Rewrite the Technological Landscape

"Some people may be familiar with 3D transistors, but in fact, this is not a new concept. We have seen the earliest depletion-type poor-channel transistors in 1989, and then multi-gate transistors based on DELTA technology became the industry's first an important research direction.

For example, Intel publicized their 3D transistor design as early as 2002, and Dafeng Group also began to invest heavily in the research and development of 3D transistors very early.

Why so many people are staring at this direction, to put it bluntly, is because when the size of the transistor shrinks below 25nm, the traditional screen size can no longer be reduced, and then problems will arise.

Let's watch the big screen together. "

Several technical analysis diagrams appeared on the big screen, "Two-dimensional structure transistors have been used since the 1960s and have been used for nearly half a century. However, we have noticed that as the gate length becomes smaller and smaller, the source The closer the distance to the drain, the thinner the oxide under the gate, which increases the possibility of leakage.

At the same time, whether electrons can flow from the source to the drain is controlled by the gate voltage, but the smaller the gate length, the smaller the contact area between the gate and the channel, that is, the influence of the gate on the channel The force becomes smaller.

These problems are especially evident when the gate length shrinks below 20 nanometers.

When the original source and drain are pulled up to become a three-dimensional plate structure, the channel between the source and drain becomes plate-shaped, and the contact area between the gate and the channel suddenly becomes larger.

In this way, even if the gate length is reduced to less than 20 nanometers, a large contact area can still be retained, and whether electrons can flow from the source to the drain can still be controlled. It can be said that the carrier channel of the multi-gate transistor is affected by contacting each plane. The gate control provides a better way to control the leakage current.

At the same time, since the multi-gate transistor has higher intrinsic gain and lower channel modulation effect, it can also provide better performance in the field of analog circuits, thereby reducing power consumption and improving chip performance.

Our experimental data shows that the 32nm three-dimensional transistor can bring a performance improvement of up to 40% compared with the 32nm planar transistor, and the power consumption is reduced by half under the same performance. "

At this time, the person standing on the stage in charge of giving the lecture was named Yang Peidong.

When it comes to 3D transistors, especially when it comes to FinFET, that is, Fin Field Effect Transistor, the first reaction of more people must be Hu Zhengming. After all, Hu Zhengming has always been known as the inventor of FinFET.

But in fact, FinFET was not invented by Hu Zhengming alone, but by a team, and there are three core figures, Hu Zhengming is one, and the other two are Jin Zhijie and Yang Peidong.

In order to develop 3D transistors, Meng Qian really approached Hu Zhengming, but perhaps because Hu Zhengming was still a consultant of TSMC after leaving TSMC, and the relationship between Dafeng Group and TSMC was very tense in the past few years, Hu Zhengming did not accept Meng Qian’s offer. invite.

So Meng Qian went to find Yang Peidong and Jin Zhijie. At the same time, Dafeng Semiconductor still has many top talents dug from other places. Today's success is also the result of the team.

This Yang Peidong can be regarded as a magical figure, with the three titles of "World's Top 100 Young Inventors", "World's Top 100 Chemists", and "World's Top 100 Materials Scientists".

And when Yang Peidong is mentioned, people will definitely think of the word hard work, and all the people who have worked with Yang Peidong will not miss this evaluation.

But Meng Qian and Yang Peidong were particularly speculative when they communicated in private back then. Their views on this issue were that hard work is of course a positive thing, but many times desperate people like Meng Qian and Yang Peidong simply really want to go. To do one thing, in the eyes of outsiders, it becomes desperate, but in fact, it is just because I want to do it.

There are two ways to promote hard work, one is to sell anxiety, the other is to sell love, and Meng Qian prefers the latter.

Like Hu Zhengming, Yang Peidong has worked in many universities and companies. At Dafeng Semiconductor, he is only a consultant and not often in the company. However, he really contributed a lot to this project, and finally decided to let him be the keynote speaker.

In addition to his reputation in the industry, Yang Peidong also has an indescribable self-confidence that is very suitable for the lecture. Just like now, after explaining the technology, Yang Peidong said in a flat but loud voice, "With the efforts of the company team, 3D Transistors are finally moving from the lab to the market.

It can be said that this is the greatest milestone invention in the history of transistors, and it can even be said that the transistor has been reinvented. We can finally say with hope that the era of 3D transistors has really come.

The entire semiconductor industry will officially enter the 3D era, and Gale Semiconductor will break the bottleneck of Moore's Law.

The semiconductor market will usher in another dynamic development, and I believe that this will be the most important technological breakthrough in the semiconductor industry for at least the next five years.

We should be happy for Dafeng Semiconductor, and even more so for the entire semiconductor industry, because Dafeng Semiconductor has given this industry new hope. "

It is because Yang Peidong is not only a consultant of Dafeng Semiconductor, so it is not so unreasonable for him to evaluate from the perspective of the industry, and this kind of promotion makes this technological breakthrough more meaningful and valuable, and the audience applauded.

At this time, Intel's Paul was shaking while watching the live broadcast, because this honor should belong to Intel.

The world's first 3D transistor Tri-Gate that came out of the laboratory and could be marketed was released by Intel on May 6, 2011.

Therefore, Intel has once again consolidated its position as the leader in the semiconductor industry, but all of this is gone. Even if Intel releases Tri-Gate tomorrow, Intel can only be the second child.

Paul couldn't accept this fact for a while, and even sent the intelligence department to investigate whether Gale Semiconductor had come to steal technology.

Because Intel announced this technical direction in 2002, but it has been considering this technical direction since the 1990s. At that time, Dafeng Group had not yet been established.

Paul couldn't understand or believe that Gale Semiconductor had come from behind to catch up with Intel and matured the technology. There should be at least a 5-year R\u0026D cycle gap, which is theoretically impossible.

In fact, many people in the market have this doubt, and the final conclusion is that Dafeng Group has already determined this direction and focused on it.

When the technology was still in the era of 90nm+, even Intel did not fully confirm the future direction, because technology has two sides, and there are problems in engaging in 3D transistors.

It's like killing an analog or a designer, because a designer with a traditional process has to go through fewer variables to achieve the desired electrical response.

On the one hand, this requires tool innovation to keep up. On the other hand, it greatly increases the difficulty of design. During the entire development process of 3D transistors, some people have always suspected that this direction is wrong, so continue to use two-dimensional transistors, through material innovation and two-dimensional transistors. Dimensional structure innovation is also a mainstream direction.

This is just like whether the lithography machine should be immersion or not. Everyone is not sure. Everyone is trying a little bit, but the reborn is very sure. There is no way. When Intel is still hesitating and hesitating, Meng Qian The direction has been determined and desperate, and the core talents have been dug out, and everything just happened.

At this time, Intel was very depressed. It wanted to rely on this technology to save Intel's embarrassing situation in the mobile chip market, and to seize the mobile chip market by achieving leading chip performance through technological leadership.

However, what Intel has done in the field of mobile processors is really useless. It can be said that Intel used to be 3 years ahead of this process, and Intel said at the time that the main purpose of this process was to go to mobile processors. Yes, it also increased the investment of nearly 5 billion yuan in three years, but it just didn't make the mobile processor well.

Not to mention this life, Intel's 3-year lead is gone now, and I don't know if Intel can come up with any way to change its fate.

But no matter what Intel plans to do next, the market may not trust Intel so much at the moment. After the press conference, the market reaction is the most real. On the same day, the stock price of ARM in the UK fell by 6.5%, and Intel’s stock price fell by 8.6%.

And Samsung next door fell directly by 9.8%.

Thanks for the reward of The Return of the Incredible Hulk!