Returns this value minus one.
The result rounds to Float precision and can remain unchanged for large finite values.
the resulting Float value.