dkrc Chapter 181 Extraordinary
Chapter 181 Extraordinary
Thursday, December 7th.
On the internet, a subtle change is taking place.
Zhihu.
The article questioning the 300mm ambition of the Vilan Microsystem, which has garnered over 800,000 views, has received 37 new comments in the past 24 hours.
Most of them remained the same as before—some supported the doubters' views, while others spoke up for Su Chen, arguing back and forth.
But one comment was different from the others.
Posted by user "Semiconductor Observation Room" at 4:17 PM on December 6th:
"Please note the change in the agenda for the Vilan Microsystems Technology Sharing Session on December 5th. What was originally planned as one presentation session has been changed to two. The title of the second presentation is 'Latest Achievements Release.' This is the only piece of information; please use your own judgment."
This comment only has seven likes and is on the second page of the comments section. But if you look closely, you'll find that more than one similar comment has appeared in the past twelve hours.
In a Weibo discussion forum about MEMS technology, someone posted a screenshot—the technical sharing conference agenda page from Vilan Microsystems' official website. The screenshot highlighted the second presentation session in red, accompanied by the caption: "Guess what this 'latest achievement' is?"
In the semiconductor forum on Baidu Tieba, a user with the ID "Shenzhen Chip Worker" posted a thread titled "Anyone who attended the Vilan Technology Sharing Session on December 5th? Please let me know." The post received two replies:
"I signed a confidentiality agreement, so I can't say."
"I can only say it's more exaggerated than you imagine."
These scattered pieces of information are like undercurrents beneath the surface of water, invisible to ordinary people, but those within the industry have begun to take notice.
10:00 AM on December 7th.
The user on Zhihu, who goes by the name "Calm Analyst," has updated a long article.
The title is: "Some speculations about the Vilan Microsystems technical sharing session on December 5th".
The article is short, but every sentence is very restrained:
First of all, I want to clarify that I did not attend the technical sharing session on December 5th. The following content is based entirely on speculation from publicly available information.
First, the official website of Vilan Microsystems updated the technical sharing conference agenda on the afternoon of December 3. The originally planned one-report session was split into two, with the second report titled 'Latest Achievements Release'.
Second, the agenda was updated on the afternoon of December 3, which coincides closely with the time of the previously circulated news that 'Su Chen's team is conducting a 300mm trial etching'.
Third, if the 300mm trial etching fails, there is no reason to temporarily modify the agenda to add a 'results presentation' segment.
Fourth, the 23 participating alliance companies signed confidentiality agreements. If the content to be released is unimportant, a confidentiality agreement is not required.
Based on the above four points, my conjecture is that the 300mm trial etching has been successful and the result is important enough that the pace of information dissemination needs to be controlled through a confidentiality agreement.
If this speculation is true, then all previous articles questioning the Vilan Microsystems 300mm plan—including the long article with 80 views—can be rendered invalid.
You can question an experiment. You cannot question a theoretical system.
If 12.07 degrees at 300mm is true, it's not an engineering result, it's a verification of a theoretical prediction.
Theoretical predictions can be extrapolated.
Within an hour of its publication, the article received over 400 likes and quickly rose to third place on Zhihu's trending semiconductor topic list.
The comments section exploded.
What does 12.07 degrees mean?
"Wait, 300mm in one go? How is that possible?"
"They've signed a confidentiality agreement, so it must be true, right?"
"If it could be extended to 400mm..."
Of course, there are also dissenting voices:
"It's all speculation; there's no concrete evidence."
"Confidentiality agreements can also be a means of covering up failures."
"Let's wait for the paper; everything is speculation until it's published."
But the wind has changed.
The comments section of that skeptical article, which garnered 800,000 views, began to shift. More than half of the new comments were variations of the same phrase:
"Let's wait for news on December 5th."
December 7th, 11:00 AM.
Lin Wei's office.
Sitting in front of her were two people—Chen Yuqing, Marketing Director of Vilan Microsystems, and Zhao Guoping, Head of Supply Chain.
"First, the good news," Lin Wei said, flipping through her notebook, "As of 10:00 AM today, six new companies have submitted applications to join the alliance. Combined with the existing twenty-three, the alliance will expand to twenty-nine members."
Chen Yuqing nodded: "Three of them are packaging companies, which greatly helps our supply chain."
"Mr. Zhao, can you do the math?" Lin Wei looked at Zhao Guoping, "If all three packaging companies join in, plus our existing production capacity, how much of the impact of MicroPort Sensors' supply reduction can be offset?"
Zhao Guoping had clearly already calculated this; he directly stated the figures: "MicroPort Sensors will reduce its supply by one million units in December. The three new companies can provide a combined packaging capacity of 5.5 million units. This not only completely offsets the reduction but also leaves a surplus of 4.5 million units."
"Four and a half million units," Lin Wei repeated, "which means that even if MicroPort Sensors continues to reduce production by half a million units per month as planned, our production capacity shortfall will reach—"
"It won't appear until April next year," Zhao Guoping said. "And by then, Hongyuan Precision's equipment should already be in place. If CR Micro's yield rate can increase to over 95% as planned, the shortfall in April can be covered."
Lin Wei drew a timeline in her notebook.
December: MicroPort reduced supply by 100 million units → Three packaging companies made up for it with 550 million units → Net increase of 450 million units.
January: MicroPort reduced supply by 150 million units → Packaging companies stabilized capacity → Net increase of 400 million units.
February: MicroPort reduced supply by 200 million units → Hongyuan Precision Equipment began installation → Net increase of 350 million units.
March: MicroPort reduced supply by 250 million units → CR MicroPort's yield rate is improving → Net increase of 300 million units.
April: MicroPort reduced supply by 300 million units → Hongyuan equipment went into production + CR Microelectronics increased supply by 95% → the shortage was covered.
May: Self-built packaging line goes into production (Nanshan District, 8000 million RMB investment) → Completely eliminates the need for minimally invasive sensing.
She looked at the timeline, a slight smile playing on her lips.
"The supply reduction plan from Minimally Invasive Sensing was originally our biggest threat," she said. "But once the 300mm chip was developed, three packaging companies proactively joined the alliance, which effectively resolved this threat for us ahead of time."
She closed her notebook and looked at Chen Yuqing: "Have any media outlets contacted us recently about the content of the sharing session?"
Chen Yuqing nodded: "Yes. Three tech media outlets and two financial media outlets. I'll follow your instructions and give a uniform reply: 'No interviews during the confidentiality agreement period.'"
"Keep it up," Lin Wei said. "Before the paper is published, all information related to 300mm will only circulate within the alliance. If anyone outside asks, the unified response will be: 'The technical sharing session is a regular event; the content is not publicly disclosed.'"
"clear."
"But," Lin Wei changed the subject, "have you noticed the shift in public opinion?"
Chen Yuqing took out her phone and opened Zhihu: "'Calm Analyst' posted a speculative article today, which has already made it onto the semiconductor topic trending list. His analytical logic is very clear. Although there was no leak of any internal information, the four lines of reasoning combined are basically correct."
Lin Wei glanced at her phone screen and said, "Ignore him. His speculation is based on reasonable deduction from publicly available information; we haven't leaked any information. Besides—"
She paused for a moment.
"Moreover, this speculation is advantageous to us. Without revealing specific data, it helps us generate subtle public buzz. When the paper is officially published, the market will be more receptive."
Chen Yuqing understood: "So we don't need to respond or delete the post; let it ferment naturally."
"right."
December 7th, 2 PM.
An office building in Beijing.
In He Zhiqiang's office, the curtains were drawn, and only the light from the computer screen illuminated his face.
He's browsing Zhihu.
He read the article by "The Calm Analyst" twice from beginning to end.
Then he opened another window, which was the technical sharing session page of the Vilan Microsystems official website. The agenda had indeed been changed—the original one-report session had become two, and the second one was indeed titled "Latest Achievements Release".
He leaned back in his chair, his fingers tapping unconsciously on the table.
The MEMS gyroscope project of TianShu Chip was originally scheduled to be taped out in May 2021. The technology route chosen is the traditional SOI-DRIE process, 250mm wafer, with a target accuracy of ±0.04 degrees.
This metric was competitive six months ago—Veland Microsystems publicly stated that its 250mm accuracy was ±0.025 degrees, while Tianshu's ±0.04 degrees, although slightly worse, was still within an acceptable range and had a lower cost.
But if Su Chen really did achieve 300mm, and with a precision even higher than 250mm...
He Zhiqiang closed his eyes and did some mental calculations.
Since its inception, TianShu Chip has invested 12 million yuan in its MEMS project. If the tape-out is successful in May 2021, an additional 8 million yuan is expected to be needed for mass production verification, bringing the total investment to 20 million yuan.
问题是：如果苏辰的300mm在明年上半年发论文，市场对250mm产品的兴趣会大幅下降。天枢的±0.04度在300mm的±0.018度面前，完全没有竞争力。
He opened his eyes and opened an Excel spreadsheet on his computer—the 2021 budget for the TianShu chip project.
He didn't make any changes. But he added a line to the notes section of the MEMS project:
"Pay attention to the publication date of Wei Lan's paper. If it's published in Q1, reassess the project's feasibility."
Then he saved the file and turned off the computer.
December 7th, 9 PM.
Tokyo, Japan, a district in Honjo.
Akira Ishikawa sat in his study, a new email on his laptop in front of him.
From: Su Chen.
He clicked on it.
"Professor Akira Ishikawa, thank you for your letter. I would be happy to communicate with you after our paper is published. Su Chen."
Akira Ishikawa read it twice.
Two sentences. There were no unnecessary formalities, no academic pleasantries like "I've long admired your name."
They did not refuse.
"After the paper is published."
Akira Ishikawa repeated the sentence softly.
He understood two meanings.
First layer: Su Chen is writing a thesis, and the content of the thesis involves things he doesn't want to discuss in advance.
The second point: He is willing to communicate after the paper is published. This is not perfunctory, but a clear commitment to a specific timeframe.
Akira Ishikawa closed his laptop and pulled a journal from the bookshelf—the 9th issue of the 2020 edition of the *Journal of Microelectromechanical Systems*. Su Chen's 250mm paper was published in this issue.
He turned to the last page of the paper and looked at the "Acknowledgements" section. Su Chen thanked three people: his supervisor, Professor Zhou Zhiyuan; Ms. Lin Wei, CEO of Vilan Microsystems; and "all colleagues in the laboratory".
No thanks were given to any external scholars or companies.
Ishikawa Akira put the magazine back on the shelf, then picked up his phone and opened his contacts. Su Chen's number was still in his favorites, in the same spot he'd put it in yesterday.
He did not make the call.
"Let's wait for the paper," he said to himself, his tone calm, but his eyes shining.
Friday, December 8th.
Tsinghua University, Zhou Zhiyuan's office.
The first revised draft of the paper lay spread out on the table, covered with Su Chen's annotations. Zhou Zhiyuan spent an entire morning revising all seventeen of the points.
He leaned back in his chair, staring at the ceiling.
"Don't cite too many Bosch references."
"Don't use the word 'breakthrough'."
"I'll write Part Six."
Zhou Zhiyuan's lips curled up slightly. Su Chen three months ago wouldn't have said such things. Su Chen three months ago would have cited all the relevant literature, used the most accurate rather than the most strategic wording, and written every part of the paper himself.
But Su Chen did not achieve 12.07 degrees three months ago.
He picked up his phone and sent a message to Su Chen:
"The first part is revised, please take a look. I'll start writing the second part today, and I expect to send it to you on Sunday. At this pace, we should have a final draft in two weeks."
Su Chen replied quickly: "Okay. Submit to Nature Materials."
Zhou Zhiyuan was stunned for a moment.
Nature Materials.
It is not JMEMS, not Sensors and Actuators, not any professional journal in the field of MEMS.
It is Nature Materials.
Impact factor 47.9. A top journal in the field of materials science, on par with Nature and Science.
Zhou Zhiyuan typed a message: "Are you sure?"
Su Chen: "Confirmed. 300mm is suitable for this journal. And—"
He paused for a few seconds, then continued typing:
"This paper is not merely a report on improvements to MEMS processes. The third-order correction model is a universal theoretical tool applicable to any micro/nano fabrication process involving thermoelastic coupling. This scope extends far beyond the field of MEMS."
Zhou Zhiyuan stared at the screen for a long time.
Then he replied: "Okay. I need to rewrite the theoretical foundation chapter in Part Two. It was originally written in the JMEMS style, but if we're submitting to Nature Materials, I need to explain the universality of the theoretical framework more clearly."
Su Chen: "Yes. That's why I wanted you to write the first five parts. You're better at explaining the theory than I am. I'll write the outlook for the sixth part, because I know how far this model can go."
Zhou Zhiyuan put down his phone and picked up his pen again.
He underlined the title of the second part of the paper and changed the original "Theoretical basis of the equivalent thermoelastic approximation method" to "Third-order nonlinear extension of thermoelastic coupling effect: a general theoretical framework".
Then he started writing.
On the same day, in the afternoon.
Shenzhen, Su Chen's laboratory.
Su Chen sat in front of the computer, and what was displayed on the screen was not a paper, but a computational model.
The parameter extrapolation interface for the third-order modified model.
He entered a number in the "Wafer Diameter" field: 400.
Unit: millimeters.
Click "Run".
Calculation begins. Estimated time: 3 hours and 12 minutes.
He didn't wait at the computer. He stood up, walked to the window, and looked down at the park roads. The December sun shone brightly in Shenzhen, and the leaves of a few banyan trees swayed slightly in the wind.
He didn't know what the calculation result would be for 400mm.
But he knew that if the third-order model was correct—correct at 250mm, correct at 300mm, and more accurate at 300mm than at 250mm—
Then the result of 400mm would only be better.
This is not a guess.
It's physics.
He returned to his computer and opened a new document next to the computational model.
Document title: Part VI: Discussion and Outlook (Draft)
He began to write the first sentence:
The third-order nonlinear extended model proposed in this paper exhibits a counterintuitive property that increases prediction accuracy with increasing processing scale. This property stems from—
He stopped, deleted the words "counterintuitive," and changed them to "extraordinary."
Then continue writing.
The sunlight outside the window slowly moved westward, lengthening the shadows of the trees. The progress bar of the computational model climbed from 12% to 37%.
Su Chen didn't look at the progress bar.
He is writing the sixth part of his thesis.
That was the part that only he could write.
Because only he knew that 12.07 degrees was not the end.
It is the starting point.