Intel's x86 microprocessor architecture has dominated large swaths of computing for three decades. Here's why.
(Computerworld) Thirty years ago, on June 8, 1978, Intel Corp. introduced its first 16-bit microprocessor, the 8086, with a splashy ad heralding "the dawn of a new era." Overblown? Sure, but also prophetic. While the 8086 was slow to take off, its underlying architecture -- later referred to as x86 -- would become one of technology's most impressive success stories.
"X86" refers to the set of machine language instructions that certain microprocessors from Intel and a few other companies execute. It essentially defines the vocabulary and usage rules for the chip. X86 processors -- from the 8086 through the 80186, 80286, 80386, 80486 and various Pentium models, right down to today's multicore chips and processors for mobile applications -- have over time incorporated a growing x86 instruction set, but each has offered backward compatibility with earlier members of the family.
In the three decades since the introduction of the 8086, the x86 family has systematically progressed from desktop PCs to servers to portable computers to supercomputers. Along the way, it has killed or held at bay a host of competing architectures and chip makers. Even some markets that had seemed locked up by competitors, such as Apple's use of Motorola PowerPCs in the Macintosh computer, have yielded to x86 in recent years.
How did Intel's architecture come to dominate so much of the computing world? Let's take a look.
In the beginning
Intel's first microprocessor was the 4-bit 4004, which was made for a Japanese calculator in 1971. That was quickly followed by the 8-bit 8008 and in 1975 by the 8-bit 8080 chip. The 8080 went into the Altair 8800 PC, which was sold as a mail-order kit. (Bill Gates and Paul Allen founded Microsoft Corp. to sell their version of Basic for the Altair 8800.)
Three years later, the 16-bit 8086 made its debut. IBM's selection of the 8088, an 8086 variant, to power its PC in the early '80s gave the x86 architecture tremendous momentum and helped it become an industry standard that persists today.
Patrick Gelsinger, electrical engineer, chip designer and now executive vice president at Intel, says the critical turning point for the PC industry -- the thing that really sent the industry into overdrive -- was the introduction of the 32-bit 80386 in 1985. It was not obvious at the time that the x86 needed to be upgraded from the 16-bit address space of the earlier models, he says. "People said, 'What do you mean 32 bits? That's for minicomputers and mainframes.' They derided us at the time for being extravagant."
At about the same time, Compaq Computer Corp. announced a 386-based PC, which lessened IBM's death-grip control of the personal computer market. The IBM PC at the time ran the 16-bit 80286, which was more than three times slower than the 386.
According to Intel, IBM spurned the 386 because there was not yet any 32-bit software to take advantage of it. IBM was also developing a proprietary 16-bit operating system called OS/2.
"IBM owned the architecture from top to bottom. It was their applications, their operating system and their hardware design," says Gelsinger, who was a member of the 386 design team. "When they went to the next generation, they would be the only company able to offer the top-to-bottom solution, with no guarantee of compatibility from one generation to the next."
All that changed with the advent of the 386, Gelsinger says. "We moved from a vertical industry to a horizontal industry, and that really opened up the world."
The 386 was followed by the 486 in 1989. Finding that it couldn't trademark numbers, however, Intel broke from its earlier naming convention in 1993, when it named its fifth-generation processor the Pentium rather than the 586. Numerous generations of chips have carried on the Pentium brand (e.g., Pentium Pro, Pentium II and Pentium D), and Intel has since added the low-end Celeron and the high-end Core 2 brands to its x86 offerings.
Despite the name changes -- not to mention design improvements that led to exponential increases in speed, power and efficiency -- all of these chips are based on the x86 instruction set that began with the 8086 and continues to expand today.
Ingredients in a recipe for success
Why has the x86 been so successful for so long, beating back and in some cases completely vanquishing competing microprocessor architectures? For starters, the x86 came along at just the right time. By 1978, computing had been migrating from huge, expensive mainframes to smaller, cheaper minicomputers for several years. The desktop was the logical next frontier.
Moreover, the x86 demonstrated a property that had been predicted in 1965 by Gordon Moore, who would one day become Intel's chairman and CEO. Moore said, in essence, that microprocessors would double in performance every two years at no increase in cost. His prediction, later dubbed Moore's Law, proved to be correct, and the x86 went on to dominate large swaths of computing, from the data center to the workplaces and homes of end users.
And the 8086 and its successors continued to cement the relationship between two early giants of the desktop computer industry. Bill Gates and Paul Allen had tried but failed to develop their Basic programming language for the wimpy 8008 processor in 1972. But they made it work on the more powerful 8080 that they soldered into the Altair microcomputer in 1975.
That marked the beginning of a de facto partnership between Intel and Microsoft that would create a gargantuan base of software that continues to drive the industry today. Of all the factors that have led to the success of the x86 architecture, probably none is so important as that software inventory -- and no example better demonstrates this fact than the RISC processor scare.
The RISC risk
In the late 1980s and early 1990s, a serious threat to the x86 arose in the form of reduced instruction set computing (RISC) processors such as the Sun Sparc, the IBM/Apple/Motorola PowerPC and the MIPS processors. The idea was that a processor could be made to run blindingly fast if it worked on very simple instructions, with one instruction executed each clock cycle, rather than with the elaborate, multicycle instructions used in complex instruction set computing (CISC) processors like the x86.
Pundits, the press and Intel competitors widely predicted the demise of CISC at the time. "It was a difficult time for us," Gelsinger acknowledges. Indeed, Intel rushed to develop its own RISC workstation processor, the i860. But neither the 860 nor any other RISC processor came close to dislodging the hegemony of the x86.
Here's why, according to Gelsinger, who was the lead architect for the 80486 processor: "The day before the 486 was announced [April 10, 1989], there was already billions of dollars of software waiting to run on the chip. Even though the [x86 CISC] architecture was a little bit slower, by the time you could develop software for the RISC machine, we could make the [x86] machine that much faster. We had an overwhelming economic advantage because we had so much of an installed base and so many people developing. The RISC machine could never catch up."
Ironically, the lack of software for RISC machines -- plus big performance gains on the 80486 and Pentium processors -- doomed Intel's i860 along with other RISC processors. Trying to introduce a second major microprocessor architecture was a mistake, Intel would later admit.
But RISC spurred much innovation, says David Patterson, a computer science professor at the University of California, Berkeley, and one of the key RISC innovators in the 1980s. "The [Digital Equipment Corp.] VAX architecture, for example, could not keep up with RISC, and it more or less disappeared. But Intel was able to incorporate the ideas that were becoming popular in RISC while maintaining their old architecture with its large software base. And they did that in part with their superior manufacturing."
(Computerworld) Thirty years ago, on June 8, 1978, Intel Corp. introduced its first 16-bit microprocessor, the 8086, with a splashy ad heralding "the dawn of a new era." Overblown? Sure, but also prophetic. While the 8086 was slow to take off, its underlying architecture -- later referred to as x86 -- would become one of technology's most impressive success stories.
"X86" refers to the set of machine language instructions that certain microprocessors from Intel and a few other companies execute. It essentially defines the vocabulary and usage rules for the chip. X86 processors -- from the 8086 through the 80186, 80286, 80386, 80486 and various Pentium models, right down to today's multicore chips and processors for mobile applications -- have over time incorporated a growing x86 instruction set, but each has offered backward compatibility with earlier members of the family.
In the three decades since the introduction of the 8086, the x86 family has systematically progressed from desktop PCs to servers to portable computers to supercomputers. Along the way, it has killed or held at bay a host of competing architectures and chip makers. Even some markets that had seemed locked up by competitors, such as Apple's use of Motorola PowerPCs in the Macintosh computer, have yielded to x86 in recent years.
How did Intel's architecture come to dominate so much of the computing world? Let's take a look.
In the beginning
Intel's first microprocessor was the 4-bit 4004, which was made for a Japanese calculator in 1971. That was quickly followed by the 8-bit 8008 and in 1975 by the 8-bit 8080 chip. The 8080 went into the Altair 8800 PC, which was sold as a mail-order kit. (Bill Gates and Paul Allen founded Microsoft Corp. to sell their version of Basic for the Altair 8800.)
Three years later, the 16-bit 8086 made its debut. IBM's selection of the 8088, an 8086 variant, to power its PC in the early '80s gave the x86 architecture tremendous momentum and helped it become an industry standard that persists today.
Patrick Gelsinger, electrical engineer, chip designer and now executive vice president at Intel, says the critical turning point for the PC industry -- the thing that really sent the industry into overdrive -- was the introduction of the 32-bit 80386 in 1985. It was not obvious at the time that the x86 needed to be upgraded from the 16-bit address space of the earlier models, he says. "People said, 'What do you mean 32 bits? That's for minicomputers and mainframes.' They derided us at the time for being extravagant."
At about the same time, Compaq Computer Corp. announced a 386-based PC, which lessened IBM's death-grip control of the personal computer market. The IBM PC at the time ran the 16-bit 80286, which was more than three times slower than the 386.
According to Intel, IBM spurned the 386 because there was not yet any 32-bit software to take advantage of it. IBM was also developing a proprietary 16-bit operating system called OS/2.
"IBM owned the architecture from top to bottom. It was their applications, their operating system and their hardware design," says Gelsinger, who was a member of the 386 design team. "When they went to the next generation, they would be the only company able to offer the top-to-bottom solution, with no guarantee of compatibility from one generation to the next."
All that changed with the advent of the 386, Gelsinger says. "We moved from a vertical industry to a horizontal industry, and that really opened up the world."
The 386 was followed by the 486 in 1989. Finding that it couldn't trademark numbers, however, Intel broke from its earlier naming convention in 1993, when it named its fifth-generation processor the Pentium rather than the 586. Numerous generations of chips have carried on the Pentium brand (e.g., Pentium Pro, Pentium II and Pentium D), and Intel has since added the low-end Celeron and the high-end Core 2 brands to its x86 offerings.
Despite the name changes -- not to mention design improvements that led to exponential increases in speed, power and efficiency -- all of these chips are based on the x86 instruction set that began with the 8086 and continues to expand today.
Ingredients in a recipe for success
Why has the x86 been so successful for so long, beating back and in some cases completely vanquishing competing microprocessor architectures? For starters, the x86 came along at just the right time. By 1978, computing had been migrating from huge, expensive mainframes to smaller, cheaper minicomputers for several years. The desktop was the logical next frontier.
Moreover, the x86 demonstrated a property that had been predicted in 1965 by Gordon Moore, who would one day become Intel's chairman and CEO. Moore said, in essence, that microprocessors would double in performance every two years at no increase in cost. His prediction, later dubbed Moore's Law, proved to be correct, and the x86 went on to dominate large swaths of computing, from the data center to the workplaces and homes of end users.
And the 8086 and its successors continued to cement the relationship between two early giants of the desktop computer industry. Bill Gates and Paul Allen had tried but failed to develop their Basic programming language for the wimpy 8008 processor in 1972. But they made it work on the more powerful 8080 that they soldered into the Altair microcomputer in 1975.
That marked the beginning of a de facto partnership between Intel and Microsoft that would create a gargantuan base of software that continues to drive the industry today. Of all the factors that have led to the success of the x86 architecture, probably none is so important as that software inventory -- and no example better demonstrates this fact than the RISC processor scare.
The RISC risk
In the late 1980s and early 1990s, a serious threat to the x86 arose in the form of reduced instruction set computing (RISC) processors such as the Sun Sparc, the IBM/Apple/Motorola PowerPC and the MIPS processors. The idea was that a processor could be made to run blindingly fast if it worked on very simple instructions, with one instruction executed each clock cycle, rather than with the elaborate, multicycle instructions used in complex instruction set computing (CISC) processors like the x86.
Pundits, the press and Intel competitors widely predicted the demise of CISC at the time. "It was a difficult time for us," Gelsinger acknowledges. Indeed, Intel rushed to develop its own RISC workstation processor, the i860. But neither the 860 nor any other RISC processor came close to dislodging the hegemony of the x86.
Here's why, according to Gelsinger, who was the lead architect for the 80486 processor: "The day before the 486 was announced [April 10, 1989], there was already billions of dollars of software waiting to run on the chip. Even though the [x86 CISC] architecture was a little bit slower, by the time you could develop software for the RISC machine, we could make the [x86] machine that much faster. We had an overwhelming economic advantage because we had so much of an installed base and so many people developing. The RISC machine could never catch up."
Ironically, the lack of software for RISC machines -- plus big performance gains on the 80486 and Pentium processors -- doomed Intel's i860 along with other RISC processors. Trying to introduce a second major microprocessor architecture was a mistake, Intel would later admit.
But RISC spurred much innovation, says David Patterson, a computer science professor at the University of California, Berkeley, and one of the key RISC innovators in the 1980s. "The [Digital Equipment Corp.] VAX architecture, for example, could not keep up with RISC, and it more or less disappeared. But Intel was able to incorporate the ideas that were becoming popular in RISC while maintaining their old architecture with its large software base. And they did that in part with their superior manufacturing."
