Binary coded decimal (BCD) numbers allow calculations on large numbers without rounding errors. The 8087-family coprocessors can do fast calculations with packed BCD numbers. See Section 6.4.2.2 for details. The 8086-family processors can also do some calculations with packed BCD numbers, but the process is slower and more complicated. See Section 6.4.2 for details.
This section explains how to define BCD numbers and then how to use them in calculations.