The history of CPU development for microcomputers from INTEL Corporation

The history of the CPU, also known as the microprocessor, dates back to 1971, when INTEL introduced the world's first microprocessor, the 4004. it was a 4-bit microprocessor for calculators, containing 2300 transistors. Since then, INTEL has been associated with microprocessors. The following is an example of the development of microprocessors with INTEL's 80X86 series.

1978 and 1979, INTEL has launched the 8086 and 8088 chips, they are 16-bit microprocessors, containing 29,000 transistors, clock frequency of 4.77MHz, the address bus is 20-bit, can use 1MB memory. Their internal data bus is 16-bit, the external data bus is 8-bit for the 8088 and 16-bit for the 8086. 8088 chips were first used in IBMPCs in 1981, ushering in a whole new era of microcomputers. The earliest i8086/8088 was packaged in the form of dual in-line direct insertion (DIP), and square BGA flat packages (soldered) were used from i80286, and square PGAs (pins) were used from i80386 to Pentiumpro. In 1982, INTEL introduced the 80286 chip, which contained 134,000 transistors, and the clock frequency was The 80286 has two modes of operation, real mode and protected mode.

In 1985, INTEL introduced the 80386 chip, which is the first 32-bit microprocessor in the 80X86 series, containing 275,000 transistors and clocked at 12.5MHz, and then increased to 20MHz, 25MHz, and 33MHz. Its internal and external data buses are 32-bit, and its address bus is 32-bit, and it can address 4GB of memory. It can address up to 4GB of memory. In addition to its real and protected modes, it added a mode of operation called virtual 86, which provides multitasking capability by emulating multiple 8086 processors at the same time.

In addition to the standard 80386 chip (called 80386DX), for different market and application considerations, INTEL has launched a number of other types of 80386 chips: 80386SX, 80386SL, 80386DL, and so on.

The 80386SX, launched in 1988, is a chip positioned in the market between the 80286 and the 80386DX. The difference between the 80386DX and the 80386DX lies in the fact that the external data and address buses are the same as those of the 80286, which are 16-bit and 24-bit (i.e., the addressing capability is 16MB).

The 80386SL and 80386DL, introduced in 1990, are both low-power, energy-efficient chips that are mainly used in portables and energy-efficient desktops. 80386SL differs from 80386DL in that the former is based on the 80386SX, and the latter is based on the 80386DX, but both add a new way of doing things: the System Management Method ( SMM). When entering SMM, the CPU automatically reduces the operating speed, controls the display and other components such as the hard disk to suspend work, or even stops running and enters into a "hibernation" state in order to achieve energy saving.

In 1989, INTEL introduced the 80486 chip, which broke the boundary of 1 million transistors and integrated 1.2 million transistors. Its clock frequency from 25MHz gradually increased to 33MHz, 50MHz. 80486 is the 80386 and math coprocessor 80387 and an 8KB cache integrated in a chip, and in the 80X86 series for the first time in the RISC technology, can be in a clock cycle to execute an instruction. It also utilized a burst bus approach, which greatly increased the speed of data exchange with memory. As a result of these improvements, the 80486 offered a fourfold increase in performance over the 80386DX with the 80387 math coprocessor.

Like the 80386, the 80486 came in several types. The original type described above was the 80486DX. 1990 saw the introduction of the 80486SX, a lower-priced model of the 486 type, which differed from the 80486DX in that it did not have a math coprocessor.

The 80486DX2 uses clock doubling technology, where the chip's internal operating speed is twice that of the external bus, i.e., the chip operates at two times the speed of the system clock internally but still communicates with the outside world at the original clock speed. 80486DX2's internal clock frequencies are mainly 40MHz, 50MHz, 66MHz, etc. The 80486DX4 is also the first chip with clock doubling technology. chip that employs clock doubling technology, which allows its internal units to run at two or three times the speed of the external bus. To support this increased internal operating frequency, its on-chip cache is expanded to 16KB. 80486DX4 is clocked at 100MHz and operates up to 40% faster than the 66MHz 80486DX2.

The 80486 is also available in an SL-enhanced type with a system management approach for use in portables or energy-efficient desktops.INTEL followed up in 1993 with the 80586, formally known as the PENTIUM.The PENTIUM contained 3.1 million transistors and was clocked initially at 60MHz and 66MHz, and later increased to 200MHz. The 66 MHZ PENTIUM microprocessor more than tripled the performance of the 33 MHZ 80486DX, while the 100 MHZ PENTIUM was six to eight times faster than the 33 MHZ 80486DX.

The sensation caused by the PENTIUM was not yet over, and INTEL introduced a new generation of microprocessors, the P6. The P6 contained 5.5 million transistors and was clocked at 133 MHZ, which was almost twice as fast as the 100 MHZ PENTIUM. the P6's first-level (on-chip) cache was 8 KB of instruction and 8 KB of data. Notably, the P6 includes a 256KB L2 cache chip in addition to the P6 chip in one package, and the two chips are interconnected by a high-bandwidth internal communication bus. The P6 is most notable for an innovative technology called "dynamic execution," which is another leap forward after the PENTIUM's breakthrough in superscalar architectures.

In 1997, there was a new development based on the Pentium (P54C) and the P6, a Pentium (P54C) with 57 multimedia instructions, resulting in the versatile Pentium (P55C), which was improved in the following ways compared to the P54C: (1) support for a new instruction set called the MMX multimedia extensions, with 57 new instructions for efficiently processing graphics, video, and audio data; (2) the internal Cache has been increased from 16KB to 32KB; and (3) the CPU's execution core has been optimized.

In order to make up for some of the shortcomings of the P6 chip, Intel developed two variants based on the P6: Klamath (i.e., Pentium II) and Deschutes to complement it.The Pentium II, which uses MMX and AGP technologies, has a system bus speed of 66MHz, a first-level Cache containing a 16KB instruction Cache and a 16KB data Cache, and a second-level Cache containing a 16KB instruction Cache and a 16KB data Cache. 16KB data Cache, and a secondary Cache of 512KB, with a 0.35-micron process and a CPU operating voltage of 2.8V; and Deschueses (PII350+ CPUs), a 0.25-micron version of the Pentium II, with a lower supply voltage and an external frequency of 100MHz.The PentiumII changed the The Pentium II changed from the previous PGA ceramic package and integrated the processor chip, L2 cache, and TAGPAM (to manage the L2 cache) on a single board, which was then packaged in a new SEC (SingleEdgeContact) package. Because of the new SEC package, the PentiumII had to be inserted in a 242-line SLOT1 slot, meaning that the PentiumII was not compatible with the Socket 7 architecture.

In July 1998, Intel introduced the PentiumIIXeon for servers and workstations, which utilized the new P6 microprocessor architecture, 0.25-micron fabrication, a minimum frequency of 400MHz, and 512K or 1M L2 cache internally. The PentiumII Xeon uses a 330-line SLOT2 slot to synchronize the L2 cache with the CPU main frequency, which improves system performance, but of course, the volume is slightly larger than the SLOT1 PentiumII.

The PentiumII Celeron is a PentiumII class processor released by Intel in April 1998 for the low-end market. it uses the PII core and removes the L2 cache from the PII processor, which reduces the cost, but also taxes its integer performance. inter realized this, and in the subsequent release of the 300MHz and 333MHz Celeron integrated 128K L2 cache, although less than the PentiumII's 512K, but due to the Celeron's 128K L2 cache is running at the same frequency as the CPU, so the performance is almost equal to the same main frequency PentiumII, sometimes even better than PentiumII. The price is only one half of the same frequency PentiumII, which is a great value.

On January 5, 1999, Intel introduced the Socket 370 Celeron, which still uses the Slot1 architecture of the Celeron core, but only with a new PPGA package to reduce production costs. Socket370 Celeron processor is very similar to PentiumMMX in appearance, but it has one more pin than PentiumMMX, 370 pins, while PentiumMMX has only 321 pins. So users of the old Socket7 who want to use the Socket370 Celeron, must buy a Socket370 socket motherboard, while users of the Slot1 socket motherboard, you can choose a conversion card, you can use the new Socket370 Celeron.

On February 26, 1999, Intel officially released the PentiumIII processor, which fired the first shot in the CPU war in 1999.The core of PentiumIII is roughly the same as that of PentiumII, with only the addition of 70 new SSE (StreamingSIMDExtensions, Single Instruction to SSE (StreamingSIMDExtensions) instruction set was added to the PentiumIII kernel, which enhanced the CPU's floating-point capability and improved the CPU's efficiency in executing floating-point-intensive applications. Another thing is about the serial number of PentiumIII. Since Intel has implanted a fixed serial number on the silicon of each PentiumIII, it is possible to identify the user of a computer on the Internet by the serial number of the PentiumIII. This is done to improve the security of e-commerce, but at the same time more people are concerned about their privacy being exposed online. To solve this problem, you can turn off the serial number using Intel's Serial Number Control software, or you can turn off the serial number directly in the BIOS.

The current PentiumIII runs at 450MH and 500MHz, 0.25 micron process, 32K L1 cache, 512K L2 cache also running at half the CPU frequency, 2.0V core voltage, and still uses the Slot1 slot. It is important to note that there is currently very little software that supports the SSE instruction set, which does not reflect the advantages of the SSE instruction. With the major software vendors supporting the SSE instruction set, the performance of PentiumIII will be greatly improved.

Soon after the launch of PentiumIII, Intel introduced the PentiumIII Xtreme processor, with a frequency of 500MHz and 550MHz, a core voltage of 2.0V, a Slot2 slot, and an L2-level Cache built into the chip, with 1M, 2M, or 2M+ versions. In the microprocessor market, although the Intel Corporation with its absolute size, production capacity and outstanding work design to become the industry leader, but its products are still a gap to multiply, many companies with the strength of the microprocessor is crowded into the market, to Intel issued a strong challenge, AMD's K6-2, K6-III processor, and K7 processor, they are in some respects AMD's K6-2, K6-III processors, and K7 processors, which in some ways are fully comparable to Pentium II and Pentium III, have created a complex microprocessor market.

The advent of the microprocessor was a great industrial revolution, and the evolution of the microprocessor in just a quarter of a century, from 1971 to 1999, has been incredible. The current PENTIUM is more than 300 times faster than the 8088, which was used for the first PC in 1981. It is fair to say that no other human invention has developed as rapidly and far-reachingly as the microprocessor.