First Generation Processors

First generation processors were used on the earliest of machines, the original IBM PC and XT, and the first clones. These machines were primitive in most ways compared to modern PCs, and the processors they used were of limited capabilities. First generation processors run at system bus speed and use the oldest processor technologies.

Intel 8088

The first chip used in PCs was Intel's 8088. This was not, at the time it was chosen, the best available CPU, in fact Intel's own 8086 was more powerful and had been released earlier. The 8088 was chosen for reasons of economics: its 8-bit data bus required less costly motherboards than the 16-bit 8086. Also, at the time that the original PC was designed, most of the interface chips available were intended for use in 8-bit designs. It's ironic, isn't it, that Intel's first production chip was in a way, the "8086SX"? It originally shipped at 4.77 MHz and a "turbo" version was later produced that ran at 8 MHz (Woo.. )

This original chip used what would be considered today to be archaic technology. The 8088 offers performance less than one-thousandth that of a modern processor, showing just how far we have come in 15 years.

Note: The NEC V20 is an 8088-compatible but actually delivers slightly better performance due to a slightly more efficient internal design.

Note: The original design of this chip was in NMOS, with newer ones in CMOS (80C88) for lower power consumption.

Look here for an explanation of the categories in the processor summary table below, including links to more detailed explanations.

Intel 8086

The Intel 8086 chip is almost identical to the 8088, except that it uses a 16-bit-wide data bus instead of the 8-bit bus of the 8088. It was actually invented before the 8088 but was not chosen for the original PC because IBM wanted to keep costs down by going with an 8-bit bus design.

The 8086 was used in some PC clones and later on in early IBM PS/2 models, but was never a popular choice for PCs. This is probably because by the time it started to be used, the 80286 was introduced, offering much improved performance over the 8086. The 8086 is architecturally identical to the 8088 aside from the wider data bus.

Note: The NEC V30 is an 8086-compatible but actually delivers slightly better performance due to a slightly more efficient internal design.

Note: Original designs were made in NMOS, newer ones in CMOS (80C86) for lower power consumption.

Look here for an explanation of the categories in the processor summary table below, including links to more detailed explanations.

Second Generation Processors

The only chip generally considered second generation is the 80286. Intel also made a chip between the 8088 and the 80286, called the 80186. Largely due to a lack of performance enhancements compared to the 286, in addition to compatibility problems with support chips, the 80186 was never really used in PCs. It has however found a strong niche role in the design of embedded microcontrollers, even to this day.

Intel 80286

The 80286 was the chip used in IBM's AT (advanced technology) system. The 286 was the first major step up in PC processors, providing significant performance increases over the 8088 and 8086--double or more performance at the same clock speed. The 286 also widened the address bus to allow access to 16 MB of memory, and introduced protected mode operation. It was originally available in 6 MHz and 8 MHz versions, but was later expanded to faster versions, all the way up to 20 MHz.

The 286 opened up the PC world to many users, but still was used mainly as the equivalent of a "turbo-charged 8088". At this time DOS was still the virtually exclusive operating system, and the protected mode the 286 offered was largely ignored.

The 80286 is of course considered obsolete today.

Look here for an explanation of the categories in the processor summary table below, including links to more detailed explanations.

Third Generation Processors

The third generation processors (the 386 family) represent another step forward from the second generation class. Intel started with this generation of chips to create "subfamilies" of related chips with different capabilities, using the "DX" and "SX" designations.

Intel 80386DX

The Intel 80386DX was the first true 32-bit processor used on the PC platform. Its internal register size was increased to 32 bits, and its data and address buses were as well, doubling data path width to the processor and increasing addressable memory to 4 GB theoretical. The 80386 family of chips offered more performance than the 80286s they replaced, largely through processor speed increases. The 386 did not offer the very large improvements over the 286 that the 286 did over the 8088.

The 386DX's increased power and the improved processor modes it offered (including full protected mode and virtual real modes) spurred the introduction of GUI-based operating systems on the PC, such as Microsoft Windows (although they are still quite slow on 386 chips). The instruction set of the 386 has set the standard for what is now called "x86" and hasn't changed very much since it was introduced. The 80386DX was the first to use pipelining to allow much improved processor performance through the use of much higher clock frequencies.

Invented by Intel, the 386 was also "cloned" by AMD and Cyrix. These are very good copies with no real compatibility problems. Intel only produced the 386DX up to 33 MHz (presumably to prevent overlapping into the 486's performance range) while AMD and Cyrix produced a 40 MHz version as well. This latter chip produced very good performance (for the time), comparable to many lower-end 486s. They are still however obsolete by today's standards, of course.

Look here for an explanation of the categories in the processor summary table below, including links to more detailed explanations.

Intel 80386SX

The 80386SX is a "lite" version of the 80386DX chip. It uses only a 16-bit data bus, the same external bus width as the 80286. It also only can address 16 MB of memory, like the 80286. The SX version of the 386 chip was in fact released well after the DX, and in some ways was intended to move the market away from the 286 since it had roughly the same interfaces but better performance. It was actually introduced several years after the DX.

Note: It is a common myth that the 80386SX can be substituted in place of an 80286 in a 286 motherboard. While the chips are compatible in terms of their external interfaces they use different packaging. A 386SX could be put into a 286 motherboard if a proper adapter were used.

The narrower data bus width of the 386SX creates a reduction in performance of about 20-25% compared to an equivalent-speed 386DX; a significant difference but not excessive. The 80386SX chip was a popular choice for the first small notebook computers, especially the 386SL variant that introduced the SMM power management features to the Intel line. The 386SX is still a 32-bit processor internally and will run 32-bit software (really slowly ). It is available in speeds from 16 to 33 MHz; a 40 MHz version was not produced for the SX, and is also available in cloned versions from AMD and Cyrix.

The 386SX chip is today considered obsolete.

Look here for an explanation of the categories in the processor summary table below, including links to more detailed explanations.