SK hynix, as the market leader in HBM, targets to begin mass production of its GDDR7 chips in the fourth quarter of 2024, the company said on 13th June.

In the meantime, Micron also announced the launch of its GDDR7 graphics memory at Computex, which is currently being sampled. According to AnandTech, Micron not only plans to start mass production for GDDR7 this year, but also aims to do so early enough for some customers to ship finished products by year-end, with major applications range from AI and gaming to high-performance computing.

Samsung, on the other hand, is the first among the Big Three to present its GDDR7 products. According to its press release, Samsung has completed development of the industry’s first GDDR7 DRAM in July, 2023, a 16-gigabit  product, after its development of the industry’s first 24Gbps GDDR6 DRAM in 2022. According to AnandTech, Samsung is already sampling GDDR7 memory with the aim of launching it in 2024.

According to a report from AnandTech, SK hynix already has sample chips available for partners to test. Currently, the company plans to produce both 16Gbit and 24Gbit chips, with data transfer rates of up to 40 GT/s. As Samsung and Micron both expect to begin with 16Gbit chips running at 32 GT/s for their GDDR7 products, whether SK hynix could win customers’ favor by its faster speed attracts attention, AnandTech noted.

Read more

• [News] SK Hynix Reportedly Raises Prices Again, with DRAM Products to Increase by 15-20%
• [News] New Standard for DDR6 Memory to Come out Soon

(Photo credit: SK hynix)

As the standard DRAM market experiences an unprecedented cycle of supply-demand imbalance, the shortage of DDR3 production capacity has become even more severe.

According to a report from the Economic Daily News, with leading manufacturers like Samsung exiting DDR3 production, while demand for DDR3 from AI and edge computing devices continuing to increase, the storage capacity per single device is rising sharply. This is expected to drive a rebound in DDR3 prices, potentially benefiting related Taiwanese manufacturers such as Winbond, Elite Semiconductor Microelectronics Technology (ESMT), and Etron.

In response to the shift of operational focus to high-bandwidth memory (HBM) and DDR5, the world’s top three memory manufacturers are gradually phasing out the DDR3 market.

Reportedly, Samsung has informed customers that it will cease DDR3 production by the end of the second quarter. SK Hynix had already converted its DDR3 production at its Wuxi plant in China to DDR4 by the end of last year. Meanwhile, Micron has significantly reduced its DDR3 supply to expand its DDR5 and HBM production capacity.

As per industry sources cited in the same report, it’s said that as the reduction in production by major DRAM manufacturers continues to take effect, it has driven standard DRAM prices up from the second half of 2023 to the present, with further increases expected.

Thus, prices for niche memory like DDR3 tend to lag behind standard DRAM by one to two quarters. For Taiwanese manufacturers such as Winbond, ESMT, and Etron, which focus on DDR3, the benefits of DDR3 price increases will gradually become apparent this quarter and next.

The industry sources cited by the same report also point out that DDR3 applications remain quite widespread. For example, WiFi 6 devices predominantly uses DDR3, and the next generation, WiFi 7 devices, will still primarily use DDR3/DDR4. Additionally, edge computing devices would continue to adopt DDR3. With supply significantly decreasing while demand remaining strong, DDR3 prices are expected to continue their upward trend.

Read more

• [News] Controlled Market Demand Amid Earthquake Impact, Second Quarter DRAM Prices Expected to Rise by Over 20%
• [News] Increasing HBM Capacity to Sustain Market Share, SK Hynix Rumored to Upgrade Wuxi Plant

(Photo credit: Samsung)

Due to the impact of international situations and uncontrollable factors, the global semiconductor supply chain is undergoing a shift. According to a report from WeChat account DRAMeXchange, the Southeast Asian region, with its advantages in labor and development conditions, has become the preferred location for major global companies. Countries such as Malaysia, India, and Singapore have been targeted by many manufacturers, who are rapidly setting up operations to secure a foothold.

On June 5, Taiwan-based contract chipmaker Vanguard International Semiconductor Corp. (VIS) announced to team up with Netherlands-based semiconductor supplier NXP Semiconductors N.V. to set up a joint venture, VisionPower Semiconductor Manufacturing Company (VSMC), and build a 12-inch fab in Singapore.

The fab will have an investment of approximately USD 7.8 billion. VIS will invest USD 2.4 billion and take a 60% stake, with NXP to invest USD 1.6 billion and a 40% share. The fab will be operated by VIS.

Besides, both parties have promised to allocate a total of USD 1.9 billion of long-term capacity security deposit and usage fees, with the remaining funds (Loans included) to be provided by third parties.

VSMC will run as an independent wafer manufacturing service provider, offering a certain proportion of its capacity to both partners. By 2029, the fab’s monthly 12-inch wafer capacity is expected to reach 55,000 pieces, which is projected to create around 1,500 jobs in Singapore. Following the successful mass production of the first fab, both sides will consider building a second one.

This fab will use 130nm to 40nm technologies to produce mixed-signal, power management, and analog products for markets including automotive, industrial, consumer electronics, and mobile terminals. Relevant technology licensing and transfers are expected to come from TSMC. VSMC will commence construction of the first fab in 2H24 , pending approval from relevant regulatory authorities, and it is expected to start mass production in 2027.

Currently, VIS has five 8-inch fabs, respectively located in Taiwan and Singapore. Three of them are based in Hsinchu (Taiwan) and one in Taoyuan (Taiwan). In 2023, the average monthly capacity was about 279,000 8-inch wafers.

On this collaboration with NXP, VIS Chairman Fang Leuh stated that both parties wish to own a 12-inch fab as they currently only have 8-inch fabs. More than half of the new fab’s capacity has already reserved upon long-term commitments from customers, including NXP. He also noted that setting up a fab in Singapore offers several advantages.

Since VIS is held by TSMC, industry experts believe that the establishment of the new VIS fab is driven in part by the need to meet the demands of TSMC’s mature process customers. Mature processes above 90nm account for a small single-digit percentage of TSMC’s revenue but retaining all customers is also necessary to match orders from various manufacturing capacities.

As such, VIS will take over TSMC’s customer orders. Influenced by multiple factors, the order transfer effect is expanding, and VIS has recently received new orders from several customers, like Qualcomm and MPS. That means order transfer effect in 2H24 has become evident.

It is worth noting that Singapore is being seen as a critical hub of the Asian semiconductor industry. It currently boasts a complete semiconductor industry chain, covering design, manufacturing, packaging, test, equipment, materials, and distribution, with more than 300 semiconductor-related companies already established.

According to another report from WeChat account DRAMeXchange, multitudes of semiconductor companies, including Texas Instruments, STMicroelectronics, Infineon, Micron, GlobalFoundries, TSMC, UMC, VIS, and ASE, have set up branches or expanded production in Singapore.

Read more

• [News] TSMC Affiliate VIS and NXP to Invest USD 7.8 Billion for 12-Inch Fab in Singapore, Expected Mass Production by 2027
• Taiwanese Chipmakers Expand Overseas to Capitalize on Geopolitical Shifts and De-Sinicization Benefits, Says TrendForce

(Photo credit: VIS)

According to a report from TechNews, South Korean memory giant SK Hynix is participating in COMPUTEX 2024 for the first time, showcasing the latest HBM3e memory and MR-MUF technology (Mass Re-flow Molded Underfill), and revealing that hybrid bonding will play a crucial role in chip stacking.

MR-MUF technology attaches semiconductor chips to circuits, using EMC (liquid epoxy molding compound) to fill gaps between chips or between chips and bumps during stacking. Currently, MR-MUF technology enables tighter chip stacking, improving heat dissipation performance by 10%, energy efficiency by 10%, achieving a product capacity of 36GB, and allowing for the stacking of up to 12 layers.

In contrast, competitors like Samsung and Micron use TC-NCF technology (thermal compression with non-conductive film), which requires high temperatures and high pressure to solidify materials before melting them, followed by cleaning. This process involves more than 2-3 steps, whereas MR-MUF completes the process in one step without needing cleaning. As per SK Hynix, compared to NCF, MR-MUF has approximately twice the thermal conductivity, significantly impacting process speed and yield.

As the number of stacking layers increases, the HBM package thickness is limited to 775 micrometers (μm). Therefore, memory manufacturers must consider how to stack more layers within a certain height, which poses a significant challenge to current packaging technology. Hybrid bonding is likely to become one of the solutions.

The current technology uses micro bump materials to connect DRAM modules, but hybrid bonding can eliminate the need for micro bumps, significantly reducing chip thickness.

SK Hynix has revealed that in future chip stacking, bumps will be eliminated and special materials will be used to fill and connect the chips. This material, similar to a liquid or glue, will provide both heat dissipation and chip protection, resulting in a thinner overall chip stack.

SK Hynix plans to begin mass production of 16-layer HBM4 memory in 2026, using hybrid bonding to stack more DRAM layers. Kim Gwi-wook, head of SK Hynix’s advanced HBM technology team, noted that they are currently researching hybrid bonding and MR-MUF for HBM4, but yield rates are not yet high. If customers require products with more than 20 layers, due to thickness limitations, new processes might be necessary. However, at COMPUTEX, SK Hynix expressed optimism that hybrid bonding technology could potentially allow stacking of more than 20 layers without exceeding 775 micrometers.

Per a report from Korean media Maeil Business Newspaper, HBM4E is expected to be a 16-20 layer product, potentially debuting in 2028. SK Hynix plans to apply 10nm-class 1c DRAM in HBM4E for the first time, significantly increasing memory capacity.

Read more

• [News] Reasons for Samsung’s HBM Chips Failing Nvidia Tests Revealed, Reportedly Due to Heat and Power Consumption Issues
• [News] HBM Boom May Lead to DRAM Shortages in the Second Half of the Year

(Photo credit: SK Hynix)

Driven by the rapid growth in demand for high-bandwidth memory (HBM) fueled by artificial intelligence (AI), memory manufacturers are vying for market opportunities. According to a report from CNA, Micron has announced its target to achieve a 20% to 25% market share in HBM by 2025.

Targeting the swiftly growing demand for HBM, along with its better product pricing and profitability, the three major memory manufacturers—SK Hynix, Micron, and Samsung—are all aggressively advancing in this area. Currently, SK Hynix holds the leading position, but Micron is also making significant progress.

Micron stated that its progress of HBM3e could be contributed to the company’s advanced packaging and design capabilities, along with the integration of its own processes. The company is also developing the next generation HBM4 products.

Regarding Micron’s global capacity expansion plan, the memory heavyweight has been considering Hiroshima, Japan, as one of the potential sites. The company’s HBM capacity in fiscal year 2024 has already been sold out, of which is expected to contribute hundreds of millions of dollars in revenue.

As for its ambition regarding HBM, Micron stated that it aims to capture 20-25% market share by 2025.

Notably, per a previous report from the South Korean newspaper “Korea Joongang Daily,” following Micron’s initiation of mass production of the latest high-bandwidth memory HBM3e in February 2024, it has recently secured an order from NVIDIA for the H200 AI GPU. It is understood that NVIDIA’s upcoming H200 processor will utilize the latest HBM3e, which are more powerful than the HBM3 used in the H100 processor.

During the press conference on 5 June, Micron announced the launch of its GDDR7 graphics memory, which is currently being sampled. Utilizing Micron’s 1-beta technology, GDDR7 offers more than a 50% improvement in energy efficiency compared to the previous generation GDDR6, effectively addressing thermal issues and extending battery life.

Micron highlighted that the GDDR7 system bandwidth is increased to 1.5TB per second, 60% higher than GDDR6. Its applications range from AI and gaming to high-performance computing. The product is expected to start shipping in the second half of this year.

Read more

• [News] Heated Competition Driven by the Booming AI Market: A Quick Glance at HBM Giants’ Latest Moves, and What’s Next
• [News] Micron Slightly Raises Capital Expenditure for 2024, HBM Expected to Further Drive Revenue Growth in 2025

(Photo credit: Micron)