The market started worrying about the oversupply in semiconductor 2023, when the demand will start growing again depends on two factors: the situation of the macroeconomy and the inventory status.

Since foundries’ capacity utilization rates started drop in 3Q22, chip supply as a whole has decreased significantly. This, in turn, has helped limit inventory growth across the supply chain. However, the global economy is still at risk of a mild recession, so consumers may allocate more of their spending to daily necessities. They may also spend more on tourism due to easing of the pandemic. This could lead to weak sales for consumer electronic products.

Not to mention that most consumers already purchased the electronic products that they need for working or studying at home during the pandemic. Assuming that the overall inventory level of the supply chain will return to a healthier level, TrendForce believes that chip demand will begin to rebound to a certain extent in 2Q23. Then, the demand growth will become more obvious from 3Q23 onward. Nevertheless, this demand growth may not be too strong due to uncertainties in the global economy.

If we observe the situation from the perspective of the foundry industry, smartphones represent the largest application segment in terms of wafer consumption. The smartphone supply chain started inventory correction earlier, so demand rebound might be more obvious initially for smartphone-related chips compared with chips used in other consumer devices. On the other hand, with different benchmarks, the demand for HPC chips will show more significant growth compared with the demand for smartphone chips.

The U.S. Department of Commerce announced new semiconductor restrictions on October 7 in the United States. In addition to existing restrictions on the logic IC sector, this new update extends to the memory category. In addition to Chinese-funded enterprises, the extent of these restrictions stipulates foreign-owned production centers located in China will also need to apply for approval on a case-by-case basis in order to continue to obtain manufacturing-related equipment. The US ban has far-reaching effects and may extend to the global chip industry.

U.S. ban hobbles China’s semiconductor industry, affecting foundry and memory industries

The U.S. Department of Commerce announced a series of chip export control measures on the 7th, which mainly restrict China’s ability to obtain advanced computing chips, develop supercomputers, and manufacture advanced semiconductors.

However, relevant restrictions also prohibit third-country companies such as TSMC from using US-made equipment to service Chinese customers without U.S. approval in some cases. According to TrendForce, a market research agency, the ban will expand the scope of these restrictions. In the future, it will target American companies, including CPUs, GPUs, and AI accelerators, used in HPC fields such as datacenters, AI, and supercomputers. All of these items will require review before export to China. In addition, foundries may no longer be able to manufacture any of the above-mentioned HPC-related chips for any Chinese IC design house.

TrendForce believes, regardless of whether the client is a Chinese or American IC design house, most HPC-related chips are currently manufactured by TSMC with mainstream processes at the 7nm, 5nm, or certain 12nm nodes. In the future, whether the situation is American factories no longer being able to export to the Chinese market or Chinese factories being unable to initiate projects and mass produce wafer starts, it will all have a negative impact on the future purchase order status of TSMC’s 7nm and 5nm processes.

In terms of memory, according to the new specifications announced by the U.S. Department of Commerce, the DRAM portion of sanctions will be limited to the 18nm process (inclusive) and equipment must be reviewed by the Department before import. This move will greatly restrict or delay the sustainable development of China’s DRAM sector and China’s memory manufacturers will be the first to bear the brunt of these sanctions.

TrendForce indicates that CXMT possesses the largest memory market share for a Chinese company in the domestic Chinese market. Since 2Q22, the company has been committed to moving from the 19nm process into the 17nm process. Although the purchase of machinery to fulfill future needs had been accelerated before the ban, volume is still insufficient. CXMT continues to build new plants, including Phase 2 in Hefei and SMBC (SMIC Jingcheng), which is in discussion with SMIC. All of these projects will face difficulties in obtaining equipment in the future.

The C2 plant of SK hynix’s DRAM production center in Wuxi is also affected by the restriction order. The factory accounts for approximately 13% of the world’s total DRAM production capacity and its process has evolved to 1Ynm and more advanced nodes.

In terms of NAND Flash, TrendForce indicates that the import of NAND production equipment into China will be further restricted in the future, especially for equipment used in the manufacture of product of 128 layers and above (inclusive), requiring prior approval before import. It is estimated that this ban will significantly impact the long-term plans of China’s YMTC to upgrade its factory campuses, restrict YMTC from further expanding its customer base as the ban may will greatly limit non-Chinese customers’ adoption and consideration of YMTC products, and impact Samsung’s Xi’an plant and Solidigm’s process migration plan in Dalian.

U.S. temporarily exempts several suppliers as ban disrupts supply chains

In order to mitigate excessive impact of the U.S. imposed China chip ban on the semiconductor industry, the U.S. recently exempted several semiconductor companies (including in the United States, Taiwan, and South Korea) from certain restrictions.

According to Wall Street Jounal, Intel, SK Hynix, and Samsung have all received one-year exemptions. SK Hynix also issued a statement stating that the company has completed negotiations with the U.S. Department of Commerce and has obtained approval to provide equipment and items required for the development and production of DRAM semiconductors in Chinese manufacturing plants without additional licensing requirements. The authorization period is one year.

In addition, Nikkei Asia News also quoted sources as saying that TSMC has also received a one-year exemption to continue ordering U.S. chip manufacturing equipment to expand its Chinese plant. According to people familiar with the matter, the U.S. government has assured TSMC that the equipment will be shipped to its Nanjing fab, which means the company’s China’s development plan remains unchanged and is progressing smoothly.

（Image credit: iStock）

According to TrendForce, as the United States continues to expand the content of various lists, successively pass anti-China bills, and explicitly prohibit the export of certain products to China, the two countries have gradually drifted apart and this antagonistic relationship will continue if no drastic changes occur between the two parties in the next 6-8 years.

In the face of U.S. encroachment, all sectors in China must continue to look for escape routes if the country wishes to tear down the many walls built by the U.S. and move towards industrial autonomy. China’s top priority is to make breakthroughs in the semiconductor field. As far as current development is concerned, there are still many companies in China’s domestic IC design industry moving towards advanced manufacturing processes even after leading manufacturers such as Huawei, Changsha Jingjia Microelectronics, and Goke Microelectronics were placed on the entity list. At the same time, semiconductor manufacturers such as SMIC, CXMT, and Yangtze Memory Technologies have repeatedly developed advanced process technologies while Hua Hong Group has gradually expanded in the field of mature processes. If this trend continues, it will not be difficult for China to realize semiconductor autonomy in processes above 10nm.

If U.S. effectively enforces EDA ban and does not expand controls, impact on China will emerge in 2025

The U.S. Department of Commerce’s export restrictions on Chinese manufacturers are escalating but the autonomy of China’s domestic semiconductor industry is also gradually increasing. As the confrontation between the United States and China intensifies, the United States has launched a new wave of export control measures. On August 12, 2022, the U.S. Department of Commerce announced that it will restrict the export to China of EDA software required to design integrated circuits with GAAFET structure. Since GAAFET is a structure that is used in processes below 3nm, this move is equivalent to setting an advanced threshold for China’s semiconductor development.

Domestic Chinese IC designers who are committed to the development of SoCs, cloud computing chips, and GPUs are destined to move to more advanced manufacturing processes in order to meet the iterative needs of product upgrades and are expected to move toward the 4nm manufacturing process in the next 2 to 4 years. If the U.S. effectively implements the EDA software ban and does not expand the scope of EDA software restrictions, the impact of the ban on China’s semiconductor industry is expected to gradually emerge in 2025, not only delaying the development schedule of some domestic Chinese IC designers but even causing developmental stagnation.

（Image credit: Pixabay）

Although the overall economy is unstable, the use of automotive MCUs is still increasing gradually due to electric and smart vehicle trends. In order to meet market demand, IDMs have strengthened their investment in production resources. There will also be volume and price growth in 2022. Overall market size is estimated to reach US$8.58 billion, with an annual growth rate of 25.7%.

Automotive MCU market dominated by major international IDMs, 32-bit penetration rate will reach 80.1% in 2022

NXP, Renesas, and Infineon account for approximately 70% of global automotive MCU market share. In 2022, NXP will focus on the development of its S32 series and presented a S32M test chip featuring TSMC’s 5nm process, symbolizing a major milestone in the development of automotive chips. Renesas is focusing on its RH850 series, supplemented by the Low Power RL78 to stabilize development. Infineon’s automotive MCU development is focused on its AURIX series which features a self-developed TriCore core and is designed to perform mid-to-high-level automotive system control.

In general, major international IDMs have a complete line of automotive MCUs. With the increasing number of automotive functions, requirements for MCU computing power have advanced. Considering the optimization of major manufacturers’ product portfolios, the penetration rate of 32-Bit MCUs will also increase year by year and is forecast to grow to 80.1% in 2022.

Nuvoton ranks among top ten MCUs producers worldwide, Taiwanese manufacturers’ operations suffer headwinds after tide of shortages recedes

Taiwanese MCU manufacturers are represented by Nuvoton, Holtek, and Sonix. There are other manufacturers such as Generalplus, Nyquest, Hycon, and Megawin but their revenue scale is small and proportion of MCU is low. Overall, only Nuvoton is an IDM with a MCU market share ranked among the top ten in the world and readily available automotive MCU products.

Taiwanese manufacturers mainly focus on mid-to-low-end consumer electronics applications with low barriers to entry. Most of them are fabless manufacturers, meaning the barriers to entry for capital are also low. Therefore, it is difficult to compare their product portfolios with major international manufacturers. After the shortage of semiconductors subsided, operation in 1H22 inevitably encountered headwinds and demand for consumer electronics in 2H22 will continue to be weak, signaling the arrival of a cold winter for the consumer MCU market.

（Image credit: Pixabay）

Although current semiconductor process technologies have evolved to the 3nm and 5nm nodes, SoC (system on a chip) architecture has yet to be manufactured at these nodes, as memory and RF front-end chiplets are yet to reach sufficient advancements in transistor gate length and data transmission performance. Fortunately, EDA companies are now attempting to leverage heterogeneous integration packaging technologies to link the upstream and downstream semiconductor supply chains as well as various IP cores. Thanks to this effort, advanced packaging technologies, including 2.5D/3D IC and SiP, will likely continue to push the limits of Moore’s Law.

While SoC development has encountered bottlenecks, EDA tools are the key to heterogeneous integration packaging

As semiconductor process technologies continue to evolve, the gate length of transistors have also progressed from μm (micrometer) nodes to nm (nanometer) nodes. However, the more advanced process technologies are not suited for manufacturing all semiconductor components, meaning the development of SoC architectures has been limited as a result. For instance, due to physical limitations, memory products such as DRAM and SRAM are mostly manufactured at the 16nm node at the moment. In addition, RF front-end chiplets, such as modems, PA (power amplifiers), and LNA (low noise amplifiers) are also primarily manufactured at the 16nm node or other μm nodes in consideration of their required stability with respect to signal reception/transmission.

On the whole, the aforementioned memory, and other semiconductor components cannot be easily manufactured with the same process technologies as those used for high-end processors (which are manufactured at the 5nm and 3nm nodes, among others). Hence, as the current crop of SoCs is not yet manufactured with advanced processes, EDA companies including Cadence, Synopsys, and Siemens (formerly Mentor) have released their own heterogeneous integration packaging technologies, such as 2.5D/3D IC and SiP (system in package), in order to address the demand for high-end AI, SoC architecture, HPC (high performance computing), and optical communication applications.

EDA companies drive forward heterogeneous integration packaging as core packaging architecture and integrate upstream/downstream supply chain

Although the current crop of high-end semiconductor process technologies is still incapable of integrating such components as memory, RF front-end, and processors through an SoC architecture, as EDA companies continue to adopt heterogeneous integration packaging technology, advanced packaging technologies, including 2.5D/3D IC and SiP, will likely extend the developmental limitations of Moore’s Law.

Information presented during Semicon Taiwan 2021 shows that EDA companies are basing their heterogeneous integration strategies mainly on the connection between upstream and downstream parts of the semiconductor supply chain, in addition to meeting their goals through chip packaging architectures. At the moment, significant breakthroughs in packaging technology design and architecture remain unfeasible through architectural improvements exclusively. Instead, companies must integrate their upstream chip design and power output with downstream substrate signal transmission and heat dissipation, as well as other factors such as system software and use case planning. Only by integrating the above factors and performing the necessary data analysis can EDA companies gradually evolve towards an optimal packaging architecture and in turn bridge the gap of SoC architectures.

With regards to automobiles (including ICE vehicles and EVs), their autonomous driving systems, electronic systems, and infotainment systems require numerous and diverse semiconductor key components that range from high-end computing chips to mid-range and entry-level MCUs. As such, automotive chip design companies must carefully evaluate their entire supply chain in designing automotive chip packages, from upstream manufacturers to downstream suppliers of substrates and system software, while also keeping a holistic perspective of various use cases. Only by taking these factors into account will chip design companies be able to respond the demands of the market with the appropriate package architectures．

（Image credit: Pexels）