TheIntel80387SX (387SX ori387SX) is the mathcoprocessor, also called anFPU, for theIntel 80386SXmicroprocessor. Introduced in 1987,[1] it was used to performfloating-point arithmetic operations directly in hardware. The coprocessor was designed only to work with the 386SX, rather than the standard386DX. This was because the original80387 could not communicate with the altered 16 bitdata bus of the 386SX, which was modified from the original 386DX's 32 bit data bus. The 387SX uses a 68-pinPLCC socket, just like some variants of the80286 and the less common80186 CPU, and was made in speeds ranging from 16 MHz to 33 MHz, matching the clock speed range of the Intel manufactured 386SX. Some chips like the IIT 3C87SX could get up to 40 MHz, matching the clock speeds of the fastest 386SX CPUs.
With the 386SX/387SX combination, it provides three to five times the floating-point performance compared to the 80286/80287 combination.[2]
Intel released the low power version of 387SX coprocessor.[3]
There was boxed version of 387SX for USD $506.[4]
In 1991, the 25-MHz Intel387 SX math coprocessor was available in manufacturer's suggested retail price of USD $189.[5]
In 1993, the 33-MHz Intel387 SX math coprocessor was available in manufacturer's suggested retail price of USD $109.[6]