The Intel 80386, otherwise known as the Intel386, i386 or just 386, is a microprocessor which has been used as the central processing unit (CPU) of many personal computers and workstations since 1986.
As the original implementation of the 32-bit form of the 8086-architecture, the i386 instruction set, programming model, and binary encodings is still the common denominator for all 32-bit x86 processors. As such, it has remained virtually unchanged for over 20 years, enabling modern processors to run most programs written for earlier chips, all the way back to the original 16-bit 8086 of 1978.
Successively newer implementations of this same architecture have become several hundred times faster than the original i386 chip during these years (or thousands of times faster than the 8086). A 33 MHz i386 was reportedly measured to operate at about 11.4 MIPS.
The i386 was launched in October 1985, but full-function chips were first delivered in 1986. Mainboards for 386-based computer systems were at first expensive to produce but were rationalized upon the 386's mainstream adoption. The first personal computer to make use of the 386 was designed and manufactured by Compaq
In May 2006 Intel announced that production of the 386 would cease at the end of September 2007. Although it had long been obsolete as a personal computer CPU, Intel, and others, had continued to manufacture the chip for embedded systems, including aerospace technology.
The processor was a significant evolution in a long line of processors that stretched back to the Intel 8008. The predecessor of the 80386 was the Intel 80286, a 16-bit processor with a segment-based memory management and protection system. The 80386 added a 32-bit architecture and a paging translation unit, which made it much easier to implement operating systems which used virtual memory. It also has hardware debugging support.
The 80386 featured three operating modes: real mode, protected mode and virtual mode. The protected mode which debuted in the 286 was extended to allow the 386 to address up to 4 GB of memory. The all new virtual 8086 mode (or VM86) made it possible to run one or more real mode programs in a protected environment.
Though Intel would shortly introduce the 80486 and eventually the Intel Pentium line of processors, the support in the 386 for the 32-bit flat memory model would arguably be the most important feature change for the x86 processor family until AMD released x86-64 in 2003.
Chief architect in the development of the 80386 was John H. Crawford. He was responsible for the 32-bit extension of the 80286 architecture and instruction set, and then lead the microprogram development for the 80386 chip.
In 1988, Intel introduced the i386SX, a version of the 386, designed to enable low-cost systems software-compatible with the 386.
Cost was reduced by narrowing the external data bus to 16 bits from the internal 32, which in some designs can halve the number of RAM chips, and eliminating economically unusable pins such as those address lines required to support over 16 MB of RAM. (too expensive for the low-cost systems this processor targeted).
Performance suffered both due to the narrower external databus, and the typical lack of cache memory in i386sx systems.
The original i386 was subsequently renamed i386DX to avoid confusion, though this would rather cause confusion later when the DX in the name i486DX instead indicated floating-point capability. The i387SX was a compatible i387 part (i.e. with a 16-bit databus) available as an optional math-coprocessor. The 386SX was packaged in a surface-mount QFP, and rarely offered in a socket to allow for an upgrade.
A similar cost reduction was done for the 8088 CPU used in the original PC.
The i386SL was introduced as a power efficient version for laptop computers. The processor offered several power management options (e.g. SMM), as well as different "sleep" modes to conserve battery power. It also contained support for an external cache of 16 to 64 KB. The extra functions and circuit implementation techniques caused this variant to have over 3 times as many transistors as the i386DX. The i386SL was first available at 20 MHz clock speed , with the 25 MHz model later added.
Andy Grove, Intel's CEO at the time, made the decision not to encourage other manufacturers to produce the processor as second sources, a decision that was ultimately crucial to Intel's success in the market.. From a business perspective, the i386 was significant because it was the first significant microprocessor to be single-sourced – it was available only from Intel Corp (at least initially, IBM later became a second source). Prior to this, the difficulty of making chips and the uncertainty of reliable supply required that any mass-market semiconductor be multi-sourced, that is, made by two or more manufacturers, the second and subsequent ones manufacturing under license from the designer. Single-sourcing the i386 allowed Intel greater control over its development and substantially greater profits in later years. However, AMD introduced its compatible Am386 processor in March 1991 after overcoming legal obstacles, thus ending Intel's monopoly on 32-bit 386-compatible processors.
Intel originally intended for the i386 to debut at 16MHz, but due to poor yields was instead introduced at 12MHz.
Early in production, Intel found a bug that could cause a system to unexpectedly halt when running 32-bit software. Not all of the processors already manufactured were affected, so Intel tested its inventory. Processors that were found to be bug-free were marked with a double-sigma (ΣΣ), whilst affected processors were marked "16 BIT S/W ONLY" and sold as otherwise-good parts, since at the time 32 bit software was not relevant for most users. Such chips are now extremely rare.
The i387 Math-Coprocessor was not ready in time for the introduction of the i386, and so many of the early 386 motherboards instead provided sockets to make use of an 80287. In this configuration the FPU would operate asynchronously to the CPU, usually with a clock rate of 10MHz. The original Compaq Deskpro 386 is an example of such design. However, this was an annoyance to those who depended on FPUs as the performance of the 287 was nowhere near that of the 387.
A wide range of 3rd party upgrades were also available for both SX and DX systems. The most popular ones were based on the Cyrix 486DLC/SLC core, which typically offered a substantial speed improvement via more efficient instruction pipeline and 1kb (or sometimes 8kb in the TI variant) internal L1 SRAM cache. Some of these upgrade chips (such as the 486DRx2/SRx2) were marketed by Cyrix themselves, but were more commonly found in kits offered by upgrade specialists such as Kingston, Evergreen and Improve-It Technologies. Some of the fastest CPU upgrade modules featured IBM SLC/DLC family (notable for its 16kb L1 cache), or even the Intel 486 itself. Many 386 upgrade kits were advertised as being simple drop-in replacements, but oftentimes required complicated software to control the cache and/or clock doubling. Overall it was very difficult to configure upgrades to produce the results advertised on the packaging and oftentimes less than 100% stable/compatible.