operands need not
share a data type: an INT/UNS field compared against a PACKED, ZONED,
binary or float operand - or against a literal with a fractional part - keeps
the partner's decimal positions rather than truncating them. So with I an
INT(10) holding 2 and P a PACKED(5:2) holding 2.5, I < P and
I < 2.5 are both true and I = 2.5 is false.
An UNS(20) operand compares by its full unsigned magnitude, so a value at or
above 2^63 is the largest, not the smallest.