Moore's Law is about to expire. Although it is a crisis, it is also an opportunity for global reshuffling. The importance of this opportunity is equivalent to the invention of integrated circuits that year.
Yang Lin-Bai Moore's Law will die?
One of Intel’s founders, Gordon Moore, proposed in the early 1960s that the number of transistors on an integrated circuit would double every 18 months and the performance would double. For half a century, this law has witnessed the advancement of human information technology, and it has also bred the chip giant Intel.
However, this law, which has been considered by the IT industry as a criterion, has shown its fundamental bottleneck: the principle of Moore's Law, the chip will no longer be able to develop at the same speed. It is estimated that around 2015 to 2020, the chip upgrade rate will slowly stop.
Zhang Shouyi is one of the major players in the academic community to deal with this issue. As the youngest professor of Stanford, he formed the IBM-Stanford Spintronics Research Center in 2003 and completed a preview of its new concept based on the future of the chip industry in 2006. In 2007, this concept was proved by a German experimental team. In the same year, Zhang’s discovery was rated as one of the top 10 important scientific breakthroughs in the world.
Speaking of Moore's Law is about to fail, its fundamental reason is: If you still work according to the principles of the past, every triode on the chip will generate a certain amount of heat for every calculation of the computer; and every 18 months, the triode needs Double. Once, this is an exponential growth. The exponential growth is terrible. Although a triode emits only a small amount of heat, putting too many triodes will cause the temperature to be too high and the efficiency to be hindered.
In fact, the issue of how to reduce power consumption is imminent. Now the challenge of the industry—or the opportunity of the scientific community—is whether or not an innovative idea can be found, and the industry is working on a completely new principle, so that the triode will not emit so much heat.
Professor Zhang’s solution was to reduce energy consumption through electronic rotation. In simple terms, due to the basic disorder of electron movement in semiconductors, energy consumption occurs due to continuous collisions when current flows from one end to the other. Zhang's team found a law of motion of electrons, that is, the direction of electron rotation and the direction of current actually have certain rules, which makes electrons have a certain tacit understanding in the movement and avoid mutual collision.
However, the technical upgrading is not the most shocking to me. A deep conversation with Prof. Zhang is a reflection of the implementation of the “combination of production, education and research†in this case.
With regard to the impending demise of Moore's Law, the U.S. government has seen this very early and has already had a number of investments. Research institutions have also done a lot of layouts. Among the interesting ones are the input of Intel, AMD and other companies to Zhang Shouyi's project. . This investment does not only give some money. Through this move, companies actually made a choice in direction. But why are these competitive companies appearing on the investment list at the same time?
This is because what Zhang's team is doing is a strategy for intellectual property, which will be shared among several large companies. Although companies have a competitive relationship, this competition is mainly based on how to make products better. In other words, they naturally do not want to see their competitors abandon themselves.
At the same time, this kind of cooperation model has another advantage for enterprises: most of the students who participate in Stanford study will go to the company after graduating. Once this is done, the final result will be in the company.
The phrase "combination of production, education, and research" has been heard in China for many years, but it may not have been performed so deeply. Similar technological innovations at the frontier between the Chinese business community and American colleges and universities also seem to have never happened.
I tossed this doubt to Professor Zhang. His opinion is: This is not because there are restrictions. The main reason is that the company itself is not doing well enough. If the company is busy with a price war all day long, there is no way to have the foresight to consider how to turn itself over.
He also pointed out that although Moore's Law is about to expire, it is a crisis, but it is also an opportunity for global reshuffling. The importance of this opportunity is equivalent to the invention of integrated circuits that year. "Although the U.S. industry is still basically striving for more progress under the original model, it is all united to provide funds for universities and scientific research institutions. I believe that China's semiconductor industry must also have such a sense of advancement and cannot play all day long. Price war, because this is not a very successful model."
Yang Lin-Bai Moore's Law will die?
One of Intel’s founders, Gordon Moore, proposed in the early 1960s that the number of transistors on an integrated circuit would double every 18 months and the performance would double. For half a century, this law has witnessed the advancement of human information technology, and it has also bred the chip giant Intel.
However, this law, which has been considered by the IT industry as a criterion, has shown its fundamental bottleneck: the principle of Moore's Law, the chip will no longer be able to develop at the same speed. It is estimated that around 2015 to 2020, the chip upgrade rate will slowly stop.
Zhang Shouyi is one of the major players in the academic community to deal with this issue. As the youngest professor of Stanford, he formed the IBM-Stanford Spintronics Research Center in 2003 and completed a preview of its new concept based on the future of the chip industry in 2006. In 2007, this concept was proved by a German experimental team. In the same year, Zhang’s discovery was rated as one of the top 10 important scientific breakthroughs in the world.
Speaking of Moore's Law is about to fail, its fundamental reason is: If you still work according to the principles of the past, every triode on the chip will generate a certain amount of heat for every calculation of the computer; and every 18 months, the triode needs Double. Once, this is an exponential growth. The exponential growth is terrible. Although a triode emits only a small amount of heat, putting too many triodes will cause the temperature to be too high and the efficiency to be hindered.
In fact, the issue of how to reduce power consumption is imminent. Now the challenge of the industry—or the opportunity of the scientific community—is whether or not an innovative idea can be found, and the industry is working on a completely new principle, so that the triode will not emit so much heat.
Professor Zhang’s solution was to reduce energy consumption through electronic rotation. In simple terms, due to the basic disorder of electron movement in semiconductors, energy consumption occurs due to continuous collisions when current flows from one end to the other. Zhang's team found a law of motion of electrons, that is, the direction of electron rotation and the direction of current actually have certain rules, which makes electrons have a certain tacit understanding in the movement and avoid mutual collision.
However, the technical upgrading is not the most shocking to me. A deep conversation with Prof. Zhang is a reflection of the implementation of the “combination of production, education and research†in this case.
With regard to the impending demise of Moore's Law, the U.S. government has seen this very early and has already had a number of investments. Research institutions have also done a lot of layouts. Among the interesting ones are the input of Intel, AMD and other companies to Zhang Shouyi's project. . This investment does not only give some money. Through this move, companies actually made a choice in direction. But why are these competitive companies appearing on the investment list at the same time?
This is because what Zhang's team is doing is a strategy for intellectual property, which will be shared among several large companies. Although companies have a competitive relationship, this competition is mainly based on how to make products better. In other words, they naturally do not want to see their competitors abandon themselves.
At the same time, this kind of cooperation model has another advantage for enterprises: most of the students who participate in Stanford study will go to the company after graduating. Once this is done, the final result will be in the company.
The phrase "combination of production, education, and research" has been heard in China for many years, but it may not have been performed so deeply. Similar technological innovations at the frontier between the Chinese business community and American colleges and universities also seem to have never happened.
I tossed this doubt to Professor Zhang. His opinion is: This is not because there are restrictions. The main reason is that the company itself is not doing well enough. If the company is busy with a price war all day long, there is no way to have the foresight to consider how to turn itself over.
He also pointed out that although Moore's Law is about to expire, it is a crisis, but it is also an opportunity for global reshuffling. The importance of this opportunity is equivalent to the invention of integrated circuits that year. "Although the U.S. industry is still basically striving for more progress under the original model, it is all united to provide funds for universities and scientific research institutions. I believe that China's semiconductor industry must also have such a sense of advancement and cannot play all day long. Price war, because this is not a very successful model."
Panel Light,LED Panel Lighting,Square LED Panel Light,Plastic LED Panel Light
LED Bulb Light Bracket Co., Ltd. , http://www.challleds.com