Fractions - fixed point

Another way of storing fractions (non-integers), is with one byte per digit. This code has a fixed decimal point, and a fixed number of digits.
The format is FIXED(t,d) where t is the total number of digits, and d is the number of decimal places. eg. FIXED(4,2) means 4 digits in all, and 2 of them after the decimal point. eg. 23.87
FIXED(4,2) can also be written as (99.99), the 9s indicating digits. When using this, care must be taken to ensure that the sizes specified are large enough, otherwise overflow, underflow, rounding and truncating errors can occur.
ie. using (99.999)
- 0.001 / 9.99 = 00.000 UNDERFLOW
- 93.5 + 15.455 = 08.955 OVERFLOW
- 24.4565 = 24.457 ROUNDING (up)
- 24.4563 = 24.456 ROUNDING (down)
- 24.4565 = 24.456 TRUNCATING

Example;

In COBOL when a variable is created it is given its picture
e.g. 99V999 is a 5 byte variable with 3 decimal places.
Microsoft Access uses a currency data type allowing for up to 15 digits before the decimal point and 4 after.
Warning: Some databases use a format of;
<digits before the decimal place>, <number of decimal places>,
so 99.999 is represented as 2,3.