Numbers in scripts are interpreted as either int or float, depending on decimal point. This must be kept in mind when assigning very large or very small constants or constant expressions to a variable. You can find details in the manual under "Variables". This has nothing to do with the bit size of a double.