Global Monopoly of Technology

Just now, a group of scholars exchanged academics, and everyone talked freely, but Luo Sheng was almost ignored.

And it was only at this time that he came back to his senses and realized that Luo Sheng is not only an entrepreneur who is proficient in business, but also has a greater hidden attribute, that is, a technical master.

Luo Sheng obviously didn't finish his sentence, he stared at the revised mathematical model and continued: "One more thing, since we are all taking the silicon of the depletion layer and inserting an oxide layer, then why do you have to put a bunch of useless ones? What about the silicon underneath? Make another gate directly under the oxide layer, with the channel on both sides, wouldn't it be cooler?"

Everyone was stunned for a while, and when they came back to their senses, they all applauded, and Luo Sheng added: "That's not enough, since that's the case, why do you have to bury the oxide layer in the silicon? I can totally get the silicon out. , wrap the insulating layer like a sandwich, and put the grid on the outside, isn't it cool?"

A semiconductor researcher said excitedly: "It's a genius idea, this scheme not only greatly reduces the leakage current, but also because there is one more gate, the two gates are generally connected together, Therefore, it is equivalent to greatly increasing the capacitance of the insulating layer mentioned above, which is to greatly improve the switching performance of the transistor. I dare to make a decision. This tri-gate transistor is definitely a revolutionary progress in architecture. , not only 45 nanometers, but also 32 nanometers, 22 nanometers, and 14 nanometers have great imagination space, which is another victory of Moore's Law."

It is worth mentioning that the evolution of process nodes in the semiconductor industry can be predicted, and Moore's Law has emerged as the times require, that is, how many nanometers in the next generation of chips.

For example, 130 nm, 90 nm, 65 nm, 45 nm, 32 nm, 22 nm, 14 nm, 10 nm, 7 nm (5 nm), you will find that this is a proportional sequence with a ratio of about 0.7, which is why .

However, this is just a naming convention, and there is still a gap with the actual size.

At this moment, all the research scholars at the meeting were very excited, but after the excitement, some people have returned to rationality.

A scholar tactically held the mirror and said: "This is a genius design. Basically, I dare to judge that it is feasible without supercomputing simulation verification, but the difficulty is to realize this idea. Why? Because of the erected part Silicon, the silicon used for the channel, is too thin, less than 10 nanometers, not only much smaller than the minimum size of a transistor (currently), but also much smaller than the smallest size that can be engraved by the most sophisticated lithography machines today , how to get this plan out is the real problem."

As soon as these words came out, the other excited researchers also calmed down.

Yes, this problem cannot be avoided. In the end, it came back to the problem of the lithography machine. At present, the most sophisticated lithography machine cannot be solved, not to mention that it is impossible for Huaguo to buy the most advanced ones. Lithography equipment.

At present, Huaxin International is the domestic leader in this field. It is frustrating that foreigners are already able to mass-produce 32-nanometer chips, while Huaxin International cannot even mass-produce 65-nanometer chips. This is still in the process of introducing In the case of a lithography machine, I can't build a lithography machine with a 65-nanometer process.

Luo Sheng smiled and said: "Everyone understands it wrong. It takes a lot of process flow charts to describe it in detail, but the basic principle is that this part of silicon is not lithographically etched, but grown. It is first etched with ordinary precision lithography. A bunch of 'shelves' are formed, and then a layer of silicon is deposited, then a thin layer of silicon will grow on the edge of the shelf, and then selectively etched to get rid of the excess material, and the rest is these vertical layers. It's as simple as that, ultra-thin silicon."

When people hear such a set of methods,

Everyone was stunned for a long time.

Kneeling completely, this IQ is too overwhelming.

Seeing that everyone was speechless, Luo Sheng couldn't help but smiled and added: "Actually, this physical model of tri-gate and gate-all-around transistors can be traced back to around 2000, when Professor Hu Zhengming of Berkeley first proposed it. , but no one has realized it so far, but this program can be realized.”

Xu Zhijun said excitedly: "Don't say anything else, let's talk about other issues clearly and clearly."

This may be one of the few solutions that Lens Semiconductor can come up with to compete with the world's top semiconductor manufacturers. It is the FinFET solution, that is, a fin field effect transistor, a new cutting-edge complementary metal oxide semiconductor. transistor.

Patents are required.

Lens Semiconductor can be said to be the leader among domestic semiconductor companies, but it is still a younger brother compared with international manufacturers. The weak patent background can be reflected, and there are basically no top patents that can be obtained.

Accumulated too little.

In the days that followed, under the leadership of Luo Sheng, whether it was Lens Semiconductor or the Riviera Lab, the researchers went to war in an all-round way. Luo Sheng worked out the process flow diagrams one after another. After verification, the R\u0026D team began to use violent methods to find many kinds of strange materials to test one by one.

Otherwise, what is the method of violence?

There is no way to do this. Except for the few companies that have mastered the technology, no one in the outside world knows about it. They can only try violently. Graphene materials are a very good option, and stanene materials are also very good.

However, the current technical conditions cannot be prepared at all, and the cost cannot be reduced, so it is naturally not competitive at present.

If these materials can be produced, it is no less than a subversive change initiated by new materials, and the fourth industrial revolution may be coming ahead of schedule.

New materials must be developed. Whether it is graphene or stanene, the research and development of these new materials is a part of Luo Sheng's future layout, but the meal still has to be eaten one bite at a time.

It is still the era of silicon chips, and the performance of silicon materials has not been brought to the extreme. At present, the mainstream technology is still at the 45-nanometer stage, and the 32-nanometer technology has not been fully rolled out. Moore's Law is still strong.

The R\u0026D team has made a lot of trial and error, trying more than a dozen kinds of materials almost every day and coming up with the results.

...

Science and Technology Complex Building.

In the supercomputing room on the ground floor of the building, hundreds of engineers are busy here. Most of these people are from Dawning Technology, and Luo Sheng is also present.

The focus of recent work is undoubtedly the chip design and production of the third-generation Azure mobile phone, but Luo Sheng will not be immersed in a module 24 hours a day, and he has to take care of the submarine data center. Almost all core technologies have his presence in it.

It's no wonder why Luo Sheng feels that there will never be enough talents, especially high-end talents.

At this time, a person in charge of Dawning Technology came to Luo Sheng and said with a worried expression: "Mr. Luo, is it really feasible to do this?"

Because Luo Sheng overthrew the solution of Dawning Technology, which made them very worried, if the computer fails, it will cause great trouble.

Luo Sheng was staring at a liquid crystal display panel, tapped a few times on the keyboard, called up a column of data to view, and said without raising his head: "Just do as I said, you don't need to worry about the rest, it's a mess. I won't file a lawsuit with you, it's my fault, the contract has been signed, what are you worried about?"

When the other party heard him say that, he was at a loss for words and didn't say anything.

It can only be said that it is rich and willful. This is a supercomputer, not a civilian PC computer.

Okay, you said so, anyway, this is your personal supercomputer, you have to fool around and do not follow Party B's suggestion, when the computer is broken by you, Dawning Technology will not take the blame.

In a sense, if Luo Sheng breaks his supercomputer, it will be a good thing for Dawning Technology, because it must spend more money, then Dawning Technology will definitely make extra money.

Since Luo Sheng had said this for his sake, the technical person in charge just shut up.

All right, Party A has the final say!

"Mr. Luo, President Zhang of Huaxin International came to visit, do you want to meet him?" Just as Luo Sheng was arguing about his supercomputing upgrade, the housekeeper of the complex came to him and whispered to him.

Luo Sheng nodded: "Well, I see, you will receive it first, let him wait, and say I'm dealing with things."

There are more people who want to see him now, just wait.

...