As we all know, our country is a big chip consumer country. Whether it is mature technology or advanced technology, there is a big gap in the industry. Due to the early backwardness, although the chip industry has developed at an alarming rate in recent years, there is still a big gap. An important reason for this situation is the lack of lithography machines.
As an indispensable high-end equipment in the chip manufacturing process, without it it is equivalent to "talking on paper". Naturally, the giants in China's semiconductor industry also know this truth. So as early as 2018, SMIC ordered a state-of-the-art EUV lithography machine from ASML in the Netherlands.
However, due to restrictions in the United States, the high-end equipment has been delayed in delivery, which has also plunged my country's chip industry into a quagmire and cannot extricate itself. Huawei, which we are most familiar with, has suffered a deterioration in its mobile phone business due to a lack of "high-end chips" and has lost its previous leading position; while SMIC failed to release 7nm products as expected after catching up with Platform Semiconductor Manufacturing Co., Ltd. in terms of 14nm process yield rate. A large part of the reason for the relevant progress is that EUV lithography machines are not in place.
It is true that ASML has been showing favor to the Chinese market on the surface and pretending to be "helpless", but for it, it mainly covets the huge market. Once there are other options, it will never It is impossible to take risks to provide EUV lithography machines. No, ASML has recently reported new news that it will build a lithography equipment remanufacturing factory and equipment operation engineer training center in South Korea. It seems that the Korean market has been regarded as an "alternative option."
At the same time, chip giants such as TSMC and Samsung have also stated that they will build advanced process production lines in the United States to cope with the global "chip shortage" situation. Faced with the choices of other giants in the industry chain, where should my country's "high-end chips" go?
It is said that "the first bird is shot first." Huawei's strength is destined to face such a difficult situation today, but even so, it still has not given up on breakthroughs in "high-end cores."
Huawei executives have previously stated that as long as Huawei can afford it, HiSilicon will always exist. The implication is that even as Huawei continues to lay out the basic chip industry chain, Huawei has not stopped the research and development of high-end chips because it cannot produce them.
Recently, it was reported that Huawei’s 3nm chip is already on the way, and it has applied for a trademark called “Kirin” chip. In addition, another good news also came. According to Kuai Technology, Huawei has provided Hongmeng developers with a development board called Hi3861, which means that this is a chip designed based on the RISC-V architecture.
It is worth mentioning that this chip is not used in mobile terminals, but to help developers better build the Hongmeng ecosystem. This is also in response to ARM being acquired by Nvidia. If the dust settles, it is still unclear whether ARM's open licensing services will be damaged. This chip can also be used for "unforeseen needs."
It seems from various sources that even if HiSilicon loses its profitability, Huawei will always stick to the path of self-research, at least not lagging behind the world's advanced level in chip design.
It seems that Huawei's attitude is very decisive, and we all hope that it can break the situation, but the reality is cruel, which makes Huawei look a little "tragic".
Recently, Wu Hanming, an academician of the Chinese Academy of Engineering, said this when talking about EUV lithography machines at the 5th International Conference on Big Data Science and Engineering. This kind of machine covers 100,000 parts and costs more than 100 million US dollars. There are more than 5,000 suppliers of high-end equipment with high cost. It is a global product. It is unrealistic for one country or region to do it.
The academician’s views are very objective and reasonable, but it is like pouring cold water on Huawei’s head, because it means that even if Huawei can achieve success in design, because of the current manufacturing of lithography machines Some of the technology comes from the United States, so after all, it still can't be manufactured.
Of course, all walks of life in China are also trying their best to overcome the problem of lithography machines, including developing quantum chips and carbon-based chips in an attempt to bypass EUV lithography machines. However, after all, it is quite difficult and it will take a lot of time.
Judging from the current situation of all parties, we have only one way left, which is to avoid rushing for quick results and base ourselves on the development of current basic technologies.
Earlier, the Chinese Academy of Sciences also stated that the current main domestic demand is still for 55nm chips. Achieving independence in the industrial chain on this process is more valuable than developing high-end chips.
SMIC also slowed down the progress of the 7nm process due to problems with its photolithography machines, and chose to spend 70 billion to focus on the development of the 28nm process. The current shift in the focus of Huawei's business also illustrates this point. The shift from mobile phone business to software and cloud services just shows that the "high-end core" problem cannot be solved in the short term.
The author couldn't help but think of what my mother often said when I was a child, that is, "You have to eat one bite at a time." Indeed, nothing can be rushed. We always believe that "domestic cores" are independent That day will come eventually, don’t you think?