Technology: breaking the hegemony that monopolizes the world

A foreign heavyweight media reported that industry chain sources revealed that TSMC's 3nm process technology has entered the experimental production stage at TSMC's 18th Wafer Factory, and is expected to start large-scale mass production in the fourth quarter of 1022 in the Xiayin Star Calendar.

In the follow-up, TSMC CEO Zhe Xuejia revealed in an interview: "The 3nm process technology will carry out risk trial production this year, and achieve mass production in the second half of next year. Compared with the 5nm process chip, the theoretical density of the 3nm process transistor will be 70%. The operating speed is increased by 15%, and the energy efficiency is increased by 30%.

The evidence of the two news, the light and the dark, has greatly improved the authenticity of the news.

Some media even published "TSMC has taken a key step in the 3nm foundry process, Han Xing is lagging behind again, and Xiaxin Technology will have to struggle for two more years to catch up with the progress. " reported the matter.

In the current bipolar friction, the semiconductor chip manufacturing process has always been the focus of everyone's attention.

In terms of hard standards, 3nm is the next process node after 5nm. Even though Han Xing has reached the 5nm process by "self-naming", Qualcomm said on Han Xing's car that this 5nm is really not as good as TSMC.

Han Xing even shouted a slogan, "To surpass TSMC in 1030 of the Xiayin star calendar."

The result is obvious. When everyone broke the news and made hot spots, TSMC stabbed Han Xing in the back.

Li Zarong's reaction was also very quick, and he immediately arranged for senior executives to come forward, pretending to accidentally reveal that Hanxing's 3nm process will be mass-produced in 1022 of the Xiayin star calendar.

However, this interview appeared to be very feeble in the subsequent series of operations.

Both Pingguo and Intel said that they will become the first customers of TSMC's 3nm process.

Cook also said: "Pingguo is currently TSMC's largest customer and its main source of revenue. The A15 bionic chip on Pingguo 13 is the top of the industry. I believe that after using the 3nm process in the future, Pingguo will achieve greater breakthrough."

Although the current CEO of Intel has been complaining about TSMC before, this time he followed up TSMC's 3nm process uncharacteristically.

From the perspective of the global situation, the old capital led by Qualcomm has now formed a blood transfusion to Han Xing, and the emerging capital represented by Pingguo and Intel has a blood transfusion to TSMC, so as to improve the research and development progress of the two processes and suppress it. Live the development of Xiaxin.

Although Moore's Law is no longer a golden rule in cutting-edge manufacturing processes, it still has traces to follow.

As strong as TSMC, if you want to achieve mass production of the 3nm process, you also need to overcome some difficulties, such as complex chip design, soaring foundry costs, and increased procurement costs for EUV lithography machines.

Although Xiaxin has been secretly upgrading, but its high-level thinking is to wait until the 5nm process can be stabilized before making an announcement. After all, its own 7nm process technology has certain advantages over other people's fake 5nm process technology. Thinking that TSMC and Han Xing are so fierce, and they will soon go to 3nm, the pressure of public opinion came to me all at once.

Of course, there are also many sensible people who know that Xia Xin started late, and the speed of catching up with the manufacturing process is already very fast, but it is impossible to become a big fat man with one bite, so there is no fire.

But is the current Internet public opinion environment rational?

"Can Xiaxin do it? Everyone else has gone to 3nm!" Such remarks still exist.

In this regard, Xiaxin executives also quickly organized a press conference to answer.

As the executive vice president and company secretary of Xiaxin Technology, Gao Shan speaks very well.

"We must accept criticism, because it can help us get out of the illusion of narcissism, so that we will not be self-indulgent in morality and knowledge for a long time, and go to destruction in narcissism."

Coming up is to express your attitude sincerely, and then carry out science popularization.

"Actually, Xiaxin, TSMC, Hanxing and other companies have differences in chip foundry technology, and this difference also brings about how many nm processes do not represent the same level.

TSMC chose the traditional Fin FET transistor architecture. In order to surpass TSMC in the 3nm process, Samsung chose GAA transistor technology very aggressively.

As we all know, Xiaxin started from scratch in the field of high-end manufacturing, so we use the special technology provided by Kyushu Technology.

The special process is currently in a state of confidentiality, so it cannot be announced.

But what I can guarantee is that in terms of technology, the technology of Kyushu Technology is more advanced than the technology used by Hanxing and TSMC. In theory, chips manufactured using this process have stronger performance and lower power consumption.

Of course, what I need to remind everyone is that both we and Hanxing have adopted a new architecture, so there are many technical difficulties to be overcome. The Fin FET transistor architecture adopted by TSMC is relatively more mature in technology and less difficult to develop.

But it is only a little lower. If we want to achieve mass production of the 3nm process, we need to overcome problems such as complex chip design, high foundry costs, and increased procurement costs of lithography machines. "

At the end, he said with a smile on his face: "Yes, after the upgrade of Daxia Semiconductor's entire industry chain, Xiaxin Technology will achieve three major advancements: 7nm process mass production, 5nm process limited production, and 3nm process risk trial production. .”

That is, a few hours after Xiaxin Technology’s press conference ended, Xiawei Yu Chengdong opened his mouth: “Xiawei Mate50 is coming back soon, HiSilicon Kirin will also use Xiaxin’s 5nm process and 3nm process foundry in the future. chip."

And Dong Qi from Suiren Company even said directly: "Jiuzhou Technology provides key technologies for industrial upgrading, such as light sources, foundry equipment and even process design. Jiaolong mobile phones will still use Xiaxin OEM in the future. When the production capacity increases, More powerful and cheaper chips, so stay tuned."

This week, the breaking news is one after another, and the hot news is almost changed every few hours.

Pingguo’s first thought was to make great efforts to increase the sales of Pingguo 13, but never thought that under the chain effect, Xiaxin Technology, Xia Wei and Suiren Company would eventually become peach pickers.

And in the back, these meat-smelling media rushed to the Jiuzhou Science and Technology Park in Rongcheng again. After the application was approved, they squatted in the small towns around the park to guard the employees of Jiuzhou Technology and the executives who occasionally went out.

The Chengdu Municipal Bureau is open to this. After all, these reporters usually take photos of the surrounding environment to show the new look of Chengdu to the outside world.

But it is a pity that the ideas of these reporters are doomed to fail, because the executive team of Kyushu Science and Technology is almost always in the state of nerds because of Boss Gu and relevant departments often talking about safety issues. People go to these super-large parks like Titanium and Tantalum to take a stroll.

The current titanium and tantalum star park can almost be called a small world park.