news

한어Русский языкEnglishFrançaisIndonesianSanskrit日本語DeutschPortuguêsΕλληνικάespañolItalianoSuomalainenLatina

【Text/Guardian Network columnist Qing Lan】

"This discussion is not based on facts or data, but on ideology... If ideology directly affects this (the interests of our business partners), I would be unhappy about it."

"We advise the government to build its own factories. Because it is impossible to make production in China more difficult without creating alternatives yourself... There is no point in preventing others from producing what you need."

In recent days, ASML's two former CEOs, Peter Wennink and Christophe Fouquet, have publicly criticized China's semiconductor technology controls, clearly conveying the lithography giant's helplessness and unwillingness in the face of the industry's "macro-environment."

At the same time, the US officials who shape the "big environment" are still enthusiastically forming alliances and brewing more sanctions, trying to weave this export control network more closely. In a recent interview, US Secretary of Commerce Gina Raimondo clearly stated: "They (China) cannot have our most sophisticated leading-edge AI chip, or the equipment that makes those chips... We will continue to update regulations to ensure the security of the United States." (They cannot have our most sophisticated leading-edge AI chip, or the equipment that makes those chips)

On the one hand, there is the intensified long-arm jurisdiction, and on the other hand, there is the booming Chinese market demand. The "lithography giant" who is struggling to find a balance on the tightrope will probably have to endure for a long time until the instigator wakes up from his dream.

We're listening, Jerry.

ASML would never have thought that the bullet hitting it right between the eyebrows today also has a profound connection with itself forty years ago.

In 1985, this fledgling small Dutch company was anxious about how to enter the US market. Although it was confident in its R&D capabilities, in the semiconductor equipment market where "seeing is believing and doing what you are familiar with rather than what you are new to" is extremely important, ASML faced many troubles similar to those of its Chinese counterparts today: it was difficult to impress customers without actual shipment results, and it was impossible to iterate products without customer feedback. The limited start-up capital was quickly consumed as the company operated.

Fortunately, the sudden escalation of the US-Japan "semiconductor war" brought ASML an opportunity to cross the valley of death. This year, AMD founder Jerry Sanders went all out at an industry conference, criticizing the quality of US-made semiconductor equipment and warning that high-quality and low-cost Japanese lithography machines would sweep the market.

Sanders' "outrageous remarks" immediately attracted the attention of Chuck Roberts, ASML's US marketing director. As the competition between the US and Japan in the semiconductor industry is increasingly infused with ideological colors, the pure Western European "bloodline" has inadvertently brought ASML a unique advantage in expanding the US market. Roberts excitedly said: "Sanders scolded GCA and Perkin-Elmer (two major US lithography machine manufacturers) and he doesn't want to be forced to buy Japanese products. This is our opportunity!"

Realizing this, ASML quickly launched a round of targeted advertising in industry media with the title "We're Listening, Jerry", which introduced point by point how its products were in line with Sanders' demands.

Sanders's attitude is largely due to AMD's current business difficulties, so much so that the industry expects it to be acquired soon like National Semiconductor.

This advertisement became the first stepping stone for ASML to enter AMD's supply chain. After repeated tests of business negotiations and prototype selection, and even resorting to the trick of climbing over the wall and sneaking into the customer's factory in the middle of the night to repair, ASML finally won AMD's bulk order and won the first heavyweight customer. This breakthrough also marked that it had initially gained a firm foothold in the lithography machine market and had passed the most fragile start-up stage.

Actively catering to the needs of the US market has become a main theme of ASML's early development. This diligence has also brought it rich rewards. From taking over SVG and then leveraging Intel's orders, to joining EUVL, acquiring Cymer, and "killing" Nikon's patent infringement lawsuit, all of these are closely related to the "convenience" provided by the United States, and have also become key boosts in ASML's development history.

However, after experiencing the baptism of the subprime mortgage crisis, as the focus of the global semiconductor industry accelerated its shift to Northeast Asia, a meaningful reversal of status gradually emerged between ASML and its American partners, and its major customers, who were willing to listen, also quietly replaced them with East Asian faces.

In 2012, in order to completely solve the EUV light source development problem and complete the acquisition of Cymer, ASML launched the so-called customer co-investment plan and issued additional shares to Intel, Samsung and TSMC for financing. Intel's number of shares subscribed was the largest, exceeding the total of Samsung and TSMC. Spending real money on this batch of non-voting stocks shows Intel's "sincerity". More importantly, Intel's move also sent a signal of firm standing to the industry, completely extinguishing the confidence of Japanese manufacturers such as Nikon to follow up on EUV lithography machines.

However, in July 2013, a wafer manufacturing "Big Three" meeting planned by ASML announced the sudden death of Intel's 18-inch technology, which became a symbolic event of the "rise of the East and decline of the West" in the semiconductor industry. Bill Holt, an Intel veteran who left the stage in frustration at the time, may have felt that ASML was "cheating and attacking".

ASML, which thought it had grown strong, ultimately failed to break free from the shackles that bound its freedom.

Since 2020, under the manipulation of "advisors" such as Jason Matheny, Tarun Chhabra, and Saif Khan who have deep roots in the "long-termist" circle in Silicon Valley, a "chip war" between China and the United States with old wine in new bottles has been successfully provoked. The trendsetters who have been invincible in the golden age of globalization suddenly discovered that the rules of the game of free trade have been abandoned by their makers.

From the "semiconductor war" to the "chip war", in the past forty years, the powerful ASML is no longer a flying knife, but has become the prey itself.

Of course, the visible hand of the American "market economy" has also not spared Samsung and TSMC. In order to satisfy the appetite of their "partners", the two foundry giants have planned a high-end chip production capacity of less than 5 nanometers in the United States of nearly 100,000 pieces per month.

This number itself indicates a blatant conspiracy to "dig up the roots".

A counterintuitive fact is that of the current global semiconductor manufacturing industry's total annual production capacity of about 100 million 12-inch wafers (converted number), logic chips with nodes below 5 nanometers that require EUV lithography machines account for only about 1%, and this proportion will not change by an order of magnitude in the foreseeable future.

In the "tight balance" high-end production capacity supply structure, TSMC is able to calmly adjust pricing and enjoy the fattest cream on the semiconductor manufacturing cake. However, as new production capacity in the United States that is sufficient to meet all demand takes shape, this structure will inevitably be shattered, and the bargaining power between TSMC and high-end customers will be reversed.

Faced with such a dangerous situation, the two major East Asian foundry giants dared not even make ASML-style complaints.

Bulova's Metaphor

Although the "chokepoints" in the industrial chain have been rectified, attempting to replicate the glorious victory of the "semiconductor war" against China is destined to be just a pipe dream for the U.S. government and the public.

In the semiconductor field, there is a profound disconnect between the "chip war" goals set by the United States and its capabilities. In a nutshell, this imposed "war" can neither stop the development of China's semiconductor industry nor realize the fantasy of the revival of the United States' high-end manufacturing industry.

It is true that long-arm jurisdiction has disrupted the development of China's advanced process technology. From the US government to the media of various countries, topics such as "how many generations behind China is in process", "when will EUV break through", and "how many years behind EUV" are generally the core of discussion. However, as mentioned above, such discussions have huge blind spots. The extremely low production capacity of high-end chips means that instead of struggling with this "extra question", there are many more important and urgent "basic questions" that need to be answered.

Objectively speaking, after experiencing the storm, the ecological development of China's semiconductor industry has greatly improved. From equipment materials to special processes and product categories, more and more blank spots have been filled. The high-quality customers that overseas manufacturers have "handed over" in recent years have provided local suppliers with previously unimaginable import opportunities, and the horizontal and vertical connections of the ecological network have been significantly deepened.

Ye Tianchun, Secretary General of the China Integrated Circuit Innovation Alliance, recently shared a group of progress in the localization of semiconductor materials:

The coverage rate of materials for 12-inch 45-28 nanometer processes exceeds 70%; the coverage rate of materials for advanced storage processes exceeds 75%; large silicon wafer products have achieved comprehensive breakthroughs, achieving full coverage of the main processes of domestic FAB plants; photoresists have made breakthrough progress, i-line glue market share exceeds 20%, KrF glue market share reaches 10%, ArF dry glue begins to be used in batches, and some varieties of ArFi immersion glue begin to be used in small batches; sputtering targets are fully supplied to domestic FAB plants, and some products have an international market share of more than 40%; local enterprises have become the main suppliers of CMP polishing materials, special electronic gases, and process chemicals; the main materials for traditional packaging are independently supplied......

Apart from the big tree of advanced process, the entire forest of China's semiconductor industry has become very lush.

Of course, it is still far from the time to celebrate. Looking at the industrial ecosystem of 7nm and above process, there is still a long way to go to develop each node from scratch and from existence to excellence. But no matter what, from resources to applications, the initiative is firmly in the hands of the Chinese themselves. The unremitting efforts of the semiconductor industry will surely steadily achieve its goals.

As for catching up with advanced processes, on the one hand, we do need to face up to its difficulty and talk less about it. Taking EUV lithography machines as an example, its LPP light source involves three major engineering challenges: lasers, tin droplet generators, and collecting mirrors. The breakthrough of ASML's Cymer has benefited to a large extent from the knowledge accumulation of LLNL and other US national laboratories in inertial confinement fusion, which is also the basis for the US's long-arm jurisdiction over ASML. If Chinese engineering teams want to learn from the existing system framework, they face a series of practical problems such as the lack of talent in gas laser research.

On the other hand, even with the full support of core suppliers such as Zeiss and Trumpf, ASML has gone through a full twenty years of hardship from launching EUV lithography research to delivering products to large production lines. However, the technological potential that can be tapped by LPP-EUV may not support the same twenty years of polishing and iteration for latecomers. Hyper-NA EUV, which forcibly increases the numerical aperture, faces multiple constraints such as too small depth of focus and three-dimensional mask effects. Another major change in the lithography technology system may occur in the middle of catching up.

Judging from the updated IRDS roadmap, free electron laser (FEL) may be the most promising candidate to become the basis for the next generation of 6.X nanolithography light sources. In this field that receives little public attention, Chinese researchers' innovative capabilities are not inferior to those of their European and American counterparts.

Our competitiveness in this potential track is not only reflected in the specifications and number of large FEL devices. In 2021, the State Key Laboratory of High Field Laser Physics of the Shanghai Institute of Optics and Precision Mechanics, Chinese Academy of Sciences, made breakthrough progress in the research of miniaturized free electron lasers based on laser accelerators, and was the first to complete the experimental verification of the principle of desktop free electron lasers, establishing my country's global leading position in this field.

my country's miniaturized FEL technology leads the world

In short, faced with the numerous difficulties that stand in the way of the limits of lithography, our efforts not only require hard work, but also call for "smart work" that adapts to local conditions and the times.

As for another major purpose of the "chip war", which is to drive the revival of high-end manufacturing, U.S. Secretary of Commerce Raimondo has a straightforward explanation: "We will make hardware manufacturing sexy again. Wouldn't it be nice to get a decent, glamorous, high-paying job manufacturing hardware in the United States? All of us will work together to rebuild our industrial base, enhance America's innovation capabilities, and create hundreds of thousands of high-paying jobs."

Raymondo's dream stems from her childhood experience. In that story, which she has told many times, the watchmaker Bulova moved its factory to China, destroying the community where the watchmakers lived.

Bulova, whom she mentioned, once made a groundbreaking contribution to the development of quartz watch technology. However, such a company with unique technology gradually fell behind in industrialization and was eventually forced to sell itself during the so-called "quartz crisis."

A close look at Bulova's history reveals that what brought it down was not competition from China or Japan, but rather the Americans themselves who missed the good old days. There was an inherent tension that was difficult to reconcile between "high-paying jobs" and the rigid downward trend of the manufacturing cost system.

Bulova is a metaphor for the fate of American manufacturing.

Take it easy, reflective freaks.

In response to the earnest appeals from ASML and other members of the semiconductor industry, an incomprehensible voice has always echoed in the Chinese public opinion field, cheering for the barbaric and brutal American-style regulation, and implying that the Chinese companies were at fault for what happened, thus forcing the "civilized world" to fight back. The hypocritical interpretation of "why not bully others but only you" is prevalent.

Before enlightening the masses and introspecting deeply, these "cultured people" who have insight into the world's affairs need to first make up for some general knowledge of their "spiritual motherland".

In the protracted US-Japan trade war and technology war in the 1980s, Americans had already demonstrated their "civilized world" colors.

In 1982, the death of a young Chinese technician, Chen Guoren, shocked the entire United States. Just because of a few quarrels, two Detroit auto workers who had never met him before beat him to death with baseball bats. Chen Guoren was mistaken for a Japanese and unfortunately became the target of the murderers' hatred for the Japanese auto industry.

Afterwards, an American judge sentenced the murderer to three years of probation and a fine of $3,000. In the verdict, he wrote confidently: "You cannot make the punishment fit the crime, you should make the punishment fit the criminal (character)."

During this period, the constantly honed "Japan Threat Theory" inspired many small town shop owners to hang up "No Japs Allowed" signs; a small clinic in Ohio offered employees a $400 subsidy for buying American-made cars, and the "Buy American" movement became popular. From haircuts to gas, customers who drove domestic cars could always get discounts, and even traffic violations could be pardoned due to the judge's "sympathy"; the Los Angeles Metro attempted to purchase Japanese-made train sets, which caused a wave of telephone "cyberbullying"...

Faced with such a large number of "individual cases", some domestic intellectuals probably do not dare to label them as the "Boxer Rebellion".

Compared with this surging "public sentiment" at the grassroots level, the performance of the American elite is equally impressive.

Lane Kirkland, head of the AFL-CIO, said bluntly: "Hearing the Japanese plead for free trade is like hearing the word 'love' from a prostitute's mouth"; House of Representatives leader Jack Brooks said bluntly that it was a pity that Truman only dropped two nuclear bombs on Japan, "he should have dropped four", and more members of Congress are considering imposing a 25% tariff on all Japanese imports; in the economic world, well-known scholar Lawrence Summers admitted that "the majority of Americans who now believe that Japan poses a greater threat to the United States than the Soviet Union may be correct", and Alan Weber, editor-in-chief of Harvard Business Review, concluded: "The cold peace with Japan is replacing the cold war with the Soviet Union."

In the field of public opinion, the cutting-edge Washington think tanks at the time gave Mansfield and other older generation country research experts the title of "Chrysanthemum Kisser", sometimes exaggerating the thrilling stories of "hundreds of Washington elites receiving Japanese subsidies" and sometimes describing the impending military conflict with trade rivals.

Of course, when it comes to writing prowess, we have to look to the legendary journalist Theodore White.

In his classic report "The Danger from Japan" published in 1985, Theodore White outlined the old accounts and new grudges of Japan's semiconductor industry in just a few strokes: "This year, the Ministry of International Trade and Industry's main target is clearly the U.S. electronics industry. Last year, the U.S. electronics trade deficit with Japan exceeded the automobile trade deficit, reaching $15.4 billion. The U.S. semiconductor industry is under attack. The Japanese are ruthlessly proposing to obliterate the U.S. dominance in the industry, an advantage based on our own research and inventions. Hitachi, now an exporter of electronics, was blown to pieces by our bombers as one of Japan's leading weapons manufacturers in the last two weeks of World War II. But Hitachi is now fighting back, and its tone has become aggressive again. The company is trying to take the semiconductor market away from Intel and AMD."

At the same time, Theodore White dissected the root causes of American national anxiety with astonishing frankness and coldness:

"If everyone in the world enjoyed equal rights to free trade, then those countries whose workers were grateful for meagre wages could conquer the markets of developed countries like the United States, where workers demand the highest wages in the world. Asia's living standards must rise, or ours will fall until a balance is reached... Behind Japan are the 'Four Little Dragons', with China and India close behind, eager to raise their living standards - at the expense of Americans, if necessary... All countries want to emulate Japan, even at the expense of American jobs. Conscience will not allow Americans to exclude the poor of the world who are striving to improve their lives, but common sense will not allow us to take a path that would improve their lives at the expense of our poverty... The Japanese enjoy the same peace, under our protection, at almost no cost, yet they reserve the right to drive the American livelihood into extinction."

As Theodore White revealed, in a certain American way of thinking, there is an unspoken "high-voltage line" between the American living standards and the industrialization of developing countries.

In the current "chip war" performance, it is not difficult to see that the script at the end of the Cold War is copied pixel by pixel. As long as we do not stop the pace of development and sink into the "lost years" like Japan, the hysteria of the US government and the public will probably not stop for a day.

However, times have changed. Such narrow-mindedness and overbearing practices will not only fail to block China's path to technological self-reliance, but will instead block the development space for "partners" such as ASML. When the United States wakes up from its dream one day, ASML's dominant position will probably be gone forever.

This article is an exclusive article of Guancha.com. The content of the article is purely the author's personal opinion and does not represent the platform's opinion. It cannot be reproduced without authorization, otherwise legal liability will be pursued. Follow Guancha.com WeChat guanchacn to read interesting articles every day.