
Intel 80387
Floating-point coprocessor for the Intel 80386 CPU
Released in 1987 by Intel, this chip added hardware math to the 386 processor. It was the first from the company to fully meet the IEEE 754-1985 standard for floating-point arithmetic.
- First Intel coprocessor fully compliant with IEEE 754-1985
- Released in 1987 for the Intel 80386 processor
- Operated at speeds up to 33 MHz
- Included a low-power mobile version for laptops
What it is
The 80387 is a math coprocessor designed to work alongside the Intel 80386 microprocessor. It handles complex numerical tasks in hardware rather than relying on slow software routines. This chip marked a significant step forward by adhering strictly to the IEEE 754-1985 standard for binary floating-point arithmetic. It offered faster trigonometric functions and better overall speed compared to earlier models like the 8087 or 80287.
How it came about
Intel launched this unit in 1987, two years after the main 386 processor arrived. Before its release, the 80287 was often used with the 386, but it lacked full compatibility and performance. The new chip addressed these gaps. It became the preferred partner for the standard 386 CPU, replacing older options that were slower and less capable in their instruction sets.
Inside the chip
This component uses an eight-level stack structure for its registers. It supports single, double, and 80-bit extended precision formats. The design allows for direct access to stack positions and includes instructions for exchanging register contents. Trigonometric calculations are more accurate here than in previous generations, which limited argument ranges. The chip connects via a 32-bit bus, matching the main processor it serves.
How it was made
Intel manufactured the device using processes that allowed for speeds up to 33 MHz. A military-grade version met strict environmental standards, including temperature cycling and hermetic sealing. This rugged variant operated at 16 MHz and came in specific packages like the 68-lead pin grid array. Standard commercial versions were also produced for general computing use.
Where it went
It paired exclusively with the standard 386 chip that featured a 32-bit data bus. Systems without this coprocessor had to emulate math operations through software, which was much slower. The hardware addition accelerated scientific and engineering applications. It was not compatible with the cost-reduced 386SX, which required a different companion chip due to its narrower 16-bit bus.
Versions and family
Several versions existed for different needs. The 387SX worked with the 16-bit 386SX processor. A low-power model called the 387SL targeted laptops, featuring idle modes to save battery life. Military variants were tested for extreme temperatures. Speeds ranged from 16 MHz to 33 MHz, with the fastest achieving 3.4 megawhetstones per second in performance tests.
Why it matters
This chip set a new benchmark for compliance with industry standards for numerical precision. It demonstrated the value of dedicated hardware for floating-point tasks in personal computers. By integrating better algorithms and higher speeds, it influenced how subsequent processors handled math. It paved the way for future integration of these units directly into the main CPU die.




Questions people ask about Intel 80387
What does the 80387 do?
It performs floating-point arithmetic in hardware for the Intel 80386 CPU. This makes mathematical calculations much faster than using software routines alone.
Is it compatible with the 386SX?
No. The 80387 requires a 32-bit data bus. The 386SX has a 16-bit bus and needs the separate 80387SX coprocessor instead.
When was it released?
Intel introduced the 80387 in 1987. It arrived two years after the launch of the main 80386 microprocessor.
Good reading and sources
The facts on this page come from the Wikipedia article "X87", written by its contributors and used under the CC BY-SA 4.0 licence, and from its Wikidata record (CC0). EveryChip rewrote the text in its own words, with AI assistance, and compares every page with the article to make sure no passage is copied. It may be reused under the same CC BY-SA 4.0 licence with a link back.
It may contain mistakes, and figures change over time. Nothing here is investment or buying advice. If something is wrong, tell us. Last refreshed 2026-11-14.
Like Intel 80387
All CPUs and microprocessors
🇺🇸 United StatesIntel 80387SX
Floating-point coprocessor for the Intel 80386SXCPU · x86 · 1987
🇺🇸 United StatesI486
32-bit microprocessor with on-chip cache and FPUCPU · x86 · 1989
More from Intel
All of them
🇺🇸 United StatesIntel i960
RISC microprocessor for embedded systems and military useCPU · RISC · 1988
🇺🇸 United StatesIntel 8284
Clock generator chip for early Intel processorsAnalog · 1986
🇺🇸 United StatesI386
First 32-bit x86 microprocessor by IntelCPU · x86 · 1985