According to the word length and function of microprocessor, its development can be divided into the following stages.
Stage 1
1 stage (1 971-1973) is the era of 4-bit and 8-bit low-end microprocessors, which is usually called1generation. Its typical products are Intel4004 and Intel8008 microprocessors and MCS-4 and MCS-8 microcomputers composed of them respectively. The basic features are PMOS technology, low integration (4000 transistors/chip), simple system structure and instruction system, mainly using machine language or simple assembly language, few instructions (more than 20), and the basic instruction cycle is 20~50μs, which is used for simple control occasions.
1969, Intel began to develop the first microprocessor for a project of Japanese computer manufacturer Busicom, and developed various chips for a series of programmable computers. Finally, Intel introduced the 4004 microprocessor to the global market with197115. At that time, the price of each Intel 4004 processor was $200. 4004 is Intel's first microprocessor, which lays the foundation for the future development of system intelligence and personal computers. Its number of transistors is about 2300.
Second phase
The second stage (1974- 1977) is the era of 8-bit middle and high-grade microprocessors, which is usually called the second generation. Its typical products are Intel8080/8085, Motorola, Zilog Z80 and so on. Their characteristic is NMOS technology, which improves the integration by about 4 times and the operation speed by about 10~ 15 times (the execution time of basic instructions is 1~2μs). The instruction system is relatively complete, with typical computer architecture and control functions such as interrupt and DMA. In terms of software, besides assembly language, there are high-level languages such as BASIC and FORTRAN, corresponding interpreters, compilers, and later operating systems.
1974, Intel introduced the 8080 processor as the computing core of Altair personal computer. Altair is the destination of the spaceship Enterprise in the TV series Starship Adventure. At that time, computer enthusiasts could buy an Altair suite for $395. It sold tens of thousands of sets in a few months, becoming the first model manufactured after placing an order in history. The number of Intel 8080 transistors is about 6000.
Stage 3
The third stage (1978- 1984) is the era of 16-bit microprocessor, which is usually called the third generation. Its typical products are Intel's 8086/8088, Motorola's M68000 and Zilog's Z8000. Its characteristic is that HMOS technology is adopted, and its integration (20,000 ~ 70,000 transistors/chip) and operation speed (the execution time of basic instructions is 0.5μs) are one order of magnitude higher than those of the second generation. The instruction system is richer and more perfect, with multi-level interrupts, multiple addressing modes, segmented storage mechanism, hardware multiplication and division components and a software system. The famous microcomputer products in this period include IBM's personal computer. 198 1 year, the personal computer introduced by IBM uses 8088CPU. Then 1982 introduced the extended personal computer IBM PC/XT, which expanded the memory and increased the hard disk.
80286 (also known as 286) is Intel's first processor that can execute all proprietary software of the old processor. After compatibility, the software became the registered trademark of Intel's full range of microprocessors. During the six-year sales period, it is estimated that a total of150,000 286 personal computers have been installed worldwide. The number of transistors in the Intel 80286 processor is 134000. 1984, IBM introduced the 16-bit enhanced personal computer IBM PC/AT with 80286 processor as the core. Because IBM adopted the strategy of opening technology when developing personal computers, personal computers are popular all over the world.
The fourth stage
The fourth stage (1985- 1992) is the era of 32-bit microprocessors, also known as the fourth generation. Its typical products are Intel's 80386/80486, Motorola's M69030/68040 and so on. It is characterized by adopting HMOS or CMOS technology, the integration level is as high as 6.5438+0 million transistors/chip, and it has 32-bit address lines and 32-bit data buses. 6 million instructions (MIPS) can be completed per second. The function of microcomputer has reached or even surpassed that of super microcomputer, and it is fully capable of multi-task and multi-user operation. At the same time, some other microprocessor manufacturers (such as AMD and Texas) also introduced 80386/80486 series chips.
80386DX has 32-bit internal and external data bus and 32-bit address bus, which can address 4GB of memory and manage 64TB of virtual storage space. In addition to real mode and protection mode, its operation mode also adds a virtual 86 working mode, which can provide multi-task capability by simulating multiple 8086 microprocessors at the same time. 80386SX is a cheap and popular CPU introduced by Intel to expand its market share. Its internal data bus is 32 bits, and its external data bus is 16 bits. The 16-bit I/O interface chip developed for 80286 can be accepted to reduce the cost of the whole machine. After its launch, 80386SX has been widely welcomed by the market, because its performance is much better than that of 80286, and its price is only one third of that of 80386. Intel 80386 microprocessor contains 275,000 transistors, which is 0/00 times more than the original 4004. This 32-bit processor supports multitasking design for the first time, and can execute multiple programs at the same time. The number of Intel 80386 transistors is about 275000.
1989, the 80486 chip that we are all familiar with is introduced by Intel. The greatness of this chip, which took four years to develop and invested $300 million, is that it actually broke the boundary of 6,543.8+0,000 transistors for the first time, integrated 6,543.8+0.2 million transistors, and adopted a manufacturing process of 654.38+0 micron. The clock frequency of 80486 is gradually increased from 25MHz to 33MHz, 40MHz and 50MHz.
80486 integrates 80386, math coprocessor 80387 and 8KB cache on one chip. The digital operation speed of 80487 integrated in 80486 is twice that of the previous 80387, and the internal cache shortens the waiting time of microprocessor and slow DRAM. Moreover, the 80x86 series adopts RISC (Reduced Instruction Set) technology for the first time, and one instruction can be executed in one clock cycle. It also adopts the burst bus mode, which greatly improves the data exchange speed with the memory. Because of these improvements, the performance of 80486 is four times higher than that of 80386 DX with 80387 math coprocessor.
The fifth stage
The fifth stage (1993-2005) is the era of Pentium series microprocessors, which is usually called the fifth generation. Typical products are Intel Pentium series chips and compatible AMD K6 and K7 series microprocessor chips. It adopts superscalar instruction pipeline structure internally, with independent instruction and data cache. With the appearance of MMX microprocessor, the development of microcomputer in networking, multimedia and intelligence has reached a new height.
The Pentium II processor introduced by 1997 combined with Intel MMX technology can efficiently process movies, sound effects and drawing data. Single-sided contact (S.E.C) box package with built-in cache is adopted for the first time. This chip allows computer users to capture and edit digital photos and share them with friends and relatives through the network, edit and add text, music or transitional effects of making home movies, and use videophones to transmit movies through standard telephone lines and the Internet. The number of transistors in the Intel Pentium II processor is 7.5 million.
The Pentium III processor launched by 1999 has added 70 new instructions. The SIMD extension set of Internet streaming named MMX can greatly improve the performance of advanced images, 3D, streaming music, movies, speech recognition and other applications. It can greatly improve the experience of using the Internet, enabling users to browse realistic online museums and shops and download high-quality movies. Intel introduced the 0.25 micron process for the first time, and the number of Intel Pentium III transistors was about 9.5 million.
In the same year, Intel also released the Pentium iii Xeon processor. As the successor of Pentium II Xeon, it not only adopts a brand-new design in the kernel architecture, but also inherits 70 instruction sets added by Pentium III processor, which can better execute multimedia and streaming media applications. In addition to facing the enterprise market, Pentium III Xeon has also strengthened its e-commerce application and advanced business computing capabilities. There are also many improvements in cache speed and system bus structure, which greatly improves the performance and is designed for better multi-processor cooperation.
In 2000, Intel released the Pentium 4 processor. Users can use personal computers based on Pentium 4 processors to make professional-quality movies, transmit TV-quality images through the Internet, conduct real-time voice and image communication, perform real-time 3D rendering, quickly perform MP3 encoding and decoding operations, and run various multimedia software when connecting to the Internet.
Pentium 4 processor integrated 42 million transistors, and the improved Pentium 4 (Northwood) integrated 55 million transistors. And began to use 0. 18 micron for manufacturing, and the initial speed reached 1.5GHz?
Pentium 4 also provides SSE2 instruction set, adding 144 new instructions. In SSE, the data compressed by 128bit can only be processed in the form of four single-precision floating-point values, while in SSE2 instruction set, the data can be processed in various data structures:
Four single-precision floating-point numbers (SSE 2); Corresponding to two double-precision floating-point numbers (SSE 2); Corresponding to 16 bytes (sse2); Corresponding to 8 words (words); Corresponding to 4 double words (SSE 2); Corresponding to two four-word numbers (sse2); Corresponding to 1 integer, 128 bits (SSE2).
In 2003, Intel released the Pentium M (mobile) processor. In the past, although there were mobile versions of Pentium II, III and even Pentium 4-M products, these products were still designed based on desktop computer processors, adding some new functions of energy saving and management. Even so, the energy consumption of Pentium III-M and Pentium 4-M is much higher than that of CPU specially designed for mobile computing, such as Quanmeida's processor.
Intel Pentium M processor combines 855 chipset family and Intel PRO/Wireless2 100 network connection technology, and becomes the most important part of Intel Centrino mobile computing technology. Pentium M processor can provide the main frequency speed up to 1.60GHz, and includes various performance enhancement functions, such as 400MHz system bus for optimizing power supply, micro-operating system and dedicated stack manager. These tools can quickly execute instruction sets and save power.
In 2005, Intel introduced Pentium D and Pentium Extreme Edition, and also introduced 945/955/965/975 chipset to support new dual-core processors. These two new dual-core processors produced by 90nm technology adopt LGA 775 interface and have no pins, but the number of chip capacitors at the bottom of the processors has increased and the arrangement is different.
The processor of the desktop platform, code-named Smithfield, is officially named Pentium D processor. In addition to removing Arabic numerals, English letters represent the generation alternation of dual-core processors, and the letter D is more reminiscent of the meaning of dual-core.
Intel's dual-core architecture is more like a dual-CPU platform, and Pentium D processors continue to use Prescott architecture and 90nm production technology. In fact, the Pentium D kernel consists of two independent Prescott kernels. Each kernel has an independent L2 cache and an execution unit of 1mB, which adds up to 2MB. However, because the two cores in the processor have independent caches, it is necessary to ensure that the information in each secondary cache is completely consistent, otherwise there will be operational errors.
In order to solve this problem, Intel handed over the coordination between the two cores to the external MCH (North Bridge) chip. Although the data transmission and storage between caches are not huge, it will undoubtedly bring some delay to the whole processing speed because of the need to coordinate processing through external MCH chips, thus affecting the overall performance of the processor.
Due to Prescott kernel, Pentium D also supports EM64T technology and XD bit security technology. It is worth mentioning that the Pentium D processor will not support hyper-threading technology. The reason is obvious: it is not easy to correctly distribute data streams and balance computing tasks between multiple physical processors and multiple logical processors. For example, an application needs two computing threads. Obviously, each thread corresponds to a physical kernel, but what if there are three computing threads? Therefore, in order to reduce the complexity of the dual-core Pentium D architecture, Intel decided to cancel the support for Hyper-Threading Technology in the Pentium D for the mainstream market.
Both come from Intel, Pentium D and Pentium Extreme Edition. The different names indicate that the specifications of these two processors are also different. The biggest difference between them is the support for hyper-threading technology. Pentium d does not support hyper-threading technology, and Pentium Extreme Edition does not have this restriction. With Hyper-Threading enabled, the dual-core Pentium Extreme Edition processor can simulate two other logical processors and can be recognized as a quad-core system by the system.
Pentium EE series is marked with three digits in the form of Pentium EE8xx or 9xx, such as Pentium EE840. The larger the number, the higher the specification or the more functions supported.
Pentium EE 8x0: indicates that this product is Smithfield core, with 1MB L2 cache per core and 800MHzFSB. The only difference between it and Pentium D 8x0 series is that it only adds support for hyper-threading technology, and other technical features and parameters are the same.
Pentium EE 9x5: indicates that this product is a Presler core, with 2MB L2 cache per core, 1066MHzFSB. The difference between it and Pentium D 9x0 series is that it adds support for hyper-threading technology and improves the front-end bus to 1066MHzFSB. Other technical characteristics and parameters are exactly the same.
Single-core CPUs such as Pentium 4, Pentium 4 EE and Celeron D, and dual-core CPUs such as Pentium D and Pentium EE are all packaged in LGA775. Different from the previous socket 478 interface CPU, the bottom of LGA 775 interface CPU has no traditional pins, but 775 contacts, that is, contacts with 775 contacts in corresponding LGA 775 slots to transmit signals. LGA 775 interface can not only effectively improve the signal strength and frequency of the processor, but also improve the yield of the processor and reduce the production cost.
The sixth stage
The sixth stage (from 2005 to now) is the era of Core series microprocessors, which is usually called the 6th generation. Core is a leading energy-saving new micro-building. The starting point of the design is to provide excellent performance and energy efficiency, and improve the performance per watt, which is the so-called energy efficiency ratio. The early Core was based on notebook processors. Core 2: English name is Core 2 Duo, which is a new generation product system based on Core micro-architecture launched by Intel in 2006. Published on July 27th, 2006. Core 2 is a cross-platform architecture, including server version, desktop version and mobile version. Among them, the development code of the server version is Woodcrest, the development code of the desktop version is Conroe, and the development code of the mobile version is Merom.
Core microarchitecture of Core 2 processor is a new generation Intel architecture improved by Intel Israel design team on the basis of Yonah microarchitecture. The most significant change lies in the strengthening of various key parts. In order to improve the efficiency of data exchange between two cores, a design of * * * shared secondary cache is adopted, and two cores * * * enjoy up to 4MB of secondary cache.
Following the LGA775 interface, Intel launched the LGA 1366 platform for the first time, positioning the high-end flagship series. The first processor with LGA 1366 interface was code-named Bloomfield, which adopted an improved Nehalem core, based on 45nm process and native quad-core design, and built-in 8- 12MB three-level cache. Intel Hyper-Threading Technology is introduced into LGA 1366 platform again, and QPI bus technology replaces the front-end bus design that has been used since Pentium 4 era. The most important thing is that LGA 1366 platform supports three-channel memory design, which greatly improves the actual performance, which is also a big difference between LGA 1366 flagship platform and other platforms in positioning.
As the representative of high-end flagship, the processors of LGA 1366 interface mainly include 45nm Bloomfield Core i7 quad-core processors. With the 20 10 Intel entering the 32nm process, the representative of the high-end flagship was replaced by the Core i7-980X processor. The brand-new 32nm process solves the six-core technology and has the most powerful performance. For users who are ready to build a high-end platform, LGA 1366 still occupies the high-end market, and Core i7-980X and Core i7-950 are still good choices.
Core i5 is a quad-core processor based on Nehalem architecture, which adopts integrated memory controller and three-level cache mode, L3 reaches 8MB, and supports the new processor computer configuration of Turbo Boost and other technologies. The main difference between it and Core i7(Bloomfield) is that the bus does not use QPI, but uses mature DMI (Direct Media Interface), and only supports dual-channel DDR3 memory. LGA 1 156 interface is adopted in the structure, and i5 has turbo technology, which can overclock under certain conditions. The processors with LGA 1 156 interface cover different users from entry-level to high-end, and the 32nm process brings lower power consumption and better performance. The mainstream representative is Core i5-650/760, and the middle and high-end representative is Core i7-870/870K. We can clearly see Intel's positioning distinction in product naming. But overall, the high-end LGA 1 156 processor is more worth buying than the low-end entry. Faced with AMD's low-cost strategy, Intel Core i3 series processors can't match it in cost performance. The performance of LGA 1 156 high-end products is more eye-catching.
Core i3 can be regarded as a further simplified version (or castrated version) of Core i5, and there will be a 32nm process version (development code-named Clarkdale, based on Westmere architecture). The biggest feature of Core i3 is the integration of GPU (graphics processor), which means that Core i3 will be encapsulated by two cores: CPU and GPU. Due to the limited performance of integrated GPU, users can add graphics cards if they want to get better 3D performance. It is worth noting that even in Clarkdale, the manufacturing process of the display core will still be 45 nanometers. The biggest difference between i3 i5 is that i3 has no turbo technology. Represented by Core i3-530/540.
20 10 in June, Intel once again released the revolutionary processor-the second generation Core i3/i5/i7. The second-generation Core i3/i5/i7 belongs to the second-generation intelligent Core family, all of which are based on the brand-new Sandy Bridge micro-architecture. Compared with the first generation products, it mainly brings five important innovations: 1, and adopts a brand-new 32nm Sandy Bridge micro-architecture, with lower power consumption and stronger performance. 2, built-in high-performance GPU (core graphics card), video coding, graphics performance is stronger. 3, Turbo Boost Technology 2.0, smarter and more efficient. 4. A new ring architecture is introduced, which brings higher bandwidth and lower delay. 5. Brand-new AVX and AES instruction sets strengthen floating-point operation and encryption and decryption operation.
SNB(Sandy Bridge) is a new generation of processor microarchitecture released by Intel at the beginning of 20 1 1. The greatest significance of this architecture lies in redefining the concept of "integrated platform", and the "core graphics card" seamlessly integrated with the processor ends the era of "integrated graphics card". This pioneering work benefits from the brand-new 32 nm manufacturing process. Because the processor under the framework of Sandy Bridge adopts a more advanced 32nm manufacturing process than the previous 45nm process, the power consumption of CPU is further reduced theoretically, and the circuit size and performance are significantly optimized, which creates favorable conditions for the integrated graphics core (core graphics card) and CPU to be packaged on the same substrate. In addition, the second-generation Core also added a brand-new high-definition video processing unit. The speed of video conversion and decoding is directly related to the processor. Due to the addition of high-definition video processing unit, the video processing time of the new generation Core processor is at least 30% longer than that of the old processor. The new generation of Sandy Bridge processor adopts a new interface design of LGA 1 155, which is not compatible with LGA 1 156. Sandy Bridge is a new micro-architecture, which will replace Nehalem, but will still adopt 32 nm process. What's more attractive is that this time, Intel no longer uses "glue" to stick the CPU core and GPU core together, but really combines them into one.
20 12 Beijing planetarium on the afternoon of April 24th, Intel officially released the Ivy Bridge (IVB) processor. 22nm Ivy Bridge will double the number of execution units to a maximum of 24, which will naturally bring about a further leap in performance. Ivy Bridge will add an integrated graphics card supporting DX 1 1. In addition, the newly added XHCI USB3.0 controller * * * enjoys four channels, so it can provide up to four USB3.0, thus supporting native USB 3.0. The production of cpu adopts 3D transistor technology, and the power consumption of CPU will be reduced by half. Ivy Bridge products with 22nm technology will continue the service life of LGA 1 155 platform, so users who intend to buy LGA 1 155 platform will not have to worry about interface upgrade for at least one year.
On June 4th, 20 13, Intel released the fourth-generation CPU "haswell". The pins (CPU sockets) of the fourth-generation CPU were named Intel LGA 1 150, and the motherboards were named eight series chipsets such as Z87, H87 and Q87, among which Z87 was an overclocking player and a high-end customer base. Haswell CPU will be used for laptop, desktop CEO package and DIY component CPU, replacing the current third-generation Ivy Bridge.