Impeccable work personal computer and its performance depends mainly on the processor it is equipped with. Therefore, when buying a computer, it is simply necessary to pay attention to which company made its processor.
The main manufacturers of PC processors today are Intel and AMD. They, of course, compete with each other. Here are short characteristics of the main families of processors of these brands, knowledge of which may be useful when choosing them. So,
Intel Processors
There are four main families of Intel processors:
Single-core and dual-core processors of the Celeron family. The former are traditional and proven, but when choosing, it is better to give preference to the latter, since they are more productive and their price is not much higher than single-core ones.
Pentium is a family of single-core processors (among which it is better to choose models of the sixth series with a 2MB cache) and their dual-core modifications.
Core2 is a whole line of multi-core processors with two, three and quad-core modifications. When choosing such a processor, you need to pay attention to the cache size and bus frequency. And, of course, on your financial capabilities.
Core i7 – quad-core processors for high-performance computers.
AMD Corporation Processors
Sempron is an analogue of the budget Celeron processor.
Athlon is an analogue of Pentium, used in medium-power computers.
Phenom is a family of powerful processors designed for creating gaming computers.
Phenom II is the most powerful processor produced by AMD Corporation.
These are the main manufacturers of PC processors and their most used products presented on the modern computer market.
To choose good smartphone, it is important to rely not only on appearance gadget, but also on its “stuffing”. A powerful processor is an undoubted advantage for a device, but when choosing a smartphone, a buyer cannot always determine exactly how good the processor installed in it is. Often this happens due to the fact that people simply do not know which processor manufacturing companies are top. In this article we will try to clarify this issue in detail.
One of the undisputed leaders in the modern smartphone processor market is Qualcomm. It was founded in 1985 in San Diego, California, by two MIT professors, Irwin Jacobs and Andrew Viterbi. The company was engaged in research in the field of wireless communications, as well as the development of single-chip circuits (SoC). Qualcomm has collaborated with corporations such as Ericsson, Kyocera and Atheros.
Qualcomm's range of activities included the production of mobile processors and communication solutions for smartphones. The line of processors is based on the ARM architecture and has a wide the lineup, divided into several classifications: earlier Qualcomm S1, S2, S3 and S4 processors, and modern Qualcomm 200, 400, 600 and 800.
Most powerful processor at the beginning of 2015 is the Snapdragon 810, which first appeared in the LG G FLEX2 smartphone. It has 8 nuclear processor Qualcomm Snapdragon 810 (MSM8994), clocked at up to 2 GHz. 
The previous version of Snapdragon 805 is used in Samsung smartphones Galaxy S5, Google Nexus 6, LG G3. The number of “points” when testing using Antutu applications Benchmark – 37780.
Nvidia was born in 1993 in Santa Clara, California, where its headquarters are still located. The founder of the company is businessman and electronic technology specialist Huang Zhen Xun. 
The name Nvidia is known to almost every personal computer user, as it is the manufacturer of the popular line of video cards for PCs and laptops, Nvidia GeForce. The company is also developing processors for mobile devices (tablets, smartphones, etc.) based on ARM, united in the general Tegra line (Tegra 2,3, 4, K1, etc.).
The latest generation of Tegra processors is Nvidia Tegra K1. Its characteristics are 2.3 GHz frequency and four cores. This processor is used in Google Nexus, Lenovo and Acer devices. Antutu points – 43851. 
The South Korean company Samsung was founded back in 1938 as a food supply company. However, by the end of the 60s, the company reformed quite extensively and switched to the production of electronics, which is still its main area of activity. The headquarters is located in Seoul.
Samsung produces a very wide range of devices: mobile phones, smartphones, tablets, monitors, DVD players, etc. Of course, being one of the world's largest smartphone manufacturers, the company could not ignore the production of processors for these devices.
The Samsung processor line is called Exynos. The base is the ARM architecture. At the end of 2014, the most modern are Samsung processors Exynos 5 Octa 5420 (1.9 GHz, four cores) and Samsung Exynos 5 Octa 5422 (2.1 GHz, four cores). Used in a row Samsung devices Galaxy: S5, Note 3, etc. Apple and Samsung also agreed to cooperate and in 2015 smartphones and Apple tablets will be released with processors produced at the Samsung plant.
Antutu points for Exynos 5 Octa 5420 – 34739. 
MediaTek MT
Founded in 1997 by Chinese businessmen and electronics specialists Zai Minggai and Zhuo Jingzhe, the company is based in Taiwan's Hi-Tech Park in Xinchu (though has many branches around the world) and develops data storage systems, components for mobile phones, smartphones and tablets.
This company is most widely known for its production of processors for mobile devices in different price categories. Mediatek is called Qualcomm's main competitor. The most productive processors for smartphones at the end of 2014 are MT6595 (2 GHz, 4 cores), MT6735 (1.5 GHz and 4 cores) and MT6592M (8 cores and 2 GHz). MT processors are used by many smartphone manufacturing companies, from Sony to LG. Antutu rating for MT6592 is 30217.
The choice of smartphones is quite wide, as is the range of characteristics. The buyer just needs to choose the right one! Be careful when choosing a smartphone, and it will serve you faithfully for a long time.
You will also like:
Four magnificent compact smartphone With powerful batteries
What is a good and inexpensive smartphone? Review of three ultra-budget models
Feature Comparison: Samsung Galaxy Note 8 vs Galaxy S8+ vs LG G6 vs iPhone 7 Plus
Russian processor Elbrus-8S
Good afternoon, dear readers. Today's topic will be very interesting to avid patriots. Go Russia!!! And today we’ll talk about Russian processors “ Elbrus" And " Baikal" It’s a shame that the article really can’t be called “ Russian-made processors", because in fact they are produced in East Asia (like most of the world's leading electronics), and not in Russia. But we can be proud that Russia is one of the few countries in the world that is capable of developing its own microprocessors, because the future lies behind them.
Are there those among you who, to search for an article, entered the phrase “ Russian processors"? If we talk about people, then “ Not all Russians are Russians" And if we talk about processors, then they Russian. Info 100%, I checked!
So what do we have for today? And today we are in the first half of 2017 and Russian processors are developing relentlessly.
Russian processors "Processor-9" with support for DDR4 memory
What do we see in the subtitle? With the support ! This means nothing more than that Processor-9 will be in direct competition with existing giants Intel and AMD. Here you can really be proud of Russia.
What is Processor-9? This is the code name of a top Russian processor Elbrus-16S from the MCST company. It is planned to begin production in 2018. There will be two processor options with 8 and 16 cores. In general, the processor characteristics are:
Main technical characteristics of the Elbrus-16S processor (Processor-9)
Previously, computers based on Russian Elbrus processors were already sold. 4 C, but they cost an exorbitant amount of money. This was due to the fact that mass production of processors had not been established. These computers were rather experimental models, and therefore cost up to 400,000 rubles. In the case of Elbrus-16S, the situation will be corrected by mass production of processors in Taiwan. In addition, the manufacturer must understand that at such a price there can be no talk of any competitiveness.
Why don't we compare information about the entire line of Elbrus processors? It's interesting.
| Elbrus-2C+ | Elbrus-4S | Elbrus-8S | Elbrus-16S | |
| Year of issue | 2011 | 2014 | 2015-2018 (revisions) | 2018 (plan) |
| Clock frequency | 500 MHz | 800 MHz | 1300 MHz | 1500 MHz |
| Bit depth | I don't know | 32/64 bit | 64 bit | 64/128 bit |
| Number of cores | 2 | 4 | 8 | 8/16 |
| Level 1 cache | 64 KB | 128 KB | — | — |
| Level 2 cache | 1 MB | 8 MB | 4 MB | 4 MB |
| Level 3 cache | — | — | 16 MB | 16 MB |
| RAM support | DDR2-800 | 3 x DDR3-1600 | 4 x DDR3-1600 | 4 x DDR4-2400 |
| Technical process | 90 nm | 65 nm | 28 nm | 28 nm (or 16) |
| Power consumption | 25 W | 45 W | 75-100 W | 60-90 W |
There were also developments of processors that did not pass state certification. But that was a long time ago and not true.
What do you think about Russian processors? Would you buy a computer for 400,000 just because it is Russian? Write, let's talk about this topic.
Russian Elbrus processors compared to Intel
I know that many people are interested in comparing Russian processors with Intel processors. This is not surprising, Russians are a proud people, and therefore we want to compare our achievements with the best. And Intel is exactly like that in the world of computer processors.
In general, there is a certain tablet floating around the network comparing Elbrus processors with Intel, but decide for yourself how reliable it is. As I understand, this table is not new, because the comparison is not with the newest Intel processors, but some of them still cannot be called old. Moreover, some of them are powerful servers Intel processors Xeon. In the table you can compare the main technical characteristics, as well as the performance of processors in Gigaflops.
In general, here is the processor comparison table itself. I am inserting it in the form in which I found it, do not judge strictly. It’s a pity that there is only a comparison between Elbrus and Intel, and there are no Baikal processors there, but I think there will still be enthusiasts who will correct this shortcoming.
Russian processors Elbrus: comparison with Intel
Russian processors Baikal-T1 and Baikal-M
If Elbrus processors are intended purely for computers and are ready to compete with other manufacturing companies, then Baikal processors are intended more for the industrial segment and will not face such tough competition. However, Baikal-M processors are already being developed, which can be used for desktop PCs.
Processor Baikal-T1
According to Baikal Electronics, processors Baikal-T1 can be used for routers, routers and other telecommunications equipment, for thin clients and office equipment, for multimedia centers, CNC systems. But the processors Baikal-M can become the heart of work PCs, industrial automation and building management. Already more interesting! But detailed information O technical specifications Not yet. We only know that it will run on 8 ARMv8-A cores and will have up to eight ARM Mali-T628 graphics cores on board and, what is also important, the manufacturers promise to make it very energy efficient. Let's see what happens.
While I was writing the article, I made a request to Baikal Electronics JSC, and the answer was not long in coming. Dear Andrey Petrovich Malafeev (public relations and corporate events manager) kindly shared with us the latest information about the Baikal-M processor.
The company plans to release the first engineering samples of the Baikal-M processor this fall. And then I quote, so as not to distort the essence of the information in any way:
— Start of quote —
The Baikal-M processor is a system on a chip that includes energy-efficient processor cores with ARMv 8 architecture, a graphics subsystem and a set of high speed interfaces. Baikal-M can be used as a trusted processor with extensive data protection capabilities in a number of devices in the B2C and B2B segments.
Areas of application of Baikal-M
- monoblock, automated workstation, graphic workstation;
- home (office) media center;
- video conference server and terminal;
- microserver;
- Small enterprise level NAS;
- router/firewall.
The high degree of integration of the Baikal -M processor allows the development of compact products in which the main share of added value comes from the domestic processor. Availability complete information O logical circuit and the physical topology of the chip, in combination with trusted software and appropriate hardware solutions, allows the processor to be used as part of systems designed to process confidential information.
Applicable software
The widespread use of the ARMv8 (AArch64) architecture allows the use of a huge amount of ready-made application and system software. Supported OS Linux and Android, including at the level of binary distributions and packages. Numerous devices are available that connect to PCIe and USB buses. The software package supplied by Baikal Electronics includes the Linux kernel in source and compiled form, as well as drivers for the controllers built into Baikal-M.
Main characteristics of the Baikal-M processor
- 8 ARM Cortex-A57 cores (64 bit).
- Operating frequency up to 2 GHz.
- Hardware support for virtualization and Trust Zone technology at the level of the entire SoC.
- Interface with RAM– two 64-bit DDR3/DDR4-2133 channels with ECC support
- Cache – 4 MB (L2) + 8 MB (L3).
- Eight-core Mali-T628 graphics coprocessor.
- Video path providing support for HDMI, LVDS
- Hardware video decoding
- Built-in controller PCI Express supports 16 PCIe G en lanes. 3.
- Two 10-Gigabit controllers Ethernet networks, two gigabit Ethernet controllers. The controllers support virtual VLANs and traffic prioritization.
- Two SATA controller 6G, providing data transfer speeds of up to 6 Gbit/s each.
- 2 USB v.3.0 channels and 4 USB v.2.0 channels.
- Support for trusted boot mode.
- Hardware accelerators supporting GOST 28147-89, GOST R 34.11-2012.
- Energy consumption – no more than 30 W.
— End of quote —
What do you say, friends? Did Russian processors impress you or leave you indifferent? Personally, I believe in the great future of Russian digital technologies!
Did you read to the very end?
Was this article helpful?
Not really
What exactly did you not like? Was the article incomplete or false?
Write in comments and we promise to improve!
The production of microcircuits is a very difficult matter, and the closedness of this market is dictated primarily by the features of the dominant photolithography technology today. Microscopic electronic circuits are projected onto a silicon wafer through photomasks, the cost of each of which can reach $200,000. Meanwhile, at least 50 such masks are required to make one chip. Add to this the cost of “trial and error” when developing new models, and you will understand that only very large companies can produce processors in very large quantities.
So what to do scientific laboratories and high-tech startups that need non-standard designs? What should we do for the military, for whom purchasing processors from a “probable enemy” is, to put it mildly, not comme il faut?
We visited the Russian production site of the Dutch company Mapper, thanks to which the production of microcircuits can cease to be the lot of celestials and turn into an activity for mere mortals. Well, or almost simple. Here, on the territory of the Moscow Technopolis, with the financial support of the Rusnano Corporation, a key component of the Mapper technology is produced - the electron-optical system.
However, before understanding the nuances of Mapper maskless lithography, it is worth remembering the basics of conventional photolithography.
Clumsy Light
On a modern processor Intel Core i7 can accommodate about 2 billion transistors (depending on the model), each of which is 14 nm in size. In pursuit of computing power, manufacturers annually reduce the size of transistors and increase their number. The likely technological limit in this race can be considered 5 nm: at such distances quantum effects begin to appear, due to which electrons in neighboring cells can behave unpredictably.
To deposit microscopic semiconductor structures on a silicon wafer, they use a process similar to using a photographic enlarger. Unless his goal is the opposite - to make the image as small as possible. The plate (or protective film) is covered with photoresist - a polymer photosensitive material that changes its properties when irradiated with light. The required chip pattern is exposed to a photoresist through a mask and a collecting lens. The printed wafers are typically four times smaller than the masks.
Substances such as silicon or germanium have four electrons in their outer energy level. They form beautiful crystals that look like metal. But, unlike metal, they do not conduct electricity: All their electrons are involved in powerful covalent bonds and cannot move. However, everything changes if you add to them a little donor impurity from a substance with five electrons in the outer level (phosphorus or arsenic). Four electrons bond with the silicon, leaving one free. Silicon with a donor impurity (n-type) is a good conductor. If you add an acceptor impurity from a substance with three electrons at the outer level (boron, indium) to silicon, “holes” are formed in a similar way, a virtual analogue positive charge. In this case, we are talking about a p-type semiconductor. By connecting p- and n-type conductors, we get a diode - semiconductor device, passing current in only one direction. p-n-p combination or n-p-n gives us a transistor - current flows through it only if a certain voltage is applied to the central conductor.
The diffraction of light makes its own adjustments to this process: the beam, passing through the holes of the mask, is slightly refracted, and instead of one point, a series of concentric circles are exposed, as if from a stone thrown into a pool. Fortunately, diffraction is inversely related to wavelength, which is what engineers take advantage of by using ultraviolet light with a wavelength of 195 nm. Why not even less? It’s just that the shorter wave will not be refracted by the collecting lens, the rays will pass through without focusing. It is also impossible to increase the collecting ability of the lens - spherical aberration will not allow it: each ray will pass through the optical axis at its own point, disrupting focusing.
The maximum contour width that can be imaged using photolithography is 70 nm. Chips with more high resolution They are printed in several stages: 70-nanometer contours are applied, the circuit is etched, and then the next part is exposed through a new mask.
Currently in development is deep ultraviolet photolithography technology, using light with an extreme wavelength of about 13.5 nm. The technology involves the use of vacuum and multilayer mirrors with reflection based on interlayer interference. The mask will also not be a translucent, but a reflective element. Mirrors are free from the phenomenon of refraction, so they can work with light of any wavelength. But for now this is just a concept that may be used in the future.
How processors are made today
A perfectly polished round silicon wafer with a diameter of 30 cm is coated with a thin layer of photoresist. Centrifugal force helps distribute the photoresist evenly.
The future circuit is exposed to a photoresist through a mask. This process is repeated many times because many chips are produced from one wafer.
The part of the photoresist that has been exposed to ultraviolet radiation becomes soluble and can be easily removed using chemicals.
Areas of the silicon wafer that are not protected by photoresist are chemically etched. In their place, depressions form.
A layer of photoresist is again applied to the wafer. This time, exposure exposes those areas that will be subject to ion bombardment.
Under the influence of an electric field, impurity ions accelerate to speeds of more than 300,000 km/h and penetrate the silicon, giving it the properties of a semiconductor.
After removing the remaining photoresist, finished transistors remain on the wafer. A layer of dielectric is applied on top, in which the holes for the contacts are etched using the same technology.
The plate is placed in a copper sulfate solution and a conductive layer is applied to it using electrolysis. Then the entire layer is removed by grinding, but the contacts in the holes remain.
The contacts are connected by a multi-story network of metal “wires.” The number of “floors” can reach 20, and the overall wiring diagram is called the processor architecture.
Only now the plate is cut into many individual chips. Each “crystal” is tested and only then installed on a board with contacts and covered with a silver radiator cap.
13,000 TVs
An alternative to photolithography is electrolithography, when exposure is made not by light, but by electrons, and not by photo-resist, but by electroresist. The electron beam is easily focused to a point of minimal size, down to 1 nm. The technology is similar to a cathode ray tube on a television: a focused stream of electrons is deflected by control coils, painting an image on a silicon wafer.
Until recently, this technology could not compete with the traditional method due to its low speed. For an electroresist to react to irradiation, it must accept a certain amount of electrons per unit area, so one beam can expose at best 1 cm2/h. This is acceptable for single orders from laboratories, but is not applicable in industry.
Unfortunately, it is impossible to solve the problem by increasing the beam energy: like charges repel each other, so as the current increases, the electron beam becomes wider. But you can increase the number of rays by exposing several zones at the same time. And if several are 13,000, as in Mapper technology, then, according to calculations, it is possible to print ten full-fledged chips per hour.
Of course, combine 13,000 in one device cathode ray tubes it would be impossible. In the case of Mapper, radiation from the source is directed to a collimator lens, which forms a wide parallel beam of electrons. In its path stands an aperture matrix, which turns it into 13,000 individual rays. The beams pass through the blanker matrix - a silicon wafer with 13,000 holes. A deflection electrode is located near each of them. If current is applied to it, the electrons “miss” their hole and one of the 13,000 beams is turned off.
After passing the blankers, the rays are directed to a matrix of deflectors, each of which can deflect its beam a couple of microns to the right or left relative to the movement of the plate (so the Mapper still resembles 13,000 picture tubes). Finally, each beam is further focused by its own microlens and then directed to an electroresist. To date, Mapper technology has been tested at the French microelectronics research institute CEA-Leti and at TSMC, which produces microprocessors for leading market players (including Apple iPhone 6S). Key components of the system, including silicon electronic lenses, are manufactured at the Moscow plant.
Mapper technology promises new prospects not only for research laboratories and small-scale (including military) production, but also for large players. Currently, to test prototypes of new processors, it is necessary to make exactly the same photo masks as for mass production. The ability to prototype circuits relatively quickly promises to not only reduce development costs, but also accelerate progress in the field. Which ultimately benefits the mass consumer of electronics, that is, all of us.
The roots of our digital lifestyle definitely come from semiconductors, which have enabled the creation of complex transistor-based computing chips. They store and process data, which is the basis of modern microprocessors. Semiconductors, which are now made from sand, are a key component of almost any electronic device, from computers to laptops and cell phones. Even cars now cannot do without semiconductors and electronics, since semiconductors control the air conditioning system, the fuel injection process, the ignition, the sunroof, the mirrors and even the steering (BMW Active Steering). Today, almost any device that consumes energy is built on semiconductors.
Microprocessors are without a doubt among the most complex semiconductor products, with the number of transistors soon to reach one billion and the range of functionality already astonishing today. Dual core ones coming out soon Core processors 2 on Intel's almost finished 45 nm process technology, and they will already contain 410 million transistors (although most of them will be used for the 6 MB L2 cache). The 45nm process is named for the size of a single transistor, which is now about 1,000 times smaller than the diameter of a human hair. To a certain extent, this is why electronics begins to control everything in our lives: even when the transistor sizes were larger, it was very cheap to produce not very complex microcircuits, the budget for transistors was very large.
In our article we will look at the basics of microprocessor manufacturing, but we will also touch on the history of processors, architecture and look at different products on the market. You can find a lot of interesting information on the Internet, some of which are listed below.
- Wikipedia: Microprocessor. This article discusses different types processors and provides links to manufacturers and additional Wiki pages dedicated to processors.
- Wikipedia: Microprocessors (Category). See the section on microprocessors for even more links and information.
PC Competitors: AMD and Intel
The headquarters of Advanced Micro Devices Inc., founded in 1969, is located in Sunnyvale, California, and the "heart" Intel, which was formed just a year earlier, is located a few kilometers away, in the city of Santa Clara. AMD today has two factories: in Austin (Texas, USA) and in Dresden (Germany). The new plant will come into operation soon. In addition, AMD has joined forces with IBM in processor technology development and manufacturing. Of course, this is all a fraction of Intel's size, as the market leader now operates nearly 20 factories in nine locations. About half of them are used to produce microprocessors. So when you compare AMD and Intel, remember that you are comparing David and Goliath.
Intel has an undeniable advantage in the form of huge production capacity. Yes, the company today is a leader in the implementation of advanced technological processes. Intel is about a year ahead of AMD in this regard. As a result, Intel can use more transistors and more cache in its processors. AMD, unlike Intel, has to optimize its technical process as efficiently as possible in order to keep up with its competitors and produce decent processors. Of course, the design of processors and their architecture are very different, but the technical manufacturing process is built on the same basic principles. Although, of course, there are many differences in it.
Microprocessor manufacturing
The production of microprocessors consists of two important stages. The first is the production of the substrate, which AMD and Intel carry out in their factories. This includes imparting conductive properties to the substrate. The second stage is substrate testing, assembly and packaging of the processor. The latter operation is usually performed in less expensive countries. If you look at Intel processors, you will find an inscription that the packaging was carried out in Costa Rica, Malaysia, the Philippines, etc.
AMD and Intel today are trying to release products for maximum number market segments, moreover, based on the minimum possible range of crystals. A great example is the Intel Core 2 Duo processor line. There are three processors here with code names for different markets: Merom for mobile applications, Conroe - desktop version, Woodcrest - server version. All three processors are built on the same technological basis, which allows the manufacturer to make decisions at the final stages of production. You can enable or disable features, and the current level of clock speeds gives Intel an excellent percentage of usable crystals. If the market demand for mobile processors, Intel can focus on releasing Socket 479 models. If demand for desktop models increases, the company will test, validate and package chips for Socket 775, while server processors are packaged for Socket 771. Even quad-core processors are created this way: two dual-core chips installed in one package, so we get four cores.
How chips are created
Chip production involves depositing thin layers with complex “patterns” onto silicon substrates. First, an insulating layer is created that acts as an electrical gate. Photoresist material is then applied on top, and unwanted areas are removed using masks and high-intensity irradiation. When the irradiated areas are removed, areas of silicon dioxide underneath will be exposed, which is removed by etching. After this, the photoresist material is also removed, and we obtain a certain structure on the silicon surface. Additional photolithography processes are then carried out, with different materials, until the desired three-dimensional structure is obtained. Each layer can be doped with a specific substance or ions, changing the electrical properties. Windows are created in each layer so that metal connections can then be made.
As for the production of substrates, they must be cut from a single cylinder monocrystal into thin “pancakes” so that they can then be easily cut into individual processor chips. At every step of production, complex testing is performed to assess quality. Electrical probes are used to test each chip on the substrate. Finally, the substrate is cut into individual cores, and non-working cores are immediately eliminated. Depending on the characteristics, the core becomes one or another processor and is packaged in a package that makes it easier to install the processor on motherboard. All function blocks undergo intensive stress tests.
It all starts with the substrates
The first step in manufacturing processors is done in a clean room. By the way, it is important to note that such high-tech production represents an accumulation of enormous capital per square meter. The construction of a modern plant with all the equipment easily costs 2-3 billion dollars, and test runs of new technologies require several months. Only then can the plant mass produce processors.
In general, the chip manufacturing process consists of several wafer processing steps. This includes the creation of the substrates themselves, which will eventually be cut into individual crystals.
It all starts with growing a single crystal, for which a seed crystal is embedded in a bath of molten silicon, which is located just above the melting point of polycrystalline silicon. It is important that the crystals grow slowly (about a day) to ensure that the atoms are arranged correctly. Polycrystalline or amorphous silicon consists of many different crystals, which will lead to the appearance of undesirable surface structures with poor electrical properties. Once the silicon is molten, it can be doped with other substances that change its electrical properties. The entire process takes place in a sealed room with a special air composition so that the silicon does not oxidize.
The single crystal is cut into “pancakes” using a diamond hole saw, which is very accurate and does not create large irregularities on the surface of the substrate. Of course, the surface of the substrates is still not perfectly flat, so additional operations are required.
First, using rotating steel plates and an abrasive material (such as aluminum oxide), a thick layer is removed from the substrates (a process called lapping). As a result, irregularities ranging in size from 0.05 mm to approximately 0.002 mm (2,000 nm) are eliminated. Then you should round the edges of each backing, since sharp edges can cause layers to peel off. Next, an etching process is used, when using various chemicals (hydrofluoric acid, acetic acid, nitric acid) the surface is smoothed by about 50 microns. The surface is not physically degraded since the entire process is completely chemical. It allows you to remove remaining errors in the crystal structure, resulting in a surface that is close to ideal.
The last step is polishing, which smoothes the surface to a maximum roughness of 3 nm. Polishing is carried out using a mixture of sodium hydroxide and granular silica.
Today, microprocessor wafers are 200mm or 300mm in diameter, allowing chip makers to produce multiple processors from each one. Next step There will be 450mm substrates, but you shouldn't expect them before 2013. In general, the larger the diameter of the substrate, the more chips of the same size can be produced. A 300mm wafer, for example, produces more than twice as many processors as a 200mm wafer.
We have already mentioned doping, which is performed during the growth of a single crystal. But doping is done both with the finished substrate and later during photolithography processes. This allows you to change the electrical properties of certain areas and layers, and not the entire crystal structure
The addition of the dopant can occur through diffusion. Atoms of the dopant fill the free space inside the crystal lattice, between the silicon structures. In some cases, it is possible to alloy the existing structure. Diffusion is carried out using gases (nitrogen and argon) or using solids or other sources of alloying substance.
Another approach to doping is ion implantation, which is very useful in changing the properties of the substrate that has been doped, since ion implantation is carried out at normal temperatures. Therefore, existing impurities do not diffuse. You can apply a mask to the substrate, which allows you to process only certain areas. Of course, we can talk about ion implantation for a long time and discuss the depth of penetration, activation of the additive at high temperatures, channel effects, penetration into oxide levels, etc., but this is beyond the scope of our article. The procedure can be repeated several times during production.
To create parcels integrated circuit, the photolithography process is used. Since it is not necessary to irradiate the entire surface of the substrate, it is important to use so-called masks that transmit high-intensity radiation only to certain areas. Masks can be compared to black and white negatives. Integrated circuits have many layers (20 or more), and each of them requires its own mask.
A structure of thin chrome film is applied to the surface of a quartz glass plate to create a pattern. In this case, expensive instruments using an electron beam or a laser write the necessary integrated circuit data, resulting in a chromium pattern on the surface of a quartz substrate. It is important to understand that each modification of an integrated circuit leads to the need to produce new masks, so the entire process of making changes is very expensive. For very complex circuits masks take a long time to create.
Using photolithography, a structure is formed on a silicon substrate. The process is repeated several times until many layers (more than 20) are created. Layers can consist of different materials Moreover, you also need to think through connections with microscopic wires. All layers can be alloyed.
Before the photolithography process begins, the substrate is cleaned and heated to remove sticky particles and water. The substrate is then coated with silicon dioxide using a special device. Next, a coupling agent is applied to the substrate, which ensures that the photoresist material that will be applied in the next step remains on the substrate. Photoresist material is applied to the middle of the substrate, which then begins to rotate at high speed so that the layer is evenly distributed over the entire surface of the substrate. The substrate is then heated again.
Then, through the mask, the cover is irradiated with a quantum laser, hard ultraviolet radiation, x-rays, beams of electrons or ions - all of these light or energy sources can be used. Electron beams are used mainly to create masks, X-rays and ion beams are used for research purposes, and industrial production today is dominated by hard UV radiation and gas lasers.
Hard UV radiation with a wavelength of 13.5 nm irradiates the photoresist material as it passes through the mask.
Projection time and focus are very important to achieve the desired result. Poor focusing will result in excess particles of photoresist material remaining because some of the holes in the mask will not be irradiated properly. The same thing will happen if the projection time is too short. Then the structure of photoresist material will be too wide, the areas under the holes will be underexposed. On the other hand, excessive projection time creates too large areas under the holes and too narrow a structure of photoresist material. As a rule, it is very labor-intensive and difficult to adjust and optimize the process. Unsuccessful adjustment will lead to serious deviations in the connecting conductors.
A special step-by-step projection installation moves the substrate to the desired position. Then a line or one section can be projected, most often corresponding to one processor chip. Additional micro-installations may introduce additional changes. They can debug existing technology and optimize the technical process. Micro installations usually work on areas smaller than 1 square meter. mm, while conventional installations cover larger areas.
The substrate then moves to a new stage where the weakened photoresist material is removed, allowing access to the silicon dioxide. There are wet and dry etching processes that treat areas of silicon dioxide. Wet processes use chemical compounds, while dry processes use gas. A separate process involves removing residual photoresist material. Manufacturers often combine wet and dry removal to ensure that the photoresist material is completely removed. This is important because the photoresist material is organic and if not removed can cause defects on the substrate. After etching and cleaning, you can begin to inspect the substrate, which usually happens at each important stage, or transfer the substrate to a new photolithography cycle.
Substrate testing, assembly, packaging
Finished substrates are tested in so-called probe testing installations. They work with the entire substrate. Probe contacts are applied to the contacts of each crystal, allowing electrical tests to be carried out. The software tests all functions of each core.
By cutting, individual kernels can be obtained from the substrate. On this moment Probe control installations have already identified which crystals contain errors, so after cutting they can be separated from the good ones. Previously, damaged crystals were physically marked, but now there is no need for this, all information is stored in a single database.
Crystal mount
The functional core must then be bonded to the processor package using adhesive material.
Then you need to make wire connections connecting the contacts or legs of the package and the crystal itself. Gold, aluminum or copper connections can be used.
Most modern processors use plastic packaging with a heat spreader.
Typically the core is encased in ceramic or plastic to prevent damage. Modern processors are equipped with a so-called heat spreader, which provides additional protection for the crystal, as well as a large contact surface with the cooler.
CPU testing
The last stage involves testing the processor, which occurs at elevated temperatures, in accordance with the processor specifications. The processor is automatically installed in the test socket, after which all necessary functions are analyzed.
Loading...
