The Economist on Chipmakers (II)

choi

(3) Japan's semiconductors (1) | Roaring Back?  The country used to be a chipmaking powerhouse. Can it be one again?
https://www.economist.com/asia/2 ... -into-the-chip-wars

Quote:

[paragraph 1:] (reporting from Chitose) KOIKE ATSUYOSHI * * * The 73-year-old semiconductor engineer * * * his latest company [meaning employer] Rapidus. Founded in 2022, the firm opened its massive semiconductor factory, or "fab", last year in Chitose, a small city on Hokkaido, Japan's northernmost main island. In December Rapidus became the first Japanese entity to acquire an extreme ultraviolet lithography (EUV) system from ASML, the Dutch company that makes the unique devices; Rapidus had the complex up and running within months. In mid-July, Mr Koike announced the successful pilot production of two-nanometre (2nm) transistors, the thinnest, most advanced chips yet. * * *

[paragraph 2:] Rapidus is the most ambitious element of a broader effort to revive the semiconductor industry in Japan. In the boldest industrial policy push in a generation, the Japanese government ploughed ¥3.9trn ($27bn) into support for semiconductors between early 2020 and early 2024. As a share of GDP, that amounts to a bigger commitment than America made to its semiconductor industry through the CHIPS Act. Japan wants both to revive its domestic champions and to attract foreign ones, such as TSMC, the Taiwanese semiconductor giant, which now makes chips in southern Japan. * * *

[paragraph 3:] Japan once dominated the semiconductor industry. In the 1980s, Japanese firms accounted for more than half [51%, to be precise] of the global market, and an even bigger share of the cutting-edge chips of the time. But trade friction with America led to limits on Japanese chip exports, creating an opportunity for rivals in Taiwan and South Korea. Japanese companies also struggled to shift to an era of increasing specialisation in semiconductor production. * * *

[paragraph 7:] * * * Big subsidies helped entice TSMC to set up shop in Kyushu. Its first fab there produces chips of 12-28nm -- the most advanced type of semiconductor to be produced in Japan so far, but still well behind its [TSMC's] state-of-the-art models. * * *  

[paragraph 8:] Micron, an American memory chipmaker, has also received more than bn in subsidies to expand its chipmaking facilities in Hiroshima. * * *

[paragraph 9:] * * *The firm [Rapidus] is the high- est-risk and highest-reward bet of the bunch. Born of a partnership with IBM, which developed a new method for making of next-generation transistors, a type electrical component, Rapidus hopes to leapfrog across a generation of semiconductor engineering and catch up with global pace-setters. It has attracted investment from a consortium of eight blue-chip Japanese firms, including Sony, Toyota and SoftBank. The government has also bankrolled much of the initial cost, to the tune of ¥1.72trn ($12bn) through early 2025.

[paragraph 10:] * * * Rapidus has had to rely largely on older specialists who came of age during Japan's first chip boom; the average age of its recruits was initially over 50. Roughly 150 top engineers were sent to train at IBM's research facility in New York.

[paragraph 11:] * * * Rapidus is positioning itself as a boutique option, able to make smaller lots of specialised chips, rather than large batches of one-size-fits-all offerings. "We have no intention of directly competing with TSMC-the markets are different," Mr Koike says. * * *

Note:
(a)
(i) KOIKE Atsuyoshi 小池 淳義 is not a mere engineer at Rapidus. He is its president 社長, whose official title in its website is 代表取締役社長兼CEO.

Japanese-English dictionary:
atsu-i 篤い 【あつい】 (adj): "kind, cordial, hospitable, warm, faithful"
(ii) Rapidus
(A) The English adjective rapid came from Latin adjective masculine rapidus (feminine rapida) of the same meaning.
(B) Rapidus Electronics has NO Japanese name. In Japanese=language media, the company is mentioned in English.
(C) Press release: Rapidus Begins Installation of Japan’s First NXE:3800E EUV Lithography Machinery for Semiconductor Mass Production. Rapidus Corp, Dec 18, 2024 (from ASML for 2nm).

I could find a photo of Rapidus building its (so far the only) fab in 2023 with a score (20) cranes. But I could not find, in the entire Web, any information that Rapidus made or makes chips, not to mention sell them. It turns out to be true. In other words, Rapidus attempts to leapfrog from absolutely nothing directly to 2 nm, by buying ASML EUV equipment!  See (b) below.
(iii) Chitose
(A) Chitose, Hokkaido  北海道 千歳(市)
https://en.wikipedia.org/wiki/Chitose,_Hokkaido
(B) 千歳
https://ja.wikipedia.org/wiki/千歳)
("千歳（ちとせ）は、日本において「鶴は千年、亀は万年」の言い伝えにちなんだ縁起の良い言葉"_
So the term has nothing to do with emperor or aristocrats. The chi and tose are both Japanese pronunciations of kanji 千 and 歳 or 年, respectively. 歳 or 年 usually has the Japanese pronunciation toshi, with the pronunciation tose as poetic.

my translation: The lend is handed down 言い伝える that a crane lives 1,000 years and a turtle 10,000 years.


(b)
(i) Damien Chang, The Return of Japan’s Semiconductor Industry: Rapidus and the Pivot Towards an Ecosystem of Innovation. Harvard Technology Review, Apr 7, 2025.
https://harvardtechnologyreview. ... stem-of-innovation/

Quote:

(A) "Japan, like the US, thus faces significant challenges in its journey to reclaim its place at the cutting edge of the semiconductor industry. * * *

(B) "The story of Japan’s semiconductor industry is closely intertwined with its postwar economic growth miracle and its former leadership in exports of consumer electronics. From the invention of the transistor in 1947 by Bell Laboratories, the US and Japan led innovation and expansion of manufacturing capacity in the semiconductor industry through 1986, the year Japan’s market share surpassed the US for the first time. With juggernaut firms NEC, Toshiba, and Hitachi leading production, over 50 percent of the world’s chips—high-quality and low-cost—owed its origins to Japanese manufacturers, which benefited from export-driven government policy and large firm-coordination via the Very Large Scale Integrated Research Project (VLSI). The pooling of government R&D subsidies and imposition of firm reorganization—in order to compete with IBM—led to an explosion in industry-related patent applications and cross-firm research collaboration.

"US lawmakers and chip producers, however, began to sound the alarm in the 1980s and pressured Japan into signing trade agreements that protected American firms and raised the costs of Japanese imports—especially those of the latter’s dynamic random access memory (DRAM) chips, which would be crucial for the 1990s PC revolution. By 1993, the US overtook Japan’s leadership in industrial market share and regional competitors South Korea and Taiwan forged ahead innovation-wise as Japanese firms suffered from underinvestment (in the aftermath of the economic bubble burst) and inefficient vertical management structures.

(C) "While Japan still holds significant market share in some semiconductor categories, its current capabilities (40 nm fabrication process) are far from the most advanced standards (TSMC, by contrast, is producing at the 3 nm node level). * * *

(D) "Rapidus: Bold Bet or Reckless Endeavor? [which is sectional heading:]
"Based in Japan's northern island of Hokkaido, Rapidus’s aspirations are to break through to the 2 nm fabrication process by circumventing costly production advancements in fin-field effect transistor (FinFET) architecture. Because firms like TSMC, Samsung, and Intel now pursue logic chip development on an alternative architecture, known as gate all-around (GAA) technology, Rapidus believes concentrating on GAA development for its logic chips will offset its competitive disadvantage in FinFET-designed chips—which it failed to pursue previously. The GAA architecture reorients conducting channels so as to increase logic cell area density and reduce power consumption—all of which make for higher-quality chips.
* * *
"In contrast to TSMC's mass-scale industrial production model, Rapidus's approach is to serve as a foundry producing bespoke chips in small quantities—aiming at profitability by avoiding incurring the large costs of operating at scale. Skeptics reasonably question the prudence of pinning national hopes on such a high-stakes endeavor that will require—enormous levels of investment aside—a highly skilled labor force and the advanced technology and industrial processes required for vaulting from the 40 nm process (which yields general-purpose logic chips) straight to the 2 nm process (which is intended for production of specific-purpose computing such as those required by AI).

• Harvard Technology Review (or Harvard Tech Review, HTR for short) is a magazine.

Adam Zewe, Exploring Tech Intersections Student Writers Dig Deep into Innovation's Impacts in the Harvard Technology Review. John A Paulson School of Engineering, Harvard University, June 24, 2020
https://seas.harvard.edu/news/20 ... -tech-intersections
("Since its inception in January, 2019, the Harvard Technology Review has published more than 50 articles penned by student writers. The board maintains an open editorial process and invites all undergraduates to apply to write for each publication cycle")

Even so, this Harvard article is worth reading in its entirety.
• John Paulson
https://en.wikipedia.org/wiki/John_Paulson
(ii) All right, what is this Japanese company that produces 40-nm chips? It took me a couple of hours to find out. Viewing the ranking next led me to guess it was
Renesas Electronics
https://en.wikipedia.org/wiki/Renesas_Electronics
(2002- ; based in Tokyo; name origin; merger of Hitachi, Mitsubishi and later NEC (non-DRAM units only) )
and it is.
(A) Top 10 Semiconductor Companies in Japan (2025). Blackridge Research & Consulting (market research; 2021- ; based in Hyderabad, India), last updated on May 14, 2025
https://www.blackridgeresearch.c ... manufacturers-japan
("rank  Company                 2024 Revenue (USD billion)  Key Products & Focus
1          Kioxia Holdings        11.43                                NAND Flash, SSDs, BiCS 3D NAND
2        Renesas Electronics   9.63                                Automotive MCUs, SoCs, Power ICs")

MCU = microcontroller unit (the same as microcontroller)
(B) Kioxia
https://en.wikipedia.org/wiki/Kioxia
(based in Tokyo; "The company was spun off from the Toshiba conglomerate in June 2018 * ** it is currently majority owned by Bain Capital, which holds a 51.1% stake, while Toshiba holds a 30.5% stake. Hoya holds another 3% stake.[5]   Kioxia is a combination of the Japanese word kioku meaning memory [記憶, where ki and oku are both Chinese pronunciations] and the Greek word axia meaning value.[4]")
• Modern Greek-English dictionary:
* αξία (noun feminine; romanization axía): "value"
https://en.wiktionary.org/wiki/αξία
• Kioxia in Japanese is written in katakana.


(c)
(i) Hugh Grant-Chapman and Tom McGee, Japan's Chip Challenge: Semiconductor Policy for the Data Centre Era. CETAS, Alan Turing Institute, June 7, 2024.
https://cetas.turing.ac.uk/publi ... icy-data-centre-era

Quote:

(A) "Japan used to be the world's leading semiconductor manufacturer. In the late 1980s, Japan accounted for over 50% of the global semiconductor manufacturing industry, and was home to the world's three largest semiconductor firms – NEC, Toshiba and Hitachi [as well as six in the top ten (largest by revenues); they (Japanese firms) were all IDMs: both designing and making their own (memory) chips]. Today, however, Japan accounts for about 10% of semiconductor manufacturing and has no firms in the global top ten.

(B) "Trends in semiconductor end-use markets – such as cars, computers and mobile devices – have a major impact on the size and shape of the semiconductor industry. Japan learned this the hard way in the 1990s, when its firms struggled to adapt to the growth of the personal computer (PC) end-use market. The PC era led to a shift in the dominant type of semiconductor – from DRAM chips to logic chips – and the dominant business model – from integrated device manufacturers [IDMs] to fabless-foundry partnerships. In this way, the PC revolution created new winners and losers in the semiconductor industry; Japanese manufacturers, ill-positioned to react to these shifts, declined in market share. Similarly, the smartphone revolution in the 2010s benefited some countries and semiconductor firms, while others lost out.

"Today, data centres – the infrastructure that underpins artificial intelligence (AI) and performs large computational tasks – are beginning to cause profound changes in the semiconductor industry. Advances in AI technology are driving investments into data centres, with the value of the global data centre market projected to grow from $220 billion in 2022 to $418 billion in 2030. Semiconductors are the critical input for any data centre, and revenue from data centre semiconductors is projected to grow 12% per year from 2022 to 2030.

"Data centres rely on a wide variety of semiconductor inputs, but this article focuses on central processing units (CPUs), graphics processing units (GPUs) and accelerators, which together perform most data centre operations. Although these types of semiconductors are also found in some consumer electronics, the specific varieties required for data centres are larger, more expensive, and often more challenging to manufacture.

" * * * The Taiwan Semiconductor Manufacturing Company (TSMC) facility in Kumamoto, which opened in February 2024 [but started making chips on Dec 27, 2025 with 40, 28, 22, 16, and 12nm nodes]

(C) "Data centres require highly advanced semiconductors, capable of processing computationally intensive algorithms for next-generation generative AI systems. * * *

(D) " * * * At the fabrication stage, researchers are developing new technologies – such as gate-all-around (GAA) transistors [from FinFET] – to overcome physical limits to shrinking transistors, and breakthroughs in materials and equipment are also required to support advanced fabrication and packaging techniques. This could be a source of concern for Japanese firms, which typically spend less on R&D (as a proportion of operating revenue) than their international counterparts (Figure 1). * * * Today, the construction costs for a leading-edge 3-nanometer (nm) fab – a semiconductor manufacturing plant – are approximately $20 billion; a next-generation 2nm fab is expected to cost $28 billion. Capital expenditure amongst Japanese front-end manufacturers is well below international firms (Figure 2). * * *

(E) "Rapidus faces steep technological challenges. Prior to recent foreign investments [namely, TSMC in Kumamoto], Japan’s most advanced fabs could only manufacture semiconductors at the 40nm node, a technology that is more than 15 years old. Rapidus aims to catapult Japan’s semiconductor manufacturing from 40nm directly to 2nm, skipping at least eight generations of node technology in a very short time period in what would amount to an “unparalleled technological feat.” Nevertheless, Rapidus could become technologically competitive due to extensive R&D support from IBM – the first company in the world to develop the 2nm technology in 2021 – and imec – a leading R&D centre for nanoelectronics.

"The biggest challenge facing Rapidus is its commercial viability, specifically its ability to attract and retain customers. Many potential customers are hesitant to redesign their semiconductors to be compatible with different foundry processes. Rapidus is an unproven fab, and uncertainty about the yield of its manufacturing process increases the already high costs of switching manufacturing partners, dissuading prospective customers. Additionally, Rapidus’s atypical business model means that it will manufacture ‘hot lots’, small volumes of specialty semiconductors with fast cycle times. A lack of customers and small production volumes will challenge Rapidus’s ability to benefit from economies of scale and learning by doing, two factors that are essential for a fab’s commercial success.

• This report is worth reading in its entirety.
• Centre for Emerging Technology and Security (CETaS) is a research centre based at The Alan Turing Institute.11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111
• Alan Turing Institute
https://en.wikipedia.org/wiki/Alan_Turing_Institute
(based in British Library in London)


//////////////////sidebar
Semiconductor industry is roughly divided into two camps: logic chips and memory chips. Japan surpassed Intel in memory chips; the latter had produced both but abandoned memory chips to make logic chips only. Japanese semiconductor makers were never good at logic chips. When personal computers arose)also known as microcomputers back then), American makers of logic chips (also known as microprocessors) controlled almost the entire market, even for those PCs that were sold in Japan.


(1)
(a) personal computer
https://en.wikipedia.org/wiki/Personal_computer


introduction: "Since the early 1990s, Microsoft operating systems (first with MS-DOS and then with Windows) and CPUs based on Intel's x86 architecture – collectively called Wintel – have dominated the personal computer market, and today the term PC normally refers to the ubiquitous Wintel platform, or to Windows PCs in general (including those running ARM chips), to the point where software for Windows is marketed as 'for PC.'[5]


section 2 History:
section 2.2 1960s: "The personal computer was made possible by major advances in semiconductor technology. In 1959, the silicon integrated circuit (IC) chip was developed by Robert Noyce at Fairchild Semiconductor,[13] and the metal–oxide–semiconductor (MOS) transistor was developed by Mohamed Atalla and Dawon Kahng at Bell Labs.[14] The MOS integrated circuit was commercialized by RCA in 1964,[15] and then the silicon-gate MOS integrated circuit was developed by Federico Faggin at Fairchild in 1968.[16] Faggin later used silicon-gate MOS technology to develop the first single-chip microprocessor, the Intel 4004, in 1971.[17] The first microcomputers, based on microprocessors, were developed during the early 1970s.


section 2.3: "1974 saw the introduction of what is considered by many to be the first true personal computer, the Altair 8800 created by Micro Instrumentation and Telemetry Systems (MITS).[24][25]


(b) microprocessor
https://en.wikipedia.org/wiki/Microprocessor
(section 3 History, section 3.First projects: last item is "Intel 4004 (1971)" [click it and you will find this is "4-bit" in table), section 3.2  8-bit designs: The Intel 4004 was followed in 1972 by the Intel 8008, intel's first 8-bit microprocessor.[52]"/ section 3.3  12-bit designs (Intel 6100); section 3.5  32-bit designs: Intel 80386 in 1985)
(A) character (computing)
https://en.wikipedia.org/wiki/Character_(computing)
("In general, a character is a symbol (such as a letter or number) that represents information")
(B) word (computer architecture)
https://en.wikipedia.org/wiki/Word_(computer_architecture)
("After the introduction of the IBM System/360 design, which uses eight-bit characters and supports lower-case letters, the standard size of a character (or more accurately, a byte) becomes eight bits. Word sizes thereafter are naturally multiples of eight bits, with 16, 32, and 64 bits being commonly used")
(c) microprocessor chronology
https://en.wikipedia.org/wiki/Microprocessor_chronology  
(section 2 1980s: "Another change was the move to CMOS gates as the primary method of building complex CPUs. * * * CMOS greatly reduced this problem and quickly took over the market.[44] This was aided by the uptake of CMOS by Japanese firms while US firms remained on nMOS, giving the Japanese industry a major advance during the 1980s.[45]")

In the top figure of this Wiki page, one can see it was all-American.
(d) pentium by WikiChip
https://en.wikichip.org/wiki/intel/pentium
(name origin; table: Word size  32 bit' process: 800nm in 1993 (when it was launched) to 10nm in 2023 (eg, Pentium® Silver N6000 of 2021; when it was discontinued); Technology  CMOS)


(2)
(a) semiconductor industry in Japan
https://en.wikipedia.org/wiki/Semiconductor_industry_in_Japan
("In the 1970s, the Ministry of International Trade and Industry (MITI) identified semiconductors as a priority industry and orchestrated large public-private initiatives to boost competitiveness. Notably, in 1976 the government funded the Super LSI project [ja], a joint research consortium uniting major tech companies Fujitsu, Hitachi, NEC, Mitsubishi Electric, NTT, and Toshiba.[4] These rival firms collaborated at a common lab in Kanagawa to develop cutting-edge chip technologies while preventing know-how from leaking abroad.[4] Such cooperation, unprecedented at the time, led to breakthroughs – after four years, the consortium had developed advanced electron-beam lithography techniques for chip fabrication, a leap that helped Japan produce more complex semiconductors at scale.[4] By commercialising innovations in manufacturing (for example, Nikon and Canon brought lithography steppers to the market), Japan dramatically improved its semiconductor production capabilities.  As a result, Japan's share of global semiconductor sales skyrocketed from about 15% in the early 1970s to roughly 50% in the 1980s, overtaking the United States, where the semiconductor industry originated in the 1950s.[5] Japanese companies achieved several technical milestones that cemented their leadership in this era. In memory technology, they came to dominate DRAM (dynamic RAM) chips, which were crucial for computers. By the mid-1980s, Japanese DRAMs were known for high yields and reliability, and Japanese firms controlled the majority of the world DRAM market.[5] In 1987, Toshiba engineers led by Fujio Masuoka invented NAND flash memory, a new form of non-volatile storage. Toshiba publicly launched NAND flash at the 1987 IEEE Electron Devices Meeting, introducing a technology that would later enable flash drives and solid-state storage to dominate.[6] These innovations, from high-density memory chips to manufacturing equipment, gave Japan a comprehensive edge. By 1989, six of the top ten semiconductor companies in the world were Japanese.[7]")
(i) I did not click "Super LSI project [ja]," and went on my own research. "Super LSI" (without "project") led me to "Super LSI technology research association 超 LSI 技術研究組合." I returned to this en.wikipedia.org to re-read it and found "[ja]" is a hyperlink to ja.wikipedia.org for "超 LSI 技術研究組合."
(ii) "By the mid-1980s, Japanese DRAMs were known for high yields and reliability, and Japanese firms controlled the majority of the world DRAM market."

Japanese players Hitachi, NEC, and Fujitsu all focused on memory chips. This, together with a 1985 market crash of memory chips (which were DRAM only, Nand was yet to be invented) in 1985 compelled Intel to exit the DRAM market into logic chips (microcessors). Meanwhile, US government intervened, leading to the 1986 US-Japan Semiconductor Agreement. Still Japan retained dominance in Dram, until the rise of DRAM fabs in South Korea and Japan. The latter bowed out of that market due to massive losses in down years. (DRAM boom-bust cycles happened every two to three years.)
(iii) Fujio MASUOKA  舛岡 富士雄
https://en.wikipedia.org/wiki/Fujio_Masuoka

In Japan, 舛 pronounced as masu is used in name only - neither Chinese nr Japanese pronunciation. 舛 is the radical 部首 of 舞.
(b) Integrated Circuits. Semiconductor History Museum of Japan 日本半導体歴史館, undated
https://www.shmj.or.jp/english/ic.html

"1959: Invention of the planar IC [as opposed to transistor TUBE (invented in 1958)] (Robert Noyce, Fairchild Semiconductor, U.S.A.)

"Mid-1960s: Mass-production of ICs for calculators begins [with MOS ICs]

"1970s: Manufacturers in Japan enter the DRAM market and integration densities are improved: Manufacturers in Japan followed Intel into the DRAM market. In 1971, NEC developed a 1-Kbit DRAM, adopting an NMOS [N for negative] design for its high speed capability. The degree of integration of DRAM rapidly improved from 1 Kbit to 4 Kbits and then 16 Kbits. In 1976, the VLSI Technology Research Association [in Japan] was established and actively promoted various developments, including basic technology. These activities greatly contributed to the development of DRAM by Japanese manufacturers.

"1970s: Development and evolution of microprocessors[:] In the 1970s, following Intel’s lead, Japanese semiconductor manufacturers such as NEC Corporation, Toshiba Corporation, and Hitachi, Ltd. developed and mass-produced microprocessors, starting with 4-bit devices and then advancing to 8- and 16-bit devices.

"1970: 1-Kbit DRAM (Intel, USA)[:] In 1970, Intel released the world's first 1-Kbit DRAM, the 1103. The 1103 was a PMOS [p for positive] DRAM with three-transistor memory cells. DRAM rapidly replaced the core memory that had previously been the major form of computer memory.

"When the Japanese semiconductor industry reached its peak in the 1980s, DRAM memory circuits were the driving force of the world semiconductor business."

"1980s: DRAM capacity increases, the shift to CMOS advances, and Japan dominates the market