AMD 9511 (AM9511) is a Floating Point Unit designed for usage with different types of microprocessors, usually as a peripheral device. The 9511 performed arithmetic and trigonometric operations with 16-bit and 32-bit fixed point numbers, and with 32-bit floating point numbers. It could also convert fixed-point number to floating point numbers and vice versa. To transfer commands and data to/from the FPU, the main processor could use either I/O or direct memory access.
Enhanced version of the FPU, AMD 9512, was released shortly after 9511. Now fully compatible with IEEE format, the AMD 9512 supported both single and double precision floating point numbers.
The 8231 is the Intel clone (yes Intel cloned an AMD part) of the 9511.