To ensure that the huge semiconductor integrated circuit processing related industrial chain will achieve industrial upgrading within 2 years, the risk is also quite large, and the barrel effect will be vividly reflected here.

"Of course, the problem of the rupture of the supporting industrial chain required for the upgrading of the process technology has not yet appeared, and as long as enough funds are invested in the future, the problem may be avoided, but if we return to the semiconductor integrated circuit product itself, according to the current development speed, what level will the process technology reach in 30 years?"

Directly pulling the topic to thirty years later, the jump in the timeline is indeed quite large, but it does not prevent everyone from being interested in this issue, taking out their portable calculators, and directly starting to do the simplest multiplication calculations.

In a two-year cycle, between one cycle and the next, the semiconductor process accuracy is multiplied by 0.7 to the next generation of process accuracy, so .....

On the podium, Ni Guangnan, as a spokesperson, also helped everyone calculate the final answer in advance very early.

"7 nanometers, 30 years later, if Moore's Law does not fail during this period, then the semiconductor process will enter the 7 nanometer era after 30 years. But I have a problem here, at present, the most widely used semiconductor semiconductor in the semiconductor integrated circuit industry is silicon semiconductors, and the diameter of a single silicon atom is only 0.22 nanometers. ”

At this time, the scene was completely quiet, and everyone's eyes when they looked at Ni Guangnan on the stage changed, and their expressions were extremely solemn.

As leaders in scientific research, academia, and industry, everyone who participated in the conference had a solid foundation and knew what it means when a silicon atom with a diameter of 0.22 nanometers and a process accuracy of 7 nanometers exist at the same time.

Thirty years from now, if the semiconductor processing process is 7 nanometers, what level will it reach after 40 years, according to Moore's Law, by that time it will theoretically reach the era of 1 nanometer process, and the question will come.

The current semiconductor integrated circuit industry is based on silicon materials, and all semiconductor integrated circuit processing is based on the premise of "not destroying silicon atoms", so the theoretical processing accuracy limit data of silicon semiconductor integrated circuits comes out.

Looking down at the silence under the stage, this feeling is really good, wasn't it still very awesome just now, why don't you speak now?

Reaching out and patting the microphone slightly, the venue's sound equipment made a low popping sound while the keynote speech continued.

"The diameter of the silicon atom is 0.22 nanometers, and we consider that there is still a certain gap between the atoms, so the theoretical silicon semiconductor integrated circuit process accuracy can reach the maximum threshold of nanometers, and this data can never be exceeded. ”

Gordon Moore predicted the development and update speed of semiconductor integrated circuits, and now Professor Ni Guangnan has stood up and finally reached the development limit of silicon semiconductor integrated circuits through scientific calculations and speculations, and concluded that after 40 years, Moore's law will definitely fail, and silicon semiconductor integrated circuits will also reach the limit.

Moore's Law is an oracle enshrined by Intel Corporation, and now it was suddenly stirred up by Ni Guangnan, and almost everyone in the venue found that maybe this so-called law could not be held for decades.

So, is it necessary for everyone to continue along this road and follow the direction planned by Intel's big stick and small flag?

It was also in 1991 that Paul Gargini, a senior researcher at Intel Corporation, who also stood here at the ISSCC conference to give a keynote speech, advocated the establishment of a semiconductor R&D planning blueprint association in the whole industry to coordinate the development and upgrading of the global semiconductor industry.

If Chief Engineer Ni Guangnan does not stand up this time and does not shave his ears on Moore's Law, then the entire semiconductor industry may really gradually be subsumed into the association, and this association is obviously controlled by Intel, and in the future, Intel will gradually establish an unchallenged huge advantage in the field of semiconductor integrated circuits.

At the upper level, until 1998, semiconductor industry associations from Europe, Japan, Taiwan and South Korea joined the Semiconductor Industry R&D Planning Blueprint Association, and the association became an international organization.

Although Moore's Law has been fooled this time, it can at least continue to adhere to it for a few more decades, which is not controversial.

The most fundamental reason why the ABC-PC system pushed Kaiyang Semiconductor out and let Chief Engineer Ni Guangnan stir up the semiconductor industry R&D planning blueprint association advocated by Intel is to prevent Intel from continuing to expand its influence in the whole industry.

At this point in the development, everything is also played out within the framework of a pre-arranged plan, and then the water army should take the stage to direct the focus of the scene.

"Theoretically, the processing limit of silicon semiconductor integrated circuits is 0.5 nanometers, this is the conclusion you deduced, as the top semiconductor integrated circuit research scholars in the Chinese Republic, I believe there should be a corresponding solution, isn't it?"

The speaker is none other than IBM's original plan, but the content of this guy's speech is different from what he planned earlier.

It is clear that this IBM company is not a good person, and the chief engineer Ni Guangnan on the stage only pasted Intel's big ear scraper, and now IBM immediately stood up and made it difficult in every way by asking this kind of question.

I go, God knows how to solve the problem of semiconductor integrated circuit processing in more than thirty or forty years.

Raising this question is nothing more than crossing the river and tearing down the bridge, the goal of cracking down on Intel's proud Moore's Law has been achieved, IBM certainly does not want Kaiyang Semiconductor to become famous after this war, and it is appropriate to embarrass Chief Engineer Ni Guangnan on stage through this tricky question.

The trick is quite in line with the usual style of the Americans, and although it is unexpected, it is also reasonable.

Fortunately, Wang Zhengguo talked with Chief Engineer Ni Guangnan before about how to continue to improve the performance of the processor under the stagnation of the semiconductor manufacturing process, and now he happens to use it to sell it now.

After a few seconds of surprise, he smiled and nodded, signaling for the friend to sit down.

"First of all, I am not the top researcher in the semiconductor integrated circuit industry in China, and there are many young and capable researchers in China. However, when it comes to how to solve the problem that the development of semiconductor integrated circuits in the future is subject to processing technology, I have some ideas to share with you. ”

Now that we have departed from the content of the speech that I planned earlier, I should let go of my hands and feet and give an impromptu speech.

"To solve this problem, we first need to abandon the current crude practice of reducing the size of a single transistor by constantly updating the processing accuracy of the semiconductor process per unit area, and finally achieving the integration of a larger number of transistors. Now we need a more ingenious solution, and this involves the concept of spatial dimension, which may be a bit of a fantasy for me below, but I'm sure you will understand it. ”

Thinking of the proton computer that Wang Zhengguo pulled out before, this is a sincere smile, testing the IQ of the world's top scientific researchers at these ISSCC conferences, I am looking forward to it!