novelones Chapter 196 Gathering
Chapter 196 Gathering
February 20, 2021.
Suzhou, Vilan Microsystems.
Zhao Guoping knocked on the door and entered Lin Wei's office.
"Mr. Lin, the production of the first batch of sensors is complete. 150,800 units. The yield rate is 93.6%. Packaging and testing are expected to be completed by February 24th."
"Advance?"
"Two days ahead of schedule. Hongyuan's new equipment is more stable than expected. Xiao Wang's debugging team made some minor adjustments to the process parameters, and the output of the etching station increased by about 8% compared to the initial estimate."
Lin Wei nodded.
When does the pre-shipment inspection take place?
"February 25th. Based on the orders from the eight core companies, I have finalized the allocation plan—40,000 units for Zhenxin Microelectronics, 30,000 units for Huarui Optoelectronics, and 10,000 to 20,000 units each for the other six companies. All will be shipped through the alliance's standard logistics channels."
What about after-sales support?
"Each company will be assigned a technical liaison. Free integration technical support will be provided for the first three months. I have already confirmed the list of liaisons for the eight core companies with the technical department."
"Okay." Lin Wei glanced at Zhao Guoping. "Is there anything else?"
Zhao Guoping hesitated for a moment.
"There's something I need to tell you. Starcraft Packaging officially informed us last week that they've decided not to renew their contract. They'll be leaving the alliance after it expires at the end of March."
"reason?"
"No specific reasons were given. They only mentioned 'strategic adjustment.'"
Lin Wei's expression remained unchanged.
"Follow the established procedures. Just confirm that there are no outstanding issues with the contract terms. There's no need to try to salvage the situation."
"clear."
"What about the other members?"
"Stable. Ruicheng Precision and Haina Microelectronics haven't contacted us since inquiring about the paper's progress. The eight core companies are even proactively asking when the shipment will be delivered—they're more anxious than we are."
Lin Wei's lips curled up slightly.
"It's a good thing that the eight core companies are in a hurry. It means their customers are waiting."
"Yes. Mr. Chen from Zhenxin Microelectronics called last week to say that his downstream customers are already urging us. They need to produce a batch of inertial navigation modules for autonomous driving and urgently require our sensors for prototype verification."
"Then let's make sure there are no problems with the pre-shipment inspection on February 25th. The quality of this first batch is crucial—this is the first official commercial delivery since the alliance was formed. Any quality issues could shake our confidence."
"Don't worry. My quality control team and I will inspect each batch."
"Okay. Go ahead and get busy."
After Zhao Guoping left, Lin Wei turned on her computer.
She crossed out the words "Starry Sky Encapsulation" on the calendar. The alliance went from twenty-nine to twenty-eight members—one less observer member.
This was within her expectations.
The league doesn't need wavering people.
She opened another document—commercial data forecasts for the first shipments.
The shipment consists of 150,000 sensors, with a unit price ranging from 32 to 45 yuan depending on the member's level. The total revenue from the first batch of shipments is expected to be between five and six million yuan.
This number is not large—for a company with monthly operating costs exceeding two million, the revenue from the first shipment is far from enough to cover costs.
But this is not the point.
The important point is that this is the first time.
The first true commercial delivery. The first demonstration that a 300mm/±0.018° sensor could not only be produced on a production line, but also sold.
"Sold out"—these two words are more convincing than any academic paper, any data, or any Zhihu post.
Because customers are willing to spend real money to buy your product—that's the most direct market validation.
Lin Wei closed the document and glanced out the window.
In late February in Suzhou, the weather began to warm up. The workshops in the factory were brightly lit—the production lines were doing the final packaging and testing for the first batch of shipments.
February 25th.
Five days later.
……
February 22nd.
Milan, Italy.
STMicroelectronics headquarters.
Bertoli's monthly MEMS evaluation report was submitted to the department vice president.
The core content of the report is the ongoing tracking of Vilan:
"Weilan's NM paper is still under review (results expected in March or April). No internal review information has been leaked at this time. External evaluation is based on publicly available information."
New developments:
Pratt (University of Illinois) posted a detailed technical analysis of the Villan third-order model on Twitter, which has attracted widespread attention in the academic community.
Vogt (Technical University of Munich) mentioned Villan in the preview of the IEEE MEMS 2021 panel discussion—this indicates that the European MEMS academic community is beginning to pay formal attention.
He Wentao (Fudan University) published an analysis article in Chinese industry media, concluding with 'cautious optimism'.
Legal assessment (Francesca, updated): The conclusion that third-order models, as mathematical theories, are not protected by patent law remains unchanged. The process patent barrier approach is still not feasible.
Technical Assessment (Bertori): Based on Pratt's public analysis and our own internal derivation, the third-order model is theoretically self-consistent. The key issue remains experimental verification—especially the 400mm extrapolation. If Wei Lan's paper is accepted by NM, we need to complete the internal assessment and decide on our response strategy within six months.
Recommendation: Continue to await the peer review results. Simultaneously prepare two contingency plans—Plan A (paper accepted) and Plan B (paper rejected). Plan A focuses on quickly obtaining application authorization for the third-order model or establishing technical cooperation. Plan B focuses on maintaining the existing 200mm route and accelerating process optimization.
After Bertoli submitted the report, the vice president wrote a single word on it: "Noted."
The report was then archived.
In large multinational corporations like STMicroelectronics, this is usually the fate of an assessment report—to be read, annotated, and archived.
But Bertoli knew that the report would be brought up again the day the peer review results came out.
"Noted" won't be enough then.
What will be needed then is "Act".
……
February 23rd.
Su Chen received a message from Zhang Li:
"The last three sets of experiments are complete. Data from sets thirteen through fifteen has been exported. All data packages will be sent to you tonight."
Su Chen was having lunch in the cafeteria when he saw the message. He put down his chopsticks and immediately returned to the lab.
The data packet arrived at 2 PM.
Su Chen spent four hours importing all fifteen sets of data into the analysis program and running a complete statistical analysis.
At 6 p.m., the final results were released.
400mm Edge Validation Experiment – Statistical Results of Fifteen Sets of Complete Data:
S-shaped temperature gradient transition: Group 15/15 confirmed to exist.
T1-T2区域（边缘0-5mm）：偏差均值1.3%，标準差0.28%。95%置信区间[1.15%, 1.45%]。
T2-T3区域（边缘5-10mm）：偏差均值2.5%，标準差0.44%。95%置信区间[2.25%, 2.75%]。
Model comparison R² value: 0.978.
S-shaped turning point deviation: deviation from model prediction < 0.3 mm (mean of 15 groups).
0.978.
It's higher than the 0.973 from group twelve.
The data quality of the last three sets was better—Zhang Li made further optimizations in temperature control, reducing the deviation in the T2-T3 region from 2.6% to 2.2%.
Fifteen sets of data.
The S-shaped reversal is 100% confirmed.
R²=0.978.
Su Chen stared at the screen for a full ten minutes.
Then he started writing the data analysis report.
He wrote for three hours.
The report is divided into five parts: experimental methods, data presentation, statistical analysis, model comparison, and conclusions.
In the conclusion, he wrote:
The data from fifteen sets of edge validation experiments consistently show that:
The third-order model's prediction of the temperature gradient distribution in the edge region of a 300mm wafer is in high agreement with experimental data (R²=0.978).
The S-shaped temperature gradient transition is a characteristic feature of the third-order correction term in the edge region, and it was observed in all fifteen sets of experiments. This feature does not exist in the classical second-order model.
在实验条件下，三阶模型对400mm尺度边缘区域的外推预测方向正确，数值精度在可接受範围内（T1-T2偏差＜1.5%，T2-T3偏差＜3%）。
The significant deviation in the T2-T3 region is primarily due to the accuracy limitations of the temperature control equipment (±0.3°C), rather than deficiencies in the model itself. Using a more precise temperature control device is expected to further reduce the deviation.
In summary, these results provide indirect but strong evidence supporting the applicability of the third-order model on a 400mm scale.
After finishing writing the report, Su Chen sent it to Zhou Zhiyuan and Lin Wei.
Zhou Zhiyuan replied that evening: "Very good. I have completed the framework for the revision materials. Just waiting for the reviewers' comments."
Lin Wei replied the next morning: "Received. Well done."
Su Chen leaned back in his chair as he looked at the two replies.
All around.
It started on January 27th and was completed on February 23rd. To be precise, it took 27 days.
In those 27 days, he obtained 15 sets of data, an R² value of 0.978, and an S-shaped temperature gradient transition that had never been reported in the field of MEMS.
These data are his trump card.
Regardless of what the reviewers said—whether it was requesting additional experiments or questioning the 400mm extrapolation—he already had an answer.
Moreover, this answer wasn't pieced together on the spot. It was planned, prepared in advance, and based on solid data.
Defense is the best form of offense.
We held on.
The offensive has already begun.
……
February 25th.
Suzhou, Vilan Microsystems.
The first batch of 300mm sensors is expected to be shipped.
Zhao Guoping and his quality inspection team started work at eight o'clock in the morning, checking each box of packaged sensors.
每箱一千颗，包装编号从WL-300-2021-001到WL-300-2021-151。每箱都有独立的质检报告、测试数据和追溯编码。
"The yield rate is 93.6%, confirmed," the quality control team leader reported to Zhao Guoping. "All products meet the shipping standards."
What about the logistics arrangements?
"The goods from Zhenxin Microelectronics and Huarui Optoelectronics will be shipped via SF Express tomorrow. The other six orders will be shipped out in batches the day after tomorrow. All orders are expected to be delivered before March 1st."
Zhao Guoping nodded and took out his phone to take a picture of the boxed items.
The first batch.
150,000.
300mm/±0.018°.
From academic paper to product. From data to commodity. From laboratory to customer.
He sent the photos to Lin Wei, along with the message: "First batch of shipments pre-inspection complete. All items passed inspection. Shipping will begin tomorrow."
Lin Wei replied with a single word: "Okay."
Then she forwarded the photo to Su Chen and Zhou Zhiyuan.
The accompanying sentence read: "Product line complete. Academic line continues."
When Su Chen saw the photo, he was organizing the final data package for the 400mm experiment.
He stared at the neatly arranged boxes in the photo, momentarily stunned.
Then he laughed.
150,000 sensors.
Each one has a precision of ±0.018°.
Each one serves as proof of a third-order model.
150,000 proofs.
This is more convincing than any paper, any data, or any R² value.
Because customers don't care whether your theory is a "paradigm shift". What customers care about is whether your product is easy to use, whether its accuracy is high enough, and whether its delivery is on time.
Vilan's answer was: easy to use, high-quality, and punctual.
150,000.
This is the answer.
……
February 26th.
The first batch of sensors has been officially shipped.
Zhao Guoping confirmed the order status of all eight core members in the logistics system—all were "shipped".
At the same time, Su Chen opened the submission system.
Status: Under Review.
no change.
He closed the page.
February 26th.
It has been two months since I submitted the paper.
According to the normal review cycle—plus the delay due to rejection—the reviewers' comments should be available between mid-March and early April.
Two to five weeks to go.
Su Chen was no longer anxious.
He had been anxious a month ago—during those two days when the submission system's status suddenly changed. But that's all in the past now.
He now has fifteen complete sets of data for 400mm edge verification, an R² value of 0.978, and an S-shaped temperature gradient transition that has never been reported before.
Akira Ishikawa was waiting in Kyoto to discuss the data with him after the paper was published.
Zhou Zhiyuan has already prepared the framework for the revision materials.
Lin Wei's first batch of sensors is already on its way.
Bosch's Albrecht is waiting for the paper's results.
STMicroelectronics' Bertoli is preparing both Plan A and Plan B.
He Zhiqiang, the owner of the Tian Shu chip, was granted a four-month grace period.
The discussion continues on Zhihu. The discussion continues on Twitter. The analysis continues in industry media.
Professor He Wentao said, "Let us wait for the paper."
The "calm analyst" said, "Wait for the paper."
Everyone is saying "waiting for the paper".
it is good.
Then wait.
Su Chen shut down his computer, stood up, and walked to the window.
In Suzhou at the end of February, the sun came out. The first warmth of spring was gently but surely arriving, breaking through the chill of late winter.
He looked at the sunlight outside the window and thought of Lin Wei's words—
"Our goals for April remain unchanged. If the paper is delayed, we will adjust the pace, but not the goals."
It's the end of February now.
The product line is complete – the first batch of sensors has been shipped.
The academic line is ready – 400mm data is in hand, and the revision material framework is complete.
Cooperation is on the horizon – Bosch, STMicroelectronics, and Infineon are all observing.
The competition is in full swing – the Tian Shu chip has gained a buffer, but time is limited.
Public opinion is spreading – from Zhihu to Twitter to industry media, the third-order model has become a hot topic in the MEMS community.
All the lines—everything—are converging towards one point.
The results of the peer review for the paper.
Su Chen didn't know when that outcome would arrive.
But he knew one thing—
He was ready when it arrived.
Everyone is waiting for the signal.
Su Chen already held the answer in his hand.
He wasn't waiting for an answer—he already had the answer.
He was waiting for the moment when the whole world would hear the answer.