On July 1st, according to a report from Reuters, the South Korea’s Ministry of Trade, Industry and Energy has announced that exports had grown for the ninth consecutive month in June. The sustained and even increasing demand for chips overseas led to a record high in chip export value for June. This export growth also propelled South Korea’s Manufacturing Purchasing Managers’ Index (PMI) to its highest level in over two years.

In June, South Korea’s exports grew by 5.1% year-on-year to USD 57 billion, while imports decreased by 7.5% to USD 49 billion. This resulted in a trade surplus of USD 8 billion, surpassing market expectations of USD 5.24 billion and marking the largest surplus since September 2020, as well as a 13 consecutive month of surplus.

Reportedly, chips are the most significant driver of export growth, with a 50.9% year-on-year growth in June to USD 13.4 billion, marking eight consecutive months of year-on-year growth.

The Trade Ministry believes that the robust demand for high-performance memory chips, such as DRAM for AI servers, has driven the recovery in exports.

As per another report from The Korea Times, South Korea’s exports of IT products such as displays, computers, and wireless communication devices have been growing consecutively for four months. In June, these three categories saw year-on-year export increases of 26.1%, 58.8%, and 3.9%, respectively.

In terms of export markets, South Korea’s exports to the United States in June rose by 14.7% year-on-year to USD 11 billion, reaching a new high for the same month in historical records. This marks continuous monthly export highs to the U.S. since August last year. Meanwhile, exports to mainland China increased by 1.8% year-on-year to USD 10.7 billion, marking four consecutive months of growth.

For the first half of the year, South Korea’s total exports amounted to USD 334.8 billion, a 9.1% increase from the same period last year. Imports, on the other hand, decreased by 6.5% year-on-year to USD 311.7 billion, resulting in a trade surplus of USD 23.1 billion, the highest for the same period since 2018.

(Photo credit: Samsung)

On July 1st, according to a report from Reuters, the French antitrust authority plans to file charges against NVIDIA, accusing the company of engaging in anti-competitive practices, making France the first country to take such action against NVIDIA.

The French competition regulator had raided NVIDIA’s local offices in September last year. At the time, they did not disclose the details of the investigation or the company involved, only stating it was related to the graphics card sector.

However, as per a previous report from Bloomberg, NVIDIA claimed that the French agency collected information from them regarding their business and competition in the graphics card and cloud service provider market as part of an ongoing inquiry into competition in those markets.

Sources cited by Reuters’ report indicated that last year’s raid was part of a broader investigation into cloud computing. With the surge in global chip demand following the advent of ChatGPT, NVIDIA, as the world’s largest manufacturer of AI and computer graphics cards, has naturally attracted close scrutiny from antitrust authorities in Europe and the United States.

NVIDIA previously disclosed in regulatory filings that both EU and French regulators had requested information about its graphics card products. The French antitrust authority has been actively investigating to understand NVIDIA’s key role in AI processors, its pricing policies, chip shortages, and the impact on prices.

Last Friday, the French authorities released a report on competition in generative AI, highlighting the risk of chip suppliers abusing their power. The report pointed out concerns about the chip industry’s heavy reliance on NVIDIA’s CUDA software for chip programming. Additionally, NVIDIA’s focus on investing in AI cloud service provider CoreWeave has also raised significant concerns among the authorities.

Reportedly, it is understood that companies violating French antitrust rules could face fines of up to 10% of their global annual revenue, though they can choose to make concessions to avoid penalties.

Moreover, the European Commission is currently gathering informal feedback to determine if NVIDIA has breached its antitrust rules, although it has not yet launched a formal investigation into anti-competitive behavior.

On the other hand, the New York Times reported on June 5th that the U.S. Department of Justice and the Federal Trade Commission (FTC) have reached an agreement, led by senior officials of both agencies, over the past week. The DOJ will investigate whether NVIDIA has violated antitrust laws, while the FTC will examine the conducts of OpenAI and Microsoft.

Read more

• [News] NVIDIA’s H200 Order Delivered from Q3, Boosting Server Supply Chain with Strong Demand
• [News] US Government Reportedly Investigating Microsoft, NVIDIA, and OpenAI for AI Industry Monopolization

(Photo credit: NVIDIA)

In late March, NVIDIA CEO Jensen Huang personally delivered the first high-end DGX H200 AI server to customer OpenAI. According to a report from Commercial Times, following this, the upstream chips for the H200 AI-GPU entered mass production in late Q2, with large-scale deliveries expected from Q3 onwards.

Downstream companies, including Inventec, Quanta (QCT), Hewlett Packard Enterprise (HPE), Supermicro, and Lenovo, have reportedly listed the H200 products as ready for shipment, with deliveries anticipated to begin in the second half of the year.

The same report, citing sources, indicates that current pending orders are still largely focused on the H100 in the HGX architecture, with the H200’s share remaining limited. The H200 shipments expected in Q3 will primarily be NVIDIA’s DGX H200. As for the B100, there is already some visibility, with shipments expected in the first half of next year.

Despite the CoWoS production capacity catching up and a significant easing in the supply of AI GPUs, due to the allocation system, the delivery time for the main H100 GPU shipments from various end-system partners can still reach up to 20 weeks.

However, major Taiwanese manufacturers such as Quanta, Inventec, Wistron, Gigabyte, and ASUS have seen a substantial boost in their overall server operations, driven by AI server business in the first half of the year. They are reportedly optimistic about AI server shipments in the second half of the year, expecting strong demand to continue, making the server business effectively free from any off-season throughout the year.

Among them, Wistron holds a significant advantage in the production and supply of H100 series substrates and the subsequent B100 series GPU modules and substrates. Starting from the second quarter, Wistron’s AI server-related business has shown high visibility, providing strong support for its overall operations.

On the other hand, the H20 series, an AI chip tailored for China due to U.S. chip restrictions, has also seen demand in the Chinese market. As NVIDIA recently secured a rush order for the H20 series, Taiwanese companies, including Wistron and Inventec, are expected to benefit.

Read more

• [News] Taiwanese AI Server Supply Chain to Benefit from Apple’s Potential Collaboration with Baidu and Alibaba
• [News] Rumored Sharp Drop in H100 Server Black Market Prices in China Raises Concerns Over Market Stability

(Photo credit: NVIDIA)

According to a report from the Korean media outlet The Chosun Daily, Chinese company Huawei plans to collaborate with memory manufacturer Wuhan Xinxin Semiconductor Manufacturing Co. (XMC) to produce High Bandwidth Memory (HBM) semiconductors. Additionally, Jiangsu Changjiang Electronics Technology (JCET) and Tongfu Microelectronics, which are developing CoWoS advanced packaging technology, are also participating in the project.

CoWoS is a high-precision technology that integrates graphics processing units (GPUs) and HBM on a single substrate. This enhances computational performance, saves space, and reduces power consumption. Currently, the CoWoS technology developed by leading foundry TSMC is used in the production of AI chips for GPU giant NVIDIA.

In May 2023, according to another report from Reuters, China’s leading DRAM company CXMT (ChangXin Memory Technologies) collaborated with Tongfu Microelectronics to develop HBM chip samples. Additionally, tech media outlet The Information reported earlier that a series of Chinese companies, led by Huawei, plan to mass-produce HBM and increase China’s HBM output by 2026.

Furthermore, in March 2023, XMC announced the construction of an advanced HBM manufacturing plant, which is expected to produce 3,000 12-inch wafers per month.

The report further emphasizes that China is still in the early stages of HBM development. However, under the technological restrictions imposed by the United States in the semiconductor and artificial intelligence sectors, Huawei and a series of Chinese semiconductor companies’ move into HBM production has attracted close attention.

Currently, South Korean companies SK Hynix and Samsung Electronics control most of the global HBM market share, indicating that Huawei’s plan to develop HBM still has a long way to go.

Per TrendForce’s data, the three major HBM manufacturers held market shares are as follows: In 2023, SK Hynix and Samsung each held around 47.5%, while Micron’s share was roughly 5%. Still, forecasts indicate that SK Hynix’s market share in 2024 will increase to 52.5%, while Samsung’s will decrease to 42.4%.

Read more

• [News] Micron’s HBM Expansion at Full Throttle: Rumored to Expand in US, Malaysia Also an Option
• [News] XMC Initiates IPO Plan, Potentially Becoming China’s First HBM Foundry

(Photo credit: XMC)

As semiconductor companies led by TSMC accelerate their pace for capacity expansion, benefiting the local supply chain, industry electricity consumption has emerged as a tough challenge for Taiwan. According to the latest report by the Economic Daily News, Taiwan can only accommodate 20 more large plants, with the power supply possibly reaching its limit in two years.

As its price of industrial electricity is comparatively lower to international rates, the Taiwan market not only attracts semiconductor companies but also international giants to establish data centers, the report noted. In addition to cloud service providers like AWS, Google, and Microsoft, Apple has recently planned to set up a data center in Taiwan.

Citing engineering companies familiar with high-tech industries, the report indicated that more than 10 new data centers are expected to be constructed in Taiwan. With the recent wave of announcements by semiconductor companies to launch advanced nodes and packaging capacities in Taiwan, it is estimated that once the 2nm and 1.6nm factories are fully operational, approximately 10 more semiconductor plants will result in a power supply challenge.

That is to say, Taiwan could accommodate around 20 more tech plants to be built by 2026 in total, the report said.

Take TSMC as an example. According to earlier reports by Commercial Times, the foundry giant’s 3nm plant in Tainan plans to begin mass production in the third quarter, while EUV (Extreme Ultraviolet Lithography) machines will be introduced progressively at another 3nm plant, P8 in Hsinchu, next year. On the other hand, TSMC’s advanced 2-nanometer process capacity is set to begin mass production in 2025.

Commercial Times noted that the EUV machines, crucial for advanced processes, will see over 60 units delivered this year and next to TSMC.

However, EUV machines are considered “electricity-consuming monsters.” According to an earlier report by BITS&CHIPS, ASML’s EUV machine consumes about a megawatt to produce 160 wafers per hour. Since a chip must go through twenty passes in this scanner, this results in an additional energy consumption of about 0.2 kWh per square centimeter of the chip, totaling 1.6 kWh per square centimeter. An earlier report by Bloomberg estimated that because of the vast amount of power needed to run EUVs, TSMC is expected to use 12.5% of Taiwan’s entire electricity supply by 2025.

Read more

• [News] TSMC Achieved a New Step in the Construction of its Third 2nm Fab
• [News] TSMC Reportedly Invests over USD 12.3 Billion in EUV, Advancing in 2nm Production

(Photo credit: TSMC’s P8, DACIN Construction)