“…”

"The wafer test line officially started trial production at the beginning of the month. The yield and stability of the lithography machine are very unsatisfactory and require certain adjustments;

According to reports, 65nm process trial production is expected next month.”

It seems that Fang Nian will have doubts. Speaking of this, Wen Ye specially added a few words: "DUV immersion lithography can actually support lithography to 7nm process;

According to the professor in charge of the wafer test line:

The traditional international semiconductor technology blueprint defines the minimum metal spacing of the lithography process technology node.

According to a Rayleigh formula: CD=K1×(λ/NA), reducing the wavelength λ, increasing the numerical aperture NA of the lens, and reducing the comprehensive factor K1, the lithography process can be continuously evolved.

DUV laser light sources have been stagnant at 193nm wavelength since the late 1990s;

However, some people have proposed a new solution, an engineering simple method: add 1mm thick water to the wafer photoresist. The water can refract the wavelength of 193nm to 134nm, and then achieve a half-period of 65nm;

Anyway...

The professors in the wafer laboratory are confident that they can make a breakthrough to 65nm lithography in a short time, which may be a little longer afterwards, but it won’t be too long.”

After finishing the script in one breath, Wen Ye did not pause for much and continued to explain.

“Although the development of EUV lithography led by Zhihu Laboratory is very slow, it is due to various technologies acquired from European and American countries, etc.;

It is very helpful to promote the research and development of DUV lithography machines."

“…”

Finally, Wen Ye summarized: "As of now, these projects have achieved definite results."

"The actual experiment of the Shengyu 4G networking plan that Oguri is responsible for will take a little time, and it should be completed by the end of April."

After listening to Wen Ye's complete speech, Fang Nian pondered for a moment: "In other words, DUV lithography, which did not pay attention to at all, made rapid progress. It invested 7 billion in the last quarter of last year, and in the first quarter of this year, it invested 10 billion in EUV progressed rapidly?"

"It can also be understood this way." Wen Ye replied helplessly, "Whether it is a trait or a coordinated unit in China, the accumulation in EUV can be called 0."

Fang Nian smacked her lips: "It's really a bit unexpected."

Then Fang Nian looked at Wen Ye and changed the subject: "But Wen Mi, you can't even understand the script?

If the DUV lithography machine in the laboratory is hard, it can produce 65 nanometers now. However, without adjusting the lens and other hardware, the yield will explode, it is nothing more than multiple lighting."

Wen Ye: "…"

She couldn't help but complain: I thought everyone was like you, a big guy (B) and knew everything?

Fang Nian didn't want Wen Ye to answer, so he continued to ask himself: "Is there a more specific report on the failure of Shenlong 511?"

Wen Ye replied: "Su Zifeng's complete report explains that the performance indicators of Shenlong 511 are twice that of Shenlong 1, but in fact they are not so good. After actual verification by the laboratory, it was found that there was no problem with the architecture, but there were some optimizations."

Fang Nian hurried and asked again: "Where is the server CPU?"

Wen Ye replied: "The performance of a single chip is far inferior to that of mainstream server CPUs, such as Intel Xeon E5 series;

Strictly speaking, although it is also a desktop-level CPU, its performance requirements are not the same level at all, and the usage environment is very different, so Bai Ze can actually be said to have zero foundation;

Therefore, Bai Ze’s large CPU team really hopes that this CPU can be used in the actual environment to collect more actual data for research and development and design.”

“…”

After listening to Wen Ye's words, Fang Nian was silent for a moment before slowly speaking: "How much has the R&D investment from the large CPU project team so far?"

Most people don't prepare these numbers.

Fortunately, Wen Ye is not an ordinary person. She knows very well what Fang Nian wants, so she reports without hesitation: "The two academicians of the big CPU project are from the Academy of Sciences. Their style... How to say it is very frugal;

He is not so sensitive to the concept of time and has never achieved any results. He is a bit cautious in all aspects;

So to date, the large CPU project team has only spent 210 million R&D expenses."

Fang Nian frowned and looked at Wen Ye: "Hmm? Didn't I re-arrange it?"

Wen Ye said with a tough forehead: "Mr. Fang, this is a confidentiality project. You have told me that Su Zifeng should not intervene for the time being;

We have also supervised many times to achieve the current progress efficiency;

Of course, there are also reasons why some of the current research is based on the White Dragon series, and it does not require much funding;

The efficiency of the slitting film is quite fast this time.”

Fang Nian thought about it and decided to say, "You go to Luzhou in person and talk to the frontline leaders of Luzhou;

On-site inspection of Su Zifeng's performance, if appropriate, you can open this project to her;

Transfer my meaning, it’s not about making them perform the secret to save money!”

The finance department approved 2.5 billion yuan, the first batch was 1.5 billion yuan, and the second batch had not yet moved, but this thing was so annoying that it actually spent 210 million yuan to create a high-performance CPU.

Wen Ye nodded quickly.

Fang Nian said again: "You go and confirm again now, can they determine the stability of this CPU?"

“…”

Wen Ye hurried to call.

A few minutes later, he came to report: "Academician Han gave a guarantee to ensure stability!"

Upon hearing this, Fang Nian no longer hesitated: "How many servers are there in Luzhou Frontier Data Center now?"

"Ten thousand five." Wen Ye replied.

Fang Nian said: "Let's do this, the CPU usage of 10,000 servers, a small number of Luzhou data centers are used in the new data center in Guiyang."

"It is calculated based on the server usage. For example, the single-chip processing performance indicator of a server planned is 10, and the single-chip processing performance of our own is 2.5, so we use 4 pieces to replace it;

Anyway, the installation and debugging have not started in Guiyang, so you can modify some motherboard orders. Doesn’t the server’s motherboard have a single support for up to 4 CPUs?

In short, we must ensure that the performance metrics of a single server do not decrease."

Wen Ye confirmed: "Is the ratio of Luzhou and Guiyang 1:9?"

"2:8, the sample number is slightly balanced, which is convenient for reference to the experiment." Fang Nian determined.

Luzhou's cutting-edge data centers are more important. In fact, replacing one thousand servers is more limiting, and two thousand is even more radical.

But time doesn't wait for anyone, and who knows how many times it will take to be iterated before being qualified to put it on the table.

Finally, Fang Nian emphasized: "You will go to Luzhou early tomorrow. I need an efficient team, not a group of cautious engineers who are not creative and imaginative!

In a word, I hope to hear the big CPU project team dare to use the name of the planning before August!

Instead of now, a CPU that doesn’t even deserve a name!”

"I understand!" Wen Ye said seriously.

Then she heard Fang Nian's murmur.

"The frontier is not very strong yet, so I have the meaning of not leaving the office. Alas..."

"There are too many restrictions on basic science and too backward, and the necessary confidentiality measures are necessary. It really sets obstacles for yourself."

"well……"

Wen Ye felt deep in her heart.

She, the cutting-edge CEO, also has many things that have not been done well.

Thinking of it, Wen Ye gritted his teeth gently...

-

=======
Chapter completed!