News

In 2024, the global MLCC market shows a steady growth trend, with a market size of approximately 100.6 billion yuan, a year-on-year increase of 5.0%. The booming development of artificial intelligence (AI) has driven the increasing demand for consumer electronic products such as servers, smartphones, and personal computers, which has put higher demands on the usage of MLCC. It is expected that the global MLCC market size will continue to steadily grow to approximately 105 billion yuan by 2025. By 2029, the market size is expected to reach 132.6 billion yuan, with a compound annual growth rate of 5.7% between 2024 and 2029, demonstrating the long-term growth potential of the market. Below we summarize the sales situation of major MLCC enterprises in China, Japan, South Korea, and the United States. The relevant data is evaluated based on the official sales data provided by 2024-2025 and is for reference only. Japanese company 1. Murata: As a leading enterprise in the MLCC industry in Japan and even globally, Murata is in a leading position in technological innovation and market share. The annual sales of various capacitors exceed 40 billion yuan, accounting for more than 50% of the total revenue. Among them, MLCC has the highest global market share, exceeding 30%. Murata, with its profound technical expertise, has started to produce industry-leading 0402 size and 47uF capacity multilayer ceramic capacitors. This product can be widely used in various civilian and high-performance IT equipment, meeting the urgent market demand for miniaturization and high-capacity MLCCs. 2. Solar induced electricity: MLCC's annual sales revenue is approximately 12 billion yuan. The MLCC business of solar induced electricity is gradually growing, and related products can be used in engine ECU and other powertrain systems, achieving a maximum temperature range of 150 ° C, meeting the strict requirements of automotive electronics for stable operation in high-temperature environments. 3. Kyocera: Kyocera has consistently maintained high R&D investment and production scale in its MLCC business. Its products are widely used in multiple fields, and with stable product quality and diversified product lines, it strives to maintain competitiveness in the global MLCC market, occupying a certain market share in communication, industrial control and other fields. TDK: The sales revenue of capacitors is around 12 billion yuan, and its MLCC products have long held an important position in the automotive market, with a market share second only to Murata, accounting for over 70% of the market share, and a higher proportion in higher-level applications. Faced with the rapid changes in the market, TDK actively adjusts its business layout, increases investment in research and development in the automotive market, continuously launches new products that meet the needs of the automotive market, and continues to consolidate its market position in areas such as automotive electronics and AI to cope with fierce industry competition and constantly changing market demands. South Korean company Samsung: Samsung Electric has performed outstandingly in the field of passive components (such as MLCC), with MLCC annual sales exceeding 10 billion yuan, especially occupying the top market share in the consumer electronics sector. Samsung has successfully achieved the micro high capacitance and high voltage resistance characteristics of MLCC through dielectric ceramic powder nano technology. Stable in the global MLCC market. Taiwan, China enterprise 1. Guoju: known as the world's third largest passive parts supplier, the annual sales of MLCC business is about 6 billion yuan. Guoju continues to increase its research and development innovation efforts, launching the embedded CE series MLCC, which can achieve high-frequency circuit integration and meet the demand for high-frequency and high-performance MLCC in emerging fields such as 5G communication and artificial intelligence. 2. Huaxin Technology: Its business is quite extensive, with annual sales of around 8.4 billion yuan, and MLCC is one of its main businesses. The company focuses on research and development investment and actively expands market share. In the fields of consumer electronics, communication, etc., Huaxin Technology's MLCC products have gained recognition from customers for their good cost-effectiveness and stable quality. VISHAY, an American company, focuses on the manufacturing of electronic components and has a unique technology and product layout in the MLCC product line. The annual sales of MLCC are around 3.5 billion yuan. VISHAY's MLCC products have excellent performance and can meet the diverse needs of different customers, and are widely used in industrial, automotive and other fields. With its global R&D and production system, the company is able to quickly respond to market changes and provide customers with high-quality products and services. Chinese Mainland enterprise 1. Fenghua: MLCC's annual sales volume is about 1.8 billion yuan to 2 billion yuan. It is an old brand enterprise and the first MLCC production line in China. Stack over 1000 layers on MLCC products, covering mainstream specifications ranging from 0201 to 2220 sizes. Fenghua High tech actively expands its market, increases research and development investment, improves production capacity and technological level, occupies an important position in the domestic MLCC market, and gradually expands into the international market. 2. Third Ring: MLCC's sales are on par with Fenghua's, and we continue to invest in research and development, committed to improving the performance and quality of MLCC products. Through continuous innovation, the products of Sanhuan Group have gradually gained widespread recognition in the market and their market share continues to expand. The company stands out in the competition of the domestic MLCC market, and its products are not only applied in consumer electronics, communication and other fields, but have also made certain breakthroughs in high-end fields such as automotive electronics. 3. Micro Rong: Through continuous investment in new production capacity, the current sales revenue is around 1.5 billion yuan, with an annual growth rate of over 20%, actively expanding in the automotive MLCC market. The production capacity of microcapacitors has increased from 500 billion pieces in 2018 to around 720 billion pieces today. After the completion of the new factory in 2025, the production capacity is expected to exceed 900 billion pieces, and the plan to reach 1.2 trillion pieces by 2030 is gradually being promoted in an orderly manner. 4. Yuyang: MLCC's annual sales are around 400 million yuan, focusing on the consumer market, especially the smartphone market. Continuously improving technological level, enriching product categories, and striving to enhance competitiveness in the MLCC market. Yuyang's products are applied in multiple fields, gradually expanding their market influence through continuous technological innovation and market expansion, and achieving certain technological advantages in miniaturization and high-capacity MLCC products. 5. Torch Electronics: Continuously cultivating in various capacitor businesses such as military and consumer industries, with annual sales of around 2.8 billion yuan. The company actively expands its market channels, enhances brand awareness and product reliability, and occupies an important position in the domestic MLCC market, especially in high-end fields such as military and aerospace, providing key electronic component support for the development of national defense and high-end manufacturing. 6. Hongyuan Electronics: Annual sales are around 1.8 billion yuan. Focusing on the research and development, production, and sales of MLCC products for military use, continuously optimizing product structure to meet different customer needs. Hongyuan Electronics has won a good reputation in the domestic market with its professional technology and high-quality services in the MLCC field, especially in the military industry where it has a high market share, providing strong support for the development of military electronic equipment in China. 7. Dali Kaipu: With annual sales of around 400 million yuan, it focuses on the research and development of RF microwave MLCC production. With unique technology and product advantages, it gradually stands out in the market. Dali Kaipu continues to improve product performance and market competitiveness, increase research and development investment in high-end MLCC products, strive to break the technological monopoly of international giants, actively layout in emerging fields such as 5G communication and new energy vehicles, and is expected to occupy a place in the global MLCC market. Major enterprises, relying on their technological advantages, market strategies, and innovation capabilities, are fiercely competing in the global market, jointly driving the continuous development of the MLCC industry to meet the growing market demand and emerging application scenarios.

In the past decade, NAND flash memory has been regarded as a star in the storage industry. Whether it is the popularization of smartphones, the trend of PC replacement, or the rise of cloud computing, NAND has played a crucial role. Its high growth has also driven a global semiconductor industry expansion boom, forming several oligopolistic patterns such as Samsung, SK Hynix, Micron, and Kaixia.             However, since 2024, the industry atmosphere has taken a sharp turn for the worse. NAND prices have experienced drastic fluctuations, putting pressure on corporate profits. Major manufacturers have coincidentally slowed down production expansion or even reduced investment, marking the arrival of a new cycle. At the same time, the rise of AI and high bandwidth memory (HBM) has gradually shifted the market's focus to the DRAM field, redefining the position of NAND in the storage industry landscape. The adjustments made by major giants in NAND research and development production indicate that this industry has shifted from "high-speed expansion" to "cautious investment", and a slow yet intense change is quietly taking place.     The slowdown and transfer of the two giants in South Korea     As a long-term dominant player in the global NAND market, Samsung has been known for its aggressive investments in the past. At this year's International Solid State Circuit Conference, Samsung announced the upcoming release of its V10 (10th generation) NAND flash memory. It is reported that Samsung's V10 NAND flash memory has over 400 active layers and an interface speed of 5.6 GT/s. It also adopts Samsung's exclusive and pioneering Hybrid Bonded Peripheral Unit (CoP) architecture. However, Samsung's V10 NAND mass production journey has not been smooth. It was originally expected to start mass production by the end of this year, but in June, there were reports in South Korea that Samsung was expected to conduct an evaluation of its supply chain composition before the second half of this year, and the actual large-scale production investment may not be carried out until the first half of next year. According to reports, as of June this year, Samsung has not yet finalized its supply chain for NAND core devices, including etching. The reason is that the market demand for high-level stacked NAND is unclear, and the cost-effectiveness issues brought by the introduction of new processes hinder investment promotion. The so-called "etching" refers to the process of removing unwanted substances on a wafer. In the past, when etching channel holes (similar to small holes), it was necessary to perform at a low temperature environment of about -20 ℃ to -30 ℃; But in V10 NAND, it is expected to require an ultra-low temperature environment of -60 ℃ to -70 ℃. The lower the temperature, the weaker the chemical reactivity, and the more precise etching can be achieved without the need for additional protective film. To this end, Samsung Electronics has introduced ultra-low temperature etching equipment from major front-end equipment manufacturers such as Lam Research in the United States and Tokyo Electric (TEL) in Japan, and conducted trial production and quality evaluation. However, the actual evaluation results show that ultra-low temperature etching technology is still difficult to directly apply to mass production. So Samsung is negotiating with Lam Research and TEL to try to conduct equipment evaluation again by increasing some etching temperatures. It is understood that the investment cost brought by the introduction of new devices is also considered one of the important reasons for Samsung's delay in investing in V10 NAND mass production. Samsung still mostly uses Lam Research's equipment in NAND etching processes. If TEL is included in the supply chain, it means diversification of equipment, but at the same time, it will face the problem of decreased usage of existing Lam Research equipment and the need to improve compatibility between the two devices. The difficult production of V10 NAND ultimately affected Samsung's plans for upgrading other production lines. According to South Korean media reports, Samsung Electronics has been promoting conversion investments in the Pingze P1 factory and Xi'an NAND factory since the beginning of this year, with the core being to shift the originally mass-produced 6th and 7th generation NAND to 8th and 9th generations. Compared to fully constructing a new production line, the cost of conversion investment is lower, and partial renovation and continued use of existing equipment result in higher efficiency. However, the conversion investment speed of the most advanced NAND is slowing down recently. Although the 8th generation NAND conversion of P1 factory is proceeding as planned, there have been reports of a delay in the investment for the 9th generation NAND conversion, which was originally scheduled to start as early as the second quarter of this year. The situation at the Xi'an factory is similar. The X1 production line, which is undergoing the 8th generation conversion, is nearing completion, while the X2 production line, which is undergoing the 9th generation conversion, only plans to invest in a monthly scale of 5000 wafers in the third quarter of this year, which is almost the minimum scale required for mass production of memory products. A semiconductor industry insider said, 'Samsung Electronics plans to continue mass producing older generation NAND such as V6 on its X2 production line before the first quarter of next year, and the true implementation of the 9th generation conversion will not take until at least the middle of next year,' explaining that 'this is because the demand for advanced NAND is still sluggish.'. However, Samsung remains cautious about investing in next-generation NAND and related technologies. It had originally considered applying hybrid bonding technology to V9 NAND on the Xi'an X2 production line, but has recently decided to put it on hold. Behind the delay in NAND technology is a lack of market confidence and a reconsideration of capital utilization. Compared to Samsung's high-intensity investment in HBM and DRAM, its conservative attitude in the NAND field is more prominent. Equally lacking in confidence is SK Hynix, which made it clear during the conference call that its NAND business will remain cautious and prioritize profitability. It is worth noting that SK Hynix completed the acquisition of Intel's NAND business this year and is currently restructuring its factory and related assets in Dalian, China. However, the facility investment for its second factory, which began construction about three years ago, is still on hold. Even before acquiring Intel's NAND business, SK Hynix had been discussing building a second factory in Dalian. In 2022, the company held a groundbreaking ceremony in Dalian and announced plans to continue expanding its 3D NAND production capacity in China. The industry initially expected SK Hynix to introduce infrastructure, including cleanrooms, according to its construction plan starting in mid-2023. However, due to geopolitical factors making it difficult to introduce advanced semiconductor manufacturing equipment into China, and the weak performance of the NAND market, the investment plan has been postponed. Korean media pointed out that SK Hynix Dalian Second Factory has not made any equipment investment in the past three years since its foundation. A semiconductor industry insider said, "Although SK Hynix has obtained the 'Certified End User (VEU)' qualification in the United States and investment restrictions in China have been relaxed, the NAND business still faces great uncertainty due to the economic downturn and the entry of newcomers from China. ”He added, "Therefore, there have been no discussions on the specific investment for the new factory in Dalian at present In fact, over the past few years, despite acquiring Intel's NAND business, SK Hynix has still been at a relative disadvantage in this field, and now the AI driven HBM demand has exploded, allowing it to rediscover its core growth points. At present, SK Hynix holds a leading position in the HBM field, almost monopolizing NVIDIA's AI acceleration card supply chain, and its profitability has significantly improved. In contrast, its V10 NAND development progress is relatively lagging behind, and the company has also focused its resources on advanced DRAM and HBM internally. This strategic shift clearly indicates that NAND is gradually being marginalized within SK Hynix and will not be the main investment direction, at least in the short term.   American strategic contraction, Japanese pressure, domestic breakthrough     Micron's actions also confirm the difficulties faced by the NAND industry. In August of this year, Micron announced that due to the continued weak financial performance of mobile NAND products in the market and the slowdown in growth compared to other NAND opportunities, Micron will cease the development of future mobile NAND products globally, including terminating the development of UFS5 (fifth generation universal flash storage). Micron pointed out that this decision only affects global mobile NAND product development work. Micron will continue to develop and support other NAND solutions, including SSD, automotive, and other terminal NAND solutions. At the same time, Micron will continue to support the mobile DRAM market and provide DRAM product combinations. Faced with profit pressure, Micron directly withdrew from competition with Samsung and SK Hynix in the consumer market. The company will shift its focus to the enterprise SSD, automotive, and data center markets, which, although not as large as smartphones, have more stable demand and higher profit margins. At the same time, Micron has significantly increased its research and development, as well as capacity building, in HBM and DRAM, in an attempt to carve out a bigger share of the pie in the AI era. According to its financial report, since the second half of 2024, Micron has continuously raised its revenue and gross profit margin guidance, resulting in significantly better market performance than its NAND business. This differentiation pattern of "NAND sluggish, DRAM strong" reflects Micron's strategy of shrinking NAND to free up resources for future AI memory. Compared to Samsung and Hynix, the situation of Armor Hero is even more difficult. As the world's third-largest NAND supplier, Kaixia has long relied on technology and production capacity cooperation with Western Digital. However, the merger between the two companies, which has been brewing for many years, has yet to materialize, making it difficult for Kaixia to achieve economies of scale that can compete with Korean giants. The lack of scale advantage means that Kaixia is in a passive position in cost control, technological iteration, and market discourse power. At the same time, the drastic fluctuations in NAND prices directly dragged down Kaixia's financial performance, causing its performance to hover on the edge of profit and loss for years. Faced with the pressure of the capital market, Kaixia needs to maintain basic R&D and production investment to maintain competitiveness, but lacks sufficient financial flexibility to promote larger scale expansion or technological upgrades, falling into the dilemma of "unable to advance and unable to retreat". On the other hand, China's Yangtze River storage has chosen to increase investment against the trend and embarked on a differentiated development path. With the support of domestic market demand, Changjiang Storage has been able to maintain its expansion momentum during the global industry's widespread "scale reduction" phase, winning a strategic window for future competition. However, from a global perspective, the NAND market is still dominated by oligopolies, with established manufacturers such as Samsung, SK Hynix, Micron, and Kaixia still firmly controlling most of their production capacity and channel resources. Even though the Yangtze River storage is rapidly catching up, it is difficult to completely shake this structure in the short term. The overall atmosphere remains cautious, with companies placing greater emphasis on risk control and capital returns in investment and expansion, awaiting the arrival of the next growth cycle with clear demand.     The 'cold wave' of equipment manufacturers     The most direct victims of the slowdown in NAND investment are undoubtedly semiconductor equipment manufacturers. In the past few years, local equipment companies in South Korea, such as SEMES (a subsidiary of Samsung), Jusung Engineering, Dongjin Semichem, etc., have relied heavily on Samsung and Hynix's expansion projects to maintain growth. Now, with the two giants delaying the mass production plans of advanced NAND projects, related equipment orders have significantly declined, and cash flow and profitability have been severely impacted. Global device giants have not been spared either. ASML's EUV lithography machines mainly serve the logic and DRAM processes, and their DUV equipment shipments are under pressure during low NAND demand; TEPCO Electronics (TEL) has a high dependence on NAND business in the fields of thin film deposition and etching, and delayed orders directly affect its performance; Applied Materials and Lam Research have felt the most obvious impact in the deposition and etching processes, especially in the demand for ALD, CVD, and high aspect ratio etching equipment for high-level NAND. This has forced both companies to lower their shipment expectations. Kokusai Electric、 Japanese companies such as Hitachi High Tech and SCREEN Semiconductor have also been hit by chain reactions. Faced with this dilemma, equipment manufacturers have to accelerate their "pivot". With the explosive demand for AI driven HBM and high-performance DRAM, as well as the continuous increase in advanced logic processes (3nm/2nm), equipment giants are gradually shifting Jusung Engineering's focus from NAND to logic chips and HBM/DRAM related devices to fill the NAND investment gap. Dongdian Electronics has repeatedly emphasized the stability of its DRAM business in its financial report, while Kelei and Applied Materials are actively promoting the layout of advanced logic and wafer level packaging equipment. At the same time, the popularization of the "conversion investment" model is also changing the industry ecology. More and more wafer fabs are choosing to purchase second-hand equipment and carry out remanufacturing or renovation to extend the service life of the equipment, in order to reduce overall capital expenditures. This trend has driven a rapid rise in the second-hand equipment market, leading to business growth for companies such as Surplus GLOBAL, AG Semiconductor, MTM, and others. However, this has further compressed the market demand for new devices, causing equipment manufacturers that previously relied on large-scale NAND expansion to feel an unprecedented chill in the short term.     NAND， How to break the deadlock     The NAND market entering a downturn is not due to a single reason, but a combination of multiple factors: On the one hand, the global shipment growth of smartphones is sluggish, the PC replacement cycle is prolonged, and the demand for traditional mobile NAND is under long-term pressure. On the other hand, the explosive growth in demand for HBM and DDR5 driven by AI has tilted capital and R&D manpower towards these emerging fields, and NAND is gradually being marginalized. In addition, NAND is also facing capital pressure. Over the past three years, the total investment in the semiconductor industry has reached a historic high, forcing manufacturers to control costs and invest limited resources in businesses with faster returns. Finally, we cannot ignore the technological challenges. Currently, the number of NAND stack layers is approaching the limit of 400, and process difficulty, yield issues, and cost control have become bottlenecks, greatly increasing investment risks. In the foreseeable future, the NAND market will continue to maintain a low speed, investment will be cautious, and manufacturers will allocate more resources to DRAM and HBM. However, in the medium term, with the popularity of AI training, edge computing, high-capacity SSDs and other scenarios, NAND may rediscover growth opportunities. For example, in the storage systems required for future generative AI, high-capacity and low-cost NAND may have unique value. Of course, NAND manufacturers are also looking for breakthrough opportunities like HBM, such as the recent collaboration between Sandisk and SK Hynix aimed at standardizing "high bandwidth flash" (HBF). HBF is a memory technology based on NAND flash memory, built into a package similar to HBM. This marks the first solid step in the industry towards integrating flash memory and DRAM like bandwidth into a single stack, with the potential to fundamentally change the way AI models access and process data on a large scale. But in the long run, the dual breakthroughs in technology and market will truly determine whether the NAND industry can reach the next turning point. On the one hand, new processes such as hybrid bonding and>400 layer stacking need to mature in order to achieve a qualitative leap; On the other hand, new application bursts must emerge on the demand side in order to free NAND from the fate of marginalization. Otherwise, this former 'star storage' is likely to continue hovering on the edge of the industry.

On August 19th, Intel Corporation and Japan's SoftBank Group announced that SoftBank will invest $2 billion (approximately 14.376 billion yuan) in Intel! According to the agreement, SoftBank will purchase Intel's common stock at a price of $23 per share. Affected by this news, Intel's stock rose 4% in after hours trading. This investment is seen as an important vote of confidence in Intel. In recent years, Intel has failed to fully seize the opportunities brought by the artificial intelligence boom in the advanced semiconductor field, resulting in poor stock price performance. In 2024, Intel's stock price fell by 60%, marking the worst annual performance since the company went public more than half a century ago. However, as of the close of this Monday, Intel's stock price has risen by 18% in 2025. Intel, as the only company in the United States capable of producing the most advanced chips, holds a crucial position in the US semiconductor supply chain. Recently, Intel has become a focus of discussion in the Washington political arena as the company is seen as a key chip supplier to the United States. However, Intel's foundry chip manufacturing business has not yet received significant customer orders, which is crucial for the stability and expansion of its business. Last month, Intel stated that it would further invest in its foundry business after receiving customer orders. At the industry level, the arrival of this funding coincides with Intel's critical transformation period. According to the second quarter financial report, Intel achieved revenue of $12.9 billion, which was basically the same as the same period last year, but suffered a net loss of approximately $2.9 billion due to restructuring, impairment, and one-time costs. To alleviate pressure, the company's management is reducing operating expenses while continuously promoting organizational streamlining, and is shifting space for wafer foundry manufacturing, chip products, and AI routes through a "cost reduction+focus" approach. Currently, the progress of capital expenditures, customer commitments, and manufacturing outsourcing business has become an important signal for the outside world to observe Intel's future trends. However, in the fiercely competitive semiconductor industry, relying solely on financial subsidies or cost control is not enough to reverse the situation. What is more crucial is technological iteration and gaining more customers. Especially Intel's decision to continue investing in its wafer foundry business (IFS) is crucial in identifying key customers and long-term foundry orders. This is related to whether the utilization rate and gross profit margin of the production line can be improved, thereby establishing a sustainable business path and forming a positive cycle for the subsequent research and mass production of nodes such as 14A/18A. SoftBank's investment has also sparked speculation in the market about the potential synergy between Arm, a subsidiary of SoftBank, and Intel. Intel is promoting its foundry business, hoping to provide wafer manufacturing services to more chip customers. If the two can collaborate, Arm's IP ecosystem and Intel's manufacturing capabilities may complement each other.

On the 15th, US President Trump stated that he will impose semiconductor tariffs within the next two weeks, with rates possibly as high as 200% or even 300%, symbolizing his readiness to intensify efforts to force chip manufacturing to return to the United States. According to reports, while flying to Alaska to meet with Russian President Putin, Trump said on Air Force One, "I will impose tariffs on steel and chips next week and the following week." Trump has repeatedly promised that tariffs on chips and drugs will be implemented within a few weeks, but has not yet announced them. The US Department of Commerce has launched an investigation into the chip and pharmaceutical industries since April, which is a prerequisite for Trump's imposition of tariffs on national security grounds. This program can be quite complex, and investigations often take months or even longer to complete. Manufacturers and artificial intelligence (AI) companies have been eager for clearer semiconductor tariff plans as chips are widely used in various modern consumer products. Last week, Trump announced at an event with Apple CEO Cook that he plans to impose a 100% tariff on semiconductors, but if manufacturers relocate production to the United States, their products will not be affected. The White House has not further explained how the exemption mechanism operates, but Trump has hinted that Apple, which has committed to a $600 billion Made in America program, may receive an exemption. On the 15th, Trump pointed out that semiconductor tariffs will initially be set at a lower level, allowing companies time to set up factories and build production capacity in the United States. After a period of time, the tariff rate will be significantly increased, possibly reaching 200% or 300%. Trump said in an uncertain tone, 'Will the tax rate I want to set be 200% or 300%?' He said he is confident that companies will set up factories in the United States instead of choosing to pay high tariffs.

In response to the recent layoffs in China by Micron, Micron has officially responded to the CFM flash memory market:   Given the continued weak financial performance of mobile NAND products in the market and the slower growth compared to other NAND opportunities, we will cease the development of future mobile NAND products globally, including terminating the development of UFS5.   Micron stated that this decision only affects the development of global mobile NAND products. Micron will continue to develop and support other NAND solutions, such as SSDs and NAND solutions for the automotive and other end markets.   Micron will continue to develop and support the mobile DRAM market globally, and provide an industry-leading DRAM product portfolio. In recent years, as large technology companies have increased their investment in artificial intelligence data centers, orders for high bandwidth memory (HBM) chips from semiconductor manufacturers such as Micron have surged due to their powerful data processing capabilities.   The company currently expects quarterly revenue of $11.2 billion, with a fluctuation of $100 million, compared to the previous forecast of $10.7 billion, with a fluctuation of $300 million.   Micron also raised its adjusted gross profit margin forecast for the fourth quarter to 44.5%, with a fluctuation of 0.5%, compared to the previous forecast of 42%, with a fluctuation of 1%.   The company stated that the revised forecast reflects an improvement in pricing, particularly for dynamic random access memory or DRAM products.   Sumit Sadana, Chief Commercial Officer of Micron, said at an industry conference on Monday, "We focus on all different end markets around the world, and price trends have been strong. We have achieved great success in raising prices   EMarketer analyst Jacob Bourne stated that supply constraints in HBM production and strong demand for artificial intelligence have enabled Micron to set higher prices for its products, representing a shift in memory chip manufacturers' history towards thinner profit margins.   In June of this year, Micron announced that it would increase its investment in the United States by $30 billion, bringing the total to $200 billion

US President Trump has stated that the CEO of Intel has a serious conflict of interest and must resign immediately. There is no other solution to this problem.   This statement also cast a heavy shadow over the future direction of Intel, which is currently undergoing changes, directly leading to a nearly 5% drop in Intel's stock price in pre-market trading.   Futurum Group's Chief Semiconductor Analyst Ray Wang said, "This is a crisis point for how Intel will handle its relationship with the US government in the future." He warned that Trump's interference could make the situation worse. Phil Blancato, CEO of Ladenburg Thalmann Asset Management, a financial company, expressed concern that the incident could set a bad precedent, saying, 'You don't want the President of the United States to decide who will manage the company.'. According to the data, Chen Liwu was born in Malaysia in 1959, grew up and received education in Singapore, graduated from Nanyang University with a Bachelor's degree in Physics, and then majored in Nuclear Engineering at the Massachusetts Institute of Technology in the United States, obtaining a Master's and Doctoral degree. He later founded the venture capital company Walden International.   In 2017, the data analysis company Relationship Science selected Chen Liwu as one of the most connected executives in the technology industry and gave him a perfect score of 100 in the "Power Rating"   65 year old Chen Liwu took over as CEO of Intel from his predecessor Pat Gelsinger in March this year, and he is also the first Chinese CEO in Intel's history. After taking office, he proposed a series of reform measures, hoping to once again turn Intel into a customer-centric and engineering centric company. In order to achieve this goal, Chen Liwu streamlined about 50% of the management hierarchy, making management more flat, while divesting non core businesses and initiating layoffs, thus focusing resources more on core businesses. Including plans to reduce the company's workforce to 75000 by the end of the year, a decrease of approximately 22%. In addition, Intel has promised to adopt a more cautious strategy in manufacturing investments. Intel once dominated the field of chip manufacturing, but in recent years it has lagged behind its Taiwan, China rival TSMC in manufacturing technology, and has almost no share in the AI chip market, which is currently dominated by Nvidia. When Chen Liwu took over Intel, the industry generally believed that with his abilities, he could save Intel from the current "crisis". At present, Chen Liwu has only been serving as CEO of Intel for less than 5 months, and the reform of Intel is still actively underway. If Chen Liwu is forced to resign at this time, it will obviously be extremely unfavorable for Intel's future development. At the same time, this controversy undoubtedly adds new uncertainty to Intel's ongoing strategic transformation. In response to this, Intel issued a statement rejecting Trump's call for CEO Chen Liwu to resign and promising to make "significant investments" in the president's "America First" agenda.   Intel, the Board of Directors, and Chen Liwu are firmly committed to advancing the national and economic security interests of the United States and making significant investments in accordance with the President's America First Agenda, "Intel said in a statement released on Thursday Intel has been manufacturing in the United States for 56 years. We will continue to invest billions of dollars in domestic semiconductor research and manufacturing, including our new factory in Arizona, which will adopt the most advanced manufacturing process technology in the United States, and we are the only company to invest in leading logic process node development in the United States. We look forward to continuing to work with the government   In addition, regarding Trump's request for Intel CEO Chen Liwu to resign, Stacy Rasgon, a senior analyst at Bernstein Research, commented, "Chen Liwu's activities in China are not secret. He is a legendary figure in the semiconductor industry." "Intel is in a difficult situation, and when he accepted this job, they were in a difficult position, which is part of the reason why he accepted it. And if he turns losses into profits, it will be a difficult process Bernstein analyst Stacy Rasgon believes that Trump's dissatisfaction may stem from Chen Liwu's failure to establish a personal relationship with him, as well as Trump's possible dissatisfaction with Intel's cuts in capital expenditures and wavering in cutting-edge processor manufacturing. Intel is a significant beneficiary of the Chips and Science Act funding, having received nearly $8 billion in grants. However, the Trump administration has been trying to use these grants to encourage businesses to commit to more investment. In addition, Intel announced this year that it will postpone the production of its chip manufacturing plant in Ohio until at least the 2030s, which may contradict the government's goal of expanding semiconductor production capacity in the United States. Trump's close ally, Ohio Republican Senator Bernie Moreno, also called for Chen Liwu to resign. Chen Liwu is not the first corporate leader to resign after a conflict with Trump. One important reason for Chen Liwu's employment is his profound industry expertise and network. If he resigns, it is currently unclear who can replace him.

TSMC has ambitious plans in the United States, one of the most important of which is to produce cutting-edge N2 technology to help the country take a leading position in the chip industry. TSMC's cutting-edge chip technology is entering the United States and is currently unstoppable. Since the Trump administration took office, TSMC has seen great interest in the region, partly due to the "tariff threat" faced by this chip giant.   According to reports, TSMC is currently preparing a 2-nanometer production line at its Fab P3 factory in Arizona. It is reported that the plant is under construction and may be put into production as early as 2026, which is nearly a year later than the production line in Taiwan, China. Establishing chip factories in the United States has clearly become a trend among companies, mainly because the US government is making the bill a part of tariff agreements with countries such as South Korea. For TSMC, the company's start was indeed difficult, especially with many issues in culture and logistics. However, now, with the support of increased investment, this chip company seems to be more determined in its ambitions in the US market. TSMC is indeed aggressively attacking the US chip supply chain, indicating its desire to dominate this market. Due to the need for large technology companies to maintain close contact with the current US government, NVIDIA、 Companies such as Apple and AMD are ready to transfer their supply chains to the United States through hundreds of billions of dollars in investments, ranging from establishing production facilities to assisting supply chain partners in the transition. Since all these companies currently rely on the chips of TSMC, it is crucial for the Taiwan, China giant to establish a strong chip network in the United States. By transferring the 2-nanometer process to the United States, TSMC has indeed chosen its next business location. It is expected that in a few years, the domestic demand for chips in the United States will increase significantly, and this achievement is mainly attributed to the current government's prioritized self-reliance policy. The future development of TSMC relations in the United States is worth paying attention to, but currently, everything depends on the United States

On July 28th, according to the latest report from market research firm Counterpoint Research, the global pure semiconductor wafer foundry industry's revenue is expected to reach $165 billion by 2025, a year-on-year increase of 17%. During the period of 2021-2025, the compound annual growth rate of this industry is 12%. This growth is mainly due to the promotion of advanced process nodes. Among them, the revenue of 3 nanometer nodes is expected to increase by over 600% year-on-year, reaching $30 billion, while the revenue of 5/4 nanometer nodes will exceed $40 billion. These advanced nodes will contribute more than half of the total revenue of pure wafer fabs by 2025. The report points out that the increasing demand for high-end smartphones, AI PC solutions, AI ASICs, GPUs, and high-performance computing (HPC) solutions is the main factor driving the growth of advanced process revenue. In terms of corporate competition, TSMC has an advantage in advanced nodes, followed closely by Samsung and Intel. At the same time, UMC, Gexin, and SMIC still have strong demand in other nodes, although they may not be able to keep up with the pace of advanced nodes in terms of revenue growth rate. In addition, the backend packaging process is constantly innovating and generating revenue. For example, technologies such as HBM memory integration and migration to chip level packaging are bringing new growth opportunities to the industry. These innovations not only enhance the performance and reliability of products, but also open up new sources of revenue for semiconductor foundries.

According to SK Hynix's latest financial report, the operating profit for the second quarter was KRW 9.21 trillion, a year-on-year increase of 68.5%, exceeding analysts' expectations of KRW 8.93 trillion; Revenue was KRW 22.23 trillion, a year-on-year increase of 35.4%, exceeding analysts' expectations of KRW 20.56 trillion. In addition, on a quarterly basis, the second quarter revenue increased by 26% compared to the previous quarter, and the operating profit increased by 24%. SK Hynix stated, "As major global technology companies actively invest in the field of artificial intelligence (AI), the demand for AI oriented memory continues to grow. The company's DRAM and NAND flash memory shipments have exceeded expectations, resulting in the highest historical performance. ”The company also stated that it has expanded the sales of 12 layer HBM3E products in the DRAM business field, and increased sales in various application areas of NAND flash memory. With the highest level of competitiveness in the AI oriented storage industry and profit oriented business activities, it has continued its good performance trend. ” Thanks to this performance achievement, as of the end of the second quarter, cash and cash equivalents reached KRW 17 trillion, an increase of KRW 2.7 trillion compared to the previous quarter. The debt to net debt ratio was 25% and 6% respectively, with a significant decrease of KRW 4.1 trillion in net debt compared to the previous period. In addition, while customers increased their storage orders in the second quarter, they also increased their finished product production and maintained a stable inventory level. SK Hynix predicts that new products from customers will be launched soon in the second half of the year, and storage demand will continue to grow. SK Hynix will double the year-on-year growth in the HBM field based on its existing HBM3E product performance and mass production capabilities, thereby achieving stable performance results. HBM4 also plans to prepare timely supply according to customer requirements and continuously strengthen its competitiveness at the highest level in the industry.

The ranking of the top 50 semiconductor suppliers by 2024 in the McLean Report is shown in Figure 1. This ranking is based on calendar year sales and covers integrated circuit (IC) and O-S-D (optoelectronics, sensors/actuators, and discrete devices) equipment. If the company's fiscal year does not match the calendar year, the sales amount has been adjusted to January to December. The list includes 19 US headquarters enterprises, 11 Japanese enterprises, 7 suppliers in Taiwan, China, 6 European enterprises, 4 suppliers in Chinese Mainland and 3 Korean enterprises.   picture Figure 1 Top 50 Semiconductor Sales Leaders by 2024 (including foundries, in millions of US dollars) In 2024, the total semiconductor sales of the top 50 suppliers will increase by 26%, which is 4 percentage points higher than the 22% growth in the global semiconductor market. Winbond, Powerchip and Vanguard returned to the top 50 list, and CXMT, a DRAM supplier from Chinese Mainland, performed amazingly when it was first shortlisted, ranking 34th.   Overall, 44 out of the top 50 companies have experienced changes in their rankings. The largest increase was seen by fabless IC supplier Realtek, which jumped 10 places to 29th place thanks to its strong sales of connected media and network ICs. In addition, Monolith Power Systems rose 5 places to 37th place due to a surge in sales of power management ICs for AI and server applications.   The company with the most significant decline in ranking in 2024 is Sharp, which dropped 10 places to 43rd place; Microchip has slipped from 19th place in 2023 to 25th place in 2024.   The ranking of the McLean Report covers integrated device manufacturers (IDMs), fabless companies, and foundries, but does not include semiconductor sales data from system vendors such as Apple, Amazon, Google, Meta, etc. These companies design their own chips and specialize in their own systems, without selling ICs to the public market.   The top 50 rankings in Figure 1 also include 8 pure OEM factories. The McLean Report includes foundries in the top 50 semiconductor suppliers ranking, as it consistently considers this ranking as a "top supplier list" rather than a "market share ranking," and explicitly points out that some semiconductor sales may be subject to double counting.   Figure 2 Top 50 semiconductor suppliers excluding foundries The following companies will replace the 8 pure OEM factories:   US and Taiwan Semiconductor (Dior), $1.31 billion   Taiwan, China Nanya, US $1.06 billion Socionext, Japan, $1.03 billion Chinese Mainland GigaDevice, 1.02 billion US dollars Nuvoton, Taiwan, China, US $981 million Semtech, USA, $890 million Sanken Electronics, Japan, 832 million US dollars Taiwan, China based Macronex, 806 million US dollars picture Figure 2 Top 50 Semiconductor Sales Leaders by 2024 (excluding foundries, in millions of US dollars)   As shown in Figure 3, in 2024, US headquarters companies accounted for 57% of the top 50 semiconductor suppliers' sales excluding foundries, with Korean companies ranking second with a 21% share. The reason why the Korean market share is particularly impressive is that its 21% share is contributed by only three companies - Samsung, SK hynix, and LX Semicon. It is worth noting that, if the pure OEM factory is not included, enterprises in Taiwan, China account for 5% of the market share; However, if the sales of pure OEM factories are included, the proportion of Taiwan, China in the top 50 suppliers will jump to 17%.   picture Figure 3 Top 50 semiconductor suppliers by headquarters location before 2024 (excluding foundries) The market share of South Korea is particularly likely to change year by year due to the cyclical fluctuations in the storage market. In 2024, South Korea will occupy 21% of the market share, an increase of 5 percentage points from 16% in 2023, thanks to the strong growth of DRAM sales of 88% and the growth of NAND flash memory market of 69%.   Supplier share of semiconductor market in 2024 Semiconductor sales are dominated by a few large enterprises, and this trend is more evident in 2024 than ever before. Excluding foundries, the top 10 leading companies in 2024 account for two-thirds of semiconductor sales, an increase of nearly 20 percentage points compared to the top 10 companies in 2010 (Figure 4). The 50 largest suppliers set a record, accounting for 92% of global semiconductor sales.   picture Figure 4 Supplier share of semiconductor market in 2024 (excluding foundries, totaling $676.6 billion) picture Top 50 companies ranked by growth rate The sales growth rate of the top 50 suppliers spans 260 percentage points, from a 214% increase in CXMT to a 46% decline in Sharp (Figure 5). Among the top 50 companies, 23 achieved sales growth, with 17 experiencing double-digit growth. Meanwhile, in 2024, 27 companies experienced a decline in sales, ranging from 2% for Intel and Powerchip to 46% for Sharp.   picture Figure 5 Leading semiconductor sales companies ranked by growth rate in 2024 Leading semiconductor sales companies in the first quarter of 2025 and outlook for the second quarter The ranking of leading semiconductor sales companies in the first quarter of 2025 in the McLean Report is shown in Figure 6. When necessary, sales have been accounted for based on the first quarter of the calendar year (January to March).   picture Leading semiconductor sales companies in the first quarter of 2025 and outlook for the second quarter   As expected, NVIDIA easily maintained its position as the world's largest semiconductor supplier in the first quarter of 2025, with sales increasing by 23% month on month and 89% year-on-year. Although the company and its products have become targets of the US government's tariff policies towards China, demand for its data center artificial intelligence processors remains strong. CEO Huang Renxun bluntly stated that the sales restrictions on Nvidia processors in the United States have failed to achieve their goals and have caused more harm than good to the US semiconductor industry. He estimates that US export controls have caused the company to lose at least $15 billion in sales revenue, which was originally intended for the research and development of next-generation artificial intelligence processors. Even if the rapid growth of sales this year slows down, Nvidia may still remain at the top of the list for the rest of 2025.   In the first quarter of 2025, only Nvidia and MediaTek, ranked 10th among the top 10 suppliers, achieved month on month revenue growth. MediaTek stated that the sales growth is mainly due to increased market demand and increased customer adoption of artificial intelligence, 5G, and Wi Fi 7 technologies.   Broadcom, ranked 8th, had flat sales in the first quarter, but all other top 10 companies experienced a decline in sales, ranging from AMD's 1% decline to Samsung's 20% decline. Sales of Samsung and SK Hynix have declined due to weak seasonal demand and easing price pressure on DRAM and NAND flash memory devices.   Similarly, Intel's 11% revenue decline is partly attributed to seasonal weakness, but it also indicates that Intel still needs a lot of work to reshape consumer and original equipment manufacturer (OEM) confidence in its current (and future) product lineup and roadmap. Intel is counting on its industry-leading 18A process to provide a much-needed boost to its profitability in the second half of 2025, while continuing to seek new external customers to drive the development of its foundry business.   Overall, the ranking of the top 10 companies has not changed compared to the previous quarter, but there have been some adjustments in the downstream rankings of the list. Despite a 9% decrease in sales, NXP's ranking still rose two places to 13th place. STMicroelectronics (ST) fell from 13th to 16th place after a 24% decrease in revenue. ST stated that the delayed recovery of industrial applications and inventory adjustments, as well as the slowdown in automotive IC sales (especially in Europe), led to weak sales in the first quarter. Kioxia's sales decreased by 35%, dropping three places to 19th place in the first quarter ranking.   Among the top 25 companies, 8 achieved quarter on quarter sales growth in the first quarter of 2025, 14 suppliers experienced a decline in sales, and Broadcom's sales remained stable.   After showing strong quarterly revenue performance at the end of 2024, most storage vendors on the list experienced a general decline in sales in the first quarter. Samsung, SK Hynix, SanDisk, and Kaixia all experienced double-digit revenue declines in the first quarter of 2025, while Micron's sales declined by 2%.   After experiencing a difficult 2024, some microcontroller (MCU) and analog device suppliers can claim that the darkest days are over. Texas Instruments (TI), Analog Devices, and Renesas all achieved a month on month growth in modem sales in the first quarter of 2025. In addition, TI and Adeno expect to continue achieving single digit growth in the second quarter.   19 out of the top 25 suppliers have provided revenue guidance for the second quarter of 2025. Nvidia (9%), Micron (9%), STMicroelectronics (8%), TSMC (7%), and Texas Instruments (7%) are expected to see a significant rebound in sales in the second quarter. However, the average guidance of the top 25 companies shows a 3% month on month increase in sales in the second quarter.   The companies providing Q2 revenue guidance have been cautious in their outlook, acknowledging market uncertainty, potential impact of tariffs, ongoing regional conflicts, high interest rates, and other geopolitical tensions that may lead to sudden changes in forecasts.   Sales in the first quarter decreased by 2% compared to the previous quarter, and with an average quarterly guidance of 3% in the second quarter, this further supports TechInsights' semiconductor market forecast that annual growth will slow down due to the "mild impact" of tariffs and trade restrictions policies.

In the eyes of most people, the global frenzy over artificial intelligence sovereignty has led to competition for GPUs, which in turn has driven demand for HBM. In a report released in March this year, renowned analysis firm Yole stated that since the emergence of ChatGPT at the end of 2022, generative artificial intelligence has flourished, driving an unprecedented 187% year-on-year increase in HBM bit shipments in 2023, and a surge of 193% in 2024.   It is expected that this growth momentum will continue. The growth rate of HBM far exceeds the overall DRAM market. Global HBM revenue is expected to grow from $17 billion in 2024 to $98 billion in 2030, with a compound annual growth rate of 33%. ”Yole continued in the report. The latest revenue data from the three major storage giants shows that HBM is indeed on the track predicted by Yole. As a new DRAM leader, driven by the surge in demand for high bandwidth memory (HBM), SK Hynix expects its operating profit to reach nearly 9 trillion Korean won (6.6 billion US dollars) in the second quarter, and HBM sales are expected to account for over 50% of SK Hynix's total DRAM revenue this year, up from over 40% in the fourth quarter of 2024. Another HBM supplier, Micron, also achieved a new high in performance driven by HBM. But recently, analysis agencies have issued a warning to HBM. Goldman Sachs: HBM will experience a significant drop According to a Goldman Sachs report cited by Taiwanese media, intensified competition and oversupply may lead to the first decline in HBM prices in 2026, posing a challenge to market leader SK Hynix. Goldman Sachs pointed out that HBM prices may experience a double-digit decline by 2026. Goldman Sachs warns that intensified pressure, intensified competition, and a shift in pricing power to major clients (in which SK Hynix has significant risk exposure) may squeeze the company's profit margins. In Goldman Sachs' view, the downward trend in HBM prices may be attributed to a significant increase in HBM chip supply from major manufacturers, which is expected to exceed demand and potentially push down the annual average selling price (ASP). Goldman Sachs stated that after years of tight supply, the HBM market is expected to experience weakness in 2026, which could lead to increased pricing pressure across the industry. Goldman Sachs also emphasized that NVIDIA's next-generation GPU Rubin will not increase its HBM capacity like the B300. Both GPUs will adopt a capacity of 288GB - Rubin will use the 12Hi HBM4, and B300 will use the 12Hi HBM3E. This means that the demand growth for GPU driven HBM is limited, which is not good news for NVIDIA's main HBM supplier SK Hynix. At the same time, Goldman Sachs also predicts that HBM's growth will significantly slow down - currently expected to increase by 25% year-on-year, compared to 45% previously. Goldman Sachs has revised its HBM Total Target Market (TAM) forecast, slightly increasing its 2025 forecast by 1% to $36 billion, but lowering its 2026 forecast by 13% to $45 billion (previously $51 billion). Korean analysts warn that SK Hynix's market share may shrink when the next generation HBM4 is launched in 2025. In addition, the report points out that although the lifting of export restrictions on NVIDIA H20 chips by the United States to China should boost HBM demand and thus help SK Hynix, it may also provide a boost to its competitors. The report also cites analysts' warnings that Samsung's HBM shipments may grow at an annual rate of 20% by 2026, which could directly put pressure on SK Hynix's profit margins. According to them, Chinese companies will also become new players in this market, bringing uncertainty to the HBM market. However, Jibang Consulting believes that with the continuous increase of HBM production capacity and the steady improvement of yield rates of various suppliers, the possibility of price reduction for mature products is unlikely. However, next year's focus will be on HBM4, which is still undergoing certification, so it is too early to determine the winner of the competition. Considering the release of the next generation HBM, TrendForce expects the overall average price of HBM to continue to rise. UBS: HBM sees breakthrough However, while Goldman Sachs is declining, UBS is highly optimistic about HBM in its latest report. UBS analysts stated in a recent report that as the demand for computing from artificial intelligence continues to reshape the memory landscape, high bandwidth memory (HBM) is expected to see a breakthrough year in 2026. Analysts point out that our channel research continues to indicate that SK Hynix is likely to gain a stable market share in the HBM market by 2026, accounting for approximately 50% of the total capacity. ”This emphasizes their expectation that Hynix will continue to control the next generation of memory, even as contract negotiations and competitors' ambitions continue to escalate. Despite some noise in the short-term memory market, particularly regarding negotiations between NVIDIA and high bandwidth memory suppliers such as SK Hynix, Samsung, and Micron Technologies, as well as the possibility of a "price correction" as Samsung approaches HBM3E certification, UBS believes that the real story is brewing. UBS reiterates. Hynix's leadership position in Nvidia will be maintained, and the recent design victories won by Google, AWS, and Microsoft's ASIC divisions indicate that Hynix will be locked in as the main or only HBM supplier in the industry. Even though competition is expected to intensify, it is unlikely to become a serious obstacle for Hynix before the end of 2026. In terms of pricing, with the addition of more suppliers, HBM3E still has some room for negotiation, but Hynix expects only a "slight to moderate decrease" in prices in 2026 compared to 2025. More importantly, with the HBM4 premium obtained through its first mover advantage, Hynix expects its HBM4 price to be about 40% higher than the upcoming generation product - even after a 50% increase in cost per bit. Overall, UBS predicts that the price per bit of hybrid HBM will increase by 18.5% year-on-year in 2026, driving HBM revenue to an expected $32.7 billion and accounting for over 70% of SK Hynix's operating profit. However, the development of HBM is not without risks. The delay in Samsung's capacity expansion may intensify competitive pressure this year, and the significant increase in HBM4 production costs may also lead to more intense price negotiations in the future. Investors are also closely monitoring capital expenditures, as Hynix's expansion plans will depend on the progress of Nvidia's next-generation Blackwell Ultra and Rubin product cycles. Despite recent uncertainties, UBS reiterates that the outlook for HBM in 2026 remains strong and expects Hynix to continue to dominate. HBM， How exactly is it? From the previous reports, we can see that the long short battle of HBM is unprecedentedly fierce. Therefore, let's take a look at the production capacity and technological layout of these three HBM giants to provide reference for everyone's HBM trend. According to analysts, Samsung Electronics and SK Hynix are expected to ensure a monthly HBM production capacity of approximately 150000 wafers by the end of 2025. 1. Samsung Electronics: Initially expected to produce 170000 wafers per month by the end of 2025, but reduced to 150000 wafers per month. Therefore, the shipment forecast has been lowered from 80 billion Gb to 60 billion Gb. 2. SK Hynix: Initially expected to produce 65000 wafers per month by the end of 2025, but raised to 150000 wafers per month. Additional expansion is planned for M15 x in 2026. 3. Micron expects to expand its production capacity to 25000 pieces per month by the end of 2024, 65000 pieces per month by the end of 2025, and 90000 pieces per month by the end of 2026.   The report further points out that the market expansion of HBM3e and HBM4 is the biggest variable. After 2025, driven by Blackwell (NVDA) and TPU (AVGO), the demand for high-end HBM3e and above is expected to increase. According to the report,. The demand for Blackwell series in 2025 is 5.3 million units, and the demand for TPU v6 is 2.2 million units. Ultimately, the main reason for the increase in demand is capacity growth: 1) The DRAM capacity in Blackwell has increased by 2 times (H200 to B300) to 2.4 times (H100 to B200), 2) The DRAM capacity in TPU v6 has doubled compared to v5p. With the arrival of HBM4, ASIC customization will also stimulate the demand for HBM. According to industry insiders, Samsung Electronics, SK Hynix, and Micron are expanding their supply of HBM products to ASIC design companies. Last month, Micron announced at its earnings conference that in addition to Nvidia and AMD, ASIC platform companies are also the four major customers for HBM's bulk shipments. According to Gao Yongmin, a researcher at DAOL Investment Securities, "This reflects the confidence brought by the growth in ASIC customer demand. ” With the surge in demand for AI model specific custom semiconductors operated by companies such as Amazon, Meta, and Google, the ASIC market has also experienced rapid growth. This is because general AI semiconductor products produced by companies such as Nvidia and AMD are expensive, and their performance to power ratio is not sufficient to run AI models. The industry expects ASIC shipments to exceed Nvidia's AI semiconductor supply next year. JPMorgan predicts that the global AI ASIC market will reach approximately $30 billion (about KRW 41 trillion) this year, with an annual growth rate of over 30%. With the rapid development of ASIC companies, memory semiconductor companies producing HBM are also expanding their supply. It is reported that SK Hynix, the market leader in HBM, is supplying HBM in bulk to companies such as Amazon, Google, and Broadcom's ASIC chips. According to reports, Samsung Electronics is also supplying the fifth generation HBM (HBM3E) to companies such as Broadcom. An industry insider pointed out that the supply of ASIC still accounts for about 10% of the entire HBM market, but in fact, the supply that used to be mainly concentrated in Nvidia and AMD is rapidly diversifying. ” LS Securities researcher Cui Yonghao said, "Starting from next year, with the continuous growth of ASIC's market share, HBM customers will present a diversified prospect. ” So, in your opinion, what would be the trend of HBM?

Intel is simultaneously using its own 18A and TSMC N2 processes to address the delivery and production capacity issues of 18A. On July 14th, it was reported that Intel had completed the Nova Lake processor computing module chip fabrication at TSMC's 2nm process node N2 a few weeks ago, and the next step will be to power it up and run it. If everything goes smoothly, it can lay the foundation for mass production in 2026. Due to the urgent demand for advanced processes in CPU cores, a 2nm process can bring about a 15% performance improvement and a 30% power optimization. Therefore, this chip fabrication is likely to be its computing module, which can fundamentally improve the overall performance of the processor. Intel plans to continue its dual source foundry strategy and place over $14 billion in 2nm orders with TSMC, covering production capacity from 2024-2025. Computing modules may use both its own 18A and TSMC N2 processes to address issues with 18A delivery or production capacity. The compatibility of 18A-P/14A with the time plan is of concern, and there is uncertainty as to whether it will be ready for mass production before the 2025 production milestone. TSMC N2 plans to start production in the middle and later stages of 2025, but the timeline is tight. Nova Lake-S is expected to be delivered in the second half of 2026, with a high probability of release in the third quarter of 2026. It integrates up to 52 cores, paired with an 8800MT/s memory controller and related graphics and media processing modules, making it difficult to manufacture. At present, the chip is undergoing power on testing, and further mass production and shipment will take time. Its market performance needs to be tested.