The meeting was really a great joy, and Mr. D arranged for the newly established General Administration of Computer Affairs to be responsible for communicating with Dr. Yu, and strongly invited Dr. Yu to join in the integrated circuit industry planning.

The establishment of the General Administration of Computer Industry originated from the report of the Ministry of Computer Industry, and in January 79, the Ministry of Computer Industry submitted to the State Council the "Report on the Request for the Establishment of the State Administration of Computer Industry", which was approved in March.

The first task of the establishment of the General Administration was to investigate what problems exist in the computer industry throughout the country, and the results of the investigation were very clear, that is, there were three major and three light -- "heavy hosts, light peripherals, heavy hardware, light software, heavy production, and light applications."

To tell the truth, 74 years later, there are a lot of domestic microcomputers, 050, 060 series one by one, but it's not like that at all, it seems that as long as it is produced, it will be a bird, and nothing else will matter.

And limited to technology, technical reasons, key components can only be imitated, for example, the CPU of the 050 microcomputer produced in 77 is imitation of 8080, which is composed of 31 integrated circuits to achieve the function of 8080, and it was not until 79 that 4 circuits were realized, and as for the single chip, it has to be planed.

A variety of systems are incompatible with each other, and the large number of models has caused the situation that computers are not widely used.

Therefore, in September 1977, the Computer Industry Administration of the Ministry of Electronics held the first microcomputer professional conference, which established the idea of accelerating the development of China's microcomputer industry by making full use of favorable opportunities and all possible conditions according to China's national conditions, making full use of favorable opportunities and all possible conditions, and directly adopting the world's new advanced technology that suits China's needs.

In layman's terms, it is to introduce, that is, to buy technology.

Dr. Yu expressed a little concern about this when they communicated with the leaders of the General Administration, and the starting point was good, and the introduction of advanced technology could be produced.

However, the development of computer technology is very fast, and the introduction cycle is very long, to say less, from the selection to negotiation to the shipment back assembly and commissioning and production will take at least two or three years, and the first product will come out in the year of the donkey, what kind of industrial production to the market?

The leaders of the General Administration were silent, and the long cycle of introduction is indeed known to everyone, and foreigners even deliberately delayed it. What's more, they agree to sell you the system, but they have to pay for it first, and then take your money to develop it themselves for two or three years, and then take the developed products and deliver them, which seems to be what the Japanese often do.

However, the introduction is a reprieve, and if you don't introduce it, you will be shot on the spot.

He didn't know much about this aspect for a long time, he only knew that he must not buy the entire production line from generation to generation, because history has proved that this is emasculating his own research and development capabilities.

Dr. Yu pondered for a moment and expressed his opinion, that is, he and the disadvantages of the domestic environment that he had discussed for a long time - the lack of R&D, market and production.

This is a big problem, generally the task is reached, the research institute began to develop, there is no market research, just keep up with the international trend, completely do not consider whether the domestic can be accepted.

The things in the laboratory cannot be directly produced in the factory, and the things produced can only be used improvised, and the research institute has no market feedback at all, which is a strange circle, and in the final analysis, it is still the system reason.

Take the integrated circuit industry as an example, since the reform and opening up, the domestic factories are lifeless, and the low yield rate is scary (only 30% at the highest), although there are some equipment reasons, but why can Taiwan's yield rate rise to more than 90% like a rocket?

This is the problem of quality control, an integrated circuit to be produced from beginning to end at least thirty or forty processes, even if each process reaches 99% accuracy, how much can be left in the end?

Dr. Yu introduced the experience of foreign chip factories, a production supervisor can work for two years at most, and then it will not work, because when he first came up, he was determined to forge ahead, and worked hard to improve the yield rate in order to exceed the results of his predecessor, and it would slow down after two years, so he had to change.

Moreover, if an employee is unqualified, the production supervisor can let him go immediately, but can the mainland?

The leader was like a mouthbreaker, and immediately exchanged advanced foreign production experience with Dr. Yu.

The two talked with great interest, and for a long time, they were like listening to books from heaven and drowsy.

Finally, it came to an end, and the leaders of the General Administration talked about business.

It turned out that in 79 years, before 0521 came out, the General Administration, together with the National Defense Industry Office and the Commission of Science, Technology and Industry, jointly carried out a selection of military computers.

At that time, the selected result was to focus on the development of "two small and two micro", that is, the minicomputer imitating NOVA and DEC, and the development of microcomputers compatible with Intel and Motorola, and it was clear that the development of microcomputers and minicomputers should be in line with the international mainstream computers, and take the road of international compatibility and openness.

At that time, this result was understandable, because the mainland's computer industry was only a bud, and the total number of large and small computers made in the whole country in a year was only about 500.

However, the long-term CPU has turned out to be a major breakthrough in domestic key components, which needs to be reconsidered. The performance of domestic imitation machines is unstable, the failure rate is high, and almost all of them are handicraft workshop-style production methods, which greatly hinders the application and popularization of computers.

Now it seems that with the 021 processor, at least the microcomputer can survive independently from external imports, as long as there is strong software support, the existing Intel and Motorola systems can be completely abandoned.

And this is exactly what the military requires, the single-chip system is stable and reliable, small in size, and almost perfect except for insufficient performance.

Therefore, the General Administration wants to use 021CPU as the basis to make a general-purpose single board computer.

Who knows that he interjected for a long time and said: "If I have the funds, maybe I can use the 021 processor to make a minicomputer, the cost should be lower than the current minicomputer, and the performance should be higher." ”

This made the leaders of the General Administration overjoyed and asked for details again and again.

Chang then described his idea of using multiple microprocessors in series to work in tandem to provide more computing power and data throughput.

Improving the performance of a computer is nothing more than increasing the speed of the processing unit, or adding the processing unit to integrate multi-channel processing. The former, represented by the Cray-1, reached the limit of speed and had to use freon cooling as a last resort.

The original representative of the latter was the ILLIAC-IV. machine made by the Americans in 71, which was an array machine with 64 processing units under unified control, and its reputation was not obvious.

However, because the more processing units, the larger the volume, the cost doubles, which was a nightmare in the era without microprocessors, so multi-processing unit computers are not too popular.

It wasn't until the invention of the microprocessor that integrated multiprocessors became a cheap and simple technology, and with the support of an operating system, an array could be assembled by almost a single person (a bit of an exaggeration, but it is now).

However, in modern times, the two are almost indistinguishable, and multi-processing technology has become a standard, and if it does not support multi-processing, it is a scum.

That's what Cao Changjiu meant, but his thinking is more radical, and if he wants to plagiarize, he has to copy thoroughly, and he directly put forward the blade-like structure, although he himself doesn't know the mystery very well.

When it comes to blade systems, another big hero has to be introduced, that is, the Chinese-American scientist Chen Shiqing.

Dr. Chen's specialty is supercomputers, and there are two peaks among the Chinese who pioneered the IT boom in the United States, that is, Wang An and Chen Shiqing.

It's just that Wang An went bankrupt and died depressed, while Dr. Chen was immortal, and he didn't taste defeat for decades, and one innovation after another came out one after another.

It's 79 years now, and Dr. Chen should have just joined the cray company to worship Patriarch Cray, and remember this paragraph clearly for a long time, because Dr. Chen is his idol (it seems that he has many ^_^ idols).

At that time, Chen Shiqing was also a well-known figure, studying under Professor David Cook, the chief designer of NASA's first supercomputer.

Clay gave Chen Shiqing two choices: one was to work with him to develop the second-generation product of Cray-1, guaranteeing funds and technology, but everything had to obey him, and the other was for Chen Shiqing to form another team to develop his own products, but there was no guarantee.

Chen Shiqing chose the latter.

The brilliant Dr. Chan used his incredible creativity to win performance, which was parallel computing. At that time, people were still accustomed to working with only one computing engine, but Dr. Chen took the lead in using two vector engines to start together, which greatly improved the computing speed.

As a result, the fastest computer record was broken by the CrayX-MP/2 developed by him, which shocked the industry.

Technical achievements are not enough for long-term worship, virtue is the key.

Supercomputers are an essential element of high-tech development and a country's defense and economic weapons. The importance of supercomputer technology is self-evident and the importance of such extensive use of supercomputer technology as military code-breaking, wind tunnel simulation experiments, the development of various new weapons such as hydrogen bombs, nuclear bombs, missiles, warships, warplanes, and combat vehicles, as well as energy, aviation, and automobiles.

Internationally, a high-tech embargo has been imposed on China, such as the Cray supercomputer, which the United States absolutely prohibits exporting to China.

After Clay's death, Dr. Chen continued to lead the company's team to develop the world's fastest computer, and he has become the company's vice president and chief designer, developing a 128-engine supercomputer.

At this time, he had the idea of returning to China, not to Taiwan, but to the mainland.

In '99, Dr. Chan met a Hong Kong businessman. The Hong Kong entrepreneur followed his father's teachings to "develop national high technology" and hit it off with Dr. Chan, who was raising funds.

Inspired by the design concept of turbine blades of supersonic aircraft engines, Chen Shiqing developed the "Super Blade Computer" and became the world's earliest developer and patent owner of this design concept.

In 2001, after the first generation of super blade computers was successfully developed in the United States, Chen Shiqing decided to return to China. Because it is easier to transfer a technology at the initial stage. Once it matures and becomes the world's most advanced technology, it will be difficult to transfer it, so Dr. Chen decided to let the technology take root, sprout and grow in China.

In 2002, Chen Shiqing established Shenzhen Shell Xingying Technology Co., Ltd. and led the new Huaxia team to continue to develop a new generation of super blade computers.

In 2004, Shenzhen Shell Xingying Technology Co., Ltd. successfully developed and can be industrialized in large quantities, with a theoretical operation speed of up to 100 trillion times per second.

Prior to this, the fastest computer in China was the Dawning 4000A, which was successfully developed in June 2004, and its computing speed was 11 trillion times per second, ranking seventeenth in the world.

At that time, the world's fastest computers were the 92 trillion times of the US Department of Energy, the 61 trillion times of NASA, and the 41 trillion times of the Japanese NEC "Earth Simulator".

Dr. Chen's return has Americans lamenting: "There is no way to transfer technology that has a stronger and more far-reaching impact than when a world-class expert leaves with all his knowledge." ”