Unsigned 16-bit integer
Signed 16-bit integer
15-bit Integer 0,0,0,0, 0, 0, 0, 0, 0,0 1,0, 0, 0, 0
A practitioner of fixed-point math would observe that there is an implied decimal point (called a radix point) to the right of the rightmost bit position and would ask, "Why must it fall there? Why can't I put the radix point somewhere else?" In other words, why must the rightmost bit represent 2°?
Figure 9.1.b shows the same 16-bit string. In this case, the programmer decides to place the radix point between the 5th and 6th bit positions, which make the rightmost bit 25. The string 0000000000010000 is now not 16, but 0.5. Another way to look at this is to say that the integer 16 has been scaled by 2-5 (multiplied by 2 5, or .03125):
Was this article helpful?