According to TechNews’ report, in the midst of production cutbacks by Samsung, SK Hynix, and Micron, NAND Flash wafer prices are surging.

As the traditional peak season for end-user stockpiling comes to an end, memory module manufacturers wish to position themselves favorably during a dip in demand. However, the reduced supply resulting from production cutbacks paradoxically elevates the demand for NAND Flash, intensifying the momentum of price rebounds. Memory module manufacturers are left with no choice but to accept the price increases imposed by memory manufacturers.

Fueled by the expectation that memory manufacturers will continue to raise prices, memory module manufacturers continue aggressive purchasing, contributing to an upward price trend in December.

Major memory manufacturers Samsung, SK Hynix, and Micron had previously announced significant production reduction plans. Samsung initiated a decrease in NAND Flash production from the second quarter and further expanded the reduction to 50% of total capacity in September, focusing mainly on products with less than 128 layers. This move instilled confidence in price hikes among industry peers.

Due to the unexpectedly substantial reduction in production by major memory manufacturers, coupled with generally low inventory levels on the client side, NAND Flash prices continue to rise.

In the latter half of this year, the demand for Mobile DRAM and NAND Flash (eMMC, UFS) has not only been driven by the traditional peak season but also stimulated by the production expansion goals of other Chinese smartphone brands, including the Huawei Mate 60 series. This sudden influx of demand is contributing to the price hikes in fourth-quarter contracts.

The most significant price surge in this wave is undoubtedly in NAND Flash wafer prices. According to the latest research from TrendForce, the current industry situation indicates that module manufacturers’ inventories have rapidly depleted due to increased orders from customers. This has prompted module manufacturers to turn to memory manufacturers, requesting expanded supply.

However, with memory manufacturers persisting in their production reduction strategies, the imbalance between supply and demand has led to a robust rebound in NAND Flash wafer prices in the fourth quarter. According to TrendForce’s data, the month of November alone witnessed a price increase of over 25% for NAND Flash wafers.

Industry sources reveal that in the current scenario of limited supply and significantly increased demand, module manufacturers have no choice but to accept the forceful price hikes imposed by memory manufacturers. The industry, anticipating that memory manufacturers will continue to raise prices, has resulted in a situation where “Everyone just keeps scrambling for inventory.”

Based on the current market conditions, TrendForce believes that in December, with tight supply, NAND Flash prices will continue to rise. However, whether prices will continue to surge significantly in the first quarter next year depends on the production reduction strategies of NAND manufacturers and the state of demand.

It is reported that there are industry rumors suggesting that some memory manufacturers are considering increasing production capacity due to the strong downstream demand. If memory manufacturers decide to increase its capacity earlier, coupled with unclear improvements in demand, the extent of price increases may be noticeably limited.

(Photo credit: Samsung)

Market reports suggest Nvidia’s new product release cycle has shortened from two to a year, sparking intense competition among major memory companies in the realm of next-gen High Bandwidth Memory (HBM) technology. Samsung, SK Hynix, and Micron are fervently competing, with SK Hynix currently holding the dominant position in the HBM market. However, Micron and Samsung are strategically positioned, poised for a potential overtake, reported by TechNews.

Current Status of the HBM Industry

SK Hynix made a breakthrough in 2013 by successfully developing and mass-producing HBM using the Through Silicon Via (TSV) architecture. In 2019, they achieved success with HBM2E, maintaining the overwhelming advantage in the HBM market. According to the latest research from TrendForce, Nvidia plan to partner with more HBM suppliers. Samsung, as one of the suppliers, its HBM3 (24GB) is anticipated to complete verification with NVIDIA by December this year.

Regarding HBM3e progress, Micron, SK Hynix, and Samsung provided 8-layer (24GB) Nvidia samples in July, August, and October, respectively, with the fastest verification expected by year-end. All three major players anticipate completing verification in the first quarter of 2024.

As for HBM4, the earliest launch is expected in 2026, with a stack increase to 16 layers from the existing 12 layers. The memory stack will likely adopt a 2048-bit memory stack connection interface, driving demand for the new “Hybrid Bonding” stacking method. The 12-layer HBM4 product is set to launch in 2026, followed by the 16-layer product expected in 2027.

Navigating HBM4, the New Technologies and Roadmaps of Memory Industry Leaders

SK Hynix

According to reports from Business Korea, SK Hynix is preparing to adopt “2.5D Fan-Out” packaging for the next-generation HBM technology. This move aims to enhance performance and reduce packaging costs. This technology, not previously used in the memory industry but common in advanced semiconductor manufacturing, is seen as having the potential to “completely change the semiconductor and foundry industry.” SK Hynix plans to unveil research results using this packaging method as early as next year.

The 2.5D Fan-Out packaging technique involves arranging two DRAM horizontally and assembling them similar to regular chips. The absence of a substrate beneath the chips allows for thinner chips, significantly reducing the thickness when installed in IT equipment. Simultaneously, this technique bypasses the Through Silicon Via (TSV) process, providing more Input/Output (I/O) options and lowering costs. 

According to their previous plan, SK Hynix aims to mass-produce the sixth-generation HBM (HBM4) as early as 2026. The company is also actively researching “Hybrid Bonding” technology, likely to be applied to HBM4 products.

Currently, HBM stacks are placed on the interposer next to or GPUs and are connected to their interposer. While SK Hynix’s new goal is to eliminate the interposer completely, placing HBM4 directly on GPUs from companies like Nvidia and AMD, with TSMC as the preferred foundry.

Samsung

Samsung is researching the application of photonics in HBM technology’s interposer layer, aiming to address challenges related to heat and transistor density. Yan Li, Principal Engineer in Samsung’s advanced packaging team, shared insights at the OCP Global Summit in October 2023.

(Image: Samsung)

According to Samsung, The industry has made significant strides in integrating photonics with HBM through two main approaches. One involves placing a photonics interposer between the bottom packaging layer and the top layer containing GPU and HBM, acting as a communication layer. However, this method is costly, requiring an interposer and photon I/O for logic chips and HBM.

(Image: Samsung)

The alternative approach separates the HBM memory module from packaging, directly connecting it to the processor using photonics. Rather than dealing with the complexity of packaging, a more efficient approach is to separate the HBM memory module from the chip itself and connect it to the logic IC using photonics technology. This approach not only simplifies the manufacturing and packaging costs for HBM and logic ICs but also eliminates the need for internal digital-to-optical conversions in the circuitry. However, careful attention is required to address heat dissipation.

Micron

As reported by Tom’s Hardware, Micron’s 8-layer HBM3e (24GB) is expected to launch in early 2024, contributing to improved AI training and inference performance. The 12-layer HBM3e (36GB) chip is expected to debut in 2025.

Micron is working on HBM4 and HBM4e along with other companies. The required bandwidth is expected to exceed 1.5 TB/s. Micron anticipates launching 12-layer and 16-layer HBM4 with capacities of 36GB to 48GB between 2026 and 2027. After 2028, HBM4E will be introduced, pushing the maximum bandwidth beyond 2+ TB/s and increasing stack capacity to 48GB to 64GB.

Micron is taking a different approach from Samsung and SK Hynix by not integrating HBM and logic chips into a single die, suggested by Chinese media Semiconductor Industry Observation. This difference in strategy may lead to distinct technical paths, and Micron might advise Nvidia, Intel, AMD that relying solely on the same company’s chip carries greater risks.

(Image: Micron)

TSMC Aids Memory Stacking       

Currently, TSMC 3DFabric Alliance closely collaborates with major memory partners, including Micron, Samsung, and SK Hynix. This collaboration ensures the rapid growth of HBM3 and HBM3e, as well as the packaging of 12-layer HBM3/HBM3e, by providing more memory capacity to promote the development of generative AI.

(Image: TSMC)

(Image: SK Hynix)

Explore more

• Manufacturers Anticipate Completion of NVIDIA’s HBM3e Verification by 1Q24; HBM4 Expected to Launch in 2026, Says TrendForce
• [News] SK Hynix Signs Prioritized Supply Agreement with NVIDIA for HBM3e, 2023 Q4 Revenue Expected to Surpass KRW 10 Trillion Again
• [News] Memory Giants Samsung & Micron Actively Embrace DDR5 and HBM as Favorable Choices
• [News] Seizing the AI Trend! Revealing Samsung and Micron’s HBM Expansion Timetable
• Continuous Rise in HBM Demand, Memory Giants Expecting HBM4 Delivery in 2025

According to South Korean media Chosun Biz’s report, the prices of all memory products, encompassing servers, mobile devices, and PCs, are on the rise. This trend, combined with the thriving development of the AI market, is expected to result in even higher profitability for High-Bandwidth Memory (HBM) than initially anticipated.

Major memory manufacturers Samsung and SK Hynix are beginning to emerge from a business downturn, leading to upward revisions in their financial outlook for the fourth quarter of 2023.

The report notes that the adjustments in Q4 financial outlook by Samsung and SK Hynix indicate a rapid increase in demand for HBM due to the thriving AI market. Additionally, the recovery of the largest sales item, DRAM, is contributing to better operational performance for both companies in the fourth quarter.

Market experts reveal that Samsung’s projected operating profit for Q4 is expected to be KRW 3.487 trillion, showing slight growth compared to the estimate from a week ago. As for SK Hynix, the expected loss in Q4 is KRW 294.4 billion, reflecting a convergence from the market estimate of KRW 335.3 billion a week earlier, despite remaining in a deficit.

In addition, Micron, the American company considered one of the three major global DRAM manufacturers along with South Korean companies Samsung and SK Hynix, has also revised its financial forecast for the first quarter of the 2024 fiscal year.

The initial estimate in November of USD 4.4 billion has been adjusted to USD 4.7 billion , while the expected Earnings per Share has been raised from USD -1.07 to USD -1.

Regarding trends in the memory industry, TrendForce indicated in a recent report that a key turning point in the third quarter for the NAND Flash market was Samsung’s decision to actively reduce production.

Previously, buyers maintained a low inventory and slow procurement strategy due to concerns about low visibility of end demand and worries about a lackluster market peak season. With the leading supply-side companies significantly reducing production, buyers, anticipating a significant reduction in supply, have shifted to a more positive procurement attitude. By the end of the third quarter, contract pricing for NAND Flash had shifted toward stabilization and even price increases.

TrendForce predicts that NAND Flash products will experience both increased volume and prices in the fourth quarter. The average selling price for all products is estimated to increase by 13%, and the overall revenue growth for the NAND Flash industry in the quarter is expected to exceed 20%.

Contrarily, in the case of DRAM, prices have been on a downward trend since 2023, but they started to rise from October. TrendForce believes that the three major global DRAM manufacturers have begun intensive production cuts, and as market demand begins to recover, the pricing power of memory manufacturers is gradually increasing.

In terms of DRAM supply in the fourth quarter, memory manufacturers have a clear upward pricing attitude, as TrendForce projects a noticeable increase of approximately 13-18% in contract prices during this period. However, the recovery in demand is not as strong as in previous peak seasons.

Overall, while there is demand for stocking up, in the current scenario, the server sector remains passive in terms of procurement due to high inventory levels. The shipment growth in the DRAM industry for the fourth quarter is expected to be limited.

Read more

• [News] Beyond Price Hikes, What Lies Ahead for the Memory Market?
• [Insights] Signals from the Latest Financial Reports of Top 5 Global Storage Giants

(Photo credit: Samsung)

China’s top panel makers, including BOE, TCL CSOT, HKC, and CEC-CHOT, are cutting production to support TV panel prices. Speculations indicate a capacity utilization drop below 60% in the first quarter, benefiting Taiwanese panel companies like Innolux and AUO. Despite these efforts, industry sources caution of a panel industry slowdown due to reduced demand, possibly resulting in lower TV panel prices, reported by TechNews.

The memory industry saw a big downturn due to major players like Samsung, SK Hynix, and Micron drastically reducing production. This resulted in a surge in memory prices and a gradual recovery for the entire memory industry. Panel makers may adopt a similar strategy to boost the overall panel industry as well.

Chinese panel companies currently command a global market share of over 70% in TFT-LCD, with the world’s top 3 LCD panel makers from China: BOE, TCL CSOT, and HKC. As China holds the largest share of panel production capacity, a reduction in production by major players could have a more substantial impact on the panel industry compared to the effects on the memory industry caused by Samsung, Micron, and others.

According to TrendForce’s report in November, BOE began adjusting its production levels in the Q3 and will continue to do so in Q4, with an estimated decrease of 7 percentage points in utilization rates. CSOT, on the other hand, maintained high utilization rates in Q3, supported by major customer stockpiling and the ramp-up of the T9 new production line. However, due to reduced procurement of TV panels by both in-house group brands and international frontline brands, CSOT’s utilization rate is expected to decrease by about 17 percentage points to 76% in Q4.

HKC, which still has two production lines not running at full capacity, anticipates a 14 percentage point decrease in its utilization rates for Q4. This is primarily due to the need to reduce production of one of its main products, the 32-inch TV panel, to alleviate inventory pressures and avoid significant price drops.

In response to this, both Innolux and AUO express their intention to dynamically adjust production capacity utilization in line with market conditions. This strategic flexibility is aimed at fostering a more robust and balanced industry order.

Overall, most panel makers are adopting a more conservative approach to production for 1Q24. Furthermore, several panel makers have indicated a two-week Lunar New Year shutdown for 2024. As a result, TrendForce anticipates that overall Gen5+ LCD panel utilization rates may be revised down to 70% or lower in 1Q24 to maintain the market supply-demand balance.

(Image: BOE)

Explore more

• [News] Samsung Ends Partnership with BOE! AUO Welcomes the Diversion Effect, but Capacity Challenges Loom
• Gen5+ LCD Panel Utilization Rates to Reach Approximately 72% in Q4, with Potential Further Decline in 1Q24, Says TrendForce
• [News] AUO Reports Strong Growth in Automotive Panel Revenue and Positive Outlook for IT Panels

India could see at least three more semiconductor fabrication units coming up in the next few months with cumulative investments to the tune of USD 8-12 billion, Union Minister for Electronics and Information Technology Ashwini Vaishnaw said on Tuesday.

India is working hard to get big semiconductor companies to build chip plants in the country with its $10 billion semiconductor program. Global semiconductor giants are considering making factories in India driven by the need to diversify supply chains beyond China and the big demand for tech talents in India.

According to Deccan Herald’s report. At the unveiling of AMD’s global design center in Bengaluru on November 28th, Ashwini Vaishnaw, Union Cabinet Minister for Railways, Telecommunications, Electronics and Information Technology of India, briefed journalists on ongoing negotiations with several local governments to determine precise locations in the following months.

AMD’s global design center plans to host around 3,000 AMD engineers in the coming years, focused on the design and development of semiconductor technology including 3D stacking, artificial intelligence, machine learning, and more. Vaishnaw pointed out that AMD setting up its biggest design center is strong evidence of how much global companies trust India. Besides, Vaishnaw highlighted the achievement of Micron’s semiconductor plant project, which was revealed in June and started construction in September. He said the success of Micron’s initial facility in India has boosted confidence globally.

Apart from global giants’ investment in India, the minister also told about the Semiconductor Lab (SCL) progress in Mohali, mentioned significant headway in the Expression of Interest (EoI) process. He anticipates finalizing matters within the next 5-6 months. The lab is planned to serve as both a research and semi-commercial facility.

The Indian government asserts that a comprehensive microprocessor chip manufacturing initiative is on the horizon in the country. Anil Agarwal, Chairman of Vedanta, anticipates the launch of “Vedanta chips made in India” by 2025, reported by icbank on weixin.

The government envisions establishing a “semiconductor city” in Dholera, Gujarat, covering an area equivalent to Singapore, with Prime Minister Modi expressing confidence in attracting both domestic and international giants to set up manufacturing plants despite the remote location.

The tech hub in Bangalore has played a pivotal role in India’s ascent as a global participant in the semiconductor network, primarily through chip design. Over the past two years, substantial government subsidies have transformed India into a hub for electronic product manufacturing.

India’s Challenges abound in the Actual Chips Manufacture

Despite initiatives like the “production-linked incentive measures” encouraging mobile assembly in India since 2020, chip manufacturing remains a formidable task.

Presently, nearly all cutting-edge chips are produced in Taiwan, with China’s chip investments surpassing those of India. Vedanta Group’s Agarwal plans to kickstart chip manufacturing within two and a half years, extending invitations to global chip manufacturing experts.

While today’s cutting-edge chips are predominantly manufactured in Taiwan, TSMC has even established a fab in Arizona with the assistance of the U.S. “Chip Act.”

India lacks a historical presence in chip manufacturing and faces a shortage of highly skilled engineers and equipment. Nevertheless, the government aims to achieve what companies like TSMC took decades to accomplish through substantial spending and capital investment promotion. However, China’s significant investments in domestic chip manufacturers, following the U.S. ban tightening China’s semiconductor development last October, far exceed India’s investments in local companies.

With the goal of establishing India’s inaugural semiconductor foundry, Mr. Agarwal of Vedanta Group aims to commence chip manufacturing within two and a half years, extending invitations to international talents like David Reed and Texas Instruments.

Agarwal aspires to attract around 300 foreign experts from East Asia and Europe to Gujarat. The manufacturing of microchips also necessitates a significant amount of customized infrastructure, with Vrishno, the responsible official, indicating that India’s largest chemical plant near Dholera can produce the specialized gases and liquids required for any chip factory’s operation.
(Image: AMD India)