chip industry has been developing for more than half a century, the goal pursued has not changed, that is, stronger performance, and the method that depends on achieving stronger performance has not changed, that is, more advanced manufacturing processes are adopted, and the method of achieving more advanced manufacturing processes has not changed, that is, the more advanced lithography machine, . In fact, even the method of continuously iterating the lithography machine has not changed, that is, the light source with shorter wavelengths and objective lenses with larger numerical apertures.
So, over the past half century, the development of the chip industry has basically formed such a main line until the chip manufacturing process enters 7 nanometers. The situation has changed because the design cost and manufacturing cost of 7 nanometers and below processes are too high, and the technical difficulty is too high, which directly leads most fabs to withdraw from the 7 nanometers and below process market. In fact, even lithography machines are almost coming to an end now, because ASML has already stated that the numerical aperture of the lens cannot be increased anymore because the cost is too high and there is no light source with a shorter wavelength.
Therefore, the industry began to look for other methods, try to use as little as possible, and even do not use these advanced processes. Obviously, the development of the chip industry is no longer moving forward along the main line that lasted for more than half a century. This main line has begun to appear, that is, a new track has appeared in the chip industry. Since this is the case, we have the opportunity to change lanes and overtake.
The main line of the chip industry development we mentioned above can actually be described in one word, that is, homogeneity and isomorphism, that is, a chip adopts the same material and the same architecture. This leads to the improvement of performance, and we must rely on more advanced processes to integrate more transistors in the chip, and exchange quantity for performance. There is no doubt that this is a stupid method. Now, we have found a smarter direction, that is, heterogeneity and heterogeneity.
It is obvious that compared with homogeneous isomorphism, heterogeneous isomerization is to use different materials and different architectures to design and manufacture chips. For example, the very hot core particles, that is, Chiplet technology, are actually part of heterogeneous isomerization. The core particles are actually homogeneous technology. Previously, Chinese academician Wu Hanming pointed out that we have achieved chips made using 40-nanometer process through homogeneous isomerization technology, which can reach the level of 16-nanometer process.
almost spans five generations from 40 nanometers to 16 nanometers. This performance improvement is very obvious, and this is only the initial stage of homogeneous and isomerized. Just imagine that we already have a 14-nanometer process. If we use a homogeneous and isomerized 14-nanometer chip, we may be able to achieve the performance level of 2-nanometer chips. Everyone is optimistic. This is only the preliminary level of homogeneous and isomerized, and we have not added heterogeneous and isomerized.
heterogeneity is based on different architectures and uses different materials. At present, the main material of the chip is silicon, but new materials such as carbon nanotubes and graphene have greater potential. Currently, Tsinghua University has already possessed outstanding strength in the field of graphene. For example, Tsinghua University used graphene to realize 0.34-nanometer transistors, while Peking University has outstanding strength in carbon nanotubes, such as the paper "High-density semiconductor carbon nanotube parallel array for high-performance electronics". Peking University was published in the international regular journal "Science" two years ago.
In addition, in the top international journal "Advanced Functional Materials", Peking University also shared relevant technologies. The dean of Peking University School of Electronics stated that the performance of carbon nanotube chips will be hundreds or thousands of times higher than that of silicon-based chips, and the effect achieved in the current stage alone has been 10 times, which is an order of magnitude.
So heterogeneous heterogeneity is not just a direction for us to overtake in lanes, but more like the new main line of the development of the next generation of chip industry. In the future, chips will not compete with a few nanometers of manufacturing process, but with material technology and heterogeneous integration technology. Among them, heterogeneous integration technology relies very much on advanced packaging. The good news is that we are already in the first echelon whether it is the packaging lithography machine or packaging technology required for advanced packaging.
So in fact, the current chip technology lithography equipment problem has also been solved. In fact, under homogeneous and heterogeneous technology, the dependence on advanced process lithography machines has been greatly reduced. The industry has made pessimistic predictions. The shipment of ASML's top lithography machines will be greatly reduced in the future. If you think it is reasonable, thank you for giving a forwarding, like and reading to let more people know this good news.
