MTL - The Science Fiction World of Xueba-Chapter 473 Self-developed architecture

September 15, 1994.

Shenzhen Airport.

Pang Xuelin and Liu Qi stood at the exit of the passenger together, and at a distance they watched Ni Guangnan and an old lady in their 60s walk out with a dozen men and women of different ages.

The old lady was dressed elegantly, although she was more than sixty, with fair skin and exquisite face, but she could vaguely see her youthful appearance.

Pang Xuelin took the initiative to greet him: "President Ni, welcome to the deep city, is this this?"

Pang Xuelin turned his eyes to the old lady beside Ni Guangnan.

Ni Guangnan said with a smile, "President Pang, let me introduce you. This is academician Xia Peisu of the Chinese Academy of Sciences. I heard that you are going to engage in HiSilicon Semiconductor. She came here to see this time."

"Hello, Academician Xia!"

Pang Xuelin and Xia Peisu shook hands and smiled.

In fact, Pang Xuelin vaguely guessed the identity of the other party before Ni Guangnan introduced it.

Xia Peisu, academician of the Chinese Academy of Sciences, the founder of the Chinese computer industry, known as the mother of the Chinese electronic computer, is also a very rare female computer expert in this era.

In 1958, she was responsible for the design and development of 107 general-purpose electronic digital computers. This was China's first self-developed general-purpose electronic digital computer. It proved with facts and actions that the Chinese had the ability and ambition to design and develop their own computers.

She proposed the principle of the maximum time difference pipeline to make the computer greatly increase the operation speed, which greatly shortened the clock cycle of the pipeline computer.

In addition, she has effectively solved the problem of waveform distortion that occurs during high-speed signal transmission in large computers.

This research result has been adopted or referenced by many domestic companies that develop large-scale high-speed computers. She also formulated specific engineering design specifications for the multiple high-performance computers she later designed and developed, and achieved good signal transmission effects.

She was responsible for the successful development of the high-speed array processor 150-AP. The operation speed of the 150 computer was 1 million times / second, and the operation speed of the 150-AP reached 14 million times / second. The calculation speed of similar products embargoed by China has received great attention internationally and has made significant contributions to China's oil exploration.

She presided over the functional distributed computer system and led the team to successfully develop the GF10 series of computers, making this direction the research focus of the Institute for a period of time.

In addition, she also founded the most influential "Chinese Journal of Computers" in China's computer field and "" published abroad; hosted the "English-Chinese Computer Dictionary".

In later generations, Academician Xia Peisu was hailed as the mother of Chinese computer and has a lofty position in the computer field.

"President Pang, hello!"

Xia Peisu's eyes looked at Pang Xuelin with some scrutiny.

She has heard many rumors about this young man, what a talented teenager, the youngest richest man, a young entrepreneur.

But what attracted Xia Peisu most was the paper written by Pang Xuelin on the outline of the development of computers and the Internet.

The paper's elaboration of the entire ICT industry is a dream of breaking through time and space and seeing through the mists of the future.

Because of this, Xia Peisu became very interested in Pang Xuelin.

It's a pity that when she wanted to talk to Pang Xuelin, she learned that Pang Xuelin went to the United States.

This delay has passed nearly half a year.

This time she accidentally learned from Ni Guangnan's mouth that Pang Xuelin will set up HiSilicon in Shencheng and invited Ni Guangnan and his team to settle in. She woke up on a whim and also came with her.

Pang Xuelin and Xia Peisu greeted a few words, then, Ni Guangnan first introduced Pang Xuelin to the team members he brought out from Lenovo.

These people are all talents from the Chinese Academy of Sciences and various universities in the field of microelectronics that Ni Guangnan has tried to find. Most of them are not well-known, but Pang Xuelin did not expect to find several future generations from these people.

After all, this era is regarded as the dark age of the Chinese semiconductor industry. Without intervention, China will have to explore for at least many years. After many setbacks, it will only slowly get on track after 2010.

The two students brought by Xia Peisu attracted Pang Xuelin's attention.

One of them is Hu Weiwu, the father of the later Godson.

This is a cow.

Hu Weiwu was born in a teacher's family in Yongkang, Zhejiang. When he was born, his body was relatively weak, and his father named him "powerful", hoping that his son would be strong, powerful, and heroic.

In September 1986, Hu Weiwu entered the University of Science and Technology of China as the champion of the college entrance examination in Yongkang County, Zhejiang.

The University of Science and Technology of that era still has the momentum above Qingbei.

Of the six or seven hundred students enrolled by HKUST, 13 are provincial champions.

This shows the strength of Hu Weiwu's high school era.

After entering the university, Hu Weiwu showed an amazing talent. In the university era, he welded an 8086 CPU with 74xx pieces.

In July 1991, after graduating from the Chinese University of Science and Technology, Hu Weiwu entered the Institute of Computer Technology of the Chinese Academy of Sciences for a doctorate without a test.

During his Ph.D. study, his tutor was Academician Xia Peisu.

From undergraduate to master's and doctorate studies, Hu Weiwu has always been in the country, and he can be called a true "native school", and he even dubbed himself as "Tu Biao school".

There are many opportunities for him to study abroad. His doctoral thesis has won the "Special Award of the Dean Scholarship of the Chinese Academy of Sciences" and the first "National 100 Excellent Doctoral Thesis" awards. One of the awards is to go abroad with public funds.

But Hu Weiwu still decided to stay in the country and willingly be a "tuchi faction."

Hu Weiwu's level is unquestionable. The two textbooks he wrote, "Computer Architecture Basics" and "Computer Architecture", are classics, and they use the usual methods to explain the computer system. .

Pang Xuelin looked at textbooks written by Hu Weiwu when he learned about computer-related materials in the real world.

It's a pity that fate dictates.

When the Institute of Computers of the Chinese Academy of Sciences decided to develop a general-purpose processor in early 2000, Hu Weiwu chose the MIPS architecture instead of the later ARM architecture.

But this is also an afterthought.

In fact, ARM didn't develop well before 2000. At that time, the strongest core design that ARM company could come up with was just the very original ARM11, no out-of-order execution, no multiple launches, and no cache system.

The original goal of Godson was desktop and supercomputing systems, which required high computing performance and could compete with the X86 system.

In contrast, the MIPS architecture implemented advanced designs such as out-of-order four launches, deep pipelines, branch prediction, register renaming, and load-store speculation in the 1990s. It is a CPU architecture that can compete with the X86 system worldwide. One, once brilliant.

In the 1990s, the popular home game consoles generally used MIPS architecture CPUs, such as the well-known SONY R3000A, later PSP R4000, and Nintendo N64 R4300i, as well as Motorola, Ericsson, Nokia, Cisco, Qualcomm, SUN , And other communication giants have used MIPS chips in switches, routers, and communication equipment.

In the case where the patent barriers that the x86 camp has operated for many years could not be broken, the choices that could be made for Chinese CPUs at that time were nothing more than MIPS, ALHPA, and ARM.

ARM is still embarrassed in its childhood. The MIPS architecture has been used as a MAINCPU for user terminal equipment by a large number of Japanese technology companies due to its excellent computing performance. This has important reference value. Therefore, Hu Weiwu, the father of Godson, chose MIPS.

But no one expected that the previously hot MIPS architecture would eventually fall apart, and Godson became the only user of this architecture.

Loongson started to operate independently early and was responsible for its own profits and losses.

In the nearly two decades of Godson's career after birth, Hu Weiwu's team relied on meager income and the support of the Chinese Academy of Sciences to provide support for the development of our country in the fields of supercomputing, military, and aerospace.

And formed a set of independent chip system.

And the total R & D expenses of these two decades are only less than 2 billion, which is equivalent to a fraction of Intel's annual R & D expenses.

Without the full support of the country and the support of related industries, Hu Weiwu can be said to have achieved the extreme of what he can do.

In the two decades of teaching and research career, Hu Weiwu also brought out many top talents in the field of microelectronics.

The most famous one is Chen Yunji, who later created the artificial intelligence chip Cambrian.

This big man led the Intelligent Processor Research Center of the Chinese Academy of Sciences in later generations and developed the world's first deep learning dedicated processor chip, which is regarded as a pioneer and pioneer in the field of artificial intelligence chips.

"Mr. Pang, what's wrong?"

Xia Peisu saw Pang Xuelin's eyes always staring at his disciples.

Pang Xuelin shook his head and smiled and said, "It's nothing."

Calculated according to time, this year should be the last academic year for Hu Weiwu to study with Xia Peisu. Such a solidly-founded and endlessly powerful man, Pang Xuelin feels sorry for not fooling him to work under his own hands.

Next, Pang Xuelin and Liu Qi took Xia Peisu and Ni Guangnan on a bus. First, they led everyone to settle in a five-star hotel in Shenzhen.

After lunch and after the wind and dust, Pang Xuelin took Ni Guangnan and Xia Peisu to the Shenyi Industrial Building in Nanshan District.

This is also the location of Huawei's headquarters.

Pang Xuelin has already spent 500 million RMB with the ability to use banknotes to completely buy this building, ready to be used as the headquarters of Warwick and Hisilicon in the next few years.

Warwick also saved itself from moving around as the scale grew.

After arriving at the Shenyi Industrial Building, Ren Zhengfei had been waiting for a long time.

Pang Xuelin first asked Ren Zhengfei to bring the people brought by Ni Guangnan to visit. He himself took Xia Peisu and Ni Guangnan to go around the building first, let them familiarize themselves with the next office environment, then, Pang Xuelin took They came to the meeting room of Warwick.

Ni Guangnan knew at this time that Pang Xuelin actually owned 49% of Warwick ’s shares, and could not help but feel a little sighed: “Pang Pang, Warwick ’s current program-controlled switches are selling well in the market, and the market occupied by the program-controlled switches developed by Lenovo It ’s almost eroded by you. I did n’t expect Warwick to be your business. Coupled with the star-ring technology that is struggling with gold, I think that the title of China ’s richest man is probably deserved. ”

Pang Xuelin smiled and said: "Nigong, these are just fake names. To be honest, making money makes no sense to me. What I value more is whether we can convert the money into the scientific research resources we need."

Ni Guangnan nodded and said, "President Pang, then I won't talk to you nonsense. If you want to develop chips for HiSilicon, do you think about the architecture? At present, there are not many on the market that can compete with the X86 architecture. "

Pang Xuelin shook his head with a smile, and said, "Ni Gong, I originally wanted to buy the ARM architecture to do it myself, but the result was unsuccessful. The British refused to sell ARM to us. So thinking about it, I think if we are really To build a chip that really belongs to us, then we must design a set of our own architecture. "

"Our own architecture system?"

Ni Guangnan and Xia Peisu looked at each other and were surprised. "President Pang, this difficulty is not as simple as you think."

The most difficult thing for the new architecture to develop is not the technical level, but the time, money, and ecosystem.

On the one hand, the R & D structure actually needs time to accumulate and a lot of financial support.

Without these two, there is basically no way to talk about the research and development of the new architecture.

But at the same time, the ecosphere is also a top priority.

Desktop processor market of x86, low power consumption market of arm, cheap market of mips.

Even if you develop a new type of instruction set, but others do n’t need it, they do n’t have the money to take it, and they ca n’t continue to develop without the money. This is an endless loop.

Therefore, in a sense, the importance of the ecosystem is far greater than technology. Once the advantages are established, it is difficult to overthrow, just like MAC's high-level, and it does not overturn the windows.

No matter how smooth wp is, it is the same as cannon fodder in front of Android.

With the new instruction set developed, how many applications are willing to optimize for you? How many people can recognize and use it?

It's a problem.

And the instruction set needs the support of the system. UU reading www.uukanshu.com You pirated someone else's Windows system, Linux system, is it useful for your own development?

After all, it is impossible to develop your own instruction set on other people's systems. If you want to develop your own good instruction set, you must first develop a good system and a good CPU.

Therefore, in a sense, the importance of the ecosphere is far greater than technology. Once the advantages are established, it is difficult to overthrow.

With the new instruction set developed, how many applications are willing to optimize for you? How many people can recognize and use it?

It's a problem.

And the instruction set needs the support of the system. Do you pirate someone else's Windows system, Linux system, is it useful for your own research and development?

After all, it is impossible to develop your own instruction set on other people's systems. If you want to develop your own good instruction set, you must first develop a good system and a good CPU.

txt download address:

phone-reading: