Give you a complete overview

CPU is the abbreviation of Central Processing Unit (central microprocessor). It is the most important part of the computer and consists of arithmetic units and controllers. If a computer is compared to a human being, then the CPU is the human brain. The development of CPU is very rapid. It only took less than twenty years for personal computers to develop from 8088 (XT) to the current Pentium 4 era.

In terms of production technology, the original 8088 integrated 29,000 transistors, while the integration of Pentium III exceeded 28.1 million transistors; the running speed of the CPU is measured in MIPS (million instructions per second). Unit, 8088 is 0.75MIPS, and by the time of high-energy Pentium, it has exceeded 1000MIPS. No matter what kind of CPU, its internal structure can be summarized into three parts: control unit, logic unit and storage unit. These three parts coordinate with each other to analyze, judge and calculate commands and data and control the coordination of various parts of the computer. Work.

CPU has a history of more than 20 years since its initial development. During this period, according to the word length of the information it processes, CPU can be divided into: 4-bit microprocessor, 8-bit microprocessor, 16-bit microprocessors, 32-bit microprocessors, and 64-bit microprocessors are being built. It can be said that the development of personal computers is advancing with the development of CPUs.

Intel 4004

In 1971, Intel launched the world's first microprocessor 4004, which was the first four-bit microprocessor that could be used in microcomputers. Contains 2300 transistors. Intel then launched the 8008. Due to poor computing performance, its market response was very unsatisfactory. In 1974, 8008 developed into 8080, becoming the second generation microprocessor. The 8080 is used as a replacement for electronic logic circuits in a variety of application circuits and devices that would not be possible without a microprocessor.

Because microprocessors can be used to complete many computing tasks that previously required larger equipment and are cheap, semiconductor companies began to compete to produce microprocessor chips. Zilog produced the enhanced Z80 of the 8080, Motorola produced the 6800, and Intel produced the enhanced 8085 in 1976. However, these chips basically did not change the basic characteristics of the 8080, and they all belonged to the second generation of microprocessors. They all use NMOS technology, with an integration level of about 9,000 transistors. The average instruction execution time is 1μS to 2μS. They are programmed in assembly language, BASIC, and Fortran, and use a single-user operating system.

Intel 8086

The 8086 produced by Intel in 1978 was the first 16-bit microprocessor. Soon Zilog and Motorola also announced plans to produce the Z8000 and 68000. This was the starting point for the third generation of microprocessors.

The 8086 microprocessor has a maximum clock speed of 8MHz, a 16-bit data channel, and a memory addressing capability of 1MB. At the same time, Intel also produced the matching math coprocessor i8087. The two chips use mutually compatible instruction sets, but the i8087 instruction set adds some instructions specifically for mathematical calculations such as logarithms, exponentials, and trigonometric functions. These instruction sets are collectively called x86 instruction sets. Although Intel successively produced second-generation and third-generation new CPUs that were more advanced and faster, they were still compatible with the original x86 instructions, and Intel continued to use the original x86 sequence in the naming of subsequent CPUs until later. Due to trademark registration issues, we gave up continuing to use Arabic numerals for naming.

In 1979, Intel developed the 8088. Both 8086 and 8088 use 16-bit data transmission inside the chip, so they are both called 16-bit microprocessors, but 8086 can transmit or receive 16 bits per cycle. bit data, whereas the 8088 only uses 8 bits per cycle.

Because most of the initial devices and chips were 8-bit, and the 8088's external 8-bit data transmission and reception were compatible with these devices. The 8088 is available in a 40-pin DIP package and operates at 6.66MHz, 7.16MHz or 8MHz. The microprocessor integrates approximately 29,000 transistors.

Soon after the 8086 and 8088 came out, Intel began to improve them. They integrated more functions on the chip, and thus the 80186 and 80188 were born. Both microprocessors work with 16 bits internally. The 80186 uses 16 bits for external input and output, while the 80188 and 8088 use 8 bits.

In 1981, the American IBM Company used the 8088 chip in the PC it developed, thus creating a new era of microcomputers. It was from the 8088 that the concept of personal computers (PC) began to develop around the world. Since the 8088 was applied to IBM PCs, personal computers have truly entered people's work and life, and it also marked the beginning of a new era.

Intel 80286

In 1982, Intel developed the 80286 microprocessor based on the 8086. The maximum main frequency of the microprocessor is 20MHz, and the internal and external data Transmissions are all 16-bit, using 24-bit internal memory addressing, and the memory addressing capability is 16MB. 80286 can work in two modes, one is called real mode and the other is called protection mode.

In real mode, the total amount of memory that the microprocessor can access is limited to 1 megabyte; in protected mode, the 80286 can directly access 16 megabytes of memory. In addition, the 80286 works in a protected mode, which can protect the operating system and prevent it from shutting down the system when encountering abnormal applications like real mode or unprotected microprocessors such as the 8086.

IBM used the 80286 microprocessor in its advanced technology microcomputer, the AT machine, which caused a great sensation. The 80286 has significant improvements over its predecessors in the following four aspects: support for larger memory; ability to simulate memory space; ability to run multiple tasks simultaneously; and improved processing speed. The speed of the earliest PC was 4MHz, and the first AT machine based on 80286 ran at 6MHz to 8MHz. Some manufacturers also increased the speed on their own, so that 80286 reached 20MHz, which meant a significant improvement in performance.

The package of 80286 is a square package called PGA. PGA is a cheap package derived from PLCC. It has an internal and external solid pin. In this package, the 80286 integrates approximately 130,000 transistors.

The bus of the IBM PC/AT microcomputer maintains the three-layer bus structure of XT, and adds high and low byte bus driver conversion logic and high byte bus. Like the XT machine, the CPU is soldered to the motherboard.

At that time, the original machine only referred to the IBM PC, and the compatible machine referred to other machines besides the IBM PC. At that time, in addition to Intel, companies producing CPUs also included AMD and Siemens, and people did not care what kind of CPU their computers used, because the CPUs produced by AMD and other companies were almost the same as Intel's. It was not until the 486 era that people Be concerned about your CPU.

The era of 8086~80286 was the era when personal computers started. At that time, very few people in China used or even saw PCs. It was a mysterious thing in people's minds. It was not until the early 1990s that computers began to become popular in China.

Intel 80386

In the spring of 1985, Intel had become a first-class chip company, and it was determined to develop a new generation of 32-bit core CPU-80386. Intel designed three technical points for 80386: using the "286-like" structure, developing the 80387 microprocessor to enhance floating-point computing capabilities, and developing cache to solve the memory speed bottleneck.

On October 17, 1985, Intel's epoch-making product-80386DX was officially released. It contains 275,000 transistors and a clock frequency of 12.5MHz, which was gradually increased to 20MHz, 25MHz, 33MHz, and finally There are also a small number of 40MHz products.

The internal and external data buses of 80386DX are 32 bits, and the address bus is also 32 bits. It can address 4GB of memory and manage 64TB of virtual storage space. In addition to real mode and protected mode, its computing mode also adds a "virtual 86" working mode, which can provide multi-tasking capabilities by simulating multiple 8086 microprocessors at the same time.

80386DX has more instructions than 80286. The 80386 with a frequency of 12.5MHz can execute 6 million instructions per second, which is 2.2 times faster than the 80286 with a frequency of 16MHz. The most classic product of 80386 is 80386DX-33MHz, which is generally what we call 80386.

Due to the powerful computing power of 32-bit microprocessors, PC applications have expanded to many fields, such as commercial office and computing, engineering design and computing, data centers, and personal entertainment. The 80386 made 32-bit CPUs the PC industry standard.

Although the 80386 did not have a complete and powerful floating-point operation unit at the time, coupled with the 80387 co-processor, the 80386 could successfully complete many tasks requiring a large amount of floating-point operations, thus successfully entering mainstream commercial computers. market. In addition, 30386 also has other rich peripheral accessories support, such as 82258 (DMA controller), 8259A (interrupt controller), 8272 (disk controller), 82385 (Cache controller), 82062 (hard disk controller), etc. In response to the speed bottleneck of memory, Intel designed a cache (Cache) for the 80386, and adopted the method of pre-reading memory to alleviate this speed bottleneck. From then on, the Cache and the CPU became inseparable.

Intel 80387/80287

Strictly speaking, 80387 is not a real CPU, but a co-processing chip that cooperates with 80386DX. In other words, 80387 can only assist 80386 It completes the functions of floating point operations and has a very single function.

Intel 80386SX

In 1989, Intel launched the quasi-32-bit microprocessor chip 80386SX. This is a cheaper popular CPU launched by Intel in order to expand its market share. Its internal data bus is 32 bits and the external data bus is 16 bits. It can accept the 16-bit input/output interface chip developed for 80286. Reduce overall machine cost.

After the launch of 80386SX, it has been widely welcomed by the market because the performance of 80386SX is much better than that of 80286, and the price is only one-third of 80386.

Intel 80386SL/80386DL

In 1990, Intel launched two models of 386 chips, 80386SL and 80386DL, specifically for notebook computers. These two types of chips can be said to be energy-saving versions of 80386DX/SX. Among them, 80386DL is based on the 80386DX core, and 80386SL is based on the 80386SX core. These two types of chips not only consume less power, but also have power management functions that automatically cut off the power supply when the CPU is not working.

Motorola 68000

Motorola's 68000 was the earliest 32-bit pico processor launched in 1984. After its launch, it had outstanding performance and was favored by Apple, which was in full swing. This chip was used in the epoch-making personal computer "PC-MAC". But after the launch of 80386, it gradually declined.

AMD Am386SX/DX

AMD's Am386SX/DX is a third-party chip compatible with 80386DX. Its performance is almost the same as Intel's 80386DX, and it became one of the mainstream products at the time.

IBM 386SLC

This was designed by IBM based on research on 80386. It is fully compatible with 80386 and manufactured by Intel. 386SLC is basically a built-in Cache based on 80386SX, and also contains the instruction set of 80486SX, and its performance is also good.

Intel 80486

In 1989, the 80486 chip that we are all familiar with was launched by Intel. The great thing about this chip, which took four years of development and $300 million in capital investment, is that it breaks the 1 million transistor barrier for the first time, integrates 1.2 million transistors, and uses a 1-micron manufacturing process. The clock frequency of 80486 gradually increased from 25MHz to 33MHz, 40MHz, and 50MHz.

The 80486 integrates the 80386, the mathematics co-microprocessor 80387, and an 8KB cache into one chip. The number-crunching speed of the 80487 integrated in the 80486 is twice that of the previous 80387, and the internal cache shortens the waiting time of the microprocessor and slow DRAM. Moreover, RISC (reduced instruction set) technology is used for the first time in the 80x86 series, which can execute one instruction in one clock cycle. It also uses a burst bus method, which greatly improves the speed of data exchange with memory. As a result of these improvements, the 80486's performance is improved by a factor of 4 over the 80386 DX with the 80387 math co-microprocessor.

With the continuous development of chip technology, the frequency of CPU is getting faster and faster. However, PC external equipment is limited by process limitations and can withstand limited operating frequencies, which hinders the further improvement of CPU main frequency. In this case, CPU frequency doubling technology emerged, which makes the internal working frequency of the CPU 2 to 3 times the external frequency of the microprocessor. This is where the names of 486 DX2 and 486 DX4 come from.

Intel 80486 DX

Common 80486 CPUs include 80486 DX-33, 40, and 50. The 486 CPU is 32-bit internally and externally like the 386 DX, but the 486 CPU is the slowest It is also faster than the fastest 386 CPU. This is because the 486 SX/DX only needs one oscillation cycle to execute an instruction, while the 386DX CPU requires two cycles.

Intel 80486 SX

Because the 80486 DX CPU has a built-in floating-point co-microprocessor and is powerful, of course it is more expensive. In order to meet the needs of ordinary users, especially those who do not need to perform a lot of floating point operations, Intel introduced the 486 SX CPU. 80486 SX motherboards generally have an 80487 co-microprocessor socket. If you need the function of a floating-point co-microprocessor, you can plug in an 80487 co-microprocessor chip, which is equivalent to a 486 DX. Common 80486 SX CPUs include: 80486 SX-25, 33.

Intel 80486 DX2/DX4

In fact, the name of this CPU is related to the frequency. The internal frequency of this CPU is two/four times the motherboard frequency, such as 80486 DX2- 66, the frequency of the CPU is 66MHz, and the frequency of the motherboard only needs to be 33MHz.

Intel 80486 SL CPU

The 80486 SL CPU was originally designed for notebook computers and other portable computers. Like the 386SL, this chip uses a 3.3V power supply instead of a 5V power supply, and also has The internal circuit is cut off so that the microprocessor and some other optional components are in a sleep state when not working. This can reduce the energy consumption of notebook computers and other portable computers and extend the use time.

Intel 486 OverDrive

Upgrading 486 SX can install an 80487SX chip on the motherboard's co-microprocessor slot, making it equivalent to 486 DX, but after this upgrade, it is just Adding floating-point co-microprocessor capabilities does not increase system speed. In order to increase the speed of the system, there is another upgrade method, which is to plug a 486 OverDrive CPU into the co-microprocessor slot. Its principle is the same as the 486 DX2 CPU. Its internal operating speed can be twice the external speed. times. For example, after an OverDrive CPU is installed on a 20MHz motherboard, the internal operating speed of the CPU can reach 40MHz. The 486 OverDrive CPU also has the function of a floating-point co-microprocessor. Common ones are: OverDrive-50, 66, and 80.

TI 486 DX

As one of the world's leading semiconductor manufacturers, Texas Instruments (TI) also emerged in the 486 era. It produced its own 486 DX series CPUs, especially the 486DX2 After becoming mainstream, its DX2-80 became one of the mainstream products at the time due to its higher cost performance. The highest frequency of TI 486 was DX4-100, but it never entered the CPU market again.

Cyrix 486DLC

This is the 486 CPU produced by Cyrix. Saying it is a 486 CPU means that its efficiency is close to the 486 CPU, but it is not a 486 CPU in the strict sense. , which is determined by the characteristics of the 486 CPU. The 486DLC CPU only combines the 386DX CPU and 1K Cache in one chip. It does not contain a floating-point co-microprocessor and requires two oscillation cycles to execute an instruction. However, due to the exquisite design of the 486DLC CPU, the efficiency of the 486DLC-33 CPU is close to Intel's 486 SX-25, while the 486DLC-40 CPU exceeds the 486 SX-25, and the price of the 486DLC-40 CPU is cheaper than the 486 SX-25. The 486DLC CPU is designed to upgrade the 386DM. If you originally have a 386 computer and want to upgrade to a 486, but don't want to replace the motherboard, you can unplug the original 386 CPU and plug in a 486DLC CPU.

Cyrix 5x86

Since Intel took a different approach and developed Pentium, Cyrix quickly launched its own new generation product 5x86. It still uses the original 486 series CPU socket, but increases the main frequency from 100MHz to 120MHz. Compared with 486, the performance of 5x86 has increased, but compared with Pentium, not only the floating point performance is far insufficient, but even the integer operation performance that Cyrix has always been proud of is not that superb, giving people a feeling that it is not as good as the one above. The feeling of having more than enough. Since 5x86 can use 486 motherboards, it is generally regarded as a transitional product.

AMD 5x86

AMD 486DX is AMD's weapon in the 486 market. It has a built-in 16KB write-back cache, and has begun the era of multiple instructions in a single cycle. It also has paging virtual memory management. technology. Since TI later launched the 486DX2-80, the price was very low, Intel launched the Pentium series, and AMD launched the 5x86 series CPU in order to seize the market vacancy.

It is the product with the highest main frequency of 486 level, which is 5x86-120 and 133. It uses an integrated 16K write-back cache, 0.35 micron process, 33×4 133 frequency, performance is directly at the Pentiun 75, and power consumption is less than Pentium.

Intel Pentium

In 1993, the new generation 586 CPU, which comprehensively surpassed the 486, came out. In order to get rid of the confusion of microprocessor names in the 486 era, Intel named its new generation products For Pentium (Pentium) to distinguish AMD and Cyrix products. AMD and Cyrix also launched K5 and 6x86 microprocessors respectively to deal with the chip giants, but due to the best performance of Pentium microprocessors, Intel gradually occupied most of the market.

Pentium’s most entry-level CPUs are Pentium 60 and Pentium 66, which operate at the same frequencies of 60MHz and 66MHz respectively as the system bus frequency. There is no multiplier setting as we call it now.

Early Pentiums of 75MHz to 120MHz used a 0.5 micron manufacturing process, while later Pentiums with frequencies above 120MHz switched to a 0.35 micron process. Performance on the classic Pentium is pretty average, with both integer and floating point operations being good.

Intel Pentium MMX

In order to improve the application capabilities of computers in multimedia and 3D graphics, many new instruction sets have emerged. The three most famous ones are Intel's MMX, 3D NOW! from SSE and AMD. MMX (MultiMedia Extensions, Multimedia Extensions Instruction Set) is a multimedia instruction enhancement technology invented by Intel in 1996. It includes 57 multimedia instructions. These instructions can process multiple data at one time. MMX technology can be obtained with the cooperation of software Better performance.

The official name of Pentium MMX is "Pentium with MMX technology", which was released at the end of 1996. Starting from the Pentium Multi-Power, Intel began to lock multipliers on the CPUs it produced. However, MMX CPUs had particularly strong FSB capabilities, and could also overclock by increasing the core voltage, so overclocking was a very fashionable idea at that time. action. The term overclocking also became popular from that time.

Pentium is another successful product of Intel after Pentium, and its vitality is also quite tenacious. The multi-purpose Pentium has made major improvements based on the original Pentium, adding on-chip 16KB data cache and 16KB instruction cache, 4-way write cache, branch prediction unit and return stack technology. In particular, the newly added 57 MMX multimedia instructions make the multi-function Pentium much faster than the Pentium CPU with the same clock speed even when running non-MMX optimized programs.

These 57 MMX instructions are specially used to process audio, video and other data. These instructions can greatly shorten the waiting time of the CPU when processing multimedia data, giving the CPU more powerful data processing capabilities. Different from the classic Pentium, the multi-purpose Pentium adopts a dual-voltage design, its core voltage is 2.8V, and the system I/O voltage is still the original 3.3V. If the motherboard does not support dual voltage design, then you cannot upgrade to Pentium Multi.

The multi-purpose Pentium, code-named P55C, is the first CPU with MMX technology (integer unit execution). It has a 16KB data L1 Cache, a 16KB instruction L1 Cache, is compatible with SMM, and has a 64-bit bus. 528MB/s bandwidth, 2 clock latency, 4.5 million transistors, power consumption 17 watts. Supported working frequencies are: 133MHz, 150MHz, 166MHz, 200MHz, 233MHz.

Intel Pentium Pro

Once upon a time, Pentium Pro was synonymous with high-end CPUs. The performance of Pentium Pro surprised many people at the time, but Pentium Pro was designed with a 32-bit data structure CPU, so the Pentium Pro had mediocre performance running 16-bit applications, but was still the 32-bit winner, but later, the emergence of MMX eclipsed it.

The core architecture of Pentium Pro (high-power Pentium, 686-level CPU) is code-named P6 (also the core architecture used by PII and PIII in the future). This is the first generation product. The secondary cache has 256KB or 512KB, with a maximum secondary cache of 1MB. Working frequencies are: 133/66MHz (engineering sample), 150/60MHz, 166/66MHz, 180/60MHz, 200/66MHz.

AMD K5

K5 is AMD's first independently produced x86-class CPU, released in 1996. Because K5 encountered problems in development, its launch time was much later than Intel's Pentium. Coupled with poor performance, this unsuccessful product once caused AMD to lose a lot of market share. The performance of K5 is very average. Its integer computing power is not as good as Cyrix's 6x86, but it is still slightly better than Pentium. Its floating point computing power is far behind Pentium, but slightly better than Cyrix. Taken together, K5 is a product with relatively average strength. The low price of K5 is obviously more attractive to consumers than its performance. Low price is the biggest selling point of this CPU.

AMD K6

AMD was naturally unwilling to let Pentium dominate the CPU market, so they launched K6 in 1997. The design indicators of the K6 CPU are quite high. It has brand-new MMX instructions and 64KB L1 Cache (twice as much as Pentium MMX). Its overall performance is better than Pentium MMX and is close to the level of PII with the same main frequency. K6 can process more instructions in parallel and run at a higher clock frequency than K5. AMD has been very successful in integer operations. Where the K6 lags slightly behind is in running applications that require MMX or floating point operations. It is much worse than the Pentium with the same frequency.

K6 has 32KB data L1 Cache, 32KB instruction L1 Cache, integrates 8.8 million transistors, uses 0.35 micron technology, five-layer CMOS, C4 process reverse chip, and the core area is 168 square millimeters (the new product is 68 mm2), using Socket7 architecture.

Cyrix 6x86/MX

Cyrix can also be regarded as a veteran CPU developer. As early as the x86 era, it formed a three-power situation with Intel and AMD.

Since Cyrix merged with National Semiconductor, it finally has its own chip production line, and the finished products are becoming increasingly complete and complete. Cyrix's 6x86 was a Pentium-compatible microprocessor put on the market.

IDT WinChip

As a new CPU manufacturer joining this field, the American IDT Company (Integrated Device Technology) launched the first microprocessor product in 1997, WinChip (i.e. C6), accounting for less than 1 in the entire CPU market. In May 1998, IDT announced its second-generation product WinChip 2.

WinChip 2 has made some improvements on the basis of the original WinChip, adding a dual-instruction MMX unit and enhancing the floating-point operation function.

The performance of the improved WinChip 2 is improved by about 10% compared to the WinChip of the same frequency, basically reaching the performance of the Intel Pentium microprocessor.

Intel Pentium II

1997 to 1998 was a year of extremely fierce competition in the CPU market. The CPU chips during this period were so colorful that it was dizzying.

The Chinese name of Pentium II is "Pentium II". It has several series of products with different core structures such as Klamath, Deschutes, Mendocino, Katmai, etc. The first generation uses Klamath core and is manufactured with 0.35 micron process. There are 7.5 million transistors integrated inside, and the core operating voltage is 2.8V.

The Pentium II microprocessor adopts a dual independent bus structure, that is, one bus is connected to the secondary cache and the other is responsible for the main memory. Pentium II uses an off-chip external high-speed L2 Cache with a capacity of 512KB and runs at half the CPU clock speed. As a compensation, Intel increased Pentium II's L1 Cache from 16KB to 32KB. In addition, in order to defeat its competitors, Intel adopted the patented Slot 1 interface standard and SECC (Single Side Contact Box) packaging technology in Pentium II for the first time.

On April 16, 1998, Intel's first 350 and 400MHz CPUs code-named Deschutes that supported 100MHz rated FSB were officially launched. Pentium II microprocessors using the new core not only increase the external frequency to 100MHz, but they are also manufactured using a 0.25 micron process. Their core operating voltage has also been reduced from 2.8V to 2.0V. The L1 Cache and L2 Cache are 32KB and 512KB respectively. The supported chipset is mainly Intel's 440BX.

Between 1998 and 1999, Intel launched a CPU that was more powerful than Pentium II - Xeon (Xeon microprocessor). The core of this microprocessor is similar to that of Pentium II, with a 0.25 micron manufacturing process and supports 100MHz FSB. Xeon can be equipped with a maximum of 2MB Cache and runs at the CPU core frequency. It is different from the chip used by Pentium II and is called CSRRAM (Custom StaticRAM, customized static memory). In addition, it supports eight CPU systems; uses 36-bit memory addresses and PSE mode (PSE36 mode), with a maximum memory bandwidth of 800MB/s. Xeon microprocessors are mainly targeted at servers and workstation systems with higher performance requirements. In addition, Xeon's interface form has also changed, adopting a Slot 2 architecture that is slightly larger than Slot 1 (can support four microprocessors).

Intel Celeron (Celeron)

In order to further seize the low-end market, Intel launched a cheap CPU-Celeron (Chinese name Celeron) in April 1998. The initially launched Celeron has two versions, 266MHz and 300MHz, both of which use Covington core and are manufactured with 0.35 micron process. They integrate 19 million transistors and 32KB level one cache. The operating voltage is 2.0V and the external frequency is 66MHz. Compared with Pentium II, Celeron removes the on-chip L2 Cache. Although this greatly reduces the cost, because there is no second-level cache, the performance of the microprocessor is greatly reduced, and its integer performance is even inferior to Pentium MMX.

In order to make up for the shortcomings in the performance of Celeron microprocessors that lacked L2 cache and further undermine competitors in the low-end market, Intel released a new processor using the Mendocino core shortly after the launch of Celeron266 and 300. Celeron microprocessors—Celeron300A, 333, 366. Different from the old Celeron, the new Celeron is manufactured using a 0.25 micron process. It also uses Slot 1 architecture and SEPP packaging. It has built-in 32KB L1 Cache and 128KB L2 Cache, and works at the same core frequency of the CPU, thus greatly improving L2 performance. Cache working efficiency.

AMD K6-2

AMD officially launched the K6-2 microprocessor in April 1998. It is manufactured using a 0.25 micron process, the chip area is reduced to 68 square millimeters, and the number of transistors is increased to 9.3 million. In addition, K6-2 has a 64KB L1 Cache, the second-level cache is integrated on the motherboard, the capacity ranges from 512KB to 2MB, the speed is synchronized with the system bus frequency, the operating voltage is 2.2V, and it supports Socket 7 architecture.

K6-2 is a K6 chip plus 100MHz bus frequency and supports 3D Now! A "combination" of floating-point instructions. 3D Now! The technology is a major breakthrough in the x86 system, which greatly enhances the intensive floating-point computing performance required to process 3D graphics and multimedia. In addition, K6-2 supports superscalar MMX technology and supports 100MHz bus frequency, which means that the transfer rate of the system and L2 cache and memory is increased by nearly 50, thereby greatly improving the performance of the entire system.

Cyrix MⅡ

As the last microprocessor independently developed by Cyrix, Cyrix MII began production in March 1998. In addition to the characteristics of 6x86 itself, this microprocessor also supports MMX instructions. Its core voltage is 2.9V and has 256 bytes of instructions; 3.5X frequency multiplication; 6.5 million transistors are integrated in the core and the power consumption is 20.6 watts; 64KB level cache.

Rise mp6

Rise is an American company founded in November 1993. It mainly produces x86-compatible CPUs and launched the mP6 CPU in 1998. MP6 is not only cheap, but also has excellent performance, good multimedia performance and powerful floating point operations. MP6 uses Socket 7/Super 7 compatible sockets and has only 16KB of first-level cache.

Intel Pentium III

Just after the Spring Festival of 1999, Intel released a new generation of microprocessor using the Katmai core - Pentium III. In addition to being manufactured using a 0.25 micron process, integrating 9.5 million transistors, and having a Slot 1 architecture, this microprocessor also has the following new features: the system bus frequency is 100MHz; it uses the sixth-generation CPU core—P6 microarchitecture, targeting 32 Optimized for applications, dual independent buses; the first-level cache is 32KB (16KB instruction cache plus 16KB data cache), the second-level cache size is 512KB, running at half the CPU core speed; using SECC2 packaging; newly added to enhance SSE (Streaming SIMD Extensions, Data Streaming Single Instruction Multiple Data Extensions) instruction set for audio, video and 3D graphics effects, with 70 new instructions. The starting clock speed of Pentium III is 450MHz.

Like Pentium Ⅱ Xeon, Intel has also launched a high-performance CPU for server and workstation systems - Pentium Ⅲ Xeon Xeon microprocessor.

In addition to the early Pentium II Xeon500 and 550 using 0.25 micron technology, this microprocessor is manufactured using 0.18 micron technology, Slot 2 architecture and SECC packaging, with built-in 32KB level one cache and 512KB level two cache, and an operating voltage of 1.6V.

Intel Celeron II

In order to further consolidate its advantages in the low-end market, Intel launched Celeron II using Coppermine core on March 29, 2000. This microprocessor is also manufactured using a 0.18 micron process. The core integrates 19 million transistors and is packaged in FC-PGA. Like the Celeron Mendocino, it has a built-in 128KB L2 Cache that runs synchronously with the CPU, so its core is also called Coppermine. 128. Celeron II does not support multi-microprocessor systems. However, the FSB of Celeron II is still only 66MHz, which limits its performance to a great extent.

AMD K6-III

AMD launched the K6-III code-named "Sharptooth" in February 1999. It was the company's last model to support Super 7 The architecture and CPGA package form of the CPU use a 0.25 micron manufacturing process, the core area is 135 square millimeters, and it integrates 21.3 million transistors. The operating voltage is 2.2V/2.4V.