As we enter the second half of 2023, smartphone brands are eagerly launching their new devices, hoping to revitalize the market that is yet to fully recover. Among them, Samsung stands out with its ambitions to captivate consumers with flagship models while maintaining stability in the mid-to-low-end segment.

However, the Galaxy A and M series, targeting the mid-range to low-end market, have faced lackluster sales. To reverse this trend, Samsung is gearing up to unveil new devices in the coming months, including the highly anticipated Galaxy Fold/Flip5. This foldable flagship is seen as a game-changer that could drive up shipments in the latter part of the year.

The company’s flagship models, which encompass the Galaxy S and Fold/Flip series, account for about 20% of its total shipments. Within this segment, foldable devices contribute approximately 5% to the overall sales.

On the mid-tier front, Samsung is planning to introduce the S23FE series later this year, targeting the mid-range flagship market. Although the S23FE will feature a dual-platform of AP configuration, catering to different regional markets, it will be equipped with older-generation processors. As a result, its potential to significantly boost sales might be limited. According to TrendForce’s estimate, the total production and shipment of the S23FE series are expected to be less than 1.5 million units by the end of 2023.

A comprehensive view of Samsung’s strategy reveals that the second half of 2023 will be driven by the flagship Fold/Flip and FE series. Despite the company’s efforts, the overall performance will still be influenced by the prevailing economic conditions, posing significant challenges.

Notably, the decision to unveil the FE series in the fourth quarter marks a strategic departure from the previous year’s delay of the S21FE until the following year’s first quarter. This move by Samsung demonstrates a strong commitment to address the market gap resulting from the absence of the A7x series mid-range models. By positioning the mid-tier flagship as a compelling marketing point, Samsung aims to entice consumers to upgrade their devices, potentially bolstering the company’s sales and maintaining a robust presence in the highly competitive smartphone market.

DRAM Spot Market:

Continuing from the previous week, the influx of used chips that were stripped from decommissioned server DRAM modules is still causing spot prices of DDR4 products to drop. As for DDR5 products, there is no such issue affecting their supply. However, module houses are holding a high level of inventory for DDR5 products because they stocked up aggressively in 2Q23, while the actual demand remains in a slump. Hence, spot prices on the whole have kept falling. The average spot price of the mainstream chips (i.e., DDR4 1Gx8 2666 MT/s) dropped by 0.74% from US$1.479 last week to US$1.468 this week.

NAND Flash Spot Market:

The spot market has been lingering at sluggish transactions since July under feeble inquiries between market participants. Fortunately, the reduction of spot prices has somewhat mitigated over the past two weeks due to suppliers’ continuous intention in an on-going production cut and price increment. Certain components have not seen a further drop in prices after a consolidation at the low end, though the level of transactions remains at a rather insignificant extent, and the continuity of purchases will require further observations. 512Gb TLC wafer spots have risen by 1.21% this week, arriving at US$1.421.

South Korean media reported that Samsung is set to manufacture a new generation of Full Self-Driving (FSD) chips for Tesla’s Level 5 autonomous vehicles. These chips will be utilized in Tesla Hardware 5 (HW 5.0) onboard computers, with production expected to commence after 2025. The chips will be manufactured using Samsung’s 4nm process.

TrendForce’s analysis:

Samsung May Competing with TSMC for Tesla HW 5.0 Chips

In the early stages of Tesla’s autonomous driving technology, the company collaborated with Samsung for FSD chips used in various vehicle models, including Model 3, Model 5, Model X, and Model Y. However, in 2022, Tesla chose to work with TSMC, citing TSMC’s better yield performance in 4nm process technology at that time.

In response, Samsung has been actively improving its 3nm and 4nm process technologies within a short period. While Samsung’s 4nm process yield has reached 75%, it still slightly lags behind TSMC’s 80%. Despite this difference, given their previous collaborations, it is not ruled out that Tesla might place orders with both TSMC and Samsung this time. The main reason being Samsung’s plan to advance to the 2nm-level SF2 process technology in 2025 and further progress to the 1.4nm-level SF1.4 process technology in 2027, aligning its overall roadmap with TSMC’s. This advancement will assist Tesla in accelerating the production plan of its DOJO supercomputer, facilitating the transition to Level 5 autonomous driving.

(Photo credit: Tesla)

Looking at the impact of AI server development on the PCB industry, mainstream AI servers, compared to general servers, incorporate 4 to 8 GPUs. Due to the need for high-frequency and high-speed data transmission, the number of PCB layers increases, and there’s an upgrade in the adoption of CCL grade as well. This surge in GPU integration drives the AI server PCB output value to surpass that of general servers by several times. However, this advancement also brings about higher technological barriers, presenting an opportunity for high-tech PCB manufacturers to benefit.

TrendForce’s perspective: 

• The increased value of AI server PCBs primarily comes from GPU boards.

Taking the NVIDIA DGX A100 as an example, its PCB can be divided into CPU boards, GPU boards, and accessory boards. The overall value of the PCB is about 5 to 6 times higher than that of a general server, with approximately 94% of the incremental value attributed to the GPU boards. This is mainly due to the fact that general servers typically do not include GPUs, while the NVIDIA DGX A100 is equipped with 8 GPUs.

Further analysis reveals that CPU boards, which consist of CPU boards, CPU mainboards, and functional accessory boards, make up about 20% of the overall AI server PCB value. On the other hand, GPU boards, including GPU boards, NV Switch, OAM (OCP Accelerator Module), and UBB (Unit Baseboard), account for around 79% of the total AI server PCB value. Accessory boards, composed of components such as power supplies, HDD, and cooling systems, contribute to only about 1% of the overall AI server PCB value.

• The technological barriers of AI servers are rising, leading to a decrease in the number of suppliers.

Since AI servers require multiple card interconnections with more extensive and denser wiring compared to general servers, and AI GPUs have more pins and an increased number of memory chips, GPU board assemblies may reach 20 layers or more. With the increase in the number of layers, the yield rate decreases.

Additionally, due to the demand for high-frequency and high-speed transmission, CCL materials have evolved from Low Loss grade to Ultra Low Loss grade. As the technological barriers rise, the number of manufacturers capable of entering the AI server supply chain also decreases.

Currently, the suppliers for CPU boards in AI servers include Ibiden, AT&S, Shinko, and Unimicron, while the mainboard PCB suppliers consist of GCE and Tripod. For GPU boards, Ibiden serves as the supplier, and for OAM PCBs, Unimicron and Zhending are the suppliers, with GCE, ACCL, and Tripod currently undergoing certification. The CCL suppliers include EMC. For UBB PCBs, the suppliers are GCE, WUS, and ACCL, with TUC and Panasonic being the CCL suppliers.

Regarding ABF boards, Taiwanese manufacturers have not yet obtained orders for NVIDIA AI GPUs. The main reason for this is the limited production volume of NVIDIA AI GPUs, with an estimated output of only about 1.5 million units in 2023. Additionally, Ibiden’s yield rate for ABF boards with 16 layers or more is approximately 10% to 20% higher than that of Taiwanese manufacturers. However, with TSMC’s continuous expansion of CoWoS capacity, it is expected that the production volume of NVIDIA AI GPUs will reach over 2.7 million units in 2024, and Taiwanese ABF board manufacturers are likely to gain a low single-digit percentage market share.

(Photo credit: Google)

During the first half of 2023, the polysilicon industry experienced expansion in production capacity, resulting in an oversupply of polysilicon and a subsequent downward trend in the entire industry chain prices. However, by the end of June, the prices of polysilicon reached a near-bottom point, and both polysilicon and wafer prices stabilized, leading to a significant increase in customer demand.

Polysilicon: In the first half of 2023, the polysilicon market witnessed fluctuating prices, starting with an initial rise followed by a subsequent decline, and an underlying issue of oversupply persisted into the second half of the year.

In early January, the expansion of polysilicon production capacity coincided with weakened demand as the year-end approached, leading to a significant decline in prices. By mid-January, the polysilicon prices fell below the cost line for polysilicon enterprises. In response, leading enterprises refrained from selling polysilicon at such low prices, causing a price rebound. As February began, the operation rate of wafer production significantly increased, leading to higher procurement demands for polysilicon and consequently a sharp rise in its prices. However, by the middle of the month, most polysilicon orders for the month had been signed, dampening the stimulus for further price increases. In March, the pressure of excess inventory prompted some polysilicon enterprises to cut prices to facilitate higher shipments, resulting in a slow reduction of prices. Moving into April, the overall supply of polysilicon remained abundant, and with strong willingness among silicon enterprises to sell, prices continued to decline gradually. May witnessed a further increase in polysilicon output, causing a faster decline in prices due to inventory accumulation and pressure from crystal pulling activities. As June approached, the market faced the release of additional production capacity and a considerable accumulation of inventory. This led to polysilicon prices nearly reaching the cost line as market demand fell short of expectations. In response, some enterprises opted for temporary shutdowns, overhauls, and delayed production to reduce inventory pressure. Moreover, increased procurement volumes from crystal pulling plants helped stabilize prices temporarily.

In July, the polysilicon inventory levels experienced a decline compared to the previous period due to increased downstream demand, leading to price stabilization. However, there remains a possibility of a slight price rebound. Projections indicate that in the third quarter, polysilicon prices might rebound to more than RMB 80/KG. The second half of 2023 is expected to witness a peak in photovoltaic demand, and the current price levels within the industry chain can help stimulate this demand. However, the third quarter will also see the concentration of new production capacity from several polysilicon manufacturers entering the market. As a result, the oversupply of polysilicon is unlikely to be significantly altered. While there may be a chance for prices to rebound, both the timing and extent of such a rebound are expected to be very limited.

Monthly price trend of Polysilicon Unit: RMB/KG

Wafer: In the first half of 2023, wafer prices experienced fluctuations with an initial rise followed by a decline. As the second half of 2023 approaches, the wafer prices still face a potential downside risk.

In mid-January, downstream pullback was observed in wafer prices due to increased cost pressures and inventory consumption among polysilicon enterprises. However, early February witnessed a surge in downstream procurement demand, resulting in a slight shortage of polysilicon supply and subsequently leading to sharp price increases for wafers. By later February, the wafer market saw stability as the supply tightened due to the influence of crucible quality, leading to a decline in cell procurement speed. In early March, the supply of high-purity quartz sand remained tight, restricting overall wafer output. The rising wafer costs and limited supply provided support for a slight price increase. As April arrived, the arrival of imported sand eased the tight supply of quartz sand, leading to an increase in wafer production. However, subdued downstream demand resulted in a slight price decline. In May, to avoid losses caused by falling prices, cell companies showed reduced willingness to procure wafers, leading to an accumulation of wafer inventory. The rapid decline in polysilicon prices further contributed to a sharp drop in wafer prices. In early June, wafer enterprises responded by reducing production and cutting prices to address inventory concerns. However, the oversupply situation persisted, and with upstream silicon prices experiencing a significant decline, wafer prices followed suit and declined sharply. Towards the end of June, wafer inventory gradually rebounded to a more reasonable level, and the decline range of polysilicon prices narrowed down. As a result, cell enterprises displayed increased willingness to purchase, leading to wafer prices ceasing further declines first.

In July, there is an expected increase in the output of cell modules, and upstream polysilicon prices have stabilized. However, during this period, there are rumors of India potentially banning the export of quartz sand. Although manufacturers have confirmed that it is merely a rumor, it has still caused some short-term nervousness in the market. Consequently, wafer prices have shown slight signs of rebounding. Nevertheless, based on the current statistics from TrendForce, the effective capacity or output of wafers in a single month remains significantly higher than that of downstream cell and modules. Even if there is an explosive demand for wafers in the second half of the year, there is still a substantial amount of new production capacity waiting to be released. As a result, the market continues to face an oversupply situation, and there remains a considerable risk of declining wafer prices later on.

Monthly price trend of wafer Unit: RMB/Pcs

Cell: In the first half of 2023, cell prices experienced sharp fluctuations, with N-type cells maintaining a premium advantage.

In mid to late January, there was a significant increase in polysilicon and wafer prices, prompting cell prices to rise accordingly. Early February saw a surge in module production scheduling, driving up the demand for cells and supporting their rising costs, leading to further price increases. However, by mid and late February, the interplay between module inventory and pricing resulted in slight declines in cell prices. In March, cost pressures prompted cell enterprises to consider raising prices. However, the high inventory levels and resistance from downstream module companies, unable to sell at higher prices, led to price stability. Towards the end of the month, increased demand for G12 cells caused prices to rise slightly. Moving into April, the overall supply and demand for cells achieved a better balance, resulting in generally stable prices. G12 cells, due to tight supply and demand, commanded significantly higher prices compared to M10 cells. However, May saw a sharp decline in upstream polysilicon and wafer prices. Additionally, downstream module companies exerted pressure to reduce prices, leading to a rapid decline in cell prices. In June, as upstream raw material prices were approaching their bottom, cell prices continued their downward trend. Towards the end of the month, the stabilization of polysilicon and wafer prices, coupled with increased downstream purchasing demand, resulted in a narrower range of cell price declines.

Upstream polysilicon and wafer prices have stabilized, providing a favorable environment for the market. Furthermore, the surge in customer demand has led to a significant month-on-month increase in module production scheduling for July, which is expected to provide strong support for cell prices. If downstream demand surpasses expectations or experiences an early explosion, there is a potential opportunity for cell prices to rebound in the future market. Throughout the first half of 2023, the production capacity of N-type cells fell short of expectations, but the robust customer demand created a structural shortage of N-type products in the market. This resulted in a price gap between N-type and P-type cells. However, in the third quarter, manufacturers are gradually increasing N-type production capacity, which should alleviate the intense supply constraints of N-type products. Yet, this might further stimulate explosive demand for N-type products among customers.

Monthly price trend of cells Unit: RMB/W

Module: In the first half of 2023, overall module prices experienced fluctuations and showed a downward trend. However, expectations for the second half of 2023 are optimistic, as customer demand is anticipated to explode.

In early January, as customer projects reached their end stages, there was a gradual decrease in procurement demand. The sharp decline in upstream raw material prices impacted the modules sector, leading to price fluctuations. In February, with customer projects not yet scaling up and market demand increment falling short of expectations, module prices remained relatively stable. Come early March, customer purchasing was not active, and stable cell prices contributed to the overall stability of module prices. Towards the middle and end of the month, there was an uptick in overseas demand, but cost pressures persisted, keeping module prices stable. Throughout April, module prices continued to remain stable as costs remained unchanged. However, early May saw temporary falls in upstream polysilicon, wafer, and cell prices, which did not immediately affect module prices, allowing them to maintain stability. Nonetheless, by mid-May, the impact of the declining industry chain prices started affecting modules, leading to lower-than-expected customer demand and a significant decline in module prices. In June, although polysilicon and wafer prices gradually stopped falling, customer demand had not yet surged on a large scale, leading to a continued downward trend in module prices. In some instances, module prices even dropped below RMB 1.2/W.

Currently, module prices are experiencing irregular fluctuations, but overall, they have reached the lowest level in recent years. As upstream prices have started stabilizing, it is expected that module prices will soon follow suit, stabilizing at around RMB 1.3-1.4/W. This price point can act as a stimulus for a quick pickup in market demand. In the domestic market, positive signals of increasing demand are evident. Large-size base projects have already commenced construction, and the country has issued the second batch of project lists, with preparations underway. Additionally, plans for the third batch of projects are in progress, and the centralized market is expected to witness an explosion in demand. For the main overseas market, Europe’s inventory has reduced after a period of consumption, but the grid consumption issue has posed a challenge for demand revitalization. However, with the arrival of summer, Europe is expected to face peak electricity consumption, presenting a new opportunity for demand pickup in the region.

Regarding the U.S. market, being a high-value overseas market, it plays a crucial role for various leading module companies in their sales strategies. However, policy fluctuations in the U.S. market have always been a concern for both the supply and demand sides. The Middle East market has been notably active this year, with recent signings of large-scale sales frameworks and cooperation announcements on the manufacturing side. As a result, this market holds considerable potential for future growth and is worth close attention.
Monthly price trend of module Unit: RMB/W

In the first half of 2023, there was a continuous release of polysilicon production capacity, leading to intensified dynamics between the upstream and downstream industry chains, resulting in price fluctuations. Looking ahead to the second half of 2023, the situation of excess supply of polysilicon is anticipated to persist, making it challenging to bring about significant changes. Consequently, any potential price rebound is expected to be limited in both timing and extent. The wafer market remains oversupplied, presenting a downside risk to prices. On the other hand, the cell market is experiencing gradual release of N-type cell production capacity, with expectations of an explosive demand for N-type products. As for modules, both domestic and overseas demands for installed capacity are projected to improve, indicating a positive turn in the market outlook for the latter half of 2023.