Shulou(Shulou.com)06/03 Report--

DeepTech Shenzhen Science and Technology in conjunction with Foxconn Science and Technology Group and China Science and Technology Chuangxing held the DeepTech 2018 Semiconductor Industry trend Forum-"one mind as one, stone into crystal" Beijing Summit on June 2, 2018, during which a number of heavyweight leaders and elites of the science and technology industry will be invited to deliver keynote speeches and hold round-table discussions on the future development of China's science and technology industry.

At the forefront of historical opportunities, Chinese semiconductors need to land on the ground and become the most solid backing for the development of a new generation of science and technology industry at a critical moment. "China can only be hard if it is strong in science and technology, and it cannot be strong if it is tough."everyone says that science knows no national boundaries, but the real core technology has national boundaries." This is the call made by Dr. Miley, founding partner and co-CEO of China Science and Technology Founding Star, who is also the founder of the concept of "hard technology" at this summit.

What is hard technology? Let's see what Dr. Miley said! The following is a transcript of the speech (revised without changing the original meaning according to reading habits):

"Today's theme is the general trend of semiconductor chips. In fact, this ZTE incident has actually sounded a wake-up call for the entire Chinese people, making us disillusioned.

Everyone says that science knows no national boundaries, but the real core technology has national boundaries.

So, next, I would like to share that in fact, hard technology and chips are closely related. In fact, it is through the third information revolution that the United States has become a world science and technology power. After World War II, the United States has become the scientific center of the world, especially the information revolution represented by the integrated circuit revolution, which has really made the United States a global scientific and technological power.

In fact, the United States ranked first in the world in GDP in 1894, but it was not until after World War II that the United States became the world's science and technology center. In other words, the four great inventions of the 20th century were all born in the United States. Semiconductors and computers are a cornerstone of today's entire integrated circuit industry and information industry, while laser is the cornerstone of the entire Internet (optical communication).

We can see in which country the scientific revolution was born and which country can become the tipping point and center of industrial change, while the United States seized the third scientific and technological revolution.

Judging from the development history of Silicon Valley, it should be said that venture capital comes from MIT. The first venture capital company in the world is called ARD (American Research and Development Company), which was founded by Compton, the president of MIT, in 1946 with four universities and the government, as well as some funds.

In the 1970s, these American venture capital slowly developed to the west, and the Sequoia Capital and KPCB we know today all originated from Fairchild Semiconductor.

Well, Sequoia Capital's Tang Valentine was the first sale of Xiantong Semiconductor. After working in Xiantong Semiconductor for a long time, he came out to create his own business. KPCB invested in Intel and Valentine invested in Apple, so the whole Silicon Valley originated from the integrated circuit industry and hard technology.

Sequoia Capital later invested in Cisco and Google, that is to say, you can see that the development of the entire information industry starts from chips to personal computers, and then personal computers promote the software industry. After everyone on the personal computer has one, the Internet and routers that promote the entire optical fiber, the entire optical fiber infrastructure will come out.

After the United States completed its fiber-optic infrastructure from the 1990s to 2000, Internet companies such as Google rose in 1999.

So, historically, it must be the infrastructure of hard technology that drives the innovation of the Internet to the mobile Internet, which also originated from the birth of iPhone in 2007, so hard technology must be the foundation.

Judging from the four entrepreneurial waves in the United States, from integrated circuits to software, from optical fiber communications to the Internet, from smart phones to the mobile Internet, the next step must be an intelligent era, with the infrastructure of the Internet of things and artificial intelligence as the core.

The core of artificial intelligence is big data, now the new big data in addition to BAT, Google, they are using the original Internet big data, the new big data all come from sensors, so this is an opportunity in the future.

The problem facing the United States now is that from the 1940s to 1995, the United States mainly invested in hard technology, but since 1995, with the development of the Internet, because the return of the Internet is really fast, so now the United States also has a lot of capital pouring into model innovation and the Internet (like a lot of capital in Silicon Valley). This has also led to a certain decline in the US economy, because the Internet can only change online, and its contribution to the offline economy as a whole is limited.

A new round of scientific and technological revolution is about to be born, so we analyzed that it was from the third information revolution that integrated circuits began to detonate. after 60 years of development, basically by March 2016, Moore's Law was close to invalid. has not reached the past rapid growth, has entered a bottleneck period.

In fact, every scientific and technological change, basically, the period of rapid growth is about 50-60 years, which is also the famous Combo cycle, and it also stems from the scientific and technological revolution, because of a scientific and technological revolution, for example, from the steam engine detonated in Britain to the electrical revolution detonated in the United States, and then to the information revolution, basically its technological upgrading cycle is about 50 or 60 years, and from a core point at the same time. The cycle of spreading from one country to the whole world is also 50 or 60 years.

When it spreads to the world, hard technology will become high-tech and traditional technology, and eventually there will be overcapacity, such as China's overcapacity in steel and cement in the past few years. This is that some of the technologies of the past have lost their technological dividends in the end.

At this time, we need a new wave of scientific and technological dividends and a new wave of scientific and technological revolution, so we think that the new wave of revolution is an intelligent era, which must be brought about by optical chips, new chips, including some new technologies in life sciences.

Therefore, I have made an analogy to the four industrial revolutions. in fact, all our products are made up of four categories: light, machinery, electricity and calculation. the first industrial revolution is the mechanization revolution, the second industrial revolution is the electrification revolution, the third industrial revolution is from separate devices to integrated circuits, and the fourth industrial revolution is the revolution of light and computing.

We can see that the third revolution, from integrated circuits to software, and then from optical fiber communications to the Internet, has reached the extreme of mechanical and electrical development in the past 200 years, and basically the potential of this field has been tapped to the extreme. The next step is that optical technology, like the electrified era at that time, is evolving from separate devices to integrated optical paths, which is a new change.

At the same time, we know that the infrastructure of artificial intelligence actually comes from optics, just as the infrastructure of the Internet comes from optics, so I have been doing optical communication for 17 years.

I have been communicating with people in our optical communication circle, saying that without the efforts of optical communication people, there would be no Jack Ma, Ma Huateng, Google people today, because all the traffic is built on the optical fiber communication network, including 5G, the core is actually optical fiber communication.

Now optical fiber communication is also faced with a bottleneck, that is, the cost of optical fiber communication is falling, and there is no way to catch up with the increase in traffic. The information superhighway we built in the past is now in serious traffic jams, and now the bottleneck is the problem of the information superhighway, so now we can see, what are Google, Facebook and Elon Musk doing? They are all laying out optical fiber communication networks.

Google is working on Google Fiber, Elon Musk is studying the Internet, so this is the bottleneck they face. Zuckerberg is now everywhere trying to build some hot air balloons and drones to lay the Internet, which are hardware bottlenecks.

Therefore, the integrated optical path must be the infrastructure in the era of artificial intelligence. First of all, the driving effect of 5GMagne5G on optical communication is more than 4 times that of the original, and now it is from 100G optical module to 400G optical module. The whole drive is very huge.

In addition, we talk about artificial intelligence, now all the companies of artificial intelligence, which two companies did they promote in the end?

Shunyu Optics, is to do iPhone and other mobile phone cameras. Haekangwei TV has also made a large number of cameras.

Because if you can't see the face clearly, there is no way to do artificial intelligence recognition, so the algorithm must be based on the optical infrastructure.

Based on this principle, we can see that integrated circuits have grown at a high speed for such a long time from the first large-scale integrated circuit in 1967 to the birth of Moore's Law in 1968. Now the integrated optical path is similar to that era of integrated circuits. Now a single integrated optical path can integrate 2000 opto-mechatronics.

Therefore, if the integrated optical path is applied on a large scale in the future, it will certainly be able to achieve such a revolution as integrated circuits.

After Moore's Law expired in March 2016, I thought that the third scientific and technological revolution was driven by Moore's Law, and the fourth revolution must have a new law, so I summed up a "Mi 70" law. that is, in the future, optics must account for 70% of all hardware costs, because the 21st century is the century of optics, and the cost of optics will become higher and higher.

Like Moore's Law, this is actually an empirical formula, because now in the field of communications, 70% of every dollar we pay for Internet access is handed over to optical companies, and optical communication is the first large-scale application field of optics. the future will be the same in all fields of information, because our future is the information age, big data era, the acquisition, transmission, calculation, storage and display of information. In the past, our optics only did information transmission. this time, half of the chips banned by ZTE are optical chips, and all are optical chips and optical device companies.

Therefore, the core of the future artificial intelligence era is the acquisition of information and data. We look at self-driving companies. Today, the most expensive thing is lidar, that is, we must do information acquisition, binocular vision, and even fiber optic gyroscopes. There are a lot of optical sensors and optical chips for information acquisition.

The transmission of information is now 70%. For the calculation of information, IBM and Intel, including the relevant research institutes of the Chinese Academy of Sciences, are developing optical CPU to replace traditional integrated circuits with optical computing. Traditional integrated circuits have now become multi-core because a single core can no longer be handled.

There is also the storage of information. Now a large number of optical transceiver modules have been used in the cloud computing center. Google is now the largest purchaser of optical devices in the world, and the display of information itself is optical, so the optical cost of AR and VR display technology in the future will be very high.

I think iPhone started the era of mobile Internet in 2007, and Face ID launched by iPhoneX in 2017 verified this theory. Now it has begun to increase the cost of optics in mobile phones. Three VCSEL lasers are used in Face ID, two of which are VCSEL laser arrays, each with hundreds of VCSEL lasers.

Therefore, we can see that in order to obtain face recognition, the artificial intelligence algorithm first uses the optical chip to hit the face, and then uses the optical detector and the original DOE, which begins to increase a lot of optical cost.

I believe that after decades of rapid development of integrated optical path in the future, all future products cannot be called consumer electronics, because the era in which consumer electronics dominates the cost of products has passed. In the next 20 or 30 years, consumer photons (integrated optical path) will account for the main cost. For example, CPU will soon become an optical CPU, and all the sensors in it may be related to optics. In the future, it must be an era of consumer photons. This is also a new IP.

Therefore, I think a new era has come, and it is also a very critical moment for China, because China's economy is now facing a shift from rapid growth to high-quality growth, and our past growth model has begun to have problems. We are going to shift from a demographic dividend to an innovation dividend, from factor-driven and investment-driven to innovation-driven, not to release the physical productivity of hundreds of millions of people in the future. It is to release the mental productivity of tens of millions of researchers, so this is an opportunity for China in the future.

Moreover, China is now facing a problem, that is, core chips are still being imported. Over the past 30 years, our entire development model has been to assemble refrigerators, washing machines, televisions, computers, and mobile phones. Our nature has not changed on a large scale. Our core chips, core materials, and core technology are mainly imported.

Therefore, in the next 30 years, what China must conquer is the core technology represented by the core chip. This is the problem that China has to solve in the future. If we continue to assemble it, if the Americans do not import it to us, we will be finished. This is also the warning brought to us by the ZTE incident.

However, I think the problem facing China now is that on the one hand, we are very short of chips, on the other hand, no one is investing in chips. We saw the report on investment in Qingke in the first quarter of 2018. In Internet finance and consumption, we invested nearly 1000 billion yuan. We invested 135 million in semiconductors.

Although we can't buy so many chips, if we don't get everyone's investment, if we continue this model, in another 10 or 20 years, Trump will still impose an embargo on China, and we will still be very dangerous.

Therefore, I think the essence of the Sino-US trade war is not a trade war, but a hard science and technology war.

The nature of the Opium War was also like this. Because of the trade surplus, the British exported opium to the Qing Dynasty, and the Qing government outflowed a lot of silver. After the Qing government could not stand it, the Qing government sent Lin Zexu to ban smoking. As soon as the smoking ban was banned, Britain immediately launched a strong cannon, so this is the problem.

Therefore, I think there are three points in which the times call for hard technology:

The first point: the turning point of science and technology is now at the turning point from the third scientific and technological revolution to the fourth scientific and technological revolution. Can we seize this opportunity? This is the most important opportunity for China. For the past 50 or 60 years, we have been running behind others, but to tell you the truth, when others are 50, 60, and 100 years ahead, others have already run 9000 meters, and it is very difficult for us to catch up with the 10,000-meter race from the beginning. Any country has seized a new round of scientific and technological revolution in its rise, and whether China can seize the new round of scientific and technological revolution depends on our opportunities. The second point: demographic dividend to innovation dividend. The third point: at the turning point between China and the United States, the United States is actually going downhill. The manufacturing industry in the United States now accounts for only more than 10 percent of the United States, and China now accounts for 30 percent, but 30 percent can't be any lower. This is a red line.

Therefore, we must keep 30% of China's manufacturing industry no lower than this red line. We have ushered in a new round of scientific and technological revolution and industrial change, which is a period of transformation of the mode of development and historical interaction in our country, so this is the time for the concept of hard science and technology to be put forward.

I think hard science and technology is the fourth scientific and technological revolution, and the new technologies that can have a significant impact and change on China's future world economic structure and industrial structure are all hard science and technology.

Therefore, hard science and technology is the core technology for China to keep pace with and take the lead in the world in the future. at the same time, he is not only talking about hard technology, but I think there is also a hard spirit. What China lacks now is the spirit of craftsmen. Can we do one thing in 10 or 20 years? to be number one in the world, what we lack is the spirit of pursuing world champions. this is what we need more, our scientific and technological workers and every entrepreneur. We should strive to be world champions in our own field.

Therefore, only when China is strong in science and technology can China become hard, and it cannot be strong without being tough. When we talk about the eighth Route Army of hard science and technology, we think that many frontier areas of the fourth scientific and technological revolution are what we need to do, and of course they are not limited to this field. As long as they are new revolutionary technologies, we all think that they are hard science and technology.

Therefore, the idea of hard science and technology that we advocate is that we think that in China, China needs hard science and technology thinking more than Internet thinking, and China needs to be down-to-earth to earn slow money. As Jin Yong said in his novel, Guo Jing is very stupid and stupid. He practiced the dragon's eighteen palms and Kang's regrets, and practiced for 10 years, and after 10 years, Yang Kang could not beat him. Although Yang Kang is very smart, he practices a nine-yin white bone claw today and a toad tomorrow, but he is easily obsessed.

Our entrepreneurial and investment circles cannot play one wind a year. We cannot play O2O this year, P2P next year, Bitcoin and speculative coins this year. In that case, China will never be strong, and those who play O2O and P2P are almost dead.

Therefore, the curve of mode innovation is different from that of hard technology. Hard technology is like this blue curve. In front of it, hard technology is rewarded by every penny of hard work. Internet and mode innovation are called hard work and return very much. Money comes quickly, just like O2O and blockchain. Money comes quickly in front, but what the Chinese are best at is to follow other people's way, so that others have no way to go.

We copied the homework of our classmates from small, grew up and copied the companies of our classmates, and the best went to the United States to copy the companies of Stanford and MIT, but this era has passed, and now we have to innovate independently, because there are not many opportunities for copying other companies (opportunities for model innovation), and we can only rely on independent innovation in the future.

Therefore, many hard-tech companies, such as Huawei, are becoming more and more powerful at the back, so we should encourage this. China is also facing a turning point from model innovation to scientific and technological innovation, which is the turning point of the two modes of thinking. this turning point is very difficult, because the thinking of the whole people can not change, if everyone is still like in the past, speculating in real estate, buying mines and making fast money. If no one is willing to engage in scientific and technological innovation, China's hard science and technology industry will never be able to become an innovative power, and we may fall into the Latin American trap like Latin American countries, so this is our most dangerous moment.

We believe that scientific and technological entrepreneurship is the main theme in the next 30 years. When we break through the death valley of scientific and technological innovation, we have to invest where no one is willing to invest, that is, from scientific research to the middle of the industry. Therefore, I think that hard science and technology is a core construction principle of the modern economic system, that is, to allocate modern financial and human resources to scientific and technological innovation and the real economy, so we have become a pilot institute. Provide platforms, equipment, funds, specifically to incubate the chip of the enterprise.

So, this is some of our layout on the chip, including some LED chips of semiconductors, including photonic integrated chips, VCSEL chips, including the establishment of some platforms.

We have been building a rainforest ecosystem for hard science and technology entrepreneurship. In the future, China must cultivate countless rain forest ecology of innovation and entrepreneurship like Silicon Valley, so as to make China's hard science and technology entrepreneurship easier and better than that of the United States. Only then can China become a world scientific and technological power, and can we achieve the great rejuvenation of the Chinese nation, Chinese Dream.

Thank you! "

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.