How the computer of the future will look like. But the capacity is large⁠ ⁠

A computer of the future. as it will be, imagine?

Over the past few decades, personal computers have significantly increased their performance, but at the same time decreased in dimensions. However, from the technical side they have not undergone significant changes. In modern computers based on Intel Haswell, the same approaches are used as on the “antediluvian” Apple II. But everything can change in the near future. What will be a fundamentally new computer of the future?

The main thing that the principle of interaction between the computer and the person will change will change. Today, all “communication” is mainly through the sensory screen, mouse and keyboard. Computer networking technologies of modern Internet give an instant connection between users, and the Wi-Fi HTTP: // www radio motorist.Conetec.SU/Catalog/Wireless/Wireless-Brige/without a cable provides a high connection speed at a distance of 30 km with a guarantee of stability. That is, communication through those technologies that were developed and implemented a decade ago.

With the tasks assigned to them, they coped with “excellent” and will continue to be used. But among engineers and scientists, the opinion prevails that the future for technologies that today are only in its infancy.

The computer of the future, first of all, will be able to perceive and process the image in real time, that is, the world around him will be perceived by the machine as a person. For example, this will make it possible to create security control systems that by analyzing the environment that surrounds them will be predicted by natural disasters or terrorist attacks.

The main advantage of this system, unlike a human observer, is the lack of fatigue or loss of vigilance. In addition, high.quality “vision” will enable computers to interact with a person, better perceiving their facial expressions and gestures.

Also, future systems will learn to hear better. That is, it is not the quality of sound intake to improve, but the ability to analyze the sound series. Already today, computers are able to perceive speech, and in the future they have to improve this skill. to learn how to distinguish between the finest intonations.

In addition, the computer of the future will be trained to decipher the sounds made by animals. The system improved in this way will be able to explain to parents, the reasons for the crying of a baby. The matter is not limited to communication. based on the analysis of sounds, the computer will be able to warn about the occurrence of breakdown in various equipment.

The appearance of the possibility of transmitting tactile sensations is also expected. This will become the basis for the revolution in online sales. After all, the goods can be considered from all sides, and first “touch”.

It is simply impossible to equip mobile devices with displays with large diagonals, and projection equipment is far from always convenient to use. But you can use VRD (virtual retinal monitors). In this technology, the image will be projected directly on the retina of the eye. In this case, it will seem to the user that the monitor is “suspended” in the air in front of him.

If the image is projected only on one eye, it can be seen simultaneously with the surrounding objects. When projecting on both eyes, realistic and voluminous images will be created. The use of VRD will also reduce the load on the battery of the mobile device.

And the final stage in the development of intenses can be a direct connection between the human brain and electronics. And for this, implanting the chip under the skin will not be mandatory.

About the appearance of a computer of the future, you can only build assumptions. Futurologists are very actively engaged in this. Some dream of chips implanted under the skin, others about the imminent appearance of scraper computer, which will surpass even the best modern desktops in performance. However, as Lem said, “the future looks different than we imagine it.”.

Back in the middle of the twentieth century, few people thought about the mass distribution of the same tablets. Only one thing can be said with confidence. the overall dimensions of the future systems will decrease, and the capabilities and performance will increase significantly.

The introduction of all these achievements, according to experts, will take place gradually. And the main role here will play the power of the human habit. In full the possibility of a new one, only the next generation is always evaluated, and the previous one has only to adapt.

A computer of the future. carbon nanotubes, cubes and a foreseeable future

At the same time, scientists are already looking further. Around the world in various laboratories, experiments are being conducted on the development of fundamentally new computing systems. New systems are based on something fundamentally new, not silicon chips, tested for decades. This is due to the fact that modern computers simply have nowhere to develop further. You can increase frequencies, the number of nuclei, but all these are temporary measures. Only fundamentally new technologies will provide significant increase in performance.

Today we are increasingly talking about quantum computers capable of performing more operations per unit time, which modern systems cannot do. The main element of the quantum computer will be a quantum bit (Cubit). Cubit can accept both standard values ​​- “0” or “1”, and both of them at the same time. According to scientists, this will increase the performance of such systems by millions of times in comparison with traditional.

In addition, development is underway to create fundamentally new computers of the future. A computer of the future. carbon nanotubes, not silicon chips will be based on. The thickness of the nanotubes is several thousand times less than the human hair. It is expected that due to their use, energy consumption and the size of the systems will significantly decrease.

In addition, there are a lot of other no less curious developments, the theoretical description of which looks quite attractive. However, the problem is that these developments are still far from leaving the walls of the laboratories. The commercial use of the same quantum computers, according to the most optimistic forecasts, will begin at least a few decades later. They don’t even dream about home use of such technologies yet.

On the other hand, if you think about whether a computer of the future is needed in general to an ordinary user? Most likely, in the future they will use a different approach. PC will be some thin customers who interact with high.performance remote systems (for example with quantum computers).

Cloud computing today is actively using not only large corporations, but also individuals. In the future, with improving the quality of access to the Internet, the capabilities of the “clouds” will expand. At the same time, high.performance computers, which, for example, will be capable of real.timely processing video truly high resolutions, will only be used to professionals.

And we can say with confidence. the computer of the future will be the way we cannot imagine it now manufacturers will find what to offer!

Molecular computers

Recently, Hewlett-Packard announced the first successes in the manufacture of components, from which powerful molecular computers can be built. Scientists from the HP and the University of California at Los Angeles (UCLA) announced that they managed to force Roatoxan molecules to move from one state to another-in essence, this means the creation of a molecular element of memory.

The next step should be the manufacture of logical keys that can perform the functions and, or not. The entire such computer can consist of a layer of conductors laid in one direction, a layer of Roatoxan molecules and a layer of conductors directed in the opposite direction. The configuration of components consisting of the required number of memory cells and logical keys is created electronically. According to HP scientists, such a computer will be in 100 billion. once more economically than modern microprocessors, occupying many times less space.

The very idea of ​​these logical elements is not revolutionary: silicon chips contain billions of the same. But the advantages in the consumed energy and size are able to make computers to ubiquitous. Molecular computer the size of a grain of sand can contain billions of molecules. And if you learn how to make computers not three.layer, but three.dimensional, overcoming the restrictions on the process of flat lithography used for the manufacture of microprocessors today, the advantages will be even more.

In addition, molecular technologies promise the appearance of micromans that can move and attach an effort. over, to create such devices, even traditional etching technologies can be used. Someday these micromachines will independently assemble components of molecular or atomic size.

The first experiments with molecular devices do not yet guarantee the appearance of such computers, but this is exactly the path that is destined for the entire history of previous achievements. Mass production of the existing molecular computer may well begin somewhere between 2005 and 2015.

Biocomputer

The use of biological materials in the computing technique of biological materials will reduce computers to the size of a living cell over time. So far, this cup of Petri, filled with DNA spirals, or neurons taken by the leeches and connected to electric wires. Essentially, our own cells are nothing more than biomachines of molecular size, and of course, our brain is an example of a biocomputer.

Ihud Shapiro (Ehud Shapiro) from the Weith Institute of Natural Sciences built a plastic model of a biological computer 30 cm high. If this device consisted of real biological molecules, its size would be equal to the size of one of the components of the cell. 0.000025 mm. According to Shapiro, modern achievements in the assembly of molecules allow you to create cellular devices that can be used for biomonitoring.

traditional DNA computers are currently used to decipher the genome of living creatures. DNA samples are used to determine the characteristics of another genetic material: thanks to the rules for mating spirals, you can determine the possible location of four basic amino acids (a, c, t and g).

To give useful information, DNA chains must contain one basic element. This is achieved using a ray of light and mask. To obtain an answer to a particular question related to the genome, it may take up to 80 masks, with which a special chip worth more than 12 thousand is created. Dol. This is where the DMD chip from Texas Instruments came in handy: its microse-grinder, directing light, eliminate the need for masks.

Bill Ditto from the Georgia Technological Institute conducted an interesting experiment, connecting microdators to several leeches neurons. He found that, depending on the input signal, neurons form new relationships. Probably, biological computers consisting of neuro.like elements, unlike silicon devices, will be able to look for the necessary solutions through self.programming. Ditto intends to use the results of his work to create the brain of robots of the future.

What will computers of the near future look like. ⁠ ⁠

The 1st concept had to be done black. This is a witch hat!

The computer of the future will be a microscopic size and will stick like a blister to a person on the temple, sending impulses to the visual and auditory nerve. Input devices will be the eyes and signals that a person sends the motor nerves. There is simply no need to wave your hands, if you can slightly take your eye or think about the team and it will be completed.

The second concept delivered. Phrases torn from context. It can be curled as a roll of paper light and functional as iPhone to fold like iPhone. It does not give rest

About designers-designers-thoughts I will say a few words.These fuckers are not familiar with the technical, they just drew huynyui rejoice and even in their souls do not fuck how to realize all this.

It looks like pictures from some fantastic films, and it can become a reality! I have survived =) I feel when I will be the old one, I will not understand anything either =) and my grandchildren will try to explain to me how it all works =)

the second is chic) ​​and the first I really liked it) and you can attach a holographic console to the busty?

Well they want to replace the whole mouse so? In general, all the concepts from the category “A look for a stem battery was enough for a month and cs: go ch! And all the games were free Shtob there “

Ivan Petrovich Sidorov buys a computer. August 2028. Grotesque. Fairy tale⁠ ⁠

The hero of our history. Ivan Petrovich Sidorov. Master of the steel workshop from Nizhny Tagil. He lives in a small house of 300 kV.m. with eight bedrooms, in a near suburb, on the shore of the Upper Voysk reservoir. He has seven children, five girls and two boys.

At the end of the summer holidays of 2028, the children of Petrovich together began to ask for a new computer. A computer with the Intel Core i7 processor purchased back in 2021, with a working frequency of 4.2 GHz, RAM 32 gigabytes, SSD with a 1 terabyte drive and a NVIDIA RTX3060 video card categorically refused to provide normal parameters for modern games and interactive films.

For example, the favorite game of the older boy “Dagger, Iskander and Vanguard”, based on the events of 2023, did not go above 60fps (60 frames per second) that was completely indecent for the summer of 2028, and the younger girl was very upset when her favorite interactive film “ Pink ponies travel along 3D ”worked only in 2D And most importantly, it was very difficult to divide one computer into all possible users.

The situation has developed, time has come. The old technique has been a thing of the past realizing the need, and reconciled with the inevitable loss of the amount reaching 1000 (at the rate of August 2028. About 80,000 US dollars), Ivan Petrovich went to the electronics store, which is comfortably located in the green park, near the center of Nizhny Tagil, on Serov Street.

Petrovich called a taxi. A small five.meter Toyzhiga car, waited for the car for a couple of minutes, and after twenty minutes was in the store. At the entrance to the store Ivan Petrovich met a robot. Consultant (car dumb and ingenuous). The robot nominated a purchasing chair from his corps (where the respected Ivan Petrovich sat down), and drove along the alleys of the store, simultaneously asking Petrovich various leading questions. In a couple of minutes of travel between the ranks of refrigerators, home robots, televisions and extractors, the robot asked Petrovich with everything that a person needed to know. the seller, and transmitted the received information on 8g of air communications (frequency of 8.2GHz) to the younger organizer of the computers department. The task was non.trivial, and the usual robot. the seller would obviously not cope with her.

In order to choose the computing system to Ivan Petrovich, it was necessary to take into account the whole glad factors:

The number of users. 12, including adults. two children. Seven, two cats, dog. Accordingly, it is necessary to have two ports for 2D monitors, 7 ports for 3D monitors and 3 ports for “animal intense”.

The system should have access to the internship, at a speed of at least 100 Mbit per second.

The system should be integrated with the cloud “School”, with access to all its educational resources.

The system should be able to effective disposal of the generated heat.

The junior organizer of the computers department kindly met Ivan Petrovich, and began to introduce him to the latest achievements of the domestic electronic industry. The world’s largest super.large integral schemes (SBS) became the basis of interesting computational and robotic systems. Much larger both in size and in power than obsolete Elbruses and Baikal. The new processors of Beshtau, Belukha and Vetlan operate at frequencies from 6 to 12 GHz, have an intermediate repository of the first level (cache, if the old one) at least 512 MB, as well as a very beautiful exterior, remarkably resembling ten-kilogram gold bullion. Both in appearance and in size.

The fact is that scientists have made the right decision not to use small and expensive in production technological processes 7 and 14 nanometers for household appliances, and use easily executable and inexpensive products using 90 nanometers for this purpose. So what, that the processors turned out large and energy.intensive. For the free (including the population) electricity, this was not a problem. And to achieve performance ten times superior in the past in the past Intel Core i9 was not at all difficult, combining the fifthly core processors of Beshtau.

Having looked at several catalogs, Ivan Petrovich opted for the computer Ecova (electronically. Computational educational complex) Behtau 5×5, on a stack of five processors of Beshtau, which are produced in large series in the city of Pyatigorsk. Ecova has a sufficient number of external intenses, both for people and animals, and even for robots, and excess heat from the operation of the system is allocated to the external radiator, which is included in the heating system of the house.

Ivan Petrovich paid 1048 27 kopecks with the Card “The whole world”, and went home. And after him, an electronics store went, into which the EVK itself was loaded, 2 large 2D monitors with a resolution of a 20K with a diagonal of 2 meters, 7 children’s 3D monitors with a resolution of 8k with a diagonal of 1 meter, and a set of virtual reality glasses, as well as 3 intenses for animals, with small monitors of 2k.

I will not tell in detail how the system was brought to Petrovich home, how the whole family was upset that such an elegant apparatus, which affects exquisite finishes with natural wood and aluminum, the size of a small refrigerator, had to be put in the basement, because only there it was convenient to connect an ec to the system Heating at home, as all members of the family put monitors, keyboards, rat manipulators and joysticks on the desktop, as all of the 8G air communications, all this was quite quickly. And then the evening came.

Ivan Petrovich, sitting in a comfortable chair, loaded the good old game Fallout 3, and happily noting the excellent quality of the graphics (200 FPS at a resolution of 20k, at the ultras) decided to recall youth, and spend a couple of hours in the universe, which has never been, and ( Luckily). will never. The evening was quiet and calm, and only a light August breeze lazily moved the curtains on open windows.

P.S. Technical specification of the computer EVKS BESTAU 5×5:

Processor: 5 fifthly core processors of Beshtau 10 GHz in the stack;

RAM: bioelectronic, 512 GB, operating frequency 4.2 GHz;

The first level drive: SSD 8 TB “Ural”;

Second.level drive: HDD 256 TB “Ural”;

Graphics module: pole 8060, 256 GB of own video memory, with the ability to use graphics up to 50% of the system’s RAM;

Cube code

Seven qubits on this IBM chip consist of a chain of superconductors, in which the current takes place in both directions. Cubes are combined using a summarized waveguide

Writing a code for a quantum computer is far from usual. The stages of calculations are divided into quantum valves that simulate classical operators and, or not. For this, the cubes receive the input state and then in several stages are combined with each other.

There are still only a few algorithms optimized for quantum computers. For example, for a shore algorithm to decompose a natural number for simple factors, a classic computer needs extremely a lot of time, even if it involves thousands of processors. The quantum 100-cubic computer, according to the head of the Google laboratory, John Martinis, will cope with such calculations very quickly. A group of scientists, led by a professor of physics, has already built a 49-cubic quantum computer for Google.

The figure shows the process of programming the quantum chip IBM. Green points control the state of the cube, the blue symbols determine its connection with other cubes

However, such a quantum computer, according to Martinis, should contain a certain number of additional cubes to correct errors. The fact is that the basic units of quantum information are extremely susceptible to external influences, which leads to a distortion of the result. hence the need arises for a thorough verification. Therefore, the issue of the scalability of a quantum computer remains open: it is possible that the error correction system will require so many computing resources that the development of large quantum computers will not make sense. such a pessimistic forecast.

But if the scaling succeeds, quantum computers will be able to solve all problems faster in the future, which are reduced to the right choice of one probability of so many. Today we use supercomputers, which, however

Self.optimized code

Fifty years ago, the idea to reproduce the mental processes occurring in the human brain pushed scientists to create neural networks. Artificial neural networks process complex entered data, such as images or texts, more effective than any, even the most intelligent algorithms.

Instead of processing the entered these line behind a line or pixel behind a pixel, neural networks are divided, for example, an image into small fragments so that neurons can analyze them in parallel. Neurons are essentially small programs that form layers. They identify typical signs of a fragment, applying a filter on it, and pass them on to the next layer of neurons.

Neural networks are trained, correcting errors

Convolution Neural Networks is a real breakthrough in depth learning: they identify objects in the pictures as true as a person. A decisive role in identifying signs is played by the interaction of two layers of neurons. bundles and sample (Convolution and Pooling Layers). In the learning process, an error signal is reversed.

Convolution. Neurons use various filters to identify and highlight the characteristic features of the image fragment.

Sample. All extra data is ignored, the result is sent with several sample layers, which makes it possible to increase the size of the analyzed fragment.

Classification. All identified signs are collected: for each option, the neuron indicates its probability.

Error correction. The reverse passage of the error signal allows neurons to adjust weights: all weights are adjusted in proportion to the participation of the neuron in the recognition of the object.

The transmission signal is enhanced or weakened depending on the weight of the connection from the neuron to the neuron. To determine the weight of the connection from the neuron, the network is trained based on millions of images. For successful training after each direct passage along all layers of the network, the signal is transmitted in the opposite direction, taking into account the error in order to adjust the weight of the connections.

The next layer of neurons receives data from the neurons of the previous layer, analyzes a larger fragment and transfers the results to the third layer. And so, a layer by a layer, a network of neurons works out a fragment until the end of the analysis of the whole image. At the end, special neurons include drawings to any class and draw conclusions that it depicts.

Computers of the near future

and impressive developments come to the world of computer technology, striking more and more perfect and almost fantastic modifications, and the concepts of personal computers were destined to endure a huge number of modifications.

Following the rule that “the technique should be not only functional, but also beautiful” designers create futuristic conceptual projects that radically change our ideas about how a new generation computer should look like.

The following computer new products have not yet reached mass production, but who knows, maybe a year later they will become the object of lust with a million users around the world:

The designer-designer Yakub Wiring offered the world a curious futuristic concept of a portable computer, which allows users to control the device wherever there is a flat surface that will act as a “display”. The desktop computer will only consist of a system unit the size of a mouse. But it is equipped with cameras and a small.sized projector capable of displaying a holographic image of high quality on any convenient surface, including glass.

Rolltop is a laptop with a flexible 17-inch display, which can turn off as a paper roll, providing the user with maximum convenience when carrying. The laptop is equipped with a folding OLED display that supports sensory control with multi.line function, which makes it easy and functional as iPhone. If desired, the device can turn into a 13-inch tablet computer. In addition to sensory, a full.fledged button keyboard is built into the gadget, which is very useful for working with a set of texts. In folded form, this compact laptop can be easily worn as a bag, the belt of which is thrown over the shoulder.

The next generation concept from Hewlett-Packard called Lim (Lessis-which means “less means more”) is a small system unit resembling an empty photo album and a 19-inch absolutely transparent touch display that looks like a piece of glass. All this complements the touch panel that plays the role of a computer mouse, and a thin wireless keyboard.

This fantastic computer was developed by the Korean designer Won-Seok Lee. The membrane keyboard appears at the base of the device only at the time when it is necessary. With its help, you can enter the text, but the monitor in this gadget is not provided. Instead, the image can be projected onto any surface using a rotating projector.

This stationary computer is equipped with a flexible OLED screen. Users will no longer have to worry about the size of the monitor, because the display of this concept can be stretched to the required value.

Perhaps this novelty will be the dream of any gamer. Designer Kyle Cherry proposed the original concept of a gaming laptop with three folding OLED displays. The author proposes to use in addition to the main 10-inch OLED display (parties of the parties 16:10) two more Aux OLED panels. In a folded state, Prime looks like an ordinary 13-inch laptop. In the open gamer will receive an excellent large.format display with a diagonal of 26 inches (parties ratio 32:10). It is assumed that the laptop body will be made entirely of aluminum, and the computer itself will be equipped with a powerful cooling system.

A designer named Ewen Hallman came up with a computer concept, which will become a very useful tool in conference rooms. Its design resembles to keep for napkins, in which there are several sensory displays united into a single network. The digital handle will allow users to draw directly on the touch screen.

Sony Nextep is a futuristic device that is worn on the wrist of the hand like a bracelet. The concept has a flexible OLED touch screen, a holographic projector, a retractable keyboard panel and access to social networks. The developers make every effort so that the wrist mini-computer becomes available to buyers in 2020. The wrist concept of designer Hiromi Kiriki has become the most anticipated novelty from the Middle Kingdom.

This futuristic modular computer system consists of a thin, light portable main module with a display, the main components and a removable control module. All modules perform various functions depending on the needs of the user. Another part of the system is a wire that can be connected to the main module and use various interchangeable connectors located at its other end.

Computers of the future: how they will look? (photo text)

and impressive developments come to the world of computer technology, striking more and more perfect and almost fantastic modifications, and the concepts of personal computers were destined to endure a huge number of modifications.

Following the rule that “the technique should be not only functional, but also beautiful” designers create futuristic conceptual projects that radically change our ideas about how a new generation computer should look like.

The following computer new products have not yet reached mass production, but who knows, maybe a year later they will become the object of lust with a million users around the world:

The designer-designer Yakub Wiring offered the world a curious futuristic concept of a portable computer, which allows users to control the device wherever there is a flat surface that will act as a “display”. The desktop computer will only consist of a system unit the size of a mouse. But it is equipped with cameras and a small.sized projector capable of displaying a holographic image of high quality on any convenient surface, including glass.

Rolltop is a laptop with a flexible 17-inch display, which can turn off as a paper roll, providing the user with maximum convenience when carrying. The laptop is equipped with a folding OLED display that supports sensory control with multi.line function, which makes it easy and functional as iPhone. If desired, the device can turn into a 13-inch tablet computer. In addition to sensory, a full.fledged button keyboard is built into the gadget, which is very useful for working with a set of texts. In folded form, this compact laptop can be easily worn as a bag, the belt of which is thrown over the shoulder.

The next generation concept from Hewlett-Packard called Lim (Lessis-which means “less means more”) is a small system unit resembling an empty photo album and a 19-inch absolutely transparent touch display that looks like a piece of glass. All this complements the touch panel that plays the role of a computer mouse, and a thin wireless keyboard.

This fantastic computer was developed by the Korean designer Won-Seok Lee. The membrane keyboard appears at the base of the device only at the time when it is necessary. With its help, you can enter the text, but the monitor in this gadget is not provided. Instead, the image can be projected onto any surface using a rotating projector.

This stationary computer is equipped with a flexible OLED screen. Users will no longer have to worry about the size of the monitor, because the display of this concept can be stretched to the required value.

Perhaps this novelty will be the dream of any gamer. Designer Kyle Cherry proposed the original concept of a gaming laptop with three folding OLED displays. The author proposes to use in addition to the main 10-inch OLED display (parties of the parties 16:10) two more Aux OLED panels. In a folded state, Prime looks like an ordinary 13-inch laptop. In the open gamer will receive an excellent large.format display with a diagonal of 26 inches (parties ratio 32:10). It is assumed that the laptop body will be made entirely of aluminum, and the computer itself will be equipped with a powerful cooling system.

A designer named Ewen Hallman came up with a computer concept, which will become a very useful tool in conference rooms. Its design resembles to keep for napkins, in which there are several sensory displays united into a single network. The digital handle will allow users to draw directly on the touch screen.

Sony Nextep is a futuristic device that is worn on the wrist of the hand like a bracelet. The concept has a flexible OLED touch screen, a holographic projector, a retractable keyboard panel and access to social networks. The developers make every effort so that the wrist mini-computer becomes available to buyers in 2020. The wrist concept of designer Hiromi Kiriki has become the most anticipated novelty from the Middle Kingdom.

Koob modular computer system

This futuristic modular computer system consists of a thin, light portable main module with a display, the main components and a removable control module. All modules perform various functions depending on the needs of the user. Another part of the system is a wire that can be connected to the main module and use various interchangeable connectors located at its other end.

Found GT 730 2 GB GDDR5 memory for 7000, at the current moment a good price (for a new video card). Full namerine Gigabyte 2 GB (GV-N730D5-2GL-OEM).

“OMA Delivery. Only a map in a sealed package.”The package opens (and is also easily closed).

The left upper and lower right screws are screwed not in the center, with a shift. There are scratches on the lower right screw. Probably, they have already been unscrewed (they removed cooling and warmed the graphic chip?)

Traces of screws are visible, it has already been screwed and judging by the footsteps more than once.

The affliction on the board indicates heating during work.

But this melting bothers more strongly.

There are small traces of use on the contacts.

Damage to a consistent flash memory (AH17212 25Q21BT E76102), similar to external, mechanical damage (perhaps and will not affect work in the future).

The view from the other side is nothing unusual:

The number of transistors in GPU-Z is not known, alarms. The remaining indicators seem to correspond to the expected.

Furmark, when resolving the screen 1280×720, showed 18 FPS and a temperature of 67 degrees (dough duration 5 minutes). During the check, there were no artifacts on the screen. On YouTube, I saw Furmark for such a card 15 FPS and 63 degrees. In principle, the result is normal.

Of course, there is a warranty for 6 months, but this is still unknown how easy it will be to use it.

Now you know what to expect from a video card on ozone.

P.S.: I will probably not return, I will leave. Make your bet how long this video card will last with a minimum-medium load?

DDR SDRAM on Sata ⁠ ⁠

I walked in search of a scheme for the PC motherboard, and came across the name “Gigabyte Gc-Ramdisk-Box-RH Rev 1.one”. Thinking for a second about the word “Ramdisk”, I remembered the assemblies of the LiveCd for which a similar norm. But the scheme is for this why? I walked, looked at the picture, and there it

So, the map in the PCI port with 4 DDR SDRAM slots (DDR200, DDR266 and T.D.), battery and SATA port. Why and most importantly why? I took my interest and I downloaded the scheme to understand what that payment is doing. The first crazy assumption that came to my mind without a scheme. This is a PCI adapter on DDR SDRAM like “SSD”. That’s just so that the data is not lost when the power is loss to shove the battery and added SATA port to send SATA commands. True, it is very brilliant to shove the memory with a minimum speed of 1600 MB/s through an intense at a maximum speed of 266 MB/s. Probably there is SATA II, or even Sata III. It will be faster, and there is a port for something. We must watch the scheme.

The scheme was ruthless, and immediately showed this

That is, PCI is there only for power. And the data is apparently transmitted through SATA. Stop, why 1.5 gb/s? We leaf through below.

That all this fee. In fact, adapter with DDR SDRAM on SATA I. It generally has three chips. One on BGA for memory. I don’t know which one, Datashit does not google. SATA Bridge, moreover, brake. And Charker to control the charging of the battery.

How I neighing about it hysterically. According to the scheme, it is clear that this miracle entered the market after April 26, 2007. Disks at that time catch up to 70 MB/s. Of course, random reading is much slower and closer to the center, the speed of work is reduced. But taking into account the fact that SATA encodes data on 8/10B we get a maximum of 120 MB/s. It is very interesting for how much it was flooded. Since the board is more expensive, the second disc and Raid 0 could not only cost cheaper, but also represented a much larger volume with a higher linear speed in some cases.

In general, an amazing invention. If I suddenly see in a pile of garbage from someone, I will fix it and turn it into a collection item.

High.speed computing nuclei for computers of the future

For generation of chips size 7 nm, nano.enable transistors will be used, in which the shutter completely wrapps a fin. On the right. a prototype developed at the IMEC research center

Reducing the size of transistors will develop new form factors of home PC in the future. How they might look, I recently showed AMD, spilling light on the design of its new processor with heterogeneous architecture of the exaflopse class (EHP) with high computing power.

The EHP chip includes all the main components of PC: central and graphic processors, memory. Many central processors today, along with the computing cores of the CPU, contain a graphic core. it is enough to decode videos, but it is weak for games. In the framework of the EHP, it is implied by the allocation of a graphic core of a sufficient amount of space on the crystal to carry out high.performance calculations.

The components will be associated with the help of interposer. silicon substrate with thousands of lines etched in it, so that data exchange occurs several times faster. Near the graphic nuclei is a memory of the High Bandwidth Memory type.

High.performance PC the size of a palm. AMD introduced the EHP processor model. All components are placed on Silicon substrate and connected using thousands of lines in the interpose

The advantage of HBM over the DDR RAM modules is the location of the cells with stacks, which also accelerates data transfer (due to a wide tire) and data storage density.

Due to high heating, placing memory directly in the processor today is not possible today-the microcircuit will simply overheat. For this concept, transistors of an even smaller size with low energy consumption and high performance are required.

Smart hardware

EHP concept can be adapted to PC, but it was not created specifically for them. Tumors to a technical breakthrough do not come from servers, PC or smartphones. What will be the computer of the future, determine two other factors: artificial intelligence (AI) and the Internet of things, that is, the combination of sensors and devices into a single network.

Self.learning systems

The nodes (neurons) of the neural network are located in layers, each node performs simple calculations. As a result, the results of different rows of neurons are verified and the network gives out a probabilistic result. The figure shows a neural network recognizing images.

The term “artificial intelligence”. unsuccessful, modern AI has little in common with what is usually meant by the word “intellect”. AI, who, like a person, analyzes unpredictable situations, will be created soon.

Now, within the framework of the concept of artificial intelligence, the calculation is produced by a neural network that solves specific problems, and before that, training is underway. Some systems continue to study and improve after the completion of the preparation using feedback.

Neural networks consist of many nodes that produce parallel calculations (see. Block on the left). AI developers use video cards of the GeForce GTX 1080-it contains 2560 arithmetic logical blocks (ALU). But this is nothing more than a temporary solution: most of ALU is often not involved, since the neural network does not require parallelization on such a scale.

The Intel Lake Crest chip contains 12 clusters for neurons that will store data in HBM memory stacks (1 Tbit/s throughput). Clusters are connected to other chips through channels with a bandwidth of up to 100 Gb/s

In addition, the neural network must be constantly transferred to memory and receive a large amount of data from it. in this regard, HBM would be quickly than GDDR5X on GTX 1080. In the second half of the year, an accelerator of calculation called Lake Crest, intended for building AI, which was announced by Intel, is expected on the market.

The Lake Crest chip will carry 12 computing nodes, which Intel calls clusters, and each cluster is crushed into several ALU associated with HBM.

Artificial intelligence in the masses

Sony Xperia XZ Premium. a smartphone on the Qualcomm processor with support for AI Snapdragon 835

The hardware optimized for AI is already massively produced: for example, some smartphones operate on the new Qualcomm Snapdragon 835 chip, containing, in addition to CPU and graphic nucleus, a number of special microcircuits, one of which the manufacturer intended to support the technology of the Google Tensorflow machine learning technology.

Using TensorFlow, the speed of operations compared to the CPU rises 25 times, and compared to the graphic nucleus. eight times. Google opened the Tensorflow library for developers about a year ago, and since the beginning of 2017 it has been supported by Windows.

Micro Mote. a computer of B1 cubic millimeter with its own AI, which fits on the edge of the coin

You can use the Iron AI optimized for calculations on the Internet of things. this has already felt the American startup Cubeworks. As part of the Micro Mote project, computers are developed in a cubic millimeter, which can be renewed with a microcamer or motion or temperature sensors or temperature sensors.

Recently, the developers presented a tiny computer working on a processor with AI support, with consumption of only 300 μW. The device can, for example, determine what caused the noise outside the window. the approach of the attacker or the wind.

Raisr: Exception without loss. Google Raisr technology, using neural networks, increases photos so well that it replaces any filter that changes the size of the image

The presence of artificial intelligence in our daily life reinforces not only the optimization of hardware. Some calculations can be continued only with AI. So, in November, Google spoke about a new algorithm designed to increase images resolution.

Raisr technology (“Fast and accurate image in superdrawal”) uses a neural network and knows how to change the size of the image better than any of the filters developed by a person.

Development on the use of neural networks in encoding video already in use. in Google, the development of its own VP10 video code is in full swing. If you convert films using neural networks, the output quality will theoretically increase, since when the video is compressed, the main task is to identify the same color and brightness values ​​in different frames. Complex algorithms can do this, but II could additionally optimize the search for the same values ​​to the conditions of a particular film.

Human weaknesses in a social dilemma

In the game Wolfpack, two AI (red dots) are proposed to catch prey (blue point). “Wolves” can cooperate or hunt alone, but in the second case the probability of success is higher

Can computers lie? During the training of generative competition networks (GAN) they can not only, they should: Gan is a kind of war of two neural networks, where one network tries to deceive another, providing it with images, including generated ones, and the second task is to recognize a fake.

The end of the silicon era. The processors of the future ⁠ ⁠

The era of silicon chips is coming to an end. New processors come out hot, and everyone forgot about the law of Moore. Will the development of electronics stop? What will be the processors of the future? Is there a replacement for silicon?

So far, new technological processes are at the very least appear: in 2026M TSMC promises us 2-nm chips, or maybe a incomprehensible Intel 20a process will appear a little earlier. But you know that this is marketing. We should not wait for a high.quality reduction in the size of the transistor.

And the reason here is in physics. the minimum shutter size of the silicon transistor is 5 nm. Five silicon nanometers are the limit. With a lower meaning, it simply cannot be created by a transistor. it will not work as a switch, the electrons will freely tune through its channel without paying attention to the prohibited zone. In other words, such a transistor will always be turned on. so there will be no calculation magic. And the smaller the marketing technology process, the closer we are to this physical limit, and the more essential the effect of tunneling becomes, interfering with calculations.

Of course, chipmakers are trying to solve this problem with all their might. so, you can reduce other parts of the transistor, or make a shutter of a cunning shape. for example, in the form of a fin, from where the Finfet technology came in which transistors are essentially three.dimensional.

However, all these tricks led to the fact that the density of transistors in the chips has seriously increased, and now in a piece of silicon the size of a nail can be hundreds of billions of tiny switches that actively release heat when working on each other. And heat discharge from this sandwich is a serious problem.

Looking at all these problems caused by silicon at the sunset of life, the question arises-why not change this chemical element to something else? After all, he is hardly one in the Mendeleev table, which has already more than a hundred elements. That’s right. not one. There is such a metal as Germany from which semiconductors can also be made.

over, the first transistors in the late 40s from this metal made. Its is three times higher than electrical conductivity, less voltage-and therefore heat loss-on the P-n transition and less resistance of the open channel-in general, German seems better than silicon in semiconductors, but by the 60th of it almost completely refused it.

There were three reasons at once. Firstly, this metal is much more expensive and is less common than silicon, which is almost 30 % in the earth’s crust. Silicon there is the second after oxygen.

Secondly, Germany has much less thermal power, that is, when heated, it faster loses its characteristics, moreover, there are problems with oxidation.

Well, thirdly, he has worse thermal conductivity-that is, it is more difficult to remove heat than from Silicon Chip. All this led to the fact that the era of Germany computers turned out to be as short as the lamps. then I had to switch to silicon.

The era of pure Germany ended 60 years ago. But we live in the days of advanced alloys and complex chemical compounds. Remember T-1000 from Terminator 2! Is it really impossible to pump Germania so that it becomes better than silicon for the production of semiconductors? It turns out that it is possible, and such a substance is called Germanan. In fact, this is like a graphene, only from Germany. a thin one.day film. Its production is a separate type of art, when a layered pie from graphene and calcium is first made, after which the latter is washed out with water, which in the process gives its hydrogen, making Germanic bonds stronger and allowing to separate the single.layer films of this metal.

As it turned out, such films conduct current ten times better than silicon, and issues with cooling are not so acute here. But, of course, it is still far from commercial production. they have learned to create Germanan only in laboratories, and so far there is not a single finished chip on its base. However, Germanan is not all, there is another compound on the basis of which even managed to create a semiconductor chip.

It is called molybdenum disulfide, or. molybdenite. Now it is mainly used to create various alloys, but it has excellent semiconductor properties that work with such tiny sizes when silicon is oxidized to glass. So, scientists managed to bring the thickness of molybdenitis to 0.65 Nm with the full preservation of semiconductor properties.

Is it true that reading from the screen spoils vision?⁠ ⁠

There is a common opinion that reading from a computer, tablet or smartphone worsens vision and leads to myopia. We decided to check if there are scientific confirmations of this point of view.

(Для ЛЛ: нет никаких доказательств того, что чтение с экранов портит зрение сильнее, чем чтение с бумажных носителей)

О вреде чтения с экрана для глаз предупреждают родительские форумы и сайты многочисленных оптик. Подобные предостережения также встречаются на ресурсах, посвящённых здоровому образу жизни. В материалах утверждается, что чтение с компьютера или смартфона из-за светящегося экрана радикально снижает зрение и вызывает близорукость. В качестве основного аргумента в таких статьях приводится интервью 2013 года британского офтальмолога Дэвида Алламби, в котором он заявил, что с 1997 года количество пациентов, страдающих от близорукости и обратившихся в его клинику, увеличилось на 35%. Этот феномен он назвал специальным термином «экранная близорукость» и связал его с использованием смартфонов.

First, try to figure out how it differs from the point of view of physics reading from paper and electronic media. As is known from school lessons, the subject becomes visible when it reflects or emits elementary particles of light. photons that fall on the photosensitive cells of the retina, and from them a signal along a chain of neurons reaches the brain. Photons from the sun or artificial lighting source fall on the book sheet, black letters absorb the black letters, and white gaps reflect them directly on the retina. From the point of view of physics, it would be more correct to say not so much “I see letters”, but rather “I do not see the letter, but I see the intervals between them”. In the case of an electronic bearer, the reflected light is not required for us, the built.in illumination of the screen itself emits the required number of photons so that we can perceive the text or image.

With insufficient lighting, the human eye has the opportunity to adapt. When we try to consider something at dusk, our pupil expands so that more light falls on the retina. When it returns brighter lighting, the pupil is narrowed. If the body’s forces were not achieved by sufficient brightness, we use external adjustment capabilities: we adjust the lighting for the needs of our vision (we turn on a brighter light, we transplant closer to the light source), and in the case of an electronic media we adjust the backlight power. It is important to note that the fear that reading in insufficient lighting is harmful to vision is absolutely groundless. On the mark of the analogy of the New York ophthalmologist Richard Rosen, “it is like saying that photography in poor lighting will damage your camera”.

The largest study of the influence of reading from screens on vision, perhaps, can be called the work of scientists from the University of Ohio. In 1989, they selected 4512 children aged 6 to 13 years of different ethnic groups without signs of myopia and observed their vision for 21 years. When developing a research design among potential risk factors, scientists identified the time spent on the TV screen, and later a computer. The study showed that this factor did not eventually play a significant role in the development of myopia, unlike, for example, such an unobvious parameter at first glance, such as the time of games in the fresh air. Karl’s backdrop, the head of this study, emphasizes that “despite the fact that the time spent by the screen was considered an important factor in the development of myopia for almost 100 years, our large and ethnically representative sample did not show any connection”. Her colleague, Dr. Donald Mutti, agrees with the backdrop: “There is no convincing evidence that work at the computer increases the risk of myopia in adults compared to other forms of work associated with tension of vision”.

However, some relationship between the amount of time spent on reading and the development of myopia exists. Vision of vision due to long.term work for the monitor is caused by the fact that many do not comply with the rules of safe work near, namely, neglect the necessary distance between the text and eyes and do not make the necessary pauses for eye rest. Ophthalmologists call the most important rule “Rule of 20–20–20”: every 20 minutes of work it is necessary to take a break and translate a look at an object located at a distance of 20 feet (about 6 m) for 20 seconds). Observing it, we give the eyes the necessary rest and can continue to work without experiencing unpleasant symptoms and without harm to our vision. The neglect of this rule, most likely, explains the “screen myopia” of the patients of Dr. David Allamby.

It is also interesting to note: in 2019, scientists came to the conclusion that reading white letters from a black background stimulates the paths of information unusual for our eye and is a prevention of myopia, unlike the standard reading of black letters from the white background. It is also worth mentioning the observation of Japanese scientists: chewing chewing gum during tense reading from the screen, using various muscles of the face, reduces such symptoms of eye fatigue as dryness, a feeling of sand in the eyes, doubleness and pain.

Thus, there is no evidence that reading from the screen of a computer, tablet or smartphone more spoils vision than reading from paper. However, it is important to remember that regardless of whether the book is in front of you or a mobile phone, you should follow some rules of safe work and, possibly, following the advice of Japanese scientists, relax the muscles of the face, in parallel with reading chewing chewing gum.