Motorola 68881

The MC68881 and MC68882 from Motorola are historical floating point coprocessors could be used with the MC68020 and MC68030 processors to the coprocessor, or as a peripheral component for any other processors.

With the MC68881/MC68882 the instruction set of the MC68020 MC68030 or extended by various floating-point operations, and 8 floating-point registers, each 80 bits in length and associated status register was. The MC68881/MC68882 set the standard IEEE 754 to largely complete and correct. As a number of formats are simple (single, 32 -bit), double (double, 64 bit ) and extended (extended, 80 -bit) supports accuracy and packed BCD representation.

In addition to the basic arithmetic and trigonometric and exponential functions, and format conversions are supported.

Between the MC68881 and MC68882, there is relatively little variation. It is practically to the same chip, but the MC68882 has a revised microcode, the internal parallel execution of some functions allowed, such as format conversions can be processed in parallel with arithmetic operations. Thus, the MC68882 achieved a somewhat higher performance than the MC68881 for certain operations.

In the MC68040 processor following the MC68030 floating point unit was integrated. The instruction set is largely compatible with the MC68881/MC68882, however, such as the trigonometric functions on the MC68040 not as a command, but by means of a library of functions to be executed.

The FPUs among others MegaSTE Atari, Atari TT, Atari Falcon 030, Amiga 3000 and the Apple Macintosh II were used